Document VZrv6JYegpeL5NnYKZ4xpX58

GENERAL ELECTRIC DIOXIN AND FURAN 6-3-93 PRODUCTION BATES # NPC # 784089 - 784645 GENP# 011278 - 011834  I I I / I II I 784089 GENP'Ol 1278 Dioxins and Furans: Questions and Answers Todd Paddock Academy of Natural Sciences Nineteenth and the Parkway Philadelphia, PA 19103 Dioxins and F u m i E N P 011279 784090 1989 Academy o f Natural Sciences o f Philadelphia ISBN: 0-910006-08-3 L itany of Congress Catalog Card Number 89-84471 Printed in the United States of America ii Dioxins and Furans I : l I 784091 OENP 0 119Jtn T TABLE OF CONTENTS List of F ig u re s ..................................................................................................................................... v L in of T ab les........................................................................................................................................ v Acknowledgements ............................................................................................................................. vi Executive Sum m ery............................................................................................................................. vii Dioxins and Furans: Introduction and a Brief History .................. 1 What is the history of dioxins and fu ra n s ? ....................................................................... 2 Discovering their toxic effects on humans and animals .................................. 2 Chick edema disease .............................................................................................. 2 Concern about 2,4,5-T and related h o b ic id e s..................................................... 3 Tunes Beach and other areas in Missouri .......................................................... 4 Agent O range................................................................................................... 7 The Scveso accident .............................................................................................. 9 C o n c lu s io n ...................................... 10 Detecting Trace Amounts of Dioxins and F u ra n s .......................................................... 11 Dioxins and Furans What They Are, Where They Are Found, and How They Behave . . . What is d io x in ? ................................................................................................................... Are all dioxins and furans dangerous?............................. .............................................. What factors affect the toxicity of dioxins and furans? ............................................... Where are dioxins and furans found?............................................................................... Are dioxins and furans found in animals and plants? .................................................. How long do dioxins and furans last in the environm ent?..................... ................... To what degree do dioxins and furans spread when they are introduced into the environm ent?.................................................................... ................................... Are dioxins and furans found in hum ans?....................................................................... How do dioxins and furans reach humans? ....................................................... 13 13 13 17 19 20 23 24 26 26 Sources of Dioxins and Furans .............................................................................................. How are dioxins and furans c rea te d ? ............................................................................... What products can be contaminated with dioxins andfurans?....................................... PCBs ...................................................................................................................... Hexacfalorophene ................................................................................................. 2,4-triclorophaioxyacetic acid (2,4J-T ) ....................................................... 2,4-didUoroprienoxyaccdc acid (2 ,4 -D ).......................... Chlorophenols ....................................................................................................... Polychlorinated b e n z e n e s .................................................................................... Diphenyl ether h e ib ic u k i.................................................................................... Hexacfalorocydohexxne ....................................................................................... Paper mills and products .............................................................................. What products can produce dioxins and furans whenthey are burned?....................... PCBs ...................................................................................................................... Poiybrominaied biphenyls (PBBs) and polybrominated diphenyl ethers (PBD FEs)................................................................................................. Polychlorinated diphenyl ethers (PC D Es).......................................................... Chlorophenols ...................................................................................................... TetracMoroethytene and polychlorinated benzenes.......................................... Polyvinyl chloride ( P V Q .................................................................................... Leaded g a s o lin e .................................................................................................... The burning of dioxins and furans .................................................................... Does the burning o f paper, wood, peat, or coal producedioxins and furans?............ 29 29 30 31 32 32 33 34 36 37 37 . 37 38 39 39 40 40 40 41 41 42 42 Dioxins ri Furans lit 784092 B N P 011281 Do municipal incineraron produce dioxin* and fu ran s7 ............................................... Do other large combustion sources produce dioxin and fu ra ro ? ................................ What about dioxins and fuians from sewage plants, waste streams, and landHUs? ............................................................................................................... Are there other processes that create dioxins and furans? ........................................... Conclusions on sources of dioxins and f u r a n s ............................................................... 43 44 45 45 46 The Health Effects o f Dioxins and Furans, in Animals and H u m a n ........................................ What are the health effects on an im als?.......................................................................... Some general features ......................................................................................... Doses that produce no observable adverse health effects ............................. The short-term health effects on a n im a ls.......................................................... C a n e e n ................................................................................................................... Reproductive effects ..................................... .................................................... Mutagenic e f fe c ts ................................................................................................ Immune effects............................. ........................................................................ Effects on blood constituents............................................................................... Are dioxins and furans hazardous to organisms in the environm ent?................ What are the pharmacokinetics of dioxins and furans? ....................... What are the effects of dioxins and furans an human h e alth ? ..................................... The shan-term health effects on h u m a n s.......................................................... The long-term health effects on humans .......................................................... C ancre................................................................................................................... Abortions, birth defects ............................. ........................................... Immune eiffects...................................................................................................... Conclusions on health e ffe c ts............................................................................................ How do dioxins and furans cause the health effects that they d o ? ............................. 48 48 48 50 51 52 53 54 54 54 55 56 59 60 60 61 62 63 63 64 Regulating Dioxins and Fuians .............................................................................. ...................... How do government agencies arrive at acceptable human exposure to dioxins and furans? ............................................................................................................ What has been done about dioxins and fu ra n s ? ............................................................. 65 65 69 Conclusions and Recommendations ............................................................................................... 72 Liierenire Cited .............................................................................................................. .................. 74 A List of Documents Read but not C ite d .............................................' ........................................ 88 A ppendix.............................................................................................................................................. 91 Recommended Reading .................................................................................................................... 92 Glossuy and List of Abbreviations................................................................................................. 93 Index to A utbore.................................................................. ........................................................... .. 94 Index to Subjects .............................................................................................................................. 9 7 . iv Dioxins and Furans G E N P 011282 784093 Lfat of F T turq Figure 1. Dibenzo-p-dioxin.................... Figure 2. Oibenzofuren ................................................................................................................... Figure 3 . 2 J.7 .8 -T C D D ................................................................................................................... Figure 4. Virtually safe lifetime daily doses of 2J.73-TC D D from various agendas and countries ............................... 13 13 14 66 List of Tables Table 1. Relative toxicity of dioxins and f u r a n s ................... Table 2. Products that can be contaminated with dioxins and/or f m a n s .................................. Table 3. Chlorinated products that can produce dioxins and/or furans when bunted . . . . . . Table 4. Acutely lethal single doses of 2,3.7,8-TCDD (L D 5 0 ).................................................. Table 3. Lethal exposures to 2^,7,8-TCDD in water (L C 50)..................................................... Table 6. Lowest dose with an observed adverse effect (L O A E L )............................................. Table 7. Lowest concentration with observed adverse effect (LOAEL) .................................. Table 8. Highest concentration with no observed adverse effect (N O A E L).......................... IS 30 39 49 49 50 51 51 Dioxins ind Furans v 784094 G BN p0U283 Acknowledgem ent! I wish to extend my sincere gratitude to those people who reviewed sections of this report: Dr. Ken Burgess of Dow Chemical Company, Dr. David Firestone of the USFDA, Dr. Annette Guiseppi-Elie of Drexel University. Dr. Rolf Halting of the Univcnity of Michigan, Dr. Paul Michael of Monsanto Company, Dr. Tom Robinson of Vulcan Chemicals, Dr. Steven Safe of Texas A & M University, Dr. Fred Tscfairley, Professor Emeritus o f Michigat State University, and Mr. John Wilkinson of Vulcan Chemicals. Their comments greatly improved the report. I also wish to thank Dr. Robert Baughman of Harvard University, Dr. P.E. des Rosiers of the USEPA, and Dr. Alan Poland of the University of Wisconsin far their many explanations, ideas, and advice. These people and many others took time from their busy schedules to answer my numerous questions. I especially wish to thank Dr. Ruth Patrick and the Environmental Associates at the Academy of Natural Sciences in Philadelphia, for providing me with the resources and freedom .to write this report. I greatly appreciate the opportunity they gave me, and the constructive criticism that 1 received from than. vi Dioxins and Farm s GENP 011284 784095 i Executive Sum m inr In the U.S., dioxin is infamous because it was the substance in Agent Orange blamed for a host of riiwa-w in Vietnam veterans, and the substance that contaminated Tunes Beach, Missouri, prompting the federal government to purchase the entire town. Since then, the discovery of dioxins in incinerator emissions has caused further concern. There are actually many kinds of dioxins. The chlorinated dioxins have caused the most concern. A related family of compounds called chlorinated furans are very similar to dioxins in their distribution, toxicity, and behavior, although not as much is known about them. The chlorinated dioxins and furans are the subject of this report. One particular dioxin has received more attention than any other, because it was found in relatively large amounts in some widely used chemical products, and appears to be the most toxic. This dioxin is 2J.73-TC D D . The word dioxin is commonly used to mean 2J,7,STCDD. A great deal more research has been done on 2,3.73-TCDD than on the other dioxins and furans, and this report is based largely an studies of 2J.73-TC D D . How toxic are dioxins and furans? Based on laboratory and field studies, 20 or so o f the 210 chlorinated dioxins and furans are considered to be extremely toxic to animals. They accumulate in the tissues of organisms, and are often the only types found in organisms from the wild. Most dioxins and furans appear to be almost nontoxic, and do not accumulate; W hat factors affect the taxkity of dioxins and furans? The toxic effect of a dinxin or furan on an animal depends on more than the type of dioxin or frnaa. It also depends on the species of animal, the status of the animal, the route of exposure, and the substance in which the dioxin or funn is present Some animal species are ihnmanrfx of times mare sensitive than others to the toxic effects o f the same dioxin or furan. How do dkedas and furans reach hamass? The major routes of exposure are generally considered to be eating contaminated food and breathing contaminated particles. Panicles in air, especially near urban areas or large sources of combustion products, can be contaminated with dioxins and furans. Trace levels of Dioxins and Furans vii ^ 784096 GB^P 0 l i28s dioxins and furaru have been found in a variety of foods in Japan, Canada, and Europe, and the same is probably true of foods in the U.S. Dioxins have also been found in fish from certain an as in the U-S. Are dioxins and furans found (n humans? Dioxins and furans have been found in samples of blood and fatty tissue from persons in numerous countries, including the U.S., Canada, Japan, Vietnam, and several countries in Europe. Based on these findings, it is generally agreed that the population at large in many pans o f the world has been exposed to low levels of dioxins and furans. W hat are the health effects of dioxins and furans7 The long-term health effects of these compounds have been intensely investigated, and they are now among the most-studied substances with respect to their effect on animal and human health. Dioxins and furans cause mortality and/or cancer in many animals, but th o e is no conclusive evidence that they cause cancer or any other life-threatening health problem in humans. No documented human death has occurred because of s i exposure to dioxins or furans, People have been exposed to relatively large amounts of dioxins and furans (1 microgram or more) during industrial accidents, and other incidents of accidental contamination. The health effects were serious, but subsided with time. With the exception of severe acne, most researchers have found no serious, long-term health problems associated with exposure to dioxins, even at the highest reported levels, and after ten to thirty yean have elapsed. How are dioxins and ftirmns created? Dioxins and furans have no useful purpose and have never been msnufactixcd deliberately, except in small amounts for research purposes. They are known to be created in two major ways: (1) they are created by unwanted side reactions, which take place during the chemical processes used to manufacture useful products such as bioddes, paper, dismfectantt, and preservatives; (2) they a s created when a snbsancc containing chlorine is burned, cr when a substance is tu n e d in the presence o f chlorine. viii Dioxins and Furans GENP 011286 784097 Dioxins and/or (irons have been found as contaminants in the following m anufctuied products: Product PCBs hexachlorophene 2.4.5-T 2,4-D chlorophenols with 3 or mote chlorines polychlorinated benzenes diphenyl ether herbicides hexachlorocyclohexane paper products * I could find no analyses of these products far firons. Contaminants Furans Dioxins* Dioxins* Dioxins* Dioxins and Furans Other products derived from chlorophenols have been identified as possibly contaminated and are under investigation by the U.S. Environmental Protection Agency (USEPA) and others. W hat products can produce dioxins and (brans when they are burned? The following chlorinated products or substances can produce dioxins and/or firons when burned: Product Contaminants PCBs polybrominated biphenyls (PBBs) polybrominated diphenyl ethers (pBDEs) polychlorinated diphenyl ethen (PCDEs) chlorophenols tetracfaloroethylene polychlorinated benzenes polyvinyl chloride (PV Q Leaded gasolines paper, wood, peat, and coal dioxins and furans Furans Dioxins and Furans '" ** *" "" *" ** ** ** Note: When a material is burned in the laboratory to determine whether it is capable of producing dioxins and furans, it is typically burned under those conditions most likely to produce dioxins and furans, or that produce the largest quantity of dioxins and furans. Therefore, the same substance may produce much smaller amounts of dioxins and furans under the uncontrolled conditions o f a fire, or in the highly controlled conditions of an incinerator. Dioxins and Furans ix 784098 r, Incinerators and other large combustion sources, sewage plants, waste streams, and landfills are ilsp sources o f dioxins and furans. There are many sources of dioxins and furans. The combustion sources are widespread and numerous. Highly contaminated manufacturing and waste sites are fewer in number and usually local problems. The relative importance of the two major sources of dioxins and furans (chlcrophenols vs. combustion) is not yet known. It is generally agreed that modem industrial activity, not the burning o f wood, is the primary source o f dioxins and furans to the environment ' W here are dioxins and furans found? Dioxins and furans are present at trace levels in many places, and at relatively high levels at a small number of contaminated sites. The presence of dioxins and furans is usually associated with the production, storage, use, or disposal of chlorophenoUbased compounds. In addition, they have been detected at low levels in urban areas, where they are believed to be the product of numerous combustion processes. As our ability to detect dioxins and furans improves, it is likely that they will-be found in more pieces. Dioxins and furans have been found in argaiisms from contaminated areas, and from fish in many areas of the U.S. Plants typically contain these compounds at levels much below the surrounding environment, and only when they are grown in highly contaminated soil. Fruits do not appear to contain dkrxins, although root crops may sometimes contain them. How long do and A n n s last in the environm eet? 2J.73-TC D D can be broksi down in a imrwr of days in the environment, if it is mixed with the proper organic solvent and e xpend to ultraviolet light. On the surface of the soil, without a solvent. 2J.7.8-TCDD has an environmental halMife o f one year or less, due to slow photodegrndarion and evaporation. Underground and in sediments, this very stable compound degrades much mare slowly, with a halMife of about 10 years. Very little is known a to m the halMives o f other dioxins and furans. In general, dioxins and furans with fewer than four chlorines are broken down mare quickly than 2J.73-TC D D . The dioxin or furan with eight chlorines is mare resistant to breakdown. x Dioxins and Furans L jENP 011288 784099 V To what degree do dioxins aod faraas spread when they ara Introduced Into the eaefroM ieat? Dioxins and farm s bind very s&oo^y to the organic carbon present in panicles of soil, in sediments, and to the panicles found in the air. These panicles can be lifted and earned by wind or moving water. Dioxins snd farm s can be transported gresc distances in air, and smaller distances in water. In m il and sediments, diMiiM and forms appear to migrate very little or not at alL Areas of severe contamination have typically remained local problems, and mast are not expected to spread the contamination to nearby areas. W hat has betn done about din!TM and faring? Because they are very potent toxins and carcinogens in laboratory animals, regulatory agencies from various countries have issued guidelines for human exposure to dioxins and fumra. These guidelines vary by mare than a thousandfold, due to different interpretations of the same animal, lab o ao ry studies. As rfinvin and hum s have been trim"*"* u contaminants in products, the mronfacturer* have lowered the levels of contamination. The USEPA has cancelled or restricted the manufacture, use, and Hiprmt of many of the products enntaiwtng hazardous levels of dioxins and farm s, and the disposal of wastes from the manufacture of many products. Many contaminated sites have been cleaned up, and many mare are oa the Superfimd list. CoodasioQS and Recommendations Based oa the many studies already published, dioxins and farm s do not appear to be a haxvd to otv health when we am exposed to the low levels typically present in the environme n t Several group* o f people need special anmrinn became they may be exposed lo g re a ts thm nom al amount* of dioxins rod farms. These indude breaa-fad babies (because dioxins snd A n n s are usually present in t r a f t milk), and a o ric ai in industries that use contaminated products. We should continue wrienr the health of group* of people who w e n exposed to large amounts o f these compounds, s who were expored to smaller amounts far a long time. This will enable us to iderxify any tag-term health effects not already recognised. Now that Dioxins and Furans xi 784100 GENP 011289 wc have the ability to determine the extent of aa individual's past exposure to dioxins and finals through a blood tes^ the identification of exposed persons is much easier and mote accurate, making such studies more powerful We also need to determine how dioxins and farm s cause the effects that they do, and why their toxicity varies so greatly between species. Such an understanding would not only have preventive or therapeutic value, but is also critical far accurate assessments of the risks these compounds pose to humans and other organisms. Because of their demonstrated toxicity to laboratory animals, plus their distribution and persistence, some dioxins and furans are a potential hazard to organisms in the environment. There is a great need for more study in this area. We need to know what are unacceptably harmful levels for the various organisms in the environment, and the nature and amount of dioxins and furans from the various sources. What appears to be the largest amount of toxic dioxins and furans is buried in landfills or stored at past or present sites of manufacture or use. We need to develop safe, yet practical ways to contain, or better yet destroy, this contaminated waste. We may need to reduce the amounts of dioxins and furans occurring as unwanted contaminants of some products. Great reductions have been made in certain products in the past, and further reduction may be necessary. Unfortunately, such reduction can result in very highly contaminated waste streams, and proper disposal is difficult and expensive; Therefore, the emphasis needs to be on preventing their creation during the manufacturing process. We also need to reduce the unwarned creation o f dioxins and farm s during combustion. For example, incinerators can be o p a wed under conditions that reduce the c t i n n of dioxins and farms. More than one billion dollars has been spent in the U.S. alone far the research of dioxins and farm s, and this research has greatly cxpmrtrrf oar knowledge. With this knowledge, and with improved technologies far both reducing future sources of these com pom di and cleaning up presently contaminated areas, dioxins and farm s should be regarded as a manageable problem. xii Dioxins and Furans 784101 Dioxins and F ttraa c Introduction and a Brief History A When most people use the to m dioxin, they mean a single chemical compound, often refened to as *tbe most toxic substance ever made by m in '. In the U.S., dioxin is infamous bmiTVi it was the substance in Agent Orange blamed for a host of diseases in Vietnam veterans, and the substance that contaminated Times Beach, Missouri, prompting the Federal Government to purchase the entire town. Howevo', the dioxins (short far chlorinated dibenzo-p- dioxins) are a large family of compounds whose toxicity varies or, for many types, is unknown. * A closely related and very similar group is the furans (chlorinated dibenzofurans). Dioxins and furans have been discovered in unexpected places. They are in fish from lakes and rivers, in common paper products, in auto exhaust, and in the ash from municipal and industrial indnentors. Dioxins and Auans will continue to be found in new places. In part, this is they can be measured in extremely small amounts, and have been looked for in many places. Also, many dioxins and furans are very stable, can remain in the environment for years, and can wmimtiiaw in organisms. One particular dioxin has received more mention than any other, because it was found in relatively large amounts in some widely used chemical products, and appears to be the most toxic1. This dioxin is 2 J ,7,8-ietnchlorudibenzo-p-diaxin, also known as 2.3,7,8-TCDD, or simply TCDD. The word dioxin is commonly used to mean 2J.73-TC D D . Much of the general public's knowledge of dioxins sad funas was gained as a result of a few widely-pobtietzed inddatts o r controversies. A brief history of dioxins and frirana, with stannaries o f a few of the most publicized inddentt, can serve to remind us of how these oooipouoda became so well known. 1 Here and in the r a t of the report, "toxic* means that the substance is assoc wed with advene health effects, but not necessarily effects that are serious, or permanent. Dioxins and Furans 1 1 L 784102 W hat b Um hbto ry of dioxins to d furans? h Dioxins and fuism were synthesized by organic chemists in the 1930s and were patented in 1939 as constituents in a product used for electric insulation (cited in Huff and W asson, 1973), Different dioxins and fuians were tested and in some cases patented as agena against bacteria, inseca , and fungi, as intermediates in chemical processes, and as flame retardants, from the 1930s to at least 1970 (cited in Huff and Wassom, 1973). Discovering their toxic effects on humans and animals - The toxic effects o f dioxins and fuians in humans were mare or less unrecognized until 1957, when Dr. Karl Schultz, a dermatologist at the University of Hamburg in West Germany, tried to determine why woriceis at a nearby chemical plant were developing a severe form of acne, called chicracne (Gough, 1986: pp 29-33), After some careful tests, Schultz concluded 1 that a contaminant in one of the plant's products (2,4,5-triehIorophenoI) was responsible for the chloracne, and identified the contaminant as 23,7,8-TCDD. We now know that 2,4,5t trichloropbenol is always contaminated with 2J,7,8-TCDD, and therefore trace levels of 2J.7.8TCDD are also present in the products made from 2,4J-trichktrophenoL Schultz and colleagues published several papers in 1957, showing that several dioxins and furans induced acne when applied to rabbit ears (the standard and most sensitive test at that time), and that small doses of 2J.7J8-TCDD caused liver damage and death in rabbits: The plant d a n found a way to reduce the amount of the contaminants fam ed during the production of their 2,4J-trichlarophenol, and the chloncns problem in their workers subsided. Chick edema disease Also in 1957, scientists iu the U.S. Food and Drag Adminisuatian (USFDA) were trying to deunnine the cause of in oatfareak of millk m of deaths o f chickens in the U.S. (Fresipoe, 1973). They determined in 1958 that the cause of the disease (chick edema) was toxic conom ininu in commercial fatty adds that were pan of the chicken feed. Symptoms included excessive Quid in the heart sac and abdominal cavity, and liver damage. Although the contaminant was not yet identified, the USFDA issued a regulation in 1960 that required fatty 2 Dioxins and Furans Lk 784103 a d d s to be proven free of the contaminants through a 3-week chick-feeding bioassay. Advarees io detection techniquea mads U possible to further identify the contaminants, i n i in 1966 it was determined that l,2J.7,8.9-hesach krodibenzo-p-dioxin was one of the cnmmi>qnn and that a synthetic hcxactilorodibcnzo-p-dioxin produced the disease in chickens (Ftresume, 1973). It had been shown as early as 1936 (cited in Huff and W asson, 1973) that dioxins could be produced by heating chloropbenols, and the USFDA scientists proposal that chlorophenols were the source of the dioxins in the toxic fats being fed to chickens. Chloropbenols, including trichlorophenol and pentachlaropbenol. were widely used as agents against slimes, bacteria, termites, and as herbicides. The USFDA scientists were able to produce a variety of dioxins by heating various commercial chloropbenols, and found these same rfimnrn in the toxic fat. They tested these dioxins on chickens, found that they produced chick edana, and found that the most toxic dioxin was 2J.7JJ-TCDD. Finally, it was determined in 1972 that the source of chick edema disease was fat derived from hides (bom cattle, hogs, sheep, and other animals) that had been treated with commercial pemachlorophenol (Metcalfe, 1972, cited in Firestone, 1973). Concern abort 2.4.5-T and H ated hwhiridM In 1970, at hearings before the U.S. Congress, evidence was presented to show that the widely-used herbicide 2,4,5-trichIarophenoxyacetic acid (2,4,5-T) was capable of causing birth defects in rail and mice (cited in Huff sod Wassom, 1973). During that same year, a study showing similar results was published (Courtney et sL, 1970). It was propoud that the relarivefy high levels (ppraximxiely 30 pans per million [ppm]) of 2J.73-TC D D comominarian in 2A.J-T w o e probably responsible for the birth defects (cleft palate, cystic kydney) in mice and r e s , and a study an 2J.7JS-TCDD found that it affected reproduction in rats (S p n c h n et iL , 197Q), Based an these findings and the earlier findings of the toxicity of 2J.7.8-TCDD to humans and animals, the U.S. Depotment of Agriculture (USDA) decided in 1970 to cancel legisuMiuns far the use of 2,4,5-T in certain areas, far example its use an human food crops, near bodies o f water, and around homes (Cough, 1986: p 138). They did not caned registniioa Dioxins Fumns 3 784104 GETSfP Tor other usee of 2,4,5-T if the levels of 2J.7.8-TCDD were below 1 ppm (cited in Huff and W asson, 1973). However, the USDA id U.S. Department of Health, Education, and Welfare issued a joint statement that 2,4,5-T and 2^3,7,8-TCDD may cause birth defects in mice and rats (cited in Huff and Wassom, 1973). Also in 1970, the U.S. Department of Defense stopped using the herbicide Agent Orange in Vietnam. Agent Orange was a half-and-half mixture of 2.4.5- T and 2,4-dichlorophenoxyacetic a d d (2,4-D), and was contaminated with 2J,7$-TC D D . In 1971. a Science Advisory Council from the U.S. National Academy of Sdences recommended to the U.S. Environmental Protection Agency (USEPA) that the registration of 2.4.5- T be restored for use on forests, rangeland, and rice fields, with certain restrictions on its use (Gough, 1986: p 138). For example, they required that present stocks of 2,4,5-T have less than 0.5 ppm 2,3,7,8-TCDD and that newly manufactured 2,4,5-T have' less than 0.1 ppm (cited in Huff and Wassom, 1973). In 1973, the Science Advistny Council recommended that no t more than 0.1 ppm 2,4,5-T be allowed in water or on edible food products, if they were for human consumption (cited in Huff and Wassom, 1973). In 1979, the USEPA issued an emogency suspension of most remaining uses of 2,4,5T, based on accumulating evidence that 23,7,8-TCDD caused cancer and birth defects in laboratory animals, and because of a 1979 study that showed elevated levels of spontaneous abortions among women near Alsea. Oregon, where 2,4,5-T was sprayed (Gough, 1986: p 138145). A panel of scientists subsequently reviewed this 1979 study and concluded that because of a flawed design, the study was incapable of showing either the presence or absence of effects of s p a s m to 2,4,5-T (Coulssm and Olajos, 1980, cited in USEPA, 1985: p 9-25). From 1979 until 1984, 2,4,5-T was used on rice and sugarcane fields, pending settlement of a suit between the USEPA and manufacturers of 2,4,5-T. All companies withdrew from the suit by 1984, and the USEPA then cancelled all uses of 2,4,5-T and the closely related herbicide Silver. Times Beach and other areas in Missouri (Mach o f the following discussion is based on a chapter in Gough, 1986) Heiachlcrophene, a germicidal product used as a skin disinfectant in hospitals; in germicidal soaps, and in some veterinary products, is produced bom 2,4,5-trichkxopbenol, the 4 Dioxins and Furans 784105 > same chemical used to manufacture 2,4.5-T and related h o b tid e s. When manufactures of 2.4.5-eichlorophenol-baaed products became aware of the toxic properties of 2J.73-TCDD, they found ways to greatly lower the concentrations in their final products. However, this purification aocesi can create wastes with very high concentrations of 2.3,7,8-TCDD. If this wasre is not disposed of property, it can be a hazard to humans and other organisms. Improper disposal of such wastes was the cause of the contamination of Times Beach, Missouri A chemical plant in Verona, Missouri produced hexachlorophene from 1970 to 1972. The purification of their product (by distillation) resulted in a thick, oily, 2J.7.8-TCDDcontaminated waste called still bottoms. The plant's still bottoms were hauled away by a separate company, and sold in turn to a waste oil buyer. The waste oil buyer then used the still bottoms in a secondary business, in which he sprayed oil to keep dust down in horse arenas, private roads, and parking lots. Horses, birds, and other animals became sick and died 1 following the spraying of one of the bone arenas in 1971, and a child who played in the horse arena was hospitalized with bladder inflammation and bleeding. The U.S. Centers for Disease Control (CDC) began an investigation of the poisoning episode si that stable in August, 1971, three months after the spraying. They and local doctors HingnrxM the episode as chemically related, but could not identify the disease or the contaminant. In the meantime, the owners of the first horse arena and another arena had suspected the cause of the illness was the sprayed oil, and had removed the top layer of soil from their arenas and placed it in a nereby landfill. The CDC investigation continued, and in 1973 they la d identified 2J.73-TC D D in samples o f din from the arenas. They notified the Missouri Department of Health in 1974, and the CDC and state began a joint investigation of the wpimrift. The origins of the waste ail were quickly determined by the investigator, but identifying the many areas that h id been sprayed with the oil was more difficult. Fortunately, the owners of the fust arena bad secretly followed and kept records of the spraying done by the waste oil buyer, and these records were used to locale other contaminated areas. Arenas and trailer paries had been contaminated by spraying, and roil from the arenas was removed and I Dioxins and Furans 3 784106 GENP 011295 placed in fanrffiU*, on farms, and used a nil for building projects. Also, the waste oil buyer had sprayed 23 miles of unptvcd streets in the town of Times Beach, Missouri, from 1972 to 1976. The contamimriai of Times Beach was not confirmed by the federal government until December 1982, when their soil tests showed 5*300 parts per billion (ppb) 23.73-TCD D in the soil hum the streets o f Times Beach. Based on the results, the CDC recommended that the town of Times Beach was unsafe for its inhabitants, and in February 1983 the federal government announced it would buy the homes, properties, and businesses of Tunes Beach. The many other contaminated sites in Missouri (51 in all. according to USEPA, 1987a: pp 2J* 2.35) are being dealt with in various ways. The USEPA and CDC have removed soil, decontaminated the soil, restricted access, and advised minimal contact with soil, depending on the level of contamination and possibility of exposure (USEPA, 1987a: pp 2.18-19). Based on studies of residents of the most highly contaminated areas in Missouri, it appears that the people exposed to 23,7,8-TCDD have suffered few if any long-term effects, although several commonly used areas were contaminated with extremely high levels of 23.7,8TCDD (as much as 31 ppm in the soil o f one hone arena). The people who were exposed to high levels of 23,7,8-TCDD in the arena soil suffered from a variety of effects, including nausea, headaches, diarrhea, nosebleeds, diloracne, and severe bladder inflammation. Their health subsequently ream ed to normal (reviewed in Reggtani, 1980). A study of the residents of a trailer park that had 39*1100 ppb of 23,73-TCD D in the soil found abnormal levels of some blood constituents, that suggested long-term effects on the liver and immune system (Stehr-Greca et aL, 1987), However, no liver changes could be directly detected, and no excess of illness was reported. The authors noted, T h e effects we observed may be merely, p i n of the normal, attaptive response to a toxicologic challenge, sod well within the normal reserve capwtity of the affected argm systems (Le., they are m arte n of exposure and not signs o f diseue)." 6 Dioxins and Furata GENT? 0U 296 784107 Agent Orange Herbicides were used by the U.S. in the Vietnam War from 1962 to 1970. Agent Orange, a half-and-half mixture of the herbicides 2,4,5-T and 2,4-D. was sprayed over South Vietnam ftom 196S to 1970, accounting for mare than half of all herbicides there. Earfi- mixtures were called Agents Purple, Pink, and Green. Bared oa analyses of Agent Orange manufactured before 1970, the mixture sprayed an South Vietnam was contaminated with from 0.02-54 ppm 2^,7,8-TCDD (Esposito et al, 1980: p 98). The U.S. Air Force (USAF) estimates that A gon Orange had a mean concentration of 2 ppm Z3,7,3-TCDD. Agent Purple had a mean concentration of 33 ppm, and Agents Pink and Green were estimated to have 66 ppm (Albanese, 1988: p 3). The large-sc^lc use of herbicides in Vietnam ended in 1970 because of concern about the effects on the environment and inhabitants of South Vietnam. However, that concern has been almost entirely eclipsed by concern about the possible effects on the health of soldiers from the ILS. and other countries who were exposed to large amounts of the herbicides. Because of this concern, several large studies by the CDC and USAF have been conducted or are underway. The USAF study investigated the general health and rates of cancer and death among Vietnam veterans who were exposed to herbicides, as well as birth defeas among children o f the veterans. The CDC study investigated the health of Vietnam veterans in general, most of whom were not exposed to Agent orange, and therefore is not relevant tn this discussion. The USAF Health Study is investigating the health of USAF personnel who participated in the aerial spraying of herbicides over Vietnam. The spraying program was called Operation Ranch Hind; Ranch Hands include the men who flew the planes, end all men who worked with the herbicide and spray equipment used for the aerial spraying. The health of Ranch Hands is being compared to a carefully matched group of USAF pa s o n n d who flew in and out of Southeast Asia (but not Vietnam) during the same period. The mast recera report from this study focused oa eleven health effects that studies of ari{r"1 and humans hive associated with expastae to 2J.73-TC D D (Albanese, 1988). They reported that for 6 of the 11 effects. Dioxins and Rnans 7 784108 GENP differences were detected between the two groups, and for 5 of the effects the difference was in the direction that would be expected from current knowledge of the effects of 2J.73-TC D D . The six effects were increases in the rate of cancer, increases in the number of birth defects reponed in children of the veterans, an increase of psychological changes, an increase of liver changes, an increase of cardiovascular changes, and an increase in changes in the endocrine system. However, no differences were detected in the sperm count or proportion of normal sperm, the rate of heart disease and heart attack, the immune system, overall mortality, and other health effects that studies of animals and humans have associated with exposure to 2J,7,8-TCDD. Follow up studies, including an investigation of the birth certifcales and health records of ail children of the veterans, are now under way. Albanese (1988) concluded that while 'A t this time one cannot ascribe the observed group differences to an effect of dioxin", the study also ' d o e s not exonerate dioxin as a causative agent of these differences.* He noted that the study did not find that the health effects increased in severity with a greater exposure to Agent Orange, and that some of the characteristic effects of exposure to 2J,7,8-TCDD were not observed. He also noted that five o f the six group differences were in the direction o f 2JJ.8-T C D D effects, and that the rait-niarinn of exposure was only an estimate. Finally, be noted that the sample sizes used in the study made it possible to detect common diseases and death, but almost impossible to detect rare diseases. An important pan of the controversy aver the possible health effects of Agent Orange is the question of determining exposure; what veterans were exposed and what was the level of exposure? Resent studies indicate that 2*3,7,8-TCDD posists far decades in the fatty tissues of humans (Schecter and Ryan, 1988), and levels of 2J.73-T C D D in the blood c a t be correlated with levels in the Duty tissues (Patterson a aL, 1988). Measurements of 2J.7.8-TCDD in the blood a n therefore be used to determine a person's level of exposure to 2^,7,8-TCDD, even if the exposure occurred maoy yean before. The CDC recently reported the levels of 2,7.S<TCDD in the blood of approximately 700 veterans who were part of the Vietnam Experience Study or the Ranch Hand Study. Ranch 8 Dioxins and Furans Hand veterans had significantly higher levels of 2J.7.8-TCDD in their blood (mean of 49 pons per trillion [pptj) than did other Vietnam-era Air Force veterans (mean of 5 ppt). Seventy-id percem of Ranch Hands had blood levels above 10 ppt, and many had levels of 100-300 ppt (CDC, 1988). However, other Vietnam veterans did not have higher levels of 2J.7.8-TCDD in their blood (mean of 4.8 ppt) than did non-Vietnam veterans (mean of 4.9 ppt) (CDC, 1987). It appears that Ranch Hands, but not other Vietnam veterans, were exposed to significant amounts of Agent Orange. To determine past exposure, the CDC used an estimated half-life1 of about seven yean (CDC, 1988), and thetefoe assumed that two to four half-lives have passed since Vietnam veterans were exposed to 2,3,7,8-TCDD in Agent Orange. Recent work by Kissel and Robarge (1988) suggests that the half-life of 2J,7,8-TCDD is inversely related to the concentration of 2J.73-T C D D in the body, and that the half-life is significantly shelter when concentrations are relatively high (e.g. immediately after exposure). If so, the CDC may be underestimating Vietnam veterans' past exposure to 2J,7,8-TCDD in Agent Orange. The Severn accident In 1976, a cloud of 2,4,5-mchlorophenal and caustic lime, contaminated with 2J,7,8- TCDD, was released over the town of Seveso, Italy and nearby towns, when a chemical plant's reactor vessel went out of control. Approximately 700 seres inhabited by 38,000 people were contaminated with 2J.73-TC D D (Mastroiacovo et aL, 1988). H ie level of 2J.73-TC D D ranged Grom greater than 150 ppt in the soil of the mom contaminated area to less than SO ppt in the soil o f the k a rt area (Wipf n d Schmid, 1983). Vegetation (including ga rden vegetables mid orchard fruits) was oontammaied with from more than 1000 ppb 2 J .7 3 - TCDD to te n than 1 ppb, which was the limit of detection at that time (Wipf and Schmid, 1983). The accident resulted in the deaths of about 3000 domestic animals (almost all rabbits and chickens); same of these "*"**! died .tlmrui immediately after the accident, and their 1 The half-life is the mnnm of time during which the 2J,73*TCDD present in the body is reduced by one-halt Dioxins and Furans 9 784110 GENP w dcaUu were attributed not to 2^,7,8-TCDD, but to the other chemicals in the cloud or to delib a ste slaughter. (All animals given a lethal dose of 2,3,7,8-TCDD take at least two to three weeks to die after their exposure.) One hundred and eighty-three people, mostly children, were confirmed as having contracted chloracne, indicating that (hey were exposed to significant amounts of 2,3,7,8-TCDD. It was presumed that children were exposed to greater amounts because they played in the vegetation, and may also be more sensitive. All cases of chloracne cleared up within 5 years, although the 15 most severe cases resulted in permanent scars. Aside from chloracne, no other health effects were confirmed by an investigation of the accident (Wipf and Schmid, 1983). However, there was concern that the effects of the exposure might include birth defects. The Italian health authorities advised the inhabitants of the area to avoid becoming pregnant, and many women requested abortions after the accident. A panel set up by the Italian government investigated, from the beginning of 1977 to the end of 1984, all suspected birth defects in children bom to women who were inhabitants of the area at the time of contamination (Mastrotacovo et al., 1988). This study did not find an increased rate of birth defects in children bom to the exposed women. The authors noted that rare defects could not have been detected because of the small number of births, and that it was possible that spontaneous abortions decreased the number of children bom with defects. They cited one study that found an increase in spontaneous abortions after the accident (Sami et aL, 1983), and one study that did not (Bianco et aL, 1986). Conclusion The conaminarion o f Times Beach and Seveso, and the health effects attributed to use of the herbicide 2,4,5-T, especially from Agent Orange, have made, dioxin a household word with in infamous connotation. Much attention has been paid to the potential danger of dioxins and furans and their wide presence in very small quantities. Yet investigators of the health of the people exposed to relatively large quantities have found little evidence to justify the enomtoos reaction to these incidents. Nevertheless, dioxins and furans are present in our environment, in organisms, and in us, and can be extremely toxic, at least to other organisms. A great deal of money and other resources have been spent on tbe research o f dioxins and 10 Dioxins and Furans GENP 011300 784111 furans, and will continue (a be spent. The question then is. what do we know about dioxins and furans? Detecting T race Amounts ot Dioxins and Fnrans One of the difficulties of discussing dioxins and furans is the almost unimaginably small quantities we can detect with modem techniques: as little as one trillionth, or in some cases even a few quadrillionths of a gram. The following examples illustrate how little these quantities are. If we assume a drop of water to be equal to 0X5 millileten (20 drops to a milliliter), then one ppb is equivalent to one drop o f water in 13X00 gallons, or a pool 20-feet square and approximately 4.5-feel deep. One ppt is equivalent to one drop of water in 13X00.000 (thirteen million two hundred thousand) gallons, or a pool of water as long and wide as a football field, and approximately 28-feet deep. One part per quadrillion (ppq) is equivalent to one drop of water in 13X00X00X00 (thirteen billion, two hundred million) gallons, or a 6square-mile lake, approximately 10.5-feet deep. Dioxins and furans were among the very first compounds chemists could measure in one billionth o f a gram (in 1970) and one trillionth of a gram (in 1976), and they are now approaching the detection of one quatfcillionth o f a gram. As the new techniques for detection were developed, dioxins and h ire were used to le a the techniques because they appeared to be toxic a t extremely tow levels. Before the detection techniques were developed, researchers used animals such as young chickens to test the toxicity of dioxins and furans (pen. conurc, David F a t s , USFDA). Dioxins tend themselves to the techniqua used to detect amounts this small. They react with very few other chemicals, many last a long time, and they dissolve readily in organic solvents. Also, dioxins and fin n s are relatively simple molecules. The molecules o f many other toxins are very complex Open. comnL, David Firestone, USFDA). In order to d etea amounts this small, it is first necessary to isolate the snhstanoe you want to measure from all the o th a substances that * e present in a typical sample. This process Dioxins and Fursci 11 784112 f" is called analytic! clean-up; the chemist puts the sample through a complicated and de licate preparation process, involving extractions, washings, and scpraatkns. Once the sample has been cleaned up, the separated materials need to be detected and their amounts measured. This can be done using several techniques, including an Electron Capture Detector, Flame Ionization Detector, or Mass Spectrometer, along with a Gas Chromatograph (pen. comm., Annette Guiseppi-Elie, Drexel University). in addition to the difficulty of detecting these very small amounts, the toxicity of these compounds necessitates great care in their handling. Therefore, detecting the presence of dioiins and furans at levels approaching 1 ppt is a very time-consuming, expensive, and difficult undertaking. An experienced analytical chemist needs several months in order to prepare the laboratory for detecting a particular chemical at these extremely low concentrations. Once everything is in place, the analysis of just one sample costs about $2000 in materials, requires the attention of an experienced analytical chemist, and typically consumes a half-day of work (pen. comm., Robert Baughman, Harvard University). A much cheaper and quicker method has been developed for determining the level of toxicity of a mixture of dioxins and furans in a sample. It has been established that dioxins and furans induce a specific enzyme and bind to a specific protein in animals and humans (Poland el al, 1979). Rather than using analytical chemistry to determine the amounts of various toxic congeners' of dioxins and furans in a sample, ceil cultures are exposed to the sample and the above biochemical responses are measured. The accuracy o f these methods have been verified by comparing their results with the results of chronic toxicity tests using laboratory animals (Safe, 1987; Sawyer al. 1983).1 1 A congener is a single, particular dioxin or furan, e.g. 2J,7,8-TCDD. 12 Dioxins and Furans GENP 011302 L 784113 r Pterins sod Furans: W hat They Are, W here They Are Found, ta d H n Th*y rut.TM W hat Is dkndm? The dibenzo-para-dioxin structure (see Figure 1). which provides the Cnmework far tens of thousands of compounds, is made up o f two benzeue rings joined by two oxygen atoms (hence dibenzo dioxin). The 'pare* means the benzene rings are attached to opposite sides of the dioxin molecule, and it is usually designated as simply "p*. Benzene is a ring of six carbon atoms (one bl each comer) and six hydrogen atoms (one bonded to the carbon atom at each comer). Figure 1. Diberatvo-dioxin Figure 1. Dibenzofuran In add" to dibenzxvp-dioxifl, there is the related structure called a dibenzofunn (see F igure.2). A dihenzoftnn. like a dtbenzo>{HliaxiD, is composed of two benzene rings. H o w m , the benreno rings are joined by one oxygen a re a , instead of two. The single oxygen dux jo the two Iw hw m rings farms the A m pun of the molecule. A farm is a ring like h a * , but is made up of four carbon atoms plus one other atom (in this care, that o th a atom is the oxygen atom). Both dtorins in d A n n s have eight c o m a s (four on each bemene ring), free to react with other atoms. These other atoms can be, for example, hydrogen, chlorine, bromine, fluorine, iodine, or nitrogen. The comers can also react with groups made up of these aunuL The number o f possible combinations is enormous. Dioxins and Funns 13 784114 GENP 01303 There are 75 polychlorinated dibenzo-p-dioxins (PCDDa) and 135 polychlorinated dibeazofursns (PCDFs). FCDFs are similar to PCDDs in th d r distribution, toxic propeniex, and behavior, although not as much b known about them. While those dibenzo-p-dioxins and dibenzafurans with atoms other than chlorine on their corners (bromine, nitrogen, and so on) may have similar toxic potential (tests indicate this to be m is for some types), they not as common or numerous, and are not considered to be as important as chlorinated dibeozo-pdioxins and dibenzofuians. However, there has been some concern about the polybeaminatrd dibenzo-p-dioxins and dibenzofuians, because polybrominated compounds are used as gasoline additives, and have been widely used as flame retardants in plastics; -carpets, textiles, and so on (Buser, 1987). It has been shown that when these compounds are burned, polybrominated dioxins and lim ns can form (Buser, 1986; Haglund et aL, 1988). Almost nothing is known, however, about the extent of their presence in the environment. For the rest of this report, ooiy the chlorinated dibenzo-p-dioxins and dibenzofuians will be discussed ( v H a a s 1*22J i - m n As illustrated in Figure 3, the numbers 2 J .7 3 refer to the position on the benzene rings where the four (thus t e n ) chlorine atoms me attached. 23,7,3-TCDD is the most toxic of the dioxins o r f in e s , and can be formed in significant mounts during the production of trichloropfaenol-basod chemicals. Also, 2J,7,8*TCDD is accumulated preferentially by animah and humans, cam pued to the other types o f dioxins and f tn n s . A great deal mom research has been done on 2J.73-T C D D than on the other dioxins and fa in s , and the following riic m ic n is based largely on studies of 2J.73-TCDD. 14 Dioxins rod Furans GENP 011304 _L 784115 Are tU dioxins and Aurans daageroos? Only 20 or so o f the 210 chlorinated dioxins w d funns are considered to be extremely toxic. These have 4 to 7 chlorines, with a chlorine in each o f the 23,7, and 8 positions (23.7,8-subsunitcd dioxins and furans). Table 1 lists these very toxic types o f dioxins and funns, and all' other dioxins and furans, along with their toxicity relative to the most toxic dioxin or filian, 23.7,8-TCDD. Table 1. Relative toxicity o f dioxins and furans Compared to the most toxic, 23,7.8-TCDD. From theUSEPA National Dtoxin Study (USEPA, 1987a: p 1.7} Type of dioxin or f u n s 2 3 , 7 ,8 - tetra-CDD 133 ,7 ,8-penta-CDD 23.7.8-tttra-CDF 133.73-penta-CDF 23,4,7,8-penta-CDF 1,23,4,73-hexa-CDD 13 3 ,6 ,7 ,8-hcxa-CDD 133.7,8,9-hexa-CDD Other teoa-CDDs 12 3 .4 ,7 ,8 -h e x frC D F l33,6.7.S*hex*CDF 133.735-hexa-CDF 23*4A73-hex&CDF Other pema-CDDs 1 3 3 ,4 ,6 ,7 3-hepta-CDD Other tetn-CDFs 133.4A 73-hqjca^D F 1 3 3 ,4 ,7 3 3 -h ep t^ C D F Other penta-CDFs Other bexa-CDDs Other hexa-CDFs Other hepta-CDDs/CDFs Other CDDs/CDPs (4 chlorines) (5 chlorines) (4 chlorines) (5mchlormines) (6mchlormines) mm (4 chlorines) (6n chlormines) mm m (5 chlorines) (7 chlorines) (4 chlorines) C7 chlormines) (5 chiarina) (6 chiarina) (7 chiarina) (1*3, 8 chlorine) Relative toxicity l (Moat Toxic) 1/2 as toxic l/10tb * 1/25ih 1/lQOth 1/200th moooth 9m * m 1/2500th l/10.000th 1/100,000th Not toxic Dioxins and Funns IS 784116 GENP 011305 Dioxins and ftinns usually occur in a mixture of several or many types, cad the mixture varies with the source of the dioxins and furans. In order to anive at a number that represent the total toxicity of the mixture, relative to 2*3,7,8-TCDD, the appropriate factor is ^ U e d to the amount of each type of dioxin or fiiran in the mixture and the result then added, for all the types present. The total amount is (hen expressed in *2J,73-TCDD equivalents'. This allows the direct comparison of the toxicity of different mixtures of dioxins and A n n s, from different sources. For example, let us say we have analyzed samples of soil from two contaminated waste sites, and found the following: Siw.Qre Site Two 3 ppb 2J.7.8-TCDD 6 ppb 2J.7.8-TCDF 200 ppb other tetra-CDDs 1 ppb Z3.7.8-TCDD 4 ppb 1,23,7,8-penta-CDD 100 ppb lA3,7,8-penta-CDF How can we compare the toxicity of these two simples? The toxicity equivalent method is one way. The toxicity of these two samples would be computed as follows: Site One Factor Result 3 ppb 2,3,7,8-TCDD 6 ppb 23,7.8-TCDF 200 ppb other tetra-CDDs X I 3 ppb X 0.1 0 ppb X0.01 2 ppb 5.6 ppb Site Two Factor Result 1 ppb 23,73-TCDD 4 ppb 1,2J3,7,8-pemi-CDD 100 ppb lA3.73-peuta-CD F XI X 05 X 0.1 1 ppb 2 ppb lO roh 13 ppb r a t , the toxicity of the sod in Site One is equal to 3.6 ppb 2J.7.8-TCDD equivalents and in Site Two is equal to 13 ppb 2J,7-TC D D equivalents. The USEPA used data from several kinds o f studies to deannine these (acton (Barnes et aL, 1986). O f first importance were data concerning cancer and reproductive effects in animal, based on long-term studies, but this information is available for only a few types of 16 Dioxins and Furans GENP 011306 r dioxins and furam. Therefore, data from studies of the biochemical effects at ih* compounds oa ceil cultures w en also used. These studies, called enzyme induction and receptor binding studies, measwe certain biochemical responses that are related to the toxic effects of dioxins and fonoa. The USEPA chose not to use data from short-term animal toxicity tests, stating that such information is not useful for predicting the long-term health effects of dioxins and furans. These focton are the result of interpretation, and agencies from other countries have released somewhat different guidelines (Barnes et aL, 1986). This is an area of ongoing investigation, and the USEPA will probably revise this table. For example, a recent study iniiirmir* that octa-CDD/CDF are toxic, not non-toxic as this table shows (Couture et aL, 1988). Other types of dioxins and furans have also been investigated, and found to have toxicities higher or lower than the present focton indicate (Pleuss et aL 1988a; 1988b). The use of such focton to determine the toxicity of a mixture is somewhat controversial. The CDC, for t example, feels that this method b not scientifically valid, given our limited knowledge of the health effects of the many << and furans; especially in a mixture. However, Pleuss et al (1988b) investigated the toxicity of a mixture of dioxins and furans in rats, and found the USEPA's focton to be adequate for estimating the total toxicity. W hat factors affect the toxicity of dioxins and furans? The toxic effect of a dioxin or funu on an animal depends on more than the basic toxicity as expressed in the table above. It also depends upon the kind of animal being exposed, the condition of the retinal, and the medium that the compound is mixed with. Some lim it m ttw m w ii o f mow sensitive than oth cn to the toxic effects of the same dioxin or furen. For example, guinea pigs (the most sensitive animal tested so f a ) are ^xuxim ateiy 3000*5000 times mom sensitive to the lethal effects of 2J.7.8-TCDD than are him sten (the iM iiii i i raim ii 30 for tested). Also, they are very lipid-soluble (soluble in fats, mis, and solvents such as acetone), dioxins and fiureis have had a greater toxic effect on an Dioxins and Furans 17 784118 GENP 011307 r animal when ihey were administered in, for example, com oil, than when they were administered in dry food. Different kinds o f dioxins and floats have different behaviors within animals: some are excreted rather quickly, while others are stored in the tissues for long periods o f time: This may affect their toxicity to the animaL Laboratory studies showed that rats and carp fed a mixture of dioxins and furans accumulated higher amounts of 2.3,7.8-substituted types (these are the most toxic types of dioxins and furans) than other types (van den Berg et &L, 1983; Kuehl et aJ,, 1987a; 1987b). They also found that carp retained 2J.7.8-TCDD and its counterpart furan (2,3,7,8-TCDF) for longer periods than dioxins or furans with fewer or more chlorines. Researchers who have made congener-specific analyses of tissues Grom organisms in the wild have also found that the 2 3,7,8-substituted congnos of dioxins and furans are the only types found (Rappe et aL, 1987a; Hetda et aL, 1986). Crustaceans appear to be an exception to this general rule; Rappe et aL (1987a) detected dioxin and Kuans congeners in crabs and lobsters that were not 23.7.8-subsdtuted. Soils from different 2^,7^-TCDD coatam inated sites show greatly varying toxicides. Researchers Cram the Notional Institute of. Environmental Health Sciences and Rutgers Medical School fed guinea pigs 23.7,8-TCDD-contaminated soils Grom two different sites (McConnell et aL, 1984; Umbreit et aL, 1986). The animals showed greatly different reactions; although the amount o f 23,7,8-TCDD given to them was similar. They concluded that the u su re o f the soil affected (he toxicity of the 23,73-TCDD, presumably by determining bow much of the 2 J .7 3 TCDD was a b o rte d by the animal (this is known as bioavailability). The reaearcha s suggested that tbs difference betweoi the toxidty of the soils was caused by the difference in tbe amount of carbon in the sotL It is known that dioxins bind strongly to carbon and are thus less likely to be absorbed d o in g digestan. 18 Dioxins and Furans GENP 011308 L_ 784119 f\ r W here are Htn*lna aDd furxm found? 1 In so il water, and air There are areas contaminated with relatively high levels (1 ppb or more) of dioxins and (mans; most are sites where 2,4.5-trichlorophenol and 2,4.5-trichIorophenal-based biocides were manufactured or processed, and/or their associated waste sites. Wood-preserving fvilities that have used pentachlarophenol, and sawmills that have used chlorophenolic solutions to prevent staining, can be contaminated with dioxins and furans and in some cases have contaminated the .local environment. In addition, there are many contaminated areas in Missouri where 2J.7.&TCDD-laden oil was sprayed an roads and horse arenas, and sevaal military bases where Agent Orange was stored or heavily sprayed. Apart from these recognized sites of contamination, 2J.7.8-TCDD has not been found \ commonly in soil. As part of their National Dioxin Study, the USEPA randomly selected 221 urban sites and 142 rural sites where no previously known sources of dioxins or furans had been reconfcd. Seventeen of two hundred and twenty-one urban sites and one of one hundred and thirty-eight rural sites had detectable amounts of 2J.7.8-TCDD in the soil (USEPA, 1987a: pp 3.26*32). The limit of detection was about 1 ppt, and levels ranged from 1-11 ppL Low levels of 2^,73-TCDD have been found in areas that were sprayed with the herbicide 2,4.5-T. During their nationwide survey, the USEPA found 2J,7,8-TCDD at 15 of 26 sites where 2,4,3,-T was sprayed commercially; these sites included sugarcane and rice fields, rangeland, and forests (USEPA, 1987c pp 3.12-21). The highest levels (1-6 ppb in soils or sediment) were generally found where spraying equipment was loaded, or where the herbicide accumulated. Where the herbicide was simply sprayed, levels w o e very low or below the limit of detection (1-3 ppt). Dioxins and furans have also been found in sediments in the Great Lakes and some arsociatrd riven, and in some large river systems ocar urban c a te rs . They are ubiquitous in low ppb levels in the sediments of Lakes Erie, Ontario, Huron, and Michigan (Czuczwa and Hites, 1986). They have also been found in vatimemi* of the Niagara River (Hallet and : Dioxins and Furans 19 784120 Brooksbmk, 1986). In many areas, samples of fish or other organisms have been used to detect the presence of dioxins and fumns in the water ecosystem, rather than direct sampling o f the sediments. (See the following section oa dioxins and funns in animals and plants.) Dioxins and furans have not been detected in treated drinking w a ts in the U.S. However, furans have been found in o p water in Japan (Shiiaishi et aL, 1985). In the U.S., dioxins have been found in water of the Niagara River adjacent to the severely contaminated Love Canal area, and in groundwater near the canal (Hallet and Brooksbank. 1986). Groundwater and surface waters in the vicinity of wood-preserving facilities and sawmills, that usd chlorophenol-based solutions to treat wood, have been contaminated with dioxins and furans (USEPA, 1988a: pp 53323-4). Both groundwater and surface waters can be sources for municipal drinking water, although the water is treated first. Those people who live near contaminated sites and use untreated well water have been nodTied. Dioxins and furans have been detected in air in urban areas of the U.S. (Czuczwa and Hites, 1986), West Germany (Rappe and Kjellcr, 1987; Rappe e t al, 1988), and Japan (Nakano et aL, 1987), in air near contaminated sites in the U.S. (F u rie s et aL, 1987), and in the air emissions from incinerators. It is suspected that incinerators and automobiles burning leaded gasoline are important sources of dioxins and furans in urban areas. In conclusion, dioxins and furans are present in very low levels in many places, and at high levels at a small number of contaminated sites, Their presence is usually associated with the (induction, storage, use, or disposal of chloropbenot-bnsed compound. In addition, they have been detected in low levels in mfam areas, where they are believed to be the in d u c t of num enm combustion processes. As otv ability to detect dioxins and furans improves, it is likely that they will be found in more places. Are dlnatw and knar found In and plants? Dioxins and furans are much n x n soluble in tats than in water and therefore tend to aremimitate m the Duty tissues sad orgros of animals exposed to them. Levels of 1-200 ppt 20 Dioxins and Furans 784121 r [ 2,3.73-TCDD have been detected in fish and other organisms in some puts of the U.S., [ and fish from the Great Lakes and associated riven generally enntam higter wwitw than | elsewhere in the country (USEPA, 1987a: p 3.29*32). The USEPA detected 2^,7,8-TCDD in 17 o f 90 samples of fish bom randomly selected national monitoring sites, at levels of 1-19 ppt. Samples were also taken from 305 areas of general interest, chosen because they were near . population centers, were used for commercial or recreational fishing, or because other water quality information was already available. Of these samples, 95 of 305 bom regionally selected sites had 1*85 ppt 23,7JB*TCDD. Of fish samples from the Great Lakes, 23 of 29 had M l ppt j - 23.7.8-TCDD. i The USEPA also found as much as 85 ppt 2J.7.8-TCDD in fish, from riven that were * receiving effluents from some pulp and paper mills (USEPA, 1987a: p 3 J1 ). Since then, dioxins and finans have been found in the sludge, wastewater, and products o f pulp and paper \ mills that use a chlorine bleaching process (Amendola et aL, 1987). A study of dioxins and finans in fish and Herring Gulls from the Great Lakes (Stalling et aL, 1983) found a wide range of total levels of dioxins (undetectable to 223 ppt) and fuians (15-290 ppt) in fish from various locations in the Great Lakes watershed. 2J.7.8-substituted congeners were responsible for the bulk of the dioxins and frosns present, and 2^3,73-TCDD was the predominant congener. The two Herring Gulls analyzed had 27 and 26 ppt total finans, and 196 and 110 ppt total dioxins, including 165 and 75 ppt 2J.7.S-TCDD. 2J,7,8-TCDD has also been detected in Herring Gull eggs from the Great Lakes, St levels ranging from 9-90 ppt (N om ura et aL, 1982). Because 2J.7.8-TCDD is a contaminant of the herbicide 2,4,5-T, there was concern shorn 2J.73-TC D D being spread through the use o f this herbicide (2^4,5-T is do longer produced or used in the UJ3.) Therefore, a joint study by the USEPA and the University of Nebraska was imdeiuken to determine if ammala showed detectable a m e n ta of 2^,73-TCDD after normal application of 2,4,5-T to Oregon forests (Gross, 1980, d ied in W eaasingbe and Gross, 1985). Animal tissue and whole *** (mice, shrews, birds, and newts) were sampled Dioxins and Furans 21 ! 784122 GEbrp and, although 3 ppt were found in several samples by one laboratory, this could not be confirmed by the second laboratory. However, some studies have detected low levels of 2J.73-TC D D in organisms after normal applications of 2,4,5-T. In a 1984 study by the USEPA, deer were placed In a plot that was then sprayed with 2,4,5-T (Harless et aL, 1983). They found 1-27 ppt 2J.7.8-TCDD in the fat, muscle, and liver tissue of the deer, and the frequency of detection increased during the four-week sampling period. No 2J.7.8-TCDD was found in the bone marrow of the deer, or in a deer placed in a separate area. Also, 2J.7.8-TCDD in levels as high as 1 ppb have been detected in whole-animal samples of fish and shellfish from areas In South Vietnam (Baughman and Meselson, 1973) that w oe heavily sprayed with Agent Orange during the Vietnam War. At exceptionally contaminated sites, such as an improperly managed waste site (Heida et al., 1986), a military base where Agent Orange was repeatedly handled and sprayed (Young and Cockerham, 1985). or where 23,7,8-TCDD was released during an industrial accident (Fanelli et aL. 1980), 2J,7J8-TCDD has been found in many different organisms, such as rodents, insects, lizards, earthworms, and birds. Researchers in Italy found that plana grown in soil contaminated with 2J.7.8-TCDD accumulated the compound in their roots, and to a much lesser extent, in the above-ground portion of the plana (Facehetti et aL, 1986). However, in Scveso, one year after the accident, no traces of 2,3,7,8-TCDD were found in the flesh of fruits or in com kernels and cobs, from trees and plants grown in soil contaminated with approximately 10 ppb to o l dim ini. Dioxins were found in the peels of fruits; the researchers concluded that they were corasnunsied externally by dust, and not by dioxins in the soil (Wipf et aL, 1982). Dioxins and ftaana have been fotmd in orgm isns from cwuamimtad s e a s , and from fish in many areas o f the U.S. Most nounsIs do not appear to accumulate these compom di at levels greater than in their starounding environment, although *bioconccnureion* can occur. P Itita typically contain th e n compounds a t levels much below the surrounding environmert, and only when they are grown in highly contaminated soiL The fruits do not appear to contain dioxins; however, root oo p s may be more susceptible to cornsninatioo. 22 Dioxins and F in n s H]\rp n 1 784123 How k | do dlnxlni and farm s last la the eovfrmiMutT 1 Many (acton influence the persistence of dioxins and furena in the environment: what land ot dioxin or faren it is, whether it is prerent as a solid or gas, what it is mixed with, and what kind of environment it is exposed to, among other things. Photodegradadon (breakdown by light) is believed to be the most important environmental process for the breakdown of dioxins and furans, Crosby et aL (1971) and Crosby and Wong (1977) have demonstrated that ultraviolet (UV) light is responsible few the photodegradaiion of 2,3,7,8-'ICDD. They also found 'that the rate of breakdown varied with the material the dioxins were mixed with, and die surface to which the mixture was applied. They reported that 2,3,7,8-TCDD applied to wet and dry soil, or dissolved in water, showed no signs of breakdown after as much as four days of exposure to ultraviolet light. However, when it was mixed with the herbicide Agent Orange, the commercial herbicide Esteron, or a solvent such as methanol (wood alcohol), and applied to roil, glass, or plant leaves, the 2J.7.8-TCDD was partially or totally degraded in less than eight hours. The authors noted that these conditions would often have been met during the application of 2J,7,8TCDD-containing herbicides such as 2,4,5-T, and may explain why the normal application of 2J,7.8-TCDD-cootaifling herbicides typically does not leave persistent, detectable amounts of 2J.73-TC D D on soil or foliage. In addition to photodegmditkm, there are other processes that can break down dioxins and farm s. For example, some breakdown of these compounds in soil appears to be the result of digestion by microorganisms (M arsum tn and Beaezet, 1973). However, this breakdown occurs much more slowly thro the pbocodegradadon process mentioned above. Eflions to isolate a microorganany dire can digest 2J.73-TC D D have met with little success: only a small percentage of the substance is broken dawn, evra after months of incubation. It also appears d o t dkmins and farm s bound to policies m d dissolved in water are very slowly released to the atmosphere as a gas (Palau&y et aL, 1986; Nash and Beall. 1980), The significance of (his Dioxins and Furans 23 784124 \ f r N\ v r- process is not yet agreed upon. Once released to the air by this process, it is expected that dioxins and furans are broken down by sunlight In sum, 2,3.7,8-TCDD can be broken down in a matter o f days in the environment if it is mixed with the proper organic solvent (not water, but acetone or even com or olive oil) and exposed to ultraviolet light (Wipf et a t, 1978). On the surface of the soil, without a solvent 2J.7.8-TCDD has an environmental half-life of one year or less, due u> slow photodegradaiion and volatilization (Crosby and Wong, 1977). Underground and in sediments, this very stable compound has broken down much more slowly, with a half-life of about 10 years (DiDomenico et a t, 1980a; 1980b). Very little ts known about the half-lives of other dioxins and furans. In general, dioxins and furans with fewer than four chlorines are broken down mere quickly than 2J.7.8-TCDD (Crosby et aL 1971). The dioxin or furan with eight chlorines is more resistant to breakdown. ' To what degree do dioxins and furans spread when they are introduced Into the environment? The most important m anna o f transport o f dioxins and furans appears to be the physical movement of particles to which they are bound. Dioxins and furans have a great affinity far organic carbon; thus they bind very strongly to the gam e carbon present in panicles of soil, in sediments, and to the particles found in the air. These panicles can be lifted and carried by wind or by water. In the air, dioxins and furans can be transported considerable distances. Far instance, they have been found in the sedim ea o f a lain on Isle Royale, an island in Lake Superior (Cznczwa et aL. 1984). It is believed that dioxins and furans could reach this lake only by n s port m air. Also, dioxins and fiirans have been found in Arctic seals taken in the Arctic Circle, many thousands of ldlometen from any known sources of the compounds (Oehme et al. 1988). 24 Dioxins and Furans GENP 011314 784125 There is also evidence of the transport of dioxins m d A n n s in w e n . 2J.73-TCDD has been detected in fish downstream from known sources, and in gro u n d w m Mid rivw water near landfills in which (Mgo amounts of 2J,73-TCDD-comaminMcd w tse s were rfi<po ^r1 (H alls and Brookabank, 1986). Dioxins and fuxazu have also migrated from wood-preserving facilities and sawmills through surface wzten and groundwater (U5EPA, 1988a: pp 53323-*). Measurements of the movement of 23,7,3-TCDD through soil indicate that the process is generally a very slow one. from one to ten centimeters per year. Palausfcy et aL (1986) reported that the initial depth to which the 2J,7,S'TCDD contaminates the soil greatly . influences the resulting movement by vapor phase. This depth of contamination is related to the type of solvent in which the 2J.7.8-TCDD is carried: the volatility (at what temperature it evaporates) of the solvent, its viscosity (how easily it flows), and how it interacts with soil organic matter. Because they are so insoluble in water, only very small amounts of dioxins and furans I are washed from soil parades or released from sediments to the water above it (Isensce and j Jones, 1973). Their extremely low solubility in water makes it very hard to measure just how much dissolves; consequently, there is a wide range of reported values. The recently accepted value far 2J,73<TCDD is about 20 ppt in pure water (Marple et aL, 1986). Dioxins and furans can be transported as a gas, but Palausky e t al (1986) detected this only a t temperatures above 30oC (87*F); it was therefore concluded that this vapor phase f i transport of Z3.7*TCDD takes place only in upper layers of soil, during hot weather. la c o o d u sa i, dioxins and A m a attached to particles can migrate considerable in the air, and to a fa*** extent in w aar. But these compounds are so insoluble in water, and they bind so strongly to particles in the soil and water, they ; appear to migrate very little ones they reach sediments and soil. Areas of severe contiintnirion Dioxins and Furans 23 784126 GENP Are dionio and furana found In humans? D am ns and furans have been found in samples of human blood and Cat tissues from numerous countries, including the U.S., Japan, Vietnam, and several countries in Europe. Based on these findings, it is generally agreed that the population at Urge in many parts of the world has been exposed to low levels of dioxins and furans. In North America, far example, 2,3.7,8-TCDD is typically present in human fat, some other tissues, and breast milk, at the level of approximately 7 ppt (Ryan. 1986). In addition to ' 2J,7,8-TCDD. penta-CDD is typically found at 10 ppt. pexua-CDF at IS ppt,-and the hepta* and octa-CDDs/CDFs at levels approaching 1 ppb. In humans as in other animats, the most toxic dioxins and furans, with chlorines in the 2 J .7 and 8 positions, have been found in greats' amounts than the other dioxins and furans with the same number of chlorines. How do dioxins and furans reach humans? Although we can be exposed to enough of these compounds to detect them in our bodies, the mechanisms of this process are still under investigation. Dioxins and furans are believed to reach humans in the following ways; when we inhale contaminated dust or vapors; when we contact or ingest contaminated water, sediment, or soil; when we eat contaminated vegetable, meat, or dairy products (Mukerjce et aL, 1986). At the present time, the relative importance of these routes of exposure is not understood. However, drinking and contacting contaminated water is generally considered a minor route **" '*" dioxins and furans have not been detected in treated drinking water in the U.S., and o t r contact with um eaied in te r and sediments is normally minimal (e^j. swimming, other wateraparts). Ingestion o f and contact with soil is also considered a minor route, except far small children. The major routea of exposure, then, are generally considered to be breathing contaminated particles and eating contaminated food. Particles in air, especially near urban areas or large sources of combustion products, can contain low levels of dioxins and furans, and 26 Dioxins and Furans r [ [ | ; | ! i i tbe cciwniilatrd lifetime exposure from inhaling there panicles is considoed potentially important sm ics (M utlund et aL, 1986). Studies by Rappe et aL (1986s) aral Travis end Hjnexner'Frey (1987) cotnpued breathing to b o d conznsipooo, re d * * breathing is Tar less important, even near a large m i c e of combustion. However, very lisle b known the actual amounts contributed by breathing air panicles. Dioxins and b ra n s have been found in a variety o f foods in Japre and Canada, in dairy products in Sweden, and in selected foods in the U.S. They have been found in Osh from the Great Lakes and associated rivers, fish from many major rivers near urban centers, and fish from rivers below some pulp and jmpa mills. As s result, advisories far limiting consumption of fish in some of these areas have been issued (see Appendix). In 1984. the USEPA found ppt levels of 2J.7.8-TCDD in 3 of 85 samples of beef fat from cattle that had grazed oo land sprayed with 2,4,5-T (USEPA, 1985: p 4JO). Dioxins and brans, but not 2J.7J-T C D D , have also been found in chicken liven and chickm eggs, and in gelatin from supemuriceu (USEPA, 1985: p 4-31) In both c u es, the sources of the contamination were hides (from cattle, hogs, sheep, and other animals), treated with peatachlorophenoi, a product which b usually contaminated with dioxins and brans. Fat from the (rested hides was used in the manufacture of the chicken feed and the gelatin (Firestone, 1973). Investigators in Sweden found dioxins and brans (but not 2J.7J-T C D D or 2J.7.S -' TCDF) in levels below 1 ppb in cow (at, milk, cream, and liver (Nygrcn et aL, 1986), and investigsion in Japan (Ooo et aL, 1987) found similar amounts of dioxins and b rans in a variety of meat and dairy products. A researcher in Canaria also found ppt leveb of dioxins and b rans in fruits and vegetables (Davies, 1988). 2J.73-TC D D has not commonly been fraud in food in tbe U J 4 its presence in selected foods has been related to specific sources, as above; However, it b probable that trace amounts o f other dioxins and b rans a s present in common foods, as has been found in other commies. Similar tests have not yet beat reported for the U J . Dioxins and Furans 27 784128 GENP 011317 Breast miOc, like (any human tissues, has low levels of 2^.7,8-substituted dioxins and furans, and they can be transferred to a baby through the mother's milk (reviewed in Lindstrom, 1988: p 34). Because a baby is tiny, undergoing rapid development, and may have breast milk as its sole source of nourishment, a breast-fed baby may be subject to a greater than normal risk (Tarkowski and Yrjanheikki, 1986; Schecter and Gasiewkx, 1987). Breast-fed babies may receive levels of dioxins and furans that exceed safe guidelines (Nygren et aL, 1986). However, a study by the World Health Organization (WHO) Regional office in Europe determined that breast-fed babies receive dioxins and furans in amounts far below what are capable of causing adverse health effects (WHO, 1988, cited in Lindstrom, 1988). Young children ingest more soil than o.der children and adults, and are probably subject to greater contact exposure as well (Houk, 1986). The Center for Disease Control took this greater exposure into account when they determined that 1 ppb or more of 2,3,7,8-TCDD in the residential soil of Times Beach suggested the need for corrective action. Workers in several occupations are or have been subject to a greater potential exposure (Choudhaiy, 1983). These include: mills where wood is treated with pentachlcrophenols or where treated wood is sawed; leather and tanning industries, which have used chlorophenols as preservatives; facilities that manufacture, ship, or formulate chlorophenols; pulp and paper mills that use chlorophenoi-based fungicides; health-related facilities, where hexachlorophene is used as a bactericide; occupations where workers apply chlorophenoi-based biocides. Io sum, it is generally believed that eating contaminated food and breathing contaminated purides are the most impanxnt sources a f exposure. We need to know more abore the routes of human exposure to dioxins and furans. Also, breast-fed babies, small children, and some waricera in the industries d m use contaminated products, are subject to greater that normal potential exposure. These cases m a il special attention. 1 28 Dioxins and Furans L 784129 < Sources of Dloalni ml Parana Hawlie dterin ud ftuus anted? Dioxins m d (mans have no useful purpose and have never manulKtured deliberately, except in small amounts Cor research purposes, They am known to be created in two major ways: (1) they are created in trace quantities by unwanted side reactions, which take place during the chemical processes used to manufacture useful products such as biocides, paper, disinfectants, and preservatives; (2) they are created when s substance cotuaming ehioHn is burned, or when a substance is burned in the prcsaxz of chlorine (chlorine is a enq imm* element of many substances). In addition to the dirwin and forms from these two major sources, there is evidence due diming and fuxans can be created when certain mixtures of substances are exposed to sunlight The importance of this process is not yet known. Determining the relative or absolute importance of the various sources of dioxins and fu n o i is complicated. Simply, the question is: What kind of hazard does the source represent to humans and other organisms? In order to estimate this, at least the following information thfiM be 1) What types of dioxins and fuians are produced by the source? 2) What amounts of the different types are produced? 3) How will humans and other organisms be exposed? Far how long? To w tm levels? 4) How many hanana or other oqsanhms will bo exposed? Because this bifnrnurinn is unavailable far most o f the sources of the dioxins and farm s, it is impotable to msks a n thin s rudim ciaey qualitative estimate rtf the importance of the various sources, relative to each other. For some sotices, even this' is mnrenanted because o f lack of infonnaion. Therefore, while a dkiww of the sources is worthwhile, it should be noted that the various known sources may prove id be much more or less important Dioxins and Fuians 29 Jl _. _ ____________ __________________ 784130 GENP 011319 than our present understanding makes them appear. Also, new sources of dioxins and furans will continue to be identified. 1 What products can be contaminated with dioxins and furans? Dioxins are fonned as unwanted byproducts during the manufacture of some chlorophenols (especially those with three or more chlorines), and during the manufacture of chemical products that are derived from these chlorophenols: phenoxy herbicides such as .2,4,3-T. and biocides and preservatives such as pentachlorophenol. Manufacturers have greatly reduced the levels o f dioxins and furans in their chemical products since they were identified as contaminants in the early 1970s. In some cases this resulted in much higher concentrations of dioxins and furans in the waste from these processes. The safe disposal of such waste is t difficult, and inadequate disposal methods have resulted in environmental contamination. Dioxins and/or furans have been found as contaminants in the following products: \ Table 2. Products that can be contaminated with dioxins and/or furans Product Contaminants PCBs hexachioropbene 2 .4 J -T 2,4-D chlorophenols with 3 or more chlorines polychlorinated benzenes diphenyl ether herbicides hexachlorocydohexane paper products Furans Dioxins* Dioxins* Dioxins* Dioxins and Furans * I could find no analyses of these products far funm . N a te The above list is based oq information available is journals or reports. Under the auspices-of the Toxic Substances Consol Act (TSCA), the USEPA published a largo" list of chemical products that may be contaminated with chlorinated or bran mated itin and furans (USEPA, 1987b). For mare information, see under 'W hat has been done about dioxins and furans?', page 69. UlUfJJI! 'J-JJJMLIBUW 30 Dioxins and Furans l GENF 011320 784131 1 ECHa PCBs end PCB-comaining mixtures can be a source of furans and to a lesser n t a source of dioxins. They were widely used prior to 1977, primarily as dielectric fluids in electrical equipment such as uansfonnen and capacitors. PCBs have also been used as plasticizers, hydraulic lubricants, and in a variety of other applications. PCBs are no longer produced in the U.S. or installed in new equipment, and their use and disposal is restricted (see under 'What Has Been Done About Dioxins and Furans?', page 69). Levels of furan contamination in PCBs vary with the manufacturer and the type of PCB (Bowes et al., 1975; Rappe and Buser. 1980). A recent and particularly sensitive study of both U.S. and Japanese PCBs found up to 60 different kinds of furans, including most of the 23,7,8substituted types, in tow ppb levels (Wakimoto et al., 1988). No dioxins were detected above the limit of detection of 2 ppb. Used PCBs can have higher levels of furans. Apparently, the heat that they may be subjected to during use can promote the further formation of furans (pen. comm., P.E. des Rosiers, USEPA). However, several studies have shown that simple overheating or arcing in electrical equipment does not generate furans (des Rosien, 1987). While dioxins generally have not been detected in PCBs, a recent study found relatively high levels of dioxins in used oil (R. Adams, et aL, 1986). For many applications, such as dielectric fluids, PCBs are combined with other compounds such as chlorinated benzenes. There other compounds are believed to be responsible for the presence o f dioxins (see discussion o f chlorinated benzenes, below). Approximately 1.4 billion pounds of PCBs were purchased by U.S. industries prior to 1977. Based on there sales, an estimated 84-94 kilograms (kg) of furans were present in the PCBs produced during this period, including 8 kg of 23,73-T C D F and 14 kg of 23A7,8-peraaCDF (Hutzinger et aL, 1983). The authors noted that these were underestimates, because of missing sales information far some types o f PCBs. Most PCBs have been incinerated by licensed hazardous waste facilities or buried in designated landfills. However, a January 1988 U.S. General Accounting Office report estimates that 312 million pounds of PCBs still remain in millions of pieces of electrical equipment in the Dioxins and Furans 31 784132 TTO1 \ U.S., and that as much as ISO million pounds of PCBi have been released to the environment (pert, comm., USEPA). Low levels of PCBs are now ubiquitous in the environment, and it may be presumed that they are contaminated with trace levels of furans. In areas that are heavily contaminated with PCBs, it may be presumed that fiirans are present as welL Based on a comparison of the furan congeners in the tissues of aquatic animals (a snapping turtle, seal, and killer whale), to the furan congeners present in PCBs, several authors have concluded that PCBs were the source of the furans in the tissues (Rappe et al,, 1981; Waldmoto et al., 1988). The release of PCBs is now highly restricted, and they are probably no longer a significant source of new releases of furans (the uncontrolled burning of PCB-containing Quids can be an exception; see 'PCBs* on page 39). Hexachlorophene Hexachlorophene is a bactericide used in special soaps, veterinary medicine, and in restricted fashion as a surface disinfectant. It was once widely used in hospitals, in the cosmetics industry, and in some consumer products such as baby powder and skin cleansers. The USEPA no longer allows its use in non-prescription products. Very few measurements of dioxins in hexachlorophene have been made, but levels of firm 0.2-0.5 ppb 2,3,7,8-TCDD were found in three samples (Baughman, 1974, cited in Rappe, 1984). Present levels have been reported as equal to or less than 30 ppb (USEPA, 1985: p. 4.12). Hexachlorophene is currently not produced in the U.S. (USEPA, 1986: p 3.16). 2.4.5-tricloroohenoxyaceric acid (2A3-T\ The herbicide 2,4.5-T was used widely in the U.S. before 1980, and in a restricted fashion until 1983. It is no longer produced or used in the U.S. Dioxin concentrations as high as 100 ppm were found in 2,4,5-T poduced in the 1950s and 1960s, with an estimated average of about 1 ppm (Rappe et aL, 1982: p 496). 2J.73-TC D D was present in greater concentration than other tsomen* (Rappe, 1984). When this contamination was verified in the early 1970s, * An isomer is a single member of a homoiogue; a homologue is that group of dioxins or furans with the same number of chlorines. 32 Dioxins and Furans t GENP 011322 784133 r manufacturers lowered the level of contamination considerably, and in 1984 producers of 2,4,5-T repotted that the levels of 23,73-TCDD were below 0.1 ppm (Rxppe. 1984). 1110 authors of one study estimated this maximum amount of 2^3,7,8-TCDD that could have been present in 2,4,5-T produced bom 1960-1970 (Hutzinger et a t, 1985). Bared on total 1960-1970 production of 48.2 million kg, and a maximum concentration of 100 ppm 2J.7.8TCDD, they estimated a maximum total of 4800 kg 23,7,8-TCDD present in 2,4.5-T. The authors added that after 1970. both the level of contamination and level of production dropped sharply. The defoliant Agent Orange, which was sprayed heavily over most of South Vietnam during the Vietnam war, was composed of a 50:50 mixture of 2,4.5-T and 2,4-D. Levels of 2J.73-TC D D in Agent Orange were found to vary between 0.02 and 54 ppm, with an average of 2 ppm (reviewed in'Esposito et al., 1980: p 98). The very heavy spraying of South Vietnam resulted in significant contamination of the environment with 23,7,8-TCDD (Baughman and Meselson, 1973). In the U.S., normal spraying with 2,4,5-T appears to have contributed only trace levels of 2J,7,8-TCDD to the environment, presumably because the 2,3,7,8-TCDD 3 present in a solvent (2,4.5-T) and therefore can be quickly broken down by sunlight However, it is possible that the use of 2,4,5-T contributed to the low background levels of 2,3,7,8-TCDD in the fat of persons in the U.S. The herbicide 2,4,5-T is no longer produced or used in the U JL and is therefore no longer a source of dioxins here. However, a large amount of contaminated w as wxs created during the product of 2,4,5-T, resulting in many contaminated manufacturing and waste disposal sites (USEPA, 1987a). Some of there sites are still contributing dioxins to the covinximeru, predominandy the local area (see `What about dimtini and funns b o a sewage plants, waste streams, and landfills?", page 45). . 2.4-dichlotoohenoxvaccric.acid (2.4-Di The widely used herbicide 2,4-D is made bom dichlorophenoL Dichlarophcnol has only rarely been found to be contaminated with dioxins and funms, and 23,7,8-TCDD is not Dioxins and Funns 33 784134 oenpoii3 2 3 1 r4 formed in the manufacture of 2,4-D. However, other, much less toxic dioxin congnas have frrcn found in 2,4-D ai levels o f from less than 1000 ppm to less than 10 ppm total dioxins (Cochrane et aL, 1982). The authors of this study analyzed 2,4-D as acid, e sta , and amine formulations. The e s ta formulations contained the highest levels of dioxins. The majority of the dioxin congeners identified in 2,4-D by this study and others are considerably less toxic than 2,3.7,8-TCDD (e.g. 1,3,6,8-TCDD). or not toxic (dioxins with 1*3 chlorines). While the dichkxophenol from which 2,4-D is derived is not normally contaminated with 2J.7.8-TCDD, isolated cases of 2,3,7,8-TCDD contamination of 2,4-D may have occurred when equipment formerly used to process 2,4^ -T was used to produce 2,4-D (USEPA, 1986: pp 2.21-323). The contaminated equipment would not be expected to contribute more than very low levels of 2J.7.8-TCDD in the first few batches of 2,4-D. In Canada, an annual production of 8 million leg of 2,4-D was estimated to be the source of 9 kg of dioxins with from 2-4 chlorines (Tosine, 1983). In the U.S., 7.1 million kg o f amine formulations of 2,4-D were produced in 1986, and 6 million kg sold (U.S. Department of Commerce, 1987). No e s ta formulations were produced in the U.S. in 1986, but 3.7 million kg were sold. Because the level of production is similar to that in Canada, 1 assume it to be the source of a similar amount of dioxins. As in 2,4,5-T, dioxins in 2,4-D are present in a solvent (the herbicide) and therefore may be broken down by sunlight soon a f ta their application. This possibility, along with proper disposal of the wastes from the production of 2,4-D, make it probable that the production of 2,4-D is a relatively unimportant source of dioxins in the environment. Chlomahcnols Chloropbenoli are a principal ingredient in the processes used to manufacture many of the above chemicals. They or their derivatives an used to preserve wood and drilling muds, and as biocides f a process find cooling waters in some industries. Some of these uses may be phased out in the near future. In the past, chlorophenols and their derivatives have been used to preserve hides, textiles, paints, glues, and other materials, as disinfectants, and as biocides in fluids that aid in the eatin g of metals. 34 Dioxins andFurans GENP 011324 784135 Chlorophenols with one or two chlorines (chlorophenol and dfchkaoptaenof) have rarely been found to be contaminated with dioxins or furans. Chlorophenols with three or mare chlorines have been found to be contaminated with tu n y congeners of dioxins and furans, and 2,43-crichloropheaol is always contaminated with 23.73-TCD D (pen. comm.. Rolf Hanung, University of Michigan). Chksophenals with four ar fewer chlorines are no longer produced in the U.S. ( p e n comnL, John Wilkinsoo, Vulcan Chemicals). There is only one current manufacturer of pentachlorophenol in the U.S., and pentachlorophenol and a product made from pentachlorophenol, sodium pentachloiophenate, are used almost entirely to preserve wood or prevent wood staining (USEPA, 1986: pp 3.17-21; USEPA, 1988a; pp 33286-7). Pentachlorophenol has been commcroially produced by two basic methods: the hydrolysis of hexachlorobenzene or the direct chlorination of phenol (Esposito ct al, 1980: pp 78-88). Although both methods result in the fannadon of dioxins and furans, the hydrolysis of hexachlorobenzene can' result in higher levels of dioxins and futans that are of most concern. In the U.S,, pentachlorophenol has always been produced by the direct chlorination of phenol (Esposito e t al, 1980: pp 78-88). The USEPA has reported that solutions o f U-S.-produced pentachlorophenol in use at wood-preserving faciiirie* average approximately 300 ppb total dioxins and furans, in 2 3 ,7 3 - TCDD equivalents (USEPA, 1988a: p 33301). 2J.73-subsntm ed hexa-CDDs are responsible far moat of this calculated toxicity, and in the past were thought to be responsible for pcntachlcrophenoTi demonstrated toxicity. However, the resulu of a recent bioassay (NT?, 1988) showed that alone is a carcinogen in mice, and the presence o f 10 ppm hexa-CDDs in the pentachlorophenol made little or no difference. 23,73-nitetitutcd dioxins and A n n s with fewer than six chlorines am present in pentachlorophenol only at very tow detectable levels or below the limits o f detection. Tetra-dioxins, including 23,73-TCDD , have not been detected in numerous analyses o f U3.-produced pentachlorophenol (pen. comm., P . des R o se n , USEPA). A recent sod putualarly sensitive analysis of 10-year old pentschlorophenols and sodium poitachloropbenates. manufactured in Europe by the hydrolysis o f hexachlorobenzene^ Dioxins and Furans 33 i 784136 I'H'Ul'RP found very low ppb and ppt levels of 23.73-subttianed ctr*- and pen* dioxins and fauns, in arfrfiiifwi to 2 J ,73-substituted dioxins a id fauns with six or m a n chlorines (Hsgenmsier end Brunner, 1987). However, the low levels would have little impact an the estimated overall toxicity of the products; it would hoc differ signiftcanly from that of U-S.-produced pcntachlorophenols. Hagenmiaer and Brunner (1987) postulated that, in West Germany, the production and use of pexuachlorophenol and sodium pentachlorophcnaie may be, or have been in the past, the predominant source of dioxins and furans in their environment. The USEPA considers the wastes from wood-treatment facilities and sawmills to be sources of significant environmental contamination, and for that reason has regulated or proposed to regulate such wastes as hazardous (see 'W hat Has Been Done About Dioxins and Furans?", page 69). It should be pointed out that altonative wood preservatives (creosotes and inorganic arsenicals) are considered human carcinogens (USEPA, 1988a), and a study by Carnegie-Mellon University concluded that pcmachlorophenols are the safest wood preservatives available (Carnegie-Melloa Univeraity, 1982), Chktrophenols; especially 2,43-trichlorophenol, were probably sources of large quantities of 23,73-substituted dioxins and furans in the past (Tcaine, 1983; Hutzinger, 1985), and many contaminated sites remain from this era. However, levels of chlorophenol production and contamination have been greatly reduced. Based on a total annual production of pentacfalorophend in N crtrA m esica of 11 million kg (USEPA, 1988a), and ro avenge 23,73* TCDD- equivalent content of 300 ppb, this production would be the source o f 3 3 kg o f 23,73* TCDD equivalents annually in North America. M ott of this m o u n t is present in used peatachlorophenoi, and is mcmosied as a tm rdoua waste. Polychlorinated benzenes Etotychlorinated benzenes have been used as biocides, solvents, snd in electrical and chemical industries: They were often mixed with PCBi in dielectric fluids. Dioxins rod furans have been found in some chlorobenzenes at ppb rod ppm levels, although no infarourion is available on the specific types (reviewed in Heindl and Hutzinger, 1986). 36 Dioxins and Furans : GENP 784137 Diphenvl her herbicides Diphenyl ether herbicides can be derived from trichlorophenol and a variety of and furans w e n found in three types of diphenyl ether herbicides in one study (Yamagishi et aL, 1981, cited in Rappe. 1984: p 83a). Total concentration of dioxins and funas ranged from less than 2 ppm to about SO ppm. No 2J,7,8-TCDD was found and the few individual congeners that were identified are not considered very toxic. Hexachlorocvclohexanc Hexachlorocyclohexane is used primarily to produce the insecticide Lindane. Dioxins and furans are apparently formed by this process, and the waste has been found to be contaminated with ppm levels of dioxins and furans. including ppb levels of 2J,7.8-nibstituicd congeners (Scfaolz and Engler, 1987). The significance of this source is not known. Paper mills and products Pulp and paper mills that use chlorine bleaching process have been identified as sources of dioxins and furans. Dioxins and furana are believed to be formed when naturally occurring phenolic compounds in wood called lignins react with chlorine during the bleaching operations in the manufacture of pulp (Beck et aL, 1988). A preliminary study by the USEPA and the pulp and paper industry of five pulp and paper mills that use a chlorine' bleaching process, found that the wastewater effluent from these mills had concentrations of less than 1 ppt total dioxins and furans, arid the sludge had concentrations of less than 1 ppb total dioxins and furans (Amendota et aL, 1987). Because of the Large volume of effluents from pulp and peper mills and the potential for the acctmulation of dioxins and A n n s in organisms, even these low levels may pose significant risks to organisms o poaod to the effluents. As noted earlier in this report, low levels of 2J.7JJ-TCDD were found in fish downstream from some pulp and paper mills and these findings are o f concern to people who eat fish from these waters. The USEPA and the pulp and paper industry have a more extensive study underway. Researchers in Sweden determined that the isomeric pattern o f dioxins and final from pulp plants is easily distinguishable from that of dioxins and furans from incineraton (Swanson Dioxins and Furans 37 784138 { ' ', Vs ct al, 1988). They found that crabs and sediments near the pulp mill had dioxins and furans with an isomeric pattern similar to that in pulp, which then changed to the mrineraicr pattern as they moved away from the milL t h e authan estimated that the pulp industry in Sweden was a source of 5-15 grains of 2J.7.8-TCDD equivalents per year, and reported that the Swedish EPA estimates 1 kg of 2J.7.8-TCDD equivalents per year for all sources of dioxins and furans in Sweden. A study of West German paper products such as newsprint, coffee filters, and recycled paper found that they can contain a variety of Kama of dioxins and furans. Total concentrations of dioxins and furans were below 1 ppb with concentrations of the more toxic congenera at levels below 50 ppt (Beck et aL, 1988). A study sponsored by the National Council of the Paper Industry for Air and Stream Improvement (NCASI) determined that contact with paper products poses an insignificant health risk to humans (NCASI. 1987a). However, studies by researchers in Canada suggest that cardboard canons made of bleached pulp can transmit dioxins and furans to certain food products, such as milk (Ryan et al. 1988). Any such assessments include a large number of uncertainties, and further research is necessary. i W hat products can produce dioxins and (brans when they are burned? Dioxins and furans can be formed when many materials containing some fann of chlorine are burned, or when certain materials are burned in the presence of chlorine. The amount and kind of dioxins and furans produced varies greatly with the source and conditions. 38 Dioxins and Furans GENP 011328 I -L . 784139 V --- T ..... ......' I The following chlorinated products can all produce d io x in and/or furans when burned. Table 3. Chlorinated produce that can produce dioxins and/or furans when burned Product Contaminants PCBs Furans polybrominated biphenyls (PBBs) Dioxins and Furans i polybnxninaied diphenyl ethers (PBDEs) | polychlorinated diphenyl ethen (PCDEs) chlarophenols letrachloroethylene polychlorinated benzenes polyvinyl chloride (PVC) Leaded gasolines Note: When a material is burned in the laboratory to determine whether it is capable of producing dioxins and furans, it is typically burned under those conditions most likely to produce dioxins and furans, or that produce the largest quantity of dioxins and Turans. Therefore, the same substance may produce much smaller amounts of dioxins and furans under the uncontrolled conditions of a fire, or in the highly controlled conditions of an incinerator. EEBl PCBs are fire-resistant compounds and do not readily support combustion. However, if they are incompletely burned in fires fueled by other materials, furans may be formed. More than 1 ppra of each of the most toxic congeners of furans (4-7 chlorines -and 2J.7.3 positions filled) have been produced by bunting PCBs in the laboratory (Erickson et aL, 1984: Swanson et aL, 1985). Fires involving electrical equipment have contaminated buildings and other places with soot containing as much as 5000 parts per million total furans, including low ppm levels of 2J.7.8-TCDF (Erickson et aL, 1984; Rappe et aL, 1983: p 121), In the U.S.. PCBs are no laager produced or used in new equipment, and PCB-ccncaimng equipment in buildings is subject to stria regulation. However, large numbers of transformers and capacitors with PCB-conuminated dielectric fluids we still in use, and these contaminated dielectric fluids cao produce furans when they we burned (Narang a >1, 1988). PolvhromwMied hiphenvh fPBBri md Doivhromin^ rfinhenvl ethen fFBDPEsl Polybrominated biphenyls (PBBs) and polybrominated diphenyl ethen (PBDPEs) have been widely used as flame retardants in textile, carpets, and plastics. Although the addition of these nifastances nukes products such as clothing rod carpets less susceptible to igniting, the materials ewi bum in a building fire or incinerator. When burned in the laboratory, they have Dioxins rod Furans 39 784140 GENP 011329 produced brominated dioxins and furans, and smaller amounts of chlorinated dioxins and furans if they are burned in the presence of some form of chlorine (Buser, 1986; 1987; Thoma et aL, 1987). Polychlorinated diohenvl ethers JPCDEsl Polychlorinated diphenyl ethers (PCDEs) are used as herbicides and have been found as low-level contaminants in chloraphenols (Soikkeli et al., 1986). When bunted in the laboratory, they can produce significant amounts of dioxins and furans (Rappe, 1984: p 83a). Chloroohenols When burned, chlorophenols can produce significant amounts of dioxins and furans, especially (hose with from 4-6 chlorines, and including the toxic 2 J ,7 3 congeners (Rappe et al, 1978; Rappe et aL, 1983: p 108). An unknown amount of chlorophenol-trcated wood is eventually burned, and,m ay be a widespread source of significant amounts of dioxins and furans. Current USEPA registration for pentachlorophenol used in wood treatment specifies that the wood is not to be burned, although wood-treatment facilities may bum wood so a p in higbtemperature boilers. One study found that levels of octa-CDD were doubled when PCP-treated plywood was burned, and dioxins with six and seven chlorines were also detected (Crosby et aL, 1973). Wood-treatment facilities often collect waste pentachlorophenol in ponds, and in the past they periodically set fire to the ponds to reduce their volume. This practice generated large amounts of dioxins and furans (P.E. des Rosier, USEPA, cited in USEPA, 1986: p 3-18). Tetrachlomcthvlcnc and polychlorinated benzenes TemsdUarethylene and polychlorinated benzenes are used as cooling fluids in electrical equipment, often in combination with PCBs. When burned in the laboratory, they can produce furans and lesser amounts o f dioxins, including the most toxic congener (Erickson et aL, 1984). Polychlorinated benzenes produced less than l/10th the amount of furans produced by PCBs, and tetracfaloroethylene produced less than 1/lOOOth. However, these are significant amounts nonetheless. For example, polychlorinated benzenes are believed to be responsible for the dioxin contamination of an office building in Binghamton. New York, after a transformer fue. j 40 Dioxins and Furans GENP 011330 784141 The soot produced by this fire had 20 ppb tool dioxins, including 0.6 ppb 2J,7,g-TCDD (Rappe et aL, 1983: p 121). P p lY V iiiY lJM gid gJPV Q Polyvinyl chloride (PV Q is a commoo plastic. When burned in the labarauvy, it can produce low ppb levels of dioxins and furans (Marklund et xL, 1986: p 90). L uge mourns of PVC plastic are present in municipal waste, and are suspected of contributing to the riintim and furans emitted from incinerators. Leaded gasoline Low levels of dioxins and furans have been detected in motor oil and in the h im of c an that use leaded gasoline (Marklund et aL, 1987), but not c an that use unleaded gawimq The authors noted that dioxins and furans may be destroyed in the catalytic convenors of the c an burning unleaded'gasoline. Dioxins have been found in m uffins from diesel trucks (Bumb et aL, 1980), and dioxins and furans have been found in motor oils that are recycled from used motor and other oils (Roland et aL, 1987). Diehloro and dibromo ethane (DGE and D8E) are used as additives in leaded gasolines, and ate thought to be the precursors to dioxins and furans. Leaded gasoline is a very widely used product, and therefore even the low levels formed are potentially important. It was estimated that autos burning leaded gasoline in Sweden produce emissions with a total of 10-100 grams of 2J.7.8-TCDD equivalents per year (Marklund et aL, 1987). Using their emission values (30-540 picograms of 2^3,73-TCDD equivalcnisflrilomctcr) and assuming their left autos averaged 20 miles (32 kilometers) per gallon of jpsoline (the models were a Saab 900, a VW Golf, an Opel Kadeo, and a Volvo 245), total emissions from U.S, autos can be estimated. Based an in average figure o f approximately 57 millioa gallons of leaded gas used per day in 1988 in the U.S. (pea. coautL, American Petroleum Institute), and the values above, n tn a burning leaded gasoline in the U.5. may produce emission with from (57 mi gallocs/day x 365 days x 32 kilometets/gnllon x 30 picogrzms/kilometer) to (57 million gaUona/Uay x 365 days x 32 kilom ag/gallon x 540 picogramsfldlometer), or 20 to 360 grama of 2J.73-TC D D equivalents per yew. The use of Dioxins and Farm s 41 784142 leaded gasoline is steadily decreasing in the U.S. and much of Europe, thus reducing the contribution from this source. The burning o f dioxins and furans The incineration of dioxins and furans in the laboratory has produced other dioxins and furans with fewer chlorines. For instance, incineration of octa-CDD or octa-CDF can produce dioxins or furans with from four to seven chlorines (Maiklund et aL, 1986; Swanson el aL, 1986). This is a potentially significant process because the toxicity of dioxins and furans generally increases as the number of chlorines decreases from eight to four. Also, octa-CDD and octa-CDF are often the most numerous types of dioxin or furan in contaminated products or waste. Does the burning of paper, wood, peat, or coal produce dioxins and furans? Several studies have shown that wood, peat, and coal can produce dioxin* and furans when they are burned. Apparently, naturally occurring phenolic compounds in these substances (lignins) can react with naturally present chlorine to form chlorinated phenols; which then form dioxins and furans (Beck et al., 1988). Measurements of dioxin and furan emissions from coal- and peat-fired power plants indicate that thej emit only very small amounts of dioxins and furans, perhaps in part because the combustion conditions are carefully controlled for high efficiency (Kimble and Gross, 1980; Markltmd et aL, 1986; USEPA, 1987a: pp 4.1-26). Likewise, the combustion of paper has been found to produce only very small amounts of dioxins and furans (Olie et aL, 1982; USEPA, 1987k pp 4.1-26). The combustion of wood, on the other hand, is a potentially significant source of dioxins and furans. Dioxins and furans, including the most toxic congeners, have been detected in small amounts in emissions from wood stoves and wood boilers (Nestridc and Lamperxki, 1983; USEPA, 1987k pp 4.1-26; Thoma, 1988). Because of the large mnnher of wood combustion sources in the U.S., this could be a significant source. However, analyses of lake sediments from an island in Lake Superior did not find dioxins and furans in layers 42 Dioxins and Furans ( '' i. bom before 1940, when wood combustion was prevalent in North America (Cztxzwa a n l Hitea, 1986). Wood that has been treated with pentachlorophcnol or soditun penfachlaropheamc would be especially likely to emit dioxins and farm s when bunted. However, the major use of pentachlorophenol is for (be treaunem of utility poles, and few if any are burned in wood stoves. Do municipal incinerators produce dioxins and furans? Municipal solid-waste incinerators are recognized as important sources of dioxins and furans. They are found oa the dust and in the gas emitted through the stack; on the Qy ash and bottom ash that is collected and placed in landfills; and in the scrubber water that is part of the emission control system (Marklund et aL, 1986; USEPA, 1987a: pp 4.1-26; Hiroaka et a i, 1987). While municipal incinerators have been found to produce all types o f dioxins and furans (Karasek and Hutzinger, 1986), the amounts and distribution o f the different dioxins and furans produced by municipal incinerators vary widely and therefore so do estimates of the amounts of dioxins and furans produced in total by municipal solid-waste incineration. Incinerators operated under the proper conditions and equipped with the most modem emission control devices have greatly reduced dioxin and furan emissions (Hay et aL. 1987; USEPA, 1988b). Rappe et aL (1987a: p 1604) eatimaca that a normal-sized municipal solid waste incinerator (which bums 50,000-200,000 tons of gsb ag e per year), operating under normal conditions, emits through the stack 1-100 grains o f 2J,7,3-TCDD equivalents per year (based an a range of 1-100 nanograms of 2J.73-TC D D equivalents per cubic meter of stack effluent). This figure can thus be interpreted ss 5 millionths (1/200000) to 2 thousandths (100/50000) of a gram o f 2J.7.8-TCDD equivalent emitted through' the stack per ton o f garbage burned. According to a recent USEPA report (USEPA, 1988b), 16J million tons o f garbage are incinerated in the U.S. each year. Using Rappe's range as given above, municipal incinerators in the U.S. may be the source o f from (5 x 10-6 gramsAon x 1 6 J x 10* tons) to (2 x 10* Dioxins and Furans 43 784144 GENP 011333 r ramsrton x 16.5 x 10* ions), or from 0.0825 to 33 kg of 2J,7$-T C D D equivalents per year, distributed in their stack effluents. This estimate does not take into account the dioxins and rural jn the bottom ash, fly ash, and scrubber wmct. the majority of which an placed in landfills. In Canaria, dioxin and furan emissions from municipal incinerators has been m im aird to be 6.3 kg/year TCDD equivalent in the fly ash, and 6.1 kg/year TCDD equivalents in the stack emissions (Tosine, 1983). A study in Switzerland showed that levels of dioxins and farans in cow milk were higher in samples collected near incinerators (Rappe et aL, 1987b). However, in another study (Rappe et al., 1987c), ihe authors mention that while Sweden has 25-30 incinerators and Yugoslavia has none, similar levels of dioxins and furans are found in human milk from these countries. Apparently, incinerators are only one of many sources dial contribute dioxins and furans to the environment in these countries. Do other large combustioa sources produce dioxins and farina? There are other potentially large combustioa sources of dioxins and furans. Hazardous and hospital waste incineraton, sewage sludge incinerators, and smelters and mills that recycle metals can all produce significant amounts of dioxins and furans, including the more toxic types (USEPA. 1987a: pp 4.1-26; Clement et aL, 1987; Marfclimd et aL, 1986). The dioxin and furan emissions from other large combustioa sources have not been as well studied as those from municipal in d n en to n . The available data indicate that the emissions from a hazvdous waste or sewage sludge incinerator are of the same magnitude as a municipal incinerator (USEPA, 1987a: pp 4.M .25; Tsuji et aL. 1987; C em ent et aL, 1987). Because them are many more large industrial combustioa sources than muracipal m cm o u o n , their total i b m w of and farm s could be greater than front municipal incineraton (Rappe, 1987). 44 Dioxins and Furans G E N P 011334 784145 \ What about dioxins and furans from u w ift plants, waits streams, and tandflQs? Dioxins and furans have been found in sewage sludge tram municipal waste-water treatment plants in Germany, in levels below 100 ppb (Hagcnmaier et aL, 1986). The mjc varieties of dioxins and furans were present at levels below l ppb. Based on an analysis of the ratios of [he various congeners of dioxins and furans present, the authors proposed that pemachlorophenol and its derivatives are the source of most of the dioxins and furans in the sludge. In addition to the products themselves, the waste generated by the production of chiorophenol-based products is a potentially important source of dioxins and furans. Ironically, the contamination of the wastes is thought to have increased significantly as a result of efforts to reduce the contamination of the products: the dioxins and furans were concentrated further in the wastes. Landfills that contain the wastes from the manufacture of chlorophenols and chiorophenol-based products, that contain PCBs, or that contain fly ash, are suspected sources of dioxins and furans. The Love Canal and ncigmd chemical landfills in New York, for example; have contaminated the adjacent Niagara River with dioxins, and the Venae manufacturing and waste-disposal site in Jacksonville, Arkansas, has contaminated the nearby Bayou Meto and Arkansas River. Also, sites where chlorophenols and chiorophenol-based products were manufactured or processed are now considered sources of dioxins and furans (USEPA, 1987a: pp 3.1-26). As a result of these incidents, wastes containing dioxins and furans have been disposed o f much more carefully. Are there other processes that create dioxins and furans? It has been shown that dioxins and furans can be broken down by sunlight (this is called photolysis) to other types of and furans with fewer chlorines. Far example, the octachlorodioxin (OCDD), when exposed to UV light, can be broken down to potentially more toxic dioxins with fewer numbers of chlorines (Crosby et a)., 1971). Combustion processes Dioxins and Furans 45 I 784146 GENP 011335 V often ptodice greater amounts of octt-CDD than ail other dioxins combined, and it is often the most common dioxin in chtorophenols. However, the photolysis would then be expec ted to continue, eventually breaking down any toxic varieties into much less toxic C o n n Photolysis can also fonn dioxins and finans from other compounds. U has been shown that a mixture of polychlorinated benzenes and phenol, when exposed to UV light, c a t pnxface (u rn s (Choudhaxy et al., 1983). Both polychlorinated benzenes and phenols are common environmental contaminants. Also, the photolysis of cfalotophenols has been shown to produce octa-CDD (USEPA, 1983: p 4.17). Thus, wood and other materials preserved with chtorophenols and exposed to sunlight may result in the fonnaiion of dioxins. The significance of these photochemical processes is not yet known. T Conclusions on sources of dioxins and furans \ There are many sources of dioxins and furans. Some of them, such as s c u te s at combustion, are widespread and numerous. O th m , such as contaminated manufacturing and waste sites, are fewer in number and usually local problems. The significance of the photochemical processes that can convert less toxic compounds to dioxins and Allans is not yet known. Also, the relative importance of what appears to be the two major sources o f dioxins and fuians (chlorophenols and chlorophenol^btsod products, and sources of combustion) is not yet known. It is generally agreed that dioxins and furans in our environment come primarily from recent human activities (chemical production and use, m cm endan, automobiles). Czutzwa and Hites (1983) analyzed sediment cores from the G t o t Lakes and a Lake in Switzerland. Dioxins and furans in these sediments were at or below the Omit of detection until they reached the layers deposited after approximately 1940, a time that they say corresponds to the beginning of large-scale manufacture and disposal (often by incineration) o f chlorinated aromatic compounds. Also, levels of dioxins and furans in human fatty tissue and breast milk from North Vietnam are the lowest yet measured while levels in South Vietnam are aimiUr to the U.S., Japan, and t! 46 Dioxins and Furans 784147 1 Europe (Scbecter et iL , 1986; Rappe e t a t, 1987c p 233). This mrfieei due modem inlusoial activity, not the burning of wood, is the primary s o m e of dioxins sod l i r a s to the environment There is same evidence that the mkxzu of dioxins and fuians being released into the environment reached a peak in the tnuM970s ra d has declined since. The sediment cores analyzed by Czoczwa and Hites (1986) showed such a trend. An analysis of breast milk in Sweden also appeared to show this decline the levels of dioxins and furans in combined samples of breast milk have dropped since the study was initiated in 1972 (Karen. 1988). However, studies of diaxins and furans in breast milk in Yugoslavia (1981 to 1987) and Japan (1978 to 1984) do not show changes over time (reviewed in Undsuom. 1988: p 36). Dioxins and Furans 47 784148 GENP 011337 The Health Effects of Dioxins and Far, la Animals and Humans T W hat are the health effects on animals? Some general features Dioxins and furans have some very unusual features with respect to their toxicity to animals. Animals given fatal doses of dioxins and furans do not die immediately, several weeks elapse between the administration of the fatal dn<g and the death of the animaL Also, many of the acute (short-term) effects of less than fatal doses are temporary. When the animal stops receiving the substance, it recovers from most of the effects. The rale of recovery varies, and some animals have required more than six months to return to normal (USEPA, 1985: p 8.48). The toxicity of dioxins and furans is highly variable. They produce a wide spectrum of I effects, involving many different organs and body systems, and the affected areas usually vary with the species of animaL The toxicity of different dioxins and furans varies widely for a given species of animaL and the toxicity of a given dioxin or fuian varies widely between different species of animals. See Tables 4 and 5 far acutely lethal (causes death quickly) doses of 2 J ,7,8-TCDD for a variety of species. Finally, some individuals of a sensitive species (rats or mice, for example) do not die even when given doses as much as 100 times greater than the amount that is fatal to the other individuals in the same study. 48 Dioxins and Furans L 784149 Table 4. Acutelv_leihal single doses of 2J,7A-TCDD fLDSm (from USEPA, 1985. Table 8.1) Speda Gender Dom m iao g n m s per kg of body weight (ug/kg) Guinea Pig Male O^-Zl Guinea Pig Rat Female Male Z5-19 22 Rat Female 45 Monkey Female <70 Rabbit Rabbit Male Female 115 ' 115 Mouse Male 114-285 Mouse Female >450 Hamster Male 1157-5051 Note; The lethal values above represent dnsei that caused the death of 50 p a re nt o f the test animat. Such a value is known as an LD50, and is a standard endpoint in a toxicity test. Far aquatic animals, different concentrations of the substance in water ate used, instead of a dose. The concentration that* causes death of 50 p a re n t of the test animals is called the LC50 (see table below). Thera are also studies that use different endpoints, such as LD95 or LC95, where 95 p a re n t of the test animals die. Table 5. Lethal exposures lo 2J.7.8-TCDD in water fl-CSQ) (reviewed in Kenaga and Nanis, 1983) Spedes CoocentntkM la W ater (ug/liier) Channel Catfish Coho Salmon Guppy 0.0042 (4.2 ppO 0.0056 (5.6 ppt) 0.1-10(ppb) Duration (days) 15 1-3 5 Dioxins and Funns 49 784150 gexpoiI 3 3 9 T Generally speaking, all toxic memben of the dioxin and furan Camilla produce similar effects in a given species of animal, and several effects are common to all mammals that receive a lethal dose of these compounds. Although m o t stu d ia of the effects of dinxins and furani on laboratory animals have been done using 2,3,7,8-TCDD, these stu d ia can be used to predict the toxic effects of other dioxins and furans. The chief difference is that a g re ats dose of other dioxins or furans is necessary to produce the same effects. Many dioxins and furans. are so much less toxic than 2,3,7,8-TCDD that an experiment would require an impracticably large dose to produce lethal effects. Doses that produce no observable adverse health effects Long-term studies of mice and rats have shown that there are doses o f 2.3.7,8-TCDD that produce no observed effects on the animals, even after receiving the substance far two yean. The USEPA estimates the No Observed Advene Effects Level (NOAEL) far toxic effects other than cancer to be approximately 0.001 ug/kg/day far both mice and rats. They also cited a need for long-term stu d ia using lower doses (USEPA, 1985: p 14.10); some researchers feel that this dose can produce adverse reproductive effects (USEPA, 1985; p 14.10), and is therefore a Lowest Observed Adverse Effect Level (LOAEL). See Tabl a 6, 7. and 8 far NOAELs and LOAELs for other species. Also, the toxicity of 2,3,7,8-TCDD is a t least partially cumulative; a smaller dose than the acutely lethal dose can cause lethal effects if the smaller dose is given for a longer period of tune (McConnell, 1980: pp 110-111). l ; Table 6. Lowest d o with an observed advene effect (LOAELI (from USEPA, 1985, Tables 8-4, 8-5) Speda Doe Duration Effects Rat 0,01 u^kg/dsy 2 yen Liver damage Mouse 0.001 ug/kg/day 1 yea Several Guinea Pig 0.006 ug/kg/day 8 weeks Several (For the stud i a above, the animals received 2^,73-TC D D in their food or were face-fed 2J.73-T C D D ow e per week in a medium such as com oiL) SODioxins and Furans GEMP 011340 784151 Table 7. Lowest concentration with observed adverse effect fLOAELl Species Concentratisi la Water ug/liter (ppb) Duration days Northern Pike (eggs)* 0.0001 (0.1 ppt) 4 Rainbow Trout* 0.0001 (0.1 ppt) 4 Fathead Minnows" 0.0017 (1.7 ppt) 28 (Helder. 1982) (W. Adams, et aL, 1986) Effect Lowered Survivil Lowered Survival Lowered Survival Table-8. Highest concentration with no observed adverse effect TNOAEL1 (reviewed in Kenaga and Norris, 1983) Species Concentritioa in W ater ug/liter (ppb) Coho Salmon 0.00036 (0.36 ppt) Rainbow Trout 0.0001 (0.1 ppt) Mosquito larvae fAedes aeyvptil 0.2 Daphnia fPanhnia mama) 1.33 Alga fOedoeonium cardiacuml U3 Duration days 4 4 17 32 32 Hie short-term health effects on animals All mammals given an acutely lettaci dose o f a dioxin or Itinn suffer from wasting (losing weight), and it is sometimes the only effect leading to death. The simply waste away. This ocean primarily the do not esc enough food to minam (heir weight. Apparently, the ammali* internal weight 'setting" is lowered below m h m I, and so the animala do not eat enough food. H ow em , than is evidence that this reduction of nourishment is not entirely responsible for the weight loss. R esearch who force-fed the nimal could prevent some of the weight loss, but not all. M id the died anyway (Gasiewicz et iL , 1980). Dioxins and Furans 31 784152 G EN P011341 f N'( V. / T The thymus is affected in all mammals exposed to a toxic dose, and the liver is affected in most mammals. The thymus, and to a lesser extent the spleen and lymph nodes, typically sufTer damage and a reduction in size, while the liver typically suffers damage and an increase in size. In rodents, liver damage is suspected to be a principle causa o f death. Skin disorders have occurred in rabbits, monkeys, cattle, and hairiest mice. The effects inc lude severe acne, thickening of the eyelid, and abnormal growth or loss of hair, fingernails, toenails, or hooves. Dioxins and furans have also affected the cardiovascular system, gastrointestinal and urinary tracts, spleen, bone marrow, and gallbladder of animals. For reviews of the above health effects, see Gupta et aL, 1973; Poland and Knutson, 1982; USEPA, 1985. Cancers 2,3,7,8-TCDD is a carcinogen in mice and rats, when fed to the animal or applied to the skin. Two-year studies at Dow Chemical Corporation (Kociba et aL, 1978) and at the National Cancer Institute (NTP. 1980a; 1980b) found increased cancers of several types, resulting, from chronic (l<wg*nn) doses as low as 0.01 ug/kg/day. Rats fed this dose far two . yean had, a t the end of the Dow study, 1700 ppl 2J.7.8-TCDD in their fax. The rats fed this dose also suffered increased mortality, decreased weight gain, and increased excretion of porphyrins (a pigment produced by the liver which indicates liver damage). Rais fed 0.001 ug/kg/day suffered no health effects of ooosequenoe, and had 540 ppt 2J.7.8-TCDD in their body fix at the end of the study. Long-term studies have also been conducted using a m ix tm of two hexa-CDDs, both containing chlorine atoms in the 23,7,aod 8 positions (NTP, 1980c; 1980d. The mixture caused liver tumors in both mice and rats when administered orally in dceei of 2 J-5 .0 ug/fcg/week (far female rats), 5.0 ug/kg/wrek (for male mice and rats), and IOjO ug/kg/week (for female mice), but not at lower doses. The mixture was not esdnogem e when 0.1 tig was applied to the skin three tunes per week. Many researchers have concluded that 2J.7.8-TCDD is a promoting, rather than an initimg, carcinogen. A promoter provides a favorable environment for tumor growth, while an 52 Dioxins and Furans Lt GENP 011342 784153 initiator produces a change in DNA that causes tumorous growth. Tumor promotion is revenibifi and is not inherited, while tumor initiation is not reversiblo and is ptwed on through cell division (Paustenbacb et al, 1986). Pitot ct al (1980) used diethyltritrosamine to initiate turnon in mice, and found that 2^,7,8-TCDD promoted the growth of the initMtwt tumon. Also, Kouri et ai (1978) used 3-methyl-cholanthrene to initiate subcutaneous turnon in mice, and found that 2 J.73-TC D D promoted the growth of these tumon. Reproductive effects Dioxins and furans appear to be especially toxic to embryos and fetuses, and the .reproductive system of adult animals appears to be especially sensitive as welL Mice and rats have produced liuen with abnormally high rales of birth defects such as cleft palates, extra ribs, and deformed livers and kidneys, after receiving doses of 23,7,8-TCDD as low as 0.01 ug/kg/day, or a single dose of 1 ug/kg (reviewed in USEPA, 1985: pp 9.1-23). Reproductive effects such as smaller liuen, lower birth weights, and premature abortions have been reported in monkeys and Tenets, as well as in rats and mice, that were administered 2,3,7,8-TCDD. When 23,7,8-TCDD was fed continuously to three generations of laboratory rats, reproduction was impaired by a dose of 0.01 ug/kg/day, but not 0.001 ug/kg/day (reviewed in USEPA, 1985: p D .l). Female Rhesus monkeys suffered reduced fertility after being fed 0.0015 ug/kg/day 2J.7.8-TCDD (reviewed in USEPA, 1985: p D.l). These effects have only been observed when the embryo or mother was exposed to 2J.73-TC D D , and appear to be the result of direct exposure or transfer of the toxin from mother to embryo across the placenta Birth defects have not been associated with exposure o ily to the father, nor has reproduction been reduced when only the father was exposed to 2J.73-TC D D (reviewed in Kamrin and Matsumuia, 1985; reviewed in Silbergdd and Maoism, 1987: p 137). However, 2 J .7,8-TCDD can reduce the weight of testes, lower levels of testosterone, and impair the development of sperm, in rodents, monkeys, and chickens (reviewed in Poland and Knutson, 1982: p 524). i Dioxins and Furans 53 784154 Mutagenic effects An important question about any toxic substance ia whether it cauaea changes in the DNA of exposal organisms, because such changes can then cause cellular mutations. Such a substance is called a mutagen. A small number of early studies found 2J.73-TC D D to be a weak mutagen, while many subsequent studies have detected no evidence of mutagenicity. Most researchers have concluded that 23.7JB-TCDD is not a mutagen (reviewed in Hay, 1984; reviewed in Poland and Knutson, 1982: p 525). Immyrn effects 23.7.8- TCDD has been shown to inhibit the immune system of mice, rats, and guinea pigs, at doses much below the fatal level. The thymus, spleen, and bone marrow have all been shown to suffer loss o f tissue, and the enzyme immune system and cell immune system which normally react to a potentially toxic substance are inhibited (reviewed in Poland and Knutsoo.' 1982; p 522; reviewed in USEPA, 1985: pp 8.85-86). Similar responses have been observed with 2 J ,7 r8-TCDF. at doses thirty times higher than for 2J,7,8-TCDD. The effects of the 2J,73-T C O F disappeared six weeks after dosing ended (Vccchi a iL, 1983). EffwiLfliLblMd-conflimcnB 2.3.7.8- TCDD and 2J.7.S-TCDF have been shown to affect a variety of constituents in the blood of test animals. 2J.7.8-TCDD has been shown to reduce the number of red and white blood cells (Kociba et aL, 1978), change the level of blood lipids such as triglycerides and esters, and change the levels of thyroid hom ones such aa thyroxine and thyrotropin (reviewed ia Poland and Knutson, 1982: pp 526*529). Blood levels of progesterone and estrogen decreased in Rhesus monkeys fed 2^.73-TCDD, and blood levels o f testosterone decreased in n et (reviewed in Peterson et aL, 1984). 2J.73-T C D F has been shown to increase the level of serom globulin in mice, and decrease serum cbolesterol and senan iPamtin in monkeys (Moore aL, 1979). 54 Dioxins and Furans 784155 A n dioxins and furans hazardous to organisms in th environment? Very Utile is known about the impact of dioxins and furans on organisms in the environment, especially at low levels. Many animals died after being exposed to relatively high levels of dioxins and furans, as a result of the accidental contamination of areas surrounding Seveso, Italy, and Times Beach, Missouri (reviewed in Reggiani, 1980: pp 318, 325). It was reported that six months after the Seveso accident, normal populations of wild and domestic animals were present in the contaminated area, and appeared to be in good health (Homberger ct aU 1983, cited in OME, 1985: p 3.92). Fourteen rabbits - from the zone of highest contamination were collected, sacrificed, and examined, three months after the accident. All had more than 300 ppb 2,3,7.8-TCDD in their livers, and 5 showed evidence of liver damage. Blood constituents were normal in all 14 rabbits, and no other health problems were observed (Abbruzzi et aL, 1977, cited in OME, 1985: p 3.92). 2^,7,8-TCDD is also suspected of causing reproductive failure in fish-eadng birds of Lake Erie and Lake Ontario (Gilbertson and Fox, 1983). The reproductive failure was primarily the result o f unusually high embryo and chick mortality.' A study of several gull colonies found that the embryos and chicks were suffering from chick edema disease and hepatic porphyria, which are both indicators of poisoning by dioxins, furans, and related compounds such as PCBs. Relatively high levels of 2,3,73-TCDD were later detected in gulls from Lakes Ontario and Erie (Norarom et al., 1982). Other contaminants such as PCBs, DDT; and hexachlorobenzene were also detected in relatively high levels in these birds, and may have been responsible far these health effects. Several environmental studies o f organisms exposed to 2J.7.8-TCDD and other dioxins and furans have found little or no health effects. A 15-year saidy of beadumce at a highly contaminated site in Florida (Young et aL, 1987) found dial the 1 ppb levels of 2^,7,8-TCDD in the soil had no observable effect on the birthrate of the mice, and enlarged livers was the only observed effect. Also, NCASI sponsored a five-month study of plants and animals exposed to dioxins and furans from paper sludge spread on a pine plantation (NCASI, 1987b). The 10 Dioxins and Furans 55 784156 G N P 011345 ppt 2J.7.8-TCDD and 100 ppt 2J.73-TC D F in soil had no observable effects on the health and reproduction of the organisms studied. More studies of the environmental effects of dioxins and furans are needed. What are the pharmacokinetics of dioxins and furans? In order to understand why many dioxins and furans accumulate and persist in animals and humans exposed to them, why the processes vary with the type of dioxin or ftuan, and how these characteristics relate to the varying toxicity of dioxins and furans, it is helpful to know something about the pharmacokinetics of dioxins and furans (their absorption, distribution, metabolism, accumulation, and elimination, in animals and humans). Dioxins and 'furans arc intermediate in their affinity for fat (this is known as lipid* solubility). They are soluble enough to be readily absorbed through the walls of the digestive tract, but not so lipid-soluble that, once absorbed, they are found only in the adipose (fatty) tissues of the body. They are also found in the liver and in other organs. Adipose tissue is the V. body's storage depot, and substances there are 'in limbo". They are not interacting with the mechanisms that cause effects (toxic or otherwise), nor with the mechanisms that metabolize the substances. Therefore, the intermediate lipid-solubility of dioxins and furans makes it possible for them to persist in the body and be available to the mechanisms that cause toxic effects (Matthews and Bimbaum, 1983). Studies with rodents show that 2J,7$-TC D D is absorbed primarily through the gasnointestinal tract, and to a much lesser extent through the skin and lungs. When 2J.7.8- TCDD was administered orally in food or in c o n oil, bom 30*90% of the amount was then absorbed through the gastrointestinal tract (reviewed in USEPA, 1985: p 12). When applied to the skin in a solvent such as methanol, from 1-10% was absorbed. The absmbed 2J.7.8-TCDD is distributed rapidly to many parts of the body, but the major of storage are almost entirely the liver and the fatty tissues (reviewed in Gasicwicz et tL, 1983). The' rite of storage vary somewhat with the spedes. 23,7,8-TCDD is stored in its original fonn, not as a changed 36 Dioxins and Furans GENP 011346 784157 f T1 product of raembolism. Apparently, metabolites are excreted once they are fanned. It appears that the absorption of dioxins and funns decreases as the number of eMtyingf increases (reviewed in Matthews and Btmbcunv, 1983). The metabolism of 2J,7,8-TCDD has been studied primarily in rodents. Metabolism is affected by the number and position of chlorine atoms (reviewed in Matthews and Bimbaum, 1983). The critical factor for metabolism is the presence of two adjacent unsubstituted carbon atoms (not attached to a chlorine or other atom). When chlorines Gil the 2 J,7 , and 8 positions, no adjacent unsubstituted carbon atoms are left on the dioxin or furan molecule, and very little metabolism takes place. A year after people in Japan ingested a number o f different types of furans that were accidentally present in rice oil (known as the "Yusbo" incident), no furans with adjacent unsubstituted carbon atoms remained in their tissues. Apparently, these types had been metabolized and excreted (Rappe et aL, 1979. cited in Matthews and Bimbaum, 1983). Furans with the 2 J.7 . and 8 positions filled were still present in the tissues. In the general population, the only dioxins and furans found in people are those that are 2J3,7^-subsrituted (Ryan, 1986). The accumulation of dioxins and furans is the product of absorption and metabolism, and varies with the type of dioxin or furan. For instance, studies with carp and rodents have shown that they preferentially absorb those types with the 2 ^ ,7 jn d 8 positions filled, and they absorb a greater proportion of the available dioxin or furan as the number of chlorines decreases (Kuehl et aL, 1986; 1987a; 1987b; Opperhuizea et aL, 1986; van den Berg et aL, 1986). Alibough relatively small amounts of hepta- and octa-CDDs/CDFs are a b a t e d , these smaller amounts accumulate and persist in the body for longer periods of time; U .7.8-TC D D has been dimim md slowly from the bodies of exposed animals; the half-life varies from 10 days to approximately one yew, depending oq the specks. In humans; 23,7,8-TCDD appears to be eliminated quite slowly. Based an the rate of disappearance of one billionth o f a gram o f self-ingested 2J.73-TC D D , Poiger and. Schlatter (1986) estimate a half-life of about six years. They also found that the dioxin was distributed quickly in the blood and stared almost entirely in the fatty tissues. A hxlf-Ufe o f about seven yean was Dioxins and Furans 37 L' 784158 fay researchers from the CDC, based on measurements of 23,7,8-TCDD in the blood of Ranch Hand personnel (Pirkle et aL, cited in CDC, 1988). Recent work with pharmacokinetic models suggests that the half-life of 2J.7.8-TCDD is inversely related to its concentration in human fatty tissue (Kissel and Robarge, 1988). They predict that the half-life is 4.4 years when 2,3,7,8-TCDD is present in fatty tissue at 100 ppt, but increases to 20 years as the tissue concentration approaches 10 ppt (a typical level in a person from an industrial country). The rate of elimination of dioxins and furans increases as the number of chlorines decreases (reviewed in Matthews and Bimbaum, 1983; Kuehl et al., 1987b), and studies with rodents and fish have shown that furans are eliminated more quickly than their counterpart dioxins (van den Berg ct al., 1983; Kuehl et a t, 1987b). Apparently, 2,3,7,8-TCDD must be metabolized to different compounds in order to be excreted in the urine and bile (reviewed in Olson ct al., 1983: p 97). However, unmetabolized 2J,7,8-TCDD can be excreted in the feces (USEPA, 1983: p 7.20) and in breast milk (Moore et a t, 1976; van den Berg et a t, 1986). Studies of people exposed to furans in the Yusho incident showed that the furans were also stored in the fat tissue and in the liver, and were metabolized at a slightly faster rate than that found for 2,3,7,8-TCDD, above (reviewed in Reggiani, 1983: pp 59-61). It has also been shown that nursing women excrete both dioxins and furans in their breast milk, and there is some evidence that nursing can reduce the amourui of these compounds in the body (Noren, 1988; Ofpki et aL, 1987, reviewed in Liitdanxn, 1988: p 34). In some cases; the toxicity of a dioxin or futon has been related to the rate of metabolism. For instance, the toxicity of 2J.73-T C D F to different species was found to be inversely proportional to the rate of metabolism (reviewed in Matthews and Bimbaum. 1983). Also, a study of 23-di-CDD in goldfish found that, when the metabolism of this dioxin was prevented, it was l/60th as toxic as 2J.7.8-TCDD (Sijtn and Oppertiuizen, 1988). Normally, 2^-di-CDD is metabolized Coo quickly to accumulate, and is almost notunxie. Metabolism alone cannot explain the varying toxicides of mm and furans, however. For instance, rats metabolize 2,3,7,8-TCDD about three times faster than do guinea pigs, yet the LD50 for the rat 38 Dioxins and Furans 784159 [ is 3000 to 3000 times greater than that for the guinea pig (reviewed in Poland and Knutson, 1982: p 321). Also, octa-CDD is metabolized very slowly (Bimbaum and Couture, 1988), yet is very much less toxic than other dioxins and furans with chlorines in the 23.7,8 positions. W hat a n the effects of dioxins and flirnns on human health? Our understanding of the effects of dioxins and furans on humans is not based on controlled studies such as were done with animals. Instead, it is based on studies of people who were inadvertently exposed to these substances, as a result of an accident, their occupation, or other circumstances. This presents two major difficulties: 1) we cannot control and often do not know the level o f exposure (the dose); 2) people have almost never been exposed to these substances alone; the dioxins and furans have been mixed with other substances (PCBs, PBBs, t chlorophenols, chloropherol-based herbicides, and so on). These other substances have been shown to cause some of the same health effects as dioxins and furans, but only when people have been exposed to much larger amounts. An important difference between the toxicity of dioxins and furans and these other substances is the comparatively small amounts of dioxins and furans that can cause serious health effects. In most cases, the level of exposure to the other substances has been too low to cause such effects. The health effects o f toxic substances can be divided into two major groups: 1) acute (short-torn) effects, which occur soon after exposure and quickly subside; 2) chronic (long-term) effects, which can occur anytime during the life of the person and can be penistenL Our knowledge o f the acute effects of dioxins and furans is based on studies o f people who were exposed to relatively large amounts, usually as a result o f an industrial accident. Our knowledge of the chronic effects of dioxins and furans is based on long-term studies of these same groups o f people, or studies of people exposal to relatively small amounts far a long period of time. With these long-term studies, it is more difficult to remove the confounding effects of exposure to organic chemicals otha- than dioxins. Dioxins and Furans 59 784160 GENP 011349 TTw short-term health effects onhum ans Dioxins and fuians have caused a host of reponed acute effects on people who were exposed to relatively high levels. The reponed clinical symptoms (directly observable or felt by the patient) include nausea and vomiting, iiritaboo of the eyes, skin, and gastrointestinal tract, and a general reeling of not being w ell Pain, especially in the limbs, has been reponed as persisting for months, as has an enlarged and tender liver, and irritability and nervousness. There is no documented case of a person dying because of exposure to dioxins and furans. In addition to the clinical symptoms, there are reponed effects detectable by laboratory tests or study. They include liver damage, elevated levels of constituents of the blood such as triglycerides and lipids, and damage to the nerve fibers. Although most people recover quickly from the acute effects pf dioxins and furans, there are repans that some of these effects have lasted for months or years. For reviews of the acute human health effects of dioxins and furans see USEPA (1985: pp 8.60-8,65); Hay (1982); Reggiani (1982). The long-term health effects on humans The most common chronic effect of exposure to dioxins and furans is severe and persistent acne, called chloracne. which can occur in many pans of the body. Excessive body hair and pigmentation, and elevated levels of blood constituents such as triglycerides, cholesterol, liver enzymes, and porphyrins have also been reported several yean after the high exposure. Nerve disorders have been reported, including problems with or k m of some vision, hearing, taste, and smell. Also, general weakness and a loss o f sexual drive has been reported to occur for some years after high exposure. The reported symptoms and effects of exposure to dioxins and furans have been inconsisa n t. with the exception of chloracne. However, in most cases an enlarged and/or damaged liver and neuromuscular symptoms have also been reported. For reviews of the chronic human health effects of duxins and furans, see USEPA (1985; pp 8.65-8.69); Hay (1982); Reggiani (1982). 60 Dioxins and Furans 011350 A few epidemiologic studies lave found u association between cancers and occupational exposure to dioxins and furans. The studies that are generally considered to offer the best evidence were earned out in Sweden. Abnormally high levels o f a rare type of cancer, called soft-tissue sarcoma, were found in men who were exposed to chlorophenols and chloiopbenol-based herbicides (reviewed in HardclJ. 1983). The Swedish studies have been criticized for unreliable diagnosis of soft-tissue sarcoma (Hajdii, 1984), and for relying on the workers' memory of their exposure to the substances (Cook, 1983). In general, the diagnosis and classification of this rare type of cancer is difficult, and relying on memory for an appraisal of a subject's health is an admitted weakness. Similar studies have since been conducted in several countries, and no such association has been found (reviewed in USEPA, 1983: 11.64-91). Several o f these studies are not \ considered sensitive enough to detect the small numbers of a cancer as rare as soft-tissue sarcoma (reviewed in USEPA, 1985: 11.64-91). After the Swedish results were published, a review o f the medical history o f men in the U.S. who were occupationally exposed to chlorophenol-based chemical products showed an abnormally high number (seven cases) of soft-tissue sarcomas (Honchar and Haiperin, 1981). The authors of the U.S. review subsequently helped to reexamine the data, and they determined that only two of the seven workers were both 1) exposed to dioxin* and furan-comaminated products and 2) correctly diagnosed as having soft-tissue sarcoma (Fingerhm et aL. 1984). Nevertheless, two cases still represents a significantly higher number of soft-tissue sarcomas for this group (USEPA, 1983: p 11.91). A milnr study in Sweden found no association between rates o f soft-tissue sarcoma and exposure to pheoaxy-acid herbicides, among 330,000 Swedish agricultural or forestry workers, when compared to 1.7 million Swedish workers in other industries (WikJund and Holm, 1986, cited in Fishbein, 1987). In their 1983 Health Assessment for dioxins, the USEPA stated that the Swedish soidies offer "limited" evidence for the carcinogenicity of dioxin-contaminated chlorophenols and Dioxins and Furans 61 t 784162 OENf o i 13si 'f chlorinated phcnoxyacctic herbicides (such as 2,4.5-T), but 'inadequate" evidence for the carcinogenicity of 2J.7.8-TCDD alone (USEPA. 1985: p 11.138-140). In addition to the association between exposure to dioxin- and furan-conlaminated products and soil-tissue sarcomas, a few researchers have reported finding such an association with nasal cancer (reviewed in Hardell, 1983), stomach cancer (reviewed in USEPA, 198S: pp 98-107), and cancer of circulating cells (reviewed in USEPA, 1985: pp 92-98; reviewed in Hardell, 1983). Similar studies of worken exposed to dioxin- and furan-contaminated products have not found elevated levels of cancers (reviewed in Reggiani, 1983; p 483). The USEPA Health Assessment considered the available evidence for these other cancers to be inadequate, and this is an opinion shared by a majority of researchers. The great majority of researchers in the field of toxicology or epidemiology feel that the numerous studies of the human carcinogenicity of dioxins and fuians have provided inadequate evidence that they cause cancer in humans. Based on the epidemiological evidence and the knowledge that some dioxins and fuians arc proven carcinogens in animals, dioxins and fuians are suspected human carcinogens. Abortions, binh defects A small number of researchers have found an association between exposure to 2,3.7,8TCDD-contaminated products and abncnnaily'high levels of abortions or binh defects (reviewed in USEPA. 1985: pp 9-23-36; reviewed in Hatch, 1984; Albanese, 1988). Children of mothers who consumed rice oil contaminated with PCBs and furm s (the Yusho incident) were bam with d a rt pigmentation and deformed fingernails (reviewed in Kuratsune, 1980: p 299). Some of these babies also suffered from slowed growth, and chloncne. It has generally been concluded that farm s were responsible for these effects (Kunita.es al, 1984). Other researchers conducting similar studies have found no such association (reviewed in USEPA, 1985: pp 9223-36; reviewed tn Hatch, 1984; Mastroiacovo et aL, 1988). Many o f these studies have been largely discounted because of weaknesses in the methodology (USEPA, 1985: pp 9.23-36). There is thus Car inconclusive evidence of an associating between dioxins and fuians and abortions, and an association with binh defects was detected only in the rare rircumstance 62 Dioxins and Fuians GENP 011352 784163 ) 1 where the mother was exposed to extremely Urge amounts while pregnant (the Yusho incident). IT these substances are capable of causing such effects oa humans, then either the level of exposure must be extremely high, or they cause rare types of birth defects that researchers have not been able to detect in significant numbers. Again, this is an area where further investigation is needed. ImmiBW-fifftcu A few researchers have reported an association between long- term exposure to 2J.7.8TCDD and an impaired immune system (reviewed in Hay, .1982). Other long-term studies have found no effects on the immune system (Albanese, 1988; F a n et aU 1982). A study of people exposed to 2,3,7,8-TCDD in Missouri found evidence of an abnormal cellular immune system (Stehr-Green et a!.. 1987). However, the effects did not result in increased illness. Also, the authors noted that these effects may be part of a normal response to a toxin in the body, i.e., they could be considered indicators of exposure and not signs of disease. Conclusions on health effects The long-term health effects of these compounds have been intensely investigated, making them among the most-studied substances with respect to their' effect on animal and human health. Dioxins and furans cause mortality andfor cancer in many animals, but there is no conclusive evidence (hat they cause cancer or any other life-threatening health problem in humans. No documented human death has occurred because o f s i exposure to dioxins or furans. People have been exposed to relatively large amounts of dioxins and (mans (1 ug or more) during industrial accidents, and other incidents of accidental contamination (Yusho, Times Beach). The symptoms of these high-level exposures include dizziness, headaches, liver damage, leg pains, and reduced sex drive. These symptoms, (hough serious, have typically subsided with time. An exposure to these amounts can also cause severe acne, called chloracne, which in some cases has persisted for decades. With the exception o f chloracne, most Dioxins and Furans 63 784164 GENPO11353 1 researchers have found no serious, long-term health problems associated with exposure to dioxins, even at the highest reported levels: However, some studies have found an association between long-term diseases and exposing to dioxins and (titans. Several Swedish studies have shown an association between a person's exposure to 2J,7,8-TCDD-comaiiung herbicides and the incidence of a rare form of cancer. Other and subsequent studies have not confirmed this association, and the Swedish studies have been criticized on several grounds. Nevertheless, because they are very potent toxins and carcinogens in laboratory animals, the U.S. Environmental Protection Agency. .Centers for Disease Control, Food and Drug Administration, and other groups, have issued guidelines for human exposure to dioxins and furans, based on the results of animal studies in the laboratory (see "How do government agencies arrive at acceptable human exposure to dioxins and furans?", page 65. \ How do dioxins and furans cause the health effects that they do? We don't know yet why dioxins and furans cause the health effects that they do. But because these substances affect so many systems of the body, it is expected that the mechanism of toxicity involves metabolism at the cellular level. For example, it has been proposed that many of the toxic effects of dioxins and furans are similar to vitamin A deficiency (Hakansson et al, 1988). Some of the toxic effects of dioxins and furans (and other halogemted aryl hydrocarbons such as FCBs and PBBs) have been linked with the cellular machinery that makes and releases certain enzymes (Poland et iL , 1979). Also, 23,7,8-TCDD has been linked with changes in the outer membrane o f the c d l (reviewed in Maoumum, 1985). It is not likely, however, that these changes in the c d l are directly responsible far the toxic effects. Instead, it appearc that the toxic effects are somehow mediated by these enzyme systems and the c d l membrane: 64 Dioxins and Furans G E N P 011354 784165 RqmUtintDfades and Furan How do government agencies arrive at acceptable hamaa capusurt to dlorina and furaiu? Government agencies from many different countries have determined maximum acceptable doses of 2J.7.8-TCDD for humans, which are then used to regulate practices that expose people to dioxins and ftuans. These acceptable doses (also called tolerable, allowable, and virtually safe) vary from country to country, and can vary from agency to agency within a country (see Figure 4, following page). This variation is primarily the result of differing interpretation of the same experimental animal data, and to a lesser extent the choice of different data. In order to understand how this comes about, a brief summary of how an agency determines an acceptable dose is in order. \ First, suitable data must be chosen. The results of animal laboratory studies are relied upon most heavily and are often the only data considered. This is because human studies (epidemiological studies) are after-the-fact investigations of uncontrolled exposure, making it extremely difficult to confidently conelate a given dose of the substance with the observed health effects. Animal laboratory studies allow several controlled doses, including lethal doses, to be used, and allow the researchers to examine the health effects over most or all of the lifetime of the animal. Dioxins and Furans 65 784166 GENP 011355 9ItOdMS[Q So > 6' o M Ps s o> N The effects of different doses on long-term health aspeas such as cancer, birth and changes in the immune system are typically used by regulators, they have very serious consequences. In the United States and C a n d i, far example, date from long-term studies of cancer in rats have been used to estimate an recqxable dose of 13.7,8-TCDD. Rats were chosen as subjects because they are physiologically sm il to humans, reliable strains of rats have been developed for use in such studies, and large numbers of the animals can be n w | for the experiment Regulators must then extrapolate an acceptable dose for humans bom the chosen data. It is this extrapolation that accounts for most of the variation in acceptable doses. The extrapolation can be done using two major methods; 1) applying safety factors to the largest dose that could be correlated with no observed advene effect (the NOAEL); 2) using the various doses and their correlated adverse effects to extrapolate a dare that would be theoretically associated with an acceptably small risk of causing the effect. For example, the USEPA considers a risk of one tum o' in one million people during an average lifetime of 70 years to be an acceptably small risk. A safety factor (method 1) is used when the regularon decide that there is a threshold dose below which there is only a remote risk that the health effect will occur. The safety factor is then applied to allow for individuals that are more sensitive to the toxic effects than the average, and to allow for the possible difference between the test species and humans. Safety faetn of bom 10 to 1000 have been applied, depending on the confidence the regulato n have in the available dam. The dose extrapolation (method 2) is used when the regularon decide that any dose, no m ater bow small, increases the risk of the health effect. Recamo the an of this method is u quantify the risk and then choose an acceptable level o f risk, such an approach is called quantitative risk assessment There are many different mathematical models that can be used to extrapolate quantitative risks bom a range of dares, sod the extrapolated degrees of risk can vary considerably, depending on the model chosen. Dioxins and Furans 67 L 784168 GENP 011357 j""y I The choices outlined above account for the range of acceptable daily lifetime doses of 2J,73*TCDD as presented in Figure 4. The lowest (USEPA) and highest (Ontario Ministry of the Environment) doses in that table w o e derived from the same long-term studies o f rats by Kocibe et al (1978). The USEPA also used a similar study by the National Cancer Institute (NTP. 1980a); the Ontario Ministry of the Environment also used a similar study by Murray et aL. (1979). Because the USEPA feels that there is no threshold below which 23,7,8-TCDD does not cause an increased risk of cancer, they used a dose-extrapolation model to calculate the acceptable daily dose. The model they used to arrive at the present virtually safe dose is the most conservative of the five models they tested; it extrapolates a greater risk than do the other models at low doses. The Ontario regulators recognize 2,3,7,8-TCDD as a promoting rather than an inidating carcinogen, and therefore fee] that there is a threshold doss below which 2J.73-T C D D probably does not increase the risk of cancer. They applied a safety factor of 100 to the dose that they fell represented the NOAEL in these studies, to calculate an acceptable daily Thus, we have two very different interpretations of essentially the same da- The USEPA's recently proposed virtually safe dose is greater because it reflects the median virtually safe dose bom the different models, rather than the most conservative. As has been pointed out by Lave (1983), when (he mechanism of action o f a toxin is unknown, ail applicable models are equally valid and should be used to derive a range of risk No single model should be chosen, because undue emphasis is then placed on one estimate in what is a large range of equally relevant estimates. At the same time, regulaiora should make conservative assumptions. Then, as their knowledge abotx the toxin increases and they replace their conservative assumptions with mare accurate ones, the estimated risks should decrease. This provides an incentive for industry to conduct research on toxic substances: Because of lack of information about dioxins and furans, the USEPA and other regulatory agencies have made very conservative assumptions when estimating the health risks of these compounds. Many researchers have argued that the assumptions are overly 68 Dioxins and Furans GENP 011358 784169 r' !! V- ` conservarive, and these arguments have grown more persuasive as our knowledge of dioxins and furans has increased. For example, Paustcnbach et al (1986) critically **min the assumptions used by the CDC and USEPA to determine the risk posed by 23,73-TCDD Q joil. Their analysis indicates that the CDC and USEPA significantly overestimated the risk of 23,7,8TCDD in soiL Also, although the CDC made it clear that their risk assessment was 1) specific to the residential area in Times Beach, Missouri and 2) not applicable for industrial sites, the USEPA subsequently adapted their guideline of 1 ppb 23,7,8-TCDD in soil to determine the need for cleanup at industrial sites, where the level of exposure would be much lower. W hat has been done about dioxins and furans? The manufacture and use of 2,43-T and the relaxed herbicide Silvex in the U.S. was restricted in 1970, and prohibited in 1984. The manufacture and use of chlorophenols has been greatly reduced. No chlorophenols with four or fewer chlorines are being produced in the U.S. (pen. comm., John Robinson, Vulcan Chemicals), and hexachlorophene is not being produced because of the lack of 2,4,5-irichlonjphenol (USEPA, 1986: p 3-16). The USEPA requires that pentachtorophenol produced in the U.S. must now average no more than 2 ppm hexa-CDDs, and cannot have more than 4 ppm hexa-CDDs (Vulcan Chemicals, 1988). Approximately half of the hexa-CDDs present are 23,7,8-substituted (pen. comm., John Wilkinson, Vulcan Chemicals). The USEPA intends to cancel most non-wood uses of pentachlarophenol (USEPA, 1987b: p. 21413). Wastes from the manufacture of tri-, tetra-, and pemachlorophenols, and from the manufacture of tetra-, pent-, and hexachlorobenzene under alkaline conditions, were designated as acutely hazardous by the USEPA in 1934. The regulations also apply to discarded, unused chlorophenols or products containing them, to products made with equipment previously used to manufacture the substances above (except pentachlorophenols), and to soil that has been contaminated by these substances. The manufacturers or owners of such wastes are required to notify the USEPA. These wastes cannot be placed in a landfill if they contain more than 1 ppb Dioxins and Furans 69 784170 GENP 011359 of certab hTM and funms congeners (USEPA, 1986; p 2 5 ), rod if inrincracd . 99.9999% of the contamination must be removed (USHPA, 1986.*p 2 5 ). The USEPA has proposed that wastewaters sod tip p in g s, firm wood-preserving fariiiiifti and sawmills that use chicropheaolic solutiats, or that use eqaiproem formerly used far such solutions, be regulated as hazardous wastes (USEPA, 1988a), Also. 13.7.B-TCDD is now listed under the Clean Water Act as a compound which the USEPA must control in industrial effluents (pen. comm.. USEPA). PCBs are no longer produced in the U S., and any new use of them is strictly controlled and must be extremely well justified. A gradual phase-* of PCB-containing equipment is taking place. Tiansformen that contain fluid with more than 500 ppm PCBs are now prohibited where they can contaminate food or feed, must be registered and inspected, and leaks must be repaired within 48 hours. Capacitors with more than 500 ppm PCBs are also prohibited from areas where they might conLaminate food or feed, and large caproiton had to be placed in enclosures by November 1988. The disposal of PCBs is also highly restricted Fluid contaminated with more than 500 ppm PCBs must be incinerated with a removal efficiency of 99.9999%, and fluid contaminated with more than 50 ppm cannot be placed in a landfill. U is a violation to store PCBs for more than one year before disposal (pets, comm., USEPA). The USEPA has published lists o f chemical products that may be contaminated with chlorinated or brominated dioxins and funns (USEPA, 1987b). Mamdac&nm or importers of the chemical products on .there lists must test for and report id iho USEPA oo the presence of dioxins and funns in these products. The USEPA also published a list of precursor products, which a n not contaminated themselves bat may produce dioxins or funms if they are used to manufacture other products: Manufacturers or importers o f products made with these precursors must notify USEPA of that b e t (USEPA. 1987b). The USEPA conducted a nationwide savoy in order to determine the extent of contamination by 23,73-TCDD . The most severely contaminated shea were assigned to the Sigterfund List (most were already an the l i s t o f the presence of other toxic substances). 70 Dioxins and Funns GEN? 011360 W ho the USEPA found relatively high levels of 2J.7.8-TCDD in fish, a d v iu k s were issued to limit coraumptioo. Also, a progran was pux m place for cleaning up the many contaminated git* in Missouri, and many of the sites have since been cleaned up. n ifliiM and Furans 71 784172 GENP 011361 Conclusions and Rccoram ends tk ) We need to know mote about the unicity, distribution, and behavior of chlorinated dioxins and furans other than 2J,7,S-TCDD, especially those that tie found in relatively large amounts in the environment. For instance, current research indicates that the common oca-CDD is biologically active, not non-toxic as has been believed (Couture et al, 1988). Based on the many studies already published, dioxins and furans do not appear to be a hazard to our health when we are exposed to the law levels typically present in the environment. The epidemiological evidence indicates that dioxins and furans do not cause cancer, immune effects, or other serious, long-term health effects on humans, even at very high doses. Some groups of people, such as workers who are exposed occupationally to greats' amounts o v s a long lime, and breast-fed babies, need special attention. We should continue to monitor the health of those persons who have been exposed to laige amounts of these compounds, or who are exposed to sm a lls amounts far a long time. This will enable us to identify any long-term health effects not already recognized. We also need to detsm ine the mechanism of action. Such an underseroding would not only have preventive or therapeutic value, but is also critical for accurate assessments of the risks these compounds pose. Because of their demonstrated toxicity to laboratory animals, plus their distribution and persistence, some dioxins and furans are a potential hazani to organisms in the environment. There is a great need for mare study in this area. We need to know w h s are unacceptably harmful levels for the various organisms in the environment, and the nature and amount of dioxins and furans from the various sources. If we are to reduce the amounts o f toxic and furans in our environment, we need to deal both with the large am ow ts already created, and the processes that continue to create them. What appears to be the largest amount of toxic dioxins re d furres is buried in landfills or stored at farmer manufacturing sites; We need to develop safe yet practical ways to destroy this waste or ensure that it is contained. We may also need to reduce the amounts of dioxins and furans occurring as unwanted contaminants of same products. Greet reductions 72 Dioxins and Furans GENP 011362 784173 \ T have been made in certain products in the past, m i further reduction may be necessary. The oeeds to be on preventing their creation durinf the manufacturing process. Finally, wc need to reduce the unwanted creation of dioxins and A n n a during combtadoo. For example, ingin m inra can be operated under conditions (high temperatures, complete bunting) that reduce the creation o f dioxins ind fursns (USEPA, 1988b). Dioxins and Furans 73 784174 GENP 011363 Literatur Cited Abbmzzi, R.; Belvedere, G.: Bianchi, R., et al. 1977. 'Identification and Quantitative Determination of 2J,73-Kirachlofodiberizo-para-<liaxin in Animals from the Contaminated Areas". Fourth International Symposium on .Mass Spectrometry in Biochemistry and Medicine. Riva del Gorda. Italy. Adams, R.; Thompson, M.: Strother, Da James, R.; Miller, H. 1986. "Determination of PCDDs and PCDFs in PCB Oil from a Hazardous Waste Site". Chcmospherg. 13:1113-1121. Adams, W,; DeGraeve, G.; Sabourin, T.; Cooney, J.; Mosher, G. 1986. "Toxicity and Bioconcentration of 2,3,7,8-TCDD to Fathead Minnows fPimcohales promelas)". QtemosPbcre 15:1503-1511. Albanese, R. 1988. 'United Slates Air Force Personnel and Exposure to Herbicide OrangeInterim Report for Period March 1984- February 1988". United' Stales Air Farce School of Aerospace Medicine, Brooks Air Force Base, Texas. USAFSAM-TR-88-3. Amendola, G.; Bama, D.; Blosser, R.; LaFleur, L ; McBride, A.; Thomas, F.; Tieman, T.; Whiuemore, R. 1987. "The Occurrence and Fate of PCDDs and PCDFs in Five Bleached Kraft Pulp and Paper Mills". Presented at the Seventh International Symposium on Chlorinated Dioxins and Related Compounds, October 1987, Las Vegas, Nevada. t Barnes, D.; Beilin, J.; Geveriy, D. 1986. "Interim Procedures for Estimating Risks Associated with Exposures to Mixtures of Chlorinated Dibenzodioxins and -Dibenzofurans (CDDs and CDFs)". Cticmosohere 15:1895-1903. Baughman, R.; Meselsan, M.; 1973. "An Analytical Method far Detecting TCDD (Dioxin): Levels a f TCDD in Samples from Vietnam". Environmental Health Perspective 5:27-35. Baughman, R. 1974. Thesis. 'TetractUorodibenzo-p-dioxins in the Environment: High-resolution Mass Spectrometry at the Picogram Level". Harvard University, Cambridge, MA. Beck, H.; Eckart, K.; Mathar, W.; Wittkowski, R. 1988. "Occurrence of PCDD and PCDF in Different Kinds of Paper". Chcmosotierc 17:51-57. Bianco, W.; M easa. L.; Remoui. G. 1986. "Monitoraggio Ostetrico nella Brianza di Seveso dal 1975 al 1981". Doctimento dell* Uifickj Spcdaie di Seveso. Biinbaum, L ; Couture, L. 1988. "Disposition of Octatlorodibenzo-p><lioiun (OCDD) in Male Rate". Toxicology, and Applied Pharmacology 93:22-30. Bowes; G.; MulvihilL M.; Simoodt, B4 Burlingame, A^ Risebrough, R. 1975. "Identiflcation of Chlorinated Dibenzofurans in American Polychlorinated Biphenyls". Nature 256:305-307. Bumb, R^ Cnimm et W.; Cutie, S. 1980. "Trace Chemistries of Fire: A Source of Chlorinated Dioxins'. Science 210:385-390. Buser, H. 1986. "Polybrommated Dibenzofurans and Dibenzo-p- dioxins: Thermal Reaction Products of Poiybrominated Diphenyl Ether Flame Retardants". Environmental Science and Technology 20:404-408. Buser, H. 1987. "Brominaied and Bronunaled/Oilorinaied Dibenzodioxins and Dibenzofurans: Potential Environmental Contaminants". Chemosohete 16:1873-1876. 74 Dioxins and Furans 784175 1 Carnegie Mellon University. 1983. TemachloropbcnoL A Case Study in Pesticide Regulation'. Carnegie Mellon University, Department o f Engineering and Public Policy. Centers for Disease Consol (CDQ. 1987. 'Serum Dioxin in Vietnam-Eia Veteran* Preliminary Report". Morbidity and Mortality Weekly Report (MMWR136:470475. CDC. 1988. "Senun 2,3,7,8-Teanchlorodiberm>*p-dinxin Levels in Air Force Health Study Participants- Preliminary Report". Morbidity, and Monalitv Weekly Reoon fMMWRl 37:309- 311. Choudhary, G. 1983. "Occupational Exposure (0 Polychlorinated Dibenzodiaxins and Dibenzofurans: A Perspective". In Chlorinated Dioxins and Dibenzofurans in the Tnnl Environment. Choudhary. Keith. I_: Rappe, O , Eds. Buoerwonh Publishers, Woburn, MA. pp 335-353. Choudhary, van den Broeckc. J.; Hutzinger, O. 1983. 'Formation of Polychkxodibenzofunms (PCDFs) by the Photolysis of Polychlorobenzenea (PCBzs) in Aqueous Acetonitrile Containing Phenols". Chemosphcre 12:487492. Cement, Tosine. H.; Osborne, J,,- Ozvacic, V.; Wong, G.; Thcrndyke. S. 1987. 'Emissions of Chlorinated Organics from a Municipal Sewage Sludge Burning Incinerator". Chcmosnhere 16:1895-1900. Cochrane, W.; Singh, J.; Miles, W,; Wakefoid, B.; Scott, J, 1982. "Analysis of Technical and Formulated Products of 2.4-Di- ClUorophenoxy Acetic Acid for the Presence of Chlorinated Dibenzodioxins". In Chlorinated Dioxins and Related Compounds. Hutzinger, 0 .; Frei, R.; Meriam, E ; Pocchiari, F., Eds. Pergamon Press, Oxford, U.K. pp 209-213. Cook. R. 1983. "Soft Tissue Sarcomas: C u es and Caution". In Human and Environmental Risks of Oilorinated Dioxins and Related Compounds. Tucker, R.; Young, A.; Gray, A., Eds. Plenum Press, New York, pp 613-618. Coulston, F.; Olajos, E. 1980. "Panel Report: Panel to Discuss the Epidemiology of 2,4,5-T". Ecotoxicology and Environmental Safety 4:96-102. Courtney, iC; Gaylor, Du Hogan, M.; Falk, Evaluation o f 2.4.5-T." Science 168: 864-866. Bates, R-; Mitchell, L 1970. Teratogenic Couture, L ; Elwe11, Bimbaum, L. 1988. "Dioxin-like Effects Observed in Male Rats following Exposure to OCDD during a 13-Week Study." Toxicology and Applied Pharmacology 93:3146, Crosby, D.; Wong, A. 1977. 'Environmental Degradation of 2J,73-Tetrachlotodibenzo-p-<lioxin CTCDD)*. Science 195:1337-1338. Crosby, D^ Wong, K i Plimmer, J.; Woolson, E 1971 "Photodcconiposiiioa o f Chlorinated dibenzo-p-dioxins" Science 173:748. Crosby, Moilanen. K.: Wong, A. 1973. "Environmental Generation and Degradation of Dibenzodioxins and Diberuofurans". Environmental Health Perspective 5:259-266. Czoczwa, J.; Hites. R. 1985. "Dioxins and Dibenzofurans in Air, Soil, and Water*. In Dioxins in the Environment. Kamrin, Rodgers, P., Eds. Buoerworth Publishers, New York, pp 85-99. Czuczwa, J.; Hites, R. 1986. "Airborne Dioxins and Dibenzofurans: Sources and Fates". Environmental Science and Technology 20:195-200. Dioxins and Furans 75 784176 Czuczwa, J.; McVeety, B.; Hites, R. 1984. 'Polychlorinated Dibenzo-p-dioxins and Dibenzofurans in Sediments from Siskiwit Lake, b le Royale". Science 226:568-569. Davies, K. 1988. 'Concentration and Dietary Intake of Selected Orgmoehlarines, including PCBs, PCDDs, and PCDFs*. Chemorohcre 17:263-276. des Rosters, P. 1987, 'Chlorinated Combustion Products from Fires Involving PCB Transformen and Capacitors". Chemosohcra 16:1881-1888. DiDomenico. A.; Silano, V.; Viviano. G., Zapponi, G. 1980a. 'Accidental Release of 23.7,8Tctrachlorodibenzo-p-dioxin at Seveso, Italy (II): TCDD Distribution in the Soil Surface Layer. EcMoxicology and Environmental Safety 4:298-320. DiDomenico, A.; Silano, V.; Viviano, G,, Zapponi, G. 1980b. "Accidental Release of 23.7,8Tetrachlorodibenzo-p-dioxin at Seveso. Italy (IV): Vertical Distribution of TCDD in SoU*. Ecoioxicologv and Environmental Safety 4:327-338. Erickson, M.; Cole, C.; Flora, J.; Gorman, P.: Haile, Hinshaw, G.; Hopkins, F.: Swanson, S. 1984. 'Thermal Degradation Products from Dielectric Fluids". Office of Toxic Substances, USEPA, Washington. D.C. EPA 560/5-84-009. Esposito, M JL; Tieman, T.O.; Dryden, F.E. 1980. "Dioxins". USEPA Environmental Protection Technology Series. Document EPA-600/2-80-197. Faccbeui, Balasso, A.; Fichtncr, C.; Frare, G.; Leoni, A.: Mauri, C4 Vasconi, M. 1986. "Studies on the Absorption of TCDD by Plant Species". In Clorinated Dioxins and Dibenzofurans in Perspective. Rappe, C.: Choudhaiy, C.; Keith, L , Eds, Lewis Publishers, Chelsea, ML pp 225-235. Fairiess, B4 Bates, D.; Hudson. J.: Kloepfer, R4 Holloway, T.; Morey, D. 1987. `Procedures Used to Measure the Amount of 23,7,8-TCDD in the Ambient Air near a Superfund Site Oeanup Operation". Environ. Sci. Tech. 21:550-555. Faneili. R.; CasttUi. M.; Manelli, G.; Noseda, A.: Garatinni, S. 1980. "Presence of 23,7,8Tetrachlodibenzo-p-dioxin in Wildlife Living Near Seveso, Italy: A Preliminary Study". Bulletin oLEnvironmentai Contamination and Toxicology 24:460462. F a n , G.; Del Como, G^ Boneoi, F.; Caramasdti, Daidanoni, L ; Favaretti, C4 Giambelluca, Sm Mami, E ^ M ocarclii, P^ Montesarchio, Ea Puccinelli, V4 V dpata. C. 1982. "Chloracne after Release o f TCDD at Seveso, Italy", In Chlorinated Dioxins and Related Compounds. Hutzinger, O4 Frei, R4 Meriam, E4 Pocdiiari, F., Eds. Pergamoa Press, Oxford, UJC. pp 545-559. Fingertiut, M^ Halperin, W.; Honchar, P.; Smith, A^ Groth, D, 1984. 'Review of Exposure and Pathology Data for Seven Cases Reported as Soft Tissue Sarcoma among Persons Occupationally Exposed to Dtoxin-Coniamiiuied Herbicides'. In Public Health Risks of the Dioxins. Lowrance, W., Editor. William Kaufman, Loo Altos, California, pp 187-203. Firestone, D. 1973. 'Etiology of Chick Edema Disease*. Environmental Health Perspective pp Ftshbein, L. 1987. 'Health-Risk Fmimates for 23,73-Teirachioro-Dibenzaliaxm: An Overview". Toxicology and Industrial Health * 9 1 -I'M. * Gasiewicz. T4 Holscher, Neal, R. 1980. "The Effect of Total Parenteral Nutrition on the Toxicity of 23,73-T)rtrachloitxiibatzo-p-dioxin in the Rat". Toxicology and Armlied Pharmacology 54:469-88. 76 Dioxins and F in n s GENP 011366 784177 1 Gaewicz, T.; Olaoo, J4 Geiger, L4 Neal R. 1983. 'Absorption, Distribution, and Metabolism of 2J.73-Teoachkjrodibefux>p-dioxin in Experimental Animals'. ln Human and EnvironmA^n| Risks of Chlorinated Dioxins and Related Compounds. Tucker, fU Young, A4 Gray, A , Eds. Plenum Press, New York, pp 495-523. Gilbertson, M.: Fax, G. 1983. 'Chick Edema Disease and Hepatic Porphyria in Lah Ontario H aring Gull Embryos in the Early 1970s". In Human and Environmental Risks of Chlnrinaioil Dioxins and. Related Compounds. Tucker, R4 Young, A4 Gray, A , Eds. Plenum Press, New York, pp 341-356. Gough, Michael. 1986. Dioxin. Agent Orange. Plenum Press, New York. Gross, M. 1980. Testimony in re: The Dow Chemical Company, e t al., FIFRA Docket No. 415, eL aL, USEPA Exhibit No. 223, pp. 27-29. USEPA Hearing Clerk's Office, 401 N. Street, S.W., Washington, D.C., 20460. Gupta. B.; Vos, J.; Moore, J.; Zinkl, J4 Bullock. B. 1973. "Pathologic Effects of 23,7.8Tetrachlorodibenzo-p-dioxin in Laboratory Animals'. Environmental Health Perspective 5:125140. Hagerunaier, H.; Brunner, H.; Haag, R.; Berchtold. A 1986. ' PCDDs and PCDFs in Sewage Sludge, River, and Lake Sediments from Southwest Germany". Chemosnhere. 15:1421-1428. Hagerunaier, H,; Brunner, H. 1987, 'Isomerspecific Analysis of Peniachkuophenol and Sodium Pentachloiophenate for 23,7,8- substituted PCDD and PCDF at Sub-ppb Levels'. Giernosphere. 16:1759-1764. Haglund, P.; Egeback, K.; Jansson. B. 1988. 'Analysis of Polybrocninated Dioxins and Furan-- m Vehicle Exhaust*. Chemosphere 2129-2140. Hakansson, H.; Johansson. L.; Ahlborg, U. 1988. "Effects o f 23.7,8-teirachIorodibenzo-p-dioxin on Tissue Levels of Vitamin A and on the Distribution and Excretion of the Endogenous Pool of Vitamin A in the Marginally Vitamin A Sufficient Rat". Chemosphere 17:1781-1793. Hajdu, S. 1984. 'Classification and Pathological Diagnosis of Soft -Tissue Sarcomas*. In Public Health Risks of the Dioxins. Lowrance, W., Editor. William Kaufman. Los Altos, California, pp 173-186. Hallet, D.; Brooksbank, M. 1986. T rends of TCDD and Related Compounds in the Great L akes Hie Lake Ontario Ecosystem*. Chemosphere 15:1405-1416. HardeU, L. 1983. "Epidemiological Studies on Soft-Tissue Sarcoma. Malignant Lymphoma. Nasal, and Nasopharyngeal Cancer, and their Relation to Phenoxy A dd or Chlorophenol Exposure'. In Chlorinated Dioxins and Dibenzofurana in the Total Environment. Choudhary, G ; Keith, L4 Rappe, C,, Eds. Buoerwonh Publishers, Woburn, M A pp 367-374, Harte, RX.; Lewis, R.G4 Dupuy, A E 4 McDaniel, D.D. 1983. "Analyses for 23.7,8Teirachlorodibenzo-p-dioxin Residues in Environmental Samples*. In Human and Environmental R id nf Chlorinated Dioxins and Related Cornpounds. Tucker, R4 Young, A4 Gray, A , Eds. Plenum Press, New York, pp 161-171. Hatch, M. 1984. "Reproductive Effects o f the Dioxins*. In Public health. Risks of the Dioxins. Lowrance, W,, Editor. William Kaufman, Los Altos, California, pp 255-274. Dioxins and Furans 77 784778 Hay, A. 1982. "Exposure (o TCDD: the Health Risks*. In Chlorinated Dioxins and ttelatw^ Compounds. Hutzinger, O4 Frei R4 Meriam, E ; Pocchiari, F,, Eds. Pergamon Press, Oxford, U X pp 589-600. Hay, A. 1984. "The Mutagenic Properties of 2J.73-TCDD*. In Chlorinated Dioxins and Dibcnzofitrans in the Total Environment. Keith, L4 Rappe, C Choudhary, G,, Eds. Buoerwonh Publishers, Woburn, MA. pp 297-307. Hay, D.: Fuikelsteio, A.; Klicius, R4 Bridle, T. 1987. T h e National Incinerator Testing and Evaluations Program: Air Pollution Control Technology*. Chemorohcre 16:1923-1923. Heida, H.; Otie, K.; Prins, E 1986. "Selective Accumulation of Chlorobenzenes, Polychlorinated Dibcnzofurans and 2,3,7,8-TCDD in Wildlife of the Volgermeerpolder, Amsterdam, Holland". Chemosphere 15:1995-2000. . Heindl, A4 Hutzinger, O. 1986. "Search for Industrial Sources of PCDD/PCDF (I): Approaches In the Federal Republic of Germany". Chemosphere 15:2001-2003. Helder, T. 1982. "Effects of 23.7,8-TCDD on Early Life Stages of Two Fresh-water Fish Species". In Chlorinated Dioxins and Related Compounds. Hutzinger, O.; Frei, R4 Meriam, E ; Pocchiari, F., Eds. Pergamon Press, Oxford, U X pp 455-463. Hornberger, E ; Reggiani, E4 Sambeth, J.: Wipf. H. 1983. "The Seveso Acddenc Its Nature, Extent, and Consequences*. Annals of Occupational Hygiene. 22:327-370. Honchar, P4 Halperin, W. 1981. "2,4^-Trichiorophmol and Soft Tissue Sarcoma". Lancet 1:268-269. Houk. V. 1986. `Uncertainties in Dioxin Risk Assessment*. Chemorohae 15:1875-1881. Huff, J.; Wassom, J. 1973. "Chlorinated Dibenzodioxina and Dibenzofunns: An Annotated Literantre Collection'. Environmental Health Perspective 5:283-312. Hutzinger, O.; Berg, M.; lte, K.; Opperhuizen. A.; Safe, S. 1985. "Dioxins and Furans in the Environment". In Dioxins in the Environment. Kamrin, M4 Rodgers, P., Eds. Butterwodh Publishers, New York, pp 9-32. Isensee, A4 Jones, G. 1975. "Distribution of 2^,7.8-Tetrachlorodibcnzt>-p-<iioxin in an Aquatic Ecosystem*. Environmental Science and Technology 9:667. Kamrin, M4 Matsumar*. F. 1985. "Human Health and Toxidty- Workshop Summary*. In Dioxins in the Environment. Kamrin, M4 Rodgers, P,, Eds. Buttenrarth Publishers, New York, pp 286-287. Karesek, F.; Hutzinger, O. 1986. D ioxin Danger 6cm Garbage Incineration'. Analytical Chemistry 58:633a-642a. Kenaga, E4 Norris, L. 1983.' "Environmental Toxicity o f TCDD". In Human and Environmental Risks of Chlorinated Dioxins and Related Compounds. Tucker, R4 Young, A4 Gray, A , Eds. Plencm Press, New York, pp 277-299. Kimble, B4 Gross, M. 1980. T C D D in Stack-Collected C o d Fly Ash". Science 207:59-61. Kissel. J 4 Robarge, G. 1988. 'Assessing the Elimination o f 2J.7.8-TCDD from Humans with a Physiologically Based Pharmacokinetic Model". Chemosphere 17:2017-2027. 78 Dioxins and Furans GEISfP 011368 784179 T Koctta. R.; Keyes, Da Beyer, Ia C anton, R.; Wade, C ; Dittenber, D.; Kalnins, I t ; Frauson, L ; Parka, Cm Barnaul. S.; Hummel, Ra Humistoa, C 1978. "Result! of a Two-Ye* Chronic Toacfcy nd Oncogenicity Study of TCDD in Rats*. Toxic. App. Pharmacol 46:279-303. Komi, Ra Rude, T.; Joglckar, R.. Dansetie, P.; Jerina. Dm Adas. S.; Owens, L; Nebert, D. 1978. 2J,73-Teffachlorodibcn2i>p-d)oxin as a Co-Carcinogen Causing 3-Methylchokuuhreije-initiaicd Subcutaneous Tumors in Mice Generally "Non-Responsive* at Ah Locus'. Cancer R w a rrh 38:2777-2783. Kuchl, D.; Cook, P,; Banerman, A. 1986. 'Uptake and Depuration Studies of PCDDs and PCDFs in Freshwater Fish" Chemosohere 15:2023-2026. Kuchl, D.; Cook, P.; Banerman, A.; Butterworth, B. 1987a. 'Isomer Dependent Bioavailablility of Polychlorinated Dibenzo-p-Dioxins and Dibenzofurans*. Chemospheic 16:657-666. Kuchl, D.; Cook, P.; Banerman, A.; Lothenbach, D.; Butterworth, B, 1987b. 'Bioavailability of Polychlorinated Dibenzo-p-dioxins and Dibenzofurans bom Contaminated Wisconsin River Sediment to Carp". Chcmosoherq 16:667-679. Kunita, Na Kashimoto, T.; Miyata, H.; Fukushuna. S.; Hori, S.; Obana, H. 1984. 'Causal Agents of Y usho\ American Journal of Industrial Medicine 5:45-58. Kuratsune, M. * 1980. 'Yusho'. In Haloyenated Biphenyls. Terohenvls.. Naphthalenes, Dibenzodioxins. and Related Products. Kimbrough, R., Ed. Elsevier/North Holland, New York, pp 287-302. Lave. L. 1983. 'Risk Assessment for Regulation of Dioxin (TCDD)'. In Human and Environmental Risks of Chlorinated Dioxins and Belated Compounds. Tucker, R.; Young. A.; Gray, A., Eds. Plenum Press, New York, pp 635-638. Lindstrom, G. 1988. 'Polydilorinamed Dibenzo-p-diaxins and Dibenzofurans: Analysis of Occurrence in Milk". University of Umea, Umea. Sweden. Marklund, S.; Kjeller, L.;. Hansson, M.; Tysklind, M.; Rappe, Cm Ryan, C.; Collazo, H.; Dougherty, R. 1986. 'Determination of PCDDs and PCDFs in Incinoation Samples and Pyrolytic Products' . In Clorinaied Dioxins and_Dibenzofurans in Perspective. Rappe, C.; Choudlmy, Cm Keith, Lm Eds. Lewis Publishers, Chelsea, MI. pp 79-92. Marklund, Sa Rappe, Ca Tysklind, M.; Egeback, K. 1987, 'Identification of Polychlorinated Dibenzofurans and Dioxins in Exhausts from Cara Run on Leaded Gasoline." Chemosohere 1639-36. M aple, L,: Biunck, R^ Throop, L. 1986. 'Water Solubility of 2J,7,8-tetracltiarodtbienxo-pdioxin*. In Environmental Science and Technology 20:180-182. Mastraucows, Pm Spagnolo, A.: Mami, Ea Meazza, L ; Benollini, IL; Segni, G. 1988. 'Birth Defects in the S e v e n Area after TCDD Contamination*. Journal o f the American Medical Association 159:1668-1671 Matsumura, F. 1985. T h e Mechanism of Action o f Dioxin*. In Dioxins in the Environment Kamrin, M.; Rodgers, P,, Eds. Butterworth Publishers, New York, pp 261-265. Matsumura, F.; Benezet H. 1973. "Studies on the Bioaccumulation and Microbial Degradation of 23,73-TetrachIorodibenzo^-dioxin". Environmental Health Perspective 5:523-528. Dioxins and Furans 79 784180 GENP 011369 Maahews, H.; Bimbaum, L. 1983. 'Factors Affecting the Disposition and Persistence of Halogenaied Furans and Dioxins*. In Human. and_Environmental Risks of Chlorinated Dioxins and Related Compounds. Tucker, R.; Young, Gray, A., Eds. Plenum Press, New York, pp 463-475. McConnell, E. 1980. "Acute and Chronic Toxicity, Carcinogenesis, Reproduction, Tcratogenedty. and Mutagenesis in Animals'. In Halogenaied Biphenyls. Terohcnvls. Naphthalenes.. PibemodioxiTa. and Related Products. Kimbrough, Ed. Elsevier/Nonh Holland, New York, pp 109-150. McConnell, E.; Lucier, G.: Rumbaugh. R.; Albro, P.; Harvan, D.; Hass, J.; Hanis, M. 1984. 'Dioxin in Soil: Bioavailability After ingestion by Rats and Guinea Pigs*. Science 223:10771079. Metcalfe, L. 1972. 'Proposed Source of Chick Edema Factor". Journal of the Association of Official.Analytical Chemists 55:542. Moore, J.: Harris, M.; Albro, P. 1976. T issue Distribution of Tetrachlorodibenzo-p-dioxin in Pregnant and Neonatal Rats'. Toxicology and Applied Pharmacology 37:146-147. Moore, J^ McConnell, Dalgard, D.; Hams, M. 1979. 'Comparative Toxicity of Three Halogenaied Dibenzofmans in Guinea Pigs, Mice; and Rhesus Monkeys'. Annai nf the New YoritAcadcmv of Science* 320:151-163. Mukeijee, D.; Stara. J,; Schaum, J. 1986. "Rationale for Assessment of Risk from Exposure to 2J.73-Tctrachlorodibenzo-p-dioxm*. Chemosohere 15:1895-1813. Murray, F.; Smith, F.; Nitschke, K.; Humiston, Kocitaa. R.; Schwetz. 1979. 'Three- Generation Reproduction Study of Rats Given 2^,7,8-reaacMarodiberao-p-dioxin (TCDD) in the Diet". Toxicology and Applied Pharmacology 50:241-252. Nakano, T.; Tsuji, M.; Okuno, T.; 1987. "Level of Chlorinated Organic Compounds in the Atmosphere". Chemosohere 16:1781-1786. Narang, R.; Swami, K. Thakor, A.: Eadon, G.; Vernoy, C 1988. "Thermally-Induced Formation of Polychlorinated Dibenzofurans from Arochtor 1254 Contaminated Silicone Oil and Teoachlorethylene". Chemosohere: 2151-2160. Nash. R^ Beall, M. 1980. "Distribution of SUvex, 2,4-D, and TCDD Applied to Turf in Chambers and Field Plots". Journal of Agriculture. Food, arid Chemistry 28:614-623. National Council o f the Paper Industry for Air and Stream Improvement (NCASI). 1987a. Technical Bulletin No. 534. "Assessmait of Potential Health Risks from Dermal Exposure to Dioxin in Paper Products". National Council of tho Paper Industry far Air and Stream Improvement, Inc. New York. NCASI. 1987b. Technical Bulletin No. 526. "Land Treatment Effects on Wildlife Populations in Red Pine Plantation*". National Council of the Paper Industry far Air and Stream Improvement, Inc. New York. National Toxicology Program (NTP). 1980a. "Bioussay of 2 J .7 -tetrachlorodibenzo-p-dioxin far Possible Carcinogenicity (Garage Study)". DHHS Publication Number (NTH) 82-1765. Carcinogenesis Testing Program, NCI, NIH, Bethesda, MD, and National Toxicology Program, Research Triangle Park, NC. 80 Dioxins and Furans L GENP 011370 784181 National Toxicology Program. 1980b. "Bloassay of 2J.7j8^eaacbkndibenzo>fHlioxin Tor Possible Carcinogenicity (Detmal Study)". DHHS Publication Number (NIH) 82-1757. Cardnogenesii Testing Program, NCI. NIH, Bethesda, MD, and Nxtkxul Toxicology Program, Research Triangle Park, NC. National Toxicology Program (NTP). 1980c. 'Bioassay of 1,23,6.7,8- and 1,23,7,8.9hexachlofodibenzo-p-dioxin for Possible Carcinogenicity (Gavage Study)". DHHS Publication Number (NIH) 82-1754. Carcinogenesis Testing Program, NCI, NIH, Bethesda, MD, and National Toxicology Program, Research Triangle Park, NC. National Toxicology Program (NTP). 1980d. "Bioassay of 1.23,6,7,8- and 133,7.8,9bexachlorodibenzo-p-dioxin for Possible Carcinogenicity (Dermal Study)". DHHS Publication Number (NIH) 82-1758. Carcinogenesis Testing Program. NCI, NIH, Bethesda, MD, and National Toxicology Program, Research Triangle Park, NC. National Toxicology Program (NTP). 1988. "Toxicology and Carcinogenesis Studies of Two Pentachlorophenol Technical-Grade Mixtures in B6C3F( Mice". DHHS Publication Number (NIH) 87-2804. National Toxicology Program. Research Triangle Park, N C Nestrick, T4 Lamparski. L 1983. 'Assessment of Chlorinated Dibenzop-dioxin Formation and Potential Emission to the Environment from Wood Combustion". Chemosohere 12:617-626. Norm, K. 1988. 'Changes in the Levels of Organochlorine Pesticides, Polychlorinated Biphenyls. Dibenzo-p-dioxins, and Dibenzofurans in Human Milk from Stockholm, 1972-1985*. Chemoroherc 17:39-49. Norstrom, R.; Hallet, D.; Simon, M.; Mulvihill, M. 1982. 'Analysis of Great Lakes Herring Gull Eggs for Tetrachlorodibenzo-pMlioxins". In Chlorinated Dioxins and. Related Compounds. Hutonger, O.; Frei, R.; Meriam, E ; Pocchiari, F., Eds. Pergamoa Press, Oxford, U.K. pp 173181. Nygren, M.: Rappe, C.; Lindstrom. G.; Hansson, M.; Bergqyist, P4 Marklund, S.; Domellof, L.; Hardell, L^ Olsson, M. 1986. "Identification of 23,7,8-substituied Polychlorinated Dioxins and Dibenzofurans in Environmental and Human Samples". In Clorinated Dioxins and Dibenzofurans in Perspective. Rappe, C ; Choudhary, C ; Keith, I . , Eds. Lewis Publishers, Chelsea. ML pp 17-34. Oefame, M.: Finst, P4 Kruger, C : Meemken, HU Groebei, W. 1988. 'Presence of Polychlorinated Dibcnzo-p-diaxins, Dibenzofurans, and Pesticides in Arctic Seal from Spitsbergen". Chemosnhera 17:1291-1300. Ogaki. Takayama, K.; Miyarn, H.; Kashimoto, T. 1987. "Levels of PCDDs and PCDFs and Human Tissues and Various Foodstuffs in Japan", Chanomhera 163047-2056. Olie, K_; Berg, M.; Hutzinger, O. 1983. fo rn u tio o and Fate of PCDD and PCDF torn Combustion Processes'. Chemorohere 12^27-636. Olson, J.: Gatiewicz, T 4 Geiger, I_; Neal, R. 1983. "The Metabolism of 2 3 .7 $ Teuachlorodibenzo-p-dioxin in Mammalian System ". In Accidental Exposure to the Dioxins. Coulston, F4 Pticchiari, F., Eds. Academic Press, New York, pp 81-99. Ono, M.; Kashima, Y.; Wakimoto, T4 Tatsukawa, R. 1987. "Daily Intake of PCDDs and PCDFs by Japanese through Fbod". Chemoroherc 16:1823-1828. Dioxins and Furans 81 1 Ontario Ministry of the Environment (OME). 1989. "Scientific Criteria Document for Standard Development Number 4*84. Polychlorinated Dibenzo-p-diaxins (PCDDs) and Polychlorinated Dibenzofurans (PCDFs)". Ontario Ministry o f the Environment Canada. Opperfiuizen, A.; Wagenaar. W.; van der Wielen, R ; van den Berg, M.; Olie, Kd Gobas. F. 1986. 'l l plaice and Elimination of PCDD/PCDF Congenera by Fish After Aqueous Exposure to a Fly- Ash Extract from a Municipal Incinerator". Chemorohere 19:2049-2053. Palausky, Jd Kapila, Sd Manahan, S. Yanders, A.; Maihotrm, R.; Clevenger, T. 1986. "Studies on Vapor Phase Transport and Role of Dispersion Medium on Mobility of 2J.7.8 TCDD in Soft". Chemosohete 15:1389-1396. Patterson. D.; Needham, L d Piride, J. 1988. "Correlation Between Serum and Adipose Tissue Levels of 2J.7,8-Tetia:hkxt>dibenzo-p-dioxin in SO Persons -from Missouri". Archives of Environmental Contamination and Toxicology. 17:139-143. Paustenbach, D.; Shu, H.; Murray, F. 1986. "A Critical Examination of Assumptions Used in Risk Assessments of Dioxin Contaminated Soil". Regulatory Toxicology and Pharmacology 6:284-307. P etent. Rd Potter, G ; Moore, R. 1984. "The Wasting Syndrome and Hormonal Alterations in 2J,7,8-Teirachlorodibenzo-p-diQxin Toxicity". In Public Health R id a of the Dioxins. Lowrance, Editor. William Kaufman, Los Altos, California, pp 313-349. Pirlcle, Jd Wolfe, W.; Patterson, D. a &L "Estimates o f the Half-Life of 2J.7.8-TCDD in Ranch Hand Veterans". Presented at Dioxin *87, Las Vegas, Nevada, October 4-9, 1987. Pitot. Hd Goldsworthy, T.; Campbell, Hd Poland, A. 1980. "Quantitative Evaluation of the Promotion by 23.7,8-Tetrachlarodibenzo-p-dioxin o f Hepatocarcinogenesis from Diethylnitrosamine". Cana* Rtw atrh 40:3616-3620. PJeuss, N.; Poiger, H.; Hohbach, Cd Schlatter, C. 1988a. 'Subchronic Toxicity of Some Chlorinated Dibenzofurans (PCDFs) and a Mixture of PCDFs and Chlorinated Dibenzodioxins (PCDFs) in Rats". Chemosphere 17:973-984. Pleou, Nd Poiger, Hd Hohbach. Cd Suter, Md Schlang, G 1988b. "Subehrooic Toxicity of 2J,4,73*Pemichkxodibenzofuraa (PeCDF) in Rats". Chemosphctg 17:1099-11101 Poiger, Hd Schlauer, G 1986. "Phaimacoldncdcs of 23,73-TCD D in Man". Q cm osohere 15:1489-1494. Poland, A.; Knutson, J. 1992. "2,3,7,8-Tetrocbianlibcnzodiaxin and Related Halogenatcd Arotnatie Hydrocarbons Examination o f the Mechanism of Toxicity". Annual Review of Pharmacology and Toxicology 22317-554. Poland, Ad Greenlee, W.; Kcnde, A. 1979. "Studies an the Mechanism o f Action of the Chlorinated Dibenzo-p-diaxtni and Related Compounds". Annals of the. New Yo* AnA-my nf Science. 1979214-230. Rappe; G 1984. "Analysis of polychlorinated dioxins and farm s'. Environmental Science and Technology 18:78a-90a. Rappe, C. 1987. "Global Distribution o f Polychlorinated Dioxins and Dibenzofurans". In Solving Waste Problem s Learning from Dioxins. J, Exner, Ed. The American Chemical Society Symposium S eres,-#338. pp 20-33. 82 Dioxins and Furans GENP 011372 784183 f Rappe, C4 Buser, H . 1 9 8 0 . "Chemical properties sod analytical methods'. In H a l o g e n a t e r i BinhenvlsJTerohcnvls. Naphthalenes. Dibcraodioxim . and i w . . Kimbrough, R,, Ed. EIsevier/Natth Holland, New York, pp 41-76. Rappe, C4 Kjeller, L.;. 1987. "PCDDi and PCDFs in Environmental Samples Air, Paniculaics. Sediments, and Sail". Chemosohere 16:1775*1780. Rappe, C.; Marfclund, S.; Buser, H.; Bqsshantt, H. 1978. ' Formation of Polychlorinated Dibenzo-p-dkwtins (PCDDs) and Dibenzofurans (PCDFs) by Burning or Heating Chlorophcnate'. Chcmosphea 3:269-281. Rappe, C.; Buser, H.: Kuroki, H.: Masuda, Y. 1979. 'Identificadon of Polychlorinated Dibenzofurans Retained in Paticms with Yusho". Chemosohere 4:259-266. Rappe, G ; Buser, H4 Stalling, D.; Smith, L-; Dougherty, G 1981. 'IdenUficaiion of Polychlorinated Dibenzofurans in Environmental Samples*. Nature 292:524-526. Rappe, C.; Nygren, M.; Buser, H.; Kauppinen, T. 1982. 'Occupational Exposure to Polychlorinated Dioxins and Dibenzofurans'. In Chlorinated Dioxins and RelatcdJComoounds. Huizingcr, O.; Frei, R4 Meriam, E4 Pocchiari. F., Eds. Pergamon Press, Oxford, U.K. pp 495- 513. ' Rappe, C.; Maridund, S.; Bergqvist. P.; Hansson, M. 1983. 'Polychlorinated Dibcnzo-p-dioxins. Dibenzofurans, and Other Polynuclear Aromatics Formed During Incineration and Polychlorinated Biphenyl Fires'. In Chlorinated Dioxins and Dibenzofurans in the Total Environment. Choudhary, G ; Keith, L.; Rappe, C , Eds. Bunerworth Publishers, Woburn, MA. pp 99*124. Rappe, C.; Martdund, S4 Kjeller, L4 Tysklind, M. 1986a. 'PCDDs and PCDFs in Emissions from Various Incinerators'. Chcmosphcre 15:1213*1217. Rappe, C.; Andersson, R.; Bergqvist, P.; Brobede, C.; Hansson, M.; Kjeller, L.; Lindstrom, G.: Marfclund, S.; Swanson, S.; Tysklind, M; Wiberg, K. 1987a. 'Overview on Environmental Fate of Chlorinated Dioxins and Dibenzofurans. Sources, Levels, and Isomeric Pattem in Various Matrices*. Chcmosphcre 16:1603-1618. Rappe, G ; Nygren, M4 Lindstrom. G4 Buser, H.; B lass', 0 .; Wuthrich, C. 1987b. "Polychlorinated Dibenzofiirans and Dibenzop-diaxins and o th s Chlorinated Contaminants in Cow Milk o n Various Locations in Sw itzslaad". Environmental Science and Technology 21:964-970. Rappe, C4 Andersson, R4 Bergqvist, P4 Brohede, G ; Hansson, M.; Kjeller, L ; Linstrom, G.; Marfclund, S.; N y g ra , M4 Swanson, S.; Tysklind, M.; Wiberg, K. 1987c. 'Sources and Relative Imponance of PCDD and PCDF Emissions', w " nagemBit and Research 5:225*237. Rappe, G ; Kjeller, L.; Brockman, P4 Hackhe, K, 1988. Identification and Quantification of PCDDs and PCDFs in Urban Air". Chemosohere 17:3*20. Reggiani, G. 1980. "Localized Contamination with TCDD- Scveso, Missouri, and Other Areas'. In Haloaenated Biohenvls. Terohenvls. Naphthalenes. Dibenzodioxins. and Related Products. Kimbrough, R-, Ed. Elsevier/Nonb Holland, New York, pp 303-371. Reggiani, G. 1982. Toxicology of TCDD and Related Compounds: Observations in Man*. In Chlorinated Dioxins and Related Compounds. Hutzinger. O4 Fret, R4 Meriam, E.; Pocchiari. F., Eds, Pergamon Press. Oxford, U.K. pp 463-493. Dioxins and Furans 83 1 l i i 784184 GENP 011373 Reggixni, G. 1983. "An Overview of the Health Effects of Halogenated Dioxins and Related Compounds- The Yusho and Taiwan Episodes". In Accidental Exposure to Dioxins. Coulstoo, F4 Pocchiari. F., Eds. Academic Press, NY. pp 39-67. R o o d , W4 Christmans, W.; Launer, A4 Mann. W4 Reichert, A4 Reiss, S, Schinz, V. 1987. Occurrence of PCDD and PCDF in Motor oils. Redefined oils, and Contaminated Soils". Chemosohere 16: 1847-1849. Ryan, J. 1986. " Variation of Dioxins and Furans in Human Tissues". Chcmosphere 15:15851593. Ryan, J. Panopto, L.; Lewis, D. 1988. "Bleaching of Pulp and Paper as a source of PCDDs and PCDFs in Certain Foods". Presented at the 8th International Symposium on chlorinated Dioxins and Related Compounds, in Umea, Sweden, August 1988. Safe, S. 1987. Determination of 2J.7.3-TCDD Toxic Equivalent Factors (TEFs): Support for the Use of the in vitro AHH Induction Study". Chemosohere 16:791-802. Sand, L.; Boeri, R.; Remotli, G. 1983. "Five Years After Seven", 1:343-344. Sawyer, T.; Bandiera, S.; Safe, S. 1983. 'Bioanalysis of Polychlorinated Dibenzofuran and Dibotzo-p-dioxin Mixunes in Fly Ash": Chemosohere 12:529-535. Schemer. A.; Gasiewicz, T. 1987. "Health Hazard Assessment of Chlorinated Dioxins and Dibenzofurans Contained in Human Milk". Chemosohere 16:2147-2154. Schemer, A.; Ryan, J. 1988. "Polychlorinated Dibenzo-para-dioxin and Dibenzofuran Levels in Human Adipose Tissues from Workers 32 Yean after Occupational Exposure to 2,3,7,8-TCDD". Chemosohere 17:915-920. Schecter. A.; Ryan, I.; Constable, .J. 1986. "Chlorinated Dibenzo-pdioxin and Dibenzofuran Levels in Human Adipose Tissue and Milk Samples from the North and South of Vietnam". Chemosohere 15:1613-1620. Schoiz, B4 Engler, M. 1987. "Determination of Polychlorinated Dibenzo-p-dioxira and Dibenzafttrana in Wastes of Techocal Hexacfakmcyclohexsne*. Chemosohere 16:1829-1834. Shinishi, H4 Pilkington, N4 Otsulri, A4 Fawn. K. 1985. "Occurrence of Chlnrinairri Polynuclear Aruoutis Hydrocarbons in Tap W tu * . Environ. Sri. Technol 19:585-590. Sijm, R.; Opperhnizen, A. 1988, "Biooamfannarioii, Biaaccumuiaoon, and Lethality o f IS Dichlorodibenzo-p-dioxin: A Proposal to Explain the Biotic Fate and Toxicity of PCDDs and PCD B". Chemomhere 17: 83-99. Silbergcld, E4 Mttftison. D, 1987. "Experimental and Clinical Studies on the Reproductive Toxicology of 2J,73-TeoaehlorodibenzDiHfioxin'. American Journal of Tndusmai Madiamn 11:131-144. Soikkr.ll, J4 Tarhanen. J4 Paasivina. J4 Witick, A. 1986l "Multicomponent Analysts Method far Dimeric Chlorophenol Ethers (PCDE, PCPAs, and PCBAi) in Biological Samples*. Chemosohere 15:2103-2104, Sparschu, G4 Dunn, F.; Rowe, V. 1970. Teratogenic Study of 2J,7.S-tetrachIorodiben2o-pdioxin in the Rat". IoxicoiozY and Applied Pharmacology 17:317-318. 84 Dioxins and Furans T1 Stalling D.; Smith, L ; Petty, J4 Hogan, J,; Johnson, J.; Rappe, C.; Baser, H. 1983, 'Residues of Polychlorinated Dibeazo-p-dioiins and Dibenzofurans in L ibrarian Great i-ays Fish*. In Hitman and Fnvimmnwnal Bwlrt nf Chtflrinaiwd Dimjna and Related Compounds. Tucker, R.: Young, A4 Grey, A,, Eds. Plenum Press, New York, pp 221*240. Stehr-Green, P.: Hoffman, R,; Webb, 1C; Evans, G j Knutsea, A,; Schramm, W.; Staake, J.; Gibson, B.; Steinberg, K. 1987. 'Health Effects of Long-Term Exposure to 2J.7.8Teuachlorodibenzo-p-dioxin*. Chemosphere 162089*2094. Swanson, Sa Erickson. M4 Moody, L. 1985. Therm al Degradation Products from Dielectric Fluids'. Office of Toxic Substances, USEPA, Washington, D.C, EPA 560/5*85-022. Swanson, S.; Erickson. M.: Moody, L.; Heggem, D. 1986. Therm al Combustion o f Octachlorodibenzofuian to F o r m Lower PCDFs". In QilflTTfiird Dioxins and D i h e n z o f u r a n s in Perspective. Rappe, C.; Choudhary, G.; Keith, L.. Eds. Lewis Publishers, Chelsea, MI. pp 109119. Swanson, S.; Rappe, C.; Malmstrom, J.; Kringstad, K. 1988. *Emissions of PCDDs and PCDFs from the Pulp Industry". Chemosphere 17:681-691. Tarfcowski, S.; Yrjanheikld, E. 1986. 'PCDDs and PCDFs in Human Milk- Reasons for Concern". Chemosphere 15:1641-1648. Thoma, H. 1988. "PCCD/PCDF Concentrations in Chimney Soot from House Healing Systems". Chemosphere 17:1369-1379. Thoma. H.; Hauschultz, G.; Hutzinger, O. 1987. 'Chlorine-Bromine Exchange During Pyrolysis of 1,13,4 Tetrabnxnodibenzodioxin With Various Chlorine Donors". Chemosphere 16:15791581. Tosine, H. 1983. "Dioxins; A Canadian Perspective*. In Chlorinated Dioxins and Dibenzofurans in the Total Environment. Choudhary, C.; Keith, L ; Rappe, C., Eds. Buuerworth Publishers, Woburn, MA. p 3. Travis, C.; Hauemer-Frey, H. 1987. 'Human Exposure to 23.7J8* TCDD". Chemosphere 16:2331-2342. Tsuji, Nakano. T4 Okuno, T. 1987. 'Measurement of Combustion Products from Liquid PCB Waste Indnerauv" Chemosphere 16:1889-1894. Umbceit, T.Hu Hesse, E J.; Gallo, M.A. 1986. 'Bioavailabiliiy of Dioxin in Soil from a 2,4J -T Manufacturing Site*. Science. 232:497-499. United States Department of Commerce. 1987. "Annual Survey of Manufacturers'. United States Environmental Protection Agency (USEPA). 1985. Hf"1!! Ay*asmem Document for Polychlorinated Dibenzo-p-dioxin. USEPA, Office of Health and Environmental Assessment. Washington. D.C. EPA 600/&-84/014F. USEPA. 1986. Treatm ent Technologies for Dioxin-Containing Waste*. USEPA, Hazardous Waste Engineering Research Laboratory. Cinrinaai, OH. EPA/600/2-86-096 USEPA. 1987a. National Dioxin Study. USEPA. Office of Solid Waste and Emergency Response. Washington, D.C. EPA 53Q/SW-87/D25. USEPA. 1987b. "Polyhabgenated Dibenzo-p-dioxins/Dibenzofurans; Testing and Reporting Requirements'. Federal Register 52:21412*21452 (June 5, 1987). Dioxins and Furans 85 I L 784186 GENP 011375 USEPA. 1988a. Federal Register. Volume 53:53282-23329. (30 December, 1988). USEPA. 1988b. 'Report o f ibe Environmental Effects, Transport, sad Fate Committee: Evaluation of Scientific Isaacs Related to Municipal Waste Corabastion*. USHPA Office of the Adminisracor, Science Advisory B a u d Washington, D.C. SAB-EHTTC-88-25. USEPA. 1988c. 'A Cancer Risk-Specific Dose Estimate for 23,7,8-TCDD*. USEPA Office of Research and Development, Cincuutd, Ohio. EPA/60(V6-88/Aa\ van den Berg, M.; Otie, 1C: Hwringer. 0 . 1983. U ptake and Selective Retention in Rats of Orally Administered Chlorinated Dioxins and Dibenzofurans from Fly-Ash and Fly-Ash Extract*. Chfimosptos 12:537-544. van den Berg, M.; Heeremans, C.; Meerman, L.; Veenhoven, Ed van Wijnen, J.; Olte, K. 1986. 'Som e Pharmacokinetic Aspects of PCDDs and PCDFs in Mammals after Administration of a Fly-Ash Extract from a Municipal Incinerator*. Chemosohere 15:1477-1487. Vecchi, A.; Sironi, M.; Canegrati, Md Ganmini, 5. 1983. 'Comparison of the Immune Effects in Mice of 2,3,7,8-Tetrachlorodibenzo-p-diQxin and 2J,73-Tetrachiocodibenzofuran*. In Chlorinated Dioxins and Dibenzofurans in the Total Environment- Choudhary, 0.; Keith, L.; Rappe, C., Eds. Buuerwonh Publishers, Wobum, MA. pp 397-405. Vulcan Chemicals. 1988. 'Pemachlorophenot in Perspective: A Summary Report*. Vulcan Chemicals, Birminghaip, Alabama. Wakimoto, T.; Kannan. N.; Ono, M.; Tatsukawa, R.; Masuda, Y. 1988. `Isomer-Specific Determination of Polychlorinated Dibenzofurans in Japanese and American Polychlorinated Biphenyls*. Chemosphere 17:743-750. Weerasinghe, N.; Gross, M. 1985. "Origins of Polychlorodibenzo-p-diaxins (PCDD) and Polydilorodibenzofurans (PCDF) in the Environment", fn Dioxins in the EnvironmenL Kamrin, M.; Rodgers, P., Eds; Buttenvonh Publishers, New York, pp 133-151. Wiklund, K.; Holm, L. (1986). "Soft Tissue Sarcoma Risk in Swedish Agricultural and Forestry Workers". Journal of the National Cancer Institute 76:229-234. Wtpf, H.; Schmid, J. 1984. "Seveso- An Environmental Assessment". In Human and Environmental Risks of Chlorinated Dioxins and Related Compounds. Tucker, Rd Young, A.; Gray, A-, Eds, Plenum Press, New York, pp 255-274. Wjpf, H.; Homberger, Ed Neuner, Nd Schenker, P. 1978. "Field Trials on Fhocodegradation of TCDD oo Vegetanarj after Spraying with Vegetable Oil". In Dioxin: Toxicological and Chemical Aspects. Canabeni, Fd Cavaflaro, Ad Galli, 0 ., Eds. Spectrum Publishers, New York, pp 201-217. Wipf, Hd Homberger, Ed Neuner, Nd Ranalder, Ud Vetter, Wd Vuilleumier, J. 1982. "TCDDLevels in Soil and Plant Samples from the Seveso Area". In rhinriiwtwi Dioxins and Related Compounds. Hutzinger, Od Frei, Rd Meriam, Ed Pocchiai, F,, Eds. Pemamon Press, Oxford, UJC. pp 115-126. ' World Health Organization (WHO). 1988. "FCBs, PCDDs. and PCDFs in Breast Milk: Assessment o f Health Risks". WHO Regional Office for Europe, Copenhagen. Environmental Health Series, No. 29. Yamagishi, Td Miyazaki, T.; Akiyama. Kd Marita, Md Nakagawn. J.; Horii, Sd Kaneko, S. 1981. Chemosphere 10:1137. 86 Dioxins and Furans 1 i i t GENP 011376 784187 Young* Aa C odfgtam . L. 1985. T ale of TCDD in Field Ecosystems- Assessment and Significance for Human Exposures*. In Dioxins in the Environment. Kamrin, M ; Rodgers, P_ Eds. B utlaw ortt Publishers* New Yoric. pp 153-171. Young, A.; Cockexham, Lz `ITialtoi* C. 1987. 'A Long-tom Study of Ecosystem Contamination with 23,73-TCDD." Chemosnhere 16:1791-1815, Dioxins and Furans 87 784188 GENP 011377 A List of Documenta Read b al do* Cited American Chemical Society. 1983. 'Dioxin: A Special Issue*. Chemical and Engineering News. Volume 61June 6 . B a res, D. 1983. "Regulatory Actions on Dioxins and Related Compounds*. In Human and Environmental Risks of Chlorinated Dioxins and Relaird rrinrwn<K. Tucker. 1L; Young, A 4 Gray, A., Eds. Plenum Press, New York, pp 23*31. Branson, D.; Takahashi, I.; Parker, W.; Blau, G. 1983. 'Btocancentration Kinetics of TCDD in Rainbow Trout*. Environmental Toxicology and Chemistry 4:779-788. Buser, R.; Rappe, C. 1980. "High-Resolution Gas Chromatography of the 22 Tenachlorodibenzo-p-dioxin Isomers". Analytical Chemistry 52:2257-2262. Carier, C.; Kimbrough. R.; Liddle, J.; Oine, R.: Zadc. M.; BanheL W. 1975. 'Tetrachloradibenzodioxin: An Accidental Poisoning Episode in Horae Arenas*. Science 188:738-740. Centers for Disease Control (CDC). 1988. 'Health Status of Vietnam Veterans: Psychosocial Characteristics'. Journal of the American Medical Association 259:2701-2707. 1 CDC. 1988. "Health Status of Vietnam Veterans: Physical Health". Journal o f the American Medical Association 259:2708-2714. CDC. 1988. 'Health Status of Vietnam Veterans: Reproductive Outcomes and Child Health". Journal of the American Medical Association 259:2715-2719. Choudhiy, G.; Huuinger, O. 1982. 'Photochemical Formation and Degradation of Polychlorinated Dibenzofurans and Dibenzodkuins". Residue Reviews 84:113-161. Couture, Elwell, M.; Bimbaum. L. 1988. "Dioxin-like Effects Observed in Male Rats following Exposure to OCDD during a 13-Week Study.' Toxicology and Applied Pharmacology 93:31-46. des Rasters. P. 1987. "National Dioxin Study". In Solving Hazardous Waste Problems: Learning from Dioxins. J. Exoer, Ed. The American Chemical Society Symposium Series, 4338. pp 34- Eduljee, G. 1987. "Volatility of TCDD'and PCB 6 0 m Soil". Chcmosohere 16:907-920. Erickson , J.; Mulinare, J4 McClain, P4 Fitch, T4 James, U ; McCleam, A.; Adams, M. 1984. "Vietnam Veterans' Risks for Fathering Babies with Birth Defects". Journal of the American Medical Association 251*903-912. Field, B4 Kerr, C . 1988. "Reproductive Behaviour and Consistent Patterns of Abnormality in Offspring of Vietnam Veterans". Joumal of Medical Genetics 23:819-826. Freeman, R4 Schroy, J. 1985. "Environmental Mobility of TCDD". Chanosphere 14:873. Giri, A. 1986. "Mutagenic and Genotoxic Effects of 2J,7^-Tetrachkaodibenzo-pKlioxni". Mutation Research 168:241-248. Gramley, J, 1987. "Understanding Dioxin", New York State Joint Legislative Commission an Toxic Substances and Hazardous Wastes. Albany, New York. 88 Dioxins and F in n s "1 Hay, A. 1982. The Chemical Scvthe: Lessons of 2.4.5-T and Dioxin. Plenum Press, New York. Hiraoka, M4 Takizawa, Y4 Masuda, Y.; Takeabita, R.; Yagomc, K.; Tanaka, M.; Wannabe, Y.; Morikawa, K. 1987. Investigation on Generaban of Dioxins end Related Compounds from Municipal Incineraron in Japan'. Chcmosphcrc 16:1901-1906. Hoffman, R.; Stehr-Green, P.; Webb, K., et aL 1986. 'Health Effects of Long-term Exposure to 2J.7.8-TCDD", Journal of the American Medical A<syiariofl 255:2031-2038. Jackson, D.; Roulier, M.; Grana, H.; Rust, S.; Warner, J. 1986. 'Solubility of 2,3,7,8-TCDD in Contaminated Soils'. In Clorinated Dioxins and Dibeitzofurans in Perspective. Rappe, C.: Choudhary, C.; Keith, L., Eds. Lewis Publishers, Chelsea. ML pp 185-200. Kimbrough, R. 1980. Haloeenated Biphenyls. Terohcnvls. ^ awhalencs. Dibcnzodioxins and Related Products. Elsevier/Nonh Holland, New York. Kimbrough, R. 1983. 'Morphology of Lesions Produced by the Dioxins and Related Compounds'. In Human and Environmental Rid nf Chlorinated Dioxins and Related Compounds. Tucker, R.; Young, A.; Gray, A , Eds. Plenum Press, New York, pp 527-538. Kimbrough, R., Falk, H., Stehr, P,, Fries, G. 1984. 'Health Implications of 2J.7.8-TCDD Contamination of Residential Soil'. Journal of Toxicology and Environmental Health 14:47-93. May, G. 1983. TCDD*. A Study of Subjects 10 and 14 Years After Exposure'. Chemosohcre 12:771-778. Miller, G4 Zcpp, R. 1987. *23.7,8-TeuachIorodibenzo-p* dioxin; Environmental Chemistry' . In Solving Hazardous Waste Problems: Learning from Dioxins. J. Exner, Ed. The American Chemical Society Symposium Series, #338. pp 82-93. National Council of the Paper Industry for Air and Stream Improvement (NCASI). 1987. Technical Bulletin No. 524. 'Dioxin; A Critical Review of its Distribution, Mechanism, Impacts on Health, and the Setting of Acceptable Exposure Limits'. National Council of the Paper Industry for Air and Stream Improvement. Inc. New York. NCASI. 1987. Technical Bulletin No. 525. "Assessment of Human Health Risks Related to Exposure to Dioxin from Land Application* of Wasrewater Sludge in Maine". National Council of the Paper Industry for Air and Stream Improvement. Inc. New Y ork.. Nogrodt, L; Balischmider, K. 1986. 'Causes for, and Reduction Strategies Against Emissions of PCDD/PCDF from Waste Incineration Plants- Interpretation of Recent Measurements*. Chemosohcre 15:1225-1237. O'Keefe, P.: Meyer, C4 Smith, R4 Hilker, D4 Aldoux, JC; Wilson. L. 1986. "Reverae-Phase Absorbent Cartridge for Trapping Dioxins in Drinking W ats*. Chemosohae 15:1127-1134. Otis, K4 Lustenhouwer, J.; Hutzinger, O. 1982. 'Polychlorinated dibcnzodioiins and Related Compounds in Incinerator Effluents', In Chlorinated Dioxins and Related Compounds. Hutzinger, O 4 Fre, R4 Meriam, E.; Pocchiai. F,, Eds. Pergamon Press, Oxford, UJC pp 227244. Rappe, Cd Kjeller, L ; Maiklund, S.; Nygren, M. 1986. 'Electrical PCB Accidents, an Update". Chcmosphere 15:1291-1295. Dioxins and Furans 89 784190 fV t Schecier, A4 Ryan, J.; Gitlitz, G. 1986. 'Chlorinated Dioxin and Dibenzofuran Levels in Human Adipose Tissues from Exposed and Control Populations". In Clotinated Dioxins and Dibenzoforans in Perspective, Rappe, C; Choudbay, C ; Keith, L , Eds, Lewis Publishers, Chelsea. ML pp 51-65. Saoggin, D. 1984. "The Interaction of Science, Policy, and the Law in Agency Use of Risk Assessments for the Regulation of Carcinogens". Hazard* Waste 1:363-375. Seefeld, M4 Peterson, R. 1983. "TCDD-Induced Weight Loss: A Proposed Mechanism". In Human and Environmental Risks of Chlorinated Dioxins and Related Compounds. Tucker, R.; Young, Aa Gray, A., Eds. Plenum Press, New York, pp 405-413. Smith, A.; Pearce, N. 1986. "Update on Soft Tissue Sarcoma and Phenoxy Herbicides in New Zealand". Chcmosohcre 15:1795-1798. Smuckier. E. 1985. "Biological Effects of Dioxins and Other Halogenated Polycyciics". In Dioxins in the Environment. Kamrin, M4 Rodgers, P,, Eds. Butterwonh Publishers, New York, pp 215-223. Suskind. R- 1985. "The Health Effects of 2,4.5-T and its Toxic Contaminants'. In Dioxins in the Environment. Kamrin, M.: Rodgers. P.. Eds, Butierworth Publishers, New York, pp 231-239. I Thiess, A.; Fremzel-Beyme, R.; Link, R. 1982. "Morality Study o f Persons Exposed to Dioxin in a Trichlorophcnol-Process Accident that Occurred in the BASF AG on November 17, 1953". American Journal of Industrial Medicine 3:179-189. Toxic Substances and Disease Registry. 1987, Toxicological Profile for 23.7,8Tetrachlorodibeozo-p-dioxin". Agency for Toxic Substances and Disease Registry, U.S. Public Health Service, CAS 1746-01-6. USEPA. 1988. "A Cancer Risk-Specific Dose Estimate for 23,7,8-TCDD: Appendices A Through F". Office of Health and Environmental Assessment, Washington, D.C. EPA/60Q/688jTO7Ab. Van Strum, Ca Merrell. P. 1987. No Margin o f Safety. Greenpeace, USA. Vogg, H.; Stieglitz, L. 1986. "Thermal Behavior of PCDD/PCDF in Fly Ash from Municipal Incincraton". Cbemorohere 15:1373-1378. Webster, G4 Muidrew, D.; Graham, J4 Santa, L4 *Muir, D. 1986. "Dissolved Organic Matter Mediated Aquatic Transport of Chlorinated Dioxins". Chcmosphcre 15:1379-1386. Zack, J.; GafTey, W, 1983. "A M orality Study of Workers Employed at the Monsanto Company Plant in Nitra, West Virginia". In Human a rid E n v iro n m e n ta l H iA x of Chlorinated Dioxins and Related Compounds. Tucker, R4 Young, A.; Gray, A,, Eds. Plenum Press, New York, pp 575591. \ ! i 1 1 90 Dioxins and Ftnans GENP 011380 784191 "1 Append! Tbs Virtually Safe Lifetime Daily Dose (VSD) of 2J3.7.8-TCDD is that quantity of 2J.7.8- TCDD estimated to cause a maximum of one tumor in one million people who are exposed for 70 yean. In other words, one person faces a maximtan one in one million risk of a tumor due to 2J.7.8-TCDD during their lifetime, if they are exposed to the VSD of 2J.73-TC D D for their lifetime. Virtually Safe Lifetime Daily Dose of W .73-TC D D , from the U.S. Environmental Protection Agency 6 fenuograms/kg/day A ferntogram is 1 x 10,u grains or 1 quadrillionth of a gram. Note: The USEPA has proposed to increase this daily dose to 100 femtograms/kg/day. Virtually Safe Levels of U .73-T C D D in Soil and W ater, from the USEPA Soil: Below 1 ppb requires no immediate action (fa n the U.S. Center for Disease Control's recommendation for1Times Beach, Missouri). W ater 13 pans per quadrillion (1.3 x 10,M grams/liter). The Advisory for Fish Contaminated with 2J,73-TC D D , from the U.S. Food and Drug Administration Fish with levels of 2,3,7,8-TCDD above SO ppt should not be eaten. Fish with levels bom 25 to 50 ppt can be eaten twice a month. Fish with levels below 25 ppt can be eaten yrith no restriction. (These levels can also be used far 2J,7,8-TCDD in other foods). Dioxins and Fiirans 91 784192 OBTSTP 0 11381 Recommended Reading For a succinct, informative pamphlet that summarizes the health risks of 2^,7,8-TCDD in an casy-Uxcad style: 'Dioxin in the Environment: Its Effect on Human Health" 1986. Available from the American Council on Science and Health, 47 Maple S i, Summit, NJ 07901. Telephone (201) 277-0024. Price is S2 per copy. For a well-written synopsis of the issues surrounding 2J.7.8-TCDD and the relevant scientific information: 'D ioxin' by F. H. Tschirley, in Scientific American. February, 1986. Volume 254, Number 2, pages 29-35. Available from Scientific American. 415 Madison Ave. New York, NY 10017, Attention Rosa Davis. Cost of issue is S4. Far a book that discusser in detail the issues surrounding 23.7,8-TCDD, especially the incidents that received the most publicity in the U.S., and includes much of the relevant scientific information on 2.3,7,8-TCDD: Dioxin. Agent Orange by Michael Gough, 1986. Plenum Press, New York. Far a summary of the current research findings on dioxins and furans, consult the proceedings o f the annual International Symposium an Chlorinated Dioxins and Related Compounds, published annually in the journal Chcmosnhere (Pergamon Press). I i ! 92 Dioxins and Furans GENP 011382 I 784193 W 4-TC D D 2AD 2 .W CDC ADI ODD CDF congener DNA dioxins Furans homoiogue isomer kg LCSO LDSO LOAEL milligram m icrograra nanogram pkogram fem tognm NCASI NOAEL NOEL NTP PBBs PCBs ppm ppb ppt ppq ug USAF USDA USEPA USFDA WHO Glossary and List of Abbreviation! 2J,7,8-tetrachlorodibenzD-para-diaxin 2,4-dichloropbenoxyacetie ad d 2,4.5-crichloropSenoxyacecc acid United States Cenien for Disease Control acceptable daily intake chlorinated dibenro-p-dioxin chlorinated dibenzo furan a single, particular member of a chemical family deoxyribonucleic add chlorinated dibenzo-p-dioxins chlorinated dibenzo furans a group of dioxins or firons with the same number of chlorine atoms in their structure a single member of a homologue kilogram lethal concentration for 50% of test organisms lethal dose for 50% of test organisms lowest-observed-adverse-eifcas level 1 x ID1 grams * 1 x 10-4 grams 1 x 10* grams 1 x lO 1* grams 1 x 10`u grains National Council of the Paper Industry for Air and Stream Improvement no-observed-adverse-effects level no-observed-effect level National Toxicology Program polybrominated biphenyls polychlorinated biphenyls parts per million parts per billion parts per trillion pans per quadrillion microgram United States Air Farce United States Department of Agriculture United States Environmental Protection Agency United Slates Food and Drug Administration World Health Organization Dioxins and Furans 93 784194 ENPoil383 Inde to Authors A bbruni, R. SS Adams, R. 31 Adams, W. SI Albanese, R. 7, 8, 62, 63 Amendola, G. 21, 37 Barnes, D. 16, 17 Baughman, R. 22, 32, 33 Beall. M. 23 Beck, H. 37. 38 Bcnc2t. H. 23 Bianco, W. 10 Birnbaum, L. 56-59 Bowes, G. 31 Brooksbank, M. 20, 25 Brunner, H. 36 Bumb, R. 41 Buser, H. 14, 31, 40 Carnegie Mellon University 36 CDC 9,58 Choudhary, G. 28, 46 ' Cochrane, W. 34 Cockerham. L. 22 Cook, R. 61 Coulston, F. 4 Courtney, i t 3 Couture, L. 17, 59, 72 Crosby. D. 23, 24, 40. 45 Czuczwa, J. 19, 20, 24, 43, 46, 47 Davies, K. 27 des Rosiers, P. 31, 40 DiDotnenico, A. 24 Engler, M. 37 Erickson. M. 39, 40 Esposito, M. 7, 33, 35 Faccheui. S. 22 Fairless, B. 20 Fara, G. 63 Fingerhut, M. 61 Firestone; D. 2. 3, 27 Fishbein. L. 61 Fox, G, 55 Gasiewicz, T. 2 8 .5 1 ,5 6 Gilbertson, M. 55 Gough, M. 2-4 Gross. M. 21,42 Gupta, B. 52 Hagenmaier, H. 36, 45 Haglund, P. 14 Hajdu, S. 61 Hakansson, H. 64 Hallet, D. 2 0 .2 5 Halperin, W, 61 Haidell, L. 61.62 Karfea, R. 22 94 Dioxins and Furarti GENP 011384 L 784195 Hatch. M. 62 Hanrmer-Frey, H. 27 Hay, A. S 4 .6 0 .6 3 Hay. D. 43 H dda, H. 18,22 Hdndl, A. 36 Held, T, SI Hites, R. 1 9 .2 0 .4 3 ,4 6 ,4 7 Holm, L. 61 Hombergcr, E. SS Hoochar, P. 61 Houk, V. 28 Huff, 3. 2-4 H iazing, 0 . 31, 33, 36, 43 * Isensee. A. 25 Janes, G. 25 Kamrin, M, S3 Kaxasek, F. 43 Kenaga, E 49, 51 Kimble, B, 42 Kissel, J. 9. 58 Kjelkr, L. 20 Knutson. J. 52-54, 59 Kociba, R. 52,5 4 ,6 8 Kauri, R. S3 Kuehl. D. 1 8 .5 7 ,5 8 Kunita 62 Kuotsune, M. 62 Lamparski, L. 42 Lave, L. 68 Lindstrom, G, 28, 47, 58 Maifclund, S. 27,41-44 Maiple, L. 25 Mastroiacovo, P. 9, 10, 62 Matsumura, F. 23, 53, 64 Matthews. H. 56-58 Mattisoo, D. 53 McConnell, E 18, 50 Meselsoo, M. 22, 33 Moore, J. 54, 58 M u ta je c, D. 26 Munay, F. 68 Nakano, T. 20 N r o g , R. 39 N ash .lt. 23 NCASI 3 8 .5 5 Neanick. T. 42 Norco. K. 47,58 Nanis, L. 49,51 Nanaom , R. 21,55 NTP 35,52, 68 N ygra, M. 27,28 Oefame, M. 24 Ogaki. J. 58 OUe, K. 42 Olson. J. 58 1 i I I j j Dioxins and Fuians 95 784196 GENP 0 1 1 3 8 5 OME 55 Ono, M. 27 Opperftuizen, A. 57, 58 Palausky, J. 23, 25 Patterson, D. 8 Paustenbach, D. S3. 69 Petenon, R. 54 Piikle, J. 58 Pitot, H. 53 Pleuss, N. 17 Poiger.H. 57 Poland, A. 12, 52-54,59.64 Rappe, C. 18, 20, 27, 31-33, 37. 39-41. 43, 44, 47. 57 Reggiani, G. 6, 55, 58, 60, 62 Rbarge, G. 9, 58 Rotani. W. 41 Ryan, J. 8. 26, 57 Safe, S. 12 Santi, L- 10 Sawyer, T. 12 Scheeler, A. 8, 28, 47 ( Schlatter, C. 57 . Schmid, J. 9, 10 Scholz, B. 37 Seveso. Italy 9, 22, 55 Shiiaishi, H. 20 Sijm, R. 58 Silbergeld. E. 53 Soikkeli, J, 40 Spanchu, G, 3 Stalling, D. 21 Stehr-Green, P. 6, 63 Swanson, S. 37, 39, 42 Tarkawski, S. 28 - Thema, H. 40, 42 Tosine, H. 34, 36, 44 Travis, C. 27 Tsuji, M. 44 U.S. Department of Commerce 34 Utnbieit, T, 18 USEPA 4, 6, 15, 19-21, 25, 27. 30, 32-36, 4 a 42-46, 48-50, 52-54, 56, 58. 60-62, 66. 69 ,70, 73 van den Berg, M. 18. 57, 58 Vecchi, A. 54 Vulcan Chemicals 69 Wakimoco, T. 31, 32 W asom , J. 2-4 Weentringhe, N. 21 WHO 28 Wiklund, K. 61 Wtpf, H. 9, 10, 22, 24 Wong, A. 2 3 ,2 4 Yamagishi, T. 36 Young. A. 22, 55 Yrjanheikki, H. 28 96 Dioxins and Furans GENP 011386 1_ 784197 1 lu d to Subjects 2J.7.8-substituted isomers accumulation of 57 in human tissues 26 in organisms 18 toxicity of 15 2J.73-T C D D acceptable daily doses from various agencies and countries 66 as a mutagen 54 half-life in organism 57 half-tife in various media 24 in areas sprayed with 2,4,5-T 19, 33 in fish 21 in humans 8, 26 in organisms 20 in organisms through use of 2,4,5-T 21 in plants 22 in soil 19 levels in 2,4,5-T 32 levels in Agent Orange 7 solubility in water 25 2J.7.S-TCDD equivalents 15, 16 2,4,5-T 3-5 and 23.7,8-TCDD in cattle 27 and 2 3 .7 3 -TCDD in organisms 21 and 2J,7,8-TCDD in soils and sediment 19 as a cause o f cancer 62 as a source of 23,7.8-TCDD 32 breakdown of 2J.7J3-TCDD in 23 cancellation of 3 in Agent Orange 7 levels of 2,3,7,8-TCDD in 32 2^4,5-trichlorophenol and discovery of 2 J ,7 aS-TCDD>s toxic effects 2 and Seveso, Italy 9 and sites contaminated with 23,7,8-TCDD 19 and Tunes Beach, Missouri 4 as a source of dioxins and haans 36 manufacture of 69 2.4-D as a source of dioxins 34 dioxins in 33 hi Agent Orange 7 Abonions among women in Alsea, Oregon 4 among women in Seveso, Italy 1 0 ____ in animals after exposure to 2J.7.8-TCDD 53 A b u p tio n of dioxins and lim ns 56 Acceptable dose calculating an 65 Agent Orange 7-10 Agents Green, Pink, and Purple 7 and environmental contamination 22 breakdown o f 2J.73-TC D D in 23 Dioxins and Furans 97 784198 GElSfp Olj ^ Alsea, Oregon 2,4,5-T and abortions 4 Animals, domestic and wild accidental deaths due to dioxins and furans 5, 9, 55 Birth defects attributed to 2,3,7,8-TCDD among children in Scveso, Italy 10 among children of Vietnam veterans 8 in animals 4, 33, 55 in humans 4, 62 Blood dioxins and furans in human 8, 26 Blood constituents effects of dioxins and furans on 54, 60 Breast milk and elimination of dioxins and furans 58 dioxins and furans in 26, 28, 47 Cancer among Vietnam veterans 8 and an acceptable dose of 2,3,7,8-TCDD 67 in animals, caused by dioxins and furans 52 in humans, caused by dioxins and furans 61 CDC and 23,7,8-TCDD equivalents 17 and Times Beach, Missouri 5 calculating exposure to dioxins and furans 9 risks posed by 2,3,7,8-TCDD 69 Chick edema in chickens 2 in Gull colonies 55 Chloraoie and birth defects 62 and discovery of 2.3,7,8-TCDD's toxicity 2 and Seveso. Italy 10 and Times Beach. Missouri 6 Chkxophenols and cancer 61 and occupational exposure to dioxins and furans 28 as a source of dioxins and furans 34-36, 40,45 regulation of 69 Cool as a source of dioxins and furans 42 Dioxins and furans absorption of 56 acceptable doses o f 2,3,7,8-TCDD, from various agencies 66 breakdown of 23 description of 13 detection of 11 effects on the environment 55 from burning coal and peat 42 from burning octa-CDDs/CDFs 42 from burning treated wood 40 from landfills and waste sites 45 from non-industrial sources 46 from paper mills and paper products 37 from the combustion of various products 38 from various ptoducu 30 98 Dioxins and Furans GENP 0 1 1 3 8 8 784199 1 Dioxins and funos (cool) from wood stoves and boilers 42 huauo expense so 26 in air 20,26 in animal and plana 2 0 ,2 7 in animals a contaminated sites 22 in drinking water 20 in food 27 in humans 8, 26. 46 in sediments of lakes and riven 19 in soil 19 mechanisms of toxicity 64 metabolism of 57 occupational exposure to 28 t pharmacokinetics o f 56 ' toxidty of 15 transport of 24,25 Diphenyl ether herbicides dioxins and fiuans in 37 j Environmental effects of dioxins and (ureas 5, 9, 55 I Flame retardants i as sources 6f dioxins and (mans 39 1 Food dioxins and furens in 27 Hilf-Iife of 23,7,8-TCDD in organisms 57 of 2J.7.8-TCDD in the environment 24 Hexachlorocydohexane dioxins and (mans m 37 Heauchlorophene and occupational expostse 28 and Times Beach, Missouri 4 ! dioxins and furens in 30,32 present manufacture of 09 Immnnc System, effects o f dioxins and (mans on among people of Times Beach 6 among Vietnam veterans 8 I in snimalt 54 I in htxnana 63 Inrinawore as sources of dioxins and furens 38, 3 9 ,4 3 ,4 4 Leaded Gasolines as a source of dioxins and (mans 1 4 ,2 0 ,4 1 ,4 2 Metabolism o f dioxins and furens 57 as it relates to toxicity 58 Migration o f dioxins and (mans 2 4,25 Milk dioxins m d funos in breast m ilk- see Breast milk dioxins and furens m dairy products 27 Movement of dioxins and furens 24,25 Mutagenic effects of 2J.73-TC D D 54 Occuptiicral exposure to and (m a n 28 t Octa-CDDs/CDFs dioxins retd furens from the burning of 42 [ dioxins and furens from the photolysis of 45 preferential absorption of 57 1 Dioxins and Furens 99 784200 G B N p 0 l i 3S9 / .1 > Paper mills and occupational exposure to dioxins and A n t s 28 as a source of dioxins and ( in n s 21, 37 paper sludge, environmental effects from use of S3 Paper products dioxins and Avans from the bunting of 42 dioxins and Avans in 38 PCBj as a source of dioxins and furans 31,45 confounding health effects o f 59 dioxins and fuxans from the burning of 39 regulation of 70 Pentachlorophenols and chick edema 3 . and dioxins and furans in food 27 as a source of dioxins and furans 36, 40, 45 regulation of 35 Pharmacokinetics of dioxins and furans 56 Photodegiadation of 2,3,7,8-TCDD 23 Photolysis as a source o f dioxins and furans 46 Polybrominaicd diphenyl ethos dioxins and Avans from the burning of 39 Polychlorinated benzenes and formation o f dioxins and Avans 46 dioxins and Avans from the bunting of 40 dioxins and furaru in 36 regulation of 69 Polychlorinated diphenyl ethers dioxins and furans from the burning of 40 dioxins and Avans in 37 Polyvinyl chloride (PVC) dioxins and Avans from the bunting of 41 Ranch Hands elimination of 2 3 .7 ,8-TCDD from 58 ' exposure to 2J.73-T C D D 8 health of 7 Relative toxicity of dioxins and Avans 15 Reproductive effects o f dioxins and furans abortions among women exposed to 2.4,5-T 4 abortions among women in Seveso, Italy 10 in animals 53 in humans 62 Seveso; Italy 9 ,2 2 , 55 Sewage sludge dioxins s t d Avans in 45 SQvex 4, 69 Tcaacfafanthykne dioxins and A n n s from the burning of 40 Times Beach. Missouri 5, 28, 69 Toxic equivalency factors 15 100 Dioxins and Furans 1 ! I | | | 1 I * j ! I 784201 1 . Toxicity of dioxins and finans 15 as related to metabolism 58 factors affecting 17 1 in a mixture 16 j in different soils 18 LCSOs and LD50s for various organisms 49 I mechanism of 64 ; NOAELa and LOAELs for various organisms 50 ! of a mixture of dioxins and furana 16 relative toxicity of all dioxins and furana 15 Vietnam 2,3,7,8-TCDD contamination in 22 and use of Agent Orange 7 dioxins and furana in people of N. and S. Vietnam 46 i 'Vietnam veterans i and Agent Orange 7 , exposure to AgentOrange 8 health of 1, 7-9 ! levels of 2J.73-TC D D in 8 | Wood dioxins and furana from the burning of treated wood 40 | Wood suives and oilers i as a source of dioxins and fixons 42 Yusho incident 57, 62 II 1 I i. I l Dioxins and Furans 101 784202 GENP 011391 SENSITIVITY ANALYSIS FOR DIOXIN RISK ASSESSMENT AND IMPLICATIONS FOR DETERMINING ACCEPTABLE LEVELS OF DAILY EXPOSURE by Russell E. Keenan, Ph.D. Richard J. Wenning, M.E.M. Alan R . Parsons, Ph.D. ChemRiskTM A McLaren Company Stroudwater Crossing 1685 Congress Street Portland, ME 04102 (207) 774-0012 November 28, 1989 Ch em R isk A McLaren Company GENF 011392 784203 Ch em R isk A McLaren Company GENP 011393 784204 ChemRiakTM - A McLaren Company NOVEMBER 28, 1989 TABLEOF' CONTENTS Eaqa EXECUTIVE SUMMARY .................................................... i 1.0 INTRODUCTION .................................................. 1 2.0 CURRENT UNDERSTANDINGS OF THE TOXICOLOGICAL BEHAVIOR OF TCDD ......................................................... 3 2.1 Acute and Chronic Human Toxicity . . . . . . . . .................. 2.2 Average Daily Intake of PCDDs and PCDFs .................... 2.3 Acute and Chronic Animal Toxicity ........................... 3 4 4 3 .0 DETERMINING ACCEPTABLE LEVELS F EXPOSURE TO TCDD ........ 7 3.1 Various Approaches to the Carcinogenic Dose-Response Assessment of T C D D ................... 7 3.2 Traditional U.S. Agency Dose-Response Analyses ..... ........ 8 3.3 Determining an Appropriate Level of R i s k .................... 11 4.0 SENSITIVITY ANALYSIS FOR DIOXIN RISK ASSESSMENT .......... 14 4.1 Overview ........ 14 4.2 Histopathological Interpretation of the Kociba et al. (1978) B i o a s s a y .................... IS 4.3 Interpretation of Hyperplastic Nodules and Neoplastic Nodules 17 4.4 Principal Carcinogenic Response in Female Sprague-Dawley Rats 20 4.5 Reassessment of Carcinogenic Risk Estimates Based on Hepatocellular Carcinomas ................................ 21 4.6 Use of the MLE versus the 95% LCL of the Human R s D ......... 22 4.7 Biological Basis for Extrapolating Across Species (Interspecies Scaling Factors) .............................. 24 5.0 CONCLUSIONS .................................................... 27 5.1 Conclusions of the Sensitivity Analysis ..................... 27 5.2 Implications for Determining Acceptable Levels of Daily Exposure..................................................... 29 6.0 REFERENCES ....................... 31 7.0 GLOSSARY OF TERMS .............................................. 41 G E N P 011394 784205 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 LIST OF ...FIGURES gaga Figura 1: Risk-Specific Doses (at 10~5) and Acceptable Daily Intakes Calculated for TCDD ....................................... iia Figura 2: Risk-Specific Doses (at 10-s) and Acceptable Daily Intakes Calculated for TCDD by Various U.S. Agencies and Other Countries ...... . . . . . . . . . . . . ............. ..... ....... la Figura 3: Risk-Specific Doses (at 10~5) and Acceptable Daily Intakes Calculated for TCDD by Various U.S. Agencies and Other Countries................................................... 8a Figura 4: Terminology of Rat Liver Lesions .......................... 17a Figure 5: Risk-Specific Doses (at 10~5) and Acceptable Daily Intakes Calculated for T C D D ....................................... 2 9a G E N P oi 1395 784206 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 " o rLIST `TABLES E aaa Table 1: Plausible Alternative Risk-specific poses (RsDs) for 2.3.7.8- TCDD ............................................ via Table 2: Toxicological Criteria (pg/kg-day) for TCDD ................ lb Table 3: Average Daily Intake of TCDD (pg/kg Body Weight) ........... 4a Table 4: Maximum Acceptable Daily Intake (ADI) of T C D D .............. 7a Table 5: U.S. Agency Risk-specific Doses for 2,3,7,8-TCDD ......... 9a Table : Activities Associated with 1 x 104 Increased Risk of Death in Any Y e a r ......................................... 13a Table 7: Comparison of Risks from Selected Activities cn a Per Capita Basis ............................................. 13b Table 8: Summary of Neoplastic Lesions Produced by 2,3,7,8-TCDD in Sprague-Dawley Rats, Spartan Substrain that are Statistically Significant in at Least One S e x ........................... 16a Table 9: Tissue Lesions Observed by Kociba et al. (1978) Among Sprague-Dawley Rats Exposed to 2,3,7,8-TCDD that were Statistically Depressed BelowControls .................... 16b Table 10: Tissue Lesions Observed by Kociba et al. (1978) Among Sprague-Dawley Rats Exposed to 2,3,7,8-TCDD that were Statistically Increased Above Controls ................... 16c Table 11: Principal Lesions in Female Sprague-Dawley Rats Exposed to 2,3,7,8-TCDD Reported by Kociba et al. (1978) ......... 16d Table 12: Histopathological Interpretation by Squire (EPA, 1985) of Principal Lesions in Female Sprague-Dawley Rats Exposed to 2.3.7.8- TCDD in the Kociba et al. (1978) B i o assay....... 17b Table 13: Summary of Hepatic Lesions Observed in Osborne-Mendel Rats and B6C3F1 Mice Exposed to 2,3,7,8-TCDD in the NTP (1982) Bioassay ............. 17c Table 14: Risk Estimates Derived from the Incidence of Hepatocellular Carcinoma and from the Pooled Incidence of Various Tumor Types Reported in Kociba et al. (1978) ........................... 21a G E JS (P 0 U 3 9 6 784207 ChamRiskTM - A McLaren Company NOVEMBER 28, 1989 LIST OF .. TABLES Page Table 15: ' Comparison at a 1 x 10~5 Risk Level Between EPA (1985) Estimates of the 95% Lower Confidence Limits (LCD on the Human RsD and the Maximum Likelihood Estimates (MLE) of the Animal RsD Based on Surface Area Correction........ 23a Table 16: Comparison at a 1 x 10-5 Risk Level Between EPA (1985) Estimates of the 95% Lower Confidence Limits (LCL) on the Human RsD and the Maximum Likelihood Estimates (MLE) of the Animal RsD Based on Body Weight Correction......... 23b Table 17 : Comparison at a 1 x 10"5 Risk Level Between Estimates of the 95% Lower Confidence Limits (LCL) on the Human RsD and the Maximum Likelihood Estimates (MLE) of the Animal RsD Based on the Incidence of Hepatocellular Carcinoma Reported by Kociba et al. (1978) ............................ ....... . 24a Table 18: Comparison Between Estimates of the 95% Lower Confidence Limits (LCL) on the Human RsD and the Maximum Likelihood Estimates (MLE) of the Animal R3D Based on the Incidence of Hepatocellular Carcinoma Reported by Kociba et al. (1978) and NTP (1982) ............................................ 24b Table 19: Range of Plausible Risk-specific Doses (RsDs) for 2.3.7.8- TCDD Based on Considerations of Extrapolation Between Rats and Humans, the 95% LCL versus the MLE, and the Principal Carcinogenic Response Observed in Kociba et al. (1978) 29b Table 20: Plausible Alternative Risk-3pecific Doses (RsDs) for 2.3.7.8- TCDD ............................................ 30a G E N ? 0U 397 784208 SENSITIVITY ANALYSIS FOR * DIOXIN RISK ASSESSMENT AND IMPLICATIONS FOR DETERMINING ACCEPTABLE LEVELS OF DAILY EXPOSURE EXECUTIVE SUMMARY 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is one of the most studied environmental chemicals. The great interest in TCDD shown by scientists and laymen alike is due partly to its high toxicity in bioassays involving laboratory animals. The fact that TCDD-contaminated Agent Orange was used in the Vietnam War as a defoliant and herbicide has sparked additional public concern and debate far in excess of the interest surrounding any other environmental chemical. Considerable research has been carried out to identify the long-term adverse health e.ffects associated with exposure to TCDD. Studies of the most highly exposed human populations have failed to associate increased cancer rates with exposure to dioxin or dioxin-contaminated chemicals (Armstrong, 1983; Bond et al., 1983, 1989; Cook et al., 1905; Eriksson et al., 1984; Filipini et al., 1981; Lathrop et al., 1983, 1984; Lipson, 1983; Mastroiacovo et al., 1988; Minister of Veterans'1 Affairs, 1983; Moses et al., 1984; Nelson et al., 1979; Pocchiari et al., 1979; Reggiani et al., 1976, 1980; Smith et al., 1982; Suskind and Hartzberg, 1984; Zack and Gaffery, 1983) . In addition, the total weight of evidence currently available does not support a conclusion that any of the phenoxy herbicides, presumably contaminated with TCDD, presents a carcinogenic hazard to humans (Bond et al., 1989). Agencies throughout the world have regulated TCDD because of its presumed human carcinogenic effects. In the absence of adequate dose-response data from epidemiologic studies, extrapolations from animal toxicity bioassays have been made in an attempt to characterize the relationship between TCDD uptake and an anticipated carcinogenic response in humans. Clearly, the assumptions used in extrapolating the dose-response curve generated from these experiments can have a significant impact on estimates of human risk associated with exposure to TCDD (Paustenbach,et al., 1986; Sielken, 1987). A number of regulatory agencies in Western Europe and North America have estimated human exposure limits for TCDD based on the application of classical toxicological safety factors to either the no-observable-adverse-effect level (NOAEL) or to the lowest-observable-adverae-effect level (LOAEL) of 'exposure in rodents. The use of the classical safety factor approach is based on the weight of scientific evidence that an exposure threshold exists below which adverse effects will not occur (Shu et al., 1987; CanTox, 1989) . Using this methodology, a number of countries have developed lifetime allowable daily intakes (ADIs) ranging between 1 and 10 picograms per kilogram of body weight per day (pg/kg-day). The U.S. Environmental Protection Agency (EPA) and certain other federal policy and regulatory groups, however, employ a very different extrapolation procedure to the same animal bioassay data (EPA, 1985). Through the use of the linearized multistage (IMS) model, human exposure levels at a selected G E N P 0li39g 784209 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page ii probability of increased cancer risk are extrapolated from rodent bioassay data. The use of this model is based on an assumption that there is no threshold for carcinogenesis; i.e., any dose, regardless of the quantity, poses some level of risk. At a risk level of 1 in 100,000 (1 x 10'5), the use of the multistage model and the EPA assumptions results in a risk-specific dose (RsD) of 0.064 pg/kg-dayl. Acceptable daily intakes and risk-specific doses developed for TCDD by various U.S. agencies and other countries are shown in Figure 1 for comparative purposes. It is clear that the application of the two different extrapolation procedures to the same animal cancer bioassay data results in markedly different exposure limits" for TCDD. Shu et al. (1987) have critically reviewed the TCDD bacterial, animal, and human data on mutagenesis, carcinogenesis, and tumor promotion and have concluded that the scientific evidence does not support risk estimates for TCDD which are based on linear low-dose extrapolation models and that alternative means for evaluating risk should be investigated. CanTox (1989), in their critical review of the biological data, concluded that the most appropriate method for the estimation of permissible limits for human exposure would be the application of a safety factor to the NOAEL observed in the twoyear chronic toxicity and oncogenicity study of TCDD in Sprague-Dawley rats conducted by Kociba et al. (1978) . This resulted in a recommended ADI for carcinogenic effects of 10 pg/kg-day, in agreement with that of certain Western, industrialized nations. This report by ChemRiskTM presents a sensitivity analysis of the EPA (1985) dose-response assessment for TCDD and a reinterpretation of the biological data from the Kociba et al. (1978) bioassay prior to using the LMS model for estimating scientifically defensible RsDs for TCDD. The quantitative impacts of several choices involved in the quantitative risk assessment of TCDD were reviewed, including the definition of the carcinogenic response of concern within the experimental dataset, the pathological evaluations of the original researchers and the EPA, methods of making the fitted multistage model responsive to the data at the lower experimental doses, the choice of scaling factor between rats and humans, and the method of estimating the maximum acceptable dose. Each of these choices affects the range of plausible risk estimates predicted through the use of the LMS model. In this analysis, RsDs were determined by ChemRisk at a lifetime incremental tisk level of 1 in 100,000 (1 x 10"5), since this level of risk is generally considered to be de minimis by regulatory agencies. In fact, examination of numerous regulatory decisions has shown that for effects on small populations, regulatory action was never taken for individual risk levels below one in ten 1 U.S. agencies, including the EPA, have calculated and presented RsDs at various levels of risk. For the sake of consistency and to enhance the clarity of this discussion, the risk level of 1 in 100,000 has been used in this document when presenting RsD estimates. GENP o i l 3 9 9 784210 ChemRlskTM A McLaren Company NOVEMBER 28,1989 Page Ha 15 M 13 12 11 10 9 8 7 >> f6 sa , 5 4 3 2 1 0 Figure 1. Risk-specific Doses (at 1(F5) and Acceptable Daily Intakes Calculated for TCDD1 *<N0 X *v y -- a a 2 tiD PDA U-V-X no tkwYmk ux Q u a forDIm * G w d , U i Mtic B ate S m a * RcO (I) w n w l w l l i i r f w y (ADI atfl* FKO) ]M <rKwr<rt-Aii ADI U.^.h -- (Mb* I k | * a - (wttia* bn* ^------ AH CDC (b) 1. U-S agencies including the EPA, have ctfailirnH and presented RsDs *1 various levels of risk. For the nice of oonristEacy md to enhance the clarity of this document; the risk level of 1 in 100,000 has been used in the presentation of RsD estimates. G E N ? 0 400 784211 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page ill thousand (1 x 10~4) (Travis et al., 1987). Furthermore, it is appropriate to consider the variance associated with background risk in the human population when determining an appropriate level o de minimis incremental risk (Farber, 1989) His analysis has illustrated that incremental risks associated with chemical exposures do not alter the probability of an individual contracting a disease if the incremental risk estimate is smaller than the variance estimate for the background rate of disease incidence. In the case of exposure to TCDD, Farber (1989) has concluded that an incremental risk level of 1 in 100,000 is indeed insignificant and de minimis on both mathematical and biological grounds. The low-dose risk estimate derived by the EFA through the application of the LMS model to the Kociba et al. (1978) and the Squire (EFA, 1985) dataset3 is very sensitive to the choice of modeling assumptions. The principal conclusions of the ChemRisk sensitivity analysis are summarized below. (1) Hepatocellular carcinomas are the only indicators of a carcinogenic response in the Kociba et al. (1978) bioassay having any conceivable use in a dose-response extrapolation via the LMS model. Considering only hepatocellular carcinomas, no statistically significant differences between controls and the lowest (1,000 pg/kg-day) or intermediate (10,000 pg/kg-day) dose groups were observed. Therefore, the dose-response curve is determined by only a single data point and the underlying assumptions of the model which force a fit to the origin in order to preserve the notion of "linearity". Furthermore, the high do3e level administered tothe rats (100,000 pg/kg-day) clearly exceeded the Maximum Tolerated Dose (MTD), evident from Kociba's observations of severe liver toxicity, diminished weight gain, and increased mortality. (2) The statistically significant (95% confidence limit) increased incidence of hyperplastic nodules observed by Kociba et al. (1978) in the intermediate and highest (100,000 pg/kg-day) TCDD treatment groups should not be modeled via a nonthreshold approach because: (a) There is no evidence in this study to indicate that a hyperplastic nodule is likely to progress to a hepatocellular adenoma and then to a hepatocellular carcinoma. (b) Kociba et al. (1978) identified and reported the occurrence of adenomas in other organs and tissues, while no adenomas were reported in the liver. This clearly indicates that the observed lesions were considered to be a less serious form of hepatocellular proliferative lesion, i.e. hyperplasia. Furthermore, Kociba et al. (1978) reported that there was no evidence of metastasis of any of these hyperplastic nodules. (c) "Neoplastic nodules", a term used by Squire (EFA, 1985) in his re-evaluation of the Kociba analysis, represents terminology from a different morphological classification system developed by Squire and another researcher. Its use by Squire (EPA, 1985) does not provide convincing evidence that the hyperplasia was actually a more serious condition. The GENP 011401 784212 ChemRiakTM - A McLaren Company NOVEMBER 2 8 , 19 8 9 Page iv classification system of Squire and Levitt (1975) failed to define the exact position and significance of the "neoplastic nodule",- which included both hyperplasia and more serious lesions (EPA, 1986b) . Today, the National Toxicology Program (NTP) no longer uses the classification scheme of Squire and Levitt and neoplastic nodules are no longer identified during histopathological examinations by most toxicological laboratories, including the NTP. "Neoplastic nodule" has been replaced with two terms: hyperplasia and hepatocellular adenoma (EPA, 1986b; Maronpot et al., 1986; McConnell et al., 1988). (d) SchulteHermann et al. (1983) demonstrated that a significant incidence pf hyperplastic liver nodules are found in the livers of untreated rats. The incidence increased to 100% over the course of the animals' two-year average lifetime and lesion size increased with age. In addition, these hyperplastic nodules showed similar characteristics to those observed in rats treated with known liver carcinogens. Schulte-Hermann et al. (1983) have concluded that the high incidence of hyperplastic liver nodules in untreated aged rats precludes the distinction between chemical promoters and initiating agents during long-term carcinogenicity bioassays in laboratory rodents, particularly when liver tumor incidence data are the primary criteria. (3) The statistically significant increased incidence of hyperplastic nodules observed by Kociba et al. (1978) in the intermediate (10,000 pg/kg-day) and highest (100,000 pg/kg-day) TCDD treatment groups should be appropriately extrapolated to human exposure only through the use of a classical safety factor approach: the LOAEL divided by a safety factor of 1,000. In this case, the LOAEL for hyperplastic nodules in the rat liver (10,000 pg/kg-day) divided by 1,000 would yield an ADI of 10 pg/kg-day. This approach is identical to that taken by the United Kingdom (1989) when they derived a guideline level of 10 pg/kg-day, "which, when exceeded, should trigger investigation and appropriate measures to reduce environmental levels." (4) The frequencies of tumors in other organ systems-, specifically in the lung, tongue and palate, should not be considered in the doseresponse assessment of TCDD. The Office of Science and Technology Policy (1985) does not recommend combining tumor incidences in different organs for the purpose of estimating a cancer potency. Furthermore, Kociba (1984) has suggested that the tumor incidence observed in the lung and oropharynx is likely the result of periodically high localized concentrations of TCDD associated with surface contact of TCDD from the animal's diet, rather than an effect of systemic uptake. (5) For TCDD, the use of a body weight correction factor in the rat-tohuman extrapolation procedure is biologically more relevant than the use of a surface area correction factor. The use of a surface GENP 011402 784213 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page v area scaling factor, based on the premise that the magnitude of the adverse effect of the chemical is dependent on the basal metabolic rate of the species of concern (O'Flaherty, 1988; Clayson, 1988), is inappropriate for poorly metabolized chemicals such as TCDD. In the case of TCDD, basic pharmacokinetics suggest that the parent compound is the biologically active moiety (Shu et al., 1987; Leung et al., 1988, 1989). Despite this evidence, the EPA used a surface area correction factor of 5.4 when extrapolating doses between Sprague-Dawley rat3 in the Kociba et al. (1978) study and humans. (6) Based on this analysis and on the other scientific evidence summarized in this report, it is clear that the use of the linearized multistage model in conjunction with the Kociba et al. (197 8 ) bioassay to describe the carcinogenic dose-response of TCC" can result in a wide range of plausible risk estimates (Table 1) . The choice of 'maximum likelihood estimate (MLE) for the RsD in preference to the 95% lower confidence limit (LCD 2 estimate, body weight instead of surface area scaling factor, and a better definition or biological interpretation of the reported liver lesions will result in RsDs much different than the estimates developed by the EPA (1985, 1988) and the CDC (Kimbrough, 1984) . (7) ChemRisk further concludes that the weight of scientific evidence indicates that TCDD should be regulated as if it exhibited a threshold of carcinogenic action. The scientific evidence does not support risk estimates for TCDD which are based on linear lowdose extrapolation models. The most appropriate method for estimating permissible limits for human exposure, given the present state of knowledge in this interim period before a b i o l ogically-based pharmacokinetic model is validated experimentally, is the application of a safety factor to the NOAEL for carcinogenicity observed in the Kociba et al. (1978) study. (8) Although the use of the safety factor approach may be most appropriate for the dose-response assessment of materials such as TCDD that lack genotoxic potential, it is recognized that others may wish to use the linearized multistage model or some other nonthreshold model of low-dose extrapolation in conjunction with the Kociba et al. (1978) dataset. Since this is currently the case with U.S. regulatory agencies, then the carcinogenic behavior of TCDD in female Sprague-Dawley rats should be extrapolated to humans only on the basis of the incidence of hepatocellular carcinomas. A body weight scaling factor should be used in preference to the surface area assumptions. If the 95% LCL of the human RsD is adopted, then an RsO at a 1 x 10~* risk level of 2.S 2The 95% LCL is that dose for which there is a 95% statistical confidence that the true RsD is no lower than this value. GENP 011403 784214 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page vi pg/kg-day of TCDD is estimated. If the MLE is adopted, then the appropriate RsD at a 1 x 104 risk level is 4.0 pg/kg-day of TCDD (Table 1) . Both of these RsDs are more conservative than the ADI of 10 pg/kg-day, estimated to be protective of hepatic hyperplasia, derived via a classical safety factor approach. GENP 011404 784215 ' ChemRiskTM - A McLaren Company NOVEMBER 28,1989 Page via Table 1. Plausible Alternative Risk-specific Doses (RsDs) for 2,3,7,8-TCDD Plausible Choices* Risk-specific Dose al 1 x 10*9Incremental Carcinogenic Risk (pg/kg-day) 1. Body Weight + 93% LCL* + Hepaio. carcinomas 25 2. Body Weight + MLEc+Hepato. carcinomas 4.0 a. These assumptions are considered by ChemRiskTM (1989) to represent the m ost plausible basis for an alternative human RsD for Z3.7.8-TCDD. b. T he 93% LCL is that dose for which there is a 93% statistical confidence that the true RsD is no lower than this value. c. Maximum likelihood estimate. G B N P 011405 784216 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 1 1.0 INTRODUCTION 2,3, 7, 8 -Tetrachlorodibenzo-p-dioxin (TCDD) ia one of the moat studied environmental chemicals. The great interest in TCDD shown by scientists and laymen alike is due partly to its high toxicity in bioassays involving laboratory animals. The fact that TCDD-contaminated Agent Orange was used in the Vietnam War as a defoliant and herbicide has sparked additional public concern and debate far in excess of the interest surrounding any other environmental chemical. Considerable research ha3 been carried out to identify the long-term adverse health effects associated with exposure to TCDD. Studies of the mo3t highly exposed human populations have failed to associate increased cancer rates with exposure to dioxin or dioxin-contaminated chemicals (Armstrong, 1983; Bond et al., 1983, 1989; Cook et al., 1985; Eriksson et al., 1984; Filipini et al., 1981; Lathrop et al., 1983, 1984; Lipson,'1983; Mastroiacovo et al., 1988; Minister of Veterans' Affairs, 1983; Moses et al., 1984; Nelson et al., 1979; Pocchiari et al., 1979; Reggiani et al., 1978, 1980; Smith et al., 1982; Suskind and Hertzberg, 1984; Zack and Gaffery, 1983) . In addition, the total weight of evidence currently available does not support a conclusion that any of the phenoxy herbicides, presumably contaminated with TCDD, presents a carcinogenic hazard to humans (Bond et al., 1989) . Agencies throughout the world have regulated TCDD because of it3 presumed human carcinogenic effects. In' the absence of adequate dose-response data from epidemiologic studies, extrapolations from animal toxicity bioassays have been made in an attempt to characterize the relationship between TCDD uptake and an anticipated carcinogenic response in humans. Clearly, the assumptions used in extrapolating the dose-response curve generated from these experiments can have a significant impact on estimates of human risk associated with exposure to TCDD (Paustenbach et al., 1986; Sielken, 1987). A number of regulatory agencies in Western Europe and North America have estimated human exposure limits for TCDD based on the application of classical toxicological safety factors to either the no-observable-adverse-effect level (NOAEL) or to the lowest-observable-adverse-effect level (LOAEL) of exposure in rodents (Table 2) . The use of the classical safety factor approach is based on the weight of scientific evidence that an exposure threshold exists below which adverse effects will not occur (Shu et al., 1987; CanTox, 1989). Using this methodology, a number of countries have developed lifetime allowable daily intakes (ADls) ranging between 1 and 10 picograms per kilogram of body weight per day (pg/kg-day) (Figure 2) , The U.S. Environmental Protection Agency (EPA) and certain other federal policy and regulatory groups, however, employ a very different extrapolation procedure to the same animal bioassay data (EPA, 1985) . Through the use of the linearized multistage (LMS) model, human exposure levels at a selected probability of increased cancer risk are extrapolated from rodent bioassay data. The use of this model is based on an assumption that there is no threshold for carcinogenesis; i.e., any dose, regardless of the quantity, GENP 011406 784217 ChemRIskTM A McLaren Company NOVEMBER 28t 1989 Page la "s 15 14 13 12 11 10 9 8 f7 6 5 4 3 2 1 0 Figure 2. Risk-specific Doses (at 1(T5) and Acceptable Daily Intakes Calculated forTCDDiby Various U.S. Agencies and Other Countries CD C(b) ETA OMfonU OEC R IAV-K RO IfcwYu* !hM a Cm * tat US. fwmeim A ^acy. **> SasWCtMTio-taO Qmmb t o G a m i. U P lfc tic B n i* Serrila .1 * 0 (0 U M C M T N ll| 4 ia ^ U IW a lD n |M to M i-U > M nlbp iW irffln aT ^A D I ADIAa--km m c-- > w w ll(w r f t w adabaltfRO ) S uarfttnrY at.A M -AM W tow yrfttm ^aaA w iW tJf-- C tob-A M 1. U .S. Agencies, including the E PA , h av e c alcu lated an d presented Rj Ds a t v en o u s levels o f rule. F a r th e saJce o f consistency t a c to enhance the clarity o f this docum ent, the risk level o f 1 in 100,000 hat been u * d in the presentation o f R xD estim ates. GENP 011407 784218 ChemRskTM A McLaren Company NOVEMBER 28,1989 Page lb Table 2. Toxicological C riteria (pg/kg-day) for TCDD Health Effect Carcinogenicity (Kociba et al., 1978) Toxicological Criteria Extrapolation Factor ADI 10,000 (NOAEL) + 1000 10 Hyperplastic nodules (Kociba et al,, 1978) 10,000 (LOAEL) 1,000 (NOAEL) + 1000 + 100 10 10 Reproductive effects (Murray et al., 1979) 1,000 (NOAEL) . +100 10 GENP011408 784219 CheaRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 2 poses some level of risk. At a risk level of 1 in 100,000 {1 x 10"5), the use of the multistage model and the EPA assumptions results in a risk-specific dose (RsD) of 0.064 pg/kg-day1. Acceptable daily intakes and risk-specific doses developed for TCDD by various U.S. agencies and other countries are shown in Figure 2 for comparative purposes. Shu et al. (1987) have critically reviewed the TCDD bacterial, animal, and human data on mutagenesis, carcinogenesis, and tumor promotion and have concluded that the scientific evidence does not support risk estimates for TCDD which are based on linear low-dose extrapolation models and that alternative means for evaluating risk should be investigated. CanTox (1989), in their critical review of the biological data, concluded that tfie most appropriate method for the estimation of permissible limits for human exposure would be the application of a safety factor to the NOAEL observed in the twoyear chronic toxicity and oncogenicity 3tudy of TCDD in Sprague-Dawley rats conducted by Kociba et al. (1978). This resulted in a recommended ADI of 10 pg/kg-day, in agreement with that of certain Western, industrialized nations. It is clear that the application of the two different extrapolation procedures to the same animal cancer bioassay data results in markedly different exposure limits for TCDD. `It is certainly true that a large component of this difference between the two approaches is based upon whether one believes that the dose-re3ponse curve for TCDD is best described by either a threshold or by a nonthreshold model. However, there is also reason to suspect that other, secondary assumptions made in the dose-response assessment may be contributing to the widely varying estimates of acceptable levels of daily exposure obtained through the use of the two procedures. We hypothesized that each of these secondary choices might possibly have a substantial impact upon the range of plausible risk estimates predicted through the use of the LMS model. In this report, ChemRiskTM presents a sensitivity analysis of the EPA (1985) dose-response assessment for TCDD and a reinterpretation of the biological data from the Kociba et al. (1978) bioassay prior to using the LMS model for estimating scientifically defensible RsDs for TCDD. These alternative RsDs were then compared to the estimates of acceptable daily intake, obtained via a classical safety factor approach to human extrapolation of rat dose-response data. The quantitative impacts of several choices involved in the quantitative risk assessment of TCDD were reviewed, including the definition of the carcinogenic response of concern within the experimental dataset, the pathological evaluations of the original researchers and the EPA, methods of making the fitted multistage model responsive to the data at the lower experimental doses, the choice of scaling factor between rats and humans, and the method of estimating the maximum acceptable dose. 1 U.S. agencies, including the EPA, have calculated and presented RSDs at various levels of risk. For the sake of consistency, the risk level of 1 in 100,000 has been used in this document when presenting RsD estimates. GENP 011409 784220 CheaRiskTM - A McLaren Company NOVEMBER 28, 198 9 Pago 3 2.0 CURRENT UNDERSTANDINGS OF THE TOXICOLOGICAL BEHAVIOR OF TCDD 2.1 Acuta and Chronic Human Toxicity Numerous epidemiologic studies have examined the potential association between TCDD exposure and a number of disease or mortality endpoints in humans (NCAS1, 1987). The epidemiologic database for TCDD has been reviewed by AMA (1984), EPA (1985), Fishbein (1907), NCASI (1987), UAREP (1988), and Bond et al. (1989) . Available data on human exposure is found primarily in studies involving industrial workers, herbicide sprayers, the exposed population in the Seveso, Italy accident, residents of Times Beach, Missouri and U.S. Air Force personnel involved in the use of Agent Orange in Operation Ranch Hand. In these studies, the concentrations of TCDD to which people were exposed was much greater than would typically be encountered in the environment. The applicability of these studies to environmental exposures, therefore, is unclear. The populations evaluated in many of these studies were exposed to multiple chemicals, thereby complicating the assessment of disease or mortality endpoints potentially attributable to TCDD exposure. Chloracne is the orily consistently demonstrated long-term adverse health effect associated with human exposure to TCDD (Suskind, 1985) . This characteristic persistent dermatosis has been observed in cases of both acute and chronic exposure to significant concentrations of TCDD and can be induced following systemic uptake or dermal exposure (Kociba and Schwetz, 1982; Suskind, 1985; Kimbrough and Houk, 1987) . Due to the seven year half-life of TCDD in humans, chloracne can persist for several years following high-level occupational exposures. While chloracne is associated with exposure to a number of other chlorinated aromatic hydrocarbons (Kimbrough et al., 1984), Suskind (1985) identifies TCDD as the most potent chloracnegen. Other health effects that have been reported in individuals exposed to substances contaminated with TCDD include porphyria cutanea tarda, hyperpigmentation, hirsutism, and altered liver function, (Bleiberg et al., 1964; Pazderova-Vejlupkova et al., 1981; Singer et al., 1982; Moses et al., 1984) . Most of these effects are derived from case histories or clinical surveys that lacked an assessment of these effects in control groups and also failed to assess concomitant exposure to other chemicals. Certain Swedish studies have suggested a positive association between phenoxy herbicide or chlorophenol (presumably contaminated with TCDD) exposure and soft tissue sarcoma and malignant lymphoma (Hardell and Sandstrom, 1979; Eriksson et al., 1981; Hardell et al., 1981) . However, other epidemiologic studies involving herbicide exposure have not confirmed the positive associations in these studies (Smith et al., 1982, 1983; Wiklund and Holm, 1986; Smith and Pearce, 1986; Pearce et al., 1986). The epidemiologic database on TCDD, supports the conclusion that chloracne is the only consistently demonstrated long-term health effect associated with exposure to TCDD. Studies of the most highly exposed human populations have failed to associate increased cancer rates with exposure to dioxins or dioxin- 784221 ChemRiakTM - A McLaren Company NOVEMBER 28, 1 9 8 9 Page 4 contaminated chemicals (Armstrong, 1983; Bond et al., 1983, 1989; Cook et al., 1985; Eriksson et al., 1984; Filipini et al., 1981; Lathrop et al., 1983, 1904; Lipson, -1983; Mastroiacovo et al., 1988; Minister of Veteran's Affairs, 1983; Moses et al., 1984; Nelson et al., 1979; Pocchiari et al., 1979; Reggiani, 1978, 1980; Smith et al., 1992; Suskind and Hertzberg, 1984; Zack and Gaffery, 1983) . The total weight of evidence, reviewed in the recent paper by Bond et al. (1989), shows that TCDD, as a contaminant of phenoxy herbicides, is unlikely to be a human carcinogen at low doses. Therefore, based on the available epidemiologic data, it is not appropriate to conclude that environmental exposures to TCDD are likely to produce cancer in humans. 2.2 Average Daily Intake of PCDDs and PCDFs Human exposure to TCDD is widespread. A number of studies have reported uptake of TCDD by persons with no known exposure to dioxins. In these individuals, adipose tissue levels of TCDD ranged from 3 to 10 ppt (Ryan et al., 1985; Graham et al., 1985; Patterson et al., 1986). Several studies have been conducted to estimate average daily intakes of PCDDs and PCDFs (Table 3) . These studies propose that food intake is the primary source of PCDD and PCDF exposure for humans `(Travis and Hattemer-Frey, 1987; Ontario, 1988; Beck et al., 1988). It has been estimated that approximately 98% of the daily TCDD intake is from food (Travis and Hattemer-Frey, 1987) . Estimates of average adult daily intakes of TCDD equivalents range from 0.54 to 2.33 pg/kg-body weight (Table 3). Potential routes of dioxin exposure through the food chain include fish, beef, and dairy products (Fishbein, 1987) . 2.3 Acute and Chronic Animal Toxicity The noncarcinogenic toxicity of TCDD in animals has been studied in a number of acute, subchronic and chronic studies. This compound has been shown to be extremely toxic to certain rodent 3pecie3 that have been tested. For example, the acute LDg (the dose which is lethal to 50% of the animals tested) for guinea pigs is reported to be 0.6 pg/kg body weight. The sensitivity to TCDD toxicity is extremely variable among laboratory animal species. Kociba and Cabey (1985) have shown the LD50 of hamsters to be as high as 5,051 pg/kg; i.a., TCDD is over 8,400 times less toxic to the hamster. Clinical signs of acute toxicity in laboratory animals are severe weight loss, hepatotoxicity, chloracne, thymic atrophy, and death. Chronic exposure to TCDD has been shown to induce reproductive effects in laboratory animals, including several strains of mice (Smith et al., 1976) and rats (EPA, 1985) . The no-observed-adverse-effect-level (NOAEL) for rats was found to be 1,000 pg/kg-day (Murray et al., 1979), while in the mouse, the NOAEL was 100,000 pg/kg-day (Smith et al., 1976). In general, 1, 000 pg/kg-day is considered the NOAEL for noncarcinogenic effects. Several agencies and authorities have developed acceptable daily intakes (ADIs, also known as reference doses) to protect against noncarcinogenic effects in humans using GENP 011411 784222 _ ChemRlskTM - A McLaren Company NOVEMBER 28,1989 Page 4a Table 3. Average Daily Intake of TCDD (pg/kg Body Weight) Travis and Gilman, personal Birmingham, Hanexnercommunication et al. 1989 Frey, 1987 Air Soil Water Consumer Products Food TOTAL 0.04 0.01 <0.01-0.05 <0.01 0.49-2.0 0.54-2.1 0.07 0.02 0.002 0.005 2.32 2J3 0.014 .0065 OfiSZ 0.678 GENP 011412 784223 ChemRiskTM - A McLaren Company NOVEMBER 28, 1988 Page 5 reproductive effects in animals as the critical effect (EPA, 1984; NCASI, 1987; UK, 1989) . These AOIs have ranged from 1 to 10 pg/kg-day, A number of regulatory agencies throughout the world have regulated TCDD because of its presumed human carcinogenic effects. Although the epidemiologic database does not support this presumption, extrapolations from animal toxicity bioassays have been made in an attempt to characterize the relationship between TCDD uptake and an anticipated carcinogenic response in humans. Chronic high-level TCDD exposure has been shown to be carcinogenic in rats and mice (Kociba, 1964) . A number of different tumor types are elevated, but the liver is the primary and most important target tissue (Kociba et al., 1978). However, carcinogenicity was evident only at doses that elicited a severe noncarcinogenic toxic response in the study animals. This was evidenced by the Kociba et al. (1978) observations of severe liver toxicity, diminished weight gain, and increased mortality. Carcinogens may be generally classified as initiators or promoters. An initiator compound must bind to cellular DNA (Williams and Weisburger, 1986) and alter it in drder to initiate carcinogenesis. The initiated cell incorporates the mutated (damaged) DNA into its replicating genome, which may be locked in the cell for as long as the cell line continues to reproduce. A promoter acts by increasing the tumorigenic response of a cell to an initiator when applied after the initiator (Williams and Weisburger, 1986). Promoters require prolonged and repeated exposure or persistence in the body before tumor formation occurs in animals, whereas for tumor initiators, short-term exposure may cause tumors (Shu et al., 1987). Tumor initiation is considered an irreversible event, while tumor promotion may be reversible upon removal of the promoter (Shu et al., 1987). TCDD binding to DNA, a necessary event in mutagenesis, does not occur to any significant extent (Poland and Glover, 1979) and mutagenicity tests generally indicate that TCDD is nonmutagenic (Shu et al., 1987) . These data support the conclusion that TCDD is not genotoxic. Studies by Pitot et al. (1980) and Poland and Knutson (1982) have shown that TCDD is a potent promoter. Many promoters, including TCDD, affect cellular growth and differentiation and alter a number of cell membrane properties (Weinstein, 1984) . Unlike genotoxic (initiator) carcinogens, tumor promoters may exhibit a threshold in their dose .response (Williams and Weisburger, 1986) . Shu et al. (1987) have critically reviewed the TCDD bacterial, animal, and human data on mutagenesis, carcinogenesis, and tumor promotion and have concluded that the scientific evidence does not support risk estimates which are based on TCDD as a tumor initiator. The authors noted that risk assessments which incorporate tumor promotion activity, more accurately reflect the scientific understanding of the mechanism of action of TCDD than do those assuming a nonthreshold (initiation) mechanism. GENP 011413 784224 ChomRisfcTM - A McLaren Company NOVEMBER 28, 1989 Page 6 A EPA special advisory committee (Dioxin Update Committee, 1986) noted that TCDD acts as a potent promoting agent in at least two different tissues in two species, but. there is no evidence for initiation activity in any species. Regarding risk assessment, the EPA Committee concluded that mechanistic models should be used for quantitative risk estimation for TCDD and related compounds. Models should consider epidemiological data, sex-species susceptibility, the promoting action of TCDD, and its pharmacokinetic properties in predicting ri3ks for exposed populations. Current models used by U.S. agencies do not, however, consider these important factors in deriving risk estimates. The wide range in the sensitivity of different laboratory animals to TCDD exemplifies the problem inherent in the use of laboratory animal models to develop dose-response estimates for humans. Substantial bias can be introduced easily into potency estimates by the improper selection of an animal.model. The example of the guinea pig and hamster demonstrates this point clearly. There is in excess of 3 orders of magnitude difference in sensitivity of these species to the acute toxicity of TCDD. Such disparity will have obvious implications in the regulatory process. Issues such as similarities or differences between laboratory animals and humans in such critical aspects of metabolism as biological half-life of TCDD, systems biochemistry, and physiology should drive the process of selecting study animals. All too frequently, the rat or mouse is selected as the animal of choice for no more scientifically-based a reason than convenience. G E N P 0 1 1414 784225 ChomRiakTM - A McLaren Company NOVEMBER 28, 1989 Page 7 3.0 DETERMINING ACCEPTABLE LEVELS CW EXPOSURE TO TCDD Critical to t.he development of meaningful acceptable levels of exposure to TCDD is the determination of appropriate carcinogenic risk estimates for this compound. As discussed in Section 2.0 of this report, studies of the most highly exposed human populations have failed to associate increased cancer rates with exposure to dioxin or dioxin-containing chemicals. In the absence of adequate dose-response data from epidemiologic studies, extrapolations from animal toxicity bioassays have been made in an attempt to characterize the relationship between TCDD uptake and an anticipated carcinogenic response in humans. Clearly, the assumptions used in extrapolating the dose-response curve generated from these experiments can have a significant impact on estimates of human risk associated with exposure to TCDD (Paustenbach et al., 1986; Sielken, 1987) . 3.1 Various Approaches Assessment of TCDD to the Carcinogenic Dose-Response A number of regulatory agencies in Western Europe and North America have estimated human exposure limits for TCDD based on the application of classical toxicological safety factors to either the no-observable-adverse-effect level (NOAEL) or to the lowest-observable-adverse-effect level (LOAEL) of exposure in rodents. The use of the classical safety factor approach is based on the weight of scientific evidence that an exposure threshold exists below which adverse effects will not occur (Shu et al., 1987; GanTox, 1989) . Using this methodology, a number of countries have developed lifetime ADls ranging between 1 and 10 picograms per kilogram of body weight per day (pg/kg-day) (Table 4). The Ontario Ministry of the Environment (Ontario, 1985) calculated a maximum acceptable daily intake for humans of 10 pg/kg-day based on a NOEL of 1,000 pg/kg-day and a safety factor of 100. The State Institute of National Health (SINH) in the Netherlands derived a maximum ADI of 4 pg/kg-day based on a NOEL of 1,000 pg/kg-day and a safety factor of 250 (van der Heijden et al., 1982). A range of ADIs of 1 to 10 pg/kg-day was developed by the Federal Republic of Germany based on a NOEL of 1,000 pg/kg-day and safety factors of 1,000 or 100 (NCASI, 1987). Recently, the United Kingdom derived a guideline value of 10 pg/kg-day for TCDD and its toxic equivalents as "a level which, when exceeded, should trigger investigation and appropriate measures to reduce environmental levels" (UK, 1989). The FDA, prior to their adoption of a linear interpolation model (FDA, 1983), had originally used a safety factor approach to support advisory levels for TCDD in fish. The agency (Cordle, 1983) calculated a TCDD exposure level of 13 pg/kg-day from consuming fish containing 25 ppt TCDD at the 99th percentile of U.S. fish consumption. The FDA noted that this exposure level was less than 1/70th of the animal NOEL of 1 ng/kg-day. This approach was used to support 25 ppt TCDD as a "safe" level in Great Lakes' fish. GENP 01141$ 784226 - ChemRlskTM - A McLaren Company NOVEMBER 28,1989 Page 7a Table 4. Maximum Acceptable Daily Intake (ADI) of TCDD Jurisdiction Safety Factor Max ADI (pg/kg-day) Reference Canada Denmark Federal Republic of Germany Netherlands 100 200 100-1000 250 United Kingdom Reproductive toxicity Carcinogenicity Immunotoxicity 100 1000 100 10 5 1-10 4 1 10 60 Ontario, 1988 N A S I1987 NCASI, 1987 van der Heijden et al,, 1982 UK, 1989 GENP 011416 784227 ChemRiakTM - A McLaren Company NOVEMBER 28, 1989 Page 8 A model developed by Moolgavkar, Venzon, and Knudson (M-V-K) (Moolgavkar and Venzon, 1979;-Moolgavkar and Knudson, 1981) has recently been applied to TCDD to estimate human cancer risk based on a tumor promotion mechaniam (Thoralund, 1987). The M-V-K model ia a two-atage biological model for carcinogenesis. Thoralund assumed that TCDD exerts it carcinogenic effect by increasing the net growth rate of preneoplaatic, initiated cells. Two forms of the M-V-K model, negative exponential and log-logistic, were used to extrapolate from observed animal liver tumor rates to expected human response rates. The results using the negative exponential model can be used to calculate a riskspecific dose of 6.0 pg/kg-day at a 1 x 1 0 "5 risk level (originally reported as 0.6 pg/kg-day at a 1 x 10" risk level) (EPA, 1988). Acceptable daily intakes and risk-specific doses developed for TCDD by various U.S. agencies and other countries are shown in Figure 3 for comparison purposes. A risk-specific dose developed using the M-V-K model is also included in Figure 3. It is clear that the application of the two different extrapolation procedures to the same animal cancer bioassay data results in markedly different exposure limits for TCDD. The approach taken by EPA is at the most extreme end of the range of acceptable intake levels proposed by the regulatory agencies of Western industrialized nations. 3.2 Traditional U.S. Agency Dose-Response Analyses The EPA and certain other federal policy and regulatory groups, however, employ a very different extrapolation procedure to the same animal bioassay data (EPA, 1985) . Through the U3e of the LMS model, human exposure levels at a selected probability of increased cancer ri3 k are extrapolated from rodent bioassay data. The use of this model is based on an assumption that there is no threshold for carcinogenesis; i.e., any dose, regardless of the quantity, poses some level of risk. The Office of Science and Technology Policy (OSTP) prepared a framework for U.S. regulatory agencies involved in the process of assessing human cancer risks from chemical exposures (OSTP, 1985) . OSTP proposed that "...a single long-term (animal) study may be utilized to obtain data -on chronic toxicity, on carcinogenic potential, and for carcinogenic risk assessment." The EPA risk assessment policies are consistent with the OSTP proposal; chronic bioassay data, moat commonly from rodents, are used to calculate upperbound estimates of the human carcinogenic risk associated with chemical exposure. The EPA (1985), the Centers for Disease Control (CDC) (Kimbrough et al., 1984), and the Food and Drug Administration (FDA, 1983) have developed estimates of cancer potency for TCDD and corresponding estimates of the riskspecific dose (RsD). By definition, an RsD is defined as that dose of a GENP 011417 784228 ChemRiskTM - A McLaren Company NOVEMBER 28,1989 Page 8a Figure 3. Risk-specific Doses (at 1QT9) and Acceptable Daily Intakes Calculated for TCDD by Various U.S. Agencies and Other Countries1 13 - - 12 - 11 - 10 - - 9-- 8-- - >N I 6 -- s 5 -- 4 -3 -- 2 1 0 a S o a *g 3 n C D C (b) Z7A CW oti CDC nu Su*Crfftofe.EiD O -- ftrn -- C H rt,O A A > H rM > h H a .|iD (k) i M O M f i N f i a i r f i a r ir~ nrj B im r^iiii^iiw nr no H nTat NnlfOyFKO) k rfItaT al-A lI CHk IX 41 1. U.S. agencies, including the EPA, have calculated andprentedRsDs at various level* of risk. For the nka ofconsistency to enhance the clarity of this document, the risk level of 1 a inn nnn hm tv ^n n*^A in rfw ^ p ,p GENP 011418 784229 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 9 compound which is associated with a given probability of a toxic effect in humans. This relationship is described by the following equation: _ _ Level of Risk RsD - _______________________ Cancer Potency Factor The EPA and CDC have used the linearized multistage model (Howe and Crump, 1902) to estimate the dose-response relationship for low-level human exposures based on data obtained from rodents exposed to high doses. A linear interpolation model (Gaylor and Kodell, 1980) has been employed by the FDA (1983) . The current EPA cancer potency estimate for TCDD is 1.56 x 10s (mg/kg-day) -1, based on the combined incidence of a number of tumor types (including, the liver, lung, and oropharynx) observed in female Sprague-Dawley rats from the Kociba et al. (1978) two-year bioassay. Using the upper confidence limit (UCL) of the cancer potency estimate derived from the multistage model, an RsD associated with a risk of 1 in 100, 000 can be calculated at 0.064 pg/kg/day. This estimate is the most conservative among the RsDs calculated by the EPA (1985), the CDC (Kimbrough et al., 1904), and the FDA (1983). When the CDC evaluated the bioassay data from NTP (1982) and Kociba et al. (1978)i they presented virtually safe doses (VSDs) involving a number of plausible scenarios. Rather than simply rely on one set of assumptions about the behavior of TCDD in rodents, the CDC adopted a range of assumptions from the most sensitive species/strain/sex/tumor type combination to the least sensitive species/strain/sex/tumor type combination (Kimbrough et al., 1984). The 95% lower confidence bounds of the VSDs that would correspond to a 1 x 10-5 risk level range from 0.28 to 14.28 pg/kg/day. [Note: the term VSD is synonymous with the term RsD which is currently used by the EPA.] The calculated cancer potency estimates corresponding to the VSDs reported by the CDC range from 7.0 x IQ2 to 3.6 x 104 (mg/kg-day)"1 . The FDA (1983) has suggested a cancer potency estimate of 1.75 x 104 (mg/kgday) -l, based on the incidence of hepatic tumors in female rats in the Kociba et al. (1978) study. Their estimate corresponds to an RsD associated with a risk of 1 x 10-s of 0,57 pg/kg/day (FDA, 1983) . The FDA analysis adopted a linear interpolation model (Gaylor and Kodell, 1980), which assumed a linear dose-response relationship between the origin and the upper 95% confidence level on the doses. Cancer potency figures and corresponding risk-specific doses for TCDD are summarized in Table 5. Table 5 also includes the revised cancer potency estimate recently proposed by the EPA. Risk-specific doses are shown corresponding to incremental cancer risks of 1 in 100,000 (1 x 1 0 "5) . It is evident that the three agencies' cancer potency estimates differ considerably. The EPA (1985) potency estimate exceeds the most conservative CDC estimate by a factor of 4 and exceeds the FDA estimate by a factor of 9. Differences among the three agencies' cancer potency estimates are due to differences in GENP 011419 784230 ChemRiskTM - A McLaren Company NOVEMBER 2S, 1989 Page 9a Table 5. U.S. Agency Risk-specific Doses for 2,3,7,8-TCDD* Cancer Risk-specific Dose Potency (pg/kg-day) 1 (mg/kg-day)-1 1 x 10'* Reference EPA 1.56 x 103 0.064 EPA, 1985 EPA* 1.0 x 10* 1 EPA, 1988 CDC 3.6 x 104 to 7 x 10a .28-14.28 Kimbrough et al., 1984 FDA 1.75 x lO 4 57 FDA, 1983 i. U.S. agencies, including the EPA, have calculated and presented RsDs at various levels of risk. For die sake of consistency and to enhance the clarity of this document, the risk level of 1 in 100,000 has been used in the pre sentation of RsD estimates. b. Recommended in EPA (1988) draft document as revised RsD for TCDD. GENP 011420 784231 CheinRiakTM - a McLaren Company NOVEMBER 28, 1988 Page 10 data Interpretation and modeling analysis. Due to the size difference between humans and rats, an adjustment of the animal dose is necessary to properly predict the possible human response. It should be pointed out that the regulatory agencies do not agree on the appropriate method for making this correction. Sielken (1987, 1988) has challenged the use of the LMS model for predicting human cancer risk and, in particular, has challenged the appropriateness of adopting evidence from the Kociba rodent bioassay. Sielken concluded that when the multistage model is fitted to the animal data, trade-offs inherent in curve fitting may lead to questionable fits in the low-dose region. Compared to the response rates observed in the bioassay, the tumor response rates estimated by the LMS model are too large at the lowest nonzero experimental dose level and too small at the intermediate dose level. Sielken's analysis clearly demonstrated how the sensitivity of the low-dose extrapolation from rats to .humans is dependent on the assumptions employed in the multistage model. His analysis indicated that the extrapolation approaches which have been used by various agencies may considerably overestimate the actual cancer potency of TCDD. The statistical uncertainties of the LMS model described by Sielken (1987, 1988) have been disputed by Crump (1988) . Crump did not agree that the chisquare goodness-of-fit teat is an inadequate statistical measure of the multistage model. According to the author, the chi-square method is a universally accepted and widely used statistical test. Crump also did not agree with Sielken (1987, 1988) that the LMS model cannot handle saturationtype phenomenon at high dose levels and non-linear behavior at low doses. When the high TCDD dose group is omitted from the model, as Sielken (1987, 1988) demonstrates, Crump beiieved that the remaining data offer little assurance that the true dose-response relationship is non-linear; the gap between the lowest and middle doses is too large to discern the true shape of the dose-response curve for TCDD. Recently, a draft report prepared by a EPA Workgroup reexamined the hazard identification and dose-response assessment regarding the potential carcinogenicity of TCDD (EPA, 1988) . The Workgroup recommended on the basis of the weight of evidence that a revised RsD of 0.1 pg/kg-day (100 fg/kg-day) be adopted as the dose most likely to be associated with an increased lifetime cancer risk of 1 x 10"6 . This corresponds to a cancer potency estimate of 1 x 10* (mg/kg-day) -i. This potency estimate is 16 times lower than the cancer potency adopted by the EPA in 1985 (Table 5) . The EPA's Science Advisory BOard recently has reviewed the draft report. However, a formal statement of their conclusions has not been released to the public (Inside EPA, 1988) . The Workgroup assessed several potential approaches for estimating the cancer potency of TCDD and concluded that none of the available models, including the multistage model, adequately describe the carcinogenic behavior of TCDD at low doses. The EPA Workgroup encouraged the types of analyses generated by Sielken and the use of the M-V-K model for TCDD. However, the Workgroup did not consider these approaches sufficiently refined or accepted in the scientific community to warrant their consideration as the basis for GENP 011421 784232 ChemRiakTM - A McLaren Company NOVEMBER 28, 1989 Page 11 recommending changes in the current cancer potency or RsD estimates for TCDD. They stated that while arguments could be made that a threshold for the carcinogenicity of TCDD may exist, the evidence for such a conclusion is not compelling. The EPA Workgroup noted, however, that the EPA (1985) analysis based on the linearized multistage model is likely to have led to an overestimation of the carcinogenic risk of TCDD. The weight of evidence indicated that a more appropriate upper-bound estimate would result in an unspecified reduction in the cancer potency of TCDD. The Workgroup noted that all of the cancer potency and corresponding risk-specific dose estimates made by various federal and state agencies, including the EPA, FDA, CDC, and* the State of California, are arguably of equal scientific merit at the present time (EPA, 1988) . There appears to be little, if any, scientific basis for using a linear nonthreshold model for low-dose risk extrapolation for TCDD. According to the EPA Guidelines for Carcinogen Risk Assessment (EPA, 1986a), " . . . when pharmacokinetics or metabolism data are available, or when other substantial evidence on the mechanistic aspects of the carcinogenesis process exists, a low-dose extrapolation model other than the linearized multistage procedure might be considered more appropriate on biological grounds." 3.3 Determining an Appropriate Level o Risk It is a common misperception within risk assessment that all occupational and environmental regulations have as their goal a theoretical maximum cancer risk of one in one million (i.e., 1 in 1,000, 000 or 1 x 10"6) . In the U.S., regulatory agencies frequently adopt the cancer risk guideline of one in one million as a negligible or da minimis risk level when very large populations are likely to be exposed to a suspect carcinogen. Travis et al. (1987) demonstrated that this view is not compatible with past regulatory practice. In a retrospective examination of the level of risk which triggered regulatory action in 132 federal decisions, the authors considered three measures of risk: individual risk (an upper-limit estimate of the probability that the most highly exposed individual in a population will develop cancer as a result of a lifetime exposure), the size of the population exposed, and population risk (an upper-limit estimate of the number of additional cases of cancer in the exposed population). Travis et al. (1987) found that for exposures resulting in a small-population risk, the de manifestid level was approximately 4 in 1,000 (4 x 10~3), i.e. the level of risk above which agencies almost always acted to reduce risk. For large-population risks (the entire U.S. population) the d e "manifest is level dropped to about 3 in 10,000 (3 x 10*4) . Risks that are so low that agencies almost never act to reduce them are termed de minimis risks. For effects on small populations, regulatory action was never taken for individual risk levels below 1 in 10,000 (1 x 10"4) . For large-population effects, the GENP 011422 784233 ChamRiakTM - A McLaren Company NOVEMBER 28, 1989 Page 12 de minimis risk level dropped to 1 in 1,000,000 (1 x 10*6) . The review by Travis et al. (1987) shows that regulatory agencies have found risks far in excess of 1 in 1 ,000,000 (1 x 10*6) to be insignificant and of no concern if the size of the population is small. Farber (1989) proposes that the decisions regarding the selection of the appropriate level of incremental risk should be driven by the existing level of "background" risk. He defines background risk as that level of risk which is present regardless of whether the source of the "incremental" risk is present or not. In a real population , this "background" risk translates into incidences of disease or death which, when monitored in a population over time, are variable. The inherent variance associated with background risk estimates needs to be considered when selecting an appropriate level of incremental risk. Farber mathematically modeled the relationship between background and incremental population risk. His analysis indicates that incremental risks which are smaller than the variance estimate for the background risk do not alter the probability of an individual contracting a disease or increase case incidence. He further suggests that when the ratio of background risk/incremental risk increases beyond 1 0 0 , the additional risk no longer affects the case incidence and may be considered a "negligible" risk. Farber (1989) concluded that the use of a risk level smaller than 1 x 10-5 for deriving acceptable levels of daily exposure is almost always mathematically meaningless and unsound. The use of such a small risk level would require the estimation of small differences between two large numbers, neither of which is known with the required degree of precision. "Background" risk alone and the "background" risk plu3 "incremental" risk do not exist as precise quantities; thus, any discussion of incremental or excess risk is highly tenuous at risk levels below 1 x 1 (M or 1 x 10"5 . State regulatory agencies have not adopted uniformly a one in one million (1 x 10*6) risk criterion in making environmental and occupational decisions. The State of Maine Department of Human Services (DHS) uses a lifetime risk of one in one hundred thousand .(1 x 10"5) as a reference for nonthreshold (carcinogenic) effects in its risk management decisions regarding exposures to environmental contaminants (Maine DHS, 1988). Similarly, a lifetime incremental cancer risk of one in one hundred thousand is used by the Commonwealth of Massachusetts as a cancer risk limit for exposures to substances in more than one medium at hazardous waste disposal sites (Mass DEQE, 1988) . This risk limit represents the total cancer risk at the site associated with exposure to multiple chemicals in all contaminated media. The State of California has also established a level of risk of one in one hundred thousand (1 x 10 *5) for use in determining levels of chemicals and exposures that pose no significant risks of cancer under the Safe Drinking Hater and Toxic Enforcement Act of 1986 (Proposition 65) (California, 1986). Additionally, the states of Maryland, Michigan, Minnesota, Ohio, Virginia, and Wisconsin have employed the 1 x 10*5 level of risk in their risk management decisions (individual state offices of environmental quality, personal connunication). GENP 011423 784234 ChomRiskTM - A McLaren Company NOVEMBER 28, 198 9 Page 13 Workplace air standards developed by the Occupational Safety and Health Administration (OSHA) typically reflect theoretical ri3 ks of one in one thousand (1 x 10~3) or greater (Rodricks et al., 1987). When considering these limits it is interesting to note that many common human activities entail annual risks greatly in excess of one in one million (Tables 6 and 7). These have been discussed by Grover Wrenn (1986), former director of Federal compliance and State Programs at OSHA, as follows: "Examination of the risk3 of common human activities demonstrates . . . a lifetime risk of 1 in 100,000 or more is within the realm of, or orders of magnitude below, everyday, risks that generally do not cause undue concern. These are risks that people, while they are aware of them and may have some concern or fear over them, do not in general alter their behavior to avoid. The risks from many activities greatly exceed the level of one in 100,000. In comparison to these background risks of "everyday activities," a lifetime risk of 1 in 100,000 is relatively small. Accordingly, regulatory action will not generally be justifiable unless risks are substantially higher than this 1 in 100,000 "benchmark"." Ultimately, the selection of an acceptable and de minimis risk level is a policy decision in which both cost3 and benefits of anticipated courses of action should be thoroughly evaluated. However, actuarial data and risk estimates of common human activities, regulatory precedence, and the relationship between the magnitude and variance of background and incremental risk estimates all provide compelling support for the adoption of the de minimis risk level of 1 x IQ-5 for regulatory purposes. G E N P 011424 784235 ChemRIskTM A McLaren Company NOVEMBER 28,1989 Page-13a Table 6. Activities Associated with 1 x 1(T* Increased Risk of Death in Any Year . Activity Cause Smoking 1.4 cigarettes Drinking 0.5 L wine Eating 40 tablespoons o f peanut butter Eating 100 charcoal-broiled steaks Bicycling 10 miles Flying (jet) 1000 miles One chest x-ray Cancer, heart disease Cirrhosis Liber cancer due to aflatoxin Cancer from benzopyrene Accident Accident Cancer due to irradiation Adapted from Allman, 198S QBHp 011425 784236 ChemRlskTM A McLaren Company NOVEMBER 28,1989 Page 13b Table 7. Comparison of Risks from Selected Activities on a Per Capita Basis Activity Smoking cancer only all effects Scuba Diving Motor Vehicle Accident Boating Hunting Swimming Lightning Average Annual per Capita Risk of Mortality 1 x 10'1 3 x 10'J 4 x 10*4 2 x 10"4 5 x Iff3 3 x 10's 2 x 1Or* 5 x 10*7 A dapted from Crouch and Wilson (1982) GENP 011426 784237 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 14 4.0 SENSITIVITY ANALYSIS FOR DIOXIN RISK ASSESSMENT 4.1 Overview In the absence of adequate human epidemiological evidence, laboratory animals typically serve as models for the study of the toxicological effects of chemicals in humans, based on the assumption that the extrapolation of biological data from animals to humans is valid, at least for some physiological parameters. A critical component of the quantitative risk assessment process, interspecies extrapolation, assumes that a reasonably stable relationship exists between the potency of a chemical in laboratory animals and in humans. Moreover, it assumes that the parameters of this relationship have been adequately estimated and characterized. In the case of TCDD, low-do3e risk estimates derived by the EPA (1985) through the application of the LMS model to the incidence of hepatic lesions observed in the Kociba et al. (1978) bioassay are very sensitive to the modeling assumptions chosen by the EPA to extrapolate cancer risks from Sprague-Dawley rats to humans. Each of these assumptions contributes to the uncertainties expressed by a number of researchers (Portier et al., 1984; Kimbrough, 1984; Sielken 1987, 1988; Paustenbach et al., 1986; Shu et al., 1987; CanTox, 1989; Farber, 1969; Bonvalot et al., 1989) concerning the EPA's reliance on the LMS model to simulate the dose-response behavior of TCDD in laboratory animals and/or the appropriateness of an RsD of 0.064 pg/kg-day at a 1 x 10"5 risk level. In order to understand these uncertainties, a sensitivity analysis of the doseresponse behavior of TCDD observed in the Kociba et al. (1978) study was conducted. It was hypothesized that each of the following considerations in the dose-response assessment of TCDD might have a substantial impact upon the range of plausible risk estimates predicted through the use of the LMS model: " The histopathological interpretation of the observed tumor incidence among female Sprague-Dawley rats exposed to various doses of TCDD. Use of the term "neoplastic nodules'* in histopathological examinations of laboratory animals to describe tissue anomalies. The effect of high spontaneous rates of hepatocellular lesions in rodents on the determination of chemically-induced carcinogenic responses in the livers of Sprague-Dawley rats. Consideration of hepatocellular carcinomas as the most appropriate primary carcinogenic response in laboratory rodent bioassays. Low-dose risk estimates based on the incidence of hepatocellular carcinomas observed during the Kociba et al. (1978) bioassay. The use of Maximum Likelihood Estimates (MLE) versus the 95% Lower Confidence Limit (LCL) of the human RsD. GENP 011427 784238 ChemRiskTM - X McLaren Company NOVEMBER 28, 198 9 Page 15 The use of a body weight correction factor in the rat-to-human extrapolation procedure rather than the use of a surface area correction factor. 4.2 Histopathological Interpretation of the Kociba at al. (1978) Bioassay Among the several animal bioassays described by the EPA in their 1985 document entitled "Health Assessment Document for Polychlorinated Dibenzo-p-dioxins", the two-year study conducted by Kociba et al. (1978) has been cited as the primary evidence supporting the carcinogenicity of TCDD in laboratory animals. The EPA (1985) also cited a two-year bioassay conducted by the National Toxicology Program in 1982 as additional evidence supporting the findings of the Kociba et al. (1978) study. In the Kociba et al. (1978) study, groups of 50 Sprague-Dawley rats, Spartan substrain, of each sex were maintained for up to 24 months on diets containing 1,000, 10,000, and> 100, 000 pg/kg-day of 2,3,7,8 -TCDD. A population of 86 animals of each sex was maintained as study controls. Gross and histopathological examinations were performed on all animals. Tissues considered as possible target organs were examined in most animals from each treatment group and included liver, lungs, kidneys, urinary bladder, tongue, brain, testes/ovaries, and prostate/uterus. High early mortality was observed in all treatment groups in the Kociba et al. (1978) study. A substantial proportion of the animals in the study, including those in the control groups, died before the end of the two-year period (78% to 92% among males and 68% to 92% among females) (Kimbrough et al., 1984; Portier et al., 1984) , However, early mortality was statistically significant compared to controls only among female animals in the highest dose-group (100,000 pg/kg-day) (Kociba et al., 1978; EPA, 1985). Progressive mortality among female animals in the high-dose group began as early as the 12th month and reached 50% mortality by the 21st month of the study (EPA, 1985) . The mortality of only the group of males given the lowest dose (1,000 pg/kg-day) was significantly-different from control animals (Kociba et al., 1978). Fortyfour percent of the male animals in that group died within the first 18 months of the study (EPA, 1985). The EPA (1985) noted that the principal impact of the high early mortality rate observed in the Kociba et al. (1978) study was an overall reduction in the sensitivity of the bioassay because of a decrease in the number of animals at risk during the time of expected tumor formation. Farber (1989) and CanTox (1989) regarded the high mortality rate, coupled with depressed hematologic parameters, increased urinary levels of porphyrins, decreased weight gain, and increased activity of serum enzymes, as indicative of severe liver toxicity in rats in the highest dose-group (100,000 pg/kgday) . Similar effects were also observed in the NT? (1982) study (CanTox, GENP 011428 784239 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 16 1989). These researchers have suggested that hepatocellular toxicity would result in.marked increases in cell proliferation, which could significantly enhance the susceptibility of hepatocytes to both tumor initiators and promotors. A summary of the principal histopathological findings of the Kociba et al. (1978) study are presented in Table 8 . The lesions that are listed are those which were statistically different from control levels for at least one treatment dose and in one sex. The incidences of lesions observed among exposed animals that were found to be statistically depressed below those of control animals are listed in Table 9. In general, the lesions reduced by TCDD treatment were those that typically occur at high spontaneous rates in the particular strain of Sprague-Dawley rat employed in the bioassay (Kociba, 1984) . - These included neoplasms of the pancreas', uterus, mammary glands, and pituitary glands. Kociba (1984) found it difficult to ascertain whether the reduction in these incidences was associated with exposure to TCDD or related to nontreatment factors (e.g., diet, age, cage conditions) . Several researchers believe that the reduction in the incidence of various tissue lesions is associated with hormonal alterations or with the mediating role of one or more thyroid hormones (Potter et al., 1983; Rozman et al., 1984; Holder and Menzel, 1989; CanTox, 1989). The lesions observed among animals exposed to 2,3,7,B-TCDD that were found to be statistically increased above control levels are listed in Table 10. In the highest dose groups, 2,3,7,8 -TCDD induced statistically significant increases in stratified squamous cell carcinomas of the hard palate and/or nasal turbinates in both males and females, squamous cell carcinomas of the tongue in males, and keratinizing squamous cell carcinomas of the lungs, hepatocellular carcinomas and hyperplastic nodules in females. Kociba (1984) has argued that lesions in tissues other than the liver (lung, hard palate, and nasal turbinates) were not systemic and probably resulted from the incidental inhalation of some of the TCDD-treated food or with prolonged direct contact of the TCDD-treated food with the mouth and respiratory tract, furthermore, the pathological data presented in Kociba et al. (1978) indicate that hepatic lesions were observed in all animals that had one or more tumors at these other locations. The incidences of the principal lesions reported in female Sprague-Dawley rats by Kociba et al. (1978) that are considered by the EPA (1985) as the quantitative basis for TCDD risk extrapolations are presented in Table 11. The description of hyperplastic nodules used by Kociba et al. in 1978 to characterize this type of lesion was referred to by the EPA in 1985 as neoplastic nodules (see section 4.3). The total combined incidences of lesions in the liver, lung, hard palate, or nasal turbinate were revised by EPA (1985) to adjust for high early mortality in the first year of the study. Since hepatic lesions were observed in all animals that had one or more lesions of the lung, hard palate, or nasal turbinates, the total combined incidence of lesions presented in Table 11 represent the response frequencies GENP 011429 784240 ChemRlskTM - A McLaren Company NOVEMBER 28,1989 Page 16a Table 8. Sum m ary of Neoplastic Lesions Produced by 2,3,7,8-TCDD in Sprague Dawfcy Rats, Spartan Substrain that are Statistically Significant in a t Least One Sex Treatment Dose (^g/kg/day) Males 0 0.001 0.01 0.1 Females 0 0.001 0DI 0.1 Number of Animals Tested 85 50 50 50 86 50 50 49 Hepatocellular Hyperplastic nodules Hepatocellular carcinoma 60 20 3 0 2 1 8 3 18* 23* 1 0 2 11* Stratified squamous cell carcinoma of hard palate or nasal turbinate Keratinizing squamous cell carcinoma of lung Benign tumor of uterus 00 0 4* 00 -- 0 - 1 - 00 I 4* 00 28 12 0 11 7* 7* Subcutaneous fibroma/ fibroadenoma/lipoma 10 1* 5 6 11 0 Benign mammary gland neoplasm 00 0 1 73 35 36 Mammary gland carcinoma 00 0 0 84 4 Stratified squamous cell carcinoma of tongue 0 1 1 3* 10 0 Pituitary adenoma 26 6 11 13 43 18 13 Acinar adenoma of pancreas 14 7 5 2* 010 Adenoma of adrenal cortex 00 2 S* 96 2 Piieochromocytoma of adrenal 28 6 10 4* 72 1 Source: Kociba et *1_ 1978. * Denotes a statistically ligniant difference (p S 0.05) from control by the Fisber Exact Test. 0 24* 0* 2 12* 1 5 3 GENP 011430 784241 ChemRlskTM - A McLaren Company NOVEMBER 28,1989 Page 16b Table 9. Tissue Lesions Observed by Kociba et al. (1978) Among Sprague-Dawley Rats Exposed to 2,3,7,8'TCDD th a t W ere Statistically Depress! Below Controls Lesion Benign tumor of uterus Sex Female Treatment Group (pg/kg/day) 0.1 Benign mammary gland neoplasms Female 0.1 Mammary gland carcinomas Female 0.1 Pituitary adenomas Female 0.1 Subcutaneous fibroadenomas, fibromas, and lipomas Male 0.001 Acinar adenomas of the pancreas Male 0.1 Pheochromocytomas Souice: EPA (1984). Male 0.1 G EN P011431 784242 ChemRIskTM A McLaren Company NOVEMBER 28,1989 Page 16c Table 10. Tissue Lesions Observed by Kociba et al. (1978) Among Sprague>Dawley Rats Exposed to 2,3,7,8-TCDD that Were Statistically Increased Above Controls Lesion Treatment Group Sex (Jig/kg/day) Hepaiocellular hyperplastic nodules Female 0.01,0.1 Hepatocellular carcinoma Female 0.1 Keratinizing squamous cell carcinoma of lung Female 0.1 Stratified squamous cell carcinoma of palate or nasal turbinate Male & Female 0.1 Stratified squamous cell carcinoma of tongue Male 0.1 Adenoma of adrenal cortex Source: EPA (1984). Male 0.1 GENP 011432 784243 ChemRIskTM - A McLaren Company NOVEMBER 28,1989 Page 16d Table 11. Principal Lesions In Female Sprague-Daw ler R ats Exposed to 2,3,7,8-TCDD Reported by Kociba et al. (1978) Treatment Dose (jig/kg/day) Hepatocellular hyperplastic nodules 0 0.001 8/86(9%)" 3/50(6%) 0.01 0.1 18/50(36%)* 23/49(48%)* Hepatocellular carcinoma 1/86(1%) 0/50 (0%) 2/50(4%) 11/49(22%)* Keratinizing squamous cell carcinoma of lung 0/86(0%) 0/50 (0%) 0/49 (0%) 7/49 (14%)* Stratified squamous cell carcinoma of hard-palate or nasal turbinates (revised diagnosis in EPAt 1985) 1/54(2%) 0/30(0% ) 1/27 (4%) 5/24(21%)* Total combined incidence* Giver, lung, hard palate or nasal turbinate) 9/85 (11%) 3/48 (6%) 18/48 (37%)* 34/40(85%)* Source: Kociba et el. (1978) and EPA (1985). * Denotes a statistically significant difiaence (p<0.05) from control by the Fischer Exact Test. a. Number of rcsponses/numberof animals examined (percent response). b. The Kociba et al. (1978) datasrt was adjusted by the EPA (1985) *o eliminate the first year's data to account for high early mortality in the high-dose group. The EPA (1985) replaced the term "hyperplastic nodules" from Kociba et aL (1978) with "neoplasticnodules". These data were employed by EPA (1985) in an LMS analysis o f the Kociba et aL (1978) bioassay. 784244 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 17 adopted by the EPA in 1965 for their dose-response assessment of TCDD using an LMS model. The histopathology of the Kociba et al. (1978) 3tudy was reviewed by Or. Robert Squire, pathologist at the Johns Hopkins University Medical School and consultant to EPA's Carcinogen Assessment Group (CAG). The results of Dr. Squire's interpretation of the Kociba dataset were reported in EPA (1985) and are reproduced in Table 12. Similar to the EPA's (1985) interpretation, the total combined incidences of lesions in the liver, lung, hard palate, or nasal turbinates also were derived from an elimination of the first year of data to adjust for the high early mortality of female rats in the highest dose group. In the NTP (1982) carcinogenicity bioassay in rats and mice, considered by the EPA (1984, 1985) as significant support for the findings in Kociba et al. (197 8) , the liver also appears to be the predominant target organ of TCDD carcinogenicity. A summary of hepatic lesions observed in Osborne-Mendel rat3 and B6C3F1 mice exposed to 2,3,7,8-TCDD in the NTP (1982) study is presented in Table 13. Fifty rats of each sex and 50 male mice were exposed by gavage to 0, 0.01, 0.05, and 0.5 jig/kg-week of TCDD for 104 weeks. Animals were dosed twice each week during the study. Fifty female mice' were similarly exposed to doses of 0, 0.04, 0.2, and 2.0 |lg/kg-week of TCDD. A significant increase in follicular cell adenomas or carcinomas of the thyroid were elevated in male rats, and both neoplastic nodules and hepatocellular carcinomas were elevated in female rats in the highest dose groups. Mice of both sexes had statistically significant elevated levels of hepatocellular carcinomas at the highest treatment doses. 4.3 Interpretation of Hyperplastic Nodules and Neoplastic Nodules During chemical hepatocarcinogenesis in experimental animals, various cellular changes also are observed either in conjunction with carcinomas or at lower doses than those at which carcinomas occur. Since approximately 1975, two lesions of hepatocytes have been generally related to cancer development: altered foci and neoplastic nodules. The characterization of these lesions among pathologists has involved a range of adjectives, namely "degenerative1*, "regenerative", "hyperplastic", "preneoplastic", and "neoplastic" (Bannasch et al., 1982) . The changes in the terminology of rat liver lesions over time (Figure 4) reflects the difficulties in the understanding of the biological behavior of these lesions and their significance in hepatocellular carcinogenesis. According to the EPA (1986b), liver foci, or areas of altered hepatocytes, are generally considered to be the first indications of abnormalities in the rat liver, followed by the appearance of neoplastic nodules, and finally carcinomas. Over the last decade, this sequence of events in hepatocarcinogenesis suggested that foci were a precursor stage in the development of neoplastic nodules, which in turn were precursors of carcinomas. However, under certain circumstances, foci and nodules may regress to a condition that is morphologically indistinguishable from the GENP 011434 784245 ChemRIskTM - A McLaren Company NOVEMBER 28,1989 Page 17a Figure 4. Terminology of R at Livor Lesions Pre-1975 Preneoplastic____________ Hyperplastic Foci Basophilic Hyperplasia Enzyme-Deficient Islands \ Hyperplastic Nodule Nodular Hyperplasia Neoplastic Benign Hepatoma Hepatic Cell Adenoma Malignant Hepatom Hepatic Cell Carcinoma Trabecular Carcinoma Source EPA (1986b). Since 1975 Foci of Cellular Alteration Nodule XX Hepatocellular Carcinoma 1985 (NTP) Foci of Cellular Alteration Hyperplasia Hepatocellular ' Adenoma Hepatocellular Carcinoma G E N P011435 784246 ChemRIskTM A McLaren Company NOVEMBER 28,1989 Page 17b Table 12. HIstopathological Interpretation by Squire (EPA, 1985) of Prindpal Lesions in Fem ale Sprague-Dawley Rats Exposed to 2,3,7,8-TCDD in the Kociba et al. (1978) Bioassay Treatment Dose (pg/kg/day) 0 0.001 0.01 0.1 Keratinizing squamous cell carcinoma of lung 0/86(0% / 0/50(0%) 0/49(0%) 8/47 (17%)* Stratified squamous cell carcinoma of hard palate or nflgai turbinates 0/54(0%) 0/30(0%) 1/27(4%) 5/22(23%)* Liver neoplastic nodules* and hepatocellular carcinoma 16/86 (19%) 8/50 (16%) 27/50 (54%)* 33/47 (70%)* Total combined incidence (liver, lung, hard palate or nasal turbinates)* 16/85(19%) 8/48 (17%) 27/48(57%)* 34/40(85%)* Source; EPA (1985). * Denotes a statistically significant difference (P<0.05) from control by the Fischer Exact Test. a. Numba'o f responsea/numba' of animals examined (percent response). b. The term "neoplastic nodules' was adopted by Dr. Squire and the EPA (1985) and was not used by Kociba et aL (1978). These data were not reported separately in EPA (1985). e. These data were employed by EPA, 1985 in an LMS analysis of Squire's interpretation of Kociba et aL (1978). GENE 011436 784247 ChemRlskTM - A McLaren Company NOVEMBER 28,1989 Page 17c Table 13. Summary of Hepatic Lesions Observed in Osborne-Mendel Rats and B6C3F1 Mice Exposed to 2^,7,8-TCDD in the NTP (1982) Bioassay Treatment Dose (p.g/kg/week) Males Osbome-Mendel Rais Females 0 0.01 0.05 0.5 0 0.01 0.05 05 Number of Animals Tested Neoplastic nodule 74 50 50 50 0(0%) 0(0%) 0(0%) 3(6%) 75 5(7% ) 49 1(2%) 50 3(6%) 49 12 (24%)* Hepatocellular carcinoma 0(0% ) 0(0% ) 0(0%) 1(2%) 0(0% ) 0(0% ) 0(0%) 3(6%) Toxic Hepatitis* 0(0% ) 1(2%) 0(0%) 14(28%) 0(0% ) 0(0% ) 1(2%) 32(65%) Treatment Dose (pg/kg/week) 0 Number of Animals Tested 73 Hepatocellular carcinoma 8(11%) Males 0.01 0.05 B6C3F1 Mice 0.5 0 49 49 50 9 (18%) 8(16%) 17(34%)* 73 1(1%) Females 0.04 0.2 50 2(4% ) 48 2(4%) 2.0 47 6(13%)' 1 o Hepatocellular adenoma 7(10% ) 3(6%) 5(10%) 2(3% ) 4(8% ) 4(8%) 5(11%) Toxic Hepatitis 1(1%) 5(10%) 3(6%) 44 (88%) 0(0% ) Source: NTP (1982). * Denotes a statistically significant difference (p<0,05) fromcontrol by theFisher Exact Test, a. Identified by NTP (1982) to describe live* toxicity. 1 (2%) 2(4%) 34(73%) -ENP 01143=7 784248 ChemRiakTM - A McLaren Company NOVEMBBR 28, 1989 Page 18 surrounding hepatic tissue. Thus, an alternative interpretation is that foci and so-called "neoplastic nodules" may be independent of the development of hepatocellular carcinomas. Based on a review of a number of hepatocarcinogenic studies in laboratory rodents, the EPA (1986b) and the NTP (McConnell et al., 1988) observed that under certain experimental conditions, early proliferative lesions (i.e., foci and nodules) that have been induced by hepatocarcinogens may regress following removal of the carcinogenic agent. The EPA concluded that, to the extent that proliferative lesions regress, there may be some undulation of the potential hazard for cancer development. Nonetheless, there i3 no certain scientific evidence or consensus among pathologists regarding the biological significance of these liver nodules or their interpretation in rodent carcinogenesis studies. There was no evidence in the Kociba et al. (1978) study to indicate that a hyperplastic nodule would likely progress to a hepatocellular adenoma and then to a hepatocellular carcinoma. Kociba et al. (1978) identified and reported the occurrence of adenomas in other organs and tissues, while no adenomas were reported in the liver, clearly indicating that the observed lesions were considered to be a less serious form of hepatocellular proliferative lesion, i.e. hyperplasia. Furthermore, Kociba et al. (1978) reported that there was no evidence of metastasis of any of these hyperplastic nodules. "Neoplastic nodules", the term used by both the USEPA (1985) and Squire (EPA, 1985) in their histopathological interpretations of the Kociba et al. (1978) study, represents terminology from a different morphological classification system developed by Squire and another researcher. Its use by Squire (EPA, 1985) does not provide convincing evidence that the hyperplasia was actually a more serious condition. The classification system of Squire and Levitt (1975) failed to define the exact position and significance of the "neoplastic nodule", which included both hyperplasia and more serious lesions (EPA, 1986b). The National Toxicology Program (NTP) instituted a change in the classification of rat liver proliferative lesions in 1986 and no longer uses the scheme of Squire and Leavitt (1975) in histopathological evaluations (Maronpot et al., 1986; McConnell et al., 1988). The KTP retained the terms foci and hepatocellular carcinoma but replaced neoplastic nodule with two terms; hyperplasia and hepatocellular adenoma (Figure 4) . Hyperplasia is characterized by a mild compression of surrounding hepatic tissue that is not observed in the focus of cellular alteration. Hyperplasia is further viewed as the result of cell injury and regeneration. In contrast, hepatocellular adenoma is indicated by clear differentiation of cells from the surrounding normal hepatic tissue, including tissue compression and the loss of the normal lobular architecture of the liver. Some researchers, have reported spontaneous incidences of hepatocellular proliferative lesions in various strains of rats, including Fischer-344, Sprague-Dawley, and Osborne-Mendel strains (Hollander and Burek, 1978; Ogava ;GENP 011438 784249 ChemRiskTM - X McLaren Company NOVEMBER 28, 1989 Page 19 __ et al., 1981; Schulte-Hermann and Parzefall, 1981; Bannasch et al., 1982; Schulte-Hermann et al., 1983; EPA, 1986b). These studies indicate that the high incidence o hyperplastic liver nodules in untreated, ageing rats may often preclude the distinction between chemical treatment and nontreatment effects in the liver. For example, Schulte-Hermann et al. (1983) demonstrated that a significant incidence of hyperplastic liver nodules are found in the livers of untreated rats. The incidence increased to 100% over the course of the animals' twoyear average lifetime. The size of these lesions also increased with age. In addition, these hyperplastic nodules showed similar characteristics to those observed in rats treated with known liver carcinogens. Schulte-Hermann et al. (1983) have concluded that the high incidence of hyperplastic liver nodules in untreated aged rats precludes the distinction between chemical promoters and initiating agents during long-term carcinogenicity bioassays in laboratory rodents, particularly when liver tumor incidence data are the primary criteria. > The use of rodents as a model for liver cancer in humans and, in particular, the consideration of the incidence of hepatocellular neoplastic nodules, may considerably overestimate the potential risks associated with exposure'to TCDD (CanTox, 1989; Sielken, 1987; Shu et al., 1967). The combined incidence of hepatocellular neoplastic nodules or carcinomas in aged rats is commonly as high as 18 to 20% (Sielken, 1987) but is relatively rare in the U.S. human population (less than 5 per 100,000, or 0.005%; ACS, 1989). In a 198 6 review of proliferative hepatocellular lesions in rats and their use in risk assessment, the EPA acknowledged that the exact contribution of neoplastic nodules to the overall incidence of hepatocellular tumors in the rat remains unclear. Accordingly, the EPA adopted the following position: "Determination of carcinogenic hazard will be based upon consideration of the incidence of hepatocellular carcinoma alone, neoplastic nodule alone, and a combination of carcinoma and nodule. The range of responses could vary from cases where there are very significant increases in the incidence of carcinomas backed up by increases of nodules, through situations where only the combined frequency of animals with carcinomas or nodules is significant, to cases where increases are limited to nodules alone. (i) At one end, responses showing an overwhelming increase in carcinomas alone will be interpreted as providing sufficient evidence of animal carcinogenicity when other criteria for the sufficient category are met. In the absence of human evidence and supporting information, such evidence will generally be given a weight-of-evidence designation of Probable Human Carcinogen (Group B2), and a quantitative risk assessment is appropriate. 784250 GENP 011439 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 20 (ii) At the other end, where increases in lesions and statistical significance are restricted to neoplastic nodules alone, such data will be interpreted as only limited evidence of animal carcinogenicity. When neither human evidence nor supporting information accompany the increase in nodules, the weight-ofevidence designation is Possible Human Carcinogen (Group C), but quantitative assessment is not warranted. (iii) Situations within the two extremes will be evaluated on a case-by-case basis.... In cases where quantitation is performed, methoda of expressing the possible uncertainty in the estimated risks should be explored (e.g., expressing risk as a range between that determined from extrapolation of carcinoma incidence alone and that from a combination of carcinomas and no.dules) ." At a minimum, the EPA in their 1988 draft document (EPA, 196 8) should have reported a range of risk for TCDD in order to convey the uncertainties associated with the quantification of the carcinogenic response in female Sprague-Oawley rats by Kociba et al. (1978) . In addition to presenting an RsD based on an analysis of carcinomas and nodules combined, the EPA should have presented an alternative RsD based on the incidence of hepatocellular carcinoma alone. 4.4 Principal Carcinogenic Response in Female Sprague-Dawley Rats The EPA combined the incidences of lesions observed by Kociba et al. (1978) in the liver, lung, hard palate, or nasal turbinates of female Sprague-Dawley rats for the purposes of estimating a TCDD cancer potency factor. As stated in the Office of Science and Technology Policy's (1985) document on chemical carcinogenesis, this procedure is not recommended : "Generally, most experts agree that the incidence of total tumors at all organ sites is not a very useful expression of cancer incidence, nor is the calculation of the incidence of total benign or total malignant tumors. Most useful appears to be the number of histologically unique tumors at specific organ sites." The combining of the incidences of hyperplastic nodules (referred to as neoplastic nodules in EPA, 1985) and hepatocellular carcinomas observed by Kociba et al. (1978) also is not warranted. As discussed in Section 4.3, there was no evidence in the Kociba et al. (1978) study to indicate that a hyperplastic nodule would likely progress to a hepatocellular adenoma and then to a hepatocellular carcinoma. Kociba et al. (1978) identified and reported the occurrence of adenomas in other organs and tissues, while no adenomas were reported in the liver, clearly indicating that the observed lesions were considered to be a less serious form of hepatocellular proliferative lesion, GENP 011440 784251 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 21 i.e. hyperplasia. Furthermore, Kociba et al. (1978) reported that there was no evidence of metastasis of any of these hyperplastic nodules. Hepatocellular carcinomas are the only indicators of a carcinogenic response in the Kociba et al. (1978) bioassay having any conceivable use in a doseresponse extrapolation via the LMS model. Considering only hepatocellular carcinomas, no statistically significant differences between controls and the lowest (1,000 pg/kg-day) or intermediate (10,000 pg/kg-day) dose groups were observed (Table 10). The statistically significant increased incidence of hyperplastic nodules observed by Kociba et al. (1978) in the intermediate (10,000 pg/kg-day) and highest (100,000 pg/kg-day) TCDD treatment groups should be appropriately extrapolated to human exposure only through the use of a classical safety factor approach: the LOAEL divided by a safety factor of 1,000. In this case, the LOAEL for hyperplastic nodules in the rat liver (10,000 pg/kg-day) divided by 1,000 would yield an ADI of 10 pg/kg-day. Thi3 approach is identical to that taken by the United Kingdom (1989) when they derived a guideline level of 10 pg/kg-day as protective for potential carcinogenic effects, "which, when exceeded, should trigger investigation and appropriate measures to reduce environmental levels." 4.5 Reassessment of C a r c i n o g e n i c Risk Xstimates Based. on Hepatocellular Carcinomas Using the combined incidences of lesions in the liver, lung, hard palate, or nasal turbinate observed in female Sprague-Dawley rats from Kociba et al. (1978), the EPA in 1985 developed a cancer potency estimate for TCDD of 1.56 x 10s (mg/kg/day) with an associated 95% LCL on the RsD at a 1 x 10*5 risk level of 0.064 pg/kg-day (Table 14). These estimates were derived from the geometric mean of risk estimates developed from the EPA's interpretation of Kociba et al. (1978) (Table 11) and from the interpretation of the histopathology of the Kociba et al. (1978) dataset by Squire (EPA, 1985) (Table 12) . Both interpretations were revised in the EPA analysis to eliminate the first year's data to account for the high early mortality observed by Xociba et al. (1978) in the high-dose groups. Risk estimates derived from the pooled incidences of lesions in the liver, lung, and oropharynx by EPA (1985) and Squire (EPA, 1965) are shown in Table 14. The EPA's position in this instance appears to represent a desire to reach a statistical compromise between different histopathological interpretations of the same carcinogenic bioassay. While the differences in the interpretations of the histopathology contribute some degree of uncertainty to the calculated potency of TCDD, this approach does not represent a major source of variation. To calculate revised risk estimates for TCDD based on the incidences of hepatocellular carcinomas observed in female Sprague-Dawley rats exposed to TCDD, computer modeling was conducted using the results of the Kociba et al. (1978) bioassay and the Global*86 Linearized Multistage (LMS) Model (Howe et GENP 011441 784252 ChemRIskTM - A McLaren Company NOVEMBER 2S, 1989 Page 21a Table 14. Risk Estimates" Derived from the Incidence of Hepatocellular Carcinoma and From the Pooled Incidence of Various Tumor Types Reported in K o d b a et aL (1978) Intemretaiion: Lesions Evaluated: Animal ql* (mg/kg/day)*1 ChemRiskTM Hepatocellular * carcinoma EPA. 1985 Pooled Tumois (Liver, Lung, and Oropharynx) 3.93 x 103 2.82 x 104 Sauire, 1985 Pooled Tumors (Liver, Lung, and Oropharynx) 2.99 x 104 EPA. 1985 Geometric Mean EPA & Squire 2.88 x 10* Human q l* 4 (mg/kg/day)*1 2.12 x lO 4 1.52 x 1 0 s 1.62 x 10s 1.56 x 103 95% LCL on the Human RsD at 1 x 10** Risk Level* (pg/kg-day) 0.47 0.066 0.062 0.064 Maximum Likelihood Estimate (MLE) on Animal RsD at 1 x 1(1* Risk Levelf (pg/kg-day) 4.0 0.48 0.47 0.48 a. U.S. agencies, including the EPA, have calculated and presented RsDs at various levels of risk. For die sake of consistency, the risk level of 1 in 100,000 has been used when presenting RsD estimates. b. Risk estmales were calculated by ChcmRiskTM using the Global '86 (Howe, Crump, ft Landingham, 1986) LMS model and the incidence o f hepatocellular carcinomas observed by Kociba et aL (1978). e. The geometric mean o f risk estimates by EPA, 1985 and by Squire (EPA, 1985) from the Kociba et aL(1978) study was adopted by the EPA is the appropriate estimate of risk to humans from exposure to TCDD. d. The human ql* was calculated by multiplying theanimal ql* by a surface area correction factor of 5.4. e. The 95% LCL on the human RsD was calculated from the equation: risk potency x dose; where risk a l t 10s and potency s human q l*. f. The MLE was calculated by Global *86(Howe, Crump, ft Landingham, 1986). GENP 011442 784253 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 22 al., 1986). The Global'86 LMS model is the latest revision to an earlier version of the model, Global'92 (Howe and Crump, 1982). Both Global'86 and Global'82 are currently employed interchangeably by the EPA Carcinogen Assessment Group (CAG) to evaluate potential human carcinogens (CAG, 1989) . Data were fitted to a conventional 3-stage Global'86 LMS model (one minus the number of dose groups in the bioassay). Estimates were derived for the animal aql* and the human ql* (i.e., cancer potencies), the 95% lower confidence limit (LCD on the human RsD, and the MLE of the animal R3D. The human ql* was calculated as the product of the cancer potency in animals (aql*) and a rat-to-human surface area correction factor of 5.4. The aql* is estimated by Global'86 at the 95% confidence limit. The rat-to-human surface area correction factor of 5.4 is employed by the EPA in their dose-response assessment process. The MLE of the RsD for the rat was estimated by Global'86 from the fitted 3stage model. The 95% LCL on the human RsD at a 1 x 10~5 risk level was calculated as the ratio between a 1 x 10~s risk level and the human ql* from the equation: RsD - .Risk Level Cancer Potency On the basis of the incidence of hepatocellular carcinomas reported by Kociba et al. (1978)-, a human ql* was calculated to be 2.12 x IQ4 (mg/kg-day)-1 with an associated 95% LCL on the RsD at a 1 x 10-5 risk level of 0.47 pg/kg-day (Table 14) . The estimate of the human ql* and the 95% LCL on the RsD are approximately seven times lower and higher, respectively, than the estimates derived by the EPA (1985) . The MLE at a 1 x 10"5 risk level was calculated-to be 4.0 pg/kg-day (Table 14). 4. 6 TJse of the MLS versus the 95% LCL of the Human RsD Empirical studies have shown that the 95% lower confidence limit (LCL) of the RsD is never smaller than the best estimate of the RsD (or Maximum Likelihood Estimate, MLE) when the underlying dose-response relationship is convex or upward curving, as it is for TCDD (Sielken, 1987) . Therefore, if it is believed that the true dose-response relationship for TCDD follows one of the curves of the multistage model family, then the fitted model value, or MLE, is the more appropriate basis for a risk estimate. Comparisons between the MLE and the 95% LCL of the human RsD have been conducted by Sielken (1987) and Kimbrough et al. (1984). Large differences between the MLE and the 95% LCL of the human RsD were observed by Sielken (1987) when the hepatic tumor incidence in the high TCDD dose group (100,000 pg/kg-day) from the Kociba et al. (1978) bioassay was not included in the LMS model. The resulting fitted curve resembled an upwardly sloping line for the control and two lowest dose groups (1,000 and 10,000 pg/kg-day). The analysis of hepatic tumors in female rats by Kimbrough et al. (1984) Indicates that GENP 011443 784254 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 23 when the high TCDD dose group was not dropped from the analysis, little difference was observed between the MLE and the 95% LCL of the human RsD. The reason for these disparate results appears to lie in the mathematical function that describes the LMS model: P (d) - 1 - exp [(qid + q2d2 + ... + q**) J where P (d) is the lifetime risk (probability) of cancer at dose d. In the analysis by Kimbrough et al. (1984), when the linear term (q^) of the multistage model defines the dose-response curve from the experimental data, as in the case of the incidence of hepatic tumors in female rats from Kociba et al. (1978), there is little difference (a factor of about 1.3) between the MLE and the 95% LCL of the human RsD. In Sielken's (1987) analysis, when the high dose group was not included in the multistage model, the linear term was not adequate to describe the dose-response curve for TCDD. The upwardly sloping'fitted curve of the multistage model that was observed by Sielken (1987) suggests that the quadratic term (q2) best describes the dose-response curve for the tumor incidence observed in the control and two lowest TCDD dose groups. In this case, differences between the MLE and the 95% LCL of the human RsD can be very large. Risk estimates developed in this analysis using a 3-stage Global*86 LMS model and the incidence of hepatocellular carcinoma reported in Kociba et al. (1978) were compared to estimates developed by EPA (1985) from Kociba et al. (1978) and Squire (EPA, 1985) and to estimates calculated by Kimbrough et al. (1984) from NTP (1982). Estimates of the MLE and 95% LCL on the human R3D at a l x 10~5 risk level calculated by the EPA (1985) from Kociba et al. (1978) and from Squire (EPA, 1985) are shown in Table 14. The 95% LCL of the human RsD was calculated, in part, on the basis of surface area differences between rats and humans. In order to compare the MLE to the 95% LCL of the human RsD at a 1 x 10"5 risk level, the surface area correction factor of 5.4 was also applied to the MLE. A comparison of these estimates developed by EPA (1985) from Kociba et al. (1978) and Squire (EPA, 1985) is presented in Table 15. The comparisons in Table 15 demonstrate that the corrected MLE is consistently higher than the 95% LCL of the human RsD at a 1 x IQ"5 risk level. The increase, however, is not substantial. In all cases, the MLE, corrected for surface area differences between rats and humans, is approximately 1.3 times larger than the 95% LCL of the human RsD. The geometric mean of the MLE, corrected for surface area differences, is slightly higher, 0.088 pg/kg-day, than the 95% LCL of the human RsD at a 1 x 10"5 risk level, 0.064 pg/kg-day. Comparisons between estimates calculated by the EPA (1985) of the MLE and the 95% LCL of the human RsD were also conducted on the basis of a body weight correction factor (Table 16) . Since a body weight correction factor assumes a 1:1 relationship between the weight of a rat and a human, the estimate of the human cancer potency is assumed to be equal to the animal ql* (aql*) predicted G E N P O II444 784255 ChemRlskTM - A McLaren Company NOVEMBER 28,1989 Page 23a Table 15. Comparison at a 1 x 10'* Risk Level Between EPA (1985) Estimates of the 95% Lower Confidence Limits (LCL) on tbe H um an RsD and the Maximum Likelihood Estimates (MLE) of the Animal RsD Based on Sorface Area Correction 95% LCL on Human RsD* (pgAg-day) MLE on Animal RsD Corrected for Surface Area* (pg/kg-day) EPA, 1985 Pooled Tumors (liver, lung, oropharynx) 0.066 0.089 Squire (EPA, 1985) Pooled Tumors (liver, lung, oropharynx) 0.062 0.087 Geometric Mean EPA & Squire Interpretations of K oribaetal. (1978) 0.064 0.088 a. The 95% LCL on the human RsD was calculated from tbe equation: risk = potency x dose; where risk s i x 10** and potency s ql* from Table 14. b. The MLE was calculated by Global 36 (Howe, Crump, A Landingham, 1986k The MLE dose in the rat was corrected for surface area differences between rets and humans by dividing the MLE by a surface area correction factor of 5.4. ?8 4 2 S e GENP 011445 . ChemRlskTM - A McLaren Company NOVEMBER 28,1989 Page 23b Table 16. Comparison at a 1 x 10*s Risk Level Between EPA (1985) Estimates of the 95% Lower Confidence Limits (LCL) on the Human RsD and the Maximum Likelihood Estimates (MLE) of the Animal RsD Based on Body Weight Correction 95% LCL on Human RsD Corrected for Body Weight* (pg/kg-day) \ EPA, 1985 Pooled Tumors (liver, lung, oropharynx) 035 MLE on Animal RsD* (pg/kg-day) 0.48 Squire (EPA, 1985) Pooled Tumors (liver, lung, oropharynx) 033 0.47 Geometric Mean EPA & Squire Interpretations of Kociba et al. (1978) 0.34 0.48 a. The 95% LCL on the human RsD was corrected for body weight differences between rats and humans from the equation: risk = potency x dose; where risk = 1 x 10** and potency = aql* from Table 14. b. The MLE was calculated by Global '86 (Howe, Crump & Landingham, 1986). GENP 011446 784257 ChamRiskTM - A McLaren Company NOVEMBER 2 8 , 1989 Page 24 by the Global*86 LMS model. This approach is identical to removing the surface area correction factor from the estimates of the 95% LCL of the human RsD in Table 14. Similar to Table 15, the comparisons in Table 16 demonstrate that the MLE is approximately 1.3 times higher than the corrected 95% LCL of the human RsD at a 1 x 10~5 risk level. The geometric mean of the MLE, 0.48 pg/kg-day, is slightly higher than the 95% LCL of the human RsD, corrected for surface area differences, at a 1 x IQ-5 risk level, 0.34 pg/kg-day. The effect of employing a body weight versus a surface area correction factor also was considered for potency estimates based on the incidence of hepatocellular carcinomas observed in the Kociba et al. (1978) bioassay. This tumor type should be considered as the most appropriate carcinogenic response observed during the Kociba et al. (1978) bioassay. Regardless of whether an adjustment is made on the basis of surface area or body weight, the MLE was approximately 1.6 times higher than the 95% LCL of the human RsD at a 1 x 10"5 risk level (Table 17). In Table 17, when risk estimates were based on body weight and a 1 x 10 ^ risk level, the MLE was 4.0 pg/kg-day and the 95% LCL of the human R3D was 2.5 pg/kg-day. When these estimates were based on a correction for surface area differences between rats and humans at the 3ame risk level, the MLE was 0.74 pg/kg-day and the 95% LCL of the human RsD was 0.47 pg/kg-day. Risk estimates developed in this analysis from the incidence of hepatocellular carcinomas reported in Kociba et al. (197 8 ) were also compared to the incidences of similar lesions reported in Osborne-Mendel rats and B6C3F1 mice in NT? (1982) . Comparisons between the MLE and the 95% LCL on the human RsD calculated from Kociba et al. (1978) and NTP (1982) were conducted on the basis of a correction for body weight differences between rodents and humans. These comparisons are shown in Table 18. Based on the incidence of hepatocellular carcinomas in the liver, it appears that 56C3F1 male mice may be the most sensitive species/strain/sex to exposure to TCDD (Table 18) . Differences in the route of administration of TCDD between animals in the NTP (1982) study and Kociba et al. (1978), however, must be carefully considered before reaching a more definitive conclusion. The MLE and 95% LCL on the human RsD at a 1 x 10 risk level were calculated to be 2.6 and 1.5, respectively, in male B6C3F1 mice. In contrast, the MLE and 95% LCL on the human RsD at a 1 x 10-*5 risk level were estimated in this analysis to be 4.0 and 2.5, respectively, in female Sprague-Dawley rata. 4.7 Biological Basis for Extrapolating (Intarapacies Scaling Factors) Across Species In attempting to account for the various factors that contribute to interspeciea variation in the toxicological response to a chemical, including both carcinogenic and noncarcinogenic compounds, scientists have long sought to establish a common biological baseline for extrapolating ac,ross species. GENP 011447 784258 ChemRIskTM - A McLaren Compaaj NOVEMBER 28,1989 Page 24a Table 17. Comparison at a 1 x 10'* Risk Level Between Estimates of the 95% Lower Confidence Limits (LCL) on the H um an RsD and the Maximum Likelihood Estimates (MLE) of the Animal RsD Based on the Incidence of Hepatocellular Carcinoma R eported by Kociba et al. (1978) RsD at 1 x 10s Risk Level (pg/kg-day) 95% LCX on Human RsD Corrected for Body Weight* 2.5 Maximum Likelihood Estimate On Anim al RsD 95% LCL on Human RsD 4.0 0.47 Maximum Likelihood Estimate Corrected far Surface Area* 0.74 a. The 95%LCL on the human RsD w u corrected for body weight differeoces between n is and humans from the equation: risk potency x dose; where risk = 1 x 10*3 and potency = aql* from Table 14. b. The MLEwas calculated by Global *86 (Howe, Crump, & Landing* ham. 1986). The MLE dose in the rat was corrected for m f a r j f m differences between rets and humans by dividing the MLEby a surface area ca irctioo factorof 5.4. 784259 GENP 011448 ChemRIskTM - A McLaren Company NOVEMBER 28,1989 Page 24b Table 18. Comparison Between Estimates" of the 95% Lower Confidence Lim its (LCL) on the Human RsD and the Maximum Likelihood Estimate (MLE) of the Animal RsD Based on the Incidence of Hepatocellular Carcinomas R eported by Kociba et al (1978) and NTP (1982). ChemRiskTM Female Sprague-Dawley Rats Maximum Likelihood Estimate on RsD at 1 x 10'* Risk Level (pg/kg-day) 95% LCL on Human RsD at 1 x 10'* Risk Level, Corrected for Body Weight* (pg/kg-day) 4.0 ZS NTP Female Osbomc-Mendel Rats 18,000 6.2 NTP Male Osbomc-Mendel Rats 26,000 11.0 NTP Female B6C3F1 Mice 25.0 12.0 NTP Male B6C3F1 Mice 16 1.5 a. Risk estimates were calculated by CbemRiskTM using the Global '86 (Howe, Crump, & Landingham, 1986) LMS model. b. The 95% LCL on the human RsD was corrected for body weight differences between rets and humans from the equation: risk = potency x dose; where risk = 1 x 10s and potency = aql* from Table . 784260 GElStP 011449 ChemAiakTM - A McLaren Company NOVEMBER 28, 1989 Page 25 In the absence of detailed pharmacokinetic information on interspecies differences, the two extrapolation procedures most often considered to address physiological and toxicological differences between mammalian species are adjustments by body weight scaling or by surface area scaling. Adolph (1949) was among the first researchers to correlate a number of anatomical, biochemical, and physiological properties of mammals with their body weights using the relationship: Property - a(body weight)k His most significant finding was the empirical demonstration that a large number of physiological characteristics could be extrapolated across a wide range of mammalian species if they were first standardized in terms of some power of body weight. Estimates for the value of k tended to be less than 1.0 for relationships depicting physiological functions, indicating that as body weight increased, the physiologic function per unit of body weight increased. Fractional powers of body weight between 0.6 and 0.8 are very close to the fractional power of 0.67, which relates the surface area of cylindrical objects to their volume. Since the density of most mammalian bodies is about the same, mass or body weight can be used instead of volume in the above relationship. This is the origin of the term surface area or (body weight)0*67 scaling factor. Adjustment based on differences in body surface area is based on the premise that the magnitude of the -adverse effect of the chemical is dependent on the basal metabolic rate (BMR) of the species of concern (0*Flaherty, 1988; Clayson, 1988). Standardization of risk estimates on the basis of body surface area is not broadly accepted as the best approach to the estimation of cancer risk. For example, Krasovskii (1976) observed that expressing dosage in mg/m* in order to adjust for species surface area differences in extrapolation only seemed to be appropriate in about half of the cases considered. Furthermore, the appropriateness of body weight as a scaling factor has been substantiated by an interspecies comparison of carcinogenic potency for approximately 70 chemicals which gave reliable interspecies correlations between potencies (Crouch and Wilson, 1979). The TCDD dose extrapolation procedure used most often by the EPA (1985) is based on the assumption that dose per unit body surface area is equivalent between species. In contrast, the CDC and the FDA use dose per unit body weight when the active carcinogen is thought to be the administered compound, and only use dose per unit surface area when the active carcinogen is a metabolite of the administered compound (Bayard, 1988). Both the CDC and the FDA have chosen to use the body weight scaling factor in the case of TCDD, based on their opinion that strict dose per unit body weight considerations ought to apply since the administered compound does not have to be metabolized in order to be carcinogenic in rodents (Bayard, 1988) . G E N P 0 1 1450 784261 f \ ChemRiskTM - A McLaren Company NOVEMBER 28, 1988 Page 26 To estimate the dose-response relationship for low-level human exposure to TCOD based on the data obtained from high doses used in the Kociba et al. (1978) bioassay, the U3e of a body weight scaling factor is biologically more relevant to the pharmacokinetic behavior of TCDD than the surface area correction factor employed by the EPA. The use of a scaling factor based on surface area is inappropriate as a surrogate for relative rates of metabolism between species for poorly metabolized chemicals such as TCDD. Basic pharmacokinetics suggest that the parent compound is the biologically active moiety (Shu, et al. 1987; Leung et al., 1988, 1989). In addition, the scientific literature reports markedly different tissue distributions of TCDD between species (EPA, 1985). Thus if the liver is the organ of primary toxicological concern, a human would have to be given anywhere from 10 to 50 times the dose on a mg/kg-body weight basis to have the same liver concentrations as the rat (Bayard, 1988) . Even if one were to consider the longer relative half-life of TCDD in humans than in rats, it is the opinion of ChemRisk that the body weight scaling factor is more appropriate than surface area for this extrapolation between rats and humans. With the further development of biologically-based pharmacokinetic models for TCDD, the application of basic pharmacokinetic principles will provide the most accurate means available to extrapolate doses between species (Mordenti, --y 1985, 1986; Lutz and Dedrick, 1987; Anderson et al., 1987; Rietz et al., 1988; Travis, 1989). 784262 GENP011451 CheaRiakTM - A McLaren Company NOVEMBER 28, 1989 Page 27 5 .0 CONCLUSIONS ChemRiskTM has conducted a sensitivity analysis of the CPA (1985) dose* response assessment for TCDD, and has reevaluated the biological data from the Kociba et al. (1978) bioassay, prior to using the LMS model for estimating scientifically defensible RsDs for TCOD. The quantitative impacts of several choices involved in the risk assessment of TCDD were reviewed, including the definition of the carcinogenic response of concern within the experimental dataset, the evaluation of liver pathology by the original researchers and the CPA, the methods used to make the fitted LMS model responsive to the data at the lower experimental doses, the choice of rat-to-human scaling factor, and the method of maximum acceptable dose estimation. Cach choice affects the range of plausible risk estimates predicted using the LMS model. 5.1 Conclusions of the Sensitivity Analysis The low-dose risk estimate derived by the EPA through the application of the LMS model to the Kociba et al. (1978) and the Squire (EPA, 1985) datasets is very sensitive to the choice of modeling assumptions. The principal conclusions of the hemRisk sensitivity analysis are summarized below. Hepatocellular carcinomas are the only indicators of a carcinogenic response in the Kociba et al. (1978) bioassay having any conceivable Use in a dose-response extrapolation via the LMS model. Considering only hepatocellular carcinomas, no statistically significant differences between controls and the lowest (1,000 pg/kg-day) or intermediate (10,000 pg/kg-day) dose groups were observed. Therefore, the doseresponse curve is determined by only a single data point and the underlying assumptions of the model which force a fit to the origin in order to preserve the notion of "linearity" . Furthermore, the high dose level administered to the rats (100,000 pg/kg-day) clearly exceeded the Maximum Tolerated Dose (MTD), as evident from Kociba*s observations of severe liver toxicity, diminished weight gain, and increased mortality. The statistically significant (95% confidence limit) increased incidence of hyperplastic nodules observed by Kociba et al. (1978) in the intermediate (10,000 pg/kg-day) and highest (100,000 pg/kg-day) TCDD treatment groups should not be modeled via a nonthreshold approach because : (a) There is no evidence in this study to indicate that a hyperplastic nodule is likely to progress to a hepatocellular adenoma and then to a hepatocellular carcinoma. (b) Kociba et al. (1978) identified and reported the occurrence of adenomas in o.ther organs and tissues, while no adenomas were reposted in the liver, clearly indicating that the observed lesions were considered to be a less serious form of hepatocellular proliferative lesion, i.e. hyperplasia. GENT 011452 784263 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 28 Furthermore, Kociba et al. (1978) reported that there was no evidence of metastasis of any of these hyperplastic nodules. (c) "Neoplastic nodules", a term used by Squire (EPA, 1985) in his re-evaluation of the Kociba analysis, represents terminology from a different morphological classification system developed by Squire and another researcher. Its use by Squire (EPA, 1985) does not provide convincing evidence that the hyperplasia was actually a more serious condition. The classification system of Squire and Levitt (1975) failed to define the exact position and significance of the "neoplastic nodule", which included both hyperplasia and more serious lesions (EPA, 1986b). Today, the National Toxicology Program (NTP) no longer uses the classification scheme of Squire and Levitt, and neoplastic nodules are no longer identified during histopathological examinations by mo3t toxicological laboratories, including the NTP (Maronpot et al., 1986; McConnell et al., 1988). "Neoplastic nodule" has been replaced with two terms: hyperplasia and hepatocellular adenoma (EPA, 1986b; Maronpot et al., 1986; McConnell et al., 1968). (d) Schul'te-Hermann et al. (1963) demonstrated that a significant incidence of hyperplastic liver nodules are found in the livers of untreated rats. The incidence increased to 100% over the course of the animals* two-year average lifetime and lesion 3ize increased with age. In addition, these hyperplastic nodules showed similar characteristics to those observed in rats treated with known liver carcinogens. Schulte-Hermann et al. (1983) have concluded that the high incidence of hyperplastic liver nodules in untreated aged rats precludes the distinction between chemical promoters and initiating agents during long-term carcinogenicity bioassays in laboratory rodents, particularly when liver tumor incidence data are the primary criteria. The statistically significant increased incidence of hyperplastic nodules observed by Kociba et al. (1978) in the intermediate (10,000 P?/k?-day) and highest (100,000 pg/kg-day) TCDD treatment groups should be appropriately extrapolated to human exposure only through the use of a classical safety factor approach: the LOAEL divided by a safety factor of 1,000. In this case, the LOAEL for hyperplastic nodules in the rat liver (10,000 pg/kg-day) divided by 1,000 would yield an ADI of 10 pg/kgday. This approach is identical to that taken by the United Kingdom (1989) when they derived a guideline level of 10 pg/kg-day, "which, when exceeded, should trigger investigation and appropriate measures to reduce environmental levels." The frequencies of tumors in other organ systems, specifically in the lung, tongue and palate, should not be considered in the dose-response assessment of TCDO. The Office of Science and Technology Policy (1985) does not recommend combining tumor incidences in different organs for the purposes of estimating a cancer potency. Furthermore, Kociba (1984) 784264 ChemRislcTM - A McLaren Company NOVEMBER 28, 1989 Page'29 has suggested that the tumor incidence observed in the lung and oropharynx is likely the result of periodically high localized concentrations of TCOD associated with surface contact of TCDD from the animal's diet, rather than an effect of systemic uptake. The use of a body weight correction factor in the rat-to-human extrapolation procedure is biologically more relevant than the use of a surface area correction factor for TCDD. In the case of TCDD, basic pharmacokinetics suggest that the parent compound i3 the biologically active moiety (Leung et al., 1988, 1989). Despite this evidence, the EPA used a surface area correction factor of 5.4 when extrapolating doses between Sprague-Dawley rats in the Kociba et al. (1978) study and humans. 5.2 Implications for Determining Acceptable Levels of Daily Exposure Based on this analysis and on the other scientific evidence summarized in this report, it is clear that the use of the LMS model in conjunction with the Kociba et al. (1978) bioassay to describe the carcinogenic dose-response of TCDD can result in a wide range of plausible risk estimates (Table 19) . The choice of maximum likelihood estimate (MLE) for the RsD in preference to the 95% lower confidence limit (LCL) estimate, body weight instead of surface area scaling factor, and a better definition or biological interpretation of the reported liver lesions will result in RsDs much different than the estimates adopted by the EPA (1985, 1988) and the CDC (Kimbrough, 1984) , Acceptable daily intakes and RsDs have been developed by various U.S. agencies and other countries on the basis of some of these considerations. It is evident from Figure 5 that the application of different extrapolation procedures to the same animal cancer bioassay data has resulted in markedly different exposure limits for TCDD. Risk estimates by various regulatory agencies have ranged from 0.064 pg/kg-day (EPA, 1965) to 14 pg/kg-day (Kimbrough, 1984) when expressed at the 1 x 10~5 level of incremental risk. ChemRisk further concludes that the weight of scientific evidence indicates that TCDD should be regulated as if it exhibited a threshold of carcinogenic action. The scientific evidence does not support risk estimates for TCDD which are based on linear low-dose extrapolation models. The most appropriate method for estimating permissible limits for human exposure, given the present state of knowledge in this interim period before a biologically-based pharmacokinetic model is validated experimentally, is the application of a safety factor to the NOAEL for carcinogenicity observed in the Kociba et al. (1978) study.' Although the use of the safety factor approach may be most appropriate for the dose-response assessment of materials such as TCDD that lack genotoxic potential, it is recognized that others may wish to use the linearized multistage model or some other nonthreshold model of low-dose extrapolation in GENP 011454 784265 ChemRIskTM - A McLaren Company NOVEMBER 28,1989 Page 29a 15 - r 14 - 13 - - 12 - 11 - 10 - - 9 8 7 -- Figure 5. Rlsk-spedflcPoses (at Iff5) and Acceptable Dally Intakes Calculated for TCDD1 4 *+ 3 -4- 2 -f 1 0 < Si o <4 1 S 3 CDC(b) EM COfonte IDA M>V4 no JfcwYwt U J.Onti-- Miftw r i wA|WBy-KiD Staa C aE fM U 'taD C u n tm DIm m Com e UAIUUM l S *rh . IUD (a) U M O M M ln S iM ^ n g i <k> l J M t f w M w f l p w r f w y U J.M ain M A M M H .liD thitw tat, V i f , M m - M ) M o r i faftiS c tt Oammr *ADI (H C liH llU rM W iw (l|W rf> IM k o**hM kyPKa) SO* of Nm Y ak - ADC ADI M U a rr af O U M I M Ik M WITM . AOC U ) () O-- e w f t o r t l n t M m a i W l i l (V) M H m M iM M i1 1. U.S. agencies, including the EPA, have calculated and presented RsDa at various levels of risk. For die sake of consistency and to enhance the clarity of this document, the risk level of 1 in 100,000 has been used in the presentation of RsD estimates. GENP011455 784266 ChemRiskTM A McLaren Company NOVEMBER 28,1989 Page 29b Table 19. Range of Plausible Risk-specific Doses (RsDs) for 2,3,7,8-TCDD Based on Considerations of Extrapolation Between Rats and Hum ans, the 95% LCL versus the MLE, and the Principal Carcinogenic Response Observed in Kociba et al. (1978) Set o f Assumptions * Risk-specific Dose at 1 x 10'* Incremental Carcinogenic Risk (pg/kg-day) 1. Surface Area + 95% LCLh+ Total Hepatic Tumors* 0.064 2. Surface Area + MLE + Total Hepatic Tumors 0.088 3. Body W eight+ 95% LCL + Total Hepatic Tumors 0.34 4. Surface Area + 95% LCL + Hepato. carcinomas 0.47 5. Body Weight + M LE+ Total Hepatic Tumors 0.48 6. Surface Area + MLE + HepaL carcinomas 0.74 7. Body Weight + 95% LCL + Hepato. carcinomas4 2.5 8. Body Weight + MLE + Hepato. carcinomas4 4.0 a. Etch set of assumptions were developed from the Kociba et al. (1978) bioassay in female Sprague-Dawky rata, Spartan substrain, on the basis of three choices: 1) extrapolation from rats to humans using a surface area versus a body weight correction factor; 2) adoption of the 95% Lower Confidence Limit (LCL) of the human RsD vosus the Maximum Likelihood Estimate (MLE) of the potency in the rat; and, 3) the consideration o f hepatocellular hyperplastic nodules and carcinomas (total hepatic tumors) venus hepatocellular carcinomas as the principal response observed in the Kociba et aL (1978) bioassay. b. The 95% LCL is that dose for which thae is a 95% statistical confidence that the true RsD is no lower than this value. c. These assumptions represent the basis for the EPA's currentestimate of the human RsD for 2,3,7,8-TCDD. d. These assumptions are considered by ChemRisk to represent the most plausible basis for an alternative human RsD for 2,3,7,8-TCDD. GENP 011456 784267 ChamRiakTM - A McLaren Company NOVEMBER 28, 1989 Page 30 conjunction with th Kociba et al. (1978) dataset. Since this is currently the case with U.S. regulatory agencies, then the carcinogenic behavior of TCDD in female Sprague-Dawley rats should be extrapolated to humans only on the basis of the incidence of hepatocellular carcinomas. A body weight scaling factor should be used in preference to the surface area assumptions. If the 95% LCL of the human RsO is adopted, then the RsO at a 1 x 10~5 risk level is 2.5 pg/kg-day of TCDD. If the MLE is adopted, then the appropriate RsD at a 1 x IO-5 risk level is 4.0 pg/kg-day of TCDD (Table 20) . Both of these RsDs are more conservative than the ADI of 10 pg/kg-day, estimated by ChemRisk to be protective of hepatic hyperplasia, derived via a classical safety factor approach. 784268 ChemRlskTM - McLaren Company NOVEMBER 28,1989 Page 30a Table 20. Plausible Alternative Risk-specific Doses (RsDs) for 2 r3,'7> 8-T C D D ' Plausible Choices* Risk-specific Dose at 1 x 1G'SIncremental Carcinogenic Risk (pg/kg-day) 1. Body Weight + 95% LCL* + Hepato. carcinomas 2.5 2. Body Weight + MLEe+ Hepaio. carcinomas 4.0 a. These assumptions are considered by ChemRiskTM (1989) to represent the most plausible basis for an alternative human RsD for 2,3,7,8-TCDD. b. The 95% LCL is that dose for which there is a 95% statistical confidence that the true RsD is no lower than this value. c. Maximum likelihood estimate. GENP 011458 784269 ChemRiakTM - A McLaren Company NOVEMBER 28, 1989 Page 31 8.0 REFERENCES Aldoph, E.F. 1949. Quantitative relations in the physiological constitutions of mammals. Science 109:579-585. Allman, W.F, 198S. Staying alive in the 20th century. Science 85 6(8):29-47. American Cancer Society (ACS). 1989. Age-adjusted cancer death rates for selected sites, males, United States, 1930-1985, and age-adjusted cancer death rates for selected sites, females, United States, 1930-1985. Cancer for Clinicians 39:10-11. American Medical Association (AMA) . 1984. The Health Effects of ''Agent Grange* and Polychlorinated Dioxin Contaminants: An Update, 1884. Technical Report Updated 05-29-85. Prepared by the Council on Scientific Affairs Advisory Panel on Toxic Substances, American Medical Association. Chicago, IL. Anderson, M.E., H.L. Clewell, III, M.L. Gargas, F.A. Snath and R.H. Reitz. 1987. Physiologically based pharmacokinetics and the risk assessment process for methylene chloride. Toxicol. Appl. Pharmacol. 8 7:185-205. Armstrong, B. 1983. Australians report no link between service in Vietnam and birth defects among offspring. Epidemiol. Monit. 3(4):1. Bannasch, P ., M.A. Moore, P. Klimek and H. Zexban. 1982. Biological markers of preneopla3tic foci and neoplastic nodules in rodent liver. Toxicol. Path. 10:19-36 Bayard, S.P. 1988. Quantitative implications of the use of different extrapolation procedures for low-dose cancer risk estimates from exposure to 2,3,7,8-TCDD. Carcinogen Assessment Group. U.S. Environmental Protection Agency, (contained in EPA, 1988, Appendix A) . Beck, H., K. Eckart, W. Mathar and R. Wittkowski. 1988. Occurrence of PCDD and PCDF in different kinds of paper. Chemosphere 17(1) :51-57. Birmingham, B., B. Thorpe, R. Frank, R. Clement, H. Toaine, G. Fleming, J. Wheeler, B.D. Ripley, J.J. Ryan. 1989. Dietary intake of PCDD and PCDF from food in Ontario, Canada. Chemosphece (in press). Bleiberg, J., M. Wallen, R. Brodkin and I.L. Applebaum. 1964. Industrially acquired porphyria. Arch. Dermatol. 99:793-797. Bond, G..G., K.M. Bodner and R.R. Cook. 1989. Phenoxy herbicides and cancer: Insufficient epidemiologic evidence for a causal relationship. Fund. Appl. Toxicol. 12:172-188. Bond, G.G., M.G. Ott, F.E. Brenner and R.R. Cook. 1983. Medical and morbidity surveillance findings asaong employees potentially exposed to TCDD. Brit. J. Indust. Med. 40:318-324. 784270 ChemRlsk1* - A McLaren Company NOVEMBER 28, 1989 Page 32 Bonvalot, Y., L. Abenhaim and D. Bard. 1989. Uncertainty in quantitative risk aaaeaament procedures for 2,3# 7,8-tetrachlorodibenzo-p-dioxin. Chemosphere 29(1-6):623-628. California. 1986. Safe Drinkd:lg Water and Toxic Enforcement Act of 1986 (Proposition 65). Health and Welfare Agency, Office of the Secretary, Sacramento, CA. CanTox, Inc. Consultants in Toxicology. 1989. Biological Fisk Assessment for 2, 3, 7, 8-Tetrachlorodibenzo-p-dioxin. Prepared for The National Council of Paper Industry for Air and Stream Improvement, Inc., NCASI Technical Bulletin No. 567. New York, NY. Carcinogen Assessment Group (CAG). 1989. Personal coimunication between J. Cagliano, USEPA CAG, Cincinnati, Ohio and J. Michaud, ChemRiskTM. November 9. Clayson, D.B. 1988. Problems in Interspecies Extrapolation. In: Toxicological Risk Assessment. Volume I. Biological and Statistical Criteria. D.B. Clayson, D. Krewski and I. Munro (eds.). CRC Press, Boca Raton, FL. pp. 105-122. Cook, R.R., G.G. Bond, R.A. Olsen, M.G. Ott and M.R. Gondek. 1980. Evaluation of the mortality experience of workers exposed to the chlorinated dioxins. Chemosphere 25(9-12) :1769-1776. Cordle, F. 1983. Uses of epidemiology in the regulation of dioxins in the food supply. In: Accidental Exposure to Dioxins, Human Health Aspects. Chapter 16. F. Coulston and F. Pocchiari (eds.). Academic Press, New York, NY. pp. 245-2S6. Crouch, E. and R. Wilson. 197 9. Interspecies comparison of carcinogenic potency. J. Toxicol. Environ. Health 5:1095-1118. Crouch, E. and R. Wilson'. 1982. Risk Benefit Analysis. Chapter 7. Ballinger Publishing Company, Cambridge, MA. pp. 165-193. Crump, K.S. 1988. A critical evaluation of a dose-response assessment for TCDD. Fd. Chem. Toxicol. 2(l):79-80. Dedrick, R.L., X.B. Rischoff and D.S. Zaharko. 1970. Cancer Chemother. Rep. Part 1 54:95-101. Dioxin Update Committee. 1986. Report of the Dioxin Update Committee, Convened by EPA Office of Pesticide and Toxic Substances. Dioxin Update Conference, July 1-2. Environmental Protection Agency (EPA). 1984. Ambient Water Quality Criteria for 2,3,7,8-tetrachloro-dibenzo-p-dioxin. EPA 440/5-84-007. Office of Water Regulations and Standards, Washington, D.C. GENP 0 1 1 4 6 0 784271 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 33 Environmental Protection Agency (EPA). 1985. Health Assessment Document for Polychlorinated Dibenzo-p-Dioxins. EPA/600/8-84/Q14F. Office of Health and Environmental Assessment, Washington, D.C. Environmental Protection Agency (EPA). 1986a, Guidelines for carcinogen risk assessment. Federal Register SI (185) 33992-34003. Environmental Protection Agency (EPA). 1986b. Proliferative Hepatocellular Lesions of the Rat. Review and Future Use in Risk Assessment. EPA/625/386/011. Risk Assessment Forum Staff, Washington, D.C. Environmental Protection Agency (EPA). 1988. A Cancer Risk-Specific Dose Estimate for 2 , 3 , 7 ,8-TCDD. Review Draft. EPA/600/6-88/007 Aa and Ab. Office of Health and Environmental Assessment, Washington, D.C. Eriksson, J.D., L. Hardell, N.O. Berg, T. Holler and O. Axelson. 1981. Soft-tissue sarcomas and exposure to chemical substances: a case-referent 3tudy. Br. J . Indust. Med. 36:27-33. Eriksson, J.D., J. Mulinare, P.W. McClain, T.G. Fitch, L.M. James, A.B. McClearn and M.J. Adams, Jr. 1984. Vietnam veterans' risks for fathering babies with birth defects. JAMA. 252(7):903-912. Farber, T.M, 1989. Comments on the Georgia Department of Natural Resources* Proposed Water Quality Standard for TCDD. Submitted to the Georgia Department of Natural Resources, Augusta, GE. September 21. Filipini, G., B. Bordo, P. Crenna, N. Massetto, M. Musicco and R. Boeri. 1981. Relationship between clinical and electrophysiological findings and indicators of heavy exposure to 2, 3,7, 8-tetrachlorodibenzodioxin. Scand. J. Work Environ. Health 7:257-262. Fishbein, L. 1987. Health risk estimates for 2,3,7,8-tetrachlorodibenzo-pdioxin: An overview. 7ox and Znd. Health 3(2):91-134. Food & Drug Administration (FDA). 1983. Statement by Sanford A. Miller, Ph.D., Director, Bureau of Foods, Food and Drug Administration, before the subcommittee on Natural Resources, Agriculture, Research and Environment, Committee on Science and Technology, U.S. Bouse of Representatives. June 30. Gaylor, D.W. and R.L. Kode11. 1980. Linear interpolation algorithm for low dose risk assessment of toxic substances. J. Environ. Pathol. Toxicol. 4:3OS312. Gough, M. 1988. The most potent carcinogen? Resources for the Future 92:2-5. Graham, M., F. Hileman, D. Kirk, J. Wendling and J. Wilson. 1985. Background human exposure to 2,3,7,8-TCDD. Chemosphere 14 (6/7):925-928. GENP oi i 4 s i 784272 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 34 Hardell, L. and A. Sandstrom. 1979. Case-control study: soft-tissue sarcomas and exposure to phenoxyacetic acids or chlorophenols. Hr. J. Cancer 39:711-717. Hardall, L., M. Eriksson, P. Lenner and E. Lundgren. 1981. Malignant lymphoma and exposure to chemicals, especially organic solvents, chlorophenols and phenoxyacetic acids; a case control study. Br. J. Cancer 43:169-176. Hirota, N. and G.M. Williams. 1979. Persistence and growth of rat liver neoplastic nodules following cessation of carcinogen exposure. J. Wat. Cancer Inst. 63:1257-1265. Holder,. J.W. and H.M. Menzel. 1909. Analysis cf 2,3,7,8-TCDD tumor promotion activity and its relationship to cancer. Chemospbere 19(1-6) :861-868. Hollander, C.F. and J.D. Burek. 1978. Strain and age-associated pathology of the rat liver. In: Liver and Aging, Elsevier/Kirth Holland Biomedical Press. Amsterdam, Netherlands, pp. 39-42. Howe, R.B. and K.S. Crump. 1982. Globalf82. A Computer Program to Extrapolate Qvantal Animal Toxicity Data to Low Doses. Prepared for the Office of Carcinogen Standards, OSHA, U.S. Department of labor. K.S. Crump and Company, Inc., Ruston, LA. Howe, R.B., K.S. Crump and C.V. Landingham. 1986. Global*86. A Computer Program to Extrapolate Quanta 1 Animal Toxicity Data to Low Doses. K.S. Crump and Company, Inc., Ruston, LA. Inside EPA 9(49). December 9, 1988. Kimbrough, R., H. Falk, P. Stehr and G. Fries. 1964. Health implications of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) contamination of residential soil. J. Toxicol. Environ. Wealth 14:47-93. Kimbrough, R.D, and V.N. Houk. 1987. Effects cf. chlorinated dibenzodioxins. In: Solving Hazardous Waste Problems: Learning from Dioxins, J.H. Exner (ed.) . ACS Symposium Series 338. pp.68-79. Kociba, R.J., D.G. Keyes, J.E. Beyer, R.M. Carrera, C.E. Wade, D.A. Dittenber, R.P. Kalnins, L.E. Frauson, C.N. Park, S.D. Barnard, R.A. Hummel and C.G. Humiston. 1978. Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8-Tetrachlorodibenzo-p-dioxin in rats. Toxicol. Appl, Pharmacol, 46:279-303. Kociba, R.J. and B. Schwetz. 1962. Toxicity of 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD). Drug Metab. Rev. 13(3):387-406. Kociba, R.J. 1984. Summary and critique of rodent carcinogenicity studies of chlorinated dibenzo-p-dioxlns. In: Public Health Risks of the Dioxins. W.W. Lowrance (ed.). William Kaufmann, Loa Altos, CA. pp. 77-98. G ENP01146? 784273 \ \ ChemRiSkTM - A McLaren Company NOVEMBER 28, 1989 Page 35 ' | Koclba, R. J. and 0. Cabey. 1985, Comparative toxicity and biologic activity of chlorinated dibenzo-p-dioxins and furans relative to 2,3,7, 8-tetrachlorodibenzo-p-dioxin (TCDD) . Chemosphere 14(6/7) :649-660. Krasovskii, G.N. 1976. Extrapolation of experimental data from animals to man. Environ. Health Perspect. 13:51-58. Lathrop, G.D., P.M. Moynahan, R.A. Albanese and W.D. Wolfe. 1983. An epidemiologic investigation of health effects in Air Force personnel following exposure to herbicides. Baseline morbidity study results. June 30. USAF School of Aerospace Medicine (AFSC), Brook3 Air Force Base, TX. Lathrop, G.D., W.D. Wolfe, R.A. Albanese and P.M. Moynahan. 1984. An epidemiologic investigation of health effects in Air Force personnel following exposure to herbicides. Executive summary baseline morbidity study. In: Biological Mechanism? of Dioxin Action. Banbury Report 18. A. Poland and R.D. Kimbrough (eds.). Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. pp. 471-474. Leung, H., R.H. Ku, D.J. Paustenbach and M.E. Anderson. 1988. A physiologically based pharmacokinetic model for 2,3,7, 8-tetrachlorodibenzo-p-dioxin in C57BL/-6J and DBA/2J mice. Toxicol. Letters 42:15-28. Leung, H., D.J. Paustenbach, F.J. Murray' and M.E. Anderson. 1989. A physiological pharmacokinetic description of the tissue distribution and enzyme inducing properties of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Toxicol. Appl. Pharmacol, (in press). Lipson, A., W.R. Gaffey and F. LaVecchio. 1983. Agent Orange and birth defects. Mew England J. Med. 309(81 :491-492. Lutz, R.J. and R.L. Dedrick. 1987. Implications of pharmacokinetic modeling in risk assessment analysis. Environ. Health Perspect. 76:97-106. Maine Department of Human Services (Maine DHS). 1988. Policy for Identifying and Assessing the Bealth Risks of Toxic Substances. Norman T. Anderson, Environmental Toxicology Program, Division of Disease Control, Bureau of Health, Augusta, ME. Maronpot, R.R., C.A. Montgomery, G.A. Boorman and E.E. McConnell. 1986. National Toxicology Program nomenclature for hepatoproliferative lesions of rats. Toxicol. Pathol. 14(2):263-273. Massachusetts Department of Environmental Quality Engineering (Mass DEQE) . 1988. Draft Interim Guidance for Disposal Site Risk Characterization - In Support of the Massachusetts Contingency Plan. Office of Research and Standards, Boston, MA. GENP 011463 784274 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 36 Mastroiacovo, P., A. Spagnolo, E. Marni, L. Meazza, R. Bertollini and G. Segni. 1988. Birth defects in Che Seveso area after TCDD contamination. JAMA 259(11):1668-1672. McConnell, E.E., H.A. Solleveld, J.A. Swenberg and G.A. Boorman. 1988. Guidelines for combining neoplasms for evaluation of rodent carcinogenesis studies. In: Carcinogenicity, The Design, Analysis, and Interpretation of Long-term Animal Studies. H.C. Grice and J.L. Cimineri (eds.). ILSI Monographs. Springer-Verlag, New York, NY. pp. 183-196. m Minister of Veterans' Affairs. 1983. Case-control -Study of Congenital Anomalies and Vietnam Service (Birth Defects Study). Australian Government Publishing Service, (cited in AMA, 1984). Moolgavfcar, S.H. and D.J. Venzon. 1979. Two-event models for carcinogenesis. Incidence curves for childhood and adult tumors. Math Biosci. 47:55-77. Moolgavkar, S.H. and A.G. Knudson. 1981. Mutation and cancer: a model for human carcinogenesis^. J. Natl. Cancer Inst. 66:1073-1052. Mordenti, J. 1985. Pharmacokinetic scale-up: Accurate prediction of human pharmacokinetic profiles from animal data. J, Pharmacol. Sci. 74(10):10971099. Mordenti, J. 1986. Man versus beast: Pharmacokinetic scale-up in mammals. J. Pharmacol. Sci. 75(11):1028-1040. Moses, M., R. Lilis, K.D. Crow, J. Thornton, A. Fishbein, H.A. Anderson and l.J. Selikoff. 1984. Health status of workers with past exposure to 2,3,7,8tetrachlorodibenzo-p-dioxin in the manufacture of 2,4,5-tetra-chlorophenoxyacetic acid: Comparison of findings with and without chloracne. Am. J. 2nd. Med. 5:161-182. Murray, F.J., F.A. Smith, K.D. Nitchkee, C.G. Humiston, R.J. Kociba and B.A. Schwetz. 1979. Three-generation study of rats given 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the diet. Tox. Appl. Pharm. 50:241-252. National Council of Paper Industry for Air and Stream Improvement, Inc. (NCASI) . 1987. Dioxin: A Critical Review of its Distribution, Mechanism of Action, Impacts on Human Health, and the Setting of Acceptable Exposure Limits. NCASI Technical Bulletin No. 524. New York, NY. Nelson, C.J., J.F. Holson, H.G. Green and D.W. Gaylor. 1979. Retrospective study of the relationship between agricultural use of 2,4,5-T and cleft palate occurrence in Arkansas. Teratology 19:377-384. National Toxicology Program (NTP) . 1982. Carcinogenesis .Bioassay of 2,3, 7,8~tetrachlorodibenzo~p-dioxin (CAS Ho. 1746-01 -6> in Osborne-Mendel Rats and B6C3F1 Mice Savage Study;. Technical Report Series No. 209. NIEHS, NC. GENP 011464 784275 CheaRiakTM - A McLaren Company NOVEMBER 28, 1989 Page 37 Office of Science and Technology Policy (OSTP). 1965. Chemical Carcinogens: A Review of the Science and its Associated Principles. Federal Register 50(50):10372-10442. Ogawa, K., T. Onoe and M. Takeuchi. 1981. Spontaneous occurrence of gamma glutamyl transpeptidase-positive hepatocyte foci in 105-week old Wistar and 72week old Fischer 344 rats. J. Nat, Cancer Inst, 67:407-412. Ontario Ministry of the Environment. 1985. Polychlorinated Dibenzo-p-Dioxins (PCDDs) and Polychlorinated Dibenzofurans (PCDFs). Scientific Criteria Document for Standard Development No. 4-84. Prepared for Intergovernmental Relations and Hazardous Contaminants Coordination Branch, Toronto, Ontario, Canada. Ontario. 1988. K r a f t M i l l E f f l u e n t i n O n t a r i o . Expert Committee on Kraft Mill Toxicity. Toronto, Ontario, Canada. March 29. O'Flaherty, E.J. 1988. Differences in metabolism at different do3e levels. In: Toxicological Risk Assessment. Volume X. Biological and Statistical Criteria. D.B. Clayson, D. Krewski and I. Munro. (eds.) . CRC Press, Boca Raton, FL. pp. 53-90. Patterson, D., R. Hoffman, L. Needham, D. Roberts, J. Bagby, J. Pirkle, H. Falk, E. Sampson and V. Houk. 1986. 2,3,7,8-Tetrachlorodibenzo-p-dioxin levels in adipose tissue of exposed and control persons in Missouri. An interim report. JAMA 256(19)2683-2606. Paustenbach, D.J., H.P. Shu and F.J. Murray. 1986. A critical examination of assumptions used in risk assessments of dioxin contaminated soil. Reg. Toxicol. Pharmacol. 6:294-307. Pazderova-Vejlupkova, J., M. Nemcova, J. Pickova, L. Jirasek and E. Lukas. 1981. The development and prognosis of chronic intoxication by tetrachlorodibenzo-p-dioxins in men. Arch. Environ. Health 36(1):5-ll. Pearce, N., A.H. Smith, J.R. Howard, R.A. Sheppard, H.J. Giles and C.A. Teague. 1986. Non-Hodgkin's lymphoma, and exposure to phenoxy herbicides, chlorophenols, fencing work and meat works employment: A case control study. Br. J. Ind. Med. 43:75-03. Pitot, H., T. Goldsworthy, H. Campbell and A. Poland. 1980. Quantitative evaluation of the promotion by 2,3,7,8-tetrachlorodibenzo-p-dioxin of hepatocarcinogenesis from dimethylnitrosamine. Cancer Res. 40:3616-3620. Pocchiari, F., V. Silano and A. Zampieri. 1979. Human health effects from accidental release of tetrachlorodibenzo-p-dioxin (TCDD) at Seveso, Italy. Ann. NY Acad.Sci. 320:311-320. Poland, A. and E. Glover. 1979. An estimate of the maximum in vivo covalent binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin to rat liver protein, ribosomal RNA, and DNA. Cancer Res. 39:3341-3344. - GENP 011465 784276 ChemRisfcTM - A McLaren Company NOVEMBER 28, 1 9 3 9 Page 38 Poland, A. and J. Knutson. 1982. 2,3,7, 8-Tetrachlorodibenzo-p-dioxin and related halogenated aromatic hydrocarbons: Examination of the mechanisms of toxicity. Ann. Rev. Pharmacol. Toxicol. 22:517-554. Portier, C.J., 0.6. Hoel and J.V. Ryzin. 1984. Statistical analysis of the carcinogenesis bioassay data relating to the risks from exposure to 2,3,7,8tetrachlorodibenzo-p-dioxin. In: Public Health Risks of the Dioxins. W.W. Lowrance (ed.) . William Kaufmann, Los Altos, CA. pp. 99-119. Potter, C.L., G. Sipes and D.H. Russell. 1983. Hyperthyroxinemia and hypothermia in rats response to 2,3,7,8-tetrachlorodibenzo-p-dioxin administration. Toxicol. Appl. Pharmacol. $9:89 (cited in Kociba, 1983). Rietz, R.H., J.N. McDougal, M.W. Himmelstein, R.J. Nolan and A.M. Schumann. 1988. Physiologically based modeling with methylchloroform: Implications for interspecies, high dose/low dose, and dose route extrapolations. Toxicol. Appl. Pharmacol. 95:185-199. '\ Reggiani, G. 1978. Medical problems raised by the TCDD contamination in Seveso, Italy. Arch.Toxicol. 46:161-188. Reggiani, G. 1980. Acute human exposure to TCOD in Seveso, Italy. J. Toxicol. Environ. Health tf(l):27-43. Rodricks, J.V., S.M. Brett and G.C. Wrenn. 1987. Significant risk decisions in federal regulatory agencies. Reg. Toxicol. Pharm. 7:307-320. Rozman, K ., T. Rozman and H. Greim. 1984. Effect of thyroidectomy and thyroxine on 2,3,7,3-tetrachlorodibenzo-p-dioxin (TCDD) induced toxicity. Toxicol. Appl. Pharmacol. 72:372. (cited in Kociba, 1983). Ryan, J.J., R. Lizotte and B. Lau. 1985. Chlorinated dibenzo-p-dioxins and chlorinated dibenzofurans in Canadian human adipose tissue. CAemosphere 14(67) .-697-706. Schu Ite-Hermann, R. and W. Parzefall. 1981. Failure to discriminate initiation from promotion of liver .tumors in a long-term study with the phenobarbitaltype inducer alpha-hexachlorocyclohexane and the role of sustained stimulation of hepatic growth and monooxygenases. Cancer Res. 41:4140-4146. Schu Ite-Hermann, R,, X. Timmermann-Trosiener and J. Schuppler. 1983. Promotion of spontaneous preneoplastic cells in rat liver as a possible explanation of tumor production by nonmutagenic compounds. Cancer Res. 43:839-844. Sielken, R.L. 1987. Quantitative cancer risk assessments for 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD). Fd. Chem. Toxic. 25(3) :257-267. Sielken, R.L. 1988. A critical evaluation of a dose-response assessment for TCDD. Sielken's Response. Fd. Chem. Toxic. 26(l):79-83. GENP 011466 784277 ChemRiakTM - A McLaren Company NOVEMBER 28, 1989 Page 39 Singer, R., M. Moses, J. Valciukas, R. Lilis and I.J. Selikoff. 1982. Nerve conduction velocity studies of workers employed in the manufacture of phenoxy herbicides. Environ. Res. 29:297-311. Shu, H.P., O.J. Paustenbach and F.J. Murray. 1987. A critical evaluation of the use of mutagenesis, carcinogenesis, and tumor promotion data in a cancer risk assessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Reg. Toxicol. Pharm. 7:57-88. Smith, F.A., B.A. Schwetz and K.D. Nitschke. 1976. Teratogenicity of 2,3,7, 8-tetrachlorodibenzo-p-dioxin in CF-1 mice. Toxicol. Appl. Pharmacol. 38:517-523. Smith, A.H., D.O. Fisher, N. Pearce and C. Chapman. 1982. Congenital defects and miscarriages among New Zealand 2,4,5-T sprayers. Arch. Fnviron. Health, 37(4) -.197-200 . Smith, A.H., D.O. Fisher, H.J. Giles and N. Pearce. 1983. The New Zealand 3 0 ft tissue sarcoma case-control study. Interview findings concerning phenoxyacetic acid exposure. Chemosphere 12(4/5) :565-571. Smith, A.H. and N.E. Pearce. 1986. Update on soft tissue sarcoma and phenoxyherbicides in New Zealand. Chemosphere 15:1795-1798. Squire, R.A. and M. H. Levitt. 1975. Report on a workshop on classification of specific hepatocellular lesions in rats. Cancer Res. 35:3214-3223. Suskind, R.R. and V.S. Hertzberg. 1984. Human health effects of 2,4,5-T and its toxic contaminants. JAMA. 251 (18):2372-2380. Thorslund, T. 1987. Ouantitative dose "response model for the tumor promoting activity of TCDD. XCF-Clement Associates, Washington, D.C. Prepared for the Carcinogen Assessment Group, U.S. Environmental Protection Agency, Washington, D.C. Travis, C.C. and H .A. Hattemer-Frey. 1987. Human exposure to 2,3,7,8-TCDD. Chemosphere. 16(10-12) :2331-2342. Travis, C.C., S.A. Richter, E.A. Crouch, R. Wilson and E. Wilson. 1987. Cancer risk management: A review of 132 federal regulatory decisions. Environ. Sci. Technol. 2 1 (5):41S-420. Travis, C.C. 1989. Interspecies extrapolation. In: Biologically Based Methods for Cancer Risk Assessment. C.C. Travis (ed.). Proceedings of a NATO Advanced Research Workshop on Biologically Based Methods for Cancer Risk Assessment, June 11-16, 1988. Plenum Press, New York, NY. pp. 63-78. 784278 GENP 011467 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 40 Universities Associated for Research and Education in Pathology, Inc. (UAREP) . 1988. Human Health Aspects of Environmental Exposure to Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans. Report of the Adhoc Panel, Chairman Robert G. Stowell, M.D., Ph.D. United Kingdom (UK). 1989. Dioxins in the Environment. Report of an Interdepartmental Working1 Group on Polychlorinated Dibenzo-para-dioxins (PCDDs) and Polychlorinated Dibenzofurans (PCDFs). Pollution Paper No. 27. Department of the Environment, Central Directorate of Environmental Protection, London. Van der Heijden, C., A. Knaup, P. Kramers and M. Van Logten. 1982. Evaluation of the Carcinogenicity and Mutagenicity of 2, 3, 7, 8-Tetrachlorodibenzo-1,4-Dioxin (TCDD): Classification and No-effect Level. Report DOC/LCM 300/292. State Institute of National Health. Bilthoven, The Netherlands. Weinstein, B. 1984. Dioxins as carcinogenic promoters. In: Health Risks of the Dioxins. W.W. Lowrance (ed.). Proceedings of a Symposium, October 19-20, Rockefeller University, New York, NY. pp. 155-160. Williams, G. and J. Weisburger. 1986. Chemical carcinogens. In: Casarett and Dovll's Toxicology. The Basic Science of Poisons. Chapter 5. C. Klaassen, M. Amdur and J. Doull (eds.). MacMillan Publishing Co., New York, NY. pp.99-173. Wiklund, J. and L. Holm. 1986. Soft tissue sarcoma risk in Swedish agricultural and forestry workers. J. Natl. Cancer Inst. 76:229-234. Wrenn, G. 1986. U.S. Environmental Protection Agency Asbestos Ban and Phase out Proposal: Testimony of Grover Wrenn, Minutes of ENVIRON Corp., Waashington, D.C. Zack, J .A. and W.R. Gaffery. 1983. A mortality study of workers employed at the Monsanto Company plant in Nitro, West Virginia. Environ. Sci. Res. 26:575-591. 784279 GENP 0 1 1 4 6 8 ChamRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 41 7.0 GLOSSARY 07 TERMS Acceptable Daily Intake (ADI) The amount of a chemical to which a person can be exposed on a daily basis over an extended period of time (usually a lifetime) without suffering a deleterious effect. Adenoma A benign epithelial tumor in which the cells form recognizable glandular structures or in which the cells are clearly derived from glandular epithelium. B2 carcinogen Probable human carcinogen; SPA weight-of-evidence category for an agent which has sufficient evidence of carcinogenicity in animals, but inadequate evidence of carcinogenicity in humans. Cancer potency The incremental excess cancer risk per unit of exposure (usually in mg/kg-day) derived from the upper-bound estimate of the lowdose slope of the dose-response curve. Carcinogenic Possessing the capacity to initiate or promote cancer; producing carcinoma. Carcinoma A malignant new growth made up of epithelial cells, tending to infiltrate the surrounding tissues and give rise to metastase3. Chloracne An acneiform skin condition caused by exposure to chlorinated hydrocarbons. Da msnifastis risk Term used, to identify those risks which are of obvious or evident concern. Da miaimis risk Term used to characterize risks which are insignificant, negligible or of no concern, such that regulatory action is unwarranted. Dose The quantifiable amount of a material introduced into an animal either through a route of administration or as a result of exposure, e.g., injection, ingestion, inhalation, or dermal contact. Dose-response curve A mathematical function describing the relationship between the dose (i.e., the quantity) of the* chemical administered to the organisms and the percentage response of the test population. The curve is plotted as response versus dose. Epidemiology The study of the distribution and determinants of diseases and injuries in human populations. Exposure Contact of a receptor organism with a chemical, biological, or physical agent which can be quantified as the amount of the agent available at the exchange boundaries of the organism (e.g., skin) and available for absorption. 784280 ChamRiaJcTM - A McLaren Company NOVEMBER 28, 198 9 Page 42 7oci (hepatic) Small lesions of less than one liver lobule in size; do not show any distruptions of the normal hepatic architecture; may be part of the spectrum of lesions capable of progressing to hyperplastic nodules. The first observable morphological hepatocellular lesion following initiation. Genome The complete set of hereditary factors, as contained in the haploid assortment of chromosomes. Genotoxic DNA-reactive; induces an adverse effect on the genetic material (DNA) of living cells that may be expressed as a mutagenic or carcinogenic event. Hirsutism Abnormal, excessive growth of hair. Hepatocyte A liver cell. Hyperpigmentation Increased pigmentation. Hyperplastic Nodule A small mass of tissue, either normal or pathological that has an increase in the number of normal cells. Hyperplasia. The multiplication or increase in the number of normal cells in normal arrangement in a tissue. Initiator A carcinogen which, if not already electrophilic, undergoes metabolic transformation to an electrophile and reacts covalently with DNA such that the affected cell incorporates the DNA damage into it3 replicating genome. LCL (95% Lower Confidence Limit) Dose for which there is a 95% statistical confidence that the true RsD is no lower than that value. LDgo (median lethal dose) The statistically derived single dosage of a substance that can be expected to cause death in 50 percent of the animals tested. Lipophilic Having an affinity for fats or oils. LOAHL (Lowest-Observed-Adverse-Effect-Level) The minimum dose level, determined from chronic toxicity studies in laboratory animals, at which the first indications of adverse (toxic) effects are observed in the species tested. Metaplasia The change in the type of differentiated cells in a tissue to a form which is not normal for that tissue. Metastasis The transfer of disease from one organ or part to another not directly connected with it. The capacity to metastasize is a characteristic of all malignant tumors. GENP 011470 784281 ChemRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 43 MLS (Maximum Likelihood Estimate) Derived from the linearized multistage-model; the model's best estimate of the RsD. Mutagenic Inducing genetic mutation or change in genetic material. Mutation A permanent transmissible change, in the. genetic material, usually in a single gene; the change may be in the form of a loss (deletion), gain (translocation), or exchange (transduction) of genetic material. Neoplasia The formation of a neoplasm, i.e., the progressive multiplication of cells under conditions that would not elicit, or would cause cessation of, multiplication of normal cells. NOAEL (No-Observed-Adversa-Effeet-Level) The highest dosage administered to laboratory animals that does not produce toxic effects. N o d u l e A small mass of tissue in the form of a swelling, knot, or protuberance; either normal or pathological. Pharmacokinetic model Quantitative predictive approach used to describe and predict events occurring during the process of drug or chemical disposition throughout the body, thus yielding tissue levels of the drug or chemical. Pharmacokinetics The study of the action of a chemical in the body over a period of time, including the process of absorption, distribution, localization in tissues, biotransformation, and excretion. Physiologically-Based Pharmacokinetic (PB-PK) Modal Quantitative predictive approach which utilizes physiological parameters of the experimental animals to describe the pharmacokinetic process; allows for prediction of the relationship between administered concentrations of a drug and the resulting concentration found in target tissues. Porphyria cutanea tarda Disorder characterized by disturbances of the metabolism of porphyrins and chronic skin lesions ranging from slight skin fragility to severe scarring. Promoter A substance which, after repeated or prolonged exposure, acts to increase the tumorigenic response of a cell exposed to an initiator. Promoters are thought to affect cellular growth and differentiation and may alter a number of cell membrane properties. Promoters exhibit a threshold in their dose response and tumor promotion may be reversible upon removal of the promoter. Reference Dose (RfD) An estimate of a daily exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime. The EPA RfD is derived from the LOAEL or NOAEL by application of uncertainty factors or safety factors. GENP 011471 784282 ChamRiskTM - A McLaren Company NOVEMBER 28, 1989 Page 44 Risk A statistical concept defined as the expected frequency or probability of undesirable effects resulting from a specified exposure to known or potential environmental concentrations of a substance. A substance is considered safe if the risks associated with its exposure are judged to be de minimis. Estimates of risk may be expressed in absolute or relative terms. Absolute risk is the excess risk due to exposure. Relative risk is the ratio of the risk in the exposed population to the risk in the unexposed population. Risk-specific Dose (RsD) The dose estimated to result in an upper-bound estimate of incremental lifetime cancer risk, such as 1 in 100,000. Statistical significance An inference that the probability is low that the observed difference in quantities being measured could be due to variability in the data rather than an actual difference in the quantities themselves. The inference that an observed difference is statistically significant is typically based on a test to reject one hypothesis and accept another. Threshold concentration A concentration above which some effect (or response) may be produced and below which it will not. Toxicant An agent or material capable of producing an adverse response (effect) in a biological system, seriously injuring structure or function or producing death. Toxic Equivalency Factor (TEF) Conversion factor used to quantify toxicity of measured dioxin and furan congeners relative to 2,3,7,8-TCDD. Toxicity The inherent potential or capacity of a material to cause adverse effects in a living organism. Virtually Safa Dose (VSD) See Risk-specific Dose. Term used by CDC which is synonymous with the term RsD which is currently used by the EPA. 784283 Occupational exposure. R. Kimbrough and P. Grandjean, 1989. Purpose: A review o f occupational exposure to PCBs and other compounds, focusing on exposure potential, retention levels in human tissues and body fluids, and systemic effects. Conclusion: Exposure scenarios have been docum ented since W orld War I, o f which chlorance is the m ost characteristic clinical effect caused by PCB compounds. Several system ic effects have been linked to PCBs, including liver dysfunction and respiratory irritation. Occupational exposure recom mendations range from 0 .5 to 1 .0 m g/m 3 in other countries, while in Sweden and Japan have lim its o f 0.01 and 0.1 m g/m 3, respectively. Exposure History: In World W ar I, chlorinated naphthalenes were used for production of gas masks and material for detonators. Chlorinated biphenyls were used for insulating material for cables and condensers. These compounds were used extensively in the shipbuilding industry during World W ar II. PCBs were introduced in the early 1930s for dielectric fluids for capacitors and transformers. Repair and maintenance exposure levels for PCBs ranged up to 60 Mg/m3, while surface contamination ranged from 4-60 Mg/m2 and chronic leaking was 33 Mg/m2. Skin exposure from capacitor production was 5 Mg/cm2. PCB levels in blood from chronic exposure have been found to range from 6-3500 ng/ml. Exposure Effects: Exposure to chlorinated compounds was revealed by the appearance of chloracne, which is the most characteristic clinical effect caused by these agents. The onset of chloracne occurred within a few days of exposure, as well as, being delayed for several weeks. Lesions formed on the face, neck, skin area behind the ears, back, arms, and legs. In males, the genitals were also involved.. Chloracne remained active fo r many years, with deep-pitted scars remaining as a residual effect. Treatment has included vitamin A acid , UV light, X-ray, lancing and expressing of pustules. Temporary remissions were achieved with accutane, but possible teratogenic effects are a side effect. Chloracne was thought to be due to external contact and not the result of systemic exposure to the acnegenic agents. Accidental ingestion of chloracnegenic agents resulted in a variety o f adverse effects, including chloracne. The identity o f all agents which may be chloracnegenic in humans has not been established. Most reported cases of PCB-related chloracne involved exposure to vapors which developed when PCBs or mixtures of chlorinated terphenyls were heated. It is extremely difficult to relate particular effects to individual compounds, due to the occurrence of several compounds in occupational exposure. GENP 011473 784284 Occupational exposure, (continued) Systemic Effects: Specific effects attributed to PCBs are difficult to describe. Chloracne was present in all 17 workers engaged in PCB production. Other symptoms included loss of appetite and libido and lassitude. One study reported complaints of burning eyes, nose and throat, dry throat, nausea, and dizziness. Three fatalities from jaundice prompted a study of systemic effects o f PCBs. Recent studies include PCB-related changes in hepatocytemorphology. Liver dysfunction and chloracne occurred in workers exposed to PCB vapors for 5-14 months. Capacitor production workers complained of face and skin bums, persistent body odor, and eczematous rashes. In other studies, liver injury was found. Acute PCB exposure has caused elevated serum triglycerides, total cholesterol and phospholipids. Experimental animal studies indicated that lipid metabolism in the liver may be affected by halogenated biphenyls. Higher serum lipid levels may be directly responsible for increased serum PCB levels, since PCBs are lipid-soluble. Thus, one study found correlations between serum PCB and serum lipids to disappear when the PCB concentration was expressed on a Lipid basis. The induction of mixed-function oxidases is another effect o f PCBs on the liver. Exposed workers had a plasma antipyrine half-life significantly lower than controls. Upper respiratory irritation was reported by 48% of 326 workers in a capacitor plant, with a restrictive pattern of impairment demonstrated from spirometric studies. Immunotoxicity may provide a mechanism for some adverse effects, although the evidence available is very limited. Cancer: Research has been delineated into two main areas. First, cancer incidence rates have been determined in cohorts o f workers with documented exposures to chlorinated compounds. Second, case-referent studies have mainly linked soft-tissue sarcomas to usage of herbicides with possible dioxin contaminants. W orkers exposed to a PCB mixture in a refinery plant had significantly higher than expected rates for pancreatic cancer. A cohort study of 2567 workers from a capacitor plant followed for 15 yrs, with PCB exposure o f 2 yrs or less, resulted in increased deaths due to liver cancer, liver cirrhosis and rectal cancer. In Italy, increased cancer mortality occurred in capacitor plant workers. Prevention: In 1973, OECD requested member countries not to use PCBs except for enclosed systems in transformers, large capacitors, heat-exchangers and hydraulic systems in mines. A subsequent directive from the CEC prohibited use of products containing, more than 0.1% PCBs and PCTs. OSHA has determined a PEL of lm g/m 3 for PCBs containing 42% N p 0 , l4 7 4 784285 Occupational exposure. (continued) chlorine and 0.5 mg/m3 for PCBs"with 54% chlorine. NIOSH has recommended that exposure to PCBs in the workplace be limited to or below the minimum reliable detectable concentration of 1 Mg/m3. According to the ILO and WHO, occupational exposure to PCBs in other countries ranged from 0.5 mg/m3 to 1.0 mg/m3. In Sweden, the exposure limit for all PCBs is 0.01 mg/m3, with a short term limit o f 0.03 mg/m3; while in Japan the limit is 0.1 mg/m3. GENP 011475 784286 Specific PCB congener distribution in adipose tissue of Canadians. Jos Mes et al. 1990. Purpose: Human adipose tissue was analyzed for 34 selected PCB congeners o f which 24 were confirmed. Conclusion: Several congeners were fo u n d to be prevalent in adipose tissue o f Canadians; while other congeners were fo u n d to have a higher percentage in fem ales and different age groups. Remarks: Selected congeners represented greater than 85 % o f total PCBs in human milk fat. The possible effects of age, sex, and geographical area of sampling on PCB congener distribution were explored. Samples were taken from the abdominal and kidney region o f 81 males and 27 females, with an average age of 42 yrs (range 3-85 yrs). Results: Congeners 138, 153 and 180 made up 52% of all congeners found in adipose tissue of Canadians, therefore being the major PCB contributors to body burden. Lower total congener levels were found in the western region of Canada than the other regions. Since there were no outstanding differences among the regions, the authors assumed a common source of contamination. The percent distribution o f congeners 52, 99, and 157 was significantly higher in females than in males. Also the percentage of many tri- to hexachlorobiphenyls were significantly higher in breast milk than in adipose tissue, while many hepta- to nonachlorobiphenyls were significantly lower. Residue levels of PCB congeners 156, 180, 189, 194, 201, 203, 206 and 209 were significantly lower in the 0-25 age group than in the > 5 1 year age group. Previous observations have shown a positive correlation between PCB residue levels and age, based on total PCBs, not individual congeners. GENP0JJ47S 784287 Organochlorine pesticides and PCBs in tissues from Dutch citizens (1968 -1986). P.A. Greve and P. Van Zoonen. 1990. Purpose: To present results of monitoring data of occurrence of organochlorine pesticides and PCBs in humans (adipose tissue, milk, blood) from 1963-1986 in Dutch citizens to compare with results from other countries. Conclusion: Only a lim ited num ber o f organocholrine compounds can be m onitored satisfactorily in blood due to low concentrations. The values fro m this study can serve as 'reference' or norm al values, even though variations in blood levels are higher than those in adipose tissue. The tim e trends appear to be a good instrum ent fo r m onitoring the effect o f certain measures against the use o f certain persistent fat-soluble ,chemicals. Remarks: Earlier old methods were compared with new analytical methods to ensure comparable results over a 20 year period. Very carefully detailed and checked for accuracy of methods. Adipose tissue was collected from older adults post mortem. Milk samples were collected from mothers at 2 maternity centers and blood samples were taken from 19 year old males. Samples are not random and some groups are overrepresented. Results: One o f the objectives o f the study was to establish possible time trends for the compounds investigated. p .p '-DDT had a downward trend reflecting the ban o f this compound in many countries. The metabolite p .p '-DDE does not show a tendency to decrease in concentration due to more resistance to degradation and excretion than p .p '-DDT. HCB had an increase in the early 1970s associated with the increase in products of animal origin. Recently, measures taken to avoid contamination have lowered HCB concentration in tissue. No trends were found for dieidrin although use of dieldrin and precursor aldrin were strongly reduced in last 15 years. Little, if any, downward trend seen for B-HCH which was surprising due to prohibition of BHC mixtures in many parts of the world on edible crops and animals. The persistance o f B-HCH in fatty tissues is high. No significant downward trend was found for PCBs, although their use has been reduced over the course of the study and the effect of these prohibitory measures remains invisible from the data. Other organochlorine levels were low and levels found in milk samples had a more profound downward trend. The highest influence o f age was found for p .p '-D D E and PCBs followed by HCB and B-HCB. The levels in blood follow the same trend as the ones for adipose dssue and milk. Significant increases o f the levels in milk were observed during lactation for HCB, B-HCB and PCBs; while a significant decrease resulted for p .p '-DDE. Several other influences were GNP 011477 784288 Organochlorine pesticides and PCBs in tissues from Dutch citizens (1968 -1986). (continued) found to have no significant effect including: sex, origin (rural or urban), eating habits, or medication. GENP 011478 784289 Regional differences of PCB and PCQ concentrations in the blood and subcutaneous fat tissue of residents of Nagasaki. T. Ohgami et al. 1989. Purpose: To analyze PCB and PCQ concentrations in the subcutaneous fat tissue of residents of Nagasaki Prefecture and compare these levels between blood and subcutaneous fat tissue of people living in different regions. Conclusion: Regional differences were fo u n d fo r PCB blood and tissue levels, with fishery area sam ples having higher values. It was assum ed that dietary exposure was greater fo r fishery regions than fo r agricultural or urban regions. Remarks: Samples were obtained from seventy-one people surveyed with no history of PCB contamination living in urban, agricultural, fishery and combined fishery and agriculture areas. Results: * Regional mean PCB concentration levels ranged from 1.39 to 6.0 ppb in blood and 320 to 1942 ppb in fat tissue and were higher for fishery areas than the urban or agricultural areas. PCB concentrations in subcutaneous fat tissue were 100 times higher than that in blood samples. PCQ concentrations detected'in almost all subcutaneous fat tissue but with no regional differences. PCQ concentrations in blood were non-detectable. Discussion: The authors concluded that it is a necessity to clarify occupations, diet, and living environment in controls due to regional differences. GNP 01H79 784290 Coplaner PCBs in Swedish hum an milk. K. Noren et al. 1990. Purpose: To find the occurrence of PCB congeners (tetrachloro, pentachloro, hexachlorobiphenyl) in the Swedish population and investigate the long term trends of these concentrations. Conclusion: Non-ortho coplaner PCBs (77, 126, 169) were excreted in human milk. The concentrations o f non-coplaner PCBs were higher than those o f the most toxic PCDDs and PCDFs. The m ost toxic PCB, 3 ,3 \4 ,4 ',5 -P e C B (126) was the m ost abundant non-ortho coplaner PCB in Swedish hum an m ilk. The concentrations o f the PCB congeners decreased during the tim e period studies and only m inor changes in distribution o f congeners occurred fro m 1976 to 1989. Remarks: M ilk was taken from two mothers (25 and 36 years old) in 1989 during nursing of their first infant. The pooled samples were collected at the Mothers M ilk Center in Stockholm in 1972, 1976, 1980 and 1988/89 with 60, 204, 305 and 120 samples, respectively. Results: PCB levels were higher in milk from the older mother at 0.97 ug/g fat compared to 0.76. Total PCBs and DDT were also higher in the older mother. The distribution of congeners were similar despite the different total levels and the age o f the mothers. The average levels o f non-coplaner PCBs in milk from the M others M ilk Center decreased from 1972 to 1989. PCB 126 was the most abundant among the congeners and is considered the most toxic with a response and potency similar to TCDD. Levels o f PCB 126 are 50 times higher than those of TCD D , which were 1-6 pg/g fat. GENP 011480 784291 The relation of polychlorinated biphenyls to b irth weight and gestational age in the offspring of occupationally exposed m others. P.E. Taylor et al. 1989. Purpose: The authors studied the relation of PCBs to birth weight and gestational age among the live offspring of women occupationally exposed during manufacture of capacitors. Interviews were conducted in 1982 with 200 women having direct exposure jobs and 205 women who never held a direct-exposure related job in order to ascertain information concerning reproductive outcome. Conclusion: The authors concluded that the data indicate a significant relation between increased estim ated serum PCB level and decreased birth weight and gestational age, and that the decrease in birth weight is at least partially related to shortened gestational age. The m agnitude o f these effects is likely to be negligible except among already low birth weight or short gestation limits. Remarks: Exposure was assessed as high-homolog PCB (Aroclor 1254), a continuous exposure variable estimated from an independently derived prediction model. Direct exposure jobs were defined as those in which direct contact with PCBs occuned during the manufacturing process. Indirect-exposure jobs including those performed in office and manufacturing areas where PCBs not directly used. Results: Direct and indirect exposure groups had 172 and 1S4 births analyzed, respectively. A significant effect of high-homolog exposure was seen for birth weight. For gestational age a small but significant decrease was observed with an increase in estimated exposure. GENP 011481 784292 Trends in the levels of some chlorinated hydrocarbon residues in adipose tissue of Canadians. Jos Mes, 1990. Purpose: Presentation of results of nationwide surveys of chlorinated hydrocarbon residues from 1969, 1972, 1976, and 1985 in Canadians and possible effects of age and sex. Conclusion: In general, the adipose tissue burden o f reported residues in Canadians increased with age. A loss o ffa tty deposits during advanced age m ay have been reflected to some extent in the dim inishing num ber o f residues which are significantly different between the 26-50 and > 51 age groups. Age did not appear to affect the depletion o f residues fro m body over the years. Although rases o f elim ination may differ, the same general downward trend was observed regardless o f age group. Remarks: Data was categorized in three age groups, 0-25 yrs, 26-51 yrs, and > 5 1 yrs. Results: Statistical significant differences were observed for all residues between the years of 1972 and 1985. Residues o f dieldrin and D D T decreased steadily and was significant from 1969 to 1985 in both females and males. PCBs significantly decreased in males between 1976 and 1985. DDE residue levels in males decreased significantly until 1976. PCBs and OC1 in adipose tissue of males did not change significantly from 1972 to 1976. There were few significant differences in residue levels between males and females. In 1972, significantly higher PCB levels were recorded for males than females. In general it was found that residue levels tend to increase with age. No definite trend was observed for PCBs. A downward trend was observed in each region between 1969 and 1985. Discussion: PCB use was restricted in 1977, in Canada, with prior surveys showing no decrease in residue levels, but by 1985 adipose tissue burden had halved. There were no consistent and significant difference in residue levels between the sexes. GENP 011482 784293 PCBs and organochlorine pesticides in m ilk of Norwegian women during lactation. J.U . Skaare and A. Polder. 1990. Purpose: To investigate the variation o f organochlorine contamination levels in human milk during lactation and to assess the influence o f parity, dietary and smoking habits. Conclusions: The general tendency seem ed to be a decrease in organochlorines in m ilk fa t during lactation periods. Also findings o f a statistically significant decline during lactation o f PCB and HCB; and a decline in D D E dem onstrated that m others decrease their body burden o f these compounds by m ilk excretion. Remarks: There were a limited number o f subjects (n = 14) as the data were pooled, using data only from first child lactation period. Eight of the women were Norwegian natives while six others were from Europe with at least 7 years of residency in Norway. Results:' HCB (hexachlorobenzene) and PCB levels were significantly reduced during the period of observation. DDE levels were also reduced but not significantly. There were significant declines in the first child only group for HCB, DDE, PCB but not for the second child group. The mean levels of PCB, HCB, and DDE in `the first child group were higher than corresponding levels in milk for the second or third child groups. PCB, DDE, and HCB levels reduced 80% from first to second child lactation periods starting 3.5 years later. The highest HCB levels were found in milk from non-smoking mothers bom outside of Norway. Discussion: Overall levels o f HCB, PCB and DDE were lower than levels found earlier in Norway in 1972, 1978 and 1981. Mean levels of organochlorines in human milk are low compared to corresponding concentrations found in other European countries. Higher HCB levels in mothers bom outside o f Norway are assumed to reflect previous exposure, since they correspond to milk fat levels reported in their home countries. Also, HCB in Norway is an industrial waste product and not directly used in agriculture. The findings o f higher levels o f PCB, D D E, and HCB in first lactation agreed with earlier studies. No pronounced changes in levels during the second child lactation were attributed to short accumulation time from first child and a corresponding low contamination of these compounds resulting from a ban on DDT (1971) and restriction (1971) and ban (1980) of PCB use. GENF 011483 784294 Yu-Cheng W J . Ragan, 1989. Purpose: A review o f the literature concerning the oil disease epidemic known as YuCheng, occurring in Taiwan in 1979. Conclusion: Comparisons to the Yusko outbreak in Japan show ed that total PCBs were about 900 ppm to 30-90 ppm fo r Yu-Cheng rice oil. Although Yu-Cheng patients consumed ten tim es the amount o f contam inated oil, the same amount o f PCBs and PCDFs were fo u n d fo r both outbreak patients. Some evidence has been reported fo r a decline in PCB blood level concentrations fo r some congeners, while hexachlorinated biphenyls rem ained persistent. Background: Yu-Cheng refers to a syndrome of chloracne, hyperpigmentation, dilatation and hypersecretion of conjunctival glands caused by the ingestion of contaminated rice oil. Blood from victims as well-as oil from both the purchaser and seller contained PCBs resembling Kanechlor 500, a PCB mixture of Japanese manufacture. As of 1983, 2061 cases had been identified with 5-10% of the cases severe. The cooking oil was speculated as being contaminated from pipe heating tp remove odors and off-colors in the final processing stage. Repeated heating o f the PCB mixture would lead to formation o f PCTs, PCQs and PCDFs. Clinical Features: Eye discharge and/or disturbance of vision was the most common complaint, while general malaise, numbness of limbs and headaches were also persistent. High rates of mucocutaneous pigmentation, acne and deformed nails also occurred. A limited number of patients had neurological testing which showed some sensory and motor nerve slowing, reduced sensory nerve conduction and mildly abnormal EEGs. Examination of offspring bom to affected women showed that exposed children consistent delay in achievement o f developmental milestones. PCBs and similar chemicals induce enzymes of the mixed-function oxidase class. Placentae from four women showed large increases in ary hydrocarbon hydroxylase levels. The persistence of this phenomenon was concluded to be due to the persistence o f the chemicals, since the placental tissue was not present at time of exposure. Outbreak patients suffered more frequent severe skin and respiratory infections and a conviction that they had lowered resistance to illness. Results: Most samples contained about 30-90 ppm total PCBs, compared to 900 ppm in Yusho rice oil. The Taiwan oil had between 0.1 and 1.68 ppm total PCDFs with these amounts strictly correlated with the amount of total PCBs, with a ratio of about 0.1-0.3% . Patients consumption of oil was estimated at a rate of 1.4 kg/month for 2.7 months before they became symptomatic, and then for another six months before the oil was GENP 011484 784295 Yu-Cheng (continued) withdrawn. PCB concentrations ranged from 67-99 ppm and PCDFs from 0.21-0.40 ppm. Thus, patients consumed about 1 g of PCBs and 3.8 mg of PCDFs total. Taiwanese patients consumed roughly ten times as much contaminated oil as the Yusho patients, but got about the same amount of PCBs and PCDFs. PCB levels in blood, as o f 1985, had a mean o f 53.5 ppb. Samples drawn after only two years of exposure found a range of total PCBs from 10-720 ppb, with a mean and median of 38 and 28 ppb, respectively. The values were lower for the Japanese Yusho patients, probably due to a longer post-exposure period. Total PCB concentrations o f 15 ppb were found in samples drawn in 1985, resulting in selective elimination of certain congeners. PCDFs in blood drawn within 6 months of exposure had a median of 0.09 ppb and a range from < 0 .0 0 5 to 0.27 ppb. As of 1983, 24 deaths occurred among exposed patients and 8 deaths among 39 babies bom to affected mothers. Liver disease and hepatoma were the primary reasons for death, although in Taiwan there is an extraordinary prevalence o f hepatitis B, and that cirrhosis and liver cancer are common. Discussion: The majority of information consists of observations without controls on groups who are not well defined. Nevertheless, the evidence from Taiwan coupled with that from Japan leads to a reasonably consistent picture of the illness produced by exposure to thermally degraded PCBs. > GENP 011485 784296 Levels of dioxins, dibenzofurans and other chlorinated xenobiotics in human milk from the Soviet Union. A. Schecter et al. 1990. Purpose: To determine differences in levels between various geographical areas of ' PCDDs and PCDFs. To characterize the existence of chlorinated dioxins dibenzofurans, PCBs and other chlorinated chemicals in the Soviet Union in human tissue, food, wildlife and the environment, with results to be compared with WHO studies on dioxins in human milk. Conclusion: The study documents fo r the fir s t tim e the existence o f dioxins, dibenzofurans, and other chlorinated chem icals in the USSR in human tissue and milk. The findings suggest levels and a pattern o f dioxin and dibenzofuran congeners which perm it fin g erp rin tin g mo f Soviet tissues fro m hum ans in com parison to other countries. Regional differences were noted w ithin the Soviet Union. Remarks: All samples were pooled before analyses with the following breakdown by country: Soviet Union (24), USA (4 0 + ), W. Germany (140+ ). No statistical analyses were reported. Results: The OCDD mean value was markedly lower for Soviet Union (30-80 ppt) in samples on a fat basis, than for W. Germany (185 ppt) and USA (163-303 ppt). The average heptachlorinated dioxin value was also much lower in the Soviet Union (5-16 ppt) than for W. Germany (34 ppt) and USA (50 ppt). Three hexachlorinated dioxins held the same pattern with lower values reflecting less pollution in the Soviet Union. PnCDD values were similar for all'three countries with Siberian samples reflecting a more recent industrialization there with less accumulated pollution due to lower values. TCDD, the most toxic of the dioxins and dibenzofurans, values are similar. For OCDF, the Soviet samples are all below the levels of the USA and W. Germany. Soviet HpCDF mean levels (0.6-2.6 ppt) were less than those from USA (2.4-S.7 ppt) and W. Germany (9.9 ppt). Values were similar for 1,2,3,4,7,8-H xC D F, 1,2,3,6,7,8-H xC D F and 2,3,4,6,7,8-H xC D F in all three countries. For 1,2,3,7,8-PnCDF, the Soviet samples are higher than the USA or W, Germany, possibly due to sampling or different pollution patterns. The highly toxic 2,3,4,7,8-P nC D F was found in W. Germany to exceed the values for the USA and Soviet Union, while TCDF values were similar in all three countries. Mean levels o f dioxins (total PCDD) in the Soviet Union are lower than those in the USA and W . Germany. PCDFs showed little difference among the countries with W. Germany having the highest values. Total PC D D /Fs in human milk are lower in the Soviet Union. GENP011486 784297 Adipose tissue/serum partitioning of chlorinated hydrocarbon pesticides in humans. L.L. Needham et al. 1990. Purpose: To measure the levels o f chlorinated hydrocarbon pesticides in both adipose tissue and serum to investigate how these compounds partition between these two compartments. Remarks: More measurements are taken from serum blood than tissue. Partitioning is expressed as a ratio o f adipose tissue (lipid weight) to serum concentration (lipid weight and albumin content). Results: Data by gender found no significant differences in the concentration ratios for men and women. Concentration ratios were given for adipose tissue and serum blood samples in the following measurements: lipid weight to whole weight lipid weight to lipid weight - lipid weight to albumin weight Distributions of chlorinated hydrocarbons analytes in adipose tissue and serum are presented with concentrations being very close to the detection limits. Discussion: Mean and median concentration ratios are greater than 1 for all analytes. PCB concentration ratios should be approximately 1.0 when calculated on a per unit of lipid. Concentration ratios for various compounds are influenced by how they are transported.in blood. In an earlier study, the authors reported that nearly 100% of in vitro spiked 2,3,7,8-T C D D was found in the plasma portion of whole blood. Comments: Concentration ratios were never explained as to their importance or relevance to a discussion on chemical exposure in humans. 784298 N eurobehavioral dysfunction in firem en exposed to PCBs: possible im provem ent after detoxification. K .H . K ilbum et al. 1989. Purpose: A comparison of neurobehavioral functions of PCB exposed firemen before and after a detoxification regimen and to a comparable non-exposed group. Conclusion: Changes across the detoxification interval were selective. There were minimal changes in affective status betw een the testing sessions (exposed group) which .suggests that subjects perception o f distress d id not im prove It w as concluded that affective disorders were not causing neurobehavioral dysfunction due to only one test being significant correlated to Profile o f M ood States (POMS). The effectiveness o f the detoxification program was not answered conclusively, because memory and cognitive junction may be influenced by prior testing. ,The non-exposed group had other exposures over a 2 to 25 y e a r p e rio d discounting their validity as a referent group. This study suggests that subjects ,with neurobehavioral dysfunction m ay be aided by d ie t exercise, and sauna- induced excretion o f PCBs. Remarks: Fourteen firemen were exposed to fumes, smoke and gases from transformers containing PCBs. Inhalation time was 15 to 30 minutes without a breathing apparatus, while dermal contact occurred with skin and soaked hands with and without gloves. Symptoms began 2 days to 3 months after the fire and included: fatigue, headaches, muscle weakness, aching joints, memory loss, hypertension, insomnia, irritability and loss of balance. Four months after the fire neurobehavioral, medical and biochemical studies were performed. Afterwards a 2-3 week detoxification program was implemented including a regulated diet, exercise twice daily, and a daily sauna (heat stress). Neurobehavioral tests were repeated for the exposed group after completion of the detoxification program. A matched comparison group consisting of fourteen non-exposed firemen from the area served as the control. Results: i Firemen exposed to PCBs had poorer neurobehavioral functions than the unexposed group. There were significant differences for most of the tests performed. Serum PCB and body content of PCBs were not significantly correlated nor were they correlated with any neurobehavioral test. After the detoxification program some of the test scores improved significantly, while only one test worsened significantly. Discussion: i I There is a concern that memory and cognitive function may improve on repeat testing because of familiarity, practice, motivation and learning (although considered unlikely in this study). Another concern is whether affective disorders adversely affect neurobehavioral test scores as does post-traumatic stress. 784299 I Polychlorinated Biphenyls (PCBs): M utagenicity and Carcinogenicity S. Safe (1989) Purpose: To examine the mutagenetic and carcinogenetic properties of polychlorinated biphenyls (PCBs) in laboratory animals and humans. Conclusions: Overall, highly chlorinated PCBs (i.e.: A roclor 1254) are not m utagenic in assays done w ith Salm onella tvphimurium and E. coli. In rodents, highly chlorinated PCBs have been shown to cause hepatocellular carcinomas, adenocarcinomas, and neoplastic nodules. The carcinogenicity o f the lower chlorinated biphenyls has not yet been determined. PCBs can also act as cocarcinogens and anticarcinogens which can enhance or inhibit the tum origenic activity o f other carcinogens. In humans it appears that overall m ortalities and cancer m ortalities resulting from exposure to PCBs is lower than expected, however, there was a significant increase in the num ber o f deaths due to cancers o f the liver, gall bladder, and biliary tract combined. There may be a connection between the degree o f chlorination o f the PCB and exposure. Remarks: This paper'is based on reviews of studies done by other researchers and authors. Results: M utagenicity: As reported by Wyndham et al. (1976), the 4- chlorobiphenyl and the lower chlorinated PCB mixture, Aroclor 1221, were mutagenic to Salm onella typhim urium strain T A 1538 in the presence o f an external source of metabolic activation. The more highly chlorinated biphenyls (Aroclor 1254 and 2 ,2 \5 ,5 '-tetrachlorobiphenyi) were not mutagenic. However, studies done in the same laboratory pater on failed to reproduce the original observations. Others have also reported a lack o f mutagenicity in bacteria o f A rochlor 1254, 4-chlorobiphenyl, 3 ,3 \4 ,4 '-, 2 ,2 ',4 ,4 '- tetrachlorobiphenyl and 2 , 2 \ 4 , 4 ',6 ,6 '-hexachlorobiphenyl (Schoeny et al. 1979). In another study, Heddle and Bruce (1977) reported that Aroclor 1254 did not cause cytogenic effects in mice and Aroclors 1242 and 1254 did not cause cytogenic effects in the bone marrow o f rats. It was also determined that Cldphen A30 and A60 did not show clastogenic effects in D rosophila m elanogaster (Nilsson and Ramel 1974). In ring dove embryos, chromosomal aberrations were observed when tested with Aroclor 1254 (Peakall et al. 1972). Carcinogenicity - Laboratory Animals: In a chronic feeding studies done with Donryu rats, Kimura and Baba (1973) demonstrated that adenomatous nodules formed in the liver o f female rats but not male rats. However, it was evident that the dose levels used were toxic. Kimbrough et al. (1975) used Arochlor 1260 (at 100 ppm for 21 months) in Sherman strain female rats and G E N P 011489 784300 found that 26/184 (14%) developed hepatocellular carcinomas; 144/184 (78%) developed neoplastic nodules and 182/184 showed foci or areas of cytoplasmic alteration. This is compared to 1/173, o/173 and 28/173 controls respectively for each type o f lesion. In Sprague-Dawley rats (Norback and Weltmen 1984) Aroclor 1260 preparations caused hepatocellular trabecular carcinoma (23%), adenocarcinoma (26%), and neoplastic nodules (8%). Female rats were more susceptible than males to the hepatocarcinogenic effects o f this PCB preparation. The National Cancer Institute investigated the effects of Aroclor 1254 in male and female F344 rats in 1978. The results showed that at dose levels of 0,2,5,50, and 100 ppm in the diet for a period o f 105 weeks, there was a dosedependent increase o f hepatocellular carcinomas and hyperplastic nodules in both sexes of rats. There was also a significant increase in intestinal metaplasia (male & female combined) at the 100 ppm dose level (W ard 1985). This is in contrast to the previously mentioned study where female S.-D . rats were affected by Aroclor 1260 more than male S.-D. rats; in this study the distribution of carcinogenic effects was equal between the F344 rats. A study was done using Clophen A60 and A30 in the diet (100 ppm for up to 832 days, Schaeffer et al. 1984) on male W istar rats. The Clophens both caused neoplastic nodules to form but only Clophen A60 caused hepatocellular carcinomas in 61% of the rats which was significantly different (p < 0 .0 5 ) from controls (2%). Also significantly different was the formation of neoplastic nodules (40%(Clophen A6Q) vs 4 % (Controls)), adenofibrosis (2% vs 23% ), thymoma (0% vs 40%) and W istar nephritis (0 vs 40%). The differences between Clophen A30 and A60 lies in the markedly increased hepatocarcinogenic potency of the higher chlorinated Clophen A60 (60% by weight of Cl). The feeding,of Kanechlor 500, 400, and 300 to male dd mice for 32 weeks at dietary levels o f 500, 250, and 100 ppm showed that at the 500 ppm levels, Kanechlor 500 caused hepatocellular carcinomas (41.7% , Nagasaki et al. 1972). In BALB/CJ mice fed A roclor 1254 (300 ppm) for 11 months, hepatomas increased by 40% where this did not occur in the control animals (Kimbrough and Linder 1974). PCBs can also act as cocarcinogens or anticarcinogens which enhance or inhibit the tumorigenic activity o f other carcinogens. However it should be noted that the more highly chlorinated commercial PCBs (> 5 0 % by weight) were reported as promotors o f hepatocarcinogenesis in rodents and the lower chlorinated PCBs have not been determined. For example, Kanechlor 500 decreased the incidence o f liver tumors in rats treated with the heaptocarcinogens 3 '-methyl-4-dimethyl-aininoaxobenzene, 2acetylaminofluorene, and diethylnitrosamine in the diet. Some anitcarcinogenic activities of Aroclor 1254 and PCB congeners have been looked at in mouse GENP 011490 784301 skin models using PAHs as initiators. Results of one study by DiGiovanni et al. (1979) showed that Aroclor 1254 (100 g/mouse) given 18 hours before the initiator 7,12-dimethylbenz[a]anthracene (DBMA), significantly decreased the incidence of papilloma formation in female Charles River CD-I mice. 3 ,3 \4 ,4 '-tetrachlorobiphenyl was even more active as an anticarcinogen than Aroclor 1254. TCDD and the later mentioned PCB both significantly reduced the number of papillomas/mouse. Human Studies: Analytical studies have shown that most individuals carry significant body burdens o f PCBs in adipose tissues, ranging from 0.1 to 1.0 ppm. The residues found in these individuals come from many environmental pathways and their potential carcinogenic or anticarcinogenic effects have not been determined. The exception to this is the two major groups which have experienced very high levels o f exposure to PCBs through the consumption of contaminated rice oil in Japan and Taiwan and through occupational exposure. The most complete and comprehensive study was done by Brown and Jones (1981) and Brown (1987) on 2588 workers employed at two capacitor plants. The overall mortality and cancer mortalities were lower than expected, 295 and 62 respectively. In examining the cancer mortality data, the authors noted that the only category where there was a significant increase (p > 0 .0 5 ) in deaths was associated with liver, gall bladder, and biliary tract cancers combined. Females were more susceptible to these cancers than males and the length of employment in plant number two was also important because more highly chlorinated PCBs were used there initially. Several conclusions can be drawn from looking at the cases mentioned in the results section: 1) PCBs can covalently adduct DNA in vivo and in vitro with the more highly chlorinated PCBs being poorly metabolized and therefore binding less frequently with DNA. 2) PCB mixtures and individual compounds exhibit very little mutagenic activity in most assay tests. 3) The more highly chlorinated PCB mixtures (> 5 0 % by weight) are hepatocarcinogens in rodents whereas data on lower chlorinated PCBs appear to not be carcinogenic. 4) In some model systems, higher chlorinated PCB mixtures act as promotors of preneoplastic lesions and hepatocellular carcinomas in rodents treated with a variety of initiators. GENP 011491 784302 5) Occupational studies show that exposure to PCBs may cause an excess o f cancers at some sites but the Brown study suggests that there are no significant increases in the overall, cancer rate among workers. Since PCBs are not mutagenic and do not easily form covalent adducts with cellular DNA it is unlikely that the higher chlorinated biphenyls are genotoxic and act as promoters o f carcinogenesis in rodents. Remember that the incidence of hepatocellular carcinomas was much lower in the Aroclor 1254treated animals than Aroclor 1260 which suggests a possible difference in the carcinogenic potencies of the two mixtures related to the differences in their composition. GENP 011492 784303 Carcinogenicity of Polyhalogenated Biphenyls: PCBs and PBBs E.M. Silberixom, H .P. Glauert, and L.W . Robertson (1990) Purpose: To report on the findings of the role that PCBs and PBBs play in the carcinogenic process and how they might influence tumor production by other agents. Conclusions: Polychlorinated biphenyls are capable o f inducing preneoplastic lesions, neoplastic nodules, and hepatocellular carcinomas in rats and m ice. They are suspected o f being human carcinogens as well, but current evidence is inconclusive. PCBs do not appear to be mutagenic or genotoxic in the assays done to date. However, recem ly there has been some indication that they m ight be. PCBs are also active tum or prom oters and are antitum origenic under certain conditions. Remarks: This review will focus on the polychlorinated biphenyls (PCBs) only. Results: Carcinogenicity: After reviewing many studies that involve PCB mixtures such as Aroclor 1254, Kanechlor 500, and Clophen A30 and others for example, evidence indicates that these mixtures induce preneoplastic lesions and hepatocellular carcinoma in animals when given at appropriate doses for long periods o f time. PCB mixtures with a high chlorine content (ie: Aroclor 1254, Clophen A60, and Kanechlor 500) are more likely to induce neoplastic nodules and hepatocellular carcinomas than less chlorinated ones. Some studies have shown that female animals to be more sensitive to the tumorigenic effects o f PCBs than males. For example, in a study done by Norback and W eltman (1985) male and female Sprague-Dawley rats were fed Aroclor 1260 (100 ppm for 16 months and 50 ppm for an additional 8 months) for two years followed by a control diet for five months. The results of the study showed that 96% of the female rats had hepatocellular neoplasms where in the males it was only 15%. Malignant tumors were found in 91.5% of the females and in 4.3% of the male rats. There have been studies done on the modulation o f carcinogenesis in rainbow tro u t Results o f these studies show that PCBs are not carcinogenic in trout or other fish species nor do they contain any promoting activity in the trout model. PCBs may modulate hepatocarcinogenesis in trout but this depends on a number of factors like choice and dose of the initiating carcinogen and the modulation (PCB) and the dosing regimen of the modulator relative to .carcinogen exposure. When Aflatoxin Bt (AFBj) was coadministered with PCBs, o r if PCBs were fed prior to A FB1? it inhibited carcinogenesis in trout (Hendricks 1977; Shelton 1983, 1984) and the amount of inhibition was doserelated to Aroclor 1254. A null effect on tumor incidence has been observed when trout embryos were treated with AFBt then fed Aroclor 1254-containing diets after hatching (Hendricks 1977). GENP 011493 784304 Mutagenicity & Genotoxicitv: A number of PCB mixtures and congeners have been tested for mutagenicity using the Salm onella Ames Test. The majority of the studies done have-found that PCBs are not mutagenic in this bacterial system. Schoeny et al (1979) tested Aroclor 1254 using this test and found it inactive. Others have found similar results in other strains of Salm onella typhimurium . Arochlor 1268, Kanechlor 300 and 500 have also shown negative results for mutagenicity. However, several PCB congeners have been tested for mutagenicity and have been mutagenic to strains TA98 and TA100 either with or without an exogenous activating system (S9). These congeners were 4-chlorobiphenyl, 2 , 2 \4 ,4 ,-tetrachlorobiphenyl, 3 ,3 ',4 ,4 '- tetrachlorobiphenyl, and 2,2',4 ,4 \6 ,6 ,-hexachlorobiphenyl. Several bacterial PCB biodegradation products were also tested and found to be nonmutagenic in the Salm onella assay (Sayler et al. 1982). In some in vivo and in vitro test systems PCBs have produced negative results. Mixtures o f Aroclor 1242 and 1254 did not cause chromosomal abnormalities in bone marrow and spermatogonial cells or induce dominant lethality o f rats (Dikshith et al. 1975; Garthoff et al. 1977; Green et al. 1973, 1975). Aroclor 1254 also did not caused unscheduled DNA synthesis in primary rat hepatocyte cultures (Probst et al. 1981). When V79 Chinese hamster cells were treated with up to' 150 /ig/m l o f A roclor 1242 o r Clophen A60, neither produced mutagenic effects. Despite overwhelming negative results concerning PCBs and mutagenicity, there have been some studies with positive results. The Japanese Ministry of Health and Welfare sponsored a study that found that Kanechlor 300 produced chromosomal abnormalities in mammalian cells treated in vitro and induced repair activity in certain bacterial strains. Kanechlor 500 was found to cause chromosomal aberrations in mouse bone marrow after in vivo PCB administration. Sargent et al. (1989) reported that Aroclor 1254 at 0.011 to 1.1 n g/ml caused chromosome breakage, rearrangements and mitotic delay in human lymphocytes cultured in vitro. There have also been studies done with PCB congeners and binary mixtures but they' will not be discussed in this review. Promotion of Tumors: Hirose and associates (1981) studied the effects of PCBs on hepatic and renal carcinogenesis induced by N-ethyl-N- hydroxyethylnitrosamine (EHEN) in male F344 rats. They found that rats fed 0.05% PCBs in the diet for 32 weeks after the EHEN treatment showed a statistically significant increase (p < 0.001) in the incidence of hepatocellular carcinoma when compared to controls given EHEN alone. However, there were no significant effects on renal carcinogenesis. Anderson et al. (1986) were the first to show that tumor promotion outside the liver occurred in the lung. A single dose o f A roclor 1254 (500 mg/kg) given after initiation by dimethylnitrosamine (DMNA) in young male Swiss mice GENP 011494 784305 caused alm ost tw ice the num ber o f tum ors in the lung than w ere found in the DM NA only controls. M ice given 50 and 250 m g/kg PCBs also had increased numbers o f tum ors but it was not-significantly different from DM NA controls. The prom otion of putative preneoplastic lesions has been studied using a few two-stage liver carcinogenesis m odels. A single dose o f A roclor 1254 (500 m g/kg) prom oted enzym e-altered foci (G G T + ) in rat livers after partial hepatectom y and initiation by DENA (diethylnitrosam ine) (Pereira et al. 1982). Oesterle and Demi (1981) showed that both Clopben A50 and A30 prom oted enzym e-altered foci initiated by DENA in rat livers. D oses o f 0.2 N m /kg body w eight per w eek the C lophen's enhanced the num ber o f foci by 9 to 11fold and 13 to 17-fold respectively. These investigators also found a no-effectdose (2 m g/kg) for the prom otion by Clophen A50 in weanling rats but not adult rats, w hich may suggest evidence for a threshold dose. In another study done by O esterle and Demi (1982) it was discovered that there w ere sexdependent-prom oting effects o f PCBs on altered hepatic foci induced by DENA in the rat liver. A ntitum or Activity: Hayes and associates (1987) transplanted liver nodules generated by DENA in the resistant hepatocyte model into the spleen o f syngeneic rats previously given DENA. Subsequent exposure to A roclor 1254 reduced transplant survival to an average o f 8 versus 21% m controls. Two studies by K erkvliet and K im eldorf (1977) have dem onstrated that PCBs (A roclor 1254) have antitum or activity and inhibit tum or grow th in SpragueDawley rats inoculated with a transplantable tum or (W alker 256 carcinosarcom a). The authors speculate that PCBs may inhibit tum or growth by altering the im m une response o f the host either by increasing cellular im m unity o r by decreasing production o f blocking factor. Cocarcinogenesis and M odulation o f C arcinogenesis: M akiura et al. (1974) found that K anechlor 500 in the diet o f rats inhibited tum origenesis when treated concurrently with a known hepatocarcinogen (3*-m ethyl-4dim ethyiam inoazobenzene). ' W hen A roclor 1254 was given to pregnant Swiss C D -I m ice on day 19 o f gestation and then the new borns w ere given DENA postnatally on day four o r 14, two affects o f pretreatm ent w ere noted. O ne, it reduced lung and liver tum or incidence at som e tim es in m ice given the DENA on day 14 but, tw o, it also increased the num bers o f m ice whose livers w ere com pletely tum orous at 18 m onths w hen D EN A was given on day four. The researchers suggest that certain lung tum ors w ith long latency may be selectively responsive to the protective effect o f A roclor 1254. Intercellular Com m unication: Tum or prom oters have been shown to affect gap junction structure and/or function and to inhibit cell com m unication. Inhibition o f com m unication may disrupt norm al cellular grow th control thus allow ing preneoplastic cells to progress tow ards neoplasia. This w ill not be OENP 011495 784306 explored any further in this paper. Human Epidem iologic Studies: Brown and Jones (1981) looked at workers em ployed by two U .S. electrical plants w here PCBs were used by manufacturers- The PCBs used included A roclors 1254, 1242, and 1210. Study results showed that overall total cancer m ortality was low er than expected (39 observed vs. 43.8 expected) w hile there was excess liver cancer (3 observed vs. 1.07 expected) which was not significant. But Brown and Jones suggest that there is an association between both liver cancer and cirrhosis o f the liver and occupational exposure to PCBs. There was no increase in risk associated w ith a longer length o f em ploym ent Brown later discovered. Bertazzi et al. (1987) reported on cancer m ortality capacitor m anufacturing w orkers (2100) exposed to PCBs (A roclor 1254, Pyralene 1476 (54% C l), and Pyralene 3010 and 3011 (42% C l)) used betw een 1946 and 1982. The study revealed a statistically significant excess num ber o f cancer deaths for both sexes com pared to local population rates. M alignant tum ors accounted for 14 deaths w ith only 7 .6 expected in m ale w orkers w hile in fem ale w orkers there w ere 12 cancer deaths as opposed to 5.3 expected. Bertazzi was reluctant to make conclusions about the study because o f the sm all num ber o f deaths and other study lim itations. Zack and M usch (1979) found a statistically significant increase in deaths from circulatory diseases in w hite m ales em ployed at a U .S. PCB production facility. Gustavsson et al. (1986) exam ined Sw edish capacitor w orkers exposed to PCBs. The num ber o f cancer deaths was slightly elevated (7 observed vs. 5.39 expected) but the overall risk w as w ell w ithin the expected values. The only supporting evidence fo r chem ically induced effects was the discovery o f two relatively rare tum ors in one person who had interm ediate exposure for five years. Gustavsson concluded that his study did not indicate any excess m ortality or cancer incidence at this factor to date. D avidorf and Knupp (1979) looked a t the epidem iology o f ocular m elanom a in the state o f O hio. They w ere unable to correlate the incidence o f ocular m elanoma w ith exposure to PCBs. Chase et al. (1989) have looked at cohort m ortality studies in term s o f criteria for the association betw een PCB exposure and the developm ent o f cancer and he concludes that "There is insufficient evidence to show a causal relationship between PCB exposure and the subsequent developm ent o f any form o f cancer." GENP 011496 784307 Estim ation o f Carcinogenic RiskJo_H um ans: The U .S. Environm ental Protection Agency (EPA ), Food and D rug A dm inistration (FD A ), and the N ational Institute for OccupationaTSafety and H ealth (NIOSH) have proposed or adopted standards o r criteria fo r regulating exposure to PCBs. These guidelines have been used for the m ost part to quantify risk assessm ents for carcinogenesis by these com pounds. The federal standards and criteria for PCB exposure do not make a distinction between PCB m ixtures and individual congeners - they are all considered to be equally toxic o r carcinogenic. Based on consum ption o f 2 L o f drinking w ater and 6.5 g o f fish and shellfish per day, w ater concentrations o f PCBs o f 0.79, 0.079, and 0.0079 ng/L were estim ated to produce increm ental lifetim e cancer risks o f 10*5, 10'6, l0-7 respectively (Kim brough et al. 1975). M ore recently, (Federal R egister, Vol. 54, No. 97, May 22, 1989)the EPA has proposed a maximum contam inant level goal (MCLG) and a maximum contam inant level (M CL) for PCBs in drinking w ater. EPA has realized that an M CLG level o f zero is unrealistic so they have proposed a M CL o f 0.0005 m g/L for PCBs. This M CL level corresponds to an excess lifetim e cancer risk o f slightly less than 10*. The Am erican Conference o f Governm ental Industrial H ygienists (ACGIH) in 1980 set a threshold lim it value (TLV) for an 8-hour tim e weighted average (TW A) concentration o f 0.5 and 1.0 m g/M 3 for A rochlor 1254 and 1242 respectively. Short-term exposure lim its (STEL) w ere set at 1.0 mg/M 3 for A roclor 1254 and 1242. NIOSH has also recom m ended that w orkers not be exposed to a PCB concentration above 1.0 /xg/M3. This was a TW A concentration for a 10-hour workday and was the m ost reliable detectable lim it at that tim e, 1977. Demi and O esterle (1987) evaluated the prom oting activity o f PCBs using a rat liver foci bioassay and found the low est effective dose level to be 1 m g/kg body w eight per day (430 /xg/kg/day). They estim ate the risk o f exposure to PCBs to be very low since the LO EL was m ore than 1000-fold greater than the average daily intake for an individual as estim ated by C ordle et al. (1978) to be 5 to 10 pg o r 0.07 to 0.14 pg/kg/day for a 70 kg person. D iscussion: Polychlorinated biphenyls induce preneoplastic lesions, neoplastic nodules, and hepatocellular carcinom as in rats and m ice when given a t appropriate doses for extended periods o f tim e. PCB m ixtures that are high in their chlorine content (i.e ., A roclor 1260, Clophen A 60, K anechlor 500) are m ore potent than the lesser chlorinated PCBs in inducing neoplastic nodules and hepatocarinom as. It has also been noted that fem ale species are som ew hat m ore sensitive to the tum origenic effects o f PCBs than m ales o f the same species. Experim ental evidence appears to show that PCBs are not m utagenic o r genotoxic although there have been some recent studies that suggest otherw ise. T h Ames bacterial m utagenicity assays, as w ell as the genotoxic assays are not optim ized fo r the slow rate o f m etabolism o f individual halogenated biphenyls. 784308 Therefore it is possible that one o r m ore PCB congeners may be m etabolized to a reactive species with highly genotoxic potential. PCB m ixtures and individuals congeners are active tum or prom oters in both rats and mice when given for extended periods o f tim e after an initiating agent. The prom oting effects o f PCBs have been dem onstrated in the liver, as well as in the lung and skin. G enerally PCBs are im m unosuppressive but it has been found that PCBs contain antitum or activity w hen they axe adm inistered either before o r after tum or transplantation. PCBs also can act as cocarcinogens or m odulate caxcinogenesis. The tim ing o f the PCB dosing relative to the adm inistration o f the initiator can enhance, inhibit, o r have no effect on carcinogenesis. Human epidem iological evidence is little and for the m ost part incom plete. H ow ever, some studies, like those relating to Yusho o r Y u-Cheng, suggest that exposure to PCBs may increase the risk o f hepatocellular carcinom a. Because o f the results obtained through anim al studies on carcinogenicity, it may be reasonable to assum e that PCBs have the potential to be a human carcinogen as w ell. GENP 011498 784309 Effects of Exposure to PCBs and Related Compounds on Growth and Activity in Children J.L . Jacobson, S.W . Jacobson, -and H .B. Humphrey (1990) Purpose: th e authors exam ined 236 children from two cohorts that w ere at risk for exposure to PCBs and related com pounds at the age o f four years. They explored grow th patterns and activity levels o f these children based on com posite ratings provided by the m others and independent exam iners. Conclusions: Prenatal exposure to PCBs has effects on the growth in height and w eight as well as activity in children. Both weight and activity are negatively correlated in a dose-dependent fashion to exposure to PCBs. D epressed activity levels in children can also be attributed to postnatal exposure fro m m aternal m ilk when children are breastfe d fo r a t least 12 m onths. Prenatal exposure may cause more problem s than postnatal exposure because o f the sensitivity o f migratory cells and cells undergoing m itosis to toxic insult, incom plete developm ent o f the blood brain barrier and/or the absence o f im portant drug-m etabolizing capacities that are fo u n d posm atally. R em arks: Cord serum PCB levels and m others reported contam inated fish intake (over a six year period) from Lake M ichigan w ere associated with reduced birth w eight and shorter gestation. Cohorts are represented by 236 children chosen from 8482 women who delivered infants in four M ichigan hospitals in 1980-81 and from children who w ere exposed to PCBs through contam inated farm products (m ilk and meat) from farms whose anim als had consumed silage w ith PCBs in it. 75% o f the children assessed as infants w ere assessed at age four and four y ean and three m onths. Tests adm inistered w ere the M cCarthy Scales o f C hildren's. A bilities and reaction tim e tests. M others w ere asked to com plete the Peabody P icture V ocabulary Test-Revised (PPVT-R) and the Buss and Plom in Em otionality A ctivity Sociability (EAS) Tem peram ent Survey for Children. R esu lts: Cohort Com parisons: The farm exposure fam ilies w ere m ore likely to live in rural areas, have m ore children, and less likely to be divorced or separated. O verall socioeconom ic status (SES) was about the sam e for each cohort. There were no cohort differences in other perinatal risk factors including duration o f prenatal care, gravidity, birth w eight, gestational age, and delivery com plications. Farm exposure serum PCB levels w ere about the same as fish consum ers due to exposure from non-fish sources. D ata was pooled because o f these sim ilarities except for cord serum and m aternal m ilk levels w hich w ere not available for the farm cohort. GENP 011499 784310 Cord serum PCB levels ranged from 0 to 12.3 ng/m l (mean = 2 .5 , s.d. = 2.0 ). Four-year serum levels ranged from 0 to 23.3 ng/m l (mean - 2.1, s.d. = 3.3). M others providing m ilk sam ples breast fed for an average o f 29.6 weeks and PCB levels (fat basis) in m ilk ranged from 135.7 to 2600.0 mg/ml (mean = 835.9, s.d. = 388.4). Prenatal exposure was associated w ith low er w eight at age four in a dosedependent fashion. Children that had a cord serum level o f 5 .0 ng/m l or greater tended to weigh 1.8 kg less on the average than low er exposed children. This was significant for girls but not for boys. None o f the m easures o f PCB exposure w ere related to height o r head circum ference at four years of age. The affects o f PCBs on com posite activity levels was negatively correlated to four-year PCB level in a dose-dependent m anner. Thiity-one percent o f the children with levels o f 9 ng/m l or m ore fell in the bottom tenth percentile for activity. In the highest exposed group (9 ng/m l) 15.4% and 14.3% w ere rated "usually quiet and inactive'' on the Child B ehavior Record at the first and second testing sessions as com pared to 5.4% and 7.8% for the low est exposed group on visits one and two. i A ctivity level was also negatively related to m aternal m ilk PCB level. In m others w ith higher than average PCB levels and who breast fed for at least 12 m onths, the effect on the children's activity was the strongest. D iscu ssio n : The data seems to indicate that the effects o f in vitro exposure to PCBs and related compounds extend beyond the fetal and infant stages. There are affects on grow th in height and w eight as w ell as affects on activity in children. The four-year w eight deficit is associated with prenatal but not postnatal exposure, a pattern which held up to five m onths postpartum . A lthough larger quantities o f these com pounds axe transferred in breast feeding, m ost o f the effects o f low -dose exposure in humans have been linked to the prenatal period. This may be due to the sensitivity o f m igratory cells and cells undergoing m itosis to toxic insult, incom plete developm ent o f the blood brain barrier, and/or the absence o f im portant drug-m etabolizing capacities that are found post-natally. The effects on activity due to prenatal exposure to PCBs and related compounds are contradictory in som e cases. In som e laboratory anim al studies, increased levels o f activity has been observed as w ell as in the children involved in the Yu-Cheng episode. H ow ever, decreased activity was reported in those children exposed to PCBs in the Yusho incident and in U .S. newborns. Postnatal exposure on the other hand seem s to be m ore consistent. Reduced activity was found in two laboratory studies in rhesus m onkey juveniles exposed both prenatally and via lactation and in two studies done with rats. In this study this was also observed. These depressed activity levels may GENP 011500 784311 be associated with the duration o f breast feeding; ie: they may be negligible . unless the infant is breast fed for at least one year. Future studies may wish to explore m ore thoroughly the effects o f exposure on behavior, based on a m ore direct behavioral coding o f activity and/or autom ated recording o f the ch ild 's movem ents. This is because the effects seen here was relatively subtle and its clinical significance is uncertain. GENP 011501 784312 P olychlorinated B iphenyls an d th e D eveloping N ervous System : C ross Species C om parisons H .A . T ilson, J .L . Jacobson,^and W J . R ogan (1990) Purpose: The study looked at the effects that polychlorinated biphenyls have on the developing nervous system in various species and then to com pare those effects. Conclusions: D evelopm ental exposure to polychlorinated biphenyls results in neurobehavioral alterations in m onkeys, rats, m ice, hum ans, and other species. Prenatal exposure results in hyperactivity, decreased m otor activity, p oor visual recognition memory, low er birth w eights and shorter gestation periods in many o f the species observed. O verall, the tests o f higher cortical function in animals are affected, while in humans it is the delays in the developm ent o f m otor skills. The data fro m these studies can be evaluated fo r the developm ent o f NOAELs, LO AELs, and RfDs fo r each o f the species studied (see attached table). R em arks: Com parisons o f effects w ere made between Rhesus m onkeys, quail, rats, m ice, and humans. This study was based on a series o f laboratory studies done by other scientists to m easure behavioral effects in various species. R esu lts: Rhesus m onkeys: Bowman et al. (1978) fed 2,5 ppm A roclor 1248 to nine adult fem ale monkeys for 86-89 weeks (0.084 mg PC B /kg/day). O ut o f eight conceptions there w ere five live births, three o f which w ere tested. Birth weights o f the three-PC B-treated monkeys w ere 21% low er than controls and by the age o f two m onths had developed characteristic PCB toxicity. The anim als w ith the highest peak levels w ere m ore hyperactive at six and 12 m onths o f age. The m onkeys exposed to PCBs w ere also tested for PCBinduced alterations in learning and m em ory and changes in perform ance using the W isconsin G eneral T est A pparatus (W GTA). This test was given beginning at seven m onths and lasting until 24 m onths o f age. T he PCBtreated m onkeys showed deficits on the first tw o discrim ination reversal tasks (spatial & color). M onkeys w ere also slow in learning progressive probability shifts and the object alternation tasks. A peak PCB body burden at age four m onths corresponded significantly w ith increased errors in five o f the nine learning tasks conducted from eight to 24 m onths o f age. In a second study, Bowman took eight surviving breeding m onkeys from the first study (cohort 1) that had been o ff the PCB diet for 22 to 84 weeks before the conception o f a second set o f infants. Five infants (cohort 2) survived for testing and the average PCB content in subcutaneous fat tissue for the m others was 0 .8 0.9 ppm (SD) w as only slightly higher than controls. T he level o f PCBs in two stillborn infants was 2-2.5 /g/g tissue. In four m onths o f nursing, the infants experienced hyperpigm entation around the hairline, and showed increasing a & x oU502 784313 levels o f PCBs in skin biopsies from birth (virtually not detectable) to three months (3.31 g/g tissue). H ow ever, Bowman reported that there was very little PCB detected in subcutaneousfat o f the offspring at eight and 12 months o f age. These offspring showed hyperactivity in 24 daily 90-m inute m otor activity sessions at age 12 m onths. This is sim ilar to the first cohort. In a third cohort (new breeding m others), monkeys w ere fed A roclor 1248 three tim es a w eek at 0.3 and 1.0 ppm respectively for 65 to 102 weeks. W hen m otor activity was tested at 12 m onths and com pared to the activity levels o f control offspring in cohort 2, PCB exposure resulted in significant dosedependent increases in m otor activity. Controls and treated anim als from m others rem oved from the 2.5 ppm PCB diet (cohort 2) and the offspring from m others who received 0.5 ppm during gestation and nursing (cohort 3) w ere used in subsequent testing of schedulecontrolled operant responding. M onkeys w ere tested under a series o f fixedinterval schedules o f food reinforcem ent. There was a slight but significant low ering o f the index o f curvature in the PCB -treated monkeys. Q uail: Japanese quail chicks w ere exposed to A roclor 1254 (200 ppm) in feed beginning at seven days o f age (K reitzer and H einz 1974). Exposure continued for eight days follow ed by six days o f no exposure. A fter 24 hours o f initial dose, chicks w ere tested for avoidance response to a moving silhouette. This response was suppressed in exposed anim als even after untreated food was restored. R ats: Behavioral testing began at 12 weeks o f age after m others w ere dosed at 20 or 100 m g/kg o f K anechlor 500 on days 8-14 o r 15-21 of gestation. Open field trials and m ultiple T-m aze trials (w ith escape from w ater as an endpoint) w ere conducted. Exposure to K anechlor had no significant effect on total num ber o f im plants, num ber o f resorptions, average litter size, or num ber o f externally m alform ed fetuses. N o pups from the 100 m g/kg group survived. M ales exposed to 20 m g/kg oh days 15-21 o f gestation had increased num bers o f errors in the w ater maze. K qja et al. (1978) adm inistered K anechlor 400 (100 m g/kg orally) for six days to rats three and 20 days old. In a five m inute test period there was decreased spontaneous m otor activity in both groups. W hen three day old rats w ere dosed for 12 days at 100 m g/kg o f K anechlor 400 and w ere tested a t three and four w eeks there w as decreased spontaneous m otor activity and im paired perform ance on an inclined screen. Overm ann et al. (1987) exposed fem ale W lstar rats v ia diet to 0.02, 2 .5 , 26, o r 269 ppm o f A roclor 1254 from m ating to w eaning o f their pups. The highest doses o f PCBs decreased the num ber o f litters and pup birth weight; m ost pups delivered to those m others died w ithin a week o f birth. Low er concentrations o f PCBs had no effect on pregnancy success, pup birth w eight, GENP 011503 784314 dam body w eight, and food intake. Body weights w ere reduced and ontogeny o f negative geotaxis, auditory startle and air righting reflexes in pups from m others exposed to 26 ppm w ere altered prio r to w eaning. Pantaloni et al. (1988) subjected fem ale rats to Fenchlor 42 for five days, two weeks prior to m ating, during gestation o r during lactation. Exposure to PCBs did not affect m aternal or pup w eight up to 21 days o f age. Preconception or in utero exposure had no effect on cliff avoidance, w hile in postnatally exposed groups this m easure was suppressed. D evelopm ent o f swimm ing behavior was also affected in all three groups o f exposed rats. Preconception exposure depressed activity in an open field at 14 and 21 days o f age w hile in postnatally exposed groups, rats w ere suppressed at day 14. M ice: Chou et al. (1979) studied the effects o f 3 ,4 ,3 \ 4 '-tetrachlorobipheny 1 (4-CB) on m ice. 32 m g/kg o f 4-CB w ere adm inistered by gavage on days 10 to 16 o f gestation. A bout one-half o f the litters displayed a neurological syndrom e which included increased locom otor activity at day 15. Anim als showed jerking or rotational m ovem ent o f the head and episodes o f constant circling around the hom e cage. A ll the 4-CB treated m ice had heterotopic m otor neurons and astroglia in ventral spinal fibers and cylindrical CNS peninsulas that projected into the ventral cranial and spinal nerve roots. Tilson et al. also studied 4-CB in m ice. C D -I m ice whose m others w ere exposed to 4-CB during gestation displayed at 35 and 65 days o f age a neurobehavioial. syndrom e consisting o f interm ittent stereotypic circling and hyperactivity. In the dark phase o f the diurnal cycle, 4-CB-affected mice w ere m arkedly hyperactive and had im paired grip strength and balance. The m ice w ere also im paired in the acquisition o f a one-w ay, shock-m otivated avoidance response. In another study, Tilson found that at one year o f age, m ice exposed in utero to 4-CB had elevated levels o f m otor activity that w ere associated with decreased levels o f dopam ine and dopam ine receptor binding sites in the caudate nucleus. H um an: Tw o cohorts o f children have been follow ed in M ichigan and N orth C arolina. O f the 313 new borns in M I, 242 o f the m others had consum ed Lake M ichigan fish presum ed to be contam inated w ith PCBs. H igher cord serum PCB levels and m aternal consum ption o f contam inated fish w as thought to be related to sm aller birth size and shorter gestation. This size deficit persisted beyond the new born stage. The adm inistration o f the Brazelton N eonatal Behavioral A ssessm ent Scale in M I to 287 new borns, tested m otor coordination, m uscle tone, orientation, and state changes (arousal & self-quieting). Infants whose m others consum ed large quantities o f contam inated fish exhibited abnorm ally w eaker reflexes and w ere less responsive to the stim ulation o f the Brazelton exam ination, and showed m ore jerk y , unbalanced movem ent and m ore startles. 784315 In N C , among 802 infants follow ed from a previous study, prenatal PCB exposure was associated with poorer perform ance on the Psychom otor Index from the Bayley Scales o f Infant D evelopm ent at six and 12 months o f age. In addition, m ore highly exposed infants were hypotonic and exhibited a greater num ber o f abnorm ally w eak reflexes. Again in M I, a test o f visual recognition m em ory was adm inistered to the infants. Cord serum PCB level and m aternal consum ption o f contam inated fish w ere associated w ith poorer visual recognition memory perform ance at seven m onths am ong 123 children (a subgroup o f those tested at birth). Postnatal exposure through nursing was not related to any o f the scales used in M I or N C. These cohorts are still being studied. Com parisons: Q ualitative: "Developm ental exposure to PCB results in persistent neurobehavioral alterations in monkeys and non-prim ates, sim ilar neurological or behavioral effects are observed across species...[and] can be observed in the absence o f reduced body weights o r gross signs o f PCB intoxirication during developm ent." T he behavioral effects seen in rodents and avians exposed developm entally to PCBs w ere qualitively sim ilar to those seen in Rhesus m onkeys. t The m ost com m on finding in the anim al studies w as that developm ental exposure to PCBs resulted in behavioral hyperactivity. This was observed in m ice, male rats, and monkeys in which exposure was pre- o r perinatal. D evelopm ental exposure also decreased m otor activity in short tim e periods o f testing (five m inutes o r less). H igher cognitive processes or learning was influenced by exposure to PCBs during developm ent in rats, m ice, and Rhesus m onkeys. Learning effects w ere observed using m ore com plex discrim ination tasks and schedule-controlled perform ance in monkeys as well as unconditioned and conditioned responses in non-prim ates. In hum ans, low er birthw eight in children w ith greater exposures was seen in M I but not in N C. The only other toxicity seen in either group was neurodevelopm ental. Prim arily m otor effects w ere detected in both studies w ithin the Brazelton A ssessm ent. The other tests, the Bayley Psychom otor Index, was im paired by prenatal exposure in NC at six and 12 m onths and in M I it appeared to affect a subset o f fine m otor tasks w ithin the Bayley. There is consistent evidence w hich points to prenatal exposure to PCBs at levels encountered in the U .S . w hich produce noticeable effects on m otor m aturation and some evidence o f im paired infant learning. The authors o f this paper conclude that these effects in hum ans are not due to D D E o r social and dem ographic characteristics o f the fam ilies. O verall, the tests o f higher cortical function in anim als are affected, w hile in children it is the delays in the developm ent o f m otor skills. The evidence o f poor perform ance on visual recognition m em ory, a w ell-validated assessm ent 784316 o f infant cortical function is consistent w ith anim al data. Tests on m otor function in anim als are relatively unaffected. Q uantitative: The available data w ere subjected to the risk assessm ent model proposed by the USEPA for developm ental toxicity (1989). A NOAEL or "no-observed-adverse-effect level and LOAEL or a "low est-observed-adverseeffect level w ere determ ined from the data and from these reference doses o r RfDs w ere developed for each species o f anim al (see copy o f attached table). In looking at the table, we can see that monkeys respond to the low est doses and for rhesus monkeys, fetotoxicity, chloracne, and decreased postnatal body w eights have reference doses that are an order o f m agnitude higher than for those for m otor activity and cognitive tests. D ose inform ation is not as readily available for the women who participated in these studies so a few references m ust be made about the estim ated am ount o f PCB present in the fat o f breast m ilk at term . If they assum e that a woman weighs 60 kg, and the percentage o f that w eight which is fat, 25% , then a body burden for PCBs can be determ ined. If the theoretical level o f PCBs in breast m ilk was 3.4 ppm , then the body burden o f PCBs w ould be 5 l mg o f PCBs which the m other got from a daily dose o f 5 .6 /g/day, o r 0.093 /g/kg/day. The safety factor would be 10 for susceptibility variation am ong exposed hum ans which wouid yield a reference dose o f 9.3 x 10 ig/kg/day. F or the specific and lim ited purpose o f setting a reference dose, the rhesus m onkey data yield an estim ate o f die reference dose one order o f m agnitude higher than the estim ate calculated from the hum an data (Table 4); and the rodent data gives an RfD that is three orders o f m agnitude higher. The com parisons w ere made using that effect which results in the low est RfD among the endpoints considered. GENP 011506 784317 AND THE DEVELOPING CNS TABLE 4 REFERENCE DOSE COMPARISONS Dose - (mg/kc^day) Type of Effect LOAEL Range NOAEL Range Reference Dose* Rader.:,: Rhesus M.\".keys: NS Humar.s: * Fctotoxicity Postnatal Body Weights Reproduction Motor Activity Leaming/Memory retotoxicity and Chloracne Postnatal Body Weights Motor Activity Leanting/Memory Perfonnance Hypotonicity on the Brazclion Scale Psychomotor Scale Decrement on t2*Month Bavley Impaired Visual Recognition Memory' Carcinogenicity, Risk= l x lO" 4 iE?A quantitative estimate, lifetime exposure) l(l). p. 95] Current EPA Reference Dose (based on low- binhweicht in Rhesus monkeys) [(l), p. 94) 32-269 -- 32-269 l--32 2-32 0.0S4 0.0S4 0.014 0.014 0.014 rr1i1 O 20-26 2-10 4-26 1-5 0.014 0.014 -- -- 9.3 x I0" s 9.3 x lO*4 2.7 x I0_i 3.2x 10" 1 2.0x10" * 3.2X10" 1 2.0 x 10" ' l.Ox 1 0 '1 1.4X1Q" ' 1.4X 1 0 " 1.4X 1 0 '3 l .4 X 10" 3 1.4 X IO" 3 9.3x 10" 4 9.3 x lO" 4 2.7x 10" 4 1.3 x I0 " T l.Ox 10" ' re : animal experiments, intra- and micrspecics variability were considered. For human data, only intraspeeics variability was . consicerrd. O BN P11S07 784318 Background Levels In Humans. In Halogenated Biphenyls, Terphenyls, - Naphthalenes, Dibenzodioxins and Related Products, R.D. Kimbrough and A.A. Jensen, 1989 Purpose: To look at background levels of polychlorinated biphenyls (PCBs), polychlorinated terphenyls (PCTs), polychlorinated quaterphenyls (PCQ s), polybrom inated biphenyls (PBBs), polychlorinated dibenzo-p-dioxins (PCD D s), and polychlorinated dibenzofurans (PCDFs) in hum an blood, m ilk, adipose tissue, and other-tissues. Conclusions: There are m any factors that can affect the levels o f PCBs present in the human body such as where they live, what they eat, how old they are, and what their past exposure has been. Levels o f PCBs range fro m 0 ,5 to 4 ppm in m ilk, 0.5 to 10 ppm in adipose tissue, and 2 to 5 ppb in blood serum . H owever PCBs . are not lim ited to ju st these m edium s; they can also be fo u n d in the liver and the lungs. Levels o f PCBs tend to be higher in the urban and industrialized areas o f contries like the United States, Norway, and Sweden as w ell as higher in men than women. R em arks: Although this paper focuses on all o f the above m entioned chem icals, this review w ill focus only on polychlorinated biphenyls (PCBs). R esu lts: M ilk: There is a wide distribution o f PCBs in m ilk ranging from 0.5 to 4 ppm (m g/kg) in the industrialized nations. This average background concentration varies with donors, sam pling tim e and the analytical m ethods used. M ore than 60% o f the total PCB content in hum an m ilk is made up o f the follow ing seven congeners: 2 ,4 ,4 '-trichlorobiphenyl, 2 ,4 ,4 ',5-tetrachlorobiphenyl, 2 ,3 \4 ,4 ',5pentachlorobiphenyl, 2 ,2 ',3,4,4*,5-, 2 ,2 \4 ,4 \5 ,5 '-hexachlorobiphenyl, 2 ,2 ',3 ,3 ',4 ,4 ',5 '- and 2 ,2 * ,3 ,4 ,4 ',5 ,5 '-heptachlorobiphenyl (Y akushiji et al. 1978; Schulte and M alisch 1984; Safe et al. 1985). It should be noted that average background levels o f PCBs in hum an m ilk do not vary much between countries when sim ilar quantitation procedures are used. Levels o f PCBs can vary depending on the lactation period, m others' place o f residence (PCB levels tend to be higher in urban and industrialized areas) and the degree o f air pollution both indoors and outdoors. PCB levels tend to decrease w ith m aternal age, parity, and during the lactation period. D iet may also play a role in determ ining PCB levels in m ilk. F o r exam ple, dairy and m eat products m at be contam inated through the food chain or from feedstuffs. In some parts o f the w orld (M ichigan, Sw eden, and Japan) PCBcontam inated fish are a key source o f PCBs in m ilk. B lo o d : In the blood, PCBs are bound to lipoproteins and are m ainly in the plasm a/serum fraction. The average background level in blood plasm a or serum is estim ated to be less than 5 ppb (W asserm an et al. 1979), and it may G E N P 011508 784319 be close to 2 ppb. T he concentration o f blood serum PCB levels tends to be higher in the U nited States but variations in analytical techniques and sampling methods may affect this. ----- The most abundant PCB congeners in blood from U pstate New Y ork were 2 ,2 ',3 ,4 ,4 * ,5 '-hexachlorobipheny1 > 2 ,2 ',4 ,4 ',5 ,5 '-hexachlorobiphneyl > 2 ,2 , ,3,3*,5,6,6*-heptachlorobiphenyl > 2 ,2 ',3 ,3 ',4 ,4 '-hexachlorobipheny1 (Bush et al. 1984). Sim ilar levels w ere found in Y usho/Yucheng patients. D uring pregnancy, PCB levels in m aternal blood w ere raised to approxim ately two tim es the levels before pregnancy, and decreased back to low er levels five m onths after delivery (Kodam a and O ta 1980). PCB levels in cord blood are much low er than m aternal blood at delivery (M asuda et al. 1978; Kodama and O ta 1980; Schw artz et al. 1983; Bush et al. 1984). Some studies have shown a strong positive correlation betw een PCB levels in blood and age (Finklea et al. 1972; Kreiss et al. 1981, 1982; Schwartz et al. 1983; Stark et al. 1986; Hum phrey 1983). PCB blood levels have also been positively related to alcohol intake and serum cholesterol (K reiss et al. 1981; Stark et al. 1986) but negatively linked with obesity and enzym e-inducing m edication (Kreiss et al. 1981). A dipose Tissue: M ost data concerning PCB levels in adipose tissue is obtained from autopsy sam ples not biopsy sam ples and are usually from people much older than those who provide blood and m ilk sam ples. Background levels o f PCB in adipose tissue are reported to be 0.5 to 10 ppm (W asserman et al. 1979), which is w ithin the sam e range as m ilk levels.' D ata from various countries are not com parable prim arily because o f the different analytical m ethods used. D onor m aterials also differ especially in m ales w here levels often are m uch higher than fem ales (Solly and Shanks 1974; Kraul and K ariog 1976; G rant et al. 1976; M es et al. 1977, 1982; M ori et al. 1983; W olff et al. 1982; W illiam s et al. 1984; A nasari et al. 1986), and increase w ith age. In regard to background levels o f PCBs in adipose tissue, the order o f the different congeners is 2 ,2 \ 4 ,4 \ 5 ,5 '-hexachlorobiphenyl > 2 ,2 ',3 ,4 ,4 ',5*hexachlorobiphenyl, 2 ,2 \3 \ 4 ,4 \ 5 ,5 ,-heptachiorobiphenyl > 2 \3 ,4 ,4 ',5 'pentachloiobiphenyl (Jensen and Sundstrom 1974). In Italy it is these four congeners plus 2 ,2 ',3 ,3 \4 ,4 , ,5-hexachlorobiphenyl that constitute 60% o f the total PCB content in adipose tissue. O ther Human_Tissues: PCBs have been found in adult hum an liver and lungs from the general population in Japan. A verage levels w ere betw een 10 and 100 ppb (W antanabe et al. 1980); M asuda and Y oshim ura 1984; M asuda et al. 1985). In Norw ay, Finland, and D enm ark PCBs w ere also found in liver tissue w ith levels ranging from 1.87 ppm , 2.50 ppm , and 3.2 ppm lipid o E HP 011509 784320 basis respectively (Bjorseth et al. 1977; H attula et al. 1976; K raul and Karlog 1976). PCB levels in the brain w ere low er and averaged 0.76 ppm and 1.19 ppm in D enm ark and Finland (K raul and K arlog 1976; H attula et al. 1976). D iscu ssio n ; It is known that PCBs accum ulate in human tissues such as the liver as well as in blood and m ilk. This occurs m ost noticeably in the industrialized nations, for exam ple, N orw ay, Sweden, France, and the U nited States. It appears that the background levels o f PCBs in industrialized nations are about the same m agnitude and at present tim e no significant downward trend seems to be occurring. 784321 Genetic Toxicity. In H alogenated Biphenyls, Terphertyls, Naphthalenes, D ibenzodiaxins, and Related Products, eds. R .D . Kimbrough and A.A. Jensen, 1989 Purpose: To look at the genetic toxicity o f polychlorinated biphenyls, polybrom inated biphenyls, polychlorinated dibenzodioxins, polychlorinated naphthalenes, and polychlorinated dibenzofurans. Conclusions: Overall, the authors fe e l that polychlorinated biphenyls are not genotoxic. Test perform ed on Salm onella. E. coli. cultured m am m alian cells, rats, m ice, and other species show that exposure to Aroclor 1242, and 1254, Clophen A 30, 40, and 50, and Kanechlor 500 did not result in significant genotoxic endpoints. R em arks: This paper looked at effects in bacteria and yeast, cultured m am m alian cells, m am m als, hum ans, and other categories o f organism s. A lthough this paper looks at all the chem icals m entioned in the purpose section, this review w ill focus only on polychlorinated biphenyls. R esu lts: Bacteria & Y east: The m ajority of studies done have been with A roclor or K anechlor. Studies done with bacteria have shown with near unam inity in the literature that PCBs are nonm utagenic in Salm onella typhim urium (A roclor 1254 being the test agent). A roclor 1254 was also nonm utagenic in E. coli. W yndham et al. (1976) reported that A roclor 1254 was w eakly m utagenic in Salm onella strain TA1538 in the presence on rabbit liver S-9; the negative results mentioned above used rat, m ouse and ham ster liver S-9s. W yndham et al. (1976) also states that m utagenic activity increased as the degree of chlorination increased. A roclor 1254 did not induce error-prone DNA repair in E. coli (Q uillardet et al. 1985). Also in E, coli, K anechlor 500 was positive in a test that measured induction o f the umu operon in Salm onella (O da et al. 1985). Cultured M am m alian Cells: Tests in cultured m am m alian cells showed neither gene m utations nor chrom osom e dam age in cultured rodent cells treated with A roclor 1254, 1242 o r Clophen A40 (A m acher et al. 1979; H attula 1985). A roclor 1254 did not transform prim ary Syrian ham ster em bryo cells or induce chrom atid breaks o r m itotic inhibition when incubated w ith dividing human lym phocytes in culture (H oopingam er et al. 1972). U nscheduled DNA synthesis in prim ary rat hepatocytes and single-strand breaks w ere produced when cells w ere treated w ith A roclor 1254. This also happened in a m am m alian cell line when treated w ith 2 ,2 ',5,5'-tetrachlorobiphenyl and its epoxide and derivatives (Stadnicki et al. 1979). M amm als: In m am m als, chrom osom e aberrations w ere not induced in the bone m arrow o f rats fed A roclor 1254 for five weeks (G arthoff et al. 1977) or GENP 011511 784322 when given A roclor 1242 or 1254 by gavage in single or m ultiple doses. In m ice, there w as no induction o f m icronuclei in bone m arrow when injected with A roclor 1254 o r K anechlor 500 in ethanol. H ow ever there was a slight increase when K anechlor 500 in com oil adm inistered by gavage. A rochlor 1242 and 1254 did not produce dom inant lethal effects in rats when given by gavage or by feed, ju st A roclor 1254. H ow ever, some dom inant lethal effects w ere noted in m ale rats weaned on fem ale rats given A roclor 1254 throughout lactation. These two A roclors did not induce aberrations in rat sperm atogonia as w ell (D ikshith et al. 1975; G arthoff et al. 1977; Green et al. 1975b). O ther Species: O ther species tested for genetic toxicity w ere Drosophila larvae or adults fed the PCB m ixtures, Clophen 30 o r 50, silkworm s fed K anechlor 300 o r 500, and three carp species injected w ith A roclor 1254. N either sex chrom osom e nondisjunction or breakage was induced in D rosophila (Nilsson and Ram el 1974), nor w ere m utations induced in silkw orm s (Kawachi et al. 1980). There w ere some dose-related chrom osom e breaks and fragm ents seen in the kidneys o f the carp (A l-Sabti 1985). D iscussion: This grouR o f chem icals is w ith few exceptions not genotoxic. The positive results that do appear have rarely been duplicated in a second laboratory. The m ajority o f the studies have been done w ith PCB m ixtures so it is difficult to ascribe positive results with these biphenyl m ixtures to individual isom ers. GENP 011512 784323 PCB Reduction and Clinical Improvement by Detoxification: An Unexploited Approach? Z, Tretjak, M. Shields, and S.L. Beckman (1990) Purpose: To study a patient who has been occupationally exposed to PCBs and had elevated levels o f these chem icals in serum , adipose tissue, and nipple discharge. Then to look at the outcom e o f a detoxifying method used to rem ove the chem icals from the patient's body. Conclusions: The detoxification program presented here and adm inistered to the patient appears to have favorable results in reducing PCB levels in blood serum , adipose tissue, skin lipids, and nipple discharge. O ther symptoms present before the treatm ent began were also alleviated. This m ethod o f treatm ent works w ell in detoxifying a patient o f PCB residue build-up. R em arks: Patient w orked in a capacitor plant inspecting for the leakage o f PCBs for a period o f nine m onths. She otherw ise did not w ork in direct contact with PCBs but was w orking in an unprotected area w here PCBs w ere being used. The sum total o f PCBs detected by GC-M S was 17 congeners. O ther symptom s observed included abdom inal pain, bloating, general fatigue, m uscle pain, chloracne eruptions, sun sensitivity, hyperpigm entation o f the low er eyelids, jo in t pains and sw elling o f the lim bs, as w ell as m enstrual irregularities. R esu lts: A fter a com plete physical and chem ical exam ination, everything was norm al except for som e abnorm alities noticed in a liver biopsy. A chrom osom e study showed chrom atid and isochrom atid breaks, gaps, and m inutes in 20% o f the 200 analyzed lym phocytes. The patients history o f exposure to chem icals in the w ork place was supported by the elevated levels o f PCBs in the samples taken from the liver and analyzed m icroscopically. The highest am ounts of congeners detected w ere congeners 31/23, 74, 66, and 60. M ajor contributors to the total PCB concentration w ere the congeners nom inally substituted at the 4; 2 ,4 ; 2 ,5 ; o r 2 ,4 ,5 positions. Treatm ent was adm inistered to the patient to m obilize and enhance the rem oval o f stored lipophilic xenobiotics, including PCBs. The treatm ent consisted o f the follow ing: D aily aerobic exercise follow ed by periods o f low -heat (60-80) sauna. N iacin and polyunsaturated oil w ere given to sustain m obilization and elim ination. GENP 011513 784324 Vitamins and minerals were supplemented and daily liquid losses substituted. Body weight was maintained throughout the program Initial PCB levels were high in adipose tissue (102 mg/kg), serum (512 ^g/L), skin lipids (66.3 mg/kg), and in nipple discharge (712 g/L). The detoxification treatment lasted 23 days and resulted in reduced adipose tissue deposits of PCBs and lower blood serum levels of PCBs. Detoxification treatment reduced PCB levels to 37.4 mg/kg in adipose tissue, and 261 tg/L in serum; a 63% and 49% reduction, respectively. Skin lipids were also reduced to a level of 44.3 mg/kg (33.2%) and nipple discharge disappeared. Other symptoms were reduced or disappeared completely. Excretion of intact PCBs in sebum was good before treatment but was helped by up to five-fold with the detoxification treatment. Unchanged was hyperpigmentation of the eyelids and the loss of the sense of touch. A post-treatment chromosome study showed a decrease in the number of isochromatid and chromatid breaks and gaps to 10% in 200 examined lymphocytes. The pre- and post-treatment levels of PCB congeners in tissue samples were considered to be statistically significant (p<0.05). No adverse side-effects to the treatment were noted. At the end of six months after treatment was terminated, the patient was reported to be faring well. Discussion: Clinical problems of the patient were alliviated during treatment. Given the excess body burden of PCBs, the remaining amounts may produce further clinical consequences that require repetitive treatment. Overall, though, the attained reduction in PCB levels in adipose tissue, serum, and the nipple discharge appears to validate the treatments effectiveness. The authors state that prior studies incorporating their method of detoxification is safe and effective, with results acquired within an acceptable time frame. They feel that their results from this case involving very high PCB levels and a history of exposure to other lipophilic xenobiotics support others findings. GENP 011514 784325 Polychlorinated Biphenyl Isomers in the Blood and Biopsy Fat Specimens of a Selected Population of British Columbia (Canada) J. Mes, L. Marchand, and K. Karpinski (1989) Purpose: To report on a num ber o f selected PCB isom ers in blood and biopsy fat of residents from the town o f G olden and surrounding vicinity in British Colum bia. 1 Conclusions: None o f the PCB congeners were relaxed to the incidence o f D ow n's Syndrome. There were low levels o f PCB residues in blood (0.09% ) but rather high levels in adipose tissue (76.4 %). R em arks: This area was chosen by the B ritish Colum bia health authorities because of this area's high incidence o f D ow n's Syndrom e in order to rule out possible effects o f environm ental chem icals such as PCBs. 25 blood and biopsy sam ples w ere collected from donors with and w ithout a D ow n's Syndrom e child and w ere analyzed for 29 PCB isom ers Fat sam ples w ere taken m ost often from the abdom inal region; also one in the buttocks and two from the breast. R esu lts: M ean lipid contents o f the blood and adipose tissue w ere 0.09 and 76.4% respectively. M ean recovery o f selected PCB isom ers in blood ranged from 57-100% w ith a mean o f 79% . M ean PCB isom er recovery in adipose tissue was 83% w ith a range o f 78-86% for all selected isom ers. The coefficient o f variation (CV) for triplicate determ inations o f all detectable PCB isom ers in blood was < 20% and w as also < 20% for adipose tissue except for 2 ,4 ,4 '-tri, 2 ,3 ,2 ',5-, 2,4,2*,5', and 2 ,3 ,4 ,4 '-tetrachlorobiphenyls (CV range o f 25-55% ). The higher CV 's o f these four congeners may be due to the relatively low residue levels encountered and/or small interferences from coeluting isom ers or non-PCB-related com pounds. T he sm all differences in residue levels observed for some PCB isom ers betw een donors w ith and w ithout a D ow n's Syndrom e child w ere not statistically significant. N one o f the m axim um PCB isom er levels in blood and adipose tissue was associated w ith D ow n's Syndrom e related donors. M edian residue levels o f PCBs that w ere below GC detection w ere: 37, 44, 49, 52, 60, 87, 90, 105, 110, 180, 201, and 209. Levels o f 118,138, and 153 w ere about the sam e as reported previously by M es in a study o f blood sam ples o f children from Toronto. M ost o f the PCB isom ers detected in adipose tissue w ere also observed in blood and vice versa. The only exception to this was 3 ,4 ,4 '-trichlorobiphenyl isom er w hich was som etim es observed in blood but never in adipose tissue. GENP 011515 784326 D iscussion: The ratio profiles indicate that higher chlorinated FCB congeners have low er and m ore constant blood/adipose ratios than the low er chlorinated congeners reflecting their m ore stable m etabolic nature and a possible equilibrium between two tissues. It also appears that exposure to PCBs has no relationship with the developm ent o f Down's Syndrom e. In both blood and adipose tissue there appears to be a tendency for mean PCB isom er levels to be higher in males than in females but this was not statistically significant. GENP 011516 784327 Polychlorinated Biphenyl Congener Residues in Human Adipose Tissue Samples From Five Ontario Municipalities D.T. Williams and GLL. LeBel (1990) Purpose: To extend previous investigations on total PCB levels in human adipose tissue by. determ ining and com paring levels o f specific PCB congeners in tissue sam ples from m unicipalities in the G reat Lakes Basin. Conclusions: The 16 PCB congener levels appear to be alm ost equivalent in tissues fro m the various m unicipalities studied with the exception o f congeners 170, 180, 187, 194, and 2 0 L There appeared to be no difference between m ales and fem ales levels o f these congeners. Therefore, PCB exposure m ay be the sam e in these m unicipalities. R em arks: Human tissue sam ples obtained from unem balm ed cadavers during autopsies in five Canadian m unicipalities; C ornw all, London, St. C atharines, W elland, and W indsor. Tissue sam ples taken from the greater om entum area. 16 specific congeners o f polychlorinated biphenyls w ere analyzed by gas chrom atography-m ass spectrom etry (GC-M S). R esu lts: T he sum o f the congeners 118 to 209 would represent approxim ately 60% of the total am ount o f PCBs previously m easured in these m unicipalities. Mean residue levels for m ales w ere not significantly different (p < 0 .0 5 ) from those for fem ales for any o f the PCB congeners. Total (males & fem ales) mean residue value showed no significant differences (p < 0 .0 5 ) between m unicipalities for 11 o f the 16 PCB congeners. H ow ever, two congeners, 170 (180 ng/g for tf, 130 ng/g for 9) and 180 (260 ng/g for cr, 280 ng/g for 9) w ere significantly higher in those tissues obtained ffor* St. Catharines than in tissues from other m unicipalities. C ongeners 187, 194, and 201 w ere also significantly higher (p < 0 .0 5 ) in tissues from St. C atharines com pared to W elland tissues but tissue levels for either o f these m unicipalities w ere not significantly different from tissue levels o f these congeners for the other three m unicipalities. T here is no explanation o f these differences except for the fact that it m ight be due to non-uniform ity in donor backgrounds and lifetim e exposures, variations in individual m etabolism , and the sm all num ber o f sam ples. D iscu ssio n : There is lim ited data available on levels o f PCB congeners in human adipose tissue and the sam e congeners have not alw ays been analyzed/ H ow ever, these results are in general agreem ent with those from studies done in Sw eden, the U nited States, and Italy. PCB exposures for the general population appear to be sim ilar w orldw ide and these PCB tissue levels are not likely to cause adverse health effects. GENP 011517 784328 K idney C ancer in U tility W orkers Exposed to P olychlorinated Biphenyls (PCBs) S .L . S h alat, L .D . T ru e, L .E . Flem ing, P .E . Pace (1990) Purpose: To report on a few cases o f kidney cancer am ong utility w orkers who were responsible for maintaining electrical transm ission equipm ent including power transform ers. Conclusions: A fter examining utility workers fo r evidence o f kidney cancer, the authors fo u n d that there was conclusive evidence show ing that exposure to PCBs can result in carcinogenic growths in the kidney. The authors agree that more research needs to be done in relating PCB exposure to cancer. R em arks: The three subjects o f various ages in this report w orked as linesm en and m aintenance w orkers for a public utility com pany and w ere engaged in servicing and repairing electrical pow er transform ers. These subjects reported being exposed to organic solvents, herbicides, electrom agnetic fields, and PCBs. R esu lts: A ll tum ors w ere unifocal, located w ithin the center portion o f the kidney, and w ith a maximum dim ension o f 6-14 cm . N one o f the tum ors showed capsular or vascular invasion or nodal m etastases. All three cases showed the follow ing characteristics. Case one showed a m ajority o f solid sheets o f ceils having an eosinophilic granular cytoplasm and a m inority com ponent consisting o f cuboidal cells form ing irregular granular cytoplasm . The tum or in case two exhibited solid sheets o f cells w ith predom inantly clear cytoplasm and associated necrosis and had a prom inent vascular component.. The tum or in case three exhibited predom inantly clear cells that in some areas grew in sheets and in others form ed tubular and alveolar structures associated with broad fibrous septae. Serum concentrations for PCBs w ere not elevated in cases one and two but traces o f pesticides, herbicides, and their breakdow n products w ere found. D iscussion: The authors state that there needs to be m ore w ork done on relating kidney cancer with occupational exposure. They suggest the need for an epidem iological investigation o f renal adenocarcinom a and exposures that occur in the electrical utility industry. G B tiP o u s tf 784329 Decrease Over a Six Year Period of Dioxin and Dibenzofuran Tissue Levels in a Single Patient Foliowing Exposure A. Schecter, JJ . Ryan, and P J . Kostyniak (1990) Purpose: T his.paper is a follow -up on a patient that was m onitored over three years (extended over a six year tim e period) beginning tw o years after exposure to a PCB transform er fire. The study m onitored levels o f 2 ,3,4,7,8-P nC D F , 1.2.3.4.7.8- HxCDF, 1,2,3,6,7,8-H xC D F, 1,2,3,6,7,8-H xC D D , and 1 .2 .3 .4 .6 .7 .8 - HpCDF, and PCB blood levels as w ell. Conclusions: The study showed that over a period o f six years, blood and adipose tissue levels o f dioxin, PCDFs, and PCDDs decreased. There was also a decrease in the toxic equivalent levels fo r dioxin. The authors fe e l that data from this study w ill help others determine body burdens o f these compounds over tim e or a lack o f build up over time. R em arks: Surgical biopsy for adipose tissue was perform ed on the exposed w orker - a white male whose age ranged from the late fifties to m id-sixties during the course of the study. R esu lts: There was a decrease over six years in the blood and fat levels o f one PnCD F, two H xCD Fs, one H pCD F, and one HxCDD to a level they now think to be a plateau. The study also showed a decrease in "dioxin toxic equivalent" (TEQ) levels over tim e from 113 ppt to 64 ppt, lipid basis betw een 1983 and 1987-89 w here the TEQ also reached a plateau. O verall, total PC D D /F levels rem ained sim ilar over the years due to the usual, but high, OCDD levels found in this patient and the general population. PCB blood levels w ere m entioned but no values w ere given in the report. D iscu ssio n : The authors feel that data from this paper w ill "perm it determ ination o f increased body burden (or lack o f increase over tim e) and estim ates o f target organ dose, after potential occupational o r environm ental exposure." W ith inform ation like this it may be possible to determ ine how long after exposure these m easurem ents may be useful. The decrease in toxic equivalents appears to have reached a plateau, six years after exposure, by 1987. G eneral population tissue levels o f PC D D /Fs are believed to be derived from food as well as general am bient OCDD levels in the environm ent, and this may account for the slight increase in recently sam pled OCDD levels and a general lack o f decline in total PC D D /F levels. H ow ever, the congeners to which the patient was exposed, did decrease over tim e w hich accounts for the decline in dioxin toxic equivalents from 113 to 65 ppt, 2 years and 8 years after exposure respectively. GEKP 011519 784330 Dermatological Findings in Children Exposed Transplacentally to Heat-Degraded Polychlorinated Biphenyls in Taiwan B.C. Gladen, J.S. Taylor, Y-C. Wu, N.B. Ragan, WJ . Rogan, and C-C. Hsu (1990) Purpose: To study 123 children who w ere transplacentally exposed to polychlorinated biphenyls and dibenzofurans in Taiw an; also their parents and siblings who w ere directly exposed and 1IS control children from the sam e neighborhoods. Conclusions: The m ajor sym ptom s that m anifest them selves in children as a result o f transplacental exposure to PCBs are: growth deficiencies, developm ental delays, dental abnorm alities, and neuroectoderm al dysplasia. Other symptoms occur as well such as fin g e r and toe-nail deform ities, chloracne, hyperpigm entation, and acneiform lesions. The conclusion is that transplacental exposure to polychlorinated biphenyls results in more serious and extreme effects than direct exposure does. This suggests that the route o f exposure plays a key role in what sym ptom s appear a t a later stage. R em arks: Acute direct exposure ceased after the rice-bran cooking oil was recalled in the autum n on* 1979. Any children bom in o r after June 1978 w ere considered to have been at risk o f transplacental exposure. There w ere 132 such children bom to 74 m others ranging in age from a few m onths to ju st under seven years old. R esu lts: A t birth, exposed children had increased rates o f hyperpigm entation, eyelid sw ellings and discharge, deform ed nails, acne, natal teeth and sw ollen gums com pared to controls. The exposed children also had an increased rate o f dystrophic finger-nails and pigm ented o r dystrophic toe-nails than controls. In addition, exposed children had m ore generalized itching, localized skin infections and hair loss. These findings differ from those subjects who w ere exposed directly rather than transplacentally. O f the 57 hyperpigm ented children, 46 (aU.transplacentally exposed) w ere reportedly pigm ented over their entire body. The m ost com m on sites o f pigm entation w ere the forehead (8), trunk (8), and cheeks (7). Som e children had only one o f the above symptom s (75), others had m ultiple sym ptom s (50). Those children bom shortly after direct exposure ceased w ere neither m ore o r less likely to have these problem s than children bom much later (p > 0 .1 0 ). A physical exam ination was perform ed on the transplacentally exposed children and the controls. H yperpigm entation was persistent and occurred w ith a high prevalence in transplacentally exposed children especially in the genitalia (p = 0.0178) and feet (p = 0 .0701). N oticeable facial hair was three tim es more frequent in exposed children. Rates o f gingival hyperpigm entation w ere unaffected by transplacental exposure (p > 0 .1 0 ) but older siblings and exposed m others showed high rates. GENP 011520 784331 Rates o f dystrophic finger-nails, dystrophic toe-nails, and pigm ented toe-nails were all significantly higher in the transplacental group than the control (p < 0.001). The affected digits w ere often the first and fifth ones; 71% o f all the children w ith abnorm al finger-nails and 61 % w ith abnorm al toe-nails w ere affected in this manner. The m ost common deform ities seen in the finger-nails w ere grooves and ridges and other unspecified dystrophy (10), transverse overcurvature (pincer and tile-shaped nails), and longitudinal overcurvature (forw ard-curving and humped) also affecting 10 children. In the toe-nails the noticeable effects w ere koilonychia (spooning, 37), ridging, onychauxis (thickening, 26), onychoschizia, splitting/scaling (20), plication (flattening, 17), and transverse overcurvature. Acneiform lesions and/or scars at the tim e of exam ination w ere tw ice as common in transplacentally exposed children as in controls (p > 0 .1 0 ). They were also seen in alm ost half o f the older siblings. Seven out o f eleven exposed had follicular keratoses. The transplacentally exposed children had a higher frequency o f scaly or keratotic disorders than controls, not significant at p > 0 .1 0 , and their older siblings had an even higher rate. Exposed children also had higher rates o f sweat gland disorders although rates w ere low and not significant (p > 0.10). : In this outbreak the m ajor target organ was the skin. The transplacentally exposed children appear to have acquired neuro-ectoderm al dysplasia with dental abnorm alities, a grow th deficit, developm ental delay, and a behavior disorder. In these children w hose mode o f exposure was different and who were exposed at a critical point o f developm ent, the derm atological findings are somewhat different from those seen in adults. The observation that onehalf o f the older siblings w ith potential direct exposure had acneiform lesions and/or scars, com pared to only 17% o f the transplacentally exposed children. This observation calls into question the concept o f chloracne as the m ost sensitive biological indicator o f significant exposure to this class o f com pounds. The nail changes w ere very significant and rem arkable. The pigm entation o f nails appears to involve both the nail plate and bed. This pigm entation has been reported in PCB-exposed w orkers. In exam ining lifetim e derm atological findings as reported by the parents o f the children in the study, th e authors found that they could easily com pare the transplacentally exposed children to the controls but not to their older siblings and m others because they had m ore tim e to develop those conditions m entioned in the results section as w ell as abscesses o r boils, w arts, blistering in the sun, dark red urine, hair loss, num bness o f hands and feet, and localized thickening o f palm s and soles. T here w ere som e inconsistencies in the reported num bers o f neonatal acne, deform ed nails (both finger and toe) w here the num bers reported in the lifetim e findings w ere low er than the neonatal findings. The reason for this is not known but may be a result o f inaccurate reporting by the 784332 m others. M any reported lifetim e findings in transplacentally exposed children w ere m ore prevalent in older children than younger. The higher prevalence of reported lifetim e findings in o ld e r children may be due to being bom soon after the poisoning (m ore severely affected) as opposed to later o r because older children have had m ore tim e for the finding to occur. There is little doubt that chem ical exposure during gestation produced the abnorm alities and that PCBs w ere the m ajor contam inants o f the cooking oil ingested by these children's m others. The PCBs them selves are m ixtures o f over 200 distinct congeners. The PCBs them selves w ere heat-degraded during use and thus w ere contam inated by the PCD Fs and by polychlorinated quaterphenyls. W orkers and offspring o f w orkers exposed to relatively contam inant-free PCBs are not so severely affected. The authors suspect that m ixtures are toxic but that the PC D Fs play a key role in their toxicity. GENP 011522 784333 y . u Ida UU04 EPA/SCI.ADV.BRD. ' @0 RECEIVED T o x ic o lo g y , 6S (1990) 97--107 Ehevi Scientific Publishers Ireland Ltd. . Q, 3 , HAMILTON! '' -Correlation between acute toxicity of 2,3,7,8- tetrachlorodibenzo-p-dioxin (TCDD) and total body fat content in mammals H arald J. Geyer**, Irene Scheuntert*, Karl Rapp4, A ntonius Kettrup*" Friedheim Karte*, Helm ut Greim* and Karl Rozman6** Institut lr kologische Chemie, *Institut r Bodenoklogie, Institut f r Taxikoiagie, D-8042 ,Vuherberg fF.R.G .), '`Z nurulinsxiiut f r Versuchstierzucht, D-3000 Hannover 91 (F.P.G-), 'Unrversndat Paderborn, Angewandte Chemie, D^-790 Paderborn (F.R.G .), Technische Universitt \tineben* TVeihenstephan, Institut r Chemie, D-8050 Fretnf-fVeihenstephan fF .R .G j, University o f Kcnscs Medical Center, Department o f Pharmacology, Taxicoiogy and Therapevria, Kansas City, KS 66103 fU,S~A.) (Received May 9ch, 1990; accepted August 4th, 1990) Summary Single o n i 3(Way LD,,J of 2,3,7,S-tetxi uo^ dloxis (TCDD) were correlated with total body fat (TBF) eastern is various species mu su . .o f laboratory mammals. LDU values and TBF contests were diher obtained from the literature or determined by^experiments. A log (LD*) vs. tog (TBF) plot yielded a highly significant linear regression equation^ at 0.834, P < 0.001, n w20). It Is suggested that this correlation easts for ax least two reasons: (1) increasing TBF content in organisms represents an capacity to remove TCDD Trots the systemic circulation and (2) different TBF content reflects a differential role and regulation of fat metabolism for various organ isms. Extrapolation of this correlation to man suggests chat adult humans are among the leas native species to the acute toxicity of TCDD. K ey wo*tfxr 2.3,7,8-Tetrachlorcdibm op-dioxin; Acute toxidey; Spedes differences; Total body fat content hurodnetlo 2,3,7 (S-Tetrachlaradlbenzo-p-dioxis (TCDD) is not produced for commercial purposes and has no reported use other than as a research tool. It is farmed as a contaminant in the process of chemical synthesis o f chlorinated phenols, espe cially of 2,4,5-trichiorophenol and in smaller quantities during the combustion of municipal and industrial wastes. TCDD is a very lipophilic and persistent chemi- * T o w h o m c o rre sp o n d e n c e s h o u ld b e a d d re sse d : Harald J, Gcycr and Dr. Karl Rozmaxx, Institut fr Toxikologie d e r CSF Mnchen, Ingoisttdter Landser. 1, D-8042 Neuherbcrj, F.R.G. Q300-U3X/90/W3.30 1990 Elsevier Sdotiffc Publish Ireland Ltd. Primed and Published la Ireland GENP 011523 784334 SPA/SCl.A D V .B R D . cal. It accumulates in fat of aquatic as well as of terrestrial organisms including man (1--3]- TCOO is an extremely toxic chemical to some animal species. However, there are marked differences in the sensitivity of various species to the toxic effects of TCDD. The LDJ0 of TCDD in different species varies by nearly 4 orders of mag nitude. Male guinea pigs with an L 0 Mof about 1 pg/kg and male golden Syrian hamsters with an LDJ0 of about 5 mg/kg represent the two extremes (Table l). Like guinea pigs, American dark minks [4J and rabbits are also very sensitive, whereas adult Beagle dogs and a mouse strain (DBA/2 J) are relatively resistant to acute TCDD mrcxiorion. Interstrain differences have also been repotted for rats [5] and mice Id]. In spite o f extensive investigations in recent years, neither the cause of death nor the marked spedes differences in the acute toxicity of TCDD can be explained. Rozman [7] and Rozman er al. (8,91 suggested that species differences In TCDD toxicity may be due to different amounts of brown adipose tissue (BAT) in various spedes. BAT occurs is relatively high amounts in hibemators such as hamsters, while non-hibemators like adult guinea pigs contain little, if any, identifiable amounts of this tissue. Because the ma|" toxic effect of TCDD in ail animals studied is a loss of body weight, which is largely a result of body fat Loss, we have extended the hypothesis of Rozman et al. [7--9] to investigate if a quantitative relationship exists between the acute toxidty (L D ^ of TCDD and total body, fat content (brown adipose tis sue plus white adipose tissue) o f various mammals. Materials and methods Acute toxicity data o f TCDD The acute oral toxidty data (L D ^ of TCDD in different spedes and strains of. laboratory mammals were taken from the original literature and are compiled with references in Table I. LD^ values are influenced by many factors such as animal spedes, strain, body wright, age, sex, health, diet, food deprivation, observation period, etc. The route o f dosing (orai, intraperitoneal, dermal, sub cutaneous, etc.) has also a great impact on LD^ values of TCDD [10]. Therefore, in Table I, only single oral LDa data are given with an observation period of 30 days after dosing. In most mttaners the vehide was oil, in a few cases oil/acetont (95:5 v/v). When important information regarding body weight, age, etc. of the animal* was not reported in the original publications, such data were obtained directly from the authors. D eterm ination o f body fa t Total body fat (TBF) contest for the various sp ed was either obtained from the literature or determined experimentally (Table I). Animals used for TBF determination were o f the same strain, similar weight and/or age as those for which LDn values but so TBF contents were available in the literature. Rau, mice and hamsters were bred in the Zentralisstitui ffir Versuchsiierzuchc, Han nover, F.R.G. Male Hartley guinea pigs were obtained from Savo, Kisslegg, 98 GENP 011524 784335 ur.v av.i .AU V ,UKU . i(3C F.R.G. All animals were kept under standard laboratory conditions (12 h light/ dark cycle, 55% humidity) and had access to species-specific feed (Altromin standard diet, Lage, F.R.G.) and- water ad libitum. For each determination of TBF content three animals were used. Animals were weighed individually, sacrificed and dried to a constant weight in an oven at 110C (time: 2 days for mice, 3 days for hamsters, 4 days for rats and guinea pigs). The dried carcass was ground in a mortar, mixed with perchloroethylenc and allowed to stand for 24 h at room temperature. TBF was then extracted by the FOSS-LET method according to the operating instructions of the manufacturer Foss Electric (Skasdinavies), A /S, 1 Hugissvej, DK-3400 Hilleroad, Denmark. TBF content on a Wt five weight basis was calculated using a calibration curve (specific gravity as a function of fat content). TBF content of those species or strains obtained from the literature -- except the rhesus monkey -- was determined chemically by extraction with organic sol vents (petrolether, diethylether etc.). In the rhesus monkeys TBF content was esti mated via the tritiated water method (Bowman, R.E., pen. commun.). The FOSS-LET method for TBF determination is in good agreement with the other extraction methods taken from the Literature (Rapp, K ,, pen. commun., Herberg, L., pets, commun., 11). These data and our awn mcafliraacms are compiled with references in Table L Statistics Statistical analyses were carried out using a STATGRAPHICS (version 2.6) computer 'program of STSC, Inc., USA. The relationship between LD,, (30 day) of TCDD in different species and strains and TBF content was analyzed by linear least-squares regression analysis. Results The single oral median lethal doses (30-day LDI0) of TCDD in different species and strains were correlated with their TBF contest. Using a two variable linear regression (log (LD^) vs. log (TBF)] the following regression equation [1] was obtained: logO-Djo) * 3*30 log (TBF) - 3.22 (1) The antilog of equation (1) is LDj, - 6.03 X UT CTBFy*" (2) Correlation coefficient P =* 0.834; level of significance* P < 0.0001; standard error of the estimate (square root of the variance of estmate) of log data Syj =* 0.460; standard error of slope of equation (1) is 0.559; standard error of the intercept is 0.565 and the number of data points used in the calculation f the regression equation n a 20. Data are plotted and equation (1) is presented in Fig. 1 graphically. 99 GENP 011525 784336 TABLE 1 BODY WEIGHTS AND SINGLE ORAL lODAY LD^ OF 2.3.7.8 TETBACHLOROniBENZap DIOXIN (TCDD) IN DIFFERENT MAMMALS AND TOTAL BODY FAT CONTENT (TBF) OF THE SAME SPECIES AND STRAINS WJTII SIMILAR BODY WEIGHT trA /aci. GENP 011526 Spukt or strikt Sei lullill body wi* (g) u d /ar age (wecka) LD,,h (paAg) Reil. Body nil* (1) TBP Uefa. (W body wi) Guinea p |l M (Fkbrlghi-whilc, DunLia-Haaky)i 20 (ca. Iw) 0.7* (0.6--0.i)* 12 191.8 A 3.5 4.1 A 0.8 Own determination Guinea plgs (Hadky) M Guinea plgs (H inkyl Ribbln hirccoloced M M 211.1 l 1.2 173--21g (21 w) 200--210 0 --4 w) e i. 1000 1.71 (1.26--2.4)* 2.0 10 11, McConnell. R E., pet|. conun. 14 11, Schuhs, |C-H.( pen. comrn. Same d ill used u above. Same d in uacd u above, 2560 7.5 Own j dctcrmlniilon Own ikleim lniiion 24 Ribbln M + F e i. 2000--1000 111 16, Gcbflng, P.J. end '2100 KM 24 -4 (NcwZcilud 00 C-IO*. White) Betui. 1., put. conun. CO "vl Rhcnu monkcyj F 2100--2600 10 11, McConnell, E.B., 2140 10.3 Bowman. R.E., Xwac m u h lta (juvenik) pen. codun. pen. coouu. Rau F {Fliehet F 144) 100-110 (CA. 4--1 w) 40 17 113.1 A 2.2 10.8 A 2.3 Own determination Raii M (Fischer F 144) 110--110 47 IB, McConnell, B.E., pen, cumin. 143.1 A 5.1 10.1 A 1.4 Own dctcimlniiioQ Ril M ( S p r a g u e --Dawfe)*) 200 (6--7 vv) 50 10 18?.7 a 5.3 8.9 A 0.14 Own ticterminKflon t9o g to o K> CO CO CO CO Bail, oulbrcd F (CD, Sprague-- Dawky) 168.8 * 4.9 < 6 -1 w) 100 Itali, oulbrcd (Charles ftiver. C B /C D Sprague-- Dawlcy) M 112 t 1 * <10-- 11 w) 297 (240-- 360)- B ali (Frederick, Fischer, f/V IflN ) M 240 X 4 (11-- 12 w) 30) (210-160)* M kc (CJ7 B t/6 I> M 21.1--26.1 (10--12 w) 182 (161--201)* M kc U ( 0 7 BL/fh) 21-- 25 (9 w) 283.7 Mice (B6 D2F,J)> M 22-- 12 (1 0 -1 2 w) 296 (268-324)- Mice (D B A /2 1 ) M 27.6--28.0 (10--12 w) 2170 (2206--2912)* Duns (Beagle) M 7000-- 11 000 1000(100--3000)* H am siers M 49.) (golden Syrian) 44-14 Maocricctus (ca. 1--4 w) o awraou 1117 (829--1113)* 20 166.1 2. 11.4 0.45 Own determination 21 144,S * 1.6 9.4 0.3 Own determination 21 2)0.0 7.1 10.5 * 0.2 Own determination 2).| M 7.9 * 1.1 23 j H 6 6 16, Gcbria, P.J- and Bctso, J., pen. Camoin. 10, kon. J.K., pen. comm. 23.6 2.7 27.7 2.8 27.23 0.4 9220 2190 (7180-13 070) 46.8 * 0.3 9.1 0.2 14.1 * I V 20.0 2.6 13.8 l 4.1 Own deteraunaiioa Own determinaiion Own determination 25 9.8 X 0.8 Own determinatimi GENP 011528 8 TABLE I Hamsicrs, ouibred (golden Syslan) kitsocrkxlus Muratus M 101.0 X 10.3 70--120 (ca.lw ) 3031 22, Echo, J,, pert. comm. (3176--18407jr lUo, K.S., pas. comm. 96.2 X 0.33 17.3 X 0.4 Own determination Hamsters, ouibred (golden Syrian) M esocrkttta autetus M Same data used as above 22. Bciso, 1.. peri, coram. 127.4 6.4 Rao; |C.S., pen. comm. 20.2 4.3 26 Mean * S.E. and/or range. kAioounl requlnd to kill JOB of the animalt within 30 days post-esposure. *Cakuiatcd by the authorc. *The number in paictuhew* indicale the range of lethal dotes (0 and 100% mortality). *93% confidence. TttUlghl-whitc, Dunkln-Hanlcy slialn. (CS7 BL/4J feo k X DBA/2J male) F,. -^1 CO 4^ 03 03 CO o < 0 . 1 - 1 ------------------------------ 1----------------1------------- 1------- i 1 5 10 20 30 FAT C O N T E N T (% O F BO D Y W E IG H T ) Fig. i . b N ationthip betw een acute o ra l l o d d r y (lo f LD ,,) o f T C D D an d to ta l body fat c o m e Go* TBF) in different m am m ilt. D iscu ti * The above data demonstrate a positive linear relationship between che loga* rithm of single oral acute doses of TCDD and the logarithm of TBF content in the spedes and strains examined. The correlation is highly significant (correlation coefficient r = 0.334, significance level P < 0.001) and therefore suggests that TBF content is associated with the acme toxicity of TCDD. Adipose tissues (white and brown adipose tissue) contain large amounts of fat which is the predominant form of stored energy in an organism. Due to this large fat content, adipocytes are important storage sites for lipophilic compounds such as polychlorinated biphenyls, hetaefaiorobenzeas and TCDD. Storage of chemi cals in fat results in their removal from the systemic circulation. This in turn reduces the concentration of such substances at potential target sites of toxicity. Therefore, one might suspect that the- correlation between acute toxicity of TCDD and TBF is a result of increased storage of TCDD. Despite the attractive ness- of such a straight forward explanation, it appears to be an oversimplifi cation. First, the correlation between LDa and TBF exists on a logarithmic scale, but storage of TCDD in fat does not increase logarithmically. Second, the oral LDa for both the Sprague---Dawley rat and the rhesus monkey is 50 g/kg (Table I), buz the main storage site of TCDD for the Sprague--Dawley rat is the 103 GEN P0J!J2S 784340 liver and fat is only second in importance [27], whereas for the rhesus monkey fat represents the major site of TCDD storage [23]. Therefore, the depot function of fax stores alone is probably not sufficient to explain the herein reported strong correlation between acute toxicity of TCDD and TBF content. Recently, disturbances of major pathways of intermediary metabolism (glu cose, fat and protein metabolism) and their endocrine control have been reported In the rat after TCDD exposure [29--37]. It should be noted that intermediary metabolism is differendy regular! and the various substrates play differentia] roles in energy metabolism of various species. For example, the guinea pig is a herbivore with fermentation of cellulose to short chain fatty adds taking place in the cecum. These short chain fatty adds are the major source of energy for this spedes. In addition, guinea pigs and other herbivores have frequent bouts of feeding which reduces their need for large energy stores. Mot surprisingly this species has the lowest TBF (about 5% of body weight in young guinea pigs). Rats and rhesus monkeys are omnivors with carbohydrate bang thtir major form of energy intake. These two spedes have less frequent bouts of feeding (about two daily) and convert a substantial proportion of their energy intake into stored fat. Correspondingly larger fat depots are present in these spedes (about 10% body wt). The hamster is a hibemator with a seasonal cycle superimposed over the daily feeding pattern and a nearly complete switch from glucose to far metabo lism during hibernation. The need for large fat depot in such spedes is self evi dent (about 20% of body weight in the adult fiamstcr). These brief physiological/ teleological considerations of spedes differences indicate the differential contribu tion and role of fat metabolism and fat depots for intermediary metabolism in various spedes. It should be noted that both the acute toxidty of TCDD and TBF seem to be age dependent (Table I). Unfortunately no systematic investigations exist on age dependence of TCDD toxicity or TBF content across the herein examined species. Therefore, it is not possible to make a strong case for predicting the acute toxic ity of TCDD far children based on this correlation. However, the fa n that this strong, correlation holds for a variety of spedes and strains of different ages sug gests that an extrapolation for children may also be warranted. According to Gdgy Scientific Tables [38] the TBF content of 4--36-month-old children is 26.3 --18.3%, whereas that of newborns is about 13.6% o f body weight. The correla tion thus predicts that young children would be in the same range or slightly less sensitive whereas newborns would be about 10 times mare sensitive to an acute dose of TCDD (DLj,,: 614 pig/kg) than the adult `Reference Western Man*. We suggest that the cause of a strong correlation between acute toxicity of TCDD and TBF may be a combination of storage capacity (toxicokinetics) and o f the differential role and regulation of intermediary metabolism among species and strains. Assuming that this correlation is valid for most and perhaps all mammals, intending man, the 30 day oral LDn in the adult `Reference Western Man* of 70 kg body weight and 21% total body fat (ICRP, 39) would be 6230 g/kg. of TCDD. This prediction supports suggestions of Ayres et aL [40], Tschirley [41] and Poiger (Poiger, H., pern, commun.) that adult humans are on the less sensitive side o f the TCDD toxidty spectrum. 104 784341 We realize that the selection of studies for chis evaluation may be subject to criticism. For example, the Inclusion of 3 different guinea pig studies with the same TBF content and one hamster study with 2 different TBF contents may unduly influence the correlation coefficient and the slope of the regression line. However, omission of 2 guinea pig and one hamster studies would not change the conclusions of this paper. In that case linear regression analysis yields the equa tion LDm - 1.53 X i0-) (TBF)*-81 with /* * 0.7 and P < 0.001 predicting an LDMo f 4962 pg TCDD/kg for adult 'Reference Western Man1. In fact Table I allows for interested scientists to select any combination of these studies based on their own exclusion criteria and per form linear regression analysis. In conclusion, there is a good correlation between log (TBF) and log (L D ^ of TCDD across many species and strains. It appears that both toxicokinetic and toxicodynamic factors contribute to the existence of this relationship. Exceptions to this correlation may occur and reveal additional factors important in the acute toxicity of TCDD. Acknowledgements The authors are indebted to Professors R.E. Bowman, G.R. Hervey, L. Herberg, J.R. Olson, R.E. Peterson and to Drs. Linda Birnbaum, K. Albus, P .I. Gehxingb Joanne Betso, E.E. McConnell, K.S. Rao, H. Poiger, D. Bieniek and M. Niissei for helpful discussions, for providing body weights and/or TBF con tent data and to Dr. Kurt Bunzi for statistical analyses. The authors acknowledge Miss Cornelia Budach and Maxine Floyd for typing the manuscript. Reference* 1 K. Geyer, I. Scheune and F. Krte, Sioeonccstratioa potential (BCP) of 2,3,7,3-tetrachlorodibeazD^-dioxia (2,3,7,3-TCDD) ia to rm ia l organiiai indudlsf humani. Chemosphere, 14 (1916) 1495. 2 H. Geyv, L Scheune and F. Kette, Btoconctnrndoa potential of orguic environmental cheaieala in Humana. RtfuL Toricnl. PhanaacoL, 6 (1986) 313. 3 H- Garer, L Scbaunm and F. Kana, Comlation twwaaa tfao bioeoneentradon potential of orfanie eavtrosmcmal dienucala in Humana and ebdr a-ocaaol/wuer partition coeffldesu. Owanaphae. 16 (1917) 239. 4 JJL HocbtMtn. RJ. Aidezidi aad SJ. Burtiaa. Acura toxkityof2J,7,8-tesmdilorodibenzo-pdlcadaxaatok.Aich. Bnrlroo.Comam. ToxfcoL, 17(1911)33. 5 ft. Fohftcrim , ft. Javcaao, 3. W iaainapi. K. Raunlo aad J. Tuomlsso, Hepatie Ah-reeeptar lovitt and th effaa of 2J,7.S-wgacUorodlhanto^Kiltadn (TCDD) oa bepatie micxosoinal monnxyfnaanaaMtiaa ia a TCDD^usccptibl and-rwlmnt n i Jtnia. Toxicol. AppL PharmacolM92 (1981) 131. 6 D.E. Cuproaa and C.M . Schiller, Dma rHatrrt effeca af 2^,7,S-tetixchJorodibeaio-/><iiaxia (TCDD) In C37B1/6J aad DBA/2I tatet. Toancol. Appi. PhannacoL, 73 (1983) 147, 7 K. Roania, Hecadcease iacreuea tba tooddep of 2J.7,S^ctnehloradibeaao^dioxia (TCDD): Is 105 GENP 011531 784342 brawn adiposa tissue the primary target in TCDD-induecd wasting syndrome? Stochern. Bio- . phys. Res. Cooaua., 123 (1984) 998. 3 K. Rosnasu T. Rozmaa, E. Scheufler, T. Pazderaik and H. Greta, Thyroid hormones modulate tha toxicity of 2J,7,3-tctradilorcdibenzO'p-dioxm (TCDO). J. Toxicol. Environ. Health, 16 (1989) 481. 9 K. Rozman, D. Pereira and M.J. latropaulos, Hlstaparhology of Interscapular brown adipose tissue, thyroid and pancreas ia-^3,7,8-{rachiorodibenio*p-diffldn (TCDO)creaud m s. Toxicol. Appi. Pharmacol., 82 (1986) 591. 10 J.R. Olson, M.A. Hlscher and R.A. Neai, Toxidry of 2,3,7,S-tetrachlorodibenzo-p-djoxin in the Golden Syrian hamster. Toxicol. Appl. Pharmacol., 59 (1980) 67. 11 G. Domansid, K. AitzeissOUer and E. Jan, Schnette Fettbesdmmung in Hering, Diskussion ver schiedener Methoden: Ergebnisse mit einer Fou-Let-Apparvur. Z. Lebessm. Forsch.-Unttr- such., 154(1974) 129. 12 H. Poigcr, K. Weber and C. Schlager, Spedal aspects of metabolism and kinetics of TCDD in dogs and rats. Assessment of toxicity of TCDD-tnetabotttc(s) in guinea pigs, in O. Hutzinger, R.W. Frei, S. Meriia and F. Pocchiari (Eds.), Chlorinaird Dloxtns and Related Compounds. Impact on the Environment, Pergamon Preas, Hew York, 1982, pp. 317--325. 13 S.E. McConnell, G.W. Luder, .C. Rumbaugh. P.W. Aibro, D.J. Harvan, J.R. Hass and M.W. Harris, Dioxin in soil: BloavxUabUlty a /tc ingesdon by rau and guinea pigs. Scenes. 223 (1984) 1077. 14 E.E. McConnell, J.A. Moore, JJL. Haseman and M.W. Harris, The comparative toxidry of chlorinated dibeuop-diorina (FCDDs) in mice and guinea pigs. Toxical. Appi. Pharmacol., 44 (1978) 335. 15 K.H. Schulz, Zor Klinik und tiologie der Chlomkne. Arbeitsmed. Sorialmed. Arbcitsfayg., 3 (1968) 25. 16 B.A. Schweiz, J.M. Norris, G.X. Spirschu, V.fC. Raws, P.J. Gehring, J.L. Enanmon and C.G. Gering, Toxicology of chlorinaird dibenxo-p-dicudns. Environ. Health Perspect,, 5 (1973) 87. 1? P.W. Aibro, J.T. Corbett, M. Harris and L.D. Lawson, Effects of 2^,7,8-tetrachlorodibenz**p dioxin on lipid profiles in tissue of the Fischer rat. Chem.*31oL interact., 23 (1978) 315. 18 EJE. McConnell, J.A. Moore and D.W. Dalgard, Toxidry of 2J,7,8-tcQachlorodibenzo*p- 1 dioxin in rhesus monkeys (Affleam muiasta) following s tingle oral dose. Toxical. Appi. Pharma- coL, 43 (1978) 175. 19 J.R. Allen, J.P. van Miller and D.H. Norback, Tissue distribution, excretion and biological effects of '*C cetrxchlorodlbcnzx>p^iioxin in rats. Food Cosmet. Toxicol., 13 (1975) 301. 20 H.W. Harris, J.A. Moore, J.G. Vos and B.N. Gupta, General biological effects of TCDD in laboratory animals. Environ. Health Perspco., 5 (1973) 101. 21 R. Walden and C.M. Schiller, Comparative toxidry of 2,3,7,8-tetnchlorodibenzo-p-dloxia (TCDD) in four (sub) strains of adult male rats. Toxicol. Appi. Pharmacol., 77 (1985) 490. 22 J.W. Henefc, M.A. New, RJ . Kodbm and JLS. Rao, 2^,7,8-Tetrachlorodibenzo-p-dioxin: Acute oral toxidry in hamsters. Toxicol. Appi. Pharmacol., 59 (1981) 405. 23 M. Ahotspn and E. MintyII, Adipose tissue content as a modifier of the tissue distribution, biological effects, and exaction af a hetachioroWpbtnyi In C57BL/6J and DBA/JBOM f mice. MoL Pharmacol, 24 (1983) 464, 24 CM. Sprty and E.M. Wlddowson, Tha effea of growtit and development on the composition of mammsla. Br. J. Nutr., 4 (1950) 332. 25 H.-P. Sheng and R A Hoggins, Growth of tha beagle: la fh" " iaii compoddon. Growth. 35 (1971) 369. 26 A.M.'Kodsma, la vivo and in vitro determinations of body far and body water in the hamster. J. Appi. PhytioL, 31 (1971) 211. 27 J.Q. Rosa, J.C. Ramsey, TJf. Wenato-, 1LA. Hummd and P.J. Gehrtsg, Ths fate of 2,3,7,8- cetracfaknodibenxo-pNiioxin following single and repeated ornl dosa to tbs rat. Toxicol. Appi. Pharmacol., 36 (1976) 209. 28 J.P. Van Miller, R.J. Marlar and J.1L Allen, Tlsiua diBribution and excretion of tridated 106 GENP 011532 784343 2,3.7,8-ietracftlarodibenzo-p-diaxin in non-hum an primates and rats. Food Cosmet. Toxicol.. U (1976) 31. 29 J.R . G o n k i, M J . tatro p o u las, D. Pereira, fL A rced, G . M u d . L .W .D . W e b and K. R a m a n , Soma endocrine asd morphologic aspects o f the acute toxicity o f 2,3,7,8-cetrachIorodibenio*^ dioxin (TCDD) ia the ra t. Toxicol-. P u h a L , 16 (1938) 313. 30 J.R . G o n k i, G . M u d , L.W . W e b , 0 . Pereira, M .J. latropoulos a sd K. R o m a n . Elevated plasm a corticosterone levels and histopuhology o f the adr enals and thymuses in 2,3.7,8 -tetn chloradib*nzo-p-diojan (T C D D H reatod rats. Toxicology, 33 (1988) 19. 31 J.R . G o n k i, L .W .D . W e b an d K. R o m a n , T liiu e -jp e d /Ic a lteratio n s o f de novo farcy a d d synthesis in 2,3,7,& -tetrachl0rodibau0fM lexitt (T C D D H reased rats. A rch. Toxicol., 62 (1988) 146. 32 J.R . G onki, L.W .D . W eb and K. R o m a n , Reduced glusoneogcnesis in 2,3,7,8-tarachloro- dlbaixo-p-diaxia (TCDDVtrcatad ra u . A rch. T oxicol., 64 (1990) 66. 33 G . M u d , J . Go raid an d K . R a m a n , C om petition o f diet m odifies toxidcy o f 2 4 ,7 ,8 * e tra c h lo - rodibem o-p-dioxia in cold-adapted ra ts. A rch. T oxicol., 61 (1987) 34, 34 G. M usi, J.R . G o n k i a n d K. R o m a n , M ode o f m etabolism is altered in 2^3,7,8-tearadiion> dibenso-p-diaxifl (TCDDV treuEd ra n . Toxicol. L ea., 47 (1989) 77. 33 K. R o m a n an d H . G ra m . Toxidcy o f 2,3,7,3-tarachlorodlbenze-p-dioxin in cold-adapted r a u . Arch. Toxicol., 39 (1986) 211. 36 K. R o m a n . A critical view o f the mcchanismfs) o f toxidcy o f 2,3,7,3-tetrachIorodibeazo-p- dioxin (TCDD): im plications fo r hum an safety assessment. Occup. Environ. Dei mar. (Dermatosen in B o u f und U m w dt), 37 (1989) 81. 37 L .W . W e b , H . G ra m and K. R o m a n , M etabolism and distribution o f ["C lglucose in r a u seposed to 2j3,7,8*tetracfalorodibe&xo-p-didxin. J . Toxicol. E nviron. H ealth. 22 (1987) 195. 38 Geigy S d m a fic Tables. U nits o f M easurem ent, Body Fluids, C om position o f Body, N u a itio a, C. L eutner (Ed.), G b<G agy L td., Basle, Switzerland, 1982, p . 217. 39 IC R P, International C om m ission o n Radiological Protection. R eport o f the Task G roup bn Ref erence Man, Report N o. 23. P erg aao n Press, Oxford, New York, Toronto, Sydney, Braun schweig, 1973, p. 201. 40 S. M . Ayres, K. B. W ebb, R. G . Evans and J. Mikes, Is 2 J,7 ,3 -ceo aeh lo ro d ib i ^H iIo x iii (Dioxin) a carcinogen fo r hum ans? Environ. Health P enpect., 62 (1983) 329. 41 F .H . Tschirtey, D ioxin. S d . A m ., 234 (1986) 21. GENP 011533 107 784344 July 5, 1990 ENVIRON Prof. Dr. med. Helmut Greim Gesellschaft fr Strahlen- und Umweltforschung mbH Institute fr Toxicologie Ingolstdter Landstrae 1 D-8042 Neuherberg West Germany Dear Dr. Greim: About two and a half years ago you graciously responded to a letter I wrote to you asking two questions about a publication in Toxicology and A pplied Pharm acology (V ol.75, p. 278288, 1984) written by you, and Drs Scheaffer and Goessner. PCBs continue to be of interest to us and our clients. The reason I am writing to you now is related to some w ork that has been conducted recently in this country in which slides from a 1978 cancer study o f 2 ,3 ,7 ,8 tetrachlorodibenzo-p-dioxin (TCDD) in rats have been re-evaluated in light of more recent histopathological criteria for diagnosis of proliferative liver lesions in rats. This w ork has been conducted by Dr. Robert Squire of Johns Hopkins University and some of his colleagues. D r. Squire found that many o f the lesions that had originally been classified as "neoplastic nodule," or "hepatocellular adenoma" would, under current criteria1 be diagnosed as foci rather than as neoplasms. Because of the similarities between PCBs and TCDD, one of our clients has expressed an interest in what would be the result of re-evaluating the existing cancer studies o f PCBs using the new diagnostic criteria used by Squire. Dr. Gene McConnell, formerly of the National Toxicology Program, who is working with us on this project, has contacted the authors of two other PCB studies2 in rats to enlist their aid in performing such a re-analysis. I am writing to you to ask if you would be willing to permit Dr. McConnell to come to your lab 784345 1 Squire, R .A ., Evaluation and Grading of Rat Liver Foci in Carcinogenicity Tests. Presented at National Toxicology Program Symposium "The Significance of Foci o f Cellular Alteration in the Rat Liver," May 1989. 2 Kimbrough, R .D . et al. 1975. Induction o f liver tumors in Sherm an.strain female rats by polychlorinated biphenyl Aroclor 1260. J. Natl. Cancer Inst. 55:1453-1459; Norback, D .H . and Weltman, R.H. 1985. Polychlorinated biphenyl induction of hepatocellular carcinoma in the Sprague-Dawley rat. Environ. Health Perspect. 60:97-105. 6NV1RONC o r n o r a f l n n u . . i . k -----> - -- - 011534 T and review the slides from your study to determine whether the tumor incidences would change under Dr, Squire's diagnostic criteria. Sincerely, Duncan Turnbull, D.Phil. Senior Science Advisor GENP011535 784346 SCIENCE and the CITIZEN "A Great Poison" develop liver cancer after doses of TCDD, and male ratsdo not? Dioxin helps elucidate Researchers now know that dioxin the function o f genes works by reversiblybinding toanintra cellularreceptor, which also binds with similar compounds suchas polychlori Oneman'spoisoncanbe another man's passion. Even dioxin-- the infamous contaminant in Agent Orange-- isloved by some. This Boss Tweed as hacker, honeybee dancing, natedbiphenyls (PCBs).The receptoris soluble, that is,itis not bound to the cellmembrane. "All[ofTCDD's]toxid ty ismediated through this receptor," small and relativelysilentband of fol liquid-crystal cosmos, saysAlan P.Poland, a professorofon lowershas been probing themolecular cology atMcArdle Laboratory forCan mechanisms ofthetoxiccompound for Dead Sea Scrolls dispute cer Research at the University of Wis some 20 years,tryingtostayoutofthe consin atMadison, who isolatedthedi political and scientificjungle that en oxin receptor in 1976. The potency of velops the chemical's toxidty. ange,whichwas widelyusedduringthe dioxins or ofPCBs isa directreflection These researchers are revealing the Vietnam War.An accidentatachemical of theirabilityto bind with the recep intricate details of dioxin's activity in plantinSeveso,Italy,ledtowidespread tor-- and TCDD binds most avidly, ex cells, where it now appears that the localcontaminationin1976. plainsJames P.Whitlock,Jr.,apharma chemical mimics a hormone. Their dis Haims thattheherbicidehad injured cologistatStanfordUniversity. coveries could elucidate dioxin's di Vietnam veterans and chemical plant The receptor-dioxin complex binds verse biological effects to support workers touched offnumerous investi withDNA atwhat may be one ofsever or contest conflicting epidemiological gations. Since then, animal data have aldioxin recognition sites.This site is studiesand couldalsoprovidevaluable shown TCDD tobe lethalatsome dos a regulatory region located upstream insights into gene expression and tar es and to cause a host of different from a gene that encodes an enzyme get genes in general. "The next five effects-- including cancer, thymus and from the cytochrome P-450 family.Al years are going to reveal what genes liver damage, birth defects and im though dioxin induces the expression dioxin affects and how they relate to mune-system depression-- thatvaryby ofothergenes aswell,thisresponse is important toxic effects such as cancer spedes. Epidemiological studies find themost thoroughlydocumented. and immune suppression," predicts ingincreasedoccurrenceofsoft-tissue The cytochrome P-450 proteinworks i EllenK. Silbergeld, a toxicologistwith sarcoma andnon-Hodgkin'slymphoma todetoxifycells.This "garbage dispos the Environmental Defense Fund. "At inpeople exposed to dioxinhave been al enzyme," as Poland describes it, last we have a molecular handle on hotly contested; studies finding no helps break down fats.Increasing lev what's goingon." suchassociationshave alsobeenwide els of this enzyme, however, can lead There are 75 kinds of dioxins, but ly criticized. One uncontested long to theformationofpotentiallydanger the most potent, 2,3,7,8-tetrachiorodi- term human effect is chloracne-- a ous compounds. One researcher de benzo-p-dloxin, also called TCDD or, sometimes disfiguringskincondition. scribesP-450asa double-edgedsword. simply, dioxin, has been the principal Understandingdioxin'sbiologicalac The effect of dioxin on P-450 levels object of scrutiny. This compound is tivityentailsfinding a rnwhantsm that and theconsequence oftheselevelsre created during combustion and as a explainswhy differentspedes respond main tobe explored. contaminant in some chemical manu indifferentways.Why, forexample,are Despite dioxin'sabilitytoinitiatethe facturing processes. Trace amounts hamstersunaffectedby a dose thatcan metabolismofmany othercompounds, were presentinthedefoliantAgent Or- killa guinea pig? Why do*female rats TCDD Itself is not broken down, SOURCE: JAMES P. WHITLOCK. JR.. STANFORD UNIVERSITY 16 S a o m n c A m e r ic a n N ovem ber 1990 GENP 0 1 1 5 3 6 784347 "which la one of the reasons to be con* makeup, could modulate TCDD's ef of dioxin, but ovarlectomized rats and cemed about It," Whitlock says. TCDD fect. The region on the DNA is "like an male rats do not--even after "huge1* persists In the body and has an esti- electrical switch with a dimmer: TCDD doses. "Dioxin's carcinogenic effects mated half-life of five years. may turn It on, but something else are probably related to cell prolifera Dioxin may also regulate other target turns It up or down," Whitlock says. tion," Luder says. In turn, "these path genes. SUbergeld and her colleagues at Dioxin does not appear to damage ways are related to estrogen." the University of Maryland, where she DNA. That, along with other laboratory Although Its role In cancer remains Is a visiting professor, have Identified evidence, has led researchers to postu undear, TCDD "Is unique in that few six genes, aside from that for P-450, late that It could be a cancer promoter carcinogens bind to a specific intra whose expression is affected by dioxin. rather than trdurpr--mpantng that oth cellular receptor," Greenlee says. Track William F. Greenlee, a toxicologist at er factors could Initiate a cancer but ing It may reveal how rhpmtrai carcino the Chemical Industry Institute of Tox that TCDDwould help It along. Indeed, gens alter normal growth processes. icology In Research Triangle Park, animal experiments have shown that Recently Whitlock discovered that North Carolina, has found evidence of dioxin promotes the formation of tu TCDD may bend DNA--a potential two more. mors once a cardnogen has been intro stage In gene transolption. "In the test The dioxin receptor has been ob duced into the cell. tube, the binding of the receptor to the served In many kinds of cells from In the laboratory, TCDD has also DNA bends the DNA, distorting It," many species, In c lu d in g h u m an . and been shown to affect the normal Whitlock says. Usually DNA Is packed sharks--suggesting some evolutionary growth and differentiation of human Into the nudeus and hard to get to. significance. Although there Is some skin cells and other tissues, Including When bent, DNA Is exposed and may variation in the receptor between spe rat liver. George W. Luder, a biochemist become more accessible to proteins in cies, investigators say these differences at the National Institute for Environ volved In gene expression. "It is as if alone do not account for the varying mental Health Sciences, has found that that region has been opened up in toxidty observed between animals. TCDD will cause cancer in the rat liver some way," he observes. One explanation is that other factors, only in the presence of estrogen. Fe Whitlock and others have also deter such as the environment or genetic male rats develop cancer after doses mined the seven-nucleotide sequence of the site where the dioxin-receptor complex binds to the DNA(nucleotides are the component molecules of DNA). This pattern recurs four or five times in the same region--all within 400 nu- deotides of one another. Such repe tition is "not a random occurrence," Whitlock says, and supports the idea that the site served an evotutionarily beneficial purpose. The existence of the receptor and the binding sites has suggested to re searchers that TCDDmay be analogous to a hormone. The "true" fit to the receptor could be an as of yet undis covered compound. "The natural com pound must be very Important," Sllbes> geld says, who notes that dioxin Is hot- mondlke in that It reversibly binds to a spedfic protein receptor. The receptor, however, may not have any physiological counterpart. "Maybe the toxicological response is a vesti- gal one," Poland says--meaning that the receptor bound with something external that may no longer exist, leav ing P-450 as a response to the environ ment. But, Poland asks, why should Immune suppression or cell prolifer atian be triggered by environmental crmfaTTrinanfg 7 Whatever the outcome, though , the new TCDD data are more than support ing evidence in an epidemiological de bate. Just as morphine led to the dis- - coveryof endorphins, dioxin is leading to an understanding of gene expres sion. "This was a seaet wired part of the body. Dioxin puahed.the button, and it lit up," Poland says. "Dioxin is a great poison, and poisons ehiddatn physiology." --MargueriteHoBawcy 20 Scientific American November 1990 G E N P 011537 784348 AMERICAN INDUSTRIAL HEALTH COUNCIL SCIENCE COMMENTARY Vol. 2, Issue 2 December 1990 704349 G Eh A l 538 Background -- - ... -- - . . -- -- - In June 1988, (see Science Commentary, Vol. 1, Issue 1, Item 9), an Environmental Protection Agency (EPA) inter-office dioxin working group issued two draft reports, "A Cancer Risk-Specific Dose for 2,3.7,8-TCDD" and "Estimating Exposure to 2,3,7,8-TCDD." These reports described the EPA working group's decision to.recommend a revised standard for exposure to dioxin. In the latter draft the working group recommended changing the existing RsD (Risk-specific Dose) value for dioxin by a factor of 16, on the basis that the available scientific information did not support continued reliance on the default assumption requiring use of the linearized multi-stage (LMS) 1330 Connecticut Avenue, N.W. Suite 300 Washington, D.C. 20036-1702 (202) 659-0060 Fax (202) 659-ig(& Amaran Industria) Haaltti Coundl CHAIRMAN, Chart A. Rulbai, Amsdcan Cyanamid Company VICECHAIRMAN, procedure, and on the working group's judgement that the weight-of-the-evidence supported a less stringent R.WH*y Boums Eastman KodakCompany Eiaculivi Commlttss R. Wllsy Bourna, Chairman Eatsran KodakCompany * standard. Itwas the EPAworking group's contention that new scientific information supported a change in the dioxin exposure standard set in 1985. Shortly after the release of the working group's draft reports, the Agency Cornar IL Fay, Otputy Ctoirmtn P.L Thibaut Brian, Ph.O. Air Produca and Chsmicals, Ine. . requested that they be reviewed by EPA's ScienceAdvisory Board (SAB). An ad hoc panel was appointed by the SAB Kenneth N. Robsrison Exxon Chemical Americas to conduct such a review. Peter W. Ifland, Ph.D, Procter &Gamble Company The SAB ad hoc dioxin panel generally agreed with Bruce W. Karrt, M.D. E.I. du Pomde Nemours &Co,, Ine. the EPAworking group's interpretation ofthe science, but Thomas H. Lattone stated in their draft report that the available evidence did Monsanto Company Kaith R. UeKsnnon The DowChemical Company not compel a change in existing reliance on the LMS model. The SAB ad hoc panel noted that although "there Chartas A. Rulbal American Cyanamid Company Committee Chairs SCIENCEPOLICYCOMMUTES Ron Van Myrten are promising alternative models which may be expected to more accurately reflect the biological basis of 2,3,7,8TCDD carcinogenesis, such newer models need to be further developed and validated." In effect the ad hoc Union Carbide Corporation , panel recommended continuing the LMS procedure as Jot) R. Bandar, Ph.O., M.D.(Vlc&Chair) Owens Coming Fiberglas Corporation the default option of choice; however, the panel strongly SCIENTIFICCOMMITTEE urged EPA to build upon the working group's excellent Gerard F, Egan, Ph.0. ExxonBiomedical Sciences, Inc, review and move "to develop and validate a new risk model Donald Hughes, Ph.O. (Vice-Chair) capable of more accurately estimating the risk of hum an Procter &Gamble Company Donald E. Stevinson, Ph.0. (Vlca-Chafr) cancer caused by dioxins and related compounds." Shell Oil Company EPIDEMIOLOGYSUBCOMMITTEE M.Jana Teta, Ph.Q,, M.P.H. AIHC A ctions, UnionCarbide Corporation MUTAGENICITYSUBCOMMITTEE Robert A. LeBoeut, PtiO. (Co-Chair) Procter AGamble Company In a letter of February 23, 1989, to the chairman of the SAB ad hoc dioxin panel, AIHC supported the EPA Al U, Pti.O. (Co-Chair) working group's position for a revised dioxin standard Monsanto Comoanv REPRODUCTIVEi DEVELOPMENTAL and expanded on what the Council perceived as the EFFECTSSUBCOMMITTEE nature of the EPAworking group report and the weight- George Deaton, Ph.0. Procter &Gambia Company of-the-evidence process the group used to reach their NEUROTOXICOLOGYSUBCOMMITTEE recommendation. AIHC disagreed with an SAB panel WaynsDcughtray, PhJ). Exxon Biomedical Sciences, Inc. RISKASSESSMENTSUBCQMMTTTEE statem ent th at the change In the dioxin RsD, as recommended by the EPA working group, was a "science Donald Hughes, Ph.O. Procter &GambiaCompany* DELIVEREDDOSEWORKGROUP policy" choice. AIHC suggested that the working group's documentwas a "hazard characterization" that supported Alin G. Wilson, PfcO. (Chair) Monsanto Company San)or staff reconsideration of the exposure standard for dioxin. It Justly questioned the applicability of two default 784350 RonaldA. Lang, Prasdam assumptions: the linearized multi-stage (LSM) procedure, GaylanC. MBard, Diractor and the use ofthe standard surface area scaling factor for Mtfda G. Lawson, Managar, Communications Dans W. McMahon, MambmNp Coordinator interspedes conversion (body weight to the 2 /3 power). CaroiaJ. OTooia, Managar, Soones Policy J. David Sandtar, Managar, Sdsntific Commit* M. J. Sloan. Consuiam In verbal and written communications presented to Anna P. Santslta, Assistant Managar the SAB Executive Committee on October24,1989, AIHC Robart Barnard(Lagal Counsel) Claariy, Gottiiab, Stean &Hamtiton again supported the EPA dioxin working group's recommendations and questioned the reluctance of tbi- l-vTTk A1 1 I 3~j SAB ad hoc panel to recommend against the scientifically inappropriate LMS default option. Further, AIHC stated that the SAB ad hoc panel's recommendation seemed to imply that the Agency should not permit its scientific staff to exercise expert judgement to choose from other scientiflcaHyplausible options an alternative to the obviously inappropriate LMS default option. AIHC noted that the deliberative process used by the EPA working group was consistent with the Agency's own Guidelines for Carcinogenic Risk Assessment and with the Office of Science and Technology Policy (OSTP) principles. AIHC observed that by using this process the working group had rightly concluded that use ofthe LMS default option was not supported by the science and had chosen instead, based on the working group's scientificjudgement, one value for the RsD from among a spectrum of alternative values generated by other models and procedures. Present Status------ ----- ----- ----- ------------------------------------------------------------- - On November 28, 1989, the SAB ad hoc panel on dioxin issued their report to the chairman of the SAB. Subsequently, the SAB Executive Committee sent a summary of that report to the EPAAdministrator. Pertinent excerpts are as follows: "It should be noted that the "Dioxin" Panel was not asked, nor did it choose to address directly or in, detail, the adequacy of the Agency's 1985 cancer risk assessment for 2,3,7,8-TCDD. However, in the course ofthe current review, the Panel generally agreed with the EPA Working Group's criticism of the linear multi-stage model as applied to the specific case of 2,3,7,8-TCDD in 1985." "This criticism reflects, in part, the existence of a series of important and innovative mechanistically oriented studies which have provided new insight into the toxicological effects of2,3,7,8-TCDD and related compounds. The EC (SAB Executive Committee) joins the Panel in encouraging the Agency to support research which will incorporate this new information into risk assessment approaches currently under development, where appropriate." "The EC concludes that the existing, LMS-based risk assessm ent for 2,3,7,8TCDD lacks a firm scientific foundation. However, until the new approaches are fully developed and peer-reviewed, estimates based on othermodels are equally questionable. Unfortunately, the direction and extent of any change from the LMS-based risk estimate th at might result from the application of more appropriate models cannot be determined at this time." 784351 EPA's review of the 1985 dioxin standard, and the SAB ad hoc panel's recommendations, have provided a significant new opportunity for the Agency to look at biologically based, less ultra-conservative risk assessm ent methodology. As a result of the SAB's recommendation, EPA published a notice (55 FR 30513), "Development of a Biologically Based Model for Dioxin", which announced "a working meeting to be held by EPA's Office of Health and Environmental Assessment (OHEA) and the Institute for Evaluating Health Risks to discuss the development ofa preferred model for quantitatively expressing the carcinogenic risk of2,3,7.8-tetrachlorodibenzo- p-dioxin CTCDD, dioxin)." The topic of this meeting, which was held August 2,1990, was also discussed at the October21-24,1990, Banbury Conference on the "Biological Basis for Risk Assessment of Dioxins and Related Compounds." ^ -r-.-*TM Background-------------------------------------------------------------------------------- --- ------ In late 1989, the Health and Environment Subcommittee of the House Energy and Commerce Committee reported out H.R 3030, an amended version of the Bush Administration's Clean Air proposal. This bill deleted the "bright line" approach of Senate bill S. 1630, which required use of a numerical standard to establish what constitutes risk to a toxic air pollutant for the purpose of regulation, such as 10*4or 10*. Under H.R. 3030 the "residual risks" of toxic air pollutants (those risks remaining after "maximum achievable control technology" (MACT) is installed] would be evaluated after initial technology controls are in place and a determination made by EPA as to whether further risk reduction measures are necessary. The final version of the House bill, as reported by the full Energy and Commerce Committee on April 5, 1990, did not contain the Senate's "bright line" approach. AJHC A ction------------- --------------------------------------------------------------------------- AIHCwas concerned that amendments to H.R 3030 might be introduced during deliberations of the full Energy and Commerce Committee, which would attempt to establish a numerical "bright-line" standard of residual risk for a maximally exposed individual (MEI) of one in a million. AIHC's Scientific Committee stated their opposition to this approach in a December 22, 1989, letter to Congressman John D. Dingell, Chairman of the Energy and Commerce Committee. In their letter, the Scientific Committee supported the "residual risk" provisions of H.R. 3030, as reported by the Health and Environment Subcommittee, as providing a sound framework for risk decisions. AIHC noted that H .R 3030 was "consistent with scientific principles underlying risk assessment and would permit regulatory consideration ofanticipated advances in scientific information." AIHC cautioned that "any `bright* line approach removes EPA's discretion to address uncertainty and to use the best available science as a basis for regulatory action." On February 8, 1990, Congressman Dingell forwarded AIHC's letter to EPA Administrator William Reilly for review and comment. The House bill, as passed on May 23, 1990, did not adopt the "bright line" standard. 784352 Gfj 011541 On February 19, 1990, AIHC Board Member Emeritus Robert Barnard, Esq., addressed the 1990 Annual Winter Toxicology Forum on the scientific issues raised by the residual risk provisions in the Clean Air Act bills under consideration. He focused his analysis on the Senate Bill and misuse ofthe population-based statistical 10*4 and 10* "bright line" risk estimates to oredict the risk of anv j-- r& n line" standard had been criticized because their report had stated "the science for assessing these risks is yet too new to be locked into absolute limits." He said th Committee report had also noted that critics urged discretion and the use ofother risk dimensions and data in setting standards for cancer incidence, potency, and human exposure. Mr. Barnard stated that although the Senate report agreed that these concerns are legitimate, the Senate committee had rejected them because they felt that EPA had not followed "discretion" logic in fixing recent standards. He noted the Senate report further reasoned that "the `bright line' boundaries on acceptable risk contained in the Senate legislation are necessary to bring health considerations in the standard-setting process at the beginning." In closing, Mr. Barnard suggested that the scientific community might want to become educators of Congress on science issues in order to prevent passage of legislation with a "bright line" approach, or any other legislation that may be similarly scientifically flawed. 4>A lH C IssuesC ohceptPaperonC om blhingtheR e9idiiaIR isk oftheH ouseandSenateV ersidnsforaiiA niended i;. Background on Proposed Residual Risk Amendments to the Clean Air Act * Senate A ction--------------------------------------------------------------------------------- On April 3, 1990, the Senate passed S. 1630, its version of legislation to amend and reauthorize the Clean AirAct. The Senate bill contained the "bright line"approach and required industrial sources of 191 hazardous air pollutants to meet emission limits based on "maximum achievable control technology" (MACT) requirements to be developed by EPA and installed by industry within two to ten years. According to the Senate bill, before residual emissions could be required to be controlled further, a series of studies and reports must be completed concerning the "residual risks" from carcinogens that may remain after applying MACT. Of particular concern to AIHC were provisions in the Senate bill freezing the numerical standard "bright line" approach, and the current risk assessm ent methodology, unless Congress agreed to change the law within five years. The bill did contain the following provisions which had potential to bring about changes in existing risk assessm ent methodology: Within three months after passage, the Administration must request the NationalAcademy ofSciences (NAS) to undertake a study and make recommendations for changes in the current risk assessment methodology. A report on the NAS study would be due two years after passage of the bill. * Eighteen months after passage ofthe bill, a ten-member Risk Assessment and Risk Management Commission m ust have been appointed: three members appointed by the President, three by the House, three by the Senate, and one by NAS. GENP 0J1542 784353 nn * Following the receipt of the NAS report EPA m ust issue new risk assessment guidelines accepting or rejecting the NAS recommendations. The Risk Assessment and Risk Management Commission was charged to review the NAS report, the EPA guidelines, and undertake any necessary additional study. Areportby the Commission, with legislative recommendations, was to be made to Congress and the President not later than 42 months after passage of the bill. Congress would consider the Commission's recommendations under an accelerated schedule. House Action OnApril 5,1990, the House Energy and Commerce Committee passed H.R. 3030 and on May 23, 1990, this legislation passed the full House. The House version adopted language similar to that in the Senate bill on air toxics. Both bills required that any source emitting one or more of 189 to 191 listed air pollutants to install MACT within ten years. However, the two bills differed on residual risk --the risk remaining after MACT is applied. The House bill did not adopt the Senate's "bright line" standards and deferred the issue ofresidual risk for ten years after enactment, or eight years after MACT is installed, whichever was later. At that time Congress m ust act or not act based on recommendations developed from a study conducted by EPA and the Surgeon General. IfCongress did not legislate stricter standards, EPAcould lower the emission standard to provide an "ample margin of safety,"in accordance with the language of Section 112 of the 1977 law. AIHC A ction___________________________________________________________ AIHC has always taken the position that it is not appropriate to legislate regulatory criteria in the form ofhypothetical statistical values, i.e.f one-in-a-million risk (10'6) to the most exposed individual. This value, calculated on the unrealistic assumption of 70 years residence at constant exposure, relates to a random representative of the exposed population and was never intended to serve as an estimate of risk for any actual person. Based on these and other scientific reasons, AIHC has strongly opposed the establishment of any numerical "bright line" standard for regulating risks. Both the House and Senate proposals contained provisions for the evaluation, and, as necessary, regulation ofresidual risks remaining after mandated technological controls are in place. These provisions were to become operative within five to eight years after enactment. Both bills provided for Congressional review and had "default" provisions if Congress failed to act. 784354 E <011543 To express AIHC's views on the "bright line" approach, and to suggest a framework for the evaluation and regulation of residual risks, AIHC developed a concept paper, "Residual Risk: A Conceptual Framework for a Conference Proposal." It was hoped that AIHC's approach would aid the House-Senate conference committee in their deliberations for a compromise on the air toxics provisions in the House- and C n w rt* /. J --------- ` " - The AIHC paper addressed ways to assure that the evaluation and regulation of residual risk would be based on the latest and best scientific information. AIHC suggested that the compromise bill would be enhanced and strengthened by a combination of: ---- the House provisions and criteria on residual risk, the Senate provision for an in-depth study of risk assessm ent by the National Academy of Sciences, the Senate provision for an EPA revision of risk assessm ent guidelines in response to the Academy study report, and the Senate provision for establishment of the Bipartisan Commission on Risk Assessment and Risk Management. Present S tatu s-------------------------------------------------------------------------------------- Conferees on the House-Senate conference committee reached a compromise on the legislation during the closing days of the last session of Congress. Both the House and the Senate ratified the Clean Air Act Amendments of 1990 on October 26, 1990. The President signed the measure into law on November 15, 1990. There is evidence that several ofAIHC's recommendations were considered in the air toxics provisions of the legislation. An analysis and study of the act (750 plus pages) is still underway; however, the following summarizes the residual risk provisions which have been of direct concern to AIHC: National Academy of Sciences Study: Within three months of enactment the EPAAdministrator m ust enter into arrangements with NAS to conduct a study of risk assessm ent methodology. Within thirty months of enactment the NAS m ust submit a report to the Senate Committee on Environment and Public Works, the House Committee on Energy and Commerce, the Risk Assessment and Management Commission, and the EPA Administrator. EPA and Surgeon General Report: Not later than six years after enactment, the EPA in consultation with the Surgeon General, shall report their recommendations to Congress as to needed legislation regarding residual risk. 784355 + Default Provision: If Congress does not act on the several recommendations within eight years, EPA shall promulgate emission standards which provide "an ample margin of safety to protect public health." If the technological standards (MACD for those pollutants classified as a known, probable or possible hum an carcinogen do not reduce lifetime excess cancer risk to the . individual most exposed to less than one-in-one million (10'), then further regulation of residual risk will be considered. (Note the 10'standard is a trigger, not a "bright line"; thus EPA4s not prevented from looking 0,1 dimensions of risk, as it did in the 1989 benzene rule.l TM- 3 B ackground------------------------------------------------------------------------------------------ On December 29. 1989, the Reproductive and Cancer Hazard Assessment Section, California Department of Health Services (DHS) released for review "Draft Guidelines for Identification and Hazard Assessment ofAgents Causing Developmental and/or Reproductive Toxicity." The guidelines, prepared to satisfy the mandates of Proposition 65, were announced as providing DHS, and the "scientific community of California," with guidance "for the analysis and interpretation of information relating to the potential of an agent to cause reproductive or developmental harm." The draft guidelines addressed the first two stages of the risk assessment process, Hazard Identification and Dose-Response Assessment/Reference Dose Determination (RfD). The draft identified three types of reproductive toxicity - male, female, and developmental toxicity. It also provided rules for assessing epidemiological studies and for the proper use of these studies as well as the draft proposed rules for animal studies, DHS suggested that the draft guidelines be implemented in conjunction with pertinent EPA and National Research Council (NRC) guidance documents. The later stages of the risk assessment process. Exposure Assessment and Risk Characterization, are not covered in the draft. Comments on the draft guidelines closed May 1, 1990. After DHS review, the guidelines may be amended and re-released for public comment. AIHC A ction------------------ ---------------------------------------------------------------------- OnApril26,1990, AIHC's Reproductive and Developmental Effects Subcommittee supplied comments on California's draft Reproductive Toxicity Guidelines through the Technical Committee ofthe Environmental Working Group, a California industrysponsored organization. The AIHC comments commended DHS for recommending that a "weight-of-theevidence" approach be used to evaluate reproductive and developmental toxicity data, and supported the California agency's position to incorporate all epidemiological data from both positive and negative studies. However, AIHC cautioned that it was unclear ju st how the weight-of-the-evidence process would be used by DHS and whether certain "levels of evidence'* would be subsequently proposed for incorporating reproductive or developmental toxicants into Proposition 65 lists. AIHC also noted that it was often unclear as to whether certain DHS positions expressed in the guidelines were based on science or on DHS policy. GETSiP 011545 784356 In their comments AIHC pointed to the implications of Kamofsky's Law, which states that any compound can be positive in a developmental toxicity assay under some combination of experimental conditions. The AIHC subcommittee suggested that "any categorization of compounds needs to be qualified by the circumstances under which the compound is positive or negative/ They stated that the DHS guideline "appears to equate lack of data with hazard" and suggested that "it Is not scientifically appropriate to assume a hazard exists in the absence of data." The proposed DHS guidelines were largely based on the EPA "Guidelines for the Health Assessment of Suspect Developmental Toxicants", as reviewed by the EPA/ SAB Environmental Health Committee on October 27, 1989, and currently under revision. (See below Item 6 and Science Commentary, Vol. 1, Issue 2, Item 1, October 1989.) AIHC recommended that the DHS document not be published in final form until the EPA guidelines have been amended to reflect the Environmental Health Committee's recommendations. 6 EPA^s Environmental Health Committee Review o f /Proposed . AmCTdmCTts to the Giddelihes fbr t^ :*;7yqf Stlspect Develbp^lental T oricallts*,' Present S ta tu s-------------------------------------------------------------------------------------- As reported in the Science Commentary (Vol. 2, Issue 1, Item 6, June 1990), the EPA Science Advisory Board's (SAB) Environmental Health Committee met on October27,1989, to review EPA's document "Proposed Amendments to the Guidelines for the Health Assessment ofSuspect Developmental Toxicants. "AIHC's Reproductive and Developmental Effects Subcommittee addressed the SAB Committee at the time of their October 27 deliberations, principally on three areas of interest: the proposed weight-of-the-evidence classification scheme, the quantitative evaluation/extrapolation of data, and maternal toxicity. On April 23, 1990, the SAB and the Environmental Health Committee issued a report on the Committee's review of the proposed amended guidelines. In general the Committee found the amended document "to be adequately founded in toxicological and teratological science, and to reflect the current status in these fields." The Committee did suggest th at the Agency rethink their proposed weight-of-theevidence classification scheme "in order to avoid bonfusion with more commonly applied uses of such classifications, and to develop a more powerful conceptual approach." Further, the Committee recommended that "the Agency should begin to move away from the current use ofthe No Observed Adverse Effects/Lowest Observed Adverse Effects Level (NOAEL/LOAEL) basis for calculating the Reference Dose, to a benchmark dose/confldence limit approach, tied to empirical models of doseresponse relationships." Several ofAIHC's recommendations were given consideration in the SAB report. In particular, the Committee was in close agreement with AIHC's concern for the proposed weight-of-the-evidence classification scheme. 784357 Background___ _ - _. ______ - - Public Law95-622, signed into law on November8,1978, requires the Secretary, DHHS, to publish an annual report containing "..a list ofall substances (i)which either are known to be carcinogens orwhlch may reasonably be anticipated to be carcinogens and (ii) to which a significant number of persons residing in the United States are exposed...." In addition, the annual reports are intended to provide information "on the nature and degree of exposures to such carcinogens and the extent to which Federal regulations are effective in reducing the risk to the public health from exposures to these substances." Responsibility for preparation of the annual report was delegated by the Secretary DHHS to the Director of the National Toxicology Program (NTP) in 1979. To date five annual reports have been issued, the first in July, 1980, and the last during the fall of 1989. Some 162 substances have now been selected by NTP for inclusion on these lists. Within the past fewyears Federal regulators, and state legislators and regulators, when drafting new laws and regulatory initiatives, have, directly or by reference, included for regulatory restrictions, without further study, the list of "chemical carcinogens" as published in the Annual Report on Carcinogens (ARC). The bestknown examples of regulatory programs affected to date include the Occupational Safety and Health Administration (OSHA) Hazard Communication Standards, the Environmental Protection Agency's community right-to-know program under Title III of the Superfund Amendments and Reauthorization Act (SARA) of 1986, and the listing of carcinogens under California's Safe Drinking Water and Toxic Enforcement Act of 1986 (Proposition 65). The ARC is also a basis for listing carcinogenic or hazardous substances under right-to-know laws in Alabama, Delaware, Florida, Illinois, Maine, Massachusetts, Michigan, Pennsylvania, and Vermont. Other states indirectly adopt the ARC list by incorporating the OSHA Hazard Communication Standard into their laws or regulations by reference. 784358 This unintended, but almost automatic incorporation of the ARC lists into other laws and regulations, continues to occur even though DHHS and NTP have recently emphasized, both in a "Note to the Reader" and in the "Introduction" of the Fifth Annual Report on Carcinogens (1989), that the ARCs "are informational documents which serve as meaningful compilations of data on the carcinogenicity of the listed substances in humans and/or animals.... The reports represent the first step' in hazard identification of the substances selected for inclusion..,.The evaluation of the degree of hum an risk from substances Included in the Annual Reports requires a wider analysis than has been made in preparing the Reports. That is properly the purview of the Federal, State, and local health regulatory and research agencies authorized to Implement the laws relating to carcinogens." GENP011547 A review of p ast events sh ow s that NTP had previously stated, bu t had not widely publicized, that the ARC is based only on the first step in hazard identification. It was, and is, NTFs position that any evaluations beyond the first step is "the province of the regulatory agencies, not NTP in its preparation of the A nnual R eports....'' In a July, 1989 declaration, NTP defined the second step in hazard identification (as part of the risk a sse ssm e n t process to be performed by the regulatory agencies) as "the consideration of other data that are pertinent to risk assessm en t, i.e., assessin g the estim ated level of h u m an carcinogenic risk." NTP goes on to explain that "the 'weightof-evldence' approach is u sed in su ch an a ssessm e n t, w hich in clu d es a review of non only the epidemiological and experim ental animal carcinogenicity data, but also, if available, pharm acokinetics and m echanism s of action data. M echanism s of action studies are performed in an effort to ascertain the m echanism at a cellular and subcellular level by w hich a chem ical produces a toxic effect, e.g., cancer. It is the province of the regulatory agencies, not NTP in its preparation of the A nnual Reports, to perform this second step in hazard identification as part of the risk a ssessm en t process." In spite of these statem en ts, the NTP position w as not generally known in the regulatory community, and certainly not by the general public. Due to concerns for the m isu se of the ARC list of carcinogens, and public m isunderstanding regarding the purpose of the lists. DHHS. in December 1987. requested that the inter-agency Committee to Coordinate Environmental Health and Related Programs (CCEHRP) review how the ARCs are developed. As a result of the CCEHRP review, which considered recom m endations from several groups including AIHC, language as d iscu ssed above w as added to the "Introduction" of the ARC, and in the "Note to the Reader." w hich does help to clarify the purpose and intended use of the annual reports. Regardless of these steps toward clarification, there remains a pressing need to inform u sers of the report that the purpose of the ARC list is to convey the resu lts of the first step of hazard identification, and that the ARC is not m eant to report the results of a complete hazard identification, which can then be used as the primary basis for regulatory restrictions of listed su b stan ces. AIHC A c tio n ___________________________________________________________ Over the years. AIHC h as interacted w ith DHHS on m any different scientific and policy issu es. Since 1986 the Council has m aintained dialogue with officials in DHHS, NTP, and w ith CCEHRP concerning the im portance of estab lish in g the b est available scientific basis for the developm ent of the Annual Reports on Carcinogens, and the need to ensure that the ARCs are neither m isinterpreted nor m isused for regulatory and/or legislative purposes. In Decem ber, 1986, com m ents to NTP on the draft Fifth A nnual Report, AIHC focused their d iscu ssio n on th ose su b sta n c es that had been included on the ARC list based on experimental evidence from animal tests. To more accurately describe the b a sis NTP u sed for listing su ch chem icals, AIHC recom m ended that language be added in the ARC "Introduction" that w ould stress the need for an an alysis of the data beyond carcinogenitic activity in anim al tests. Specifically. AIHC su ggested that the wider analysis sh ould include "inform ation and data not considered in the original n'FTsTP 011548 784359 19. review and data generated subsequently"--and should consider "the physiological, pharmacological, and toxicological differences between test animals and humans." AIHC's Scientific Committee reaffirmed these views in testimony before an NTP sponsored meeting on the ARC in April, 1987, AIHC urged that the ARC state unequivocally that no evaluation had been made concerning use of the report for regulatory or'corrective action, and that the ARC also state that no decision concerning regulatory use of the lists should be made without further scientific evaluation. As alternatives, AIHC proposed that NTP either perform a full weight-ofthe-evidence hazard evaluation, including an assessment of relevance to humans, or simply continue to review positive data without a full review, but provide a prominent notice of the limited nature of the scientific review involved and place qualifications on possible regulatory or legislative use of the ARC lists. (This latter course was adopted in part by NTP stating certain limitations in the "Note to Reader" and "Introduction" of the Fifth Annual Report, as released in the fall of 1989). AIHC's views, as outlined above, were expanded in July, 1988 testimony before the CCEHRP subcommittee convened to review the criteria and process used by NTP in preparing the ARCs. In particular, AIHC suggested that the Secretary, DHHS, should assume responsibility for educating regulators and the public as to the strengths and weaknesses of the ARC, and how the ARCs should or should not be used. In spite of the progress made by NTP to clarify and explain the meaning and purpose of the ARC, AIHC believes that there remains a growing need for DHHS to initiate a more aggressive communications program directed to the users of the ARC, particularly the regulatory community and legislators. To discuss this situation, AIHC met with the DHHS Assistant Secretary for Health, Dr. Jam es Mason, and his staff, on October 17,1989. This meeting has resulted in an exchange ofcorrespondence in which AIHC has again outlined concerns, and presented further examples ofwhere the ARC continues to be a primary basis for regulatory restrictions. Also discussed at the October 17th meeting were the problems of incorporating new scientific information into the risk assessment process, and the application of negligible risk standards under the Federal Food, Drug and Cosmetic Act. a H z 4^ VO 784360 AIHC Science Commentary Order Form: (Voi. 2, Issue2) To obtain a complete copy of AIHC comments summarized in this issue check the appropriate summary number below: AIHC #1____ ; #2____ ; #3____ ; #4____ ; #5____ : #6____ ; #7____ ; N A M E ________________________________________________________________________ I A F F IL IA T IO N _________________________________________________________________ STREET ADDRESS OR BOX#__________________________________________________ CITY, STATE, ZIP CODE_____ COUNTRY OTHER THAN U.S., R eturn to: AIHC 1330 Connecticut Avenue, N.W. Suite #300 W ashington, D.C. 20036-1702 OEtiP O 550 784361 \ Neurological Effects Q. Is there epidemiological evidence to suggest that PCBs cause neurological problems in children? A- There are two studies in the U.S. (Michigan and North Carolina) that suggest that environmental exposure to organic contaminants may cause minor defects in learning abilities in children. These defects have not been clearly linked to PCBs. Other contaminants, which might be the cause, have not been measured. In the Michigan study, there was no correlation between the deficits and PCB levels. Q. What are ortho substituted congeners? A. These are PCB molecules that have chlorines in either of the four positions around the bond connecting the two rings in biphenyl. These congeners may have chlorines in some of the meta and para positions as well. Most PCBs are orthosubstituted. Q. What is the significance of the NYS finding that certain "ortho" congeners affect dopamine levels in rats and monkeys and tend to concentrate in brain tissue? A . There are numerous questions that need to be answered before the significance > can be determined: t. Exposure levels in the NYS tests were very high (500 and 1000 ppm 1254 in one study and up to 3.1 mg/kg-day 1016 in another.) These levels are much higher than humans exposed either occupationally or environmentally (e.g., eating fish). The NYS papers do not allow one to determine the response at various doses -- therefore the possible relevance of these data to human exposure is not clear. 2. The significance of dopamine level depressions of the magnitude found is unknown. Dopamine levels are depressed by other chemicals, such as alcohol, reversibly. The nature of the PCB effect may be simply anaesthetic and also reversible. 3. The in-vitro tests on rat brain ceils do not necessarily represent what happens in living mammalian systems. For example, 2,2* dichloro-biphenyl is a significant depressant for dopamine in cell studies, but is metabolized so rapidly in vivo that it will not accumulate in the brain. Q. Is there any evidence for an association of PCB exposure and Parkinson's disease? A. None of the occupational mortality or living worker studies have suggested that Parkinson's disease might be related to PCBs. If the disease does not show up with highly exposed workers, it is not likely to result from lower, environmental exposures. GENP 011551 784362 -2- NIOSH (Sinks) Study of Westinghouse Capacitor Workers (Bloomington, Indiana) Q. What were the major findings of the NIOSH study? A. 8 deaths due to malignant melanomas vs. 2 expected 5 brain cancers vs. 2.8 expected 54 deaths due to all cancers vs. 63 expected Q. Were any of the results related to PCB exposure? A. No. Attempts to relate exposure of people who died of melanoma and brain cancer did not identify a relationship. Q. Was there a relationship of latency period and exposure to PCBs? A. No. Q. Is there a biological rationale for a relationship of melanoma and brain cancer with 3 PCB exposure? ' I A. No. No rationale was presented in the paper. Q. Do the major findings from the study concur with those from other PCB mortality studies? A The absence of a relationship of cancer deaths to PCB exposures is consistent with other studies. The excesses of melanoma and brain cancer are not consistent with the results from other PCB mortality studies. For example, the study of GE workers by Taylor (NCI), which is most closely related to the NIOSH study, found no excess of melanoma and a reduction in brain cancers. Like the NIOSH study, the Taylor study found a reduction in deaths due to all cancers, combined. 784363 GENP 011552 -3- Dloxln-like PCB Congeners - TEFs Q. What are "dioxin-like PCB congeners"? A. PCB structures having both para positions chlorinated, two or more meta positions chlorinated, and no more than one ortho position chlorinated -- also known as coplanar PCBs. Q. What are TEFs? A. An approach to relating the toxicitles of dioxins other than 2,3,7,8-TCDD, dibenzofurans and coplanar PCBs to 2,3,7,8-TCDD, which is considered to be the most toxic of these compounds. Q. What is the relationship of the coplanar PCBs to 2,3,7,8-TCDD? A. The coplanar PCBs exhibit the same type of acute toxicity as TCDD -- thus include such end points as lethal dose, immunosupression, enzyme induction and skin effects. However, the relationship does not hold with respect to carcinogenicity. While TCDD is a powerful carcinogen, the most carcinogenic PCB mixture, Aroclor 1260, contains a small amount of coplanar PCBs. Aroclor 1254, which is negative as a carcinogen, contains a relatively large amount of coplanar PCBs. Q. How does this impact on the concept of regulating PCBs on the basis of TEFs? A. Since both TCDD and PCBs are regulated as carcinogens, and since carcinogenicity of PCBs appears to be related to highest level of chlorination, not TEFs, the application of TEFs to PCB risk assessment does not appear feasible. Q. What are the levels of coplanar PCBs in Hudson River sediments? A. The anaerobic dechlorination taking place in the upper river "hot spots" preferentially destroys coplanar PCBs. This finding has been determined both analytically and in biological test systems. Q. What is the cancer risk related to PCBs in upper Hudson River sediments? A. In animal tests, only 60% Cl PCB mixtures are carcinogenic. The PCBs in the sediments are derived from 1242 (42% Cl) but have been dechlorinated to a mixture that will not accumulate in humans. Workers exposed occupationally to very high levels of 1242 did not have an increased risk of cancer. The PCBs in the sediment should not represent a risk of cancer to humans. SBH 6-17-91 784364 GENP 011553 Neurological Effects Q. Is there epidemiological evidence to suggest that PCBs cause neurological problems in children? A. There are two studies in the U.S. (Michigan and North Carolina) that suggest that environmental exposure to organic contaminants may cause minor defects in learning abilities in children. These defects have not been clearly linked to PCBs. Other contaminants, which might be the cause, have not been measured. In the Michigan study, there was no correlation between the deficits and PCB levels. Q. What are ortho substituted congeners? A. These are PCB molecules that have chlorines in either of the four positions around the bond connecting the two rings in biphenyl. These congeners may have chlorines in some of the meta and para positions as well. Most PCBs are orthosubstituted. Q. What is the significance of the NYS finding that certain "ortho" congeners affect dopamine levels in rats and monkeys and tend to concentrate in brain tissue? * A. There are numerous questions that need to be answered before the significance can be determined: 1. Exposure levels in the NYS tests were very high (500 and 1000 ppm 1254 in one study and up to 3.1 mg/kg-day 1016 in another.) These levels are much higher than humans exposed either occupationally or environmentally (e.g., eating fish). The NYS papers do not allow one to determine the response at various doses -- therefore the possible relevance of these data to human exposure is not clear. 2. The significance of dopamine level depressions of the magnitude found is unknown. Dopamine levels are depressed by other chemicals, such as alcohol, reversibly. The nature of the PCB effect may be simply anaesthetic and also reversible. 3. The in-vitro tests on rat brain cells do not necessarily represent what happens in living mammalian systems. For example, 2,2* dichloro-biphenyl is a significant depressant for dopamine in cell studies, but is metabolized so rapidly in vivo that it will not accumulate in the brain. Q. Is there any evidence for an association of PCB exposure and Parkinson's disease? A . None of the occupational mortality or living worker studies have suggested that Parkinson's disease might be related to PCBs. If the disease does not show up with highly exposed workers, it is not likely to result from lower, environmental exposures. GETSIP 011554 784365 THE WALL STREET JOURNAL TUESDAY, AUGUST 6, 1! The Dioxin Un-Scare--Where's the Press? By R za I r o n 3tadn has b en described as "Hie roost m carcinogen ever tested." An tin ted cmtamhiam; in some chemicals Industrial processes. It has been the et of studies fouled by the government :e tone of SMO milUon over the past de. Claims lodged by Individuals who redly suffered or even feared serious age to their health from exposure to a have cost businesses and the gov* lent aiM tH nw al hnndrgdg of millions of rs. Now a high government official USthatan ItinrrftnatifearOfdlmrtw Just a gdPHttflq mismntennnnrifnfT 1332. the government ordered the lanon of the 2Z32 residents of Times h. Mo. because traces of dioxin were ;in the soil At that time, the Centers Isease Control believed thatIngesting Jog containing as much as one part illlon of dioxin posed a significant risk nan health. The Environmental Pro> y spent S33 million to buy up declared Itto be a dangerous - >- passing thwmgft gswo, 44 were greeted with signs ng them to keep theirwindows closed ot to stop and leave their vehicles. ? government pinned the respoirihil*the contamination of Tiroes Beach '>other Missouri sites on SyntexCarp, f Its subsidiaries had bought a plant id once supplied dioxin-taintedwaste i contractor who had sprayed it on eets of Times Beach and the other Uter years of costly litigation, Synmed a consent decree a year ago ng to clean up (he sites and incinerme 100.000 cable yards of contamlaiL Cost estimates ran as ranch as illhra over the next decade. 1after the dptnoUHoq nf httllrilngfIp Beach began this spring. Or. Ver non Book, the CDC official who had reefiimPTwtftd fhft ftw n a f ln n In lfflg. rfmppri a bombshelL At an ftnrimnpiftntai confer* eoce in Missouri, he said that be would not be concerned about the levels of dioxin at Times Beaefa because scientific studies have shown that low doses of dioxin pose minimal himlth risks. Dr. Hook, who Is director of the Center for Environmental Health and Injury Con trol at the GDC, told reporters that he now believes that the evacuation of Times Beach was unnecessary. Asked what he wouldtell the formerresidents of the town, who underwent the trauma of being torn from their hemes. Dr. Hook said: "We should have bees more upfrontwith Times WftBhiw bIb and told them. 'We're doing our best withihe estimaies of the risk, but we may be wrong/1 think we never added 'but we may be_wrong.' *' In debunking the Hat that Hiorin jj 3 potent human carcinogen. Dr. Houk at tacked the scientific theory and methodol ogy that had led him and others to what he believes was a false conclusion. Dr. Houk said the method used to assess the risk was based on an assumption that violates a fundamental rale In toxicology: The dose makes the poison. It was asumed that feeding laboratory animals the maximum dose they could tolerate would enable sci entists to determine whether trace amounts of a chemical would cause cancer In humans. One obvious problem with this Is that different animals have different re sponses to the same chemicaL Whatcan be highly toxic to gnhtea pigs may have no ef fect on rats, and whatmay came cancer In mice will not necessarily have the same ef fect on ButDr. Houk's attack went beyond fids. He added his voice to that of Brace Ames, bendof the biochemistry deportment of the University of California at Berkeley, a leadingcritic 0f themethodsused to assess the cancer risk from chemicals. Mr. Ames rrmtrmi* that the animal tests are funda mentally flawed becausethemaximum tol erated doses of the chemicals being tested m i roik due to the shfvr site of the dos age. Mr. Ames argues that this can cause rapid cell division among the surviving cells, leading to cancer-causingmutations. This suggests that risks calculated from animal tests Involving maximum tolerated doses are greatly exaggerated. Dr. Houk says that most scientists now agree with this. He d ies the dlnxln case as "a good example of why we must use both animal and human data when evaluating the potential ftflfti't* of <*hftmli*al ex- posnre fortinmans." Theepidemiologic ev idence. be says; shows that "If dlnxln is a human carcinogen, it Is a rather week one la the population exposed to high doses. . . and Is not a cardnogen In the population exposed Jo lower doses." He adds that there are no convincing data that show that exposure to dtoxin causes birth de fects, chronic diseases of the liver or of the immune, cardiovascular, or neurologic systems. Skeptics have long noted the glaring in consistency of the risk assigned to ri!^lrin based on animal tests and a<-mal human experience. What has brought scientists such as Dr. Houk around Is the mounting empirical evidence and the growing sup port for a theory that plain* why hu mans are far i* to dioxin than are guinea pigs. The theory is that for diiin to have a toxic effect It must first bind to receptors; There appears to be a dose, which varies byspedes, behnrwhich the receptors don't function: therefore there Is no risk unless that dose is reached or No one knows Just what the level is fer hum but It is apparently far higher than maximum acceptable Intake level se the EPA of Q.006 trllUonths of a gram kilogram of body weight per day. Car. and same European countries set acc able levels 170 to 1,700 times that The EPA has yet to recognize that axln's dangers have been greatly exag, aied. but Dr. Houk predicts It will eve ally come around, it plans to begin sn tng the matter soon, but acceptance of Houk's analysis won't come easily, chael Gough of the Office of Technol' Assessment says that It the EPA backs on dioxin, it will open the door to dema forreassessment ofmany other chemic: "That." he says; "Is a door they will re! tantly open." In the meantime, the cleanup of Tin Beach proceeds. Dr. Houk says there little choice but to go ahead with It. '' cause we've got the public so riled ui The media that got people riled up w scare stories about dloxln-talnted Agt Orange. Times Beach and paper-mill eff ent have done little to "unrile" them. I Houk's turnabout was reported by the : Louis Post-Dispatch under a front-pa banner headline, but U got little attend' In the East. ABC News reported it: a and NBC did non The newspapers that i finence those In Washington who cou bring the costly Times Beach boanaogg to a screeching halt burled a small A story deep on their insiri page;. Afr. frante is chnirmaa of Accuracy 1 Media lue., a. inedia icalchdog group. GENP 0 1 1 5 5 ) 784366 / '> *. .j.. * ** */-*,.* ***'M* <*** .. ', ..-..s'-VI 'WvVl.S- vj /;;>;, .Vj,V. ,*.... * 2>< -'r* ;'ft *',f ' i.fri ' :.i''VsSiiV,,ii VS ~.'i,ii';-* `: * j-i, * * /s''V-'-' >{ . # * ,'*X** it-%PW** ' - ** #, ; i ^ ;'. ' i S .z*r? f>y. ` Sv 'v ' #y : : ;,? * . ' . -, ce j t o f + v p f5 u r \ ( l e J . 784367 Os Measurement of PCOF and PCOO in Utility Equiment Sydney H. Gordon, Michael Miller 1IT Research Institute, Chicago, Illinois* Fred L. DeRoos, Marcus Cooke Battel1e Memorial Institute, Columbus, Ohio Jacques Guertin, Gil Addis Electric Power Research Institute, Palo Alto, California Introduction There 1s Increasing interest in the potential for formation of polychlorinated dlbenzofurans (PCOF) and polychlorinated dibenzo-p-dioxins (PCDD) from uncon trolled fires ihvolving polychlorinated biphenyl (PCB)-containing dielectric fluids- At the temperatures that prevail in transformer fires, PCB may react tc form- PCOF and other toxic oxidation products. Information 1s sparse, however, on the-.distribution of PCDF and PCDD 1n the PC8-contain1ng Insulating fluids used by the electric utility Industry. Even le'.s is known about the effect of service time on the contaminant concentration 1n PCB-filled electrical equipment. There is. also the possibility that abnormal operation (arcing, overheating) may create conditions that lead to the formation of PCOF and PCOO. In order to address this issue, the Electric Power Research Institute (EPRI) recently Initiated a progress to evaluate and develop compound-specific analytical procedures for characterizing PCOF and PCDD in PCB-containing insulating fluids. Although the toxicological effects of PCDF and PCOO on humans is not well under stood, tests using laboratory animals suggest that these compounds are much aare toxic than PCS- Moreover, they generally occur at very low ambient concen trations, and toxicitles can vary over a range of five orders of magnitude depending on the specific compound (1.e., degree of chlorination and location of chlorine atoms). Consequently, in order to measure specific compounds at pg/g (parts-per-trillion) levels, the EPRI-sponsored research has three basic gos U ; Improve the chemical analytical methods for individual compounds of PCOF and PCDD 784368 I SSIIOdNHO  Determine the concentration of PCDF and PCOO in utility equipment that has seen varied use Develop a'screening test that determines PCDF and PCDD concentrations rapidly in order to reduce costs and time delays -before undertaking cleanup activities after PCB fires. To this endfl the program has been divided into the following tasks: e Round-robin method evaluation ' Analysis of spiked sample matrices and In-service dielectric : fluids Development of a rapid MS/MS screening technique. This paper summarizes the chemical analytical methods used and some of the early results of these studies. Results of the MS/MS investigation will be reported elsewhere. Analytical Approach The Isomer-specific analysis of trace concentrations of PCDF and PCDD in PCBcontalnlng fluids is a challenging problem. The analytical difficulty is often compounded t y the presence- of Interfering substances such as PCB or polychlorinated, diphenyl ethers. Combined capillary column gas chromatography/mass spectrometry (GC/MS) Is the method of choice for the measurement of PCDF and PCDD 1n trace amounts. The selectivity of high resolution GC together with the sensitivity of mass spectrometry yields detection limits of 0.5 to 2 ng/g (parts-per-bi111on) for these compounds. Since GC columns cannot handle PCB-contaminated fluids (sample matrices) directly, the samples must be extracted first with a suitable solvent to separate the analytes of Interest (1.e., PCB, PCDF, PCDD) from them. This 1s followed by an enrichment step to remove co-extracted Interferences and to concentrate the- PCDF and PCDD 1n the sample extract. Finally, GC/MS analysis of the sample extracts provides, the identification and quantitative measure of the PCDF and PCDD of Interest, based on the premise that GC separation of each chlorinated compound class can be achieved so that a unique set of mass spectral indicator ions can be monitored for each of the appropriate GC retention time windows. The use of high resolution GC/h1gh resolution MS (HRGC/HRMS) improves the isolation of PCDF and PCDD from.Impurities and reduces the need for extensive sample cleanup. 784369 2 GENP 011558 To determine the recovery efficiency and Improve quantitation appropriate *3C-labeled Isotopic analogs of the ad1ox1ns* and "furans" are added to the .. samples. (i3C-conta1n1ng compounds are not present 1n significant quantities in naturally occurring materials and are clearly distinguishable from native compounds by mass spectrometry. Spiking samples with labeled compounds thus provides unique Internal standards for accurate analysis.) Round-Robin Method Evaluation the round-robin method evaluation 1s designed to validate the analytical prccedure and to determine the reliability with which specific compounds can be Identified and quantified. It Includes the synthesis of native and 1sotop1ca11y labeled standards and the analysis of spiked sample matrices using the best available inhouse procedure. Five laboratories are involved 1n this effort: Battelle Memorial Institute, IIT Research Institute (IITRI), New York State Department of Health (NYSDH), Radian Corporation, and University of Umea. Radian synthesized the standards and prepared the spiked matrix samples. These have been analyzed by the five laboratories. The laboratories use a variety of extraction-cleanup procedures and analytical GC/MS techniques, the main elements of which are sumnarized in Table I. Five baseline samples (Aroclor 1016, Aroclor 1242, Aroclor 1260, tri- and tetrachlorobenzenes, an-* aged mineral oil) were-spiked with .he isotopically labeled internal standards *3C-Z,3,7,8-TCDF, *3C-2,3,7,S-TCD0, *3C-l,2,3;7,8-PnC0F, and ^C-OCDD at concentrations of 100 ng/g each. In addition, the three Aroclors and the tri- and tetrachlorobenzene- samples were spiked with the- native isomers 1,2,3,4,7,8-HxCOF and 1,2,3,4,6,7,8-HpCDF at concentrations of 1GQ ng/g each. The. aged mineral oil sample contained severs:! additional pnlabeled isomers as well as 510 ug/g Aroclor 1260 and 530 ug/g trfand tetrachlorobenzenes. A suninary of the measured average total congener cle s s concentrations obtained by the reporting laboratories is presented in Table 2. In general, agreement between the values obtained by the various groups was found to: be within a factor of two. In the case of the tri- and tetrachlorobenzene sample, fairly good agreement was also obtained between the expected and measured amounts for the isomers spiked Into the matrix, and virtually no additional PCDF and/or PCOO was detected. By contrast, the Aroclor 1260 sample contained significant amounts of a large number of PCOF compounds. Lower, but still significant, amounts were measured for the remaining Aroclors, while the aged mineral oil sample had concentrations close to that of the spiked values. GENP 011559 Analysis of In-Service Dielectric Fluids A preliminary round-robin study was undertaken by three of the participating laboratories using four samples of dielectric fluids taken from In-service transformer and capacitor units. The average total congener class concentrations found for PCOF are shown 1n Table 3. Owing to differences 1n analytical procedures and variabilities In sample matrices, estimates of detection limit have not been Included. Sample ISL10 was a mineral oil contaminated with 100 pg/g PCB from a transformer that had failed 1n service by arcing. Sample ISL2A was an Askarel containing Aroclor 1242, which was removed from a capacitor that had bulged, but not ruptured, 1n service. Both samples had PCDF at concentration less than or equal to the limits of detection. Samples ISL3A and ISL4A both contained 70% Aroclor 1260 and 3Q% trlchlorobenzene. Sample ISL3A was taken from an Askarel transformer after 20 years of service and contained 0.5 to 2.0 ug/g of PCOF; Sample ISL4A was taken from a transformer of a different manufacturer after 31 years of service. It had substantially higher concentrations of these partial oxidation products for all congener classes except total TCDF. Work to date has developed Improved measurement methods for compound specific PCDF and PCD0 1n PCB-contam1nated Insulating fluids. Thus far, the results do not suggest a preferred analytical technique. Analysis of the four in-service fluids Indicate measurable quantities of specific PCDF compounds in 20- to 30-year old Insulating fluids, however, there is Insufficient Information for generalization to all utility equipment. Future work will focus on replicate analyses of another three baseline samples as well as 10 in-service samples. 4 784371 t a V K T 'T iO Table 1. S u m a r y of Analytical Protocols for PCDF and PCOO Mcasurteestv Laboratory Cleanup Solvent Extraction Column Chromatography Analysis Capillary Gas Chromatography* Hass Spectroststry** Battelle IITRI NYSDH Radian Umea None S102/A1203Q/A120 ^ > OB-5, CPS11-88 HRMS, El None None c h 3c n / hexane Gel perm/Al^j/ A123 SP-2330/CPS11-88 A123/C/A1203 SP-2330 S102/A1203/A1203 DB-5, SP-2340 LRMS, El HRMS, El LRMS, El None C-flber/rev elut/ Fiorisi! SP-2330 LRMS, NCI +Fused-s1l1ce capillary columns **LRMS low- resolution MS; HRMS * high resolution MS; El - electron impact; NCI negative ion chemical Ionization (methane). (All use selected ion monitoring mode MS). Table 2. Average Total Congener Class Concentrations of PCOF and PCDC 1n Spiked Baseline Sampies Sample, ug/g Congener Class Aroclor 1016 Aroclor 1242 Aroclor 1260 T H - and tetra- Aged chlorobenzenes Mineral Oil iTCDIJ iTCDF zPnCOF rHxCOF iHpCOF OCDF NO 0.01+0.01 0.010.02 0.18+0.12 0.08+0.01 0.OltO.01 NO 0.94+0.78 0.33+0.24 o.ioa.o3 0.12+0.06 0.63+1.12 NO 0.550.39 1.15+0.68 2.7342.08 1.83+0.71 2.81+1.22 NO NO NO ' 0.34+0.44 0.11+0.04 NO 0.03+0.02 0.10+0.02 0.21+0.04 0.44+0.44 0.07+0.05 0 .2 2 + 0 .IS t in IMHO NO > not detected 5 784372 <_ C Table 3. Average Total Congener Class Concentrations of PCDF In Four In-Service Dielectric Fluids Congener Class In-service Sample, ug/g ISL4A ISL3A ISL2A ISLI0 xTCOF iPnCDF rHxCDF EHpCDF OCOF Total PCOF 0.1 2.6 7.6 16.3 9.8 36.4 0.5 1.6 1.6 2.0 1.0 6.7 oo-. NO . 0.01 NO NO 0.02 ND NO ' ` NO NO NO NO NO not detected 95 U O dH 3> 784373 1 * l 4 ***,'>' * * * * **f ** *\ # t *.'* ' j > f Sponsored by i ELECTRIC P O W E R R E S E A R C H INSTITUTE October 22-25, 1985 Seattle Sheraton Hotel Seattle Washington 784374 / GENP 011563 <;ure 8,42 - Thermogram clearly pinpoints wiring with bad breaker/wire conactions. (Courtesy of Thermotest Infra-Red Surveys}. (jure 8.43 - Loose connection in high-voltage switchyard was detected from . round level. (Courtesy of Plant Engineering'# and Corporation). 'jure 8.44 - Blocked transformer cooling radiator was detected because of tr.h of heating in the radiator. In another sim ilar case history IR therrpograms '.ibfed a utility to prpve certain radiators on a new transformer were either ncked or improperly desinnad^CntuLlesv-pJ^lant Engineering ). m !0 TMI i GENP 011564 What To Do About Askarei (PCB) Transformers (Past, Present, and Future) Part 1-- S tra ig h t T a lk On A ska re ls lO hco ''fr mi Introduction airlFnoeslauss"udiiadslastscakteincanntogrtleneiqtire,a.u"leiinhdcia.ntrvgTiechapublosee,leyqntcruahuinplsosmerfidoneranfmottereidnnrsebaacipproplhyuleilcd5na0ybtlyieo(ePunassCresBrdef)oq,inrukdnhiroeoinoawgrlsntarcgahleoninsgseefherltiryocaaltinhlrldeey blPaPmdrfioeqasCreacgouskBriuetuanisdessltrtace,ie--astnttlainio-omoanfedinnndlxalpsiecnciAeondeegargtt,shepreeeientshqniarotoacunnncnfyiaopcahdnt(meleldlEityvnmhePogoensiAslc,uoftst).aaneslebitsntasaloeertgrrraytavueh,baigncelnlebeitesdss,u-lh.aateartTgeedmnhmesdidaris1usen9tsgdsda7ttauir0ansam*lttpasnopeottsrioysotrfsheoa,nteocrlasamxebngU(idediFdlo.ieriaStgvsytt.e,huiohrerEonneairnnuiesnvrff9grcoi.rto1arhoPme)upin.sCsmraeBiTamndechcuanieaudtssthsrnaseeyee,.,l pInroaddudcittiioonn, othfenEePwAabsaknanreeld-inthseulamteadnutfraacntsufroermofetrhsehflausidc,eaansdedth. us the Hdthuoaswnterivnaeol rra,mncdaulrurreteinlmittyoUvlo.aScl.aogtfioovnPesCrnB("msraefinlrrotomaredgsuterlraavntiicsoefnosirnmdoteorntsao"l--tlyroeenqnlyucilro1es0ee00daruliinenr itthsa--t arme arenqyuiroefd ttohebe reesttriomfiallteeddto 1a4lt0e,r0n0a0tiveinf-lsueidrvsi.c)e' It iassaknalirceilp-failtleedd tthraensEfPoArmteorsbeanPdCB35-cmoniltlaiomninoailt-efidlle) dmsaeyrvbiecea-aroguenddufnoirtsm(aansysuymeeadrs btoy come. 'Pending legislation may ban continued use ol PCBs in the food and teed industfies. TMI 691 PCB History Invention and D evelopm ent (Prim ary U ses) AtfCSlphiortasoee.rknmn-Luayroesprineieusanliinpssaw1y,otsa9.awMan3Tse1ithtsrrf.esiadrtTorsnicaehutsonlerefumiso.csffrTpieotmrrhdsaricmeteniryntelfralirauwnstpenisados(usDwFfloafilregao.rtrtimurFnercrgreceaarabnfps9lpoku.a4uawicMd)cge.i.tihoPtrfCooetrCrsrlosapButrh(uktbFwetbiypgaiGynasut-hteMriesneenenovsertneSevr9asnwdic.alt2aeeinEtndsatlionennaincdCpCt11rph99hi9cel.3ie23cm3Cm9)a.ioaatcicisnomaadnla,l A(Tsabsluem9m.1a):rized by Underwriters Laboratories (UL) Incorporated Transformer askarels are considered non-flammable at or odufninlciakoreymlybt.eumTshtpiebelrefairtoeurrheeasxz.paUlrodnsdiivseervemprriyaxctsutmirceaaslll.icsornedgiatirodnesd, efoxrtmreamtieolny As noted in Table 9.2, certain applications such as the pulp and paper industry have particularly relied upon askarel-filled equipment. 692 TMI TABLE 9.1 UNDEFRLWAMRIMTEARB'ISLILTAYBROARTAINTGORY (UL)* Water Transformer Askarel Silicone Transformer Oil Ether 0 2-3 4-5 10-20 100 784376 TABLE 9.2 ASKAREL TRANESSTFIOMRAMTEEDR POPULATION O M oUI1O----\i11 Classification PCB Units Per Total Units APPMuolwpUenarintUsdnPitsaper Industry 11 ooff 23560(1(14040,0,00000oof f53,050,000,000,00)00) 2 of 5* 'Non-random sampling of ten clients *See Chapter 2, Part 1, Table 2.1. TMI 693 Figure 9.2 - A typ ica l p ow er capacitor bank. (C o urte sy o f M cGrawEdison Company). Figure 9.3 - A typical capacitor as used in electronic circuits. MA TMI Figure 9.4 - A typical askarel transformer bank located inside a factory. Secondary Uses Ui(nlinsititiqeadul epinfuirroeprodreseesri,sotfaennptcreocdmaoppmaabisnislaiitnniegtsufstoheres)P:fCoBllouwsiengextdeinvdeersdefaarpbpelyicoantdionit'ss TCPCrlaaaaprsnbatsioccfnoiiztlroeemrsrssse, rciinnlaicqrslbuuyoidndntisnhpgeathfpliuecor rr(eeNssCicneRsn)atnlidghint bruabllbasetrss HLuybdrriacualnicts,fluciudtsting oils HPaeiantt-tpraignmsfeenrtsliquids Sealants APdrinhetinsigveisnks 784377 FClaouolrkiwngaxcesompounds FMDreoydiunungstitnipngagnmsage(dehneiuatsmt tfroarnsmfeicr)roscopic slides Q ffl o < 3\ Tbaebtwleee9n.31s9h31owasndtha1t97o7v.eVr iratubaillllyionallpPoCunBdssinofePxiCsBtenwceeretopdraoyduwceerde TMI 695 penrotedruecdedthesyennthveirtoicnamlleyn. tA(bToaubtle409.3p)e. rcent of this synthetic fluid has TABLE 9.3 PCB HISTORY IN THE UNITED STATES (396 TMI H o w lo Identify P C B s TgSm"oavhitvoaserektnoressaluysergtaehwbhsley"ho. bouMmylteooasinrtaespslearcihdnvoiatsmotttoe,omurslyaone,bndceoPerlnrCetdrtgBhaipedshateetarrmrtasieedadsberkewtnrefaahondmor ermewemsaoAaanurrloulekdcfsatlodocueerts^usneri.dgdeI-ndtuianwstieenathrssttehh.theTeUeagnUbSecl.nneoieat9emrni.dcd4 MU.oSs.tAaruotchloorrsit.ieIsn satnayteetvheantt,sodmeseigEnuartoiopnesanarPeCmBosstarliekemlyortoe btoexuicsetdhatno identify the coolant in a high fire point transformer. The designations "askarel" and "PCB" are rarely found on a unit's nameplate. PCB C haracteristics Molecular Structure w9Ticta(tnhy1.rhni5pee2tdeeh)ae).lc,et.P6teadTcwO0CdrewtborBtssibhoocitogyenamarnhrnltrohamudigtnelrtnereoagfadcrornstdeeruhr.seeplefehdHtslohaynorcecadamuwoernrmweneoemdsvegrbueseiiesensg--nrnteir,htsndoo1toaeof2nftpifco4nlcceya2mthhwrc,rctllsbhaooo1eorprr2rneeipin5annteph4ceeaole,tia-ttyfaosnaocptucmynreoeslhbddms.smlsobi1sotnroi2ytfiaunlt6ePhctte0ceeCht.hu*dlrBmoeelTersobahvisnplraeiepeejeorchifcnioaiunretoutsileonvesrtdemymatcllwysnstaoodolrig(lcnnyF4daeei2uignetrgh.sasu5ieetcbro4srdehe,f,, Chemical Composition 784378 CtPNuCoerevmBsesmrothof eanclrhleeylsosur,ismnetaridxateentddsrafbonirtspomfhoeerrgmniyavelseraknaadrpsekaclasrhtrileacolrurseilnaagarreteendveeissrbasceleolnynsztieitinayntleelyr(ccTahhaaabbrnlaleegcne9tdea.5--rbi)sl.emtTi.chixse. "IngeTtatebrle" f9o.5r rfreefeerehnycderoigsemnacdhelotoridaeddgilaivses(sh)o.uTldheasncyabveenpgreersaecntts aass aa result of spurious arcs. An earlier scavenger, tin tetraphenyl (1944), had a tendency to precipitate out on top of askarel as a milky-white crystalline material. *ASTM D-2283 lists seven types of askarel that have been used in transformers over the years by various manufacturers. TM' GENP 011567 697 TABLE 9.4 ASKARELS ICNOUN.TSA.AIN.ING PCBs Trademark Manufacturer ACDDAAPSNEHInysslariyhoyeeaobkrkfvl-mrac-ocaFeaotTnelrlrlselnooa-eeeotxKorromlxnllu!loohlll AWmesetrSiiKncaGWCMguanAhehoncagnlolrGgmalnCeiunsmrearSsoaeaMlnoeeCrlwrlpeAohOEEEEEoenEaNllllulreeeesreldambooacccccitstvinitttettelororrrrioteriiiiinncoccccxs1r t'rGanesnfeorrimc enra^m; me auysecdontfaoirn inPsCuBlast.i'ng liquids in capacitors and phwwa*1Ayr9iiotldc7hlgp7crcro;eloooecArrunstdayNpssrineibSusgrotnIoapndttCmosre5erAso7udyua.sNo1nrne2mcSttd.hsiQInie/naIgo0tEasi;fcnEcAce,toEollSyen/mTfAcdoirMbtSifertuiinTcosDoManrtn-ielsb2,csi8lns"eoias6sumtg4suca)kbhaln.aausttt,ireshnetsagiibc.t""ihl(slaeiIlqnEo"huyarEyiindEggdasaerSo.tsnetgeeTadserhninpcedaryactorerhohddrlmamou5vrcai9edfeto1iedca-r OVERSEAS Trademark Manufacturer FSKCKPPDP?eahyyaelKonrnernnapo;cnlenolchleoceotelnhcohcrneellrhoorlrromr ? ChemKaMknooMMgSPPnB(eoiirrCttsoogassvazddyauuoCCneeeefbblutcaarlloiiee(csshffU(ffcchhhoaaG(iEiis.rr((eSoolnF((F(orJJJ.grrmvSaa((aaalIIaapppnnatt.aakRnccaanailldaeennyyny.))))))))))) cm TMl Cl Cl Cl TABLE 9.5 -T--Y--P--I-C--A--T-L-H--TE--R--CA--NO--S-M-F-P-O--O-R-S-M-I-T-E-I-MOR--eN-A-t-h-SO-o-K-Fd-A--AR--S-E-T-L-M-S--'-D.----2-2--0--3--- OlO/NOi Trade Names Type D Type E Type G Ine7r0te-3e0n In1e0rt0e-e4n2 AP1y3raBn3oBl^3 bInygWreediigehntt)s (% - -- CCABrihhpollhoocelrroiinnnryaelteb(1d5y245W%4,eight) 70 - 60 ACrholocrlionra1?t'e1d242, BCihplhoerinnyel b(4y2W%eight) - 100 - Trichlorobenzene 30 - 40 POhxeidneoxSycparvoepnegneer SDcieapvoexnigdeer 0.18 to 0.22 0.18 to 0.22 -- 00..113155 to 'Monsanto Chemical Co. TMI 699 Tmhoisistculroeudiny thceonedairtiloynasskhaoruellds. be not interpreted as the presence of sMcoanvseanngteorshwashennotvaforiuonuds eavsikdaerneclseaoref rmeaixcetdioninbaentwyepernopthoertidoinff.erent Electrical Capability aAMh(bausVeskthAyaboo)re.enreildtM,nieto6hsli9smotrueKiPtgcVeCoh,dgB1oinn5nuizeuMnesioVdtasfAg,ttheh)t.e.heIIetnpbrherefaaasfcostctirtnine,ces,davulufelaaarrlistlnbikngieangertenthfhtleuuedisdeee7assd5irgd0loyn-en7v1l5eil9am0l5or0Ggipte'Kaes,sdtVrt3aArb4any.nrs5asfsocnKfgoirVemernm.(ec7ere.5sr, Watyshk(eaFrriegealusirsemli9mi.n6iet)e.radl ionilahpapslicaantieoxncbeellecnaut sime pouf listselisgthretnnigntgh scthriakreaccatepraisbtiicli, cOenntthoefoothilebryhavnodlu,moiel minusatsnkeavreelrbbeegminixsetdowloitwhearskitasrefilr.e"Orevseirsttawnocpee."r Relationship to D D T sckhIDDmninuomDDaomustTTiwortluesamn(l)rriaD,nieottaiiiPcevvcosCehesuforlBtmotee(hatrxehoebtipaosdyesetxti,piarciticisturoohh,ladyneseituttniarsoobbyanfnsliesgatystrtfalgifeycenieastmhcicscrtlasettoah)iyrdsl.o(opiiusfhenfhlgaetsoyhvrhtwseraetuinncamtcbaeoceele)ceebntuihnseftfmaescttacutorusptluxadcaciatcririipeiotabspbonle;afedPiwrcntCehahtneBtoueitssrpsseiePnaouaCseiasrttBseimDimsssoe.DapfbmtWaTltiieobvchirailtsneicoolgdea-ff Figure 9.6 - Comparing lightning strike capability of insulating media. 700 TMI The PCB Dilemma 784380 U.S. Hazardous Classification Only after many years of use did the PCB ecological problem become a1p97p6a.reOnbt.vTioaubslely9, .s6accornifticraiasltstriandsuel-aotfifnsghfaluvied tcarkiteenriaplbaectew.een 1968 and citwPcrcueloCraingtonBsyauntstalageaimfnhlytlyyiadoi,nsPcnaaCahssnblo;Blteewoistdene(honaevPanleoCsotrntfhB,rhgeetehbcaeweep"mcidivlmtaaoehenusoadsetsddetttilhoypeaeoxupbafimbnruotli)nihoitscsdlsoaiztasnaepnb"do;truentpnhseiosdetelyaoblmunfehtcacatheanvine,eutsstethewUivneeoo.Sntfmh.migtesaeooswannssstevosidrmraulodasmenn.dfemagUidareenartnfohsootauxatrossl dsW"lr(ePeOtiaafhcrfSsnieaeotHcrtivrefeAieiotnssy)ut"mn)rchP;oula(aot"alsCtslnilsspuldauaeltetsasffstnilhoeccteAfali"ae)hs";snOtaasttzhlconiyfaeindccrkeuda"DnpCotmioeauolawptsnoisaonsrssern.utiamBsbglsetrPShtonaoaautnitfpsoceoitfetnhnysTgesr.a("aE",nOXnPads"nnA,pdeH"oaAe"rslOot"aua,nlbRtt"ehiBsoMtn"haA-,anAd("scDCmaeaOn"i.tdnmT"lieD)s-aEatyh"rs,aaStaatsuilsnosbaidonxt' sihE"hmsppaaPooizznAlrliaae,osrr)lwoddi1san0hootns"euu.pdrssPFeo""soCuCusrnlBlibuascdsbtsosbetssmadotonatAphrcninapecmar,oeistoslihisorasaeeontnse)"n'dC,esd"asahuvaantanasedksdrg"eioeC,aoyursterahstytthe"eaetAcgo"w"lDoap.bhroOsryeaisroTszXiprfatici"crtrpoyra(divolteteopeisogrunsoinaasilbslla.ulssseFtuapau"csbintrlaolst,tahst"xoaneeinbncr"emcu"w"eetopoxtonrtou(reoaulepd,tnmulPeadbbCgeseloialBcyaarl health. DQT lists PCB as an ORM-E Substance (other regulated material), a mssi("dpRihdaeoiQpetnley)rptsrcieiifsaihdcPll,eaoCtqhtrBtiiuoahnaentaalitssnetcoudnoamonobhttbraiaipegniznrhcraeee(lruUranddtNyoemelursd2us,t3"shi1nstiat5usnab)bnDeas1yntO0admonTptatchhocreeeuklarenwlesddhhstastieez.fwinarcWisrtadhihRttsiecoQanlnrae(srsl(paseOubo.pcRreDohtlMarOtbaad-TElbqee)olsu,enqiagiqtlunsynuataDicanttyoinOtninttigTyis tt(yyF)ef4do9ellorCawFl sRRet1hg7ei2s.tE3e1Pr6,AV('sao)ld(.7e4)f,4in,1i7tNi2oo.n3. 214o6f9(b,a)A.rTeuhpgeours1tat0b2pl9eo,us1np9d7ill9re,npepo.arr5ta0ab7wl7e7aq)t.eurawnatyi TMI 701 i a b l e y.b CHANGING TRANSFORMER FLUID PRIORITIES' 1968 DIELECTRIC 1c DIELECTRIC TOXICITY TOXICITY EXPLOSION ]c EXPLOSION Courtesy o t Dow C orning Corporation. mREneecvnietr)notninmscteuludndtieeasdl PbthryoettsehecetriNeomnataiAorkgnsea:nl cRyes(Oeaffricche CofouRnecsiel acrocnhdauncdtedDefvoerlothpe hpSbpBtctircuuoloraaoiimtmstmnpseetsaaeenhadfntdoetreiirasrldaioymselnnkhe(efdePralrowxeCsaezmpaoc)nnB,oruttvd)irspclitidruhucriotpooersanebatelehmsettecuheecdatamseinovdntqatiuetnau(ibnlcgPldicepiCehpeombdxeaBnapeaatet)elitolnincatmdshtguauissrmotireneeer(ifd(cgutlotahpoahlctlceopayttcqhtaehutcnueecnepooericvatnatnroihstsrttrhiroee,soaoamnxnnlvtlmcaehdiecahledeanav.nssdlnauitewItgr.IguxnoeItpaginshsrueteesehilscodsaeuaattnleesecnbns)ddddhe., atiocna.se, PCBs might have "passed" the toxic substance evalua Cost-Benefit Evaluation wAidthilesmomciaal bceocsotms e(sTaebslpeesc9ia.7llyaanpdpa9r.e8n).t Awsheonnbeeenxeafimtspalereocfomthpeareend- 702 TMI vpeivareopnnemrsocouonrntcateal sidniiolnefgmsomcmaare,bofconoolnedstaspmaNcinCkaaRtgioicnnog.pymaptaepreiarlsanmdadperinfrtionmg rinekcsy,calerde . Effect on the Ecological Food Chain 784381 chTusccihihshemmeaeanpfsitinusnrelae.la--lftsmfioTtotehr(hnpFamicsiltmgahainsuaane,trryyceaaoosciinnrlt9otcdeh.agr7rueenii)scsy.aitmtaisrmcliIeaano.ollp)vf,prfeanoutccoourptenp,osf-itbsssPhheieoC,spdBfrhboeosaiogrcsddremasmec,dadaheyasdaaneidnbbftriihtolosioetmomywmtaaohhrtgnahdantesrtihhlmymarpofsaulaiunnmlng,inakththdthtoeheetnoierPcffmCoo(otooBhanddnes PCBs were first discovered in the Mesh of Baltic fish in 1966. In humans gtinhreaJmaappsap(ne2a0irn0a0n1cp9e6p8mofa) fsPtyeCmr Bipnlaiondnvrsiecoretfeoaniltllec(gforeondmsPulmCeBapktipionongisohoefnaaintpgtprarwonaxssifmeorabptseipelyirnv0ge.d)5. doSm(6ereo0nvirm0teerwmaFtoal)ex--sadiucinpatohtothetolayr"pcnnithhioePlyrosCsmrinBicnaosaal!w"l*tePebTdfCefhelPiicdsetivasbbet.yeat-nhplzla,rootbfdtuuhurteaconrtnesiean, l ebcwfyufhe-lippccrrthoitcdiamnuucatshtyeisodbf"cYeheuer5atsa0thei0nod"atiPicmnCucetBeis TABLE 9.7 COSTB-EBNEENFEIFTIST OEFVAPLCUBAsTION GENP 011570 TTRrreaadnnusscffootirrommn eeIrrn:SRaefleiatyb ility Property loss pRRreeeddcuuipccitttiiaootnnorisninucsoaeLsMEirdtomesoptdpsofoilcocolcalyofuolmntlniicoftteerornonoslttllsi(nlpoPgasCrsBtaiciruuplasoetledlumtiionantteelre)ctrostatic s* t7a7nj /tsiabtey-pthreo duct ma closing y o also f all resu PCB lt from incine insuffic ration s ient ites incin as of er A atio pril n and ma 18, 1978. y s u b TMI 703 ML)LU. J.O (ENVCIORSOSOTNC-MBIAEELNNCETOFAILSTTDESAVOMAFLAUGPACETBCIsOONSTS) Direct: Indirect: CRCNeooocmnnrt-meareamectriiyonccnaialatielnldgfiinsfohdofeuorpsdietarspyer Worker lost time & medical (?) RLbHoeeusicmnsrgeaoan(ft?ibo)loinrsdassl fishing & hunting of physical & mental well Figure 9.7 - Faieol PCBs in the environment. (K. Higuchi, PCB Poisoning and Pollution. Academic Press. 1976}. In the past, the EPA has stated that "...it is probably inappropriate to tuhsee ttohxeicdahtaazaIrrodmpothseedYubsyhPoCeBpsisotodehutommanask.e'**qIut awnotiutilvdebeestliimkeatceosmof pnaornin-Ug.Sa.ppPlCesBwmitahyobraenmgeosr!e K"teoexpicin" mthiannd Aamlsoertihceanretaylppeoss.*sibility that *Federal Register, Voi 42. No. 22 (Feb. 2. 1977), p. 6537. 7 0 4 TMI qthueirelodngbetelormre eafnfeycdteoffinPitCeBcoonnctlhuesiofonosdccahnabine mmaadyen.oItt ibsefakinr otowsnayfotrh2a0t wtoo3u0ldyebaerst.oThmeinreimfoirzee, dPuCrBingexthpiosstuimreetoof uthneceerntaviirnotyn,mtheentb.est practice ASSESSMENT OF PGB HAZARD t v r n f*\ AanndanTaSlCysAis* opfroPpCoBseddataguciodmelpinileesd, floelalodws intog tthhee ccroitnecrilausoifoFnIFthRaAt aPpCpBesararpeaprteicrusilsatrelyntt,oaxnicdfoarreshliokretl-ytetromaecxcpuomsuulraet,e.buPtCrBessudltos naoret subject to interpretation... TtFhhIeFeHecAonnvTicerolsuntsimniogennRrteeasbucylhtsePdCrBesgabrdaisnegdthoenpoFtIeFnRtAial rheaqzuairrdempoesnetsd tios qitnhuveeesFrttIieoFbnRraAabteglesu^,idbAeulctinutehtese.steoxsicpietycieissoabresenrvoet dreoqnuliyrefdortessotmaeniamqaulsatiicn `FfFRA - Federal Insecticide, Fungicide and Rodenticide Act. (1972) TSCA - Toxic Substances Control Act. (1976) , t - TMl emphasis 784382 y | Precaution in Handling PCB in the Field sBPehneeovcrsuiaroloudnnsnebmeeelPanCwrto.BournkreidnmgfoariwnmisthaannPyCeyBxecaserhslloetounltdcodtmiaekeleec(optrrriceincfarleuuittdrioo,nfislsluectdoh PpeCrqoButeipucmntitetshn)e.t tLhiemistuellxupriocsuarceidtoinPaCcBa!rNboatctehreym, iifcanloitshsaanfdeleadlllphreoptiemrely;, fwoirllebxuarmnpthlee, body and if dumped into a waterway, it will kill fish. arHinnudsmerasenpuPaseiCrdiBngienxtrpPaoCnsBsuforreremtnreoorrfsmil.f aT(lmhlyuacsno,ynwsohifsettnsheowsfoesrakamirnepglainwlsgiothflouPnidCEsBPaAonrdostreoslDvtieOnngTt hazardous material lists) please note: TMl 705  A v ^^d ire c t skin contact >- A ^ ^ H lire c t eye contact by use of goggles Av^w breathing of vapors for extended periods of time Avoid vapors from a severely arced askarel transformer Seeking Our Way Out of the PCB Woods oCuotnofrfotnhteedPwCBithwthoiosddsi"le(mFmigau,reAm9.e8r).ica began its efforts "to seek a way Wplahceen. tIhne1h9a72rm: ful effects of PCBs became known, five events took 1. Sfotoadndcaordnstamweinreatisoent.by U S. Food and Drug Administration for 2. Standards were proposed by the U.S. EPA for water contamina tion. 3. Mctoaotpanalslcyaitnoetronsc.vloosluendtasryilsytewmitsh--drtehwat PiCs,Bsefraolemd atlrlamnsafrokremtseresxcaenpdt In 1977: 4. Westinghouse and General Electric ceased production of PCB transformers. 1 5. aMftoenrsaacnctoepvtaoblulentsaurbilystitduitsecsonhtaindubeedenmdaenvueflaocpteudre. of askarel, TcctcqmSrhaaoeaueeppncnaetaatsltaiifoctcfyiortnniihattrooemnf4orrs5sPerri0etCrdiits-qonhB2ienduat3shlitoe.raoetoccotfmrWotwtrhnieuucieettsnsahreeit1insnl9igmwqs8omou1iaftnmihondNaoorueseanhuf,at-itigtPcorhalhvCtnoauraaneoBrfueli,krlEtdethshlsreibpppeguceloeehantfirccodnigceifrteaiaearrltel-lrltaoiSahCnpnpneeopscsodmtlftihioenoicoiefdtra(nmN-ctt1irooeEm49anor6C7nidl0aa0s)--s--n`fe21sotxd--9,ir(ncme7aavaef0)nieanpnsdordrte.iifaodtathhbntAnhaasildneyl,ttl Nme1q9Ea7uCy0iposwm.r iSemllnuatccyohisnnetmoaqtiauncnipouPmnfCateacBnitnsut rmPiefCdatBhnteuosy;dfaahcwyot.euwrreeevdemri,na>nntuohfeaPclCatuBttre-eirndshubalelalftooerfdethetehleec1t9mr7ici0da's-l Even with these several steps, voluntary regulation proved ineffective. Therefore, U.S. governmental regulations have been published. 706 TMI TMl II -^1 00 00 CO 00 707 GENP 011572 Jr i idiibioim ei maintenance By S. D. Myers J. J. Kelly R. H. Parrish TMI' TransDfoivrimsioenrA,kMSroa.innD,t.eOMnhayinoecrse, IInnsct.itute, Acts 4:12 GENP 011573 'rspective on CHLORINATED DIBENZODIOXINS And D1ENZOFURANS Chlorinated dibenzodioxins or dibenzofur,:is have been found as contaminants in uigicides (pentaciilorophenol), herbicides 2.4.5-T), polychlorinated biphenyls (PCBs). aid products from processes in which polyhiorinated phenois are used (animal fats :id tallows). Chloracne in chemical workers dso seemed due. to some degree, to chemicai processes in which trace chiorinated dioxin .evels were present. Marked variation in tox icity seemed to be associated with the numei and location of chlorines, with the 2.3. -tetrachlorodioxin being one of the most ic chemicals known. Much of the data about the^e compounds is fragmentary, in obscure sources, or un published. A conference sponsored by the XIEHS was organized in order to provide a forum where the current knowledge of these compounds could be presented and discussed by the government, industry and university community. It hopefully also served as an in dicator of where to focus future efforts. The conference was held April 2-3, 1373 at the Governors Inn. Research Triangle Park. North Carolina. Presentations included meth ods of detection, chemicai processes involved in formation, bioaccumulation, and degrada tion. toxicity, pharmacologic effects, as well as distribution and fate in man's environ ment. The papers presented at that meeting ap pear substantially as presented. The paper by Piper. Rose, and Gennng as well as the Bibliography prepared by Hurf were not part of the Conference but are included for com pleteness of the subject. Your comments on the issue are certainly welcome. J ohn A. Moore Conference Chairman G E N P 0 1 1574 September 1973 784385 Ethlogy of Chick Edema Disease ^ jtfid Firestone* fhick edema disease first came to the -cion of the Food and Drug1Administra> December, 1957, when it was learned allions of broilers died in the eastern adwestern parts of the United States. r:il groups in industry and government :a-:dy determined that the disease was due .1 toxic components in certain feed fats and that toxicity was associated with ansaponifiable portion of the fat. haracteristic symptoms included the pree of excessive fluid in the heart sac and :e abdominal cavity of chicks fed toxic -4, 5). These and other symptoms such / .subcutaneous edema and liver necrosis v"f.-re accompanied by high mortality begin ning approximately in the third week. Allen (6. 7) suggested several years later :hat the accumulation of large quantities of i'\*t:ravascular fluid in chickens might be due altered permeability of the vascular bed well as cardiac decompensation and liver jrosis. Developments Prior to Chemical Identification of the Toxic Factors Initial outbreaks of the disease in 1957 oc curred at a time of increased demands by feed manufacturers for low-cost fats to aise the caloric level of diets for food anmals. Investigations by the Food and Drug Administration in 1958 soon demonstrated that chick edema disease was caused by toxic material in by-product fatty acids ob Division of Chemistry and Physics, Office of Sci ence, Bureau of Foods, Food and Drug Administra tion, Department of Health, Education and Wel fare. Washington, D.C. 20204. tained from production of oleic and stearic acids added to certain lots of feed-grade fats. Further investigation showed that toxic ma terial was also present in various distillates and still residues obtained from several fatty acid producers who prepared commer cial oleic and stearic acids from inedible tal low (5). The most toxic samples were batch still distillates obtained after repeated dis tillation of tallow fatty acids. The general scheme used for production of commercial fatty acids is shown in Figure 1. Fatty acids TALLOW-------------FATTY ACIDS 1st RESIDUE 1st DISTILLATE 2nd fi ISTILLATE Ind RESIDUE BATCH STLL RESIDUE BATCH STILL____ ^FEED DISTILLATE FATS F ig u r e 1. Scheme for commercial production of fatty acids and by-product (batch still) distillate from tallow. obtained from hydrolysis of tallow were dis tilled in continuous stills to produce first and second (continuous still) distillates which were subjected to further processing to pro duce oleic and stearic acids. The second (continuous still) residue was then distilled in a batch still to yield by-product (batch still distillate) fatty acids which were sold as feed fats. Occasional vegetable oil fatty acid samples were also found to exhibit some chick edema activity (5), but it could not be determined whether these samples ! ptem ber 1973 59 784386 i ti -a.: I i 'Si: O were inherently toxic or were made toxic by cross contamination in the plant with toxic tallows or toxic tallow acids. __ Now that it was demonstrated that chick edema disease was caused by toxic sub stances present in the feed, the possibility existed that edible chicken flesh might con tain these toxic materials. The unsaponifia- bles isolated from the carcasses of chickens fed toxic fat were fed to chicks at various levels (0.025-1.0%) in a test ration (2). Symptoms of edema disease were observed at all levels fed. The unsaponifiables from normal control chickens, fed to chicks at 0.5% of the diet produced no abnormal symptoms in test checks. Similarly, the presence of toxic factor was demonstrated in the meat of hogs that had been fed toxic fat (0). The presence of toxic factor in commer cial still distillates and residues prompted the examination of oleic acids and stearic acids collected in 1959 from food manufac turing plants. Oleic acids from several plants showed varying degrees of chick edema toxicity (8). Tn addition, oleic acid deriva tives such as triolein and glyceryl monoole- ate*were found to be toxic. Ames et al. (10) also found chick edema factor in a number of oleic acids and glyceryl monooleates. No toxicity was found in commercial stearic acids. At this stage, the identity of the toxic factors was unknown, but evidence clearly indicated that they were chlorinated aroma tic compounds (11). Since chick edema factor was found in food grade oleic acid and derivatives, the Food and Drug Administration issued a Food Additive Regulation in 1960 for fatty acids (12) specifying that they be free of chick edema factor in accordance with a 3- week chick feeding bioassay. The need for a rapid screening test was met by develop ment of a microcoulometric gas chromato graphic method (13) involving cleanup of isolated unsaponifiables by adsorption chro matography on activated alumina prior to gas chromatographic analysis. Portions of a reference toxic fat .obtained in 1958 were made available as positive standards. The availability of methodology as_in bringing the chick'edema disease control after another outbreak in the So east in 1960. Although the structures ofTT substances causing the disease were ye* ? be determined, the typical pattern of chromatographic peaks with long retend^ times versus aldrin (Fig. 2) was found ful for detecting toxic fats. F ig u r e 2. Microcoulometric gas chromatograms of material isolated from two toxic by-product fatty acids obtained from commercial fatty acid manu facturers. GLC column temperature 248"C; R denotes retention time versus aldrin. Subsequently, electron capture-gas chro matographic procedures were developed (14-17) which were approximately 2000 times more sensitive than the microcoulo metric methods and could detect less than 10 ppb of chick edema factors in lipid sam ples. Electron-capture gas chromatography also provided increased resolution so that additional peaks with characteristic reten tion times could be observed in extracts from toxic fats (see Fig. 3). In addition, a bio assay test with increased sensitivity was de veloped (18.19) using fertile chick eggs. In jection of chick edema factors resulted in decreased hatch and development of em bryonic deformities and edema (15). Em bryos which failed to hatch exhibited mal formed beaks, eye defects, leg deformities, and lack of development of the right mesen cephalon. Embryos which hatched exhibited 60 Environmental Health Perspectives 784387 01157, ) 1 17 pounds isolated from a toxic feed fat as hexachlorohexahydrophenanthrenes on the ba sis of mass and other spectral data. Finally, in 1966, Cantrell et al. (22) showed by x-ray crystallography that one of the active crystalline materials isolated earlier from toxic fat was 1,2,3,7,8,9-hexachlorodibenzo-p-dioxin. Wootton immediate ly demonstrated that a synthetic hexachlorinated dibenzo-p-dioxin displayed prop erties similar to the material isolated from the toxic fat and, in addition, produced the chick edema disease in chickens (J.C. Woot ton, Procter and Gamble Co., private com munication. 1966). Electron-capture gas chromatogram of ..* from a toxic commercial glyceryl mono. CitX column temperature. 210s C: numbers . peaks are retention times versus aldrin. and defective feathering' and growth -Uion. "y ty of the Chick Edema Factors omatographic behavior and ultraviolet rption spectra of toxic fractions indi- i initially (-?) that the toxic factors substituted aromatic compounds, per' substituted naphthalenes or phenanles or aromatic steroids. A major break- :gh on the arduous road to identifying structure of chick edema factors oc*ed in 1960 when Harmon et al. (11. 20) 'uteri crystalline material from a toxic fat <i determined that it was an aromatic subinee containing 47% chlorine (M. Tishler, rck and Co., private communication). A nilar crystalline material which was chick .*ma-active at 0.1 ppm in the diet was iated from a sample of commercial trion found to be toxic to Cebus monkeys '>). The monkeys fed the toxic triolein hibited fatty liver, liver necrosis, pan>atic atrophy and fibrosis, bile duct prolii'ation, hemosiderosis, gross hemorrhage in e gastro-intesfcinal tract, and erythrocytongocytosis. Subsequently, Wootton and "rchene (21) characterized two com Formation and Occurrence of Chlorodibenzo-p-dioxins (Dioxins) Tomita et al. (23) had shown that chlorophenols and their salts can condense to form ehloro derivatives of dibenzo-p-dioxin. Thus, chiorophenols appeared to be the source of dioxins. Accordingly, FDA scientists (24) pvrolyzed a number of commercially avail able chiorophenols. including pentachlorophenol, and obtained chlorodioxin mixtures with GLC peaks having retention times identical to those found in toxic fats. The most toxic pyrolysis product, identified as 2.3.7.8-tetrachIorodibenzo-p-dioxin, was ob tained from 2.4.o-trichlorophenol. Hexa-. hepta-, and octachlorodioxins were obtained from pentachlorophenol. Individual compo nents were isolated by preparative GLC from a reference toxic fat. Dioxins with 3, 4, 6, and 7 chlorine atoms were toxic to chick embryos. The presence of 2,3,7-trichloroand 2.3,7,8-tetrachlorodioxin as well as hexa-, hepta-. and octachlorodioxins in the reference toxic fat indicated that the tallow from which the fat was derived was con taminated with 2,4.5-trichlorophenol as well as pentachlorophenol. Kimmig and Schulz (25) demonstrated in 1957 that 2,3,7,Stetraehlorodioxin was highly toxic and cnloracnegenic and was formed in the in dustrial production of 2,4,5-trichlorophenol by alkaline hydrolysis of 1,2,4,5-tetrachlorobenzene. Toxicological studies of individual chlorodioxins with chicks (26) and chick ember 1973 61 784388 embryos (M. J. Verrett, private communica tion, 1970) again showed that the 2,3,7,8tetrachlorodioxin was the most toxic. Hexachlorodioxins were about one fifth as toxic and 2,7-di- and octachlorodioxins were the least toxic. Pentachlorophenol and other chlorophenols have been widely used as bactericides, slimicides, defoliants, and termite control agents in industry and agriculture. They are also used in the manufacture of food-packaging materials (Subpart E, Title 21 of the Code of Federal Regulations (21 CFR), Part 121 121.2001) and as components of tides that contact food (Subpart F oi 121.2500). A list of food additive"^** 2,4,5-trichloro- and pentachlorophenol given in Table 1. Because of their spread use, a variety of commercial chlom! phenols were examined for the presence individual dioxins and related compo^^ (27) Nonacidic material isolated from the chlorophenols was chromatographed on alu. mina, and fractions were examined by com* bined gas chromatography-mass spectrom. etry. Table 1. Food additive uses for 2,4,5-trichloro* and pentachlorophenol.' Regulation No. Subpart E: 121.2001 Subpart F: 121.2505 121.2514, (b) (3) (xxxi) 121.2519 (d) (3) 121.2526 (5) 121.2534 (d) (3) 121.2550 (b) (5) 121.2556 (b) 121.2557 (d) (3) 121.2562 (c) (4) (iii) 121.2596 Specific compound TCP-Na; TCP-K; PCP-K TCP-Na; PCP-Na PCP-Na PCP-Na; TCP-Na PCP-Na PCP-Na PCP-Na; PCP-K PCP; PCP-Na TCP-Na; PCP-Na PCP-Na PCP-Na Use Slime control in manufacture of paper and paperboard Slimicides in manufacture of paper and paperboard Preservation in can end cements Defoaming agents in manufacture of paper and paperboard Paper in contact with aqueous and fatty foods Animal glue for articles holding food Closures with sealing gaskets for food containers Wood preservative Preservative for defoaming agents used in coatings Antioxidant in rubber articles used in producing, processing, or holding food Preservative for ammonium alginate used in manufacture of poiy(vinyl chlo ride) polymers that contact food *Title 21, Code of Federal Regulations. bTCP-Na=sodium trichlorophenate; TCF-K=potassium trichlorophenate; PCP= pentachlorophenol; PCP-Na= sodium pentachlorophenate; PCP-K=potassium pentachlorophenate. A variety of polychlorodibenzofurans (furans) and polychlorodiphenyl ethers (ethers) was found in the chlorophenols in addition to dioxins. Up to 6.2 ppm 2,3,7,8tetrachlorodioxin was found in six samples of 2,4,5-trichlorophenol, and up to 39 ppm hexachlorodioxins was found in eight sam ples of pentachlorophenol. In addition, chloromethoxyfurans and chloromethoxyethers were found in 2,4,5-trichlorophenol, and 62 chlorohydroxybiphenyls were found in per.tachlorophenol. Chlorodioxins, furans, and ethers found in commercial pentachlorophenols are indi cated in Table 2. Combined gas chromato graphy-mass spectrometry (GC-MS) was used to detect mixtures of these compounds which were unresolved by GLC. Character istic mass spectral peaks for a dioxin, a furan, and an ether are shown in Table 3. Tht Environmental Health Perspectives 784389 GENP 011578  ' pioxina. furans, and ethers in commercial pentachioraphenols. ' Detected by combined GC-MS Dioxins * FuransB Ethers s ++ ++ + + 4+ -f* + +++ + + , ;,,pdibenzo-p-dioxins. , .ptdibenzofurans. rmliphenyl ethers. .1* weight and fragmentation patche dioxins, furans, and ethers were ncly different so that unresolved mix. - were readily characterized by eom* i GC-MS. The results of analysis of the racial chlorophenols indicated that the -oaks of chick edema disease could have :;ited from tallows contaminated with ihenols containing preformed dioxins. ^ C haracteristic mass spectral fragments of j. dioxins, furans, and ethers. Relative intensity, Cl,- CivFragment Cl.--Dioxin s Furan e Ether c .M 78 72 22 tf-Cl 7 10 2 ,t-2Cl 3 15 54 ,1- (CO -r Cl) 35 50 -- M-2(C0 + Cl) 25 -- -- M- (CO -f 2 Cl) -- 15 -- M- (CO 4- 3 Cl) -- 75 8 *Based on "Cl; 19c or greater relative intensity. 2.3,7,8-Tetrachlorodibenzo-p-dioxin. ' Positions of chlorine substitution unknown. obable Source of Chick Edema Factors Reports in early 1960*s indicated that penachlorophenol was widely used as a preser vative in hide stripping operations, and vnight be present in the by-product tallows 'fleshing greases) recovered from hides. A number of papers in leather trades periodi cals (28--30) described the use of trichloro- pentachlorophenol as preservatives for hide curing operations. A major domestic manufacturer of chlorophenols recommends the use of the sodium salts of 2,4,5-trichlorophenol or pentachlorophenol for a variety of hide preservation and hide treatment opera tions as well as for general plant sanitizing procedures (81). Prefcanning operations in clude trimming, brining (or salt curing) of the hides, followed by soaking, liming, bat ing (neutralization with buffering salts and treatment with a proteolytic enzyme), and pickling (32). The sodium salt of penta chlorophenol is recommended for use in salt curing, brining, soaking and pickling opera tions, and the sodium salt of 2,4,5-trichlorophenol is recommended for use in soaking operations (32). By-product fats are ob tained after trimming, soaking and liming treatments (32) so that the use of chloro phenols as preservatives during processing can result in contamination of the hide greases with chlorophenols. Contaminated hide greases may have been the source of the chick edema outbreaks of 1957 and 1960. Accordingly, three commer cial oleic acids examined earlier for chick edema factor were reexamined for the pres ence of chlorophenols (33). Two of the sam ples that were previously found to be posi tive for chick edema factor were also found to be contaminated with 2,3,4,6-tetrachlorophenol and pentachlorophenol. The third sample contained a trace of pentachlorophe nol. Finally, Metcalfe (34) presented concrete evidence that the source of chick edema fac tor was fleshing grease from hides that hac been treated- with commercial pentachloro phenol. He examined samples of fleshing grease from pentachlorophenol-treated hides as well as industrial tallows containing glue emulsion with added pentachlorophenol. Chlorophenol-containing glue emulsions have been used in dry rendering operations. A fleshing grease sample was found to be ex tremely toxic when tested for chick edema factor (chick feeding bioassay) at the 16% level (34). A gas chromatogram of chloro- dioxins isolated from this fleshing grease is shown in Figure 4. (The numbers above the .ember 1973 63 784390 r~ 6 --i Origin of 1969 Outbreak of Chick Edema Disease Figure 4. Electron-capture gas chromatogram (33) of dioxins extracted from a toxic fleshing grease. The numbers above the peaks refer to the .num ber of halogen atoms in the individual dioxin molecules, GLC peaks refer to the number of halogen atoms in the individual dioxins.) The Di vision of Chemistry and Physics, Food and Drug Administration, assisted with analysis of the sample. The presence of tetra-, penta-, hexa-, hepta-, and octachlorodioxins was confirmed by combined GC-MS. Penta- and hexachiorodioxins comprised 70% of the total dioxins found. The widespread use of chlorophenols as hide preservatives prior to the middle 1960's is the probable cause of the earlier outbreaks of chick edema disease. The use of ehlorophenol-containing glue emulsions for dry rendering is another possi ble source of toxic tallows. In 1970, a survey was made of a selected number of tallow processors and fatty acid producers. Ten of 45 samples of tallow and oleic acid were found to contain up to 77 ppb of dioxins (hexa-, hepta-, and octachlor odioxins). Four non-food-grade oleic acids contained 11 to 77 ppb of dioxins and one food-grade oleic acid contained 59 ppb of dioxins. Fleshing grease from two hide processors contained low levels (5 and 11 ppb of dioxins), indicating that the use of chlorophenols for hide processing has not been entirely discontinued. In early 1969, an outbreak of chick edema disease in North Carolina resulted in the death or destruction of some 300,000 chick, ens. An additional million birds were in. volved in further outbreaks in the next sev eral weeks. The cause was traced to the use of contaminated vegetable oil by-product fatty acids in the feed. Investigation by the Food and Drug Administration disclosed that the feed fat be came contaminated at a vegetable oil re finery which also formulated antimicrobial water treatment products containing various chlorophenols. An underground pipe line was found leading from the "pesticide" product plant to traps used to collect by-product fatty acids (acidified soapstock) from the veget able oil refinery. This pipe line unintention ally transferred "pesticide" plant wash water to the traps holding acidified soapstock, re sulting in contamination of the acidified soapstock intended for sale as feed fats. The possibility of further contamination was eliminated by removal of the "pesticide" oper ation from the refinery site. Prevention of Dioxin Contamination of Foods and Feeds Since commercial chlorophenols are wide ly used as termite control and antimicrobial agents, there are numerous opportunities for direct and indirect contamination of food and food fats. Dioxin contamination of fats and fatty acids destined for use in foods or feeds can be minimized by control of sources of direct contamination with chlorophenols. These sources include the use of tri- and pentachloi*ophenol for hide preservation and other pretanning operations and in glue emul sions for dry rendering of fats. Care must also be taken in the use of chlorophenols industrially (as antimicrobials) or on the farm (as wood preservatives for fences, barns, etc.). 64 Environmental Health Perspectives T K T *~ r C AOCTTA 784391 REFERENCES 1%v>W. B** Characterization of a type of ..identified compound producing edema in -vks. J. Assoc. Offic. Agr. Chem. 42: 120 (1959). (clman, L., et al. Studies of the chick edema ;ise factor. J. Assoc. Offic. Agr. Chem. 42: *i 1959). v.-iotton, J. C., and Alexander, J. C. Some chem.:tl characteristics of the chicken edema disease factor. J. Assoc. Offic. Agr. Chem. 42: 141 :1959)- danger, V. L., et al. Alimentary toxemia in iiickens; J. Amer. Vet. Med. Assn. 133: 172 1958). Vicar. S. A., et al. The effect of a toxic sub:nce in fat on poultry. Poultry Sei. 37: 1200 :)5S). Mien. J- R. The role of toxic fat in the produc:inn of hydropericardium and ascites in chick ens. Amer. J. Vet. Res. 25: 1210 (1964). Alien J. R., and Carstens, L. A. Electron micro scopic observations in the liver of chickens fed rnxic fat. Lab. Invest. 15: 970 (1966). \ Firestone. D., et al. The examination of fats and -'atty acids for toxic substances. J. Amer. Oil Chemists' Soc. 38: 418 (1961). kFriedman, L. Progress in the chick edema prob- m. Feedstuffs, 34: March 17, 1962. Ames. S. R., et al. The occurrence of the chick pericardial edema factor in some oleic acids and products derived therefrom. J. Am. Oil Chemists' Soc. 37: 10 '(I960). Harman. R. E., et al. The isolation and charac terization of the chick edema factor. J. Amer. Chem. Soc. 82: 2078 (1960). Food Additive Regulations, April 22. 1960 (Code of Federal Regulations (CFR) Title 21, Section 121.86); see also CFR, Title 21, Section 121.1070. Firestone, D., Ibrahim, W., and Horwitz, W. Chick edema factor. III. Application of microcoulometric gas chromatography to detection of chick edema factor in fats and fatty acids. J. Assoc. Offic. Anal. Chem. 46: 384 (1963). .I. Higginbotham, G. R.. et ai. Detection of chick edema factor in fats and fatty acids by electron capture gas chromatography., J. Assoc. Offic. Anal. Chem. 50: 874 (1967), Higginbotham. G. R., Ress. J., and Firestone. D. Note on a rapid screening method for chick edema factor in fats and fatty acids. J. Assoc. Offic. Anal. Chem. 50: 884 (1967). 16. Neal P. Note on an improved cleanup method ' for the detection of chick edema factor in fats and fatty acids by electron capture gas chroma tography. J. Assoc. Offic. Anal. Chem. 50: 1338 (1967). ` 17. Assoc. Offic. Anal. Chem. Official Methods of Analysis, 11th Ed., Washington, D.C., 1970, secs. 28.109 - 28.111. 18. McLaughlin, J., et al. The injection of chemicals into the yolk sac of fertile eggs prior to incuba tion as a toxicity test. Appl. Pharmacol. 5: 760 (1963). 19. Flick, D. F., Firestone, D., and Marliac, J. P. Studies of the chick edema disease, 2. Prepara tion and biological effects of a crystalline chick edema factor concentrate. Poultry Sci. 44: 1214 (1965). 20. Yartzoff. A., et al. Studies of the chick edema factor: II. Isolation of a toxic substance. J. Amer. Oil Chemists' Soc. 38: 60 (1961). 21. Wootton, J. C.. and Courchene, W. L. A contri bution to the knowledge of the structure of two hydropericardium-producing factors from a toxic fat. J. AgT. Food Chem. 12: 94 (1964). 22. Cantrell, J. S., Webb, N. C., and Mabis. A. J. Identification and crystal structure of a hydropericardium-producing factor: 1,2.3.7,8,9-Hexachlorodibenzo-p-dioxin. Acta Cryst. B25 ( I ) : 150 (1969): see also Chem. Eng. News 45: No. 5, 10 (1967). 23. Tomita. M.t Meda, S. and Narisada, M.; Dibenzo-p-dioxin derivatives, 27. Synthesis of polyholodihenzo-p-dioxin; Yakugaku Zasshi 79: 186 (1959); Chem. Abstr. 53 I3l5d (1959). 24. Higginbotham. G. R., et al: Chemical and toxi cological evaluations of isolated and synthetic chloro derivatives of dibenzo-p-dioxins. Nature 220: 702 (1968). 25. Kimmig, J.. and Schulz, K. H.; Occupational acne (so-called chloracne) due to chlorinated aromatic cyclic ethers. Dennatologia 115: 540 (1957). 26. Flick. D. F., Firestone, D., and Higginbotham. G. R.; Studies of the chick edema disease: 9. Response of chicks fed singly administered edema-producing compounds. Poultry Sci. 51: 2026 (1972), 27. Firestone, D., et al. Determination of polychlorodtbenzo-p-dioxins and related compounds in commercial chiorophenols. J. Assoc. Offic. Anal. Chem. 55: 85 (1972). 23. Hausam, W. Progress report on the curing of hides and skins. 1937-1949. J. Soc. Leather Trades' Chemists 35: 142 (1951). 29. Bhaskaran. R., Sen. S. M., and Das, B. M. Use of antiseptics in the process of soaking hides h ^ c e m b e r 1973 65 784392 Oo and skins--1. Studies on sodium pentachlorophenate. Ball. Cent. Leather Res. Inst.. Madras 3: 115 (1956). 30. Dempsey, II., Green, G. H,, and Haines, B. M. The incorporation of antiseptic in curing salt for domestic sheep skins. J. Soc. Leather Trades' Chemists 48: 424 (1964). 31. Dow Chemical Co., Product Information Bulletin on Antimicrobial Agents for the Leather Indus try, Midland, Michigan, 1968. 32. O'Flaherty, F. Leather. In Encyclopedia of Chemical Technology, Vol. 12, R. E. Kirk an<j D. F. Othmer, Eds., Interscience, New York 1967. 33. Higginbotham, G. R., Ress. J., and Rocke, a. Extraction and GLC Detection of pentachlotophenol and 2,3,4,6-tetrachlorophenol in fats, oils and fatty acids. J. Assoc. Offic. Anal. Chem. 53: 673 (1970). 34. Metcalfe, L. D. Proposed source of chick edema factor. J. Assoc. Offic. Anal. Chem. 55: 542 (1972). 78C T T 0rTK m r> 66 Environm ental Health Perspective 784393 i Survey of the Embryotoxic Effects of TODD in Mammalian Species by D. Neubert,* P. Zens,* A. Rothenwallner,* and H.-J. M erkerf hiiroduction Only limited data are presently available :> the embryotoxic effects of 2,3,7,8-tetra::iorodibenzo-p-dioxin (TCDD). This is true -v the number of animals used per experi- utal group, especially in the higher dose ~e, as well as for the number of dif.nt species of experimental animals testu vThere has been an understandable reice to investigate these highly toxic `mpounds, because of the hazards of con':'.minating animal quarters and the risk to orsonnel. Nevertheless, we feel that the ata available allow a rough estimate of the ngers that may be encountered with ex pire to such dioxins during pregnancy. The possibility that TCDD might have cer.n types of embryotoxic effects was real:ed for the first time when Courtney et al. /) reported their data on the teratogenic rcect of 2,4,5-T. Substances like TCDD have `'een known for some time (2 - 4 ) to be con'aminants of chlorinated phenols and derivaives of such compounds and they were '.ought to be responsible for an intoxication lied chloracne (5). Some commercial mples of 2,4,5-T apparently were rather ighly contaminated with these extremely oxic substances. `Abteilung Embryonal-Pharmakologie, Pharma kologisches Institut der Freien Universitt Berlin -ondertorschun gabereich 29), Berlin, Germany. ^Abteilung Embryonal-Pharmakologie, II. An isches Institut der Freien Universitt Berlin 4erforschungsbereich 29), Berlin, Germany. September 1973 In this paper we evaluate exclusively data which have been obtained using pure TCDD. No attempt is made to analyze the rather extensive literature on results obtained with 2,4,5-T samples contaminated to various de grees with TCDD or similar compounds. Same Special Aspects of Embryotoxic Effects Induced by TCDD Some General Aspects of Embryotoxic Action Considerable confusion has been caused by many previous investigators with the nomen clature used in the field of prenatal toxi cology. Unfortunately, the term used to specify certain experimental data in this field may suggest a special mode of action (6) and, more important, may trigger con clusions on a possible hazard to be expected during human embryonic development. We, therefore, wish to suggest a system of nomenclature used by our group which is in accordance with the terms used in toxi cological research and which at the same time takes into account the special situation of prenatal development. The most comprehensive term is embryo toxic (fetotoxic) effect. A special situation of an embryotoxic effect may be characterized, e.g., a teratogenic effect, an embryolethal (fetoiethal) effect, or a retardation (re tarded growth) etc. We feel that the following definitions prove to be convenient. By embryotoxic (fetotoxic) effects we denote all transient or 67 784394 GENP 011583 permanent toxic effects induced in an embryo or fetus, regardless of the mechanism of action. By embryolethal' (fetolethal) effect or embryomortality (or fetomortality) we denote prenatal mortality at any stage of embryonic or fetal development (may be referred to as LD* -- LD9$). This term is certainly not identical with embryotoxic but refers to a special event that may occur in the course of an embryotoxic action. We define a teratogenic effect as an ab normality originating from an impairment of an event typical for embryonic or fetal development (induction process, differentia tion). The abnormality should be largely ir reversible. It may be obvious by macroscopic appearance or "hidden" (micromorphological defect or inborn error of metabolism) or result in a mental abnormality. Some examples of embryotoxic, not tera togenic effects are: general retardation of embryonic (or fetal) growth, retarded oc currence (of certain ossification centers (re versible), fetal intestinal hemorrhages (without secondary effects). Representing a teratogenic effect would be an irreversible involution of lymphatic tissues (with conse quences in postnatal life) or a severe mental defect resulting from an impairment of pre natal brain development. Some Special Aspects Connected with the Action of TCDD A brief analysis of the effects of TCDD on prenatal development may suggest that TCDD is a surprisingly specific teratogen, interfering only with a few special develop mental processes. An increased frequency of cleft palate and kidney abnormalities of a special type are the only teratogenic effects which have been observed. Interestingly enough, it has not been possible so far to induce limb or head abnormalities with this agent, despite the fact that the drug was given at the "critical period" for the induc tion of such malformations. Larger experi mental series over a sufficient dose range are necessary to permit conclusions on the specificity of the teratogenic action of TCDD, however. TCDD is certainly not a teratogenic ag exclusively, since an increased fetomorta results if the drug is given at a high enoi dose over a long enough time interval though with such a dose schedule no ohvic symptoms of a maternal toxicity becoi evident. Apparently the higher doses lead a more general toxic effect on the embryor cells, but an effect on the placenta or mate nal tissues cannot be excluded at the momer Some of the toxic signs which can , demonstrated in the fetal tissue should n be referred to as teratogenic. This includt the intestinal hemorrhages (without pe: manent secondary changes), a fatty infiltr: tion (Fig. 1 ) of fetal livers (7) (to ou knowledge not reported before with any err. bryotoxic agent), as well as subcutaneou edema and delayed ossification. Most of thes< symptoms are reversible and they are cer tainly not the result of a typical interference with developmental processes. It is interest ing that most of these symptoms can alsc be demonstrated in tissues of adult organ isms under the toxic action of TCDD. In adult mice a pronounced involution of some lymphatic tissues can be demonstrated as an early toxic sign, together with a loss of weight (8), predominantly in a drastic reduction of the size of the thymus, the spleen, and lymph nodes (9). A similar in terference has been observed in our group with the development of these systems dur ing fetal development in the presence of TCDD (20). This impaired development of lymphatic organs results in a typical post natal insufficiency and in a pronounced reduction of the chances of postnatal sur vival. Although these symptoms very much re semble changes produced in the adult organ ism by TCDD (9), we would consider these embryotoxic effects as teratogenic, since a typical developmental process is altered and the defect can be demonstrated long enough to handicap the newborn. From all our studies we have seen no indication that the general growth of the fetal mice showing a teratogenic effect was affected to a significant degree. 68 Environmental Health Perspectives 784395 GENP 011584 I "tnuRE l. Example of an electron microscopic examination of fetal rat liver (day IS of gestation). TCDD 115 jig/kg) was given once on day 17 of pregnancy. Using TCDD it has been possible to produce a fatty rmltration of the developing fetal liver as early as on day 14 of gestation. We feel that the sort of embryotoxic ef fects induced by TCDD are very interesting from a theoretical point of view and to >ome extent unique. Therefore, they justify a further, closer analysis. Results and Discussion First indications of embryotoxic effects of TCDD were reported in 1970 by Sparschu et al. (11, 12) from experiments with rats. Further data on embryotoxic effects then became available from reports of Courtney and Moore (13) and from our group (1J>) (see Table 1). Three major effects were noted in these /'idies: intestinal hemorrhages in rat fe tuses and an increased frequency of cleft palate and kidney abnormalities in mouse fetuses. Teratological studies with rats have apparently only been performed using re peated doses of TCDD while with mice ef fects produced by single as well as by re peated doses of this drug have been re ported. We shall try to evaluate the data from teratological studies available and to supple ment them with unpublished data obtained in our laboratory. The following aspects will be discussed: (1 ) dose-response relationships after repeated doses of TCDD given to rats and mice; (2 ) dose-response relationships after single doses of TCDD given to mice and Septem ber 1973 69 784396 O Co Table 1. Evaluation of the published teratogenic (embryotoxic) effects induced by TCDD in rats and mice.* Species Rat Strain effect (system) Intestinal hemorrhage Dose Minimal tested 0.125 .^gAg =EDS = 0.5 ? Time TCDD given, days Route Oral ------- Reference Sparschu et al. (5) CD Kidney abnormality 0.5 >1 6-15 SC Courtney and Moore (:) Mouse CD-I CP 1? 3 >3 6-15 SC f Kidney abnormality 1 1-3 6-15 SC DBA/2J CP 3 > 3 6-15 SC Kidney abnormality 3 >3 6-15 s c C57B1/6J CP 3 > 3 f>-lD SC Kidney abnormality 3 <3 6-15 SC - M t* t* NMRI CP 3 6.5 6-15 Oral Neubert and Dillmann (U) 9 <9 9-13 Oral n 15 40 13 Oral Neubert et al. (this paper) 5 15 11 Oral t* 19 *The smallest dose with which a significant teratogenic effect has been produced is indicated. Since sometimes only one dose level was tested this does not necessarily give the smallest dose from which a teratogenic effect could result. Routes were both oral and subcutaneous (SC). An attempt was also made to estimate the ED from the few data available. s ED: dose required to produce effect in 50% of animals. evaluation of phase specificity; (3 ) embryotoxic effects observed after application of TCDD.together with other teratogens. Dose-Response Relationship of Teratogenic Effects and Fetomortality After Repeated Doses of TCDD to Rats or Mice The first striking result which is immedi ately apparent is the extremely low dose of TCDD which is able to induce teratogenic and fetolethal effects in mice and rats. Al though these species must be considered to be comparatively insensitive during the adult status towards the toxic action of this com pound, repeated or even single doses of as little as 1 - 1 0 /Ag/kg are capable of reproducibly triggering malformations of certain types. This is by far the smallest effective dose of any teratogen known today. 70 Interestingly, of the series of chlorinated dibenzodioxins, the tetrachloro derivative ap parently is the most active one. The hexachloro derivative also seems to show some embryotoxic activity, while the dichloro and the octachloro derivatives have been report ed (15) to be nonembryotoxic. The few data available suggest that embryotoxic activity is a property of those compounds having a pronounced chioracne potency (15). Teratogenic effects induced by TCDD-- Two major types of malformation have been reported so far in fetuses of rats or mice treated with TCDD during pregnancy: cleft palate--so far reported only in mice--and kidney abnormalities of a certain type, ob served in both species after single or re peated doses of TCDD (Table 2 ). Environmental Health Perspectives GNP 011586 1 Embryotoxic effects induced by TCDD.,,(' Abnormality intestinal hemorrhage Sidney abnormalities Lethal Cleft palate Kidney abnormalities Lethal Dose producing --50% effect, msA s e 0.5 ? 1-2 ? 1-2 ? 6 1-3 7 H OD was given on days 6-15 of pregnancy. ! :fi-ature data (11, IS, i i ) . - no exact data available from the litera- Doses assumed to be in the range indi- hermore. an involution of fetal thymus pleen and other lymphatic organs can .served (10) which affects the survival birth. Intestinal hemorrhage and fatty '.n-ation of the fetal liver do not represent .uoirenic effects, as discussed above. Withho adult organism similar symptoms can -need. are convinced that looking more closely .uses affected by TCDD will reveal more f 'Nimalities ; this has been the case with i drug effect evaluated carefully enough hii the modern biochemical and micro'irphologicai methods available. Other than the reports on teratogenic and hi,)ryotoxic effects produced by TCDD in ce and rats with documented experimental a we found only a few references which .rgest that embryotoxic effects may also cur in other" animal species after a treat ment with TCDD. Although we did not find Verifiable data, TCDD apparently can also .'reduce embryotoxic effects in hamsters. Eye abnormalities and reduction of mean fetal weight--neither of which is typical of sympims seen in rats or mice after treatment with 'CDD--as well as gastrointestinal hemorrages and increased prenatal mortality are mentioned `after doses of TCDD in the /g/ kg range for 5 days (16). Unfortunately, we have no knowledge of teratological experiments performed with the highly TCDD-sensitive species guinea pig and rabbit. It would be important to know whether the comparatively high toxicity "yfn in adult animals of these species is also paralleled by a high degree of sensitivity of the embryos or fetuses when compared with that of rats and mice. Fetom ortality induced by TCDD-- Al though the teratogenic effect induced by TCDD seems to be rather specific because the development of only a limited number of special organ systems is found to be im paired, TCDD when given in multiple doses does not represent an exclusively teratogen ic agent. The occurrence of developmental abnormalities in chronic experiments is al most paralleled in a dose-response curve by the fetomortality (Fig. 2 ). However, as in many teratological experiments, there is no obligatory connection of these two para meters. In mice a high cleft palate frequency can be obtained without any apparent feto mortality (Fig. 3), just by reducing the time of treatment from 10 days (day 6-15) to 5 days (day 9-13 of gestation). Figure 2. Comparison of the teratogenic effect and the degree of fetomortality induced by TCDO. The effect of a single dose given on day 13 of preg nancy is compared with that resulting after re peated doses given during days 6-15 of pregnancy. The points are derived from the evaluation of the total fetuses per treated group of mice (at least 12 litters per dose). TCDD was given in rape-seed oil by stomach tube once a day (1 P.M.). Cleft palate frequency is evaluated in this experiment as percent of the viable fetuses. Effect (probit scale) is plotted against dose (log scale). The control values represent 2000 fetuses from mice treated with rape-seed oil for 10 days. September 1973 71 784398 fetuses affected controls 6-15 9-13, day trea ted w ith TCDD Figure 3. Comparison of teratogenic effect and feto mortality produced by TCDD. Ten pregnant mice each were treated with 9 ig/kg TCDD orally during the time interval indicated. The numbers of fetuses affected are given (M S.D.) per litter. Average number of implantation sites per Utter was 11.5 (= 100%). Controls received rapeseed oil only. CP denotes cleft palate frequency. Differences among strains in the degree of teratogenic and fetolethal effects induced by TCDD in mice--Information on differences in the susceptibility towards TCDD of dif ferent strains of experimental animals are so far available only for mice, mainly through the data of Courtney and Moore (13). Tables 3 and 4 summarize these data, supplemented by information obtained with NMRI mice in our laboratory (11). For a better comparison we have recalculated some of the data. Since for half of the strains data are only available at one dose, the com parison is made for the effect produced b 3 /xg/kg TCDD given during days 6 -1 5 c gestation. As can be seen from Table 3, with thi scheme of treatment a noticible but smal fetomortality occurs only with the CD-: mice. With the same scheme of treatment a significant increase in the cleft palate fre quency over that of the controls can be ob served with all of the four strains tested. While the incidence of this type of teratogen ic effect is roughly the same with three of the four strains (3-4% ), the C57B1 mice are clearly more susceptible to the teratogen. Further data of Courtney and Moore (;.?) lead to the conclusion that the special sensi tivity of the C57B1 mice also holds for the induction of kidney abnormalities. At the moment no clue is available for the cause of this higher sensitivity. An increased sus ceptibility of the target tissue or alterna tively a special rate of drug metabolism in this strain, leading to a higher concentration of the effective drug in the fetus, may be responsible for this special sensitivity. Fur ther data are needed to distinguish be tween these possibilities. Dose-Response Relationship of Teratogenic Effects and Fetomortality in Mice after a Single Dose of TCDD (Phase Specificity) We have started some systematic studies to clarify the question whether the terato genic effects may be induced by single doses of TCDD--and by what doses--and what ap pears to be the most sensitive interval of fetal development. The induction of deft palates was chosen as a criterion in these studies, but other types of malformation (kidney abnormalities) may also be evaluat ed fi*om these experimental series. The highest degree of malformations can be produced (14) when the drug is given to the mice on day 11 of pregnancy. Although a significant increase in the cleft palate fre quency over that of the controls can also be induced by giving the teratogen on days 10 or 12 of gestation, the effect obtained on these days is only about half of that pro duced on day 1 1 for a given strain. 72 Environmental Health Perspectives 784399 GENP 011588 , Fetomortality induced by TCDD in different strains of mice.1 Avg. fetomortality, ^ /litter" ..-o strain TCDD Controls 12.4 (144) f*L 5.3 (110) 3.6 4.3 27.0 (103) 26.1 nr 5.3 (49) 10.8 :rnD . 3 jig/kg, was given daily on days 6-15 .,[ pregnancy. \':ilues in parentheses denote percentage when mipared with the corresponding controls re viving the vehicle only. In none of the strains sted was a pronounced fetomortality ob- "-ed. 3D given subcutaneously (SCI in DMSO: .-.a of Courtney and Moore (13). ; f DD given orally in oil; data of Neubert and Diilmann i l i ) . ingly enough, cleft palates can be produced by giving different teratogens with a phase optimum which-varies considerably, indicat ing quite different modes of action of the various teratogens (Fig. 4). Because of our interest in the dose-re sponse relationship of the teratogenic effect of TCDD we compared the dose-response curves (cleft palate frequency) obtained with five cleft palate-inducing drugs (Fig. 5). The steepness of the dose-response curves varies considerably when the different drugs are When TCDD is given on day 13 of gesta:iii. an effect of only about one third of that n on day 1 1 is produced. It should be itioned that in all of these experiments . one time per day (at noon) was checked :e it was not intended to pinpoint exactly f V rime of the highest sensitivity. The maxiv.. y . m of the effect may, therefore, be somevhat higher. in order to compare the cleft paiate-induc!iir effect of TCDD with that seen after treat ment of pregnant mice with other teratogens have performed most of the following periments by giving TCDD on day 13 of station, even though the effect with TCDD rained on this day is not maximal. Interest F ig u r e 4. Phase specificity of cleft palate induction by various drugs. A rough survey is given on the optimal effect produced by different teratogens. The points are taken from large experimental se ries performed in our laboratory with NMRI mice. 100*3 indicates the maximum effect, not a cleft palate frequency of 1009r. The arrow indi cates the time at which most of our combination experiments have been performed. Not all tera togens exnibit their maximal effect at this stage of development. Table 4. Cleft palate frequency induced- by TCDD in different strains of mice.* TCDD Controls Mouse strain CD-Ie DBA e NMRI*1 C57B1* Fetuses evaluated 104 55 271 58 CP, 9 3 4 3 22 Affected litters (ac) 3/10 (30) 2/9 (22) 7/24 (29) 5/7 (71) CP. n 6 < 0.3 <1 0.7 <1 Affected litters (7e) 0/29 ( < 3) 0/23 ( < 4) 10/160 (6) 0/23 (< 4) 1TCDD, 3 tig kg, was given daily during days 6--15 of pregnancy (same experimental conditions as in Table 3). Strain C57B1 appears to be the most susceptible strain. All the other strains tested show about the same degree of sensitivity. ^ % CP = percentage of cleft palates of the total fetuses examined in this group. eTCDD given SC in DMSO; data of Courtney and Moore fJ3). 'TCDD given orally in rape 3eed oil. data of Neubert and Diilmann (14). Tptember 1973 73 784400 ask; S g e n P 011589 compared. Table 5 gives, in addition, the tan calculated from these curves as well as the dose range required for increasing the teratogenic effect from 2 % (just signi ficantly over the controls) to 50% (probit 3 to 5). ___________ . s----- 3-.; / / ;/ ./ jug kg 1 m g kg :0 jO *00 j F ig u re 5. Dose-response relationship observed after single doses of different cleft palate-inducing drugs. All drugs were given on day 13 of preg nancy. TCDD and 2,4,5-T were given per stomach tube in rape-seed oil; 6-AN, dexamethasone, and cyclophosphamide were injected subcutaneously. Each point represents the frequency observed in the fetuses from at least 12 litters, at the lower dose range of at least 20 and generally about 30 litters. The data are given as percentage of the total fetuses evaluated at the dose specified. Tan n values of the dose-response curves are given in Table 5; threshold doses derived from these curves are given in Table 6. Of the drugs tested, 6-aminonicotinamide (6-AN), a teratogen extensively studied biochemically as well as micromorphologically in our laboratory (17-20) shows the larg est increase in cleft palate frequency with increasing doses (log dose), while 2,4,5-T shows the smallest. A dose-response curve of intermediate steepness is obtained when the effect of TCDD or of dexamethasone or cyclophosphamide (Endoxan) is plotted against the dose given. The dose required for producing a 1 0 % effect with the various teratogens differs by more than four orders of magnitude ! Embryotoxic Effects Observed after Ap plication of TCDD Together with Other Teratogens Special emphasis was given to experiments Table 5. Dose-response relationship with five palate-inducing drugs/ Drug 6-AN TCDD Endoxan Dexa- methason 2,4,5-T tan a of doseresponse curve 7.6 2.3 1.5 1.4 0.8 Increase in dose necessary to in crease effect from probit 3 to 5 (2% to 509c effect) 1.3 X 3X 5X ED* (cleft palate) Tng/kg 10 0.04G 60* 6X 11X 20 2000? * *All drugs were given to the mice as a single dose on day 13 of pregnancy, (evaluation of the data shown in Fig. 5). hNo 5 0 effect obtained; value extrapolated. in which TCDD was combined with othe teratogens. We believe that such combine tion studies may serve several purposes First, very little is presently known regard ing the teratogenic effect of drug combina tions, despite the fact that single compound; very rarely act on an organism, either undei ambient environmental conditions or during treatment. Therefore, information on the chance of potentiating effects and the pos sible health hazards of combined drug ac tions on embryonic development is urgently needed. Also, almost no information is avail able today on the possibility of synergistic action of two or more drugs in producing an impairment of embryonic development which cannot be induced, even by high doses, by either substance alone. Informa tion available from studies with carcinogens suggests that the likelihood that a certain drug will induce special teratogenic ef fects may be greatly increased by varying nutritional factors (e.g. vitamins, protein uptake, heavy metal concentrations) or by blocking certain metabolic pathways. When designing experiments in which the teratogenic effect of combined drug actions is to be evaluated, several possibilities have to be considered. (1) Both (or all) drugs may be given simultaneously over an extend ed period of time. This has the advantage that all substances act on the target tissue 74 Environmental Health Perspectives 784401 GENP011590 i 1 __probably different--optimal phase of , w `;,pmenfe. The--disadvantage of such an %.,,.-imental setup may lie in the secondary interesting effects of the drugs pro: outside of the critical period (e.g., Methal effects) which could mask atogenic effect to be evaluated. (2 ) :;i or all) drugs are given only once, ... ;.r the same time. This gives more favorconditions for evaluating the mechanism action. However not all of the drugs may the chance to act at the most sensitive . (3) Both (or all) drugs are given mice, but in sequence, each at the optimal of its effect. The advantages and discages of this experimental design are us. -inch of these experimental designs has . vantages when studying special problems. K:.ch one has practical as well as theoretical .plications. We feel that in the long run ..!! rhe three models have to be tested in :\>r to allow an exact elucidation of teranic effects of drug combinations. In the ent paper we present the results of some riments performed with TCDD combi; ..ions by use of experimental designs 1 and Cleft palate frequency induced by a com bination of TCDD and 2,4,5-T given during days S and 15 of gestation--Studies with combinations of TCDD and 2,4,5-T were per formed firstr since an effect of this drug combination was of special interest. Large experimental series performed in various laboratories with 2,4,5-T preparations con taminated with dioxins to a very different ex tent led to such variable results that studies with a clear-cut combination of compounds were warranted. Figure 6 shows the results of experiments in which a teratogenically active dose of 2,4,5-T (60 or 100 mg/kg) was combined with a threshold dose of TCDD or even much smaller doses. Although we are completely aware of the fact that there is no "threshold dose" we use this term, for convenience to indicate that dose (derived from a dose-response curve) which gives just no significant in crease of cleft palate frequency over that of the controls, when 300-500 fetuses from treated mice and 2000 fetuses from controls are evaluated (1.5% cleft palate frequency, 0.7% seen with controls; under our experi mental conditions, > 3 .5 ). A clear-cut effect is seen on combining 60 mg/kg 2,4,5-T with the "threshold dose" of TCDD (2 /t/kg), and a detectable poten tiation is produced even with 1 / 1 0 of this Table 6. Just nonteratogenic doses of the drugs used in oar combination experiments (threshold doses).* "Just teratogenic" dose " "just nonteratogenic" dose "Threshold dose" as fraction of ED e TCDD, Mg/kg Day 13 Day 6-15 15 3 12 2 1/3 2,4,5-T mg/kg Day 13 250 150 Day 6-15 40 30 1/13 d 6-AN, mg/kg Day 13 7.5 7 7/10 Dexamethasone, mg/kg Day 13 4 2 1/10 Endoxan, mg/kg Day 13 15 10 1/6 d *All doses were derived from dose-response curves, with at least 300 fetuses evaluated at the low dose levels. The data refer to single* doses given on day 13 of pregnancy. "The "just teratogenic" dose gives the lowest dose found to produce a significant effect (x1 > 3.5) under our experimental conditions, 300-500 fetuses being evaluated per dose. ' Although we are aware of the fact that a "threshold dose" cannot be determined accurately, we use this term for the convenience of planning our combination experiments. In this paper this term is defined as the lowest dose found under our experimental conditions (300-500 fetuses) not to be able to produce an effect significantly different from the controls (2000 fetuses evaluated). For a more accurate definition this dose is also characterized as a fraction of the ED*. d Fetal EDw extrapolated since this dose is toxic to maternal organism or fetolethai. Septem ber 1973 75 784402 fe tu s e s affected per titter 25-; r-n over-addltlve %<2 e ffe c t controls fe tu s e s affected per litter 2 5 -i IOv^r- 10Qd- at- 2.4.S-T TCDO ippm j SO 60 22 (33) 60 60 0302 (03) 00 100 01 01 O) 100 100 002002 (Q2) 1.0 - o.*4 0.12.4, S-T TCDD (ppm ) ill t5 15 22 (133) nil, 15 15 0302 (13) i Am 30 30 30 3Q 22 (6 6 ) 02 02 '651 F ig u r e 6. Cleft palate frequency in mice produced by a combination of TCDD and 2,4,S-T. Both drugs were given in the doses indicated (mg/kg for 2,4,5-T and fig/kg for TCDD) during the days fi-lo of pregnancy once daily by stomach tube. In each group 20 litters were evaluated. The data are given as number of fetuses per litter, (V). The effect produced by 60 mg/kg 2,4,5-T -H 2 Mg/kg TCDD is highly significant, that produced by 100 mg/kg 2,4,5-T 4- 0.1 Mg/kg TCDD just significant (P = 0.01). dose of TCDD. This combination would cor respond to a "contamination" of 3.3 ppm. When the dose of 2,4,5-T--and thereby the effect produced by this compound alone--is increased, even smaller doses of TCDD (0.1 n S / k g ) lead to a detectable increase in the frequency of malformations. This combina tion would correspond to about 1 ppm. No significant potentiation can be observed when the dose of TCDD is lowered by another fac tor of 5 (to 0.02 n g / k g ) , corresponding to 0.2 ppm. Although in such an experimental setup an effect by as little as 50 ng/kg TCDD can be detected, these data clearly show that a potentiating effect in this system cannot be obtained--even if highly teratogenic doses of 2,4,5-T are used--with a combination con taining less than 0.5 ppm TCDD. Further experiments were performed to get some information on what doses of TCDD are required to induce a detectable effect when just "nonteratogenic" doses of 2,4,5-T are used. Under the conditions specified, 30 mg/kg was about a threshold dose. When this do3e of 2,4,5-T is combined with 2 ng/kg TCDD Figure 7. Cleft palate frequency in mice produced by a combination of TCDD and 2,4,5-T. Exneri. mental conditions as in Fig. 6. The effects ohserved with both 15 mg/kg 2,4,5-T 4- 2 Mg-'ktr TCDD and with 30 mg/kg 2,4,5-T 4- 2 kg TCDD are highly significant (P < 0.0027). (also a threshold dose), 66 ppm, a clear-cut potentiation can be observed (Fig. 7 ). The same dose of TCDD (2 /tg/kg) increases the cleft palate frequency over the background also when half the 2,4,5-T dose is used (15 mg/kg), corresponding to 133 ppm. But 1/10 of the TCDD dose (0.2 /tg/kg) does not in duce any detectable effect with either of the doses 2,4,5-T used in this experi mental series. This suggests that when just nonteratogenic doses of 2,4,5-T are used, there is no effedt with the system used if less than 10 -2 0 ppm TCDD is present. The results obtained with our experimental setup may give too high values since we have used neither the most sensitive strain of mice nor the malformation seen with the lowest dose in this strain (kidney abnor malities). It is furthermore interesting that with none of the combinations mentioned an increased fetomortality could be observed. Cleft palate frequency induced by giving a combination of TCDD and other teratogens simultaneously on day 13 of gestation.--In these studies ail the teratogens were given only once. Day 13 was chosen as the most convenient time of gestation (Fig. 4). In' these studies, therefore, two of the teratogens were not given at the most sensitive phase of development. Environmental Health Perspectives 7 6 c n n jk k t 784403 1 .u-o '-.4,5-T cannot be considered a very fl!i teratogen we have studied the posof an occurrence of a potentiation 1'CDD is combined with other cleft during teratogens. Special emphaplaced on the differences in dose ^ characteristics for these agents. Re- . f such experimental series are com. : in Figure 8. In this study the cleft frequency was measured when just ratogenic doses (threshold doses) of the . .irons were combined with a threshold TCDD (12 ^g/kg). ~~c "RE 8. Cleft palate frequency induced by drug mibinations. All drugs were given once on day of pregnancy (TCDD and 2,4,5-T in rape-seed il by stomach tube, all other drugs subcutan eously) at the doses indicated, which represent -he just nonteratogenic (or threshold) doses as `-xplained in Table 6. At least 12 litters were -'aluated in each group. The data are given as ercentage of the total fetuses evaluated per -roup (probit scale). It can be seen that a potentiation occurs -:ier a treatment with all the teratogens 'tudied. The degree of potentiation varied omewhat; a combination of TCDD with -.4,5-T resulted in about 15% cleft palate and *'3ut the same frequency was observed with 'mbination of TCDD with dexamethasone. With a combination of TCDD and 6-AN almost 60% of the mouse fetuses showed this teratogenic effect. It is interesting that the dose-response curve obtained with 6AN is very steep when compared with those for 2,4,5-T or dexamethasone. These data as well as data obtained with other combination {21) suggest that an espe cially pronounced potentiation is to be ex pected when two drugs showing both a rath er steep dose-response curve are combined, while less potentiation is to be expected from the combined action of two drugs re vealing a flat dose-response relationship. Of course, this prediction only holds as long as each drug does not act by interfering with the metabolism of the other. Accord ingly, the TCDD-2,4,5-T pair should be ex pected to give a moderate degree of potentia tion in the assay system used here. Cleft palate frequency after treatment with a single dose of more than two terato gens.--We have, furthermore, performed studies to elucidate the possibility of TCDD potentiating the teratogenic action of other drugs when several such teratogens are given simultaneously in "subthreshold" doses. For these experiments all the drugs were given in doses of about half of the just non teratogenic dose. Figure 9 shows that neither a combination of 6-AN and dexamethasone nor the com bination of the three drugs 6-AN, dexametha sone, and 2.4.5-T at this dose leads to an effect significantly different from that ob served in controls. When TCDD is added to this drug combination as a fourth drug, however, at about half of the just nonteraogenic dose (6 ^g/kg), a pronounced effect can be demonstrated. The cleft palate fre quency is about 30%. From these combination experiments and with regard to the model studied we wish to draw the following conclusions. (1) When doses of TCDD much lower than the threshold dose are combined with tera togenic doses of drugs that are able to pro duce the same teratogenic effect, a potentia tion may be expected, even with doses of TCDD as low as 1/20 to 1/50 of the threshold September 1973 ii 784404 i G E N pojj 5931 probit % 5-- -40 of the teratogens used is not drasticali altered during the combination studies. 1 2.4,5-T 2.4.5-T 75mg/kg TCDO 6 p g/kg F ig u r e 9. Cleft palate frequency induced by drug combinations. Experimental conditions as in Fig ure 8. except that more than two drugs were given simultaneously in some of the experiments. The doses given represent about half the threshold dose, as defined in Table 6. dose (2 fig/kg). (2) When j u s t nonteratdgenic doses of TCDD are combined with just nonteratogenic (threshold) doses of drugs which are able to produce the same teratogenic effect, a potentiation may be expected, the degree of which depends on tan a of the dose-response curve of the teratogens used. (3) When two teratogenic drugs are com bined at half the level of the threshold doses or lower, a significant effect is not to be expected unless both the teratogens exhibit very steep dose-response curves. In the case of TCDD a just significant effect is obtained with 6-AN but not in combination with 2,4,5T, dexamethasone, or Endoxan. (4) When TCDD at half the level of the threshold dose is combined simultaneously with three other teratogens, all at this dose, a clearcut potentiation is observed. (5) Our results do not give any indication that in the system used TCDD in a dose of about 1 / 1 0 of the threshold dose would be able to induce a teratogenic effect, even in combination with up to three other tera togens simultaneously present in doses of half the threshold dose or lower. All the foregoing conclusions are only valid as long as the drug metabolism of any Materials and Methods For all experiments performed in this laboratory mice of the strain NMRI (pUr. chased from Schwenke & Co., Bad Nauheim Germany) weighing 29 3 g were used! The animals were mated (20 females with 10 males per cage) for 2 hr and subsequently checked for vaginal plugs. The 24-hr period following the mating (8 A.M.) was called day 0 of pregnancy. Altromin R and tap water were given ad libitum. The animals were sacrificed on day 18 of pregnancy and the number of viable fetuses was counted. Resorbed fetuses were regarded as dead. Litters with less than five im plantation sites were not included in the evaluation. The average litter size was 11.1 1 .3 with this strain. The results presented in this paper are based on the evaluation of about 1000 treated pregnant mice (ca. 110 0 0 fetuses) and about 250 controls (ca. 2SOO fetuses). Embryotoxic effects have been evaluated in this paper as frequency per single litter, or alternatively as percentage of the total of fetuses examined. Although the litter should be considered the experimental unit for statistical evaluations in teratoiogical studies we feel that a calculation of the embryotoxic effects based on the total num ber of fetuses examined is justified, since a closer analysis of our data always has re vealed that the occurrence of embryotoxic effects is distributed randomly (14). In all general calculations of our data litters show ing more than 75% resorptions have not been included but were listed separately. Embryolethal and teratogenic effects were statistically evaluated from the total per centages or from the frequencies occurring in each litter (M S.D.) by using a -test {22) or a x:'test. Pure TCDD (Lot No. 851:142-26) was kindly provided by the Dow Chemical Co., Midland, Michigan, U.S. This extremely toxic substance was handled with utmost care. The purity (gas chromatographic analysis) in- 78 Environm ental Health Perspectives 784405 N i was 98.6%. The drug was dissolved ^ ketone or chloroform and vigorously , , . (| with rape-seed oil. Doses of 0 .1 ml/10 Gwere given by stomach tube. v 2.4,5-T (Dot SHG) was kindly pro- !y C.H. Boehringer, Ingelheim. GerThe dioxin content indicated was < ;ipm. Since this preparation was syn c e d by a special procedure it may be .-sidered dioxin-free. The drug was disveil in rape-seed oil by gently warming to i r and given to the mice by stomach tube 1 ml TO g mouse). When 2,4,5-T and 0 were given together, stock solutions mixed before giving the solution (oil) o experimental animals at a dose of .I 10 g. -AN was purchased from Calbiochem Pexamethasone (Fortecortin) was a gift E. Xerck, Darmstadt, Germany. It was "cn to the mice in aqueous solution by cutaneous injection. ' 'are cyclophosphamide (Endoxan) was .ft of Asta-Werke, Brackwede, Germany. "\ drug was given to the mice in aqueous J ion by subcutaneous injection. REFERENCES . Courtney, K. D. et al. Teratogenic evaluation of 2.4.5-T. Science 16S: 864 (1970). . Hofmann. H. T. Neuere Erfahrungen mit hoch toxischen Chlorkohlenwasserstoffen. Arch. Exp. Path. Pharmak. 232: 228 (1957). Kirnmig, J., and Schulz, K. H. Chlorierte aroma tische zyklische ther als Ursache der sogenann ten Chlorakne. Naturwiss. 44: 337 (1957). . Higginbotham. G. R., et al. Chemical and toxi cological evaluations of isolated and synthetic chloro derivatives of dlbenzo-p-dioxin. Nature 220: 702 (1968). - Teleky. Die Pernakrankheit (Chloraene). Klin. Wschr. 6: 845 (1927); ibid., 6: 897 (1927); ibid., 7: 214 (1928). Neubert. D.. and Merker. H. J. Arch. Toxicol., in press. . Becker, D. The effect of folate overdose and of 2.3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on kidneys respectively livers of rat and mice em bryos. Teratology, in press. 8. Buu-Hoi, N. F., et al. Organs as targets of "dioxin" (2,3,7,8-tetrachlorodibenzo-p-dioxin) in toxication. Naturwiss. 59: 174 (1972). 0. Fink, J., et al. in preparation. 10. Stolpmann, H.-J. in preparation. 11. Sparschu, G. L.. Dunn, F. L.. and Rowe, V. K. Teratogenic study of 2,3,7,8-tetrachIorodibenzop-dioxin in the rat. Toxical. Appl. Pharmacol. 17: 317 (1970). 12. Sparschu, G. L.( Dunn, F. L., and Rowe, V. K. Study of the teratogenicity of 2,3,7,8-tetrachlo rodibenzo-p-dioxin in the rat. Food Cosmet. Toxicol. 9: 405 (1971). 13. Courtney, K. D., and Moore, J. A. Teratology studies with 2,4,5-trichlorophenoxyacetic acid and 2.3.7,8-tetrachlorodibenzo-p-dioxin. Toxicol1. Appl. Pharmacol. 20: 396 (1971). 14. Neubert, D., and Dillmann. I. Embryotoxic ef fects in mice treated with 2,4,5-trichloro phenoxyacetic acid and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Arch. Pharmacol. 272, 243 (1972). 15. Schwetz, B. A., et al. Toxicology of chlorinated dibenzo-p-dioxins. Environ. Health Ferspect. No, 5: 87 (1973). 16. Wilson, J. G. et al. FDA Report: Report of the Advisory Committee on 2,4.5-T to the Adminis trator of the Environmental Protection Agency, May 1971. 17. Koehler, E., Barrach. H.-J., and Neubert. D. Inhibition of NADP dependent oxidoreductases by the 6-aminonicotinamide analogue of NADP. FEBS tetters 6: No. 3.225 (1970). 18. Neubert, D., et al. In: Biochemical Aspects of Teratology. (Advances in the Bicsciences. Vol. 6), Pergamon Press-Vieweg, Beaunschweig 1971, p. 575. 19. Barrach. H.-J. Inaumiral Dissertation, Free Uni versity Berlin, 1973. 20. Barrach, H. J. Effect of 6-amino-nicotinamide on the glucose metabolism of embryonic tissue. Tn Metabolic Pathways in Mammalian Em bryos during Organogenesis and Its Modifica tion by Drugs, Free University Press, Berlin, 1970, p. 365. 21. Rothenwallner, A.. Zens, P., and Neubert, D. Arch. Pharmacol., in press. 22. Patau. K.: Zur statistischen Beurteilung von Messreihen (eine neue f-Tafel). Biol. Zbl. 63: 152 (1943). .,tem ber 1973 79 784406 GENP011595 i pn*tnatal Effects of Maternal Exposure to 2, ,,7,8-Tetractilorodibenzo-p-dioxin (TCDD] by J.A. Moore,* B.N.Gupta,* J.G. Zinkl,* and I.G. Yos* nous studies reported that subcutane- The 2,3,7,8-tetrachlorodibenzo-p-dioxin Iministration of 2,3,7,8-tetrachlorodi- (>99% purity, Dow Chemical Company, , .-p-dioxin (TCDD) at a dose level of 3 Midland, Michigan) was dissolved in ace Air, in mice on days 6 through 15 of ges- tone and subsequently diluted with at least Miion produced'pups with cleft palates and 9 parts of corn oil. All mice were weighed kidney anomalies ( i) . The C57B1/6 mouse prior to dosing, and the oral dose adminis the most sensitive of the three strains tered computed on the mean weight of the j'sied to the TCDD-induced kidney effects, mice being treated. Control mice received :: -hat almost 10 0 % of the fetuses developed an equivalent amount of 0 .1 ml of an ace k :sy anomalies. The purpose of this pa- tone-corn oil preparation. ; :s to report subsequent studies which Fetuses were removed from their mother .. npt to characterize the teratogenic re- on gestation day 18 and necropsied after ( j use as seen in the C57B1/6 mouse with fixation in Bouins solution. Mice necropsied - .-pwrific emphasis on the nature and signifi in the postnatal studies were processed ac cance of the kidney anomaly. To distinguish cording to standard necropsy procedures. tween delayed kidney development, which Tissues for histologic examination were fixed transient is of diminished significance, in either Bouins or 1 0 % neutral buffered ::k1 irreversible effects, postnatal studies formalin, paraffin embedded and stained "e also conducted. Reversible delays in with hematoxylin and eosin. The incidence anephric kidney maturation in the rat of an abnormality is given as mean average e been described (2 ). percent which is derived by determining its materials and Methods percent incidence in a litter and subse quently computing the mean of these per Inbred mice of the C57B1/6 strain were cents. 'btained from either the Jackson Laboratory, -ar Harbor, Maine, or the AR Schmidt Com- Results 'rny, Madison, Wisconsin, and mated at the The effects of maternal treatment with :stitute. Detection of a vaginal plug indi ted day 0 of pregnancy. All experimental .ice were singly housed in plastic cages and .ilowed free access to food and water. TCDD on fetal palate closure and kidney development are shown in Table 1. TCDD at 3 /ig/kg, administered on gestation days 10 through 13, produced cleft palate in pups 'National Institute of Environmental Health Sci ences, National Institutes of Health, P. 0. Box 12233, Research Triangle Park, North Carolina 27709. from 12 of 14 litters with a mean average incidence of 55.4%. The mean average kid ney and bilateral kidney incidence was 95.1% and 83.1%, respectively, with all Iit- ieptember 1973 81 784407 I t I i ii I! I. I i! Ii ;! ;i I I ,I Ii ! N} I j Oi ; iIi. 1 ters affected. When the dose of TCDD ad ministered on gestation days 10 through 13 was reduced to 1 /g/kg, cleft palate inci dence decreased to a mean of 1.9%. The litter incidence of kidney anomalies per sisted at high levels; unilateral mean aver age pup incidence decreased to 58.9% and the corresponding figure for bilateral effect decreased to 36.3%. When TCDD adminis tration at 1 ig/kg was a single dose admin istered on gestation day 10 , no cleft palates were produced. The average mean incidence of unilateral kidney effects was 3 4 .3 % with anomalies occurring in 16 of 18 litters. Bi lateral kidney effects at this dose occurred in seven litters with a mean incidence of 8 .8 %. No cleft palates or kidney anomalies oc curred in controls. The fetal kidney anomaly is best de scribed as a renal papilla which is markedly reduced in size, or nonexistent in a few cases, resulting in an enlarged renal pelvis (Fig. 1 ). Giveh the stage of kidney develop ment this may reflect a retardation or ab sence of papillae development rather than loss of an already formed structure. Nephron development appears similar to that occur ring in control mice of the same age. The appearance of the affected kidneys resemble an early stage of hydronephrosis. It was fur ther noted that when the renal anomalies were unilateral, the right kidney was af fected 70.5% of the time. This preponder ance of right kidney involvement had a high statistical significance (P<0.01). Figure 1. Transverse section of a hydronephrotic kidney from an 18-day-old C57B1/6 fetus whose mother received 3 tig/kg TCDD on gestation days 10-13. To assess the impact of this kidney change on the ability of a pup to survive in an extrauterine environment, postnatal studies were conducted. Pregnant C57B1/6 mice re ceived 1 /Ag/kg TCDD on gestation day 10 and were allowed to litter. In one study lit ters from TCDD-treated mothers were fos- Table 1. Incidence of cleft palate and kidney anomalies in C57B1/6 fetuses from TCDD-treated mothers.* Treatment days 10-13 10-13 10 10, 10-13 TCDD dose, ng/ kg 3 1 1 0 No. litters 14 16 IS 27 Cleft palate No. Mean affected average litters Yc 12 53.4 3 1.9 00 00 Kidney anomalies No. affected litters Mean average % 14 95.1 15 5S.9 16 34.3 00 Bilateral kidney anomalies No. Mean affected . average Utters <?e 14 83.1 13 36.3 7 8.3 00 *No significant differences in the number of live fetuses, resorptions, fetal weight and maternal weight was found at these dose levels when compared to controls. - 82 Environmental Health Perspectives 784408 \ 'n control mothers; litters from conj.,,1 mice nursed TCDD treated mice. To --anguish a prenatal effect from a post il effect or from a combined prenatal postnatal effect, a reciprocal cross fos;jr study was conducted. The results of -e studies as they affected kidney are de.-.ed in Table 2. It was found that only one .p from a TCDD-treated mother, who ursed an untreated mother, had kidney sions when necropsied on postnatal day 14. . total of six pups, in four of the 14 litters, i nursed TCDD mothers but were born untreated mothers had hydronephrotic levs. In the four-way cross foster study, .ust all kidney lesions occurred in pups :pse mother received. 1 ig/kg TCDD and .nsequently nursed a TCDD treated -.other. Five of 7 litters contained pups with vdronephrotic kidneys with a mean aver se pup incidence of 34%. Almost one half the affected pups had both kidneys ai red. Three pups in two litters, from unated mothers who nursed a TCDD-treated ' r, also had kidney anomalies; one was 'ally involved. One pup from a TCCD- eaced female which nursed an untreated '.other had an affected kidney, as did one of ie control pups. To confirm that exposure to TCDD-treated males during the nursing period was a 'ajor factor in development of renal hyronephrosis in mouse pups, experiments 'here mothers were treated on the day of arturition were conducted. In these studies, nice received 0, 1, 3 or 10 ^g/kg at parturi tion developed hydronephrosis. A mean average of 75%~of the pups developed at least -unilateral hydronephrosis ; a mean average of 40% developed bilateral hydro nephrosis. At the 3 /ig/kg dose, a mean aver age of 71% of pups in three of three litters developed a right unilateral hydronephrosis. At the 1 ftg/k g dose level, a mean average of 1 2 % of pups in five of nine litters developed a unilateral hydronephrosis. Gross and histological studies of affected mice reveal that the lesion produced is a progressive hydronephrosis. Essentially, total atrophy of the right kidney had oc curred in several 55-day-old mice which had nursed a female that received 10 fig/kg TCDD at parturition (Fig. 2). There were Figure 2. Gross photograph of kidneys from 55-dayold C57B1/6 mice: (top row) left and right kid neys from mice who nursed a control mother, (bottom row) kidneys from mice who nursed a mother treated with 10 tig/kg TCDD. Left kidney enlarged and fluid-filled. Right kidneys with marked hydronephrosis and complete atrophy of renal parenchyma (lower right). Table 2. Incidence of hydronephrosis in C57B1/6 mouse pups following exposure to TCDD; mice treated at 1 ng/bg on day 10 of pregnancy.4 Period exposed to TCDD mother In utero Postnatal Yes No No Yes Yes Yes Yes No No Yes No No No. litters 14 14 7 5 5 7 Kidney hydronephrosis -- --............ No. Mean affected average litters % 1 2.0 4 9.6 5 34.0 1 2.2 2 7.4 1 1.7 Bilateral hydronephrosis . --------- No. Mean affected average litters % 00 1 0.8 4 17.0 00 1 0.5 00 x significant differences in pup weight, maternal weight or litter mortality was observed. Septem ber 1973 S3 784409 k Kr E N P 011598 never any indications of a hydroureter ac companying the hydronephrosis. Serial sec tions of some renal pelves at the site- of ure ter entrance did not indicate a lack of patency.. Serial sections at the site of ureter insertion into the bladder also indicated that there was no blockage. There is a weak im pression that the ureter associated with a hydronephrotic kidney has a lumen of small er diameter and that there may be a slight thickening or increase in the epithelial cells of the ureter mucosa. In those mice in which unilateral renal hydronephrosis occurred, the right kidney was affected over 90% of the time. The tendency for unilateral hydroneph rosis to involve the right kidney suggests an anatomical variation which predisposes its involvement at low dose levels. The right kidney in mice is normally superior in posi tion ; the vena cava, due to its close proximity to the right kidney, provides less room for ureter insertion into the kidney hilus. Whether these anatomical characteristics con tribute to this curious pattern of incidence remains a moot point. Failure to mention possible thymus effects in a paper dealing with TCDD exposure would be misleading. Unfortunately, data* are too meager and diverse in source to assemble specific figures for presentation; however, there appears to be a reduction in thymus weight at birth in some pups from females treated with TCDD during preg*nancy. In postnatal studies, there was a dose-related thymus weight decrease when compared to controls in pups weaned from mothers who received 10 or 3 ftg/lcg TCDD at parturition. Definitive experiments t assess thymic effect and possible impair ment of cellular immunity are planned. Discussion The results of these studies clearly cate that exposing pregnant mice to TCDD during the period when progeny are under going metanephric kidney formation or maturation leads to the development of hyd ronephrosis. The incidence of this defect shows a dose-response relationship as scored by the litter, mean average pup, and uni lateral or bilateral kidney incidence. Associa tion with a treated mother during the nurs ing period accounted for the highest inci dence of hydronephrosis in the studies de scribed. Pups probably are being exposed to TCDD by its presence in milk. The postnatal studies subsequent to pre natal exposure to TCDD at gestation day IQ would seem on first examination to discredit the significance of kidney anomalies re corded from fetuses examined on gestation day 18. However, a comparison of the salient features of the prenatal and postnatal kid ney effects reveals striking similarities: (1 ) the lesion in both cases is best described as a hydronephrosis; (2 ) there is a similar dose-response relationship of kidney effects; (3) there is a tendency for right kidney hydronephrosis at low TCDD doses. The failure to demonstrate progressive hydron ephrosis postnatally in pups exposed only in utero is likely a function of dose and Table 3. Hydronephrosis in C57B1/6 pups from mothers treated with TCDD at parturition/ Kidney hydronephrosis Bilateral hydronephrosis TCDD dose, Mg/kg 0 IQ 0 3 0 1 No. litters 4 5 4 3 a 9 Average no. pups/litter 6 fi 6 5.3 8.0 7.7 No. affected litters 0 5 0 3 0 5 Mean average % 0 75 0 71 0 12 No. affected litters 0 3 0 0 0 0 Mean average % 0 40 0 0 0 0 Pups nursing mothers treated with 10 /ig/kg TCDD were reduced in size at weaning. 84 Environm ental Health Perspectives 784410 . of target organ exposure. This con,=`Lis supported by-the findings that the -vatest incidence of hydronephrosis ocin those litters which were exposure v -t'DD both in utero and during the post period. It is hypothesized that the genesis of the hydronephrotic synae due to prenatal or postnatal maternal ..xinsure is similar. Studies which test this a\ ;iothesis are in progress. Summary '[ydronephrotic kidneys were produced in ,,-e pups that nursed a mother treated TCDD during pregnancy or at time of .irition. Variations in kidney develop ment, consistent with hydronephrosis, were observed in fetuses examined at gestation day 18. It was* hypothesized that the pre natal and postnatal kidney anomaly are of the common etiology and that the incidence and degree of hydronephrosis is a function of dose and length of target organ exposure. REFERENCES 1. Courtney, K. D., and Moore, J. A. Teratology studies with 2,4,5-T and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 20: 396 (1971). 2. Woo, D. CMand Hoar, R. M. "Apparent hydro nephrosis" as a normal aspect of renal develop ment in late gestation of rats: the effects of methyl salicylate. Teratology 6: 191 (1972). H G ENP 011600 tem ber 1973 ' 85 784411 ' Ecology of Chlorinated Dibenzo-p-dioxins H . Schwetz,* J.M. Norris,* G.L. Sparschu,* V.K. Rowe,* PJ. Gehring,' i.L. Emerson/ and C.G. Gerbigt `vere toxicological responses have been ociated with certain chlorodibenzodioxins. i ;e of these responses is chloracne, a folli--iosis first associated with skin contaminaion by chlorohydrocarbons in 1899 (i). Serious outbreaks of chloracne-like lesions ..'.-ociated with runaway reactions in the -eduction of 2,4,5-trichlorophenol* occurred Germany in the early 1950's {2). 2,4,5ichiorophenol itself does not cause acne but the contaminants which may be >med in the uncontrolled production of .5-trichlorophenol are - extremely potent icnegens (2 ). 2.3,7,8-TetrachIorodibenzo-piioxin and tri- and tetrachlorodibenzofuran were isolated from the contaminants formed :i 2,4,5-trichlorophenol production and were emonstrated to be strongly positive acneens when applied to rabbit ears (3). By ;sing the rabbit ear test, the acnegenic .)otency of 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) was confirmed in 1962 (4). In addition, 2,3,7,8-TCDD is extremely toxic in the chick embryo assay (5) and is highly embryotoxic in rats (). Another chlorodibenzodioxin, hexachlorodibenzo-pdioxin (HCDD), is known to be positive for the chick edema factor, a condition char acterized by hydropericardium, ascites, and anasarca (o. 7). Chemical Biology Research, The Dow Chemical Co.. Midland. Michigan *18640. f Human Health Research and Development Cen ter. The Dow Chemical Co., Zionsville, Indiana 46077. Experimental Materials The chlorodibenzodioxin samples used in these studies are identified and described in Table 1. Studies were limited in some cases by availability of pure samples. Acute Lethality Samples of 2,7 -dichlorodibenzo-p-dioxin, 2,3,7,8-tetrachlorodibenzo-p-dioxin, hexachlorodibenzo-p-dioxin, and octachlorodibenzo-pdioxin were evaluated for acute oral lethality in several animals as summarized in Table 2. Test materials were administered as sus pensions in corn oil or as corn oil: acetone (9:1) solutions in single doses by gavage. The animals were deprived of feed for 16 hr before dosing. After dosing, they were ob served for signs of toxicity including body weight changes for two to eight weeks. Lethality of 2,3,7,8-TCDD via skin ab sorption was tested on rabbits of mixed sexes with doses of 31.6, 63, 126, 252, and 500 fig/kg body weight. The compound was applied as a 0.0 1 % solution in acetone to the abdominal skin which had been shorn. After the acetone evaporated, the trunk of each rabbit was wrapped in cotton to pre vent ingestion. The rabbits were housed in individual holding cages and were observed for signs of toxicity including body weight changes for three weeks. Parenteral lethality was determined by injecting rabbits of mixed sexes intraperi- eptember 1973 87 784412 Table 1. Purity of samples used in the toxicology studies. Sample no. la lb lc Id 2a 2b 2c 2d Sample identification _ 2,7-Dichlorodibenzo-p-dioxin (2,7-DCDD) #104, shelf 142 AR-5708 340-2-13A 340-2-69A 2,3,7,8-TetrachIorodibenzo-p-dioxin (2,3,7,8-TCDD) Caustic insoluble isolate 1965 851-142-24 Skelly 11/11/64 340--2--o4B Source Dow Chem. Co. Dow Chem. Co. Dow Chem. Co. Dow Chem. Co. Dow Chem. Co. Dow Chem. Co. Dow Chem. Co. Dow Chem. Co. Purity 99.8% 99.6% >99% 96.4% 98% 91% >99% Tests * 1, 2, 3 3 1. 2, 3 4 1 1 1 3, 5 1, 2, 3, 5 1,2,3,4-Tetrachlorodibenzo-p-dioxin (1,2,3,4-TCDD) 3a FDA-F990 FDA 98.5% 3 Hexachlorodibenzo-p-dioxin (HCDD) 4a 252-44-12B-AL22 4b 252-44-12B-AL11 4c 340-2-82A 4d FDA-F911 Dow Chem. Co. Dow Chem. Co. Dow Chem. Co. FDA 65:35, 2 isomers 99%, 65:35, 2 isomers >99%, 89:11, 2 isomers 95.1%, 3 isomers 1, 3 3 1, 2, 3, 4, i 3 Octachlorodibenzo-p-dioxin (OCDD) 5a 251-1-142A 5b 340-2-29A 5c AR-570d 5d 340--2--o7A Dow Chem. Co. Dow Chem. Co. Dow Chem. Co. Dow Chem. Co. 98% 94% 98.86% 1, 2, 3 U3 3 1, 3, 4. 5 *Based on gas-liquid chromatographic (GLC) or GLC-mass spectrophotometric analysis. "Test identifications: 1 = LD*>; 2 = eye irritation; 3 = chloracne; 4 = tetratogenicity; 5 = chick edema. *Photolysis product of sample la. * Photolysis product of sample 5a. Test animal Rat Rat Mouse Rabbit Guinea pig Dog 88 Table 2. Evaluation of acute oral lethality. Strain Sprague-Dawiey Sherman (Spartan) Swiss Webster New Zealand albino Hartley Beagle 2.7DCDD X X Test material 237S TCDD HCDD X X X X X X OCDD X X Environmental Health Perspectives 784413 GENP011602 ---.'ally with 31.6, 63, 126, 252 and 500 i<fr of 2,3,7,8-TCDD aa a 0.01% corn oil pension; control* rabbits were injected corn oil. The rabbits were housed in . wddual holding cages and were observed dgns of toxicity for four weeks. The a were calculated by the Weil modion of the Thompson method (5, 5) or . ae Litchfield and Wilcoxon method (10). acute lethality studies were terminated v.iten it was evident that the survivors were -:rtr showing signs of toxicity. j \ e Irritation ".abbit eyes were examined prior to ex tents and found to be free from defects nutation. Approximately 2 mg of 2,7:>D, 2.3,7,8-TCDD, HCDD, or OCDD were -rilled in the conjunctival sac of one eye; *.* contralateral eye served as a control, 'he eyes were examined at various times -.er treatment for conjunctival redness and -mosis, iritis, and corneal injury. Re fuses were categorized according to in.sity. \bbit Ear Bioassay For Acnegenic Activity Acnegenic activity of 2,7-DCDD, 1 ,2,3,4TCDD, 2.3,7,8-TCDD, HCDD, and OCDD v;\s tested by applying 0.1 ml of either a -`vent solution or the supernatant of a ivent suspension of each compound to the ner surface of the rabbit's ears five days week for four weeks. The ears were exnined weekly for signs of chloracne. in animation and hyperkeratosis. The re ponses were divided into five categories: 1) none, (2) very slight, (3) slight. (4) moderate, and (5) severe. Responses in the first three categories in'iude no response to mild irritation, inreased ear thickness, slight enlargement of he follicular aperture, slight exfoliation *.nd slight crust formation. These responses alone are not considered indicative of chlorucnegenic activity. Categories 4 and 5 are indicative of acnegenic response and are characterized by comedo formation, in creased ear thickness and hyperkeratosis. Teratology Pregnant adult Sprague-Dawley (Spartan strain) female-rats weighing approximately 250 g were used to study teratogenicity of the chlorinated dibenzo-p-dioxins. The day sperm were first present in a vaginal smear was considered day zero of pregnancy. The animals were housed individually in wirebottom cages in a room controlled for tem perature, humidity, light cycle and noise. Commercial laboratory rat chow and water were provided with choice. Corn oil: acetone (9:1) solutions with varying amounts of test material were given in 2.5 ml/kg dosages by gavage. Dosages were calculated using daily body weights. Rats were treated with 100 mg of 2,7DCDD/kg-day, 0.1, 1.0, 10, or 100 & HCDD/kg-day and 100 or 500 mg OCDD/ kg-day on days 6 through 15 of gestation. Control rats received 2.5 ml/kg of corn oil: acetone (9:1) orally. All rats were observed daily throughout pregnancy and were weighed on days 6, 13, and 21 of gestation. Pregnant females were sacrificed by carbon dioxide anesthesia on day 2 1 of gestation; the uterine horns were exteriorized through a midline incision in the abdominal wall, and the number and position of live, dead, and resorbed fetuses were noted. After be ing weighed and sexed, the fetuses were examined for external anomalies; the crownrump length was measured with a vernier caliper. Half of each litter was preserved in Bouin's solution and later examined for soft tissue anomalies (11); the other half was preserved in alcohol, cleared and stained with Alizarin Red-S, and examined for skeletal abnormalities (12). A 2 x 2 contingency table was used to evaluate the frequency of anomalies and resorptions within the fetal population and between litters. Body weight and body mea surements were statistically analyzed by an analysis of variance and Tukey's test (13). In all cases, the level of significance was P<0.05. Chick Bioassay for Chick Edema Factor The bioassay for chick edema factor was ptember 1973 89 784414 conducted according to the Association of Official Agricultural Chemists method (14). Three-day-old white leghorn, single=comb cockerels were used. 2,3,7,8-TCDD, HCDD, and OCDD were the compounds studied. The diet used in the study was formulated speci fically for conducting the chick edema bio assay (Nutritional Biochemicals, Interna tional Chemical and Nuclear Corp., Cleve land, Ohio). Body weights were recorded twice weekly for the oral intubation studies and at the start and termination of the dietary study. The chicks were observed daily for signs of toxicity, and food con sumption was recorded weekly. After 20 or 2 1 days of treatment, all chickens were sacrificed by cervical dislocation and exa mined for gross lesions. The amount .of pericardial and peritoneal fluid was mea sured, and all gross lesions were recorded. If the calculated t was greater than +1.3, the mean logarithm (10 0 x ml pericardial fluid) was greater than 1.1461 for the chicks receiving the' test compound, and the mean logarithm of the negative control was less than 1.1460, the compound was considered positive for chick edema.* Pathology Toxicology studies were not designed to study the pathological changes associated with chlorodibenzodioxin administration, but in some cases, gross pathological and histopathological examinations were per formed. For microscopic examination, tis sues were fixed in 1 0 % buffered formalin and were stained with hematoxylin and eosin. Sections of fetuses of control dams and dams treated with 100 mg 2,7-DCDD/ kg-day were stained with hematoxylin and eosin, hematoxylin-phloxine-saffron, Mas "The calculated mean of logarithms of pericardial fluid voluemes of the test group and of concurrent negative control group x< and x,, respectively, is given b6y = (a, -- x,)/[(s,Vr) + {SeVn)]4*6' where n, and n< are the number of chicks in the test and control groups, respectively, and s,* and are variances of test and control groups, re spectively ( li) . son's trichrome stain, and Mallory's nh photungstic acid-hematoxylin stain. 8" Results Acute Lethality The lethality of 2,3,7,8-TCDD is Pt6. sented in Table 3. The data reveal that the single oral LD50 ranges from 0.0006 mg/kg in male guinea pigs to 0.115 mg/kg in'rab bits of mixed sex. Data on rats indicate that males are more sensitive than females; lethality is essentially the same following intraperitoneal, oral or skin administration for rabbits. Limited data show that dogs are less sensitive to 2,3,7,8-TCDD than rabbits. For female and male mice, single oral doses ranging from 0.001 to 0.130 mg/kg produced a few sporadic deaths without any definitive dose-response relationship; therefore the data are not presented in the table. Limited lethality data are available for 2,7-DCDD, HCDD, and OCDD. HCDD (sample c) killed 1 of 2 and 0 of 2 male rats given oral doses of 100 and 10 mg/kg, re spectively. No deaths occurred in four male mice given 2,0 g/kg of 2,7-DCDD (sample a or b) orally or in two female rats given l g/kg (sample a). For OCDD, oral doses of l g-'kg (sample d) to five female rats did not cause death: in four male mice, doses of 4 g/kg also did not cause death. No signs of toxicity were observed in animals treated with either 2,7-DCDD or OCDD. The only sign of toxicity among animals treated with HCDD was loss of body weight. While all species lost body weight follow ing treatment with 2,3,7,8-TCDD, other signs of toxicity were 3pecies depend ent. Ascites was seen in mice. Anorexia, dehydration, depression, emaciation, intes tinal hemorrhage and alopecia were seen in dogs. Certain rabbits treated intraperitoneally with 2,3,7,8-TCDD developed skin les ions typical of those associated with acnegens. Rabbit Eye Irritation Instillation of the chlorodibenzodioxins in to the conjunctival sac caused slight, trans- 90 Environm ental Health Perspective 784415 Table 3. Lethality of 2,3,7t3-tetrachlorodiben2o-p-dioxin ` p'i.and sex .le I'emale -vxa, pig. male -iwA pig. maIe (bit. mixed male its. female Route of Sample " . administration c Oral c Oral c Oral d Oral c Oral c Skin c Intraperitoneal c Oral c Oral Time of death, days postadminis tration LD, mg/kg 9-27 0.022 13-43 3-34 9-42 6-39 12-22 6-23 0.045 (0.030-0.066) 0.0006 (0.0004-0.0009) 0.0021 (0.0015-0.0030) 0.115 (0.038-0.345) 0.275 (0.142-0.531) -- 9-15 -- Dose, m/kg 0.008 0.016 0.032 0.063 0.032 0.063 0.126 0.252 0.500 0.30 3.00 0.03 0.10 Number deaths/ number treated 0/5 0/5 10/10 5/5 * 0/5 2/5 2/5 2/5 3/5 0/2 2/2 0/2 0/2 Responses to individual doses are given in those cases in which an LD could not be calculated. The '.Dv. for oral administration to rabbits was calculated by using the method of Litchfield and Wilcoxon the remaining values were calculated by using the Weil modification of the method of Thompson IS, IS). jtters refer to sample identification in Table 1. '' pain and conjunctival inflammation, iually. Treatment with 2,3,7,8-TCDD was .-sociated with delayed conjunctival chemo:s 13-22 days later. By day 27, the chemosis -A subsided, but the rim of the eyelid was ickened and encrusted. In rabbits treated :h HCDD, the rim of the eyelid was en:sted 27 days after treatment. Neither rneal injury nor iritis was observed in any >i the animals following instillation of the hlorodibenzodioxins in the conjunctival sac. Acnegenic Response Both 2,3,7,8-TCDD and HCDD produced cne in the rabbit ear bioassay as indicated y the formation of comedones. Solutions of ..3,7,8-TCDD (sample c) in benzene rangng in concentration from 0.04 to 400 fig/m l produced a positive response with severity increasing with concentration. A negative response was obtained with a solution of 0.004 Mg/ml. In contrast, a chloroform solu tion of 1,2,3,4-TCDD, 50 jig/ml, did not Produce a positive response. With HCDD (samples a, b, c, and d), a response was produced by solutions of 10 to 50 jug/ml in chloroform and dimethoxyethane. Chloro form extracts from 10fo suspensions of 2,7DCDD or OCDD were negative, indicating that these have a low order or possibly no acnegenic activity. Teratogenicity The effects of chlorodibenzodioxins on maternal and fetal body measurements, in cidence of fetal resorptions and anomalies are given in Tables 4 and 5. 2,7-DCDD. Rats treated with 100 mg/kgday on days 6 through 15 of gestation gained slightly more weight during pregnancy than controls but showed no toxicity. There was no effect on fetal body measurements, or in cidence of resorptions, or gross, soft tissue or skeletal anomalies. HCDD. Administration of 0.1-100 tg HC DD'leg-day was associated with a doserelated decrease in maternal weight-gain - -ember 1973 91 GENP 011605 784416 during gestation. Gross necropsy examination -at the time of cesarean section revealed'evidence of maternal toxicity only among dams receiving 100 ^g/kg-day (pale, friable liver 3/20 dams; serous atrophy of fat, 1/20 dams). Treatment with 10 or 100 xg HCDD/kgday was highly lethal to fetuses during late gestation. While the incidence of early re sorptions was not increased at any dose level of HCDD (5-7% in the treated versus 7% in the controls), there was a significant increase in late resorptions (0% at 0.1 /g/ kg-day to 79% at 100 ig/kg-day). The weight and length of surviving fetuses were significantly decreased. A significant increase in the incidence of -fetal soft-tissue and skeletal anomalies was seen following treatment of pregnant rats with HCDD at the 100 ig/kg-day dose level. The incidence of cleft palate, sub cutaneous edema, vertebrae with split or unfused centra, and split sternebrae was significantly greater than among control litters or the control fetal population. Among dams treated with 1 or 10 iff/kg-day, only subcutaneous edema occurred at a signi ficantly greater incidence than in the con trol litters or fetal population. Treatment with 0.1 ^g;kg-day of HCDD did not in crease fetal anomalies among the litters or the fetal population. The incidence of de layed ossification of sternebrae was signi ficantly increased among the fetal popula tion but not among litters. OCDD. Signs of maternal toxicity were not observed in rats given 100 or 500 mg/kg-day OCDD. Examination of the fetuses did not reveal changes in fetal body measurements, incidence of fetal resorptions, or incidence of any fetal anomaly among litters or the fetal population. At 500 mg/kg-day, the incidence of subcutaneous edema was signi ficantly increased among the fetal popula tion (23/100 compared with 8 156 in con trols) but not among litters (9.-18 compared with 6/28 in controls). Chick Edema Bioassay Chick edema was produced in groups of birds treated with 1 and 10 /g/kg-day 0f 2,3,7,8-TCDD and 10 and 100 ^/kg-day 0f HCDD (Table 6), The mean logarithm for pericardial fluid volume of the negative control groups was greater than 1.1460 and could negate the results if the guidelines for interpreting chick edema bioassay studies were rigidly followed. However, since the volume of pericardial fluid was markedly in creased by the treatments indicated above, the treatments were considered to be positive for the production of chick edema. A positive response was not observed in chicks main tained on a diet containing 0.5% OCDD. Severe dyspnea, subcutaneous edema, and distended abdomens were observed in some birds receiving 1 or 10 p.g 2,3,7,8-TCDD ' kg-day. Dyspnea and mucus accumulation in the mouth prior to death were observed in birds receiving 100 ng 2,3,7,8-TCDD/kgday. No overt clinical signs were observed in birds receiving OCDD. The gross lesions seen in chicks treated with chlorodibenzodioxins are summarized in Table 7. The most consistent gross lesions were increased pericardial and peritoneal fluid, subcutaneous and pulmonary edema, hepatomegaly and a mottled appearance of the liver. Histopathologic examination of tissues of selected birds from the 2,3,7,8-TCDD ( 1 and 10 pg) and HCDD (10 and 100 /*g) groups revealed similar lesions consisting of: atro phy ox germinal centers of the spleen, a paucity of lymphocytes in the bursa of Fabricius, pulmonary edema, intersititial edema of the myocardium, fatty degenera tion and coagulation necrosis of the liver. Many birds died as a result of pulmonary edema. Pathology Gross necropsy and histological examina tions were conducted on relatively few mam mals treated with the chlorinated dibenzop-dioxins. Therefore, the results reported here are incomplete and preliminary. The liver of animals treated with 2.3,7,8-TCDD and HCDD was most consistently affected. 92 Environm ental Health Perspectives 784417 September 1973 Table 4. Effect of treatment with chlorinated dibenzo-p-dioxin on maternal and fetal body measurements and the incidence of fetal resorption. Test compound (sample) * No. of litters Control 30 2,7-D10ie0h.0lomrogd/ikbge-ndzaoy-p-dioxin (d) 7 Maternal weight gain, g 1 Days 6--13 Days 13-21 Days 6-21 36 2 101 6 137 8 Feta) Fetal - body weight, crown-rump length, mm 6.68 0.05 44.5 0.1 Fetal resorptions, % Population 4 Litter* 7 ( 22/337) 47 (14/30) 31 1 122 4 152 6 5.80 0.09 44.2 0.2 6 ( 5/ 86) 57 ( 4/ 7) llexachlorodibenzo-p-dioxin (c) 0.1 p g /k g -d u y 1.0pg /kg -day 10.0 p g /k g -d a y IS 19 18 10D. p g /k g -d a y 19 28 2 27 3 22 3 ' 6 2* 102 6 99 6 97 5 13 7* 130 5 126 6 119 6 19 9 * 6.73 0.04 6.93 0.16 5.12 0.05 ' 3.65 0.28 * 43.8 0.1 45.7 0.5 42.6 0.2 ' 35.2 0.7 r 5 ( 10/217) 9 ( 20/218) 25 * ( 57/229) 85 1 (194/227) 47 ( 9/19) 74 (14/19) 94 ` (17/18) 100 ' (19/19) Octa1c0h0l.0ormodgi/bkegn-zdoa-py-dioxin (d) 12 500.0 mg/kg-day 17 32 2 35 3 100 8 115 4 131 7 15 6 6.73 0.09 5.69 0.05 43.6 0.4 44.5 0.2 8 < 11/131) 42 ( e / 12) 6 ( 9/199) 41 ( 7/17) Sample identified in Table 1; administered on days 6-15 of gestation as a corn oil: acetone (9:1) solution. *Mean S.E. for vurious gestation times. *Mean of litter means S.E. * % (number resorptions/nuinber implantations). 1aS%ign(nifuicmanbtelry litters with at different from least one reaorption/number litters). control by on analysis of variance and Tukey's test (measurements) or the 2x2 tions), F <0.05. ` contingency table (resorp -d 00 -fc. 00 09II0 dNaO to Table S. Effect of treatment with hexachlorodlbenzo-p-dioxin on the incidence of fetal anomalies. Incidence with treatment on days 6-16 of gestation 0 0.1 pg/Ug-tlay l.Opg/kg-day lOpg/kg-day Soft tissue anomalies Cleft palate Dilated renal pelvis Subcutaneous edema P* 00 ( 0/156) ( 0/ 28) P 0.6 ( 1/156) L P L 4 ( 1/ 28) 216 ( 8/166) ( 6/ 28) 61 00 ( 1/104) ( 1/ 19) ( 0/104) ( 0/ 19) 6 ( 6/104) 82 ( 6/ 19) 00 ( 0/99) ( 0/19) 62 ( 2/99) ( 1/19) 10505 (54/99)* (19/19)* Skeletal anomalies Split vertebral centra P 6 ( 9/168) Split sternebrae L P id 0.6 ( 5/ 27) ( 1/168) L 4 ( 1/ 27) 26 ( 2/103) ( 1/ 19) 61 ( 1/103) ( 1/ 19) Delayed ossification P 11 (18/168) 28 (29/103)* of sternebrae L 44 (12/ 27) 74 (14/ 19) 61 ( 1/99) ( 1/18) 112 ( 2/99) ( 2/18) 12 (12/99) 60 ( 9/18) ` Incidence among fetal population; % (number of affected fetuses/number fetuses examined). *Incidence among litters; % (number of affected litters/number titters examined). ' Significantly different from control by 2 X 2 contingency table, P <0,05. 00 ( 0/86) ( 0/18) 6 ( 6/86)* 17 ( 3/18) 100 (86/86)* too (18/18)* 7 ( 6/86) 29 ( 5/17) 122 ( 2/86) ( 2/17) 34 (29/86)* 71 (12/17) lOOpg/kg-day 47 ( 8/17)* 78 ( 8/11)* ; 12 ( 2/17)' i 18 ( 2/11) 110000 (17/17)* (11/11)* 31 ( 5/16)* 56 ( 5/ 9)` 6361 ( 5/16)* ( 5/ 9)` 56 ( 9/16)* 56 ( 5 / 9) 809II0 dNaO Environmental Health P ersp ectif September 1973 T ab le 6. R e su lts of chick ed em a b io a a sa y ; body w eig h t, food co n su m p tio n , a n d p e ric a rd ia l fluid volum e c a lc u la tio n s o f ch ic k s tr e a te d w ith c h lo ro d io x in s. T reatm en t (sam ple)1 n P e ric a rd ia l fluid volum e Body w eight, g " ---------------------------------------------- F o o d ----------------------------------------M ean log Day 0 D ay 21 c o n su m p tio n , g ` m l S .E . (100 x m j) C alculated (v alu e P ositive fo r chick edem a facto r based on ---------------- ;------------------------------- C alculations G ross lesions 2 ,3 ,7 ,8 -T e tra c h lo rQ d ib e n z o -p -d io x in ( d ) 1* 0 pg/kg 0.01 pg/kg 0.10 #*g/kg 1.0 PE/kg * 10.0 p/kg ' 10 10 10 2 9 46 1 44 1 45 1 42 1 42 1 Hexachlorodibenzo-p -dioxin <c)` 0 0.1 1.0 10.0 100.0 pg/kg pfi/kg pE/kg M g / kg " PE/kg * 9 10 10 9 10 38 1 42 1 43 1 38 I 36 1 Octachlorodihenzo-p-dioxin (d)* 0% of diet 0.1% of diet 0.5% of diet 12 11 11 45 1 45 1 43 1 199 6 196 7 203 7 196 24 No survivors 194 6 197 4 196 6 187 7 No survivors (Day 20) 141 8 124 6 196 10 17.4 16.7 17.2 33.7 11.2 17.4 17.8 18.2 16.7 13.3 13.3 9.5 10.9 * S a m p le id e n tifie d in T a b le 1. * M ean S.E . *Grams/chick/day. * A dm inistered orally a s a corn oil: acetone solution. * A n im als died on d ay s 9, 11, 11, 14, 15, 17, 18, an d 19 of tre a tm e n t. * A nim als died on days 3, 4, 4, 4, 5, 8, 8, 9, 12 an d 15 o f tre a tm e n t. 1 O ne a n im a l died on d a y 19 o f tre a tm e n t. k A n im als died on d a y s 6, 5, 6, 7, 8, 10, 11, 11, 15 a n d 17 o f tre a tm e n t. ' F e d in th e d ie t (0 .1 % === 100 m g /k g , 0 .5 % -- 500 m g / k g ) . 0.16 0.02 0.14 0.02 0.19 0.01 2.34 0.08 1.29 ,0.62 0.15 0.01 0.11 0.02 0.09 0.01 0.81 0.01 0.62 0.24 0.08 0.01 0.06 0.01 0.09 0.01 1.1717 1.1181 1.2688 2.3680 1.5661 1.1771 0.9978 0.9387 1.7294 1.5650 0.8053 0.7889 0.9002 -- -0.74 + 1.69 + 21.7 +1.47 -- -1.93 --4.57 + 3.82 + 2.72 -- -0.21 +0.91 -- No No No No -- No No No No -- No No -- No No Yes Yes -- No No Yea Yes -- No No ~nI 00 -P* -fc. oro cton cCnO Environmental Health Perspectiv l u b l e 7. K c iiU m o f ch ick e d e m a b i o u s a u y : s u m m a r y of g r o s s l e s io n s o b s e r v e d in c h ic k s t r e a t e d w i t h c h l o r o d i o x in s . T reatm ent (sample)* Pericardial Peritoneal M ortality fluid > 0 .2 ml fluid 2,3,7,8-Tetrachlorodibenzo-p -dioxin (d) 0 Mg/kg 0/10 * 0.01 M g/kg 0/10 0.10 Mg/kg 0/10 1.0 Mg/kg 8/10 100 Mg/kg 10/10 2/10 1/10 2/10 10/10 6/10 0/10 0/10 0/10 9/10 9/10 llexachlorodibcuzo-p-dioxin 0 Mg/kg 0 1 /<B/kg 1.0 Mg/kg 10.0 M g/kg 100.0 M g/kg (c)' 1/10 0/10 0/10 1/10 10/10 0/10 1/10 0/10 9/10 6/10 0/10 0/10 0/10 3/10 3/10 Octuchlorodibenzo-p dioxin 0% of diet 0.1% of diet 0.6% of diet (il)4 0/12 0/12 0/12 0/12 0/12 0/12 0/12 0/12 0/12 ` S a m p le identified in T ab le I. N um ber alTccted/total num ber in group. * A dm inistered orally as a corn oil; ucetone solution. 4 F ed in tlx: diet (0.1% = 100 m g /k g , 0.6% = 600 m g /k g ) . Subcutaneous Pulm onary A trophy of edem a cdeina spleen an d /o r bursa 0/10 0/10 0/10 9/10 9/10 0/10 0/10 0/10 5/10 5/10 0/10 0/10 0/10 2/10 0/10 0/10 0/10 0/10 1/10 8/10 0/10 0/10 0/10 1/10 8/10 0/10 0/10 0/10 1/10 0/10 0/12 0/12 0/12 0/12 0/12 0/12 0/12 0/12 0/12 L iv e r swollen a n d /o r m ottled 0/10 0/10 0/10 7/10 6/10 0/10 0/10 0/10 3/10 4/10 0/12 0/12 0/12 G izzard erosions 0/10 0/10 0/10 1/10 0/10 0/10 0/10 0/10 0/10 0/10 2/12 7/12 7/12 019110 dHHO .,-^.scopic examination of this organ rea highly variable pattern and degree '! :v patic necrosis with various degrees of ,. -ration and regeneration of the hepato\ depending upon the post-treatment in- .. Necrosis was observed both in the . ilobular and periportal areas. The de,,n v of necrosis of the liver was not sufficient i-cmclude that it was responsible for death. [Tepatic lesions were observed in rats, mice, .:(i>bits, and dogs. In addition to hepatic :;-volvement, other changes observed sporadv '!> include fat necrosis, periarteritis, seritrophy of fat, and ascites. C :cussion and Summary The studies reported here confirmed the :,::rh toxicity of 2,3,7,8-TCDD. In addition, ome perspective of the relative toxicities t>: 2.7-DCDD, HCDD, and OCDD has been .rained. 2,7-DCDD and OCDD failed to .ise death in female rats given orhl doses I g /k g ; even larger doses were given to ce without causing death. Limited data est that oral doses of approximately m g/kg of HCDD are needed to cause ieath in male rats. In the teratology study, no deaths occurred following administration f 100 /ig. kg of HCDD to female rats for 10 onsecutive days. 2.3,7,8-TCDD is much more toxic than he other chlorodibenzodioxins studied; the ,Dsn ranged from 0.6 n g / k g in male guinea pigs to 115 ig/k g in rabbits. Dogs appear to be less sensitive than rabbits. Others have reported 1 0 0 % mortality in rabbits treated with 10 ig/k g (15) and chick embryos treated with 0.05 jig/egg (5). Death following treatment with a lethal dose of 2,3,7,8-TCDD is often delayed for several weeks. Among the animals which died following treatment, approximately half the deaths occurred between 13 and 18 days after treatment, with one animal dying as late as 43 days after a single oral dose. In mice and rabbits, there is a marked in dividual difference in susceptibility to this compound which makes it difficult to con-`T.t acute lethality studies. ,f the results of the rabbit eye irritation test can be extrapolated to man, accidental contact o f. these" chlorodibenzodioxins with the eyes should not present a serious threat to vision. However, repeated contact with the skin of small amounts of either 2,3,7,8TCDD or HCDD may be expected to produce chloracne. Sensitivity to 2,3,7,8-TCDD was recognized by industry years ago, and pre cautions have been taken to minimize its occurrence and prevent contamination of worker's skin. HCDD is apparently a less potent acnegen than 2,3,7,8-TCDD. As previously reported, 2,3 ,7 ,8-TCDD is highly embryotoxic (5). The no-effect level for embryotoxicity was 0.03 ig/kg-dav of 2,3,7,8-TCDD. In contrast to the high em bryotoxicity of .the symmetrical 2,3,7,8TCDD, 1,2,3,4-TCDD was not embryotoxic at doses as high as 800 ^g/kg-day (16). By previously described definitions of ter atogenicity and embryotoxicity (17), HCDD is teratogenic in the rat at a 100 /ig/kg-day dose level, given orally on days 6 through 15 of gestation. Treatment of pregnant rats with HCDD caused embryotoxicity evid enced by a dose-related decrease in fetal body weight and crown-rump length and an increase in the incidence of fetal resorp tions (Table 4). Likewise, the incidence of certain soft tissue and skeletal anomalies increased in a manner related to the dose level of HCDD (Table 5). A 0.1* /ig/kg-day dosage of HCDD had no effect on embryonal or fetal development. OCDD caused embryotoxicity but was not teratogenic at 500 mg/kg-day. OCDD and 2,7-DCDD caused neither teratogenicity nor embryotoxicity at 100 mg/kg-day. Khera and Ruddick (16) reported that the ad ministration of 2 mg 2,7-DCDD/kg-day was associated with microscopic myocardial and pericardial lesions in rat fetuses. However, examination of sections of myocardium and pericardium from fetuses of dams treated with 100 mg doses in this study revealed no morphological differences from controls. Both 2,3,7,8-TCDD and HCDD give posi tive results in chick edema bioassays (Table 6 ). This HCDD result is consistent with a previous report that- the HCDD isolated Septem ber 1973 97 G E N P 01161 784422 from pentachlorophenol produced chick - -edema (5 ). These same authors reported that 2,3,7,8-TCDD was extremely toxic in the chick embryo assay but did not report that it produced chick edema. Pathological changes observed in animals treated with chlorodibenzpdioxins were in consistent from animal to animal and species to species. Hepatic lesions were observed consistently, but the nature, degree, and dis tribution of the lesions were variable. Changes in organs other than the liver were sporadic and unpredictable. Gross and mic roscopic examination of tissues after chlorodibenzodioxin treatment did not reveal the cause of death. An in-depth evaluation of the toxicity associated with chronic ex p o su re to the chlorobenzodioxins is needed. Isomers of a chlorodibenzodioxin can produce different degrees of toxicity; 2,3,7, 8-TCDD is highly embryotoxic and a potent acnegen, but 1,2,3,4-TCDD is neither em bryotoxic nor acnegenic. The toxicity of chlorodibenzodioxins other than those evaluated in this study has not been reported. Purified samples of trichIoro-, pentachloro-, and heptachlorodibenzop-dioxin which are free of tetrachloro- and hexachlorodibenzo-p-dioxin need to be syn thesized for study. However, heptachlorodibenzo-p-dioxin cannot be highly toxic, since studies on octachlorodibenzo-p-dioxin containing several per cent of heptachlorodibenzo-p-dioxin have tested the same as the pure product. Studies on the chlorodibenzodioxins have led to the following conclusions: (1) 2,7dichlorodibenzo-p-dioxin and octachlorodibenzo-p-dioxin have a low acute toxicity; (2 ) 2,3,7,8-tetrachlorodibenzo-p-dioxin has an unusually high toxicity; (3) hexachlorodibenzo-p-dioxin is highly toxic but less toxic than 2,3,7,8-tetrachlorodibenzo-p-diox in; (4) all chlorodibenzodioxins are not alike in their toxicological properties. Iso mers of the same dibenzo-p-dioxin vary in toxicological properties, making it impor tant to identify them specifically. Acknowledgement The authors are grateful to J. E. Bourne P. A. Keeler and R. W. Lisowe for their assistance in all aspects of this study. REFERENCES X. Herxheimer, K. ber chlorakne. Mnch. Med. Wochenschr. 46: 278 (1899). 2. Hofman, H. Th. New experiences with highly toxic chloro hydrocarbons. Archiv Exper. Pathol. Pharmacol. 232: 228 (1957). 3. Kimmig, J., and Schulz, K. H. Berufliche Akne (sog. Chlorakne) durch chlorierte aromatische zyklische ther. Dermatologica 115: 540 (1957). 4. Jones, E. L., and Krizek, H. A technic for test ing acnegenic potency in rabbits, applied to the potent acnegen, 2,3,7,8-tetrachlorodibenzo-pdioxin. J. Invest- Dem. 39: 511 (1962). 5. Higginbotham, G. R., et al. Chemical and toxi cological evaluations of isolated and synthetic chloroderivatives of dibenzo-p-dioxin. Nature 220: 802 (1968). 6. Sparschu, G. L., et al. Study of the terato genicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Food Cosmet. Toxicol. 9: 405 (1971). 7. Anonymous. Search for chick edema factor. Chem. Eng. News 45: 10 (Jan. 30, 1967). 8. Thompson, W. R. Use of moving averages and interpolation to estimate median effective dose. Part 1. Fundamental formulas, etc. Bacteriol. Rev. 11: 115 (1947). 9. Weil, C. S. Tables for convenient calculation of median effective dose (LD50 or ED50) and in structions in their use. Biometrics 8: 249 (1952). 10. Litchfield, J. T., and Wilcoxon, F. A simplified method of evaluating dose-effect experiments. J. Pharmacol. Exp. Therap. 96: 99. (1949). 11. Wilson, J. G. Methods for administering agents and detecting malformations in experimental an imals. In: Teratology Principles and Techniques. J. G. Wilson and T. Warkany, (Eds.), Univer sity of Chicago Press, Chicago, 1965, p. 262. 12. Dawson, A. B. A note on the staining of the skeleton of cleared specimens with Alizarin RedS. Stain Technol. 1: 123 (1926). 13. Steel, R. G. D. and Torrie, H. H. Principles and Procedures of Statistics. McGraw-Hill, New York, 1960, pp. 73, 81, 347, 349, 366. 14. Horwitz, W., Ed., Official Methods of Analysis. 10th ed. Association of Official Agricultural Chemists. Washington, D.C., 1965, Sections 26.087-26.091. 15. Milnes, M. H. Formation of 2,3,7,3-tetrachIorodibenzodioxin by thermal decomposition ox so dium 2,4,5-trichlorophenate. Nature 232: 395 <1971). 98 Environmental Health Perspectives 784423 / -J*a> K. S., and -Ruddick, J. A. Polychlorodibenzo-p-dioxihs: Perinatal effects and domi nant lethal test in Wistar rats. Advan. Chem. Ser. 121, R. F. Gould, Ed., American Chemical Society, Washington, D.C., in press. Schwetz, B. A., Sparschu, G. L., and Gehring, P, J. The effecF'of 2,4-D and esters of 2,4-D on rat embryonal, foetal and neonatal growth and development. Food Cosmet. Toxicol. 9: 801 (1971). GENP 011613 Septem ber 1973 99 784424 General Biological Effects of TCDD a Laboratory Animals In M.W. Harris,' l . i . M o m ,' I .t . Vos,' sod B.H. Sopta' 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCD) is reported to be one of the most -.oxic chemicals known. There has been vari ation in such reported toxic parameters as single oral LDTM dose, range of time interval from dosing until death, and toxic manifes tations which an animal exhibits (l - A ). It seemed most appropriate, therefore, that he mean values and ranges 6f these and ther general biological parameters be deermined under the experimental condition to which animals are exposed at this Insti tute. Materials and Methods The TCDD utilized in the studies was graciously supplied by the Dow Chemical Company and was found by analyses to con tain more than 99% TCDD. The majority of the studies reported by NIEHS scientists at this conference utilized a single stock solution prepared by dissolving 5 rag of chemical in 67.5 ml of reagent grade ace tone. Once dissolved, this TCDD acetone solu tion was added to 432.5 ml of corn oil pur chased at a local retail store. This solution was calculated to contain 10 fig TCDD/ml. Chemical analysis of an aliquot of this solu tion performed by the Dow Chemical Com pany yielded a value of 11 fig/m l 2 pg/ml. This compares favorably with the calculated value. By using appropriate dilutions of the 'National Institute of Environmental Health Sci ences, National Institutes of Health, P.O. Box 12233, Research Triangle Park, North Carolina 27709. stock solution with corn oil, the actual dos ing solution was prepared in a volume ap propriate for the requirements of the experi ment. All animals used, rats, guinea pigs, and mice, were housed in animal quarters main tained under a rigid sanitary regimen. Tem perature was maintained at 70 2F and 50 5 % RH. Food and water were avail able at all times. All animals were adminis tered the TCDD-acetone-corn oil solution via gastric intubation. The volume adminis tered ranged-from 0.2 ml to 1.6 ml in rats, from 0.17 to 0.22 ml in guinea pigs, and from 0.1 to 0.2 ml in mice. The volume ad ministered in any one experiment was the same. Control animals received an equal amount of acetone-corn oil, the actual pro portion of each used being equal that con tained in the TCDD solution. All animals were weighed at least once weekly. Rat food consumption as measured by disappearance of the blocks from a suspended stainless steel feeder was determined at least twice each week. Rats and mice were dosed accord ing to a mean average weight of the dose group. Guinea pigs were dosed according to the weight of each animal. In multiple-dose experiments, dose was recalculated each week based on the weights of the animals on the date of recalculation. All animals were as signed to a given dose group according to a table of random numbers. Rats used were of both sexes, 6 to 8 weeks of age, specific pathogen-free, and had been September 1973 101 784425 G ENP01I614 i i i acclimated to NIEHS conditions at least 2 weeks prior to use. All were random bred albino rats, CD stock, purchased from Charles River Breeding Laboratories, Wil mington, Massachusetts. Female albino guinea pigs, Hartley strain, were purchased from Carworth, Inc., New City, New York. Mice used were of two types, random bred albino CD-I stock from Charles River or inbred C57Bl/6Sch strain reared at the In stitute. The data obtained in the four experiments are presented as mean values and standard errors or deviations. Dunnett's multiple comparisons test (5) was used to make treatment control comparisons, usually twosided, except for organ weights. In addition, a nonparametric test (Jonckheere's test) (5) was used to test for monotonic doseresponse relationships. 5 | | * 1-- _1_ . --1- . 0 I 23435733 Wm ii F ig u r e 1. Body weights of female rats receiving a single dose of ( ) 0; (1) 1.0 Mg/kg; (O) 5.0 Mg/kg; (A) 25.0 Mg/kg. TCDD: There were 3 animals per dose group. All animals survived. Results Live Animal Effects In the course of various rat experiments the frequency of dose administration varied from a single administration to daily dosing for 30 consecutive days or weekly admin istration for a period of 6 weeks. The ef fects of a single dose of TCDD at 1, 0, 25, 50, or 100 p g /k g on body weight and sur vival in rats are illustrated in Figures 1-3; the actual body weights and statistical val ues are given in Tables 1-4. Male and fe male rats at the 1 or 5 ig/k g dose gained weight at the same rate as the controls At the 25 /ig/kg dose an actual weight loss was observed (for 1 week) in females (Fig. 1, Table 1). Subsequent to week 1 until the end of the experiment (week 9) animals at this dose gained weight at the same rate as controls. Male rats at the same dose had a significantly decreased weight gain 1 and 2 weeks subsequent to the TCDD dose (Fig. 2, Table 3). Their weight gain after 2 weeks was equivalent to control rats. In a later experiment (Fig. 3, Table 4), male rats showed a dose-related decrease in body weight gain in the first two weeks subse quent to a 25 or 50 i g / k g dose with the reduced weight gain observed at the higher 102 F ig u r e 2. Body weights of male rats receiving a single dose of TCDD: ( ) 0; (a ) 1.0 Mg/kg; (C) 5.0 Mg/kg; (A) 25.0 Mg/kg. There were 8 animals per dose group. All animals survived. F ig u r e 3. Body weights of male rats receiving a single dose of TCDD: ( ) 0; (a ) 25 Mg/kg; (O) 50 Mg/kg; (A) 100 Mg/kg. The number of animals is shown in parentheses. In the 100 Mg/kg group, the mean weight of the animal surviving the ex perimental period is also plotted ( ). The super script 1 denotes days on which one animal died. Environmental Health Perspectives G N p 0 116 15 784426 Table 1. Weight gain of female rata receiving a single dose of TCDD.* Single TCDD dose, ftgA g Number of animals 0S 1.0 8 5.0 8 25 8 *Mean values SD. bP < 0.01. Initial body weight, g 181.5 7.2 183.8 7.1 194.3 6.9 6 187.3 5.7 Weight gain, g Week 0-1 20.3 5.5 18.3 2.3 14.3 5.2 -6 .4 11.4" Week 1-9 85.3 16.8 81.3 12.2 S2.S 13.6 86.5 20.6 Table 2. Weight gain of female rats receiving a single dose of TCDD.* Single TCDD dose, .g/ kg 0 50 100 Number of animals 6 6" 6* Initial body weight, g 166.0 1.8 162.2 8.8 168.8 4.9 ` Mean values SD. 6One animal died on day 14. t aThree animals died between 18 and 21 days. dP <0.01. Weight gain, S Week 0-1 Week 1-6 22.5 7.8 -11.5 14.8 * -20.5 12.8d 57.2 12.3 66.2 11.4 32.7 31.8 Table 3. Weight gain of male rats receiving a single dose of TCDD.* Single TCDD dose, fig/kg Number of animals 08 1.0 8 5.0 8 25 8 *Mean values SD. *P < 0.01. Initial body weight, g 209.1 7.7 211.0 12.1 212.8 9.7 204.0 5.6 Weight gain, g Week 0-2 Week 2-8 115.3 13.0 107.0 13.6 107.6 18.5 72.3 23.61 184.4 30.8 169.5 23.1 172.9 29.3 182.1 36.0 GENP011616 Table 4. Weight gain of male rats receiving a siiigle dose of TCDD.* Single TCDD dose, ng/kg Number of animals Initial body weight, g 0 10 153.6 9.8 25 4 158.8 2.5 50 8 163.6 9.2 100 8 6 161.0 9 3 *Mean values -- 3D. "Three animals died between 14 and 21 days. " < 0.01. Weight gain.>g Week 0-2 Week 2-5 100.7 14.1 66.5 39.5 58.0 30.5* -9 .9 30.4* 113.9 22.6 122.5 27.7 1 1 Z 8 22.1 97.6 51.6 jm ber 1973 103 784427 dose significant at the 1 % level. At.the 100 t t g / k g dosage, 3 of 8 rats died between 14 and 21 days. The mean weight gain of the survivors at the high dose between weeks 2 and 5 was not significantly different from other TCDD dose levels or controls. Female rats which received 50 or 100 f t g / k g ex hibited a dose-related absolute decrease in body weight during the first week (Table 2 ). One of six and three of six rats died be tween 14 and 21 days at the 50 and 100 fig/k g dose, respectively. Weight gain be tween weeks 1 and 6 was not significantly different from that of controls; however, as the large standard deviation a t'the high dose suggests, appreciable variation among the survivors did exist. The mean time interval until death was 18.3 days in the 6 of 14 male and female rats which died at the 100 f t g / k g dose. Five of six deaths occurred between the 18th and 21st day* the female rat at the 50 t g / k g dose which died, survived 14 days. Clinical symptoms of toxicity, which were seen at the 100 t g / k g dose, were ruffled hair coat, hunched posture, and inactivity (depression). Jaundice was observed for sev eral days in those rats that died. Food con sumption in male rats which received 25, 50, and 100 f t g / k g was recorded and analy zed for the initial 2-week period subsequent to TCDD intubation (Table 5). Although the average amount of food consumed de creased in all treatment groups when com pared to controls, the difference is significant only at the 100 f t g / k g dose (P < 0 .0 5 ). If the food consumption of the three rats which subsequently died in the 100 t g / k g group is excluded and the mean recomputed, the value is no longer significantly different from controls. The standard deviation f0r food consumption in all TCDD treatment groups is relatively large which suggests marked variation within a dose group. In the daily dose experiments, female rats ' received either 0, 0.1, 1, or 10 f t g / k g TCDD for 31 consecutive days. The body weight changes which occurred in one of these ex periments are given in Table 6. Although not included in the table because it was not part of this particular experiment, the ef fect on body weight in rats receiving daily TCDD at 0.1 f t g / k g was negligible. Rats at the 10 f t g / k g dose lost 21.8 g weight during the first 7 days of the experiment. Fifteen of 16 rats died or became moribund a mean of 21.8 days after the study commenced. Weight gain at the 1 f t g / k g dose group was significantly less than that of controls for the 1-35 day time period. Weight gain dur ing the 35-63 day time period exceeded that of controls by 14 g. Table 5. Fourteen-day food consumption in male rats receiving a single dose of TCDD. Single TCDD dose, Mg/kg 0 25 50 100 100 Number of animals 10 4 8 8 5* Food consumption, . g - SD * 359.4 19.3 ,313.2 70.8 315.6 56.4 257.56 87.2 275.0 84.7 *Dose response test for food consumption sig nificant at 0.01 level. <0.05. t Three rats in group which later died excluded. Table 6. Weight gain of female rats receiving 31 daily doses of TCDD. Daily TCDD dose, Mg/kg 0 1 10 Initial no. of animals 16 12 16 g SD 183.3 6.28 185.6 5.04 184.0 s 8.00 Days 1-7 15.8 s 6.6 12.1 6.4 -21.4 * 14.9B Weight gain, g SD Days * 1-35 Days* 35-63 65.8 8.0 34.8 12.2b -- 28.4 9.7 42.0 11.9 e -- *Values based on 5 and 8 rats at 0 and 1 Mg/kg respectively; 15 of 16 rats died or were killed when moribund at mean of 21.8 days. *P < 0.01. ' P <0.06. 104 Environmental Health Perspectives 784428 Tigvre 4. Body weight of guinea pigs receiving *.*ight weekly doses of TCDD. There were 10 ani mals per dose group. All animals at the 1.0 /ig/kg dose died. Rats also received weekly doses of 0 .02, 1.0, or 5.0 /xg/kg for 6 weeks. During the losing period, decreased body weight gain occurred in the 5.0 /xg/kg dose group (7). Guinea pigs were dosed weekly for 8 weeks with 0, -0.Q0.8, 0.04, 0.2, or 1.0 p g / kg. In other experiments, guinea pigs received a single 1.0 or 3.0 /xg/kg dose of TCDD. Nine of ten animals died a mean 19.2 days after receiving the single 3 /xg/kg dose. Severe weight loss preceded death. No guinea pigs died following the single 1,0 fig/k g dose, although a reversible decreased body weight gain was observed. All guinea pigs receiving the 1.0 /xg/kg weekly dose 24-32 days after the study be gan died (mean survival time 28 days). Animals which received the weekly 0.2 /ig/kg dose weighed significantly less at the end of the experiment (7). As Figure 4 il lustrates, weight gain depression at this dose level primarily occurred during the fifth and sixth week of the study. Adult CD-I mice received a single oral dose of TCDD at 0, 1.0, 10, or 50 p g / kg. No effect on body weight was observed during the subsequent 5-week observation period. Adult C57B1/6 mice received weekly doses of 25, 5, 1, or 0.2 /xg/kg for four weeks. Significant weight loss occurred at the 25 /xg/kg dose, with one of seven mice dying on the day 25 of the study. Organ Weights In some experiments, rats were killed and organ weights recorded at predetermined intervals after a single 0, 5, or 25 /xg/kg TCDD dose. These organ weights from ran domly selected rats were obtained 1, 3, 8, 9, 16, and 28 days after TCDD intubation. Table 7. Liver and thymua weight of male rata receiving a single dose of TCDD. Time since dose, days 1 3 3 9 16 28 No. of animals 5 5 5 5 5 5 0 fig/kg 5 ng/kg 25 ag/kg --------- --------------------- -- -------------------------------------------------------------------------- Liver, Thymus, Liver, Thymus, Liver, . Thymus, g i SE mg SE * g SE mg s SE* g : SE mg SE * 9.02 2= 1.15 934 to 53 11.06 * 0.44 364 to 62 13.65 * 0.88 826 to 99 14.96 to 0.47 336 to* 107 16.22 -to 0.70 790 5 57 17.99 to 0.95 764 -to 60 10.34 2S 0.76 11.94 -to 0.34 14.08 to 0.62 14.28 -to 0.84 16.40 -to 1.27 15.61 -to 0.93 366 toe 89 724 -to 108 644 to 65 646 to* 62 570 to 55 * 664 to" 45 8.65 -to 0.67 11.14 to 0.53 13.92 to* 0.59 13.38 to 0.81 14.28 1.99 15.72 to* 1.60 768 to 29 568 * 36 334 to 25 280 to 30 264 to 45 494 to 54 ` Dose-response tests for reduced thymus significant on each day at P <0.05. ' P <0.05. P < 0.01. September 1973 105 GENP 011618 784429 OS Table 8. Liver and thymus weights of females rats receiving daily doses of TCDD. 0 Mg/kg 0.1 *g/kg No. of doses 3 6 10 13 17 24 31 No. of animals 4 4 4 4 4 4 4 Liver, g SE 7.80 0.48 8.41 0.36 3.14 0.21 7.90 0.31 9.26 0.42 9.45 0.31 9.70 0.49 Thymus, mg SB -- -- -- 630 48 615 69 525 43 668 87 Liver, g SB 8.62 0.41 10.35 0.46 b 10.78 0.41 * 10.17 0.94 ` 10.91 0.37 b 11.40 1.05 11.78 0.81 Thymus, mg SB * _ -- -- 492 81 468 22 412 28" 476 49 *All dose response tests for thymus significant at P <0.01. * P < 0.06. *P <0.01. 1.0 >*g/kg Liver, g SE 9.32 0.45 9.00 0.14 11.63 0.40* 9.76 0.62 11.33 0.29' 11.39 0.98 12.84 0.84 * Thymus, mg SE ` _ -- -- 390 58 270 11 372 28b 260 41* 10 /igAg Liver, g SE 9.58 0.43 10.84 0.97 10.63 0.36 * 9.80 1.08 7.62 1.37 4.67 0.21* -- Thymus, mg SE .-- -- -- 142 12* 110 48* 32 10* -- i - Environmental Health Perspective Table 9. Liver and thymus weights of CD-I mice treated with a single dose of TCDD,a Time since dose . weeks 1 3 5 No. of animals 4 3 3 0 /igAg Liver, g SD 1.68 0.16 1.89 0.38 1.77 0.17 Thymus, mg SD 82.5 19.8 81.7 8.1 51.0 18.0 1.0 pg/kg Liver, g SD 1.65 0.16 2.05 0.26 1.61 0.13 Thymus, mg SD 76.5 16.3 83.3 3.2 47.0 8.6 10/*g/kg Liver, g SD 1.90 0.26 1.94 0.41 1.63 0.10 Thymus, mg SD 74.2 19.2 60.8 * 6.9 64.7 6.7 50 pgAg Liver, g SD 2.27 0.16 2.43 0.19 1.83 0.04 Thymus, mg SD 60.8 17.7 60.7* 14.2 38.7 22.0 ` Dose-response test: P <0.01 (liver) at week 1; P <0.05 (thymus) at week 3. < 0.01. *P <0.05. 6I9II0dN 3O ( > e y , ^ear^> lung, spleen, adrenal, testes i thymus weights were recorded and eval uated. Significant differences in organ , weights were seen with the thymus. Wet -eights for this organ and liver are given i Table 7. There is a tendency for thymus -eights to decrease with age. A significant eduction in thymus weight was first ob served 3 days after rats received 25 ig/ kg TCDD. Thymus weights at this dose continued to decrease with the lowest weights observed on day 16. The day 28 mean value was 220 mg higher than day 16 mt still far below control values. The only ignificant (P<0.05) decrease in thymus veight in the 5 ^g/kg group occurred at lay 16. However, the average thymus weight reduction at this dose level for the entire 28-day period is significant (P < 0.01). Liver weights were not significantly differ ent from controls at either dose level. Re duced spleen weights were observed on days 9, and 16 at the 25 /tg/kg dose. Organ weights were also collected at pre determined times in some rats being in flated daily with 0, 0 .1, 1.0 , or 10 ^g/kg. 'er and thymus weights for these doses are illustrated in Table 8. Mean liver - weights were always greater at the 0.1 and l /ig/kg dose; however, the statistical sig nificance of these increases was inconsist ent. At the daily 10 /ig/kg dose livers were increased in size on the first four days organ weights were collected (days 3, 6, 10, and 13). There was a marked decrease in liver weight, falling below control mean values, on days 17 and 24. It should be remembered that body weight at this dose level markedly decreased and death occurred in 15 of 16 rats. Thymus weights were 20-25% of con trol values in the daily 10 /ig/kg dose group when recorded on days 13 and 17. Weights were further reduced at death. At the 1 , /*ff/kg level, thymus was first found signifi cantly smaller than controls at day 17; it continued to remain so through the 31-day dosing period. Thymus weights at the 0.1 Mg/kg dose were significantly reduced when evaluated for an overall dose response ef- -'s e t. Weight changes were not seen in heart, lung, or kidney. Weight decreases occurred in spleen at the high dose group as they approached death. Pathologic, hematologic, and clinical chemistry changes observed in rats, mice and guinea pigs used in these experiments are reported elsewhere (5, 6 ) . Experiments which measured humoral or cell-mediated immune capabilities are the subject of an other report (7). Table 9 lists the liver and thymus weights of CD-I mice 1, 3, or 5 weeks after a single dose of TCDD. Liver weight was signifi cantly increased after 1 week in mice which received 50 ^g/kg TCDD. Thymus weight was decreased in the 10 and 50 /g/kg groups at week" 3. The measurement of an overall dose response was significant for the liver increase at week 1 and thymus de crease at week 3. No weight effects on spleen or adrenal were observed. Guinea pig liver, kidney, thyroid, and uterus weights collected after 8 weekly doses of 0.2, 0.04, or 0.008 /g/kg TCDD were not significantly lower than controls. Decreased thymus weight was observed at the 0.04 and 0.2 /g/kg dose level (Table 10). Marked thy mic atrophy was also seen at time of ne cropsy in those guinea pigs which died in the weekly 1.0 /g/kg or single, 3.0 /g/kg dose groups. Table 10. Liver and thymus weights of female guinea pigs receiving 8 weekly doses of TCDD and killed after 56 days. Weekly dose of TCDD. **g/kg 0 0.008 0.04 0.2 No. of animals 10 10 10 J Liver giSD 29.4 3.8 29.5 4.3 26.9 i 3.6 26.9 s 3.9 1Thymus dose response P <0.01. " P <0.05. ' P < 0.01. Thymus, mg s SD ` 901 246 741 163 672 ~ 161 b 476 70' Discussion Table 11 summarizes and contrasts some TCDD effects seen in the 3 species tested. .ptem ber 1973 107 I 784431 Table 11. Summary of biological effects of TCDD. Rats ______ Guinea pigs___________ Mice Dose, pg/kg MTD, days ` Mortality6 Dose, ftg kg MTD, days * Mortalityb Dose, ^g/kg Lethal dose Single Weekly Daily 100 4 X 25 10 Body weight Lowest dose effect Single Weekly Daily No effect Single Weekly Daily 6X 30 X 6X 30 X 25 5 1 5 1 0.1 Thymus weight Lowest dose effect Single Weekly 6 X Daily 30 X 5 5 0.1 18 28 22 6/14 2/10 15/16 3 5X1 -- 18 28 - 1 8 X 0.2 -- -- 8 X 0.04 -- - -- 8 X 0.04 -- 9/10 10/10 >50 >4 X 25 4 X 25 -- 50 4X 5 -- 10 4x5 -- *Mean time to death after first exposure. " Number of animals dying/number of animals treated. Guinea pigs are most sensitive to the lethal effects of TCDD with 90% dying from a sin gle 3 f i g / k g dose. In contrast, a 100 f t g / k g dose was lethal to 6 of 14 rats. In both species the time interval until death is simi lar and a large weight loss over a period of days preceded death. We did not observe a great range in the time interval from ex posure until death which was observed by others ( 2 ) . Female rats appear to be more sensitive to TCDD than males. Although death inci dence at 100 f t g / k g was similar, one female did die at the 50 f i g / k g dose. Additionally, at the 25 and 50 f i g / k g dose females under went an actual weight loss, whereas males exhibited only a 30-40% reduction in weight gain. Body weight effects of a sin gle dose are primarily observed during the first two weeks after exposure. Subsequent ly, weight gain equates with controls, al though treated rats never narrow the ac tual weight divergence which appears dur ing the first 2 weeks. There is an overall tendency for feed con sumption to be reduced following TCDD exposure, but this depression is not suffi cient magnitude to account for the body weight changes which occurred. Food con sumption at the 100 / t g / k g dose was signifi cantly decreased primarily due to the greater anorectic state observed in rats for the several days preceding death. Administration of daily or weekly sublethal doses does not seem to raise the thres hold level of TCDD toxicity. For example decreased weight gain at the 1 f i g / k g daily dose occurred in rats once the total dose administered exceeded approximately 20 f i g / kg; decreased weight gain first occurred at the single 25 ^g/kg level. A comparison of the body weight gains of 5 f i g / k g weekly dose and the 1 i g / k g daily dose also reveals a simi lar weight pattern change. Parallel growth rates after 1 or 2 weeks in the single dose rat experiments and weight gains exceeding that of controls subsequent to 30-day ex posure at 1 /tg/kg/rat suggest that toxicity 108 Environmental Health Perspectives 784432 GETSn? 011621 :x TCDD, at least in some circumstances, /t-e reversible. As Table 11 indicates, thy[uus appears to be a most sensitive indica tor of TCDD exposure. Decreases in thymus weight consistently occurred in all species z a dose level below which body weight ef- cts occurred. Ackn ow le d g em en t The authors are grateful to Drs. J. K. Haseman and M. Hogan for statistical analyses. REFERENCES 1. Schwecz, B. A., et al. Chlorodibenzo-p-dioxin toxicology. Environ. Health Perspect, No. 5: 87 (1973). 2. Greig, J. B. Effect of 2,3,7,8-tetrachlorodibenzo1,4-dioxin on drug metabolism in the rat. Biochem. Pharmacol. 21: 3196 (1972). 3. Cunningham, H. M.f and Williams, D. T. Effect of tetrachlorodibenzo-p-dioxin on growth rate and synthesis of lipids and proteins in rats. Bull. Environ. Contamination Toxicol. 7: 45 (1972) . 4. Buu-Hoi, N. P., et al. Organs as targets of "dioxin" (2,3,7,8-tetrachlorodibenzo-p-dioxin) intoxication. Naturwiss. 59: 174 (1972). 5. Gupta, B., et al. Pathologic effects of 2,3,7,8- tetrachlorodibenzo-p-dioxin in laboratory ani mals. Environ. Health Perspect. No. 5: 125 (1973) . 6. Zinkl, J. G., et al. Hematologic and clinical chemical effects of 2,3,7,8-tetrachlorodibenzo-pdioxin in laboratory animals. Environ. Health Perspect. No. 5: 111 (1973). 7. Vos, J. G., Moore, J. A., and Zinkl, J. G. Effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin on the im mune system of laboratory animals. Environ. Health Perspect. No. 5: 149 (1973). 8. Miller, R. G. Simultaneous Statistical Inference. McGraw-HHl, New York, 1966. 9. Jonckheere, A. R. A distribution-free K-sample test against ordered alternatives. Biometrika 41: 133 (1954). GENP 011622 Septem ber 1973 109 784433 \i J Hematologic and Clinical Chemistry Effects ii 2,3,7,8-Tetrachlorodibenzo-p-dioxin in Laboratory Animals ;y l.G. Zinkl,* j.G . Vos,* I.A. M oore,* and B.H. Gupta* Introduction Chlorodibenzo-p-dioxins, especially 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD), are among the most toxic compounds known. These compounds are found as contaminants of technical chlorophenols and 'their deriv atives. A variety of pathologic condition have been associated with the injestion of ^ '.products containing chlorodibenzo-p-dioxins. ^ ` Toxic fat, the cause of chick edema dis ease (1, 2 ) and a variety of pathologic man ifestations in monkeys ( 3 ) , has been found to contain chlorodibenzo-p-dioxins, among which is TCDD (, 5). TCDD has been im plicated in a variety of other toxicoses, in cluding outbreaks of chloracne in chemical workers ( 6 ) , hepatonecrosis in rabbits (6 ), and hepatonecrosis and thymic atrophy in rats (7). However, only in the studies in which monkeys were fed toxic fat (3) and in which rats were given TCDD orally (7, 8 ) were substantial clinical pathologic analyses performed. Toxic fat caused in the monkeys, among other lesions, anemia, leukopenia and hypoproteinemia. The hypoproteinemia was pri marily due to decreased serum albumin concentration. The blood urea nitrogen 'National Institute of Environmental Health Sci ences, National Institutes of Health, P. 0. Box 12233, Research Triangle Park. North Carolina 27709. (BU N), serum bilirubin, cholesterol, sod ium and potassium concentrations, and the prothrombin times were not altered ( 3 ) . Rats given 10 mg TCDD/kg body weight had, among the other lesions, increased ac tivities of serum glutamic-oxaloacetic trans aminase (SGOT), serum glutamic-pyruvate transaminase (SGPT), lactic dehydro genase (LDH), and hydroxybuturate dehy drogenase as well as decreased arylesterase and cholinesterase activities. In addition, the rats had decreased serum glucose, sodium, and protein concentrations, and increased serum urea, lipid and bilirubin concentra tions, hemoconcentration, and neutrophilia. Serum cholesterol, potassium and chloride concentrations and aldolase and alkaline phosphatase (AP) activities were not al tered (7, S). Because the clinical pathologic and path ologic changes found in these and other studies suggest that the most profound ef fect of chlorodibenzo-p-dioxins are on the liver and various hematopoietic organs, we decided to determine the sequential clinical pathologic changes which might occur in rats given different doses of TCDD over a period of time. These methods might also determine which functions are the most sensitive to TCDD. In addition, hematologic studies were conducted as an adjuvant to studies of immunologic effects of TCDD in September 1973 111 GENP 011623 784434 guinea pigs (5) and in mice in which the effects of TCDD on oogenesis are being-de termined {J. McLachlan, personal commu nication). Materials and Methods Animals Female CD rats weighing 150-175 g were given, orally, 0.1,- 1.0, or 10.0 ug TCDD/kg daily for 30 days. Blood for the clinical pathologic studies was obtained from the heart 3, 6, 10, 13, 17, 24, and 31 days after the TCDD dosing commenced. Blood was obtained from the retroorbital sinus for hematologic studies 1, 3, and 5 weeks after 8-week-old female CD-I mice were given oral doses of TCDD of 1.0, 10.0, or 50 /ig/kg. Female Hartley strain guinea pigs weigh ing about 250 g received oral weekly doses of 0.008, 0.04, 0.2, or 1.0 fig TCDD/kg for 8 weeks. The >1.0 n g TCDD/kg group be came moribund between 3 and 5 weeks. Half the animals were injected with tetanus toxoid (group A) and half with killed M y c o b a c t e r i u m tu b e r c u l o s i s (group B) in order to determine the effects of TCDD on humoral and cell-mediated immune re sponses. The details of these experimental procedures can be found elsewhere in this issue ( 0 ) . The highest dose guinea pigs when moribund and the other animals at 8 weeks were bled from the heart for hema tologic analysis. a sulfanilic acid-caffeine-sodium benzoate method (Boehringer-Mannheim Co., New York, N. Y.), sodium and potassium by flame photometry, and chloride by coulomeirie titration. The serum enzymes, alkaline phosphatase (A P ), lactic dehydrogenase (LDH), glutamic-oxaloacetic transaminase (SGOT), and glutamic pyruvate transamin ase (SGPT) were determined by kinetic methods by use of commercially available kits (Sigma Chemical Co., St. Louis, Mis souri) in a recording spectrophotometer. Analysis of Data Statistical analysis for differences between groups was done by Dunnett's test (i.;>. Determination of a dose response was done by Jonckheere's method (14). The 5^ level was selected as the value where vari ables would be considered significantly dif ferent. Results Rats Enzymic alterations in the TCDD-treated rats consisted of increased SGPT activity in the high-dose (lO.O^g TCDD/kg-day) rats at days 17, 24, and 31 (Fig. 1 ), and increased SGOT activity at days 13, 17, 24, and 31 in the high-dose rats as well' as days 13 and 17 in the middle-dose (1.0 fig TCDD/kg-day) rats (Fig. 2). LDH was Procedures Routine methods were used for hemato logic determinations. Platelet counts were de termined in counting chambers (Unopette, Becton, Dickinson and Company, Rutherford, N. J.), fibrinogen by a heatdenaturation method ( 1 0 ); urea nitrogen was determined by a urease method (Boehringer-Mannhein Co., New York, N. Y .), cholesterol by a ferric chloride method (Hyland, Costa Mesa, California), serum proteins by the biuret technioue ( 1 1 ) , glu cose by an o-toluidine method (Hyland. Costa Mesa, California), creatinine by an alkaline picrate method ( 1 2 ) , bilirubin by Figure 1. SGPT activity in rats treated with 0.1, 1.0, or 10.0 TCDD/kg-day. High-dose rats had increased SGPT activity at days IT (P < 0.01) and 24 {P <0.05). The asterisk (*) indicates a single observation. All other points are the means of three or four observations. 112 Environmental Health Perspectives 784435 $ , ted only on day 24 in the high-dose "dl and AP was not' altered at any time. Serum cholesterol was increased on days .. 17, 24, and 31 in the high-dose rats and 24 in the middle-dose animals. Significant .'-response elevations for cholesterol con trations were also found on days 17 and .: (Fig. 3). Blood glucose concentration increased with time in all the treated groups. At day 10, glucose concentrations in all treated! animals were significantly decreased and on days 24 and 31 in the high and middle dose animals it was de creased. Significant dose-response blood glucose decreases were found at days 10, 17, and 24 (Fig. 4). Total serum protein fluctu ated with the treatment. On days 24 and 31, the high-dose rats had decreased protein ,, 200 150 100 50 h -- A. " 'O Conrroi Olyuq/fcq oiOjuq/Xg A iO O y u q /k q ------- Day* igure 2. SGOT activity in rats treated with 0.1. 1.0, or 10.0 tig TCDD/kg-day. Significant increases were found in the high-dose rats on days 13, 17, ;id 24 (P <0.011 and in the middle-dose rats on ,ys 13 and 17 (P <0.051. Significant dose re sponses were found on days 13 (P <0.011, 17 (P <0.011, and 31 (P <0.051. The asterisk (*) in dicates a single observation. All other values are means of three or four observations. 20 Days 30 F ig u r e 4. Blood glucose concentration in rats treated with 0.1, 1.0, or 10.0 jug TCDD/kg-day. Significant (P <0.051 decreases occurred on day 10 in the low- dose rats and the middle-dose rats. Highly signifi cant (P < 0.011 decreases occurred in the middledose rats on day 24 and in the high-dose rats on days 10 and 17. Significant dose responses occurred on days 10 (P <0.0l), 17 (P <0.05 and 24 (P <n.011. The asterisk (') indicates a single ob- Surumclioieiiaroi u>nc GENP 011625 F ig u r e 3. Serum cholesterol concentration in rats given 0.1, 1.0 or 10.0 ng TCDD/kg-day. Signifi cant elevations occurred on days 10 (P <0.051 and 24 (P <0.01) in the high-dose rats and on day 24 (P <0.051 in the middle-dose rats. Significant (P <0.051 dose responses were found on days 17 and 24. The asterisk (*) indicates a single ob servations. All other values are the means of three r four observations. Septem ber 1973 F ig u r e 5. Serum protein concentration in rats treated with 0.1, 1.0, or 10.0 ug TCDD/kg-day. Significant (P <0.051 increased concentrations oc curred in the middle-dose rats on days 31 and de creased concentration in the high-dose rats on day 24. The asterisk (*) indicates a single observation. All other values are the means of three or four observations. 113 784436 concentration while at day 31 it was in creased in the middle-dose rats (Fig. 5). In addition, serum bilirubin concentrations were increased in the high-dose animals on days 17, 24, and 31 (Table 1). No signifi cant changes were found for serum creatin ine, sodium, potassium, chloride, fibrinogen, and BUN concentrations. Table 1. Bilirubin concentration in TCDD-treated rats.1 Bilirubin concentration, mg/100 ml serum Time, days Control 1.0^g/kg-day 10.0 Mg/kg-day 13 0.41 0.10 0.83 * 0.28 b 17 0.25 0.08 0.22 a 0.09 1.94 0.42* 24 0.64 0.51 0.44 0.15 5.62 1.07e 31 0.52 0.15 0.40 0.09 2.16 4 ` Mean a: 1 S.D. *P <0.05. P <0.01. 4Single value. Hematologic changes were confined to hemoconcentration, as previously noted in rats (7), in the high-dose animals on days 17 and 24 (Figs. &-8 ), and to striking thrombocytopenia in all the groups. After only 3 days treatment, the high and mid dle-dose animals had depressed platelet counts which remained depressed through out the study. In the low-dose rats (0.1 p g TCDD/kg-day), platelets were decreased F ig u r e 7. Hematocrit in rats given 0.1, 1.0, or 10.Q fig TCDD/kg-day. Highly significant (P <0.0l increases occurred in the high-dose rats on days 17 and 24. AH values are the means of three or four observations. g 16 3 12 8 Ooys F ig u r e 8. Hemoglobin concentration in rats given 0.1, 1.0.' or 10.0 .tigfkg-day. Highly significant (P <0.01) increases occurred on days 17 and 24 in the high-dose rats. All values are the means of three or four observations. significantly only on day 17. Significant dose response decreases in platelets occurred throughout the study (Fig. 9). No signifi cant changes occurred in leukocyte counts or differentials (in particular, lymphocytes) in these rats. Figure 6. Erythrocyte count in rats given 0.1, 1.0, or 10.0 TCDD/kg-day. Significant increases occurred in the high-dose rats on days 17 and 24 ( P < 0.01). All values are the means of three or four observations. Mice Mice given a single oral dose of 1.0, 10.0, or 50.0 p g TCDD/kg had significantly decreased leukocyte (Table 2) and lympho cyte counts (Table 3) after 1 week. After 3 weeks, none of the treatment groups had significant differences from control mice; however, a significant dose-response lympho cyte depression remained. The leukocyte de pression was on the borderline of being a 114 Environm ental Health Perspectives iV7'fVTTA T U r r o 784437  ncRE 9. Relative platelet counts in rats treated vith 0.1, 1.0, or 10.0 mK TCDD/kg-day. Significant iP <0.05) decreases occurred in the low-dose rats *n day 17. the middle-dose rats on days 3, 10. and :;i and in the high-dose rats on day 24. Highly ,'ignificant (P <0.01") decreases occurred in the middle-dose rats on day 24 and the high-dose rats on days 10 and 17. Significant dose-responses oc curred on days 3 P <0.05), 10 (P <0.01). 17 (P <0.01), 21 (P < 0.01. and 31 (P <0.05). The mean numbers of platelets at day 3 in, the control rats were 1109 X lOVmm1 blood. All values are the means of three or four observations. significant, dose-response. After 5 weeks, no TCDD effects were detected except that the mice treated with 10 ig TCDD/kg had sig nificantly elevated erythrocyte counts (Ta ble 4). This difference is probably due to a lack of variability in the parameter rather than a real elevation and likely does not re flect an effect of TCDD. Guinea Pigs The TCDD-treated guinea pigs in the tetanus toxoid-injected group (group A) (0 ) had consistently lower leukocyte cQunts than the controls, but only for the middledose animals (0.04 Mg TCDD/kg) was this decrease significant (Table 5). The lympho cyte counts were significantly decreased in all the TCDD-treated animals in this group. A significant dose response oc curred for the leukocyte and lymphocyte decreases also, but these effects simply re flect the depression observed at all three TCDD treatment levels. \ Table 2. Leukocyte counts in mice given a single oral dose of TCDD. Leukocyte count x 10' s per mm1 Week Controls 1 ng/kg 10 Mg/kg 50 Mg/kg --------------------:------------------------------------------- ----------------------- X SD X SD X SD X SD 1 12.05 2.58 8.35 b 1.72 8.78 b ` 0.74 . 6.08 0.90 3 3.07 2.50 6.53 0.65 5.43 0.91 6.10 2.29 5 8.70 3.67 3.23 3.82 7.30 1.49 7.27 3.95 ' Pi = one-side P value. bP (one-sided) <0.05. eP (one-sided) < 0.01. Dose response * P, <0.01 Pi =0.052 P, =0.222 GENP 011627 1 Table 3. Lymphocyte counts in mice given a single oral dose of TCDD. Lymphocyte count per mm1 Controls 1 t i g / kg 10 Mg/ kg 50 Mg/kg Week X SD X SD X SD X SD 1 9233.0 1551.7 6276.5 " 1405.3 6443.5 b 1112.9 3752.2 b '378.6 3 6170.7 1872.1 5363.0 422.2 3788.3 839.8 4414.3 1764.6 5 6353.3 3261.2 6587.7 3482.6 5024.3 405.7 4864.0 2679.8 ` Pi = one-sided P value. _" P (one-sided) <0.01. September 1973 Dose response * P, <0.01 P, =0.037 P, =0.182 115 784438 Table 4. Erythrocyte counts X 10* in mice given a single oral dose of TCDD. Erythrocyte count x 10"* per mm1 Controls 1 i*g/kg 10 Mff/kg 50 *gAg Week 1 3 5 x SD X SD X SD X SD 7.28 0.320 7.30 0.141 6.90 0.356 7.33 0.479 7.63 0.115 7.57 0.306 7.43 0.058 7.33 0.252 7.53 0.058 7.73 0.153 7.80 " -- 7.83 0.569 *NS = not significant. *Pi = two-sided P value. Dose response NS * NS* Pi = 0.045 Table 5. Leukocyte, lymphocyte, and neutrophil counts in group A guinea pigs treated with TCDD for 8 weeks. Variable Leukocytes Lymphocytes Neutrophils Control Count per mm1 0.008 ig/kg-wk 0.04 tg/kg-wk 0.2 g/kg-wk X 6407.8 4159.7 1885.6 SD 1878.0 1465.3 559.2 X 5050.0 2390.9 b 2286.5 SD 1161.7 ' 726.8 1269.5 X 4850.0 * 2872.6 * 1668.7 SD 1237.6 1031.4 466.6 X 4914.3 2591.7 * 2096.7 SD 990,7 459.3 634.7 Dose response P = 0.022 ~ P = 0.021 NS e P <0.05. bP <0.01. Table 6. Hematologic parameters for group B guinea pigs treated with TCDD for 8 weeks. Control 0.008 /ttg/kg-wk 0.04 /*g/kg-wk 0.2 itg/kg-wk X SD X SD X SD X SD Leukocytes per mm1 3840.0 Lymphocytes per mm' 4270.5 Neutropil per mm1 3517.6 Platelets X 10'* 796.500 per mm* Erythrocytes X 10"* 4.16 per mm' Hemoglobin, 10.74 g/100 ml Hematocrit, % 41.70 4205.1 8230.0 * 2482.9 7260.0 2293.3 4229.1 1312.0 3717.5 2127.6 3405.9 1204.4 3104.3 103.496 -- ---- 1659.5 6140.0 * 2910.2 1169.1 3628.2 2371.7 654.8 2179.3 718.7 -- 645.000 b 84.755 0.28 4.24 0.25 4.18 0.30 4.26 0.25 0.57 11.00 0.40 11.29 0.57 10.59 0.66 1.84 42.25 1.77 41.90 2.34 40.45 2.22 Dose response P < 0.01 P = 0.024 P = 0.020 -- NS" NS NS P <0.05. bP <0.01. NS = not significant. For Group B (M y c o b a c t e r i u m t u b e r c u losis tuberculin-treated) (9) only the high est dose (0.2 fig TCDD/kg) caused signifi cantly decreased leukocyte counts (Table 6 ). Neutrophil counts were also reduced in these guinea pigs, but there was no treat ment versus control differences in lympho cyte counts. Significant dose-related de creases were found for leukocytes, neutro phils, and lymphocytes. In addition, the guinea pigs treated with 0.2 fig TCDD/kg also had lower platelet counts than the con trols ( P < 0.01). The leukocyte and lymphocyte counts of the group A and group B control guinea pigs did not differ significantly (Table 7). However, the group B controls did have higher neutrophil counts than the group B controls (P < 0.05). When the group A con trol guinea pigs were compared with the moribund 1.0 fig TCDD/kg guinea pigs, de creased lymphocyte and increased neutrc- 116 Environmental Health Perspectives 784439 GENP 011628 Table^T: Leukocyte counts of tetanus toxoid-treated (group A), Mycobacterium tuberculosis tuberculin-treated (group B), and TCDD-treated guinea pigs. Variable ,jyte count per mm1 hocyte count per mm1 rophil count per mm1 f><0.01. p<0.05. Group A controls s 6407.8 4159.7 1885.6 SD 1878.0 1465.3 559.2 . Group B controls X 8840.0 4270.5 2517.6 " SD 4205.1 2293.3 2127.6 1.0 Mg TCDD/kg-wk X 6800.0 1804.0 4686.4 * SD 4018.7 522.4 3534.7 phil counts were found in the treated ani"!s. However, this comparison may not be I because the TCDD-treated guinea pigs 3 to 5 weeks younger than the group ,-ontrols and were in a moribund state in .(ilition to not having been injected with :.L*canus toxoid. Hemoconcentration was also present in these dying animals, there being an average erythrocyte count of 5.1 x 10s per mm3 and a hematocrit of 48.9%. Plate' -t counts were decreased as well. scussion The clinical pathologic findings of this . ..d other studies (3, 7, 3 ) suggest that the major sites of the toxic action of TCDD and the chlorodibenzo-p-dioxins of toxic fat are the hematopoietic system and the liver. Mice and guinea pigs given TCDD and mon keys fed toxic fat ( 3 ) were leukopenic, hich in mice and guinea pigs is character e d by lymphopenia. In mice which re vived a single dose of TCDD the lympho penia was reversed 5 weeks after TCDD exposure. Perhaps the lymphopenia seen in mice and guinea pigs is related to the decreased cell-mediated immune response observed in these two species after TCDD exposure (5). It is interesting that lympho penia and cell-mediated immunosuppression were not found in rats treated with TCDD. Rats and guinea pigs were thrombocyto penic. The cause of this was not determined. Examination of rat bone marrow in one experiment did not reveal any differences in numbers or morphology of megakaryo cytes following TCDD treatment ( 1 7 ) , but ip another experiment in rats, spleen and le marrow megakaryocytes were degen erated and appeared to be reduced in num bers ( 1 5 ) . Therefore, further work is in dicated to determine, perhaps by radioiso tope tagging, whether the lowered platelet counts are due to decreased production or increased distruction of platelets. Neverthe less, thrombocytopenia or defects in the clotting mechanism (1 7 ) might have played a role in the production of hemorrhages that were occasionally seen in rats that died U 5). Anemia was not found in these studies, but it was seen in a previous study in mon keys fed toxic fat ( 3 ) . That anemia was accompanied by atrophic bone marrow and normal serum bilirubin suggesting that the anemia was aplastic or depression type rather than hemolytic anemia. We observed only terminal hemoconcen tration in rats and guinea pigs, similar to that previously reported in rats given TCDD (7). This is probably an effect of terminal shock and dehydration rather than increased erythrocyte production. Dehydra tion is further evidenced by the increased serum protein concentrations found in the middle-dose rats. However, hypoproteinemia was observed in the high-dose rats probably as a result of liver damage, and may have contributed `further to the hemoconcentra tion by decreasing the colloidal osmotic pressure of the blood and allowing fluid to accumulate in the tissues. Liver damage in rats was evidenced by increased SGOT and SGPT activity. The increased SGOT activity might also be due to, in part, myocardial necrosis ( 1 5 ) . Hyper bilirubinemia, hypercholesterolemia, and hy poproteinemia (in the high-dose rats) are Septem ber 1973 117 784440 I i + i On ts> VO probably other effects of the liver pathology. Since increased cholesterol concentrations have previously been seen in rats (1 8 ) and rabbits (1 9 ) which have sustained hepato cellular damage from polychlorinated bi phenyls, perhaps metabolism of cholesterol is altered in liver damage caused by these chlorinated compounds. Hypoglycemia in the TCDD-treated rats might have been due to decreased food consumption (5), but there might also be an effect on the glucon eogenic ability of the damaged liver. Al though clinical chemical parameters to as sess liver damage were not determined in guinea pigs and mice, histopathologic evi dence suggests that hepatocellular damage was minimal in these species ( 1 5 ) . It appears that hepatocellular necrosis is the main toxic action of TCDD in rats but with effects on platelets being important. The clinical pathologic changes are consist ent with this hypothesis. Ultimately, hemoconcentration due to shock and dehydra tion occurs as a terminal event in rats as well as guinea pigs. The lymphopenia observed in mice and guinea pigs are consistent with the immu nosuppressive effects of TCDD in these species ( 9 ) . Pathologic studies suggest that liver damage does not play a part in the death of guinea pigs given high doses of TCDD (15). However, in guinea pigs, atrophy of the adrenal zona glomeruiosa (9, 1 5 ) suggests that electrolyte imbalance might occur. Perhaps a study of serum elec trolyte and aldosterone concentrations in guinea pigs given lethal doses of TCDD is warranted. Acknowledgement The authors thank Drs. J. Haseman, M. Hogan, and T. Clemmer for statistical con sultation and Mrs, M. Ebron for technical assistance. REFERENCES 1. Edgar, S. A., et ai. The effect of a toxic sub stance in fat on poultry. Poultry Sci. 37: 1200 .(1958). 2. Allen, J. R. The role of toxic fat in the produc tion of hydropericardium and ascites in chickens. Am. J. Vet. Res. 25: 1210 (1964). 3. Allen, J. R,, and Carstens, L. A. Light and electron microscope observations in Maeaea mulatto, monkeys fed toxic fat. Am. J. Vet. 28: 1513 (1967). 4. Higgenbotham, G. R. et al. Chemical and toxico logical evaluations of isolated and synthetic chloro derivatives of dibenzo-p-dioxins. Nature 220: 702 (1968). 5. Metcalfe, L. D. Proposed source of chick edema factor. J. Assoc. Offic. Anal. Chemists 55: 542 (1972). 6. Schultz, K. H. Clinical picture and etiology 0f chloracne. Arbeita-med.-Socialmed.-Arheitshyg 3: 25 (1968). 7. Buu-Hoi, N. P., et al. Organs as targets of "dioxin" (2,3,7,8-tetrachlorodibenzo-p-dioxin) in toxication. Naturwiss. 4: 174 (1972). 8. Buu-Hoi, N. P., et al. Enzymatic functions as targets of the toxicity of "dioxin" (2,3,7,8tetrachlorodibenzo-p-dioxin). Naturwiss. 4: 173 (1972) . 9. Vos, J. G., Moore, J. A., and Zinkl, J. G. Effects of 2,3,7,8-tetrachIorodibenzo-p-dioxin on the im mune system of laboratory animals. Environ. Health Perspect. No. 5: 149 (1973). 10. Kaneko, J. J., and Smith, R. The estimation of plasma fibrinogen and its clinical significance in the dog. Cal. Vet. 21: 21 (1967). 11. Goraall, A. G., Bardawill, C. J. and David, M. M. Determination of serum proteins by means of the biuret reaction. J. Biol. Chem. 177: 751 (1949). 12. Henry, R. J. Clinical Chemistry Principles and Techniques. Harper and Row, New York, 1962, p. 292. 13. Miller, R. G., Jr. Simultaneous Statistical Inference. McGraw-Hill, New York, 1966. 14. Jonckheere, A. R. A distribution free K-sampie test against ordered alternatives. Blometrika 41: 133 (1954). 15. Gupta, B. N., et al. Pathological effects of 2,3,7,8tetrachlorodibenzo-p-dioxin in laboratory ani mals. Environ. Health Perspect. No. 5: 125 (1973) . 16. Harris, M., et al. General biological effects of 2.3.7.8- tetrachlorcdibenzo-p-dioxin in laboratory animals. Environ. Health Perspect. No. 5: 101 (1973). 17. Weissberg, J. B., and Zinkl, J. G. Effects of 2.3.7.8- tetrachlorodibenzo-p-dioxin on hematol ogic and blood coagulation function in the rat. Environ. Health Perspect. No. 5: 119 (1973). 18. Kuratsune, M. An abstract of results of labora tory examinations of patients with Yusho and of animal experiments. Environ. Health Per spect. No. 1: 129 (1972). 19. Roller, L. D.t and Zinkl, J, G. Pathology of poly chlorinated biphenyls in rabbits. Am. J. Path. 7C: 363 (1973). U S Environmental Health Perspectives 784441 G EN P011630 Fffects of 2,3,7,8-Tetrachlorodibenzo -dioxin upon Hemostasis and Hematologic Function in the Rat Joseph B. Weissberg* and Joseph G. Zinkl Introduction 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) has been identified as a toxic contamin.nt of chlorophenols and related hydronrbons that are widely used in agriculture .-d industry- It has been implicated in io u s diseases, including human cases of oracne (1 ) and porphyria cutanea tarda {2) and in chick edema ( 3 ) . Its teratogenic ity lias been documented in rats (4 ) and mice (5). Monkeys fed toxic fat, later found to contain TCDD and other chlorinated de rivatives of dibenzo-p-dioxin, were observed to undergo a decrease in the cellular elements of the bone marrow and the peripheral blood (d). In contrast, increases in hematocrit and leukocyte count were observed in rats treated with TCDD (7). Depression of blood platelets in TCDD-treated rats and guinea pigs has been recently observed (5). The present studies were undertaken in order to characterize hematologic changes and in particular platelet function alterations in rats exposed to TCDD. Pathologic Physiology Branch, National Institute of Environmental Health Sciences, National Insti tutes of Health, P. 0. Box 12233, Research Triangle Park, North Carolina 27709. f Animal Science and Technology Branch, National Institute of Environmental Health Sciences, Na nai Institutes of Health, P. 0. Box 12233; Reirch Triangle Park, North Carolina 27709. Methods Animals Female CD rats, weighing 150-175 g, re ceived daily oral doses of TCDD in an ace tone-corn oil mixture at a level of 10 /g/kg. Control rats received an equivalent dose of acetone and corn oil. On days 10 and 14 of treatment animals were anesthetized with methoxyfiurane, and blood was withdrawn by cardiac puncture. Marrow Megakaryocyte Studies Bone marrow preparations were made as described by Eurenius (P), and smears were prepared with a Wright-Leishman-Giemsa stain. Platelet Function Studies Bleeding times were determined on anes thetized animals. Lacerations were made on the ear with a blood lancet, and puncture sites were blotted with filter paper at 30-sec intervals until bleeding stopped. Clot re traction was determined as described in the literature ( 10). Platelet factor III activity was evaluated with the prothrombin con sumption test (11) performed on blood that was allowed to clot with and without added Inosithin. Platelet aggregation was determined macroscopically in a mixture containing, in a final volume of 0.5 ml, 0.1 onole ADP, 1.0 tmole calcium chloride, Septem ber 1973 119 784442 GENP011631 i 5.0 tmole Tris HC1, pH 7.35, and platelet-rich plasma with a final platelet concenti'ation of 200,000/m m \ ' Prothrombin times were performed as described in the literature ( 1 2 ) . Factor X assays were performed by measuring: clotting time in a mixture containing test serum, 0.1 ml, factor X-deficient plasma (Dade), 0.1 ml, and Simplastin (General Diagnos tics), 0.2 ml. Fibrinogen Degradation Products Fibrinogen degradation products were assayed in the serum of control and treated rats by using the Wellcome FDP Kit (Burroughs-Wellcome). which employs a hem agglutination-inhibition system. Other hematologic parameters were de termined by routine methods. Statistical analysis was performed by the Student's t test. Results Red cell parameters of control and treated rats are given in Table 1. Packed cell volumes ivere elevated in TCDD-treated rats at days 10 and 14 of treatment. Similar ly. erythrocyte counts were increased in the treatment as compared to control groups at days 10 and 14. Reticulocyte counts differed significantly at day 14, presumably repre senting lowered values in the control animals. Mean corpuscular volume and mean cor puscular hemoglobin were significantly de pressed in treatment groups at day iq u not day 14. TCDD-treated rats did not diff*1 significantly from controls with regard to either hemoglobin or mean corpuscular hemoglobin concentration. White cell parameters of control and treated rats are given in Table 2. Total leukocyte counts were elevated, but not significantly, in TCDD-treated animals at 10 and 14 days. Neutrophil counts were significantly increased in both treatment groups when compared to controls. Lympho cyte and monocyte counts were elevated in rats treated 10 and 14 days, and eosinophil counts were slightly depressed. These differ ences. however, were not statistically signi ficant. Table 3 demonstrates that platelet counts were significantly lowered in treated rats at days 10 and 14. Bone marrow prepara tions did not show a corresponding decrease in megakaryocyte numbers of animals treated 14 days. Megakaryocyte size and numbers of nuclei per megakaryocyte were also evaluated, and these were observed to be equivalent in control and treatment groups. Bleeding times were equivalent in treated and control rats at day 14. These results are presented in Table 4. Clot retraction was diminished in both 10-dav and 14-day treatment groups compared to consols. The velocity of platelet aggregation was not significantly altered in TCDD-treated rats. Table 1. Red cell parameters in rats administered daily oral doses of TCDD of 10 mf/kg.* Hemoglobin, g/100 ml Packed cell volume, 9c Erythrocyte count x 10-*, per mm1 Reticulocyte count, % Mean corpuscular volume, "* Mean corpuscular Hb concentration. f"r . Mean corpuscular Hb, pg/cell 10 Days Control Treated 12.3 0.7 37.4 2.3 5.0 4m 0.5 13.5 0.8 43.4 3.75.7 s 0.4" 4.2 1.2 75.8 4m 4.1 32.8 -* 1.4 24.9 1.2 4.1 1.0 65.0 * 7.031.2 -** 1.7 20.2 4 1.1" 14 Days Control Treated 12.6. * 1.2 36.6 * 2.3 5.2 * 0.3 14.8 4m 1.4 45.0 4.06.2 4 0.7* 1.5 0.6 71.1 1.8 34.5 1.3 3.3 4m 1.2* 73.0 2.2 32.9 -fc. 1.4 24.5 4m 0.9 24.3 4 1.6 s Values presented are means c: standard deviation. Four animals were tested in each group. Values marked with asterisks differ significantly from control values: *P <0.05; --P <0.01. 120 Environm ental Health Perspectives 784443 GENP 011632 Table 2. White cell parameters in rats administered daily oral doses of TCDD .__ . of 10 fig/kg/ yte count X 10 per 10 Days Control Treated 4.6 1.5 9.4 -** 4.2 14 Days Control Treated 5.1 4* 2.6 9.2 2,1 s iphil count X IO-1per 0.5 0.1 0.9 0.3* 0.6 4- 0.3 2.7 1.5* .-upiiocyte count X 10'1per mm" '.;.*::ocyte count per mm1 ..;iiophil count per mm1 3.9 1.4 137 92 61 31 8.1 3.9 254 182 21 36 4.3 2.4 89 50 77 39 6.1 1.3 243 135 65 55 - aiues presented are means standard deviation. Four animals were tested in each group. Values -..Ued with an asterisk differ significantly from control values at a level of P <0.05. Table 3. Effect of daily oral doses of TCDD of 10 /ig/kg upon marrow megakaryocytes and blood platelets in rats/ ` .:c!et count x 10'1 or mm1 .akaryocytes per 10* -- .ucieated marrow cells 10 Days Control Treated 9U 100 427 225* 14 Days Control Treated 688 58 388 124* 444 -- 75 615 219 Values presented are means standard deviation. Four animals were tested in each group. Values marked with an asterisk differ significantly from control values at a level of P < 0.01. `{owever. prothrombin consumption tests reealed markedly prolonged times, both with -:id without Inosithin, in rats from both eatment groups. This contrasted with (iiivalent prothrombin times observed in -'intro! and treated animals. No difference n factor X levels between control and rreared rats was observed at 14 days. Serum fibrinogen degradation products were not observed in either control or treat ment groups at days 10 and 14. Four animals were tested in each group. Discussion The elevations in packed cell volumes and erythrocyte counts of TCDD-treated rats reported in the present study are consistent with dehydration and consequent hemoconcentration. These results are in agreement with earlier investigations of TCDD (7). Presumably, the previously reported depres- -September 1973 sion of hematopoiesis in monkeys fed toxic fat (t>) represents species variation or the effect of a different toxic contaminant. The alterations in red cell indices and the leukocytosis with neutrophilia, lymphocy tosis and eosinopenia that have been ob served in this study are nonspecific hemato logic changes consistent with widespread toxicity of TCDD. Of interest is the ob served selective depression of blood platelets in treated rats. In view of the finding of normal marrow megakaryocytes, decreased production of platelets in treated rats is unlikely. It is possible that TCDD results in disseminated intravascular coagulation with thrombocytopenia due to the aggrega tion and incorporation of platelets into platelet-fibrin microthrombi. However, the absence of serum fibrinogen degradation products in treated animals argues against this possibility. Increased peripheral des- 121 784444 GENP 011633 I Table 4. Platelet function and procoagulant activity in rats administered daily oral doses of TCDD of 10 pg/kg.* Bleeding time, min Clot retraction, % Velocity of platelet aggregation, sec Prothrombin time, sec Prothrombin consumption test, sec Prothrombin consumption test with Inosithin, sec Factor X, sec - ,,,,10 Days Control Treated -- 73.6 5.8 (7) 18.0 1.4 (2) 11.8 0.5 22.8 6.7 -- 60.8 4.3** (6) 26.0 8.5 (2) 11.7 1.9 58.5 1.7*** 26.2 9.3. 60.0 s 0.0*** ---- 14 Cays Control Treated 3.6 0.9 81.2 6.0 4.1 2.7 69.8 4.3! 23.0 6.4 (3) 11.7 * 1.8 18.9 3.6 51.2 40.0 13.5 3.1 53,0 8.2*** 19.8 6.8 51.9 16.2* 19.0 2.0 (2) 21,0 0.0 (2) ` Values presented are means standard deviation. Numbers in parentheses indicate number of animals tested; in all other groups four animals were tested. Values marked with asterisks differ significantly from control values: *P <0.02; **P <0.01; ***P <0.001. truction of platelets due to an antibody response is a possible mechanism for the observed thrombocytopenia. Such a mechan ism is proposed in human cases of hyper sensitivity to quinidine and Sedormid ( 1 3 ) . The diminished clot retraction in TCDDtreated animals is consistent with the ob served thrombocytopenia. Otherwise, plate let function, as assessed by bleeding' time and velocity of aggregation, was not altered. Similarly, .decreased platelet factor III activity, which would be reflected in a shortening of the prothrombin consumption time that was corrected by the addition of inosithin, was not observed in treated rats. On the contrary, prothrombin consumption times were consistently prolonged in the treatment groups. In view of the normal prothrombin times, this prolongation of pro thrombin consumption times is not readily explainable. Factor X levels were normal in treated animals. Deficiency or inactivation of factor VII, which has been reported in human cases of liver and renal failure and after exposure to propylthiouracil, salicy lates and indanedione drugs (14), is a pos sibility. Prolongation of the prothrombin consumption time in factor Vll-deficient dogs has been reported ( 1 5 ) , but this 122 is accompanied by prolonged prothrombin times. Similarly, deficiency or inactivation of prothrombin might prolong the pro thrombin consumption test, but again, an abnormal prothrombin time would be ex pected. It remains to perform factor VII assays and other coagulation studies in animals exposed to TCDD, Acknowledgments The authors gratefully acknowledge the helpful advice of Doctor Thomas. Griggs of the Pathology Department, North Carolina Memorial Hospital, and the technical as sistance of Mrs. M. Ebron. REFERENCES 1. Kimmig, J., and Schulz, K. H. Occupational chloracne caused by aromatic cyclic ethers. Der matologic 115: 540 <1957). 2. Poland, A., and Glover, E. 2,3,7,3-Tetrachloradibenzo-p-dioxin; a potent inducer of J-aminoievulinic acid synthetase. Science 179: 476 (1972). 3. Higginbotham, G. R., et al. Chemical and toxi cological evaluations of isolated and synthetic chlorriderivatives of dibenzo-p-dioxin. Nature 220: 702 (1968). 4. Sparschu, G. L., Dunn, F. L., and Rowe, V. K. Study of the teratogenicity of 2,3,7,8-tetrachiorodibenzo-p-dioxin in the rat. Food Cosmet. Toxicol. 9: 405 (1971). Environmental Health Perspectives r r \T T (\ X M C tO 784445 ,, Courtney, K. D., and Moore, J. A. Teratology studies with 2,4,5-trichlorophenoxyacetic acid .1id 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol, ^ ^ p p l. Pharmacol. 20f 396 (1971). Allen, J. R., and Carstens, L. A. Light and electron microscopic observations in Macaca mulatta monkeys fed toxic fat. Am. J. Vet. Res. 23; 1513 (1967). Buu-Hoi, N. P. et al. Organs as targets of dioxin (2,3,7,8-tetrachlorodibenzo-p-dioxin) in toxication. Naturwiss. 59; 174 (1972). Zinkl, J., et al. Hematologic and clinical chemical effects of 2,3,7,8-tetrachlorodibenzodioxin in lab oratory animals. Environ. Health Ferspect. No. 5: 111 (1973). Eurenius, K., et al. Platelet and megakaryocyte kinetics following thermal injury. J. Lab. Clin. Med. 79: 247 (1972). 10. Hardisty, R. M., and Ingram, G. I. C. Bleeding Disorders. Blackwell, New York, 1965, p. 271. Owen, C. A., and Thompson, J. H. Soybean phosphatides in prothrombin-consumption and throm boplastin-generation tests. Amer. J. Clin. Path. 33: 197 (1960). 12. Quick, A. J. On various properties of throm boplastin (aqueous tissue extracts). Amer. J. Physiol. 114: 282 (1935). 13. Storworken, H., and Owen, P. A. Physiopathology of hemostasis. Sem. Hemat. 8: 3 (1971). 14. Owen, C. A., et al. Congenital deficiency of fac tor VIL Amer. J. Med. 37: 71 (1964). 15. Dodds, W. J., and Kaneko, J. J. Hemostasis and blood coagulation. In: Clinical Biochemistry of Domestic Animals. J. J. Kaneko and C. E. Cornelius, Eds., Academic Press, New York, 1971. GENP011635 'ptember 1973 Pathologic Effects of 2 ,3 ,7 ,8 -T etrachlo rodibenzo-p-dioxin in Laboratory Animals y B.N. Gupta,' l.G . Vos,* L A . Moore,' 1.6. Zinkl,* and B.C. Bullock* Introduction The technical chlorophenois and products produced from them, such as the herbicide 2.4,5-trichlorophenoxyacetic acid (2,4,5-T) have been found to contain chlorodibenzo-plioxins such as 2,3,7,8-tetrachlorodibenzo-piioxin (TC-DD). TCDD and other chlorodi;i(`nzo-p-dioxins are among the most toxic impounds known, and have been implicated in outbreaks of chloracne among chemical workers ( i ) . Hepatic and cardiac lesions and thymic involution were found in rats treated with TCDD ( 2 ) . Generalized sub cutaneous edema, ascites, hydrothorax, and hydropericardium were observed in monkeys given toxic fat (5) which purportedly con tained TCDD. It also caused focal necrosis of parenchymal cells of liver and gastric ulcers in these monkeys. The purpose of this paper is to describe the sequence of pathologic changes which occurred follow ing exposure to TCDD in rats, guinea pigs, and mice. Materials and Methods Young adult male and female random bred albino rats (CD stock, Charles River Breeding Laboratories, Wilmington, Mas *National Institute of Environmental Health Sci ences, National Institutes of Health, P.O. Box 12233, Research Triangle Park, North Carolina 27709. sachusetts) were used. They were housed individually under conditions of controlled temperature (68-72F) and 12 hr daily lighting. The diet consisted of commercially available sterilized diet. Water was provided a d l i b i t u m . Young, short-haired female al bino guinea pigs (Hartley strain) and fe male mice (CD-I) were also used to study the toxic effect of TCDD. These animals were also housed under optimal controlled conditions. TCDD (Dow Chemical Company, Mid land, Michigan) was dissolved in acetone and diluted with corn oil. All animals (rats, guinea pigs and mice) were given TCDD by gastric intubation. Control animals were treated similarly with acetone and corn oil only. Dose range and frequency of treat ments were as follows: (a) rats treated daily with 10.0, 1.0, or 0.1 n g TCDD/kg body weight and examined after 3, 6, 10, 13, 17, 24, and 31 treatments; (b) rats given a single treatment with 100.0 or 50 fig TCDD/kg body weight and ex amined when the rats became moribund or dead: (c) rats given a single treatment with 25.0 or 5.0 xg TCDD/kg body weight and examined 1, 3, 9, 10, 17, and 29 days after treatment: (d) rats given 6 weekly treat ments with 5.0, 1.0, or 0.2 xg TCDD/kg body weight and examined 10 days after the last treatment; (e) guinea pigs given 8 weekly September 1973 125 GENP 03163( 784447 ` .-eatments with 1.0, 0.2, 0.04, or 0.008 xg V'CDD/kg body weight and examined 56 *ys after the initiation of the treatment; ;) guinea pigs given a single "treatment ith 3.0 /*g TCDD/kg body weight and ex amined either after death or moribund con-:.:ion; (g) mice given a single dose with .'5j.0, 10.0, or 1.0 jttg TCDD/kg body weight z : . d examined 7, 21, and 35 days after treat ment. All animals received the scheduled eatment unless death intervened. Necropsies were performed on euthana:.zed animals and those which became morir. ,nd or died before the scheduled kill day. Tissue samples from liver, spleen, kidney, urinary bladder, heart, lung, thymus, tracr.-:a, brain, spinal cord, salivary glands, axil;;.ry and mesenteric lymph nodes, tongue, v ophagus, stomach, small and large intes:..'.es, pancreas, striated muscle from hind ;-:gs and diaphragm, bone marrow from r-.-mur, tibia, sternum, and ribs, adrenal, :r. vroid, uterus, ovary or testes, and skin v.v:re collected from most of the animals &.vi fixed in' 10% buffered neutral formal::. for histopathologic evaluation. All tissue r.f.-cimens were paraffin-embedded, sectioned 6 .v. thick and stained with Harris' hema toxylin and eosin Y (H & E ). Liver, kidney, bone, and total body of dead or moribund rats and guinea pigs were grossly examined under the ultraviolet light (Wood's lamp) for the presence of red fluorescence as an indication of porphyrin accumulation. Results Gross Pathologic Findings Most of the rats given TCDD at a level of 10.0 ^ g /k g became moribund or died be tween 17 and 31 daily treatments. Six of 14 rats given a single dose at a level of 100.0 n g / k g also died 18 to 21 days after treat ment (4). These rats had ruffled hair and appeared depressed. They usually sat in a corner of the cage without much movement. Gross pathologic changes observed in dead or moribund rats given these doses were similar and will be described together (Table 1 ). The ears, subcutaneous tissues and visceral organs appeared icteric. Oc- casionally, there were subcutaneous hemop rhages in tail, paws, and under the naiV some had a loss of hair from the ventral surface of the body. There was a loss of SUh. cutaneous and abdominal fat in TCD d" treated rats. The size of the uterus in these rats appeared to be smaller. The liver was friable and dark tan in color. The liver and spleen sizes were small. There were ulcera tions and hemorrhages in the stomach which also contained blood dots (Fig. l) . c on. gestion of meningeal vessels and submeningeal hemorrhages were also observed (Fig, 1). All thymuses were markedly atrophied! Figure 1. Gastrointestinal tract and brain of a rat given 26 daily treatments of 10.0 Mff TCDD/kg body weight. Notice congestion and hemmorrhage in the brain (arrow), stomach (S) and duo denum (D). The livers from rats killed after 10 to 17 daily intubations at 10.0 f i g / k g were dull gray (cooked appearance), swollen, and en larged. There was accentuation of lobular markings (Fig. 2 ). The small intestine and mesenteric lymph nodes appeared slightly congested and the intestinal contents were more mucinous than that of control rats. Thymic atrophy (Fig. 2), which was dosedependent, occurred in all rats examined. Significant gross pathologic changes were 125 Environm ental Health Perspectives ( CC\ T T A TKT'-r~x 784448 *t:CRE 2. Rats (left) given 10 daily treatments of 1.0 ng TCDD/kg body weight and (right) control. ,'otice the accentuation of lobular markings in ne liver (L) and thymic atrophy (arrow). lot observed in other organs of rats given i^.kO or 5.0 fig TCDD/kg single dose, 1.0 or ".l ^g/kg multiple daily doses, and 5.0, 1.0, *<r0.2 f i g / k g weekly doses. All guinea pigs given weekly treatment at a level of 1.0 ig/k g became moribund or died between 24 and 32 days after the initiation of the treatment (Table 1). Nine out of ten guinea pigs given a single dose of 3.0 fig TCDD/kg body weight died between 15 and 28 days after treatment. There was a severe body weight loss, decreased amount of subcutaneous and visceral adipose tissues, and the guinea pigs appeared to be dehydr ated. Hemorrhages were also observed in the adrenal gland, urinary bladder, gastro intestinal tract and mesenteric lymph nodes. There was marked atrophy of the thymus. Other than the relative thymic atrophy, re markable gross tissue changes were not ob served in guinea pigs given weekly doses of 0.2, 0.04 or 0.008 fig TCDD/kg. Red fluorescence under the ultraviolet light indicating the presence of excess amounts of porphyrins was not observed grossly in any organ of the rats and guinea pigs examined. Table 1. Summary of Pathologic changes in rata, guinea pigs, and mice given TCDD orally. Species Rat Rat Rat Rat Rat Rat Guinea pig Guinea pig Guinea pig Mouse Doses. Mg TCDD/kg body weight (no. of treatments) 100 (single) 50 (single) 25 (single) 5.0 (6 weekly) 10.0 (16-31 daily) 1.0 (31 daily) 3.0 (single) 1.0 (4--5 weekly) 0.2 (8 weekly) 50.0 (single) Mortality, % 43 T 0 0 94 0 90 100 0 0 Period of death, Thymic days atrophy 18-21 Severe 11 -- --. 15-31 Severe Slight/ . moderate Moderate Severe Moderate 15-28 Severe 24-32 Severe -- Slight/ moderate -- Slight Pathologic changes Liver damage Other major changes Severe Severe None Icterus, hemorrhage depletion of lymph oid organs None None Slight Severe Slight/ moderate Slight ' Slight None None Icterus, hemorrhage depletion of lymph oid organs Degenerative changes in kidney and thy roid glands Hemorrhage atrophy of adrenal zona glomerulosa, depletion of lymphoid organs Same as above, and hyperplasia of uri nary bladder epi thelium None Slight None otember 1973 127 7 8 4 4 49 r* w v - * / * S B IS S S f ^ ^ 7 2 2 fS , ; ;*, 7 * ; , M - ' . ; w i rV * > ' .Vi' >*- t ' - ' *; T i > *. > 9 fu \+ 7 t A . `7 _ , - * *+ z O -. c F ig u r e 3. Thymus of a rat treated with 10.0 vg TCDD/kg daily for 16 days. Notice the marked decrease in the number of cortical thymocytes and loss of demarcation between the cortex and medulla. E&E stain; magnification 307x . 128 Environmental Health Perspectives 784450 G E N P 011639 T * a :V ,* 't . _ . V ' ta C N rP A*' * . .. - . ^ > * - 2 ~ - r - - . \ - `* v - " i . * * . * : 4- 'v 'r - ' . f ^ ^ *><* * a v* " ' i*5*! .* ^ 1 / - -w - . *- ,.. T L. v JS . ^ ' -" . - ` - S ' ' vf t -- T -v - ^ * A A ' _ j* r * ^ r \ . -as ' * ^ M sfc ' f c .. . ** & ^ C?S ' ^ i V i ; ? -* : ^ 3 ^ - ' * ^ V ^ J A . ; * 1-- w - ^ . %* ** **Sl* 1?\ % s \* * Ci<. .:i > s CSN- 7- % > . s; 7 i ' ' * % v* ^ % t. * C F igure 4. Liver of a rat given 31 daily treatments of 10.0 Mg TCDD/kg body weight. Regenerated hpato cytes are disorganized with moderate to marked degenerative changes. H&E stain; magnification 320 X. September 1973 129 784451 GENP 011640 f 3 % r? : V n. r- Y i*- & - 4 |y Z * * $ 9 mm, K * W Il ~ M ^ r^ 5 P S 3 g B ! a Figure 5. Liver of a rat given 31 daily treatments of 10.0 ng TCDD/kg body weight Notice necrosis and tubule-like structures in the parenchymatous tissues of liver. H&E stain; magnification 320X. 130 Environmental Health Perspectives 784452 GENP 011641 <A 7<~* ` ._* f * /* ; , ^ i ; _- t',. v V v * 1* . ^ ^ <f * . '< * ' a * -' v v . - r * i - ^ J S J g S ^ i I* * -T~ 1 y ,,, -3 C i ? * ' - `i ! ? r - - " ^ S , L ZP9lwd N ao * ** *i SS g tv *r * v [JRE 6. Heart of a rat given 18 daily treatments of 10.0 Mg TCDD/kg body weight, showing massive- hem* rhages (H) separating the myocardial fibers (arrows). H&E stain; magnification 320x . September 1973 131 784453 F igure 7. Kidney of a rat given 18 daily treatments of 10.0 Mg TCDD/kg body weight. Notice hyaline drop lets,, and foamy and vacuolated cytoplasm in the renal tubules. H&E stain; 904x . Histopathologic Findings Rats-- In those rats which became mori bund oi' died, remarkable changes were consistently observed in the thymus, liver, spleen and lymph nodes (Table 1). Changes in the ovary, uterus, gastrointestinal tract, heart, and brain were frequently observed. Marked atrophy of thymus was denoted by decrease in the number of cortical thy mocytes (Fig. 3). The thymic lobules were markedly smaller, and there was no demar cation between the cortex and medulla. There was also a relative depletion of lymp hoid cells in the spleen and lymph nodes. Pyknosis of the nuclei and degenerative changes in the multinucleated megakaryocytic type giant cells of spleen and bene marrow were observed more `frequently in treated rats. There was moderate to marked distortion of the architecture of liver par enchyma with marked necrosis of hpato cytes. The heptocytes were round and large, and the hepatic cords were disorgan ized (Fig. 4). Some of the necrotic hpato cytes had tubulelike structures (Fig. 5). Rats which died or were killed when mori bund had massive hemorrhages in the heart (Fig. 6), liver, brain, adrenal gland, and gastrointestinal tract. Occasionally, organ ized thrombi were observed in the heart. 132 ,, Environmental Health Perspectives * 784454 GENP 011643 I .'S '; *' <r* ' fi l i f r *^ a ^ . - < Tv? f e .- * S T | * i r . Z ^ g P - AW F igure 8. Liver of a rat given 6 weekly treatments of 5.0 Mg TCDD/kg body weight. Notice multinucleated giant hepatocytes. H&E stain; magni fication 500 x . brain, and lung's. There were necrosis and ulceration in the glandular part of the stom ach. Atretic changes of the ovarian follicles were observed more frequently in treated than in control rats. There was also atrophy of mucosal folds and glandular structures of the uterus. The epithelial cells of renal tub ules were foamy and vacuolated and con tained numerous hyaline droplets (Fig. 7). This change was found in 3 out of 16 rats ^treated with 10.0 /*g TCDD/kg body weight ^iven daily (Table 1). September 1973 Rats given, sublethal doses of TCDD had regeneration of hepatic parenchyma de noted by megalocytosis and multinucleated hepatocytes (Fig. 8). There were increased mitoses in the parencymatous tissue of liver. Some liver contained unusually large hepa tocytes with seven to ten nuclei. Occasional ly, a part of the cytoplasm of some hepato cytes appeared more eosinophilic and homo geneous while others contained eosinophilic hyaline bodies (Fig. 9). Swelling of hepato cytes, granular appearance of the cytoplasm, aggregation and .`ondensation of cytoplas mic contents either around the nucleus or periphery of the cells, fatty infiltration, and vacuoles of different sizes and shapes were also observed in less affected livers. Pyknosis and degenerative changes were observed in the connecting and collecting renal tubules of rats given multiple daily dosage during the early and middle of the study (Fig. 10). Moderate to marked de generative changes were also observed in the epithelial cells of thyroid follicles. In addi tion, the epithelial lining was disorganized, exfoliated, or with papillary projections in the lumen of the follicle (Fig. 11). The follicular colloid was either absent or ap peared foamy and thin. There were conges tion and elongation of intestinal villi, and the goblet cells appeared to be increased only during the first half of the study. Significant microscopic changes were not observed in different organs of rats given TCDD at levels of 0,1 /ig/kg (multiple daily doses), 5.0 /ig/kg (single dos), or 1.0 and 0.2 /ig'kg (multiple weekly doses). Guinea pigs--Microscopic changes in moribund or dead guinea pigs were char acterized by a severe atrophy of the cortex of thymus with destruction of lymphocytes (Table 1). Hassall bodies were large and cystic and filled with polymorphonuclear leukocytes. There was lymphoid cell deple tion in spleen and lymph nodes. Hemorrhage 133 GENP 011644 784455 Figure 9. Liver of a rat given 10 daily treatments of 10,0 ng TCDD/ kg body weight. Notice the intracytoplasmic hyaline bodies (arrows). H&E stain; magnification 950X. in the adrenal medulla extended into the zona reticularis and sometimes into the zona fasciculata. Atrophy of zona glomerulosa, and mitotic figures and loss of lipid vacuoles were observed in zona fasciculata (Fig. 12). Liver effects were confined to a diffuse single cell necrosis of hepatoeyie;:, predominantly in the periportal area. There 134 Environmental Health Perspectives 784456 GENP 011645 GENP 011646 cube 10. Degenerative changes and pyknosis of epithelial cells of renal tnbnles of a rat given S daily . reatments of 10.0 ng TCDD/kg body weight. H&E stain; magnification 731 x . September 1973 135 784457 F igure 11, Pyknoaia of nuclei, exfoliated epithelial cells and absence of colloid from the thyroid follicles ox a TCDD-treated rat (10 daily treatments, 10.0 ?g TCDD/kg body weight). H&E stain; magnification 243 X. were diffuse hemorrhages in the urinary bladder, and subserosal hemorrhages in the gastrointestinal tract. Blood pigment, prob ably hemosiderin, was observed in the med ulla of cervical lymph nodes and lamina propria of the cecum. The transitional epi thelium of urinary bladder appeared markedly hyperplastic, and the submucosal layer was edematous (Fig. 13). Occasional ly, there were focal necrosis of smooth 136 muscle cells and infiltration of inflammatory cells in the urinary bladder. There was slight to moderate atrophy of thymic cortex of guinea pigs given weekly doses of TCDD of 0.2 ftg/kg. Mice--There was some hepatocellular swelling and a relative reduction of lymp hoid cells in the thymus of mice 21 days after receiving a single 50 fig/k g dose cf Environraental Health Perspectives 784458 GENP 011647 Figure 12. Atrophy of zona glomerulosa (G) and mitoses (arrows; in the zona fasciculata of adrenal gland of a guinea pig given 4 weekly doses of 1.0 ug TCDD/kg body weight. H&E stain; magnification 176X. TCDD. However, microscopic changes were variable from one animal to another. No changes were observed in these tissues in mice necropsied 7 or 35 days after TCDD exposure. No effects were seen in mice given lower doses. Discussion The main target organs of TCDD appear to be the liver of rats and the thymus of rats, guinea pigs, and mice. TCDD caused lymic atrophy in all three species. Liver lesions severe enough to account for death were seen only in rats. The liver lesions seen in guinea pigs and mice were mild by comparison. Other lesions such as hemor rhages, degenerative changes in kidney, thyroid, megakaryocytes in spleen, ovar ian follicles, lymphoid depletion of the spleen and lymph nodes in rats, and hemor rhages, hyperplasia of the urinary bladder mucosa, and atrophy of the adrenal zona glomerulosa in guinea pigs were also seen. The liver lesions in rats dying after TCDD exposure were strikingly similar. This hepatotoxic reaction was characterized by such degenerative changes as swelling of hepatocytes, fatty metamorphosis and ultimately necrosis in rats which received 10 fig TCDD/kg-day for 10-13 days. At this time there was also an increase in serum transaminases activities (5). There after, the hepatic lesions progressed and the weight of the liver decreased as more par enchymatous tissue was destroyed. Hyper bilirubinemia and hypoproteinemia (5) be came inadequate and death soon followed in the icteric animals. Besides these degenera tive lesions, large multinucleated giant hep atocytes were also seen in liver of TCDD treated rats. The presence of these cells, in creased numbers of mitotic figures and pleomorphism of cord cells suggest that a long term study should be done to assess the possibility of the development of hyper plastic nodules and/or neoplasm. Increases in liver weight (4) can be explained by swelling of hepatocytes, fatty metamorph osis, proliferation of smooth and rough en doplasmic reticulum as shown by electron microscopy (6), concomitant with induc tion of hepatic microsomal enzymes (7). Although a marked difference in the amount of TCDD needed to produce thymic atrophy occurred between species, the sever ity of the lesion was dose-dependent within a species (4). It was also the organ most sensitive to TCDD treatment (4). In both guinea pigs and mice, the cell-mediated im mune response was depressed; lymphopenia also occurred. In contrast, cell-mediated im munity was not suppressed and lymphopenia September 1973 137 784459 1r ~ \ J FlGUBff 13. Urinary bladder o< a guinea pig treated with 4 weekly doses oi 1.0 Mg TCTD/kg body weight: (A) Notice marked hyperplasia oi epithelial cells, congestion, hemorrhage, and edema; (B) infiltration cv inflammatory cells into the submucosal and muscular layers. H&E stain; magnification 214x . 138 E nvironm ental H ealth Perspectiv et 784460 GENP 011649 v^ J i not occur in the-rat (5). The only other lesion seen with some consistency was hemorrhage of various orr;ins in dead or dying rats and guinea pigs. A'possible cause of these hemorrhages might oe the thrombocytopenia seen in these two species (5). The thrombocytopenia might be due to the reduction in number of and necrosis of megakaryocytes seen in the spleen and bone marrow of rats. However, in another study, megakaryocyte numbers and morphology were not altered by TCDD treatment (9). The hemorrhages might be aused by platelet deficiency or abnormali ties in the clotting mechanism (9). Porphyria was not seen in rats and guinea pigs dying from TCDD which agreees with the finding that TCDD did not increase hepatic S-aminoIevulinic acid synthetase ac tivities in rats (10). However, a marked increased activity of this enzyme was de monstrated in chick embryo liver (I I). Therefore, it appears that species differ ences to the porphyrogenic action of TCDD ^list- nummary Gross pathologic and histopathologic ex aminations were performed on rats, guinea pigs and mice treated with TCDD. In rats and guinea pigs, dose ranged from no effect to one which produced death; frequency of dose ranged from a single intubation to daily or weekly administration- for several weeks. Lymphoid organs, primarily thymus, were consistently affected over a wide spec trum of dose ranges in all species examined. Atrophy of the thymus, as denoted by a dose-related decrease in weight and decrease in cortical thymocytes, was a very sensitive index of TCDD exposure. Relative depletion of lymphoid cells in spleen and lymph nodes was also observed. The liver was the only other organ in which microscopic changes were observed consistently. The degree of hepatic involve ment seemed dose-dependent, but severity of changes produced was quite variable be tween species. The most severe hepatic effects ere seen in rats which received a lethal / September 1973 dose of TCDD. These rats, which were jaundiced, had diffuse degenerative and nec rotic changes in the liver. At sublethal dose levels, transient degenerative changes in the liver were followed by megalocytosis, regen eration and unusual numbers of multinucleated giant hepatocytes. Other changes found inconsistently were hemorrhage in the gastrointestinal tract, heart, and brain as well as organized thrombi in various organs. Moderate to marked degenerative and nec rotic changes were also observed in the epi thelial cells of renal connecting and collect ing tubules and thyroid follicles. Although degenerative and necrotic changes in liver were also observed in guinea pigs and mice, th e ' magnitude of these effects were markedly diminished. To illustrate this point, hepatic changes produced at lethal TCDD levels in the rat were severe enough to be a contributing cause of death, while hepatic changes in the guinea pig receiving a lethal dose were quite mild. Acknowledgement The technical assistance of Ms. M. W. Harris, W. A. Watson, J. 0. Taylor, F. A. Talley, M. G. Matheson, and A. B. Duke is gratefully acknowledged. REFERENCES 1. Schulz, K. H. Clinical picture and etiology of chloracne. Arbeitsmed.-Sozialmed. Arbeitshyg. 3: 25 (1968).. 2. Buu-Hoi, N. P., et al. Organs as targets of "dioxin" (2,3,7,8-tetrachIorodibenzo-p-dioxin) intoxication. Naturwiss. 59: 174 (1972). 3. Allen, J. R., and Carstens, L. A. Light and elec tron microscopic observations in Macaca muiatta monkeys fed toxic fat. Amer. J. Vet. Res., 28: 1513 (1967). 4. Harris, M., et al. General biological effects of TCDD in laboratory animals. Environ. Health Perspect. No. 5: 101 (1973). 5. Zinkl, J. G., et al. Hematologic and clinical chemical effects of 2,3,7,8-tetrachlorodibenzodioxin in laboratory animals. Environ. Health Perspeet. No. 5: 100 (1973). 6. Fowler, B. A., et al. Ultrastructural changes in rat liver cells following a single injection of 139 784461 TCDD. Environ. Health Perspect. No. 5: 141 (1973). ------ 7. Lucter, G-, et al. Studies on TCDD-induced changes in rat liver microsomal and mitochon drial enzymes. Environ. Health Perspect. No. 5: 199 (1973). 8. Vos, J. G,, et al. Effect of TCDD on the immune system of laboratory animals. Environ. Health Perspect. No. 5: 149 (1973). 9. Weissberg, J,, and Zinkl, J. G. Effects of TCDD upon hemostasis and hematologic function in the rat. Environ. Health Perspect. No. 5: m (1973). 10. Woods, J, Studies of the effects of TCDD on mammalian hepatic i-arainolevulinic acid syn thetase. Environ. Health Perspect. No. 5: 221 (1973). 11. Poland, A., and Glover, E. Studies on the mech anism of action of the halogenated dihenac-pdioxins. Environ. Health Perspect. No. 5: 245 (1973). GENP011651 140 E nvironm ental H ealth Perspectives 784462 Ultrastructural Changes in Rat Liver Cells Following a Single Oral Dose of TCDD by Bruce A . Fow ler,* George W . Lucier,' Hayes W. Brown,* and O .S . McDaniel* Ultrastructural alterations of liver paren chymal cells involving the endoplasmic re ticulum are known to occur in animals ex posed to chlorinated diphenyl-p-dioxins in the diet (l-U ). The observed changes are similar to those reported for other aromatic / 'Marinated hydrocarbons (5-5) and are illy associated with induction of hepatic ... ,,rosomal enzyme systems (7, 9, 10). The present study was undertaken to cor relate ultrastructural changes in ra t liver cells with biochemical studies on liver microsomes and mitochondria for a period of 28 days following a single oral dose of 2,3, 7,8-tetrachlorodibenzo-p-dioxin (TCDD). Materials and Methods Ninety male Charles River rats were sep arated into three groups of 30 each. Ani mals in the first group received a single oral dose by gavage of 2,3,7,3-tetrachlorodibenzop-dioxin in 0.5 ml of acetone and corn oil at a dose of 5 /*g/kg while the second group was given a dose of 25 tg/kg. The third group served as controls. Five animals from each group were killed by decapitation at 1, 3, 6, 9,16, and 28 days after treatment. Blocks of liver tissue were fixed in buf- National Institute of Environmental Health Sci ences, National Institutes of Health, F.O. Box 12233, "search Triangle Park, North Carolina 27709. fered formalin, embedded in paraffin, sec tioned, and stained with hematoxylin and eosin for light microscopy. Other blocks of liver were placed in a giutaraldehyde-formaldehyde fixative described previously (11), embedded in Epon, and sectioned with dia mond knives. These sections were doublestained with lead citrate and uranyl acetate prior to examination in a Philips EM 300. Results Histologic sections of liver from all treated animals were indistinguishable from controls. Liver parenchymal cells of all animals killed on the first day after treatment exhibited ultrastructural archi tecture similar to controls (Fig. 1). By the third day, liver cells of treated animals con tained increased amounts of smooth endo plasmic reticulum (SER) which were pro minent around the periphery of cells par ticularly in areas adjacent to bile canaliculi (Fig. 2). More SER was observed in rats given the higher dose of TCDD. Some cells of treated animals also appeared to- contain more rough endoplasmic reticulum (RER) than controls. Large aggregates of SER and massive amounts of RER (Fig. 3) were observed in the liver cells of treated animals killed on days 6 and 9. This observation was most readily appreciated in rats given the v.lptember 1973 141 784463 F ig u r e l. Liver parenchymal cell irom an animal killed 1 day after dioxin treatment displaying usual aanrcdhibteilcetucraen.alNicuuclluesus{*()Na)xempitroecsheonntdirnia th(eM)c,ytroopulagshm.end1o5p,1la0s2mxic. reticulum. (RER), cytosomes (arrow) *42 Environm ental H ealth Perspectives 784464 2 * * * 011653 Figure 2. Liver cell from a rat 3 days after exposure* to dioxin. Increased amounts of smooth endoplasmic reticulum (SER) are present around a bile canaliculus (*). The RER also seems to be more prominent. 20,520 X ieptember 1973 784465 \ *"a* v iv V 95 t& - :?:- S65S5i3i3t n *- -t5Ni & fi\sw 6 sp I* r-a& S f c v ^ ; "H V W isfr V --".' i_. L-***7ij-:"'vk+V_S % 3 ? * sas E - i ' i y A t'A-V-i 3k- . _JttVtew3fe-.' -V v^i ^ g p i ^ C S "- s m * ~ - - j a . v . - -r*- - . - ? X- Ftctras 3. Cell from a rat killed 6 days after administration of dioxin. Large amounts of SER are see: around the cellular periphery and massive amounts of RER surround the nucleus. 11,080x . 144 Environm ental H ealth Perspective 784466 G E h f P 1 1655 F igure 4. Parenchymal cell from a rat 16 days after treatment. Large amonnts of RER are still promi nent in the cytoplasm but the SEE is greatly diminished. 20,520x . epteraber 1973 145 784467 Hi GENP 011656 GENP 01165 P iccoumrepa5r.abLleivetor cceolnlstroolfs. a 1r5a1t0kxi*lled 28 days after treatment containing levels of endoplasmic reticulum 146 Environmental Health Perspectives 784468 ' \ r dose of TCDD. Parenchymal cells of < killed on day 16 contained little SER ;r`iiC'[arffe amounts of RER were still evident n many cells (Fig. 4). On the day 28, most ::vi,r cells of treated animals were indis.`,,...ishable from controls (Figure 5). M, ;:i)i*inous concentric whorls of endo r m i e reticulum described in other studies ' ; ) were never observed in liver cells of u,v treated animals during the course of :he experiment. Discussion P ro liferatio n of SER in liver cells of rats . :i TCDD confirms previous reports by investigators (4-). The increase of 5ER in parenchymal cells of treated animals used in this study was closely correlated with induction of microsomal enzyme ac tivity in livers of these same animals (12). Similar observations have been made for a wide variety of other aromatic chlorinated compounds (7, 9, 10) and are indicative of a l ilular attem pt at detoxification. -- rly proliferation of SER around bile 1 /iculi of liver cells from treated animals lu.a suggest involvement of bile in the metabolism or excretion of TCDD. This idea is supported by the fact that increased bile secretion was noted (13) in other animals used in this study and rats given labeled nctachloro-dioxin excreted most of the radioctivity in the feces (14). Alterations in RER associated with the formation of membranous concentric whorls of endoplasmic reticulum (3, 4) have been noted in livers of animals exposed to chlorin ated diphenyl-p-dioxins in the diet. In the present study, proliferation of RER reached a maximum at day 6 following administra tion of TCDD and was correlated with in creased RER as determined by microsomal 'Ubfractionation (12). These observations would suggest a concomitant stimulation of both RNA and protein synthesis within liver cells of treated animals. The signi ficance and mechanism of this are unclear, but they might be related to increased synthesis of microsomal proteins (3). -'Hie lack of membranous concentric whorl formation in liver cells of TCDD-treated rats used in -this study may suggest that chronic exposure and accumulation of TCDD or its metabolites within liver cells is neces sary for the phenomenon to occur. Another possibility is that other dioxins or aromatic chlorinated compounds rather than TCDD were responsible for the membranous whorls observed in earlier studies. Investigation utilizing different doses, other purified di oxin compounds, and dioxin tissue analyses may be necessary to resolve this question. In conclusion, a single low-level dose of TCDD has been found to exert a profound effect on both the SER and RER of rat liver parenchymal cells. At present, the nature of these changes appears to be re lated to an induction phenomenon and changes in cellular RNA and protein metab olism. Many other studies are needed to ex amine the mechanisms responsible for these alterations in normal cellular function. Acknowledgement We wish to thank Mrs. Patricia Parker for her excellent technical assistance in preparing sections for electron microscopy. REFERENCES 1. Allen, J. R., and Carstens, L. A. Electron micro scopic alterations in the liver of chickens fed toxic fat. Lab Invest. 15: 970 (1966). 2. Allen, J. R., and Carstens, L. A. Light and elc tron microscopic observations in Macaco mulatta monkeys fed toxic fat. J. Vet. Res. 23: 1513 (1967). 3. Norback, D. 9., and Allen, J. R. Morphogenesis of toxic fat induced concentric membrane arrays in rat hepatocytes. Lab. Invest. 20: 338 (1969). 4. Norback, D. H., and Allen, J. R. Chlorinated aromatic hydrocarbon induced modifications of the hepatic endoplasmic reticulum: Concentric membrane arrays. Environ. Health Ferspect. No. 1: 137 (1972). 5. Ortega, P. Light and electron microscopy of dichlorodiphenyltrichloroethane (DDT) poisoning in the rat liver. Lab. Invest. 15: -657 (1966). 6. Ortega, P. Partial hepatectomy in rata fed dichlorodiphenyltrichloroethane (DDT). Amer. J. Path. 56: 229 (1969). 7. Hutterer, F., et aL Hepatocellular adaptation and injury: Structural and biochemical changes following dieldrin and methyl butter yellow. Lab. Invest. 20: 466 (1969). mber 1973 147 784469 8. Nishizmi, M. Light and electron microscopic study of chlorobiphenyl poisoning in mouse and monkey liver. Arch. Environ. Health.. 21: 620 (1970). 9. Gillett, J. W., and Chan, T. M. Cyclodiene in secticides as inducers, substrates, and inhibitors of microsomal expoxidation. J. Agr. Food Chem. 16: 590 (1968). 10. Kinoshita, F. K., Frawley, J. P., and DuBois, K. P. Quantitative measurement of induction of hepatic microsomal enzymes by various dietary levels of DOT and toxaphene in the rat. Toxic. Appl. Phara. 9: 505 (1966). 11. Fowler, B. A. The morphologic effects of dieldrin ments of rat kidney proximal tubules Path. 69: 163 (1972). Am er. 12. Lucier, G. W. et al. Studies on TCDD-induced changes in rat liver microsomal and mitochon drial enzymes. Environ. Health Perspect, No s' 199 (1973). 13. Hwang, S. W. Effect of TCDD on biliary ex. cretion of indocyanine green. Environ. Health Perspect. No. 5: 227 (1973). 14. Norback, D. H., Engblom, J. F., and Allen, J. R. Chlorinated dibenzo-p-dioxin distribution within rat tissues and subfractions of the liver. Environ. Health Perspect. No. 5: 233 (1973). k Effect of 2,3,7,8-Tetrachlorodibenzo-p-dioxin on the Immune System of Laboratory Animals by j.G . V o s / I .A . M oore/ and J.G . M l * Introduction 2,3,7,3-TetrachIorodibenzo-p-dioxin (TCDD) is one of the most toxic compounds known. TCDD and other chlorinated dibenzodioxins have been associated with occuparional chloracne in workers engaged in the .nanufacture of technical chlorophenols and sir derivatives such as the herbicide 2,4, richlorophenoxyacetic acid (2,4,5-T) (I), in lethal liver necrosis in rabbits (2), and in the chick edema disease (3). More re cently TCDD has been shown to be highly teratogenic (-4). Laboratory studies have found TCDD to cause severe atrophy of the thymus at sublethal dose levels (5, 5). Because the thymus is the central lymphoid organ for cellmediated immunity and other environment al chemicals have produced immunosuppres sion (7), studies to test the immune re sponse in TCDD-treated laboratory animals were conducted. Methods selected to assess cell-mediated immunity were delayed type hypersensitivity to tuberculin in guinea pigs and rats, and a local graft versus host reaction in mice. The effect of TCDD on the humoral immunity was also studied in guinea pigs. "National Institute of Environmental Health Sci ences, National Institutes of Health, F.O. Box 12233, Research Triangle Park, North Carolina 27709. Materials and Methods Studies on the Effects of TCDD on Humoral and Cell-Mediated Immunity in Guinea Pigs In each of two experiments, groups of 10 female Hartley strain guinea pigs received 8 weekly doses of 0, 0.008, 0.04, 0.2, or 1.0 fig TCDD/kg body weight. The TCDD in an acetone-corn oil mixture was administered orally in a volume of 1 m l/kg body weight. The animals (mean weight 256 g, ranging be tween 208 and 301) were housed in groups of five and allowed free access to food and water. Body weights were determined weekly. The guinea pigs were killed with carbon dioxide gas. Heart blood was used for de termination of total leukocyte (using a Coul ter Counter, Model B) and differential leuko cyte counts. The pathologic effects of TCDD, except for the effects on lymphoid organs and skin (second experiment) are reported elsewhere {6,8,9). In one experiment, the effect of TCDD on humoral immunity was determined by measuring guinea pig response to a sub cutaneous injection of tetanus toxoid. The purogenated toxoid (Lederle Laboratories, Pearl River, New York) was administered in a volume of 0.1 ml into the right hind foot pad at day 28 (1 Lf. tetanus toxoid, aluminum phosphate-adsorbed) and again at day 42 (1 Lf. tetanus toxoid, unadsorbed). Blood was collected (10) on days 35 and 49, ptember 1973 149 784471 and from the heart at the end of the experi ment. The serum tetanus-antitoxin concen trations were determined by using a modi fied single radial immunodiffusion technique (11); the agar contained appropriate dilu tions of tetanus toxoid and the serum (10 ,*1) was allowed to diffuse from the well. The effect of TCDD on cell-mediated im munity was studied in a second experiment by measuring the delayed-type hypersensitiv ity to tuberculin. Guinea pigs were sensi tized by a subcutaneous injection (0.05 ml) of an oil suspension containing killed Myco bacterium tuberculosis H srRa (complete H,1T Ra adjuvant) (Difco Laboratories, Detroit, Michigan) in the hind foot pad a t day 35. The delayed hypersensitivity was tested by intradermal tuberculin injections [1.25 fig tuberculin PPD (Parke-Davis, Detroit, Michigan) in 0.1 ml diluent] on days 47 and 54. The diameter (mm) of the skin reactions was determined 24 and 48 h r after the tuberculinations.. In addition, the thickness of the reaction was determined by measuring the thickness of the tuberculin reaction and of the normal skin. Subtraction of the latter from the former value gave the thickness of the tuberculin reaction. To evaluate a possible indirect immunosuppressive effect by adrenocortical hyperfunction, pooled se rum samples were analyzed for cortisol and corticosteron levels by using the double isotope derivative method (12) (New Eng land Nuclear, Boston, Massachusetts). TCDD Effect on Cell-Mediated Immunity in Rats Groups of 10 random bred female albino rats (CD stock, mean weight 185 g, ranging from 165 to 201) were given oral doses, week ly for 6 weeks, of 0, 0.2,1.0, or 5.0 fig TCDD/ kg body weight. A skin test, similar to that in guinea pigs, was performed by injecting 0.05 ml oil suspension of killed Mycobacter ium tuberculosis in the right hind foot pad at day 28 and by injecting 5 fig tuberculin PPD in 0.1 ml diluent into the skin of the shaved flank at day 42. The diameter and thickness of the reactions were measured after 24 and 48 hr. Half of the animals of each group were euthanatized and necropsied at day 45. The weight of lymphoid organs and adrenals was determined as were total and differential leukocyte counts. Tissues including the skin at the site of the tuber culin injection, were processed for histology Liver pathology is published elsewhere (8), Effect of TCDD on Cell-Mediated Immunity in Mice. In this experiment, the graft versus host activity of donor (C57B1/6) spleen cells was measured by injecting them into the right hind foot pad of hybrid recipient mice, BD=F,, obtained by mating the parental strains C57B1/6 and DBA-2 which differ from each other at the major histocompati bility [H-2] locus. In this situation, donor cells are tolerated by the hybrid because the g raft does not possess any antigens which are foreign to the host, but the donor cells react against DBA-2 antigen of host cells. The weight of the enlarged injected lymph node, which is due to the immunological reaction, compared with the weight of the uninjected left popliteal nodes (right/left ratio) was the parameter used for the graft versus host activity of the donor spleen cells (J. A. M. Kerckhaert, personal communica tion, 1972). Groups of five to seven male, 2-month-old donor mice (housed individually, mean weight 24.4 g, ranging between 22.0 and 27.2 g) were orally dosed weekly with 0, 0.2, 1.0, 5.0, or 25 fig TCDD/kg body weight (in a volume of 0.1 ml per 20 g body weight) for 4 weeks. A fter 4 weeks, the animals were killed with carbon dioxide gas. Thymuses were weighed and processed for histology. Cell suspensions were made from the pooled spleens of each donor group. Spleens were minced with scissors and the fragments were gently pressed through a nylon gauze (200 fi pore diameter). The cells were washed with a MEM (Eagle) suspension culture medium (Flow Laboratories, Rockville, Maryland) containing 100 units of penicillin and 100 fig streptomycin per milliliter. Cell suspensions were counted with a Coulter Counter. Viability was determined with ery- 150 Environm ental H ealth Perspectives 784472 GENP01 1 6 6 1 GENP 011662 F ig u r e 1. Thymus oi a control guinea pi? from the skin test experiment. Note the densely packed lyraphoxcytes in the cortex (C) and the Hassall bodies (arrow) in the medulla (SI). Hematoxylin and eoain;* 61 x jptember 1973 151 -- - ------------------------------------------------------ 784473 of lymphoid cells on microscopic examina tion of the lymph nodes. The data obtained in the four experiments are presented as mean values and standard deviations. Dunnett's multiple comparisons test (13) was used to make treatm ent con- F ig u r e 2.- Thymus of guinea pig that was killed when moribund on day 27 after receiving four weekly oral doses of 1 Mg TCDD/kg. A severe cortex atrophy (C) can be seen, with destruction of lymphocytes that are phagocytized by macrophages ("starry-sky" appearance). Large cystic Hassall bodies filled with polymorphonuclear leukocytes are present in the medulla (M). Hematoxylin and eosiu; C- >* Environm ental H ealth Perspectives 784474 GENP 011663 trol comparisons, usually two-sided, except y for skin reaction variables (one-sided). In addition, a nonparametric test (Jonckheere's test) (14) was used to test for monotonic dose-response relationships. In the case of organ weights, only the organ/body weight ratios were tested. Results Experiments with Guinea Pigs All guinea pigs treated with TCDD at the 1 fig/k g level died or were killed when mori bund between 24 and 32 days (mean 28 days). They showed severe weight loss (5), lymphopenia (9), and depletion- of the lymphoid organs, especially the thymus (5). Microscopically (Figs. 1 and 2), there was a severe atrophy of the thymic cortex with considerable destruction of lymphocytes, the nuclear debris being engulfed by macro phages ("starry sky" ). Large cystic Hassall bodies, filled with polymorphonuclear leuko cytes, were seen in the medulla. All animals at the 0, 0.008, 0.04, and 0.2 fig/'kg levels survived in both experiments. Body weights, organ weights and leukocyte counts are given in Tables 1 and 2. Weight gain was significantly lower in both 0.2 fig/ kg groups. Absolute thymus weight was sig nificantly reduced at the 0.04 (experiment 1) and 0.2 g/kg (experiments 1 and 2) dose levels, but the relative (organ to body weight ratio) weights only at the 0.2 p g /kg levels. The absolute weights of the superficial cer vical lymph nodes (experiment 1) were sig nificantly decreased in the 0.2 pg/kg group. At this level,- the relative adrenal weights were significantly increased in both studies. Total leukocyte values were significantly de creased in the 0.04 ig/k g group of experi ment 1 and in the 0.2 pg/kg group of ex- Table 1. Body weights, organ weights, and leukocyte counts of guinea pigs treated with TCDD for 3 weeks (tetanus toxid stimulation).4 Pinal body weight. g Organ weights, mg Thymus Spleen Cervical lymph nodes Right popliteal lymph node Adrenals Organ/body weight ratio x 10* Thymus Spleen Cervical lymph nodes Right popliteal lymph node Adrenals Total leukocytes x 10* per mm* Lymphocytes X 10" per mm* 0 580.2 34.3 901 * 246 1036 230 224 * 35 33.4 s :13.7 294 33 1.54 0.39 1.79 * 0.50 0.3S6 * 0.055 0.0584 0.0257 0.507 0.047 6.41 * 1.88 4.16 1.47 Weekly TCDD dose 0.008 Mg/kg 0.04 pg/kg 597.2 ** 37.5 741 -* 163 858 144 202 33 36.8 11.8 294 42 551.9 48.9 672 161' 870 - 174 199 33 38.9 12.4 302 38 1.24 0.27 1.44 0.22 0.338 -- 0.053 1.21 1.58 0.361 0.0612 0.0182 0.0704 0.490 0.049 0.548 5.05 LIB 4.85 2.39 0.73 B 2.87 0.24 0.30 0.064 0.0201 0.060 1.24 e 1.03' 0.2 Mg/kg Doseresponse test 497.1 37.8 * 476 70 6 793 -ft* 133 179 32' 30.5 12.4 285 * 36 P <0.01 0.96 0.14 0 1.59 0.23 0.362; *: 0.073 P <0.01 NS NS 0.0613 0.0268 NS 0.595 0.064 * P <0.01 4.91 -4* 0.99 P <0.05 2.59 0.46* P <0.05 = 1Mean values SD, 10 animals per group except at the 0.2 pg/kg level (7 animals). *P < 0.01. *P <0.05. ptember 1973 153 784475 priment 2. Significantly decreased lympho cyte counts were found in experimental at all 3 dose levels. Significant monotonie doseresponse relationships were determined for body weights (decrease), relative -thymus weights (decrease), relative adrenal weights (increase), and total leukocyte and lympho cyte counts (decrease). Serum cortisol and corticosteron values that were measured only in experiment 2 (Table 2) were the same in all four groups. At microscopic examination of the lymphoid organs and adrenals, no ef fects were seen except for slight cortical atrophy of the thymus at the 0.2 g/kg level. Tetanus Toxoid Stim ulation: Guinea Pigs Serum tetanus antitoxin concentrations are given in Table 3. Seven days after the first tetanus toxoid injection (day 35) there was a small but significant increase at the 0.008 and 0.04 ng/kg levels. Serum antitoxin values were significantly decreased at the 0.2 pg/kg level in the secondary response (days 49 and 56), but there was no signifi cant effect at the 0.008 and 0.04 /*g/kg dose levels. Weights and histology of the injected right popliteal lymph nodes were the same in all four groups. Skin Test: Guinea Pigs The diameter and thickness of the skin reactions, measured 24 and 48 h r after tub erculination are given in Table 4. The dia meter of the skin reactions in the 0.04 /ig/kg group were significantly reduced when meas ured 48 h r after the first, and 24 and 48 hr after the second intradermal tuberculin in jection. At the 0.2 pg/kg level, all values (dia meter and thickness skin reaction) were significantly decreased. The dose response test showed also a highly significant decrease of all data. Microscopically (Fig. 3a, 36), Table 2. Body weights, organ weights, leukocyte counts, and serum corticosteroid levels of guinea pigs treated with TCDD for 3 weeks (skin test).' Final body weight, g Organ weights, mg Thymus Bight popliteal lymph node Adrenals Organ/body weight ratios x 10* Thymus Bight popliteal lymph node Adrenals Total leukocytes x 10*per mm* Lymphocytes x 10* per mm* Corticosteroids (pooled samples), pg/100 ml Cortisol Corticosteron 0 572.9 64.5 760 A 208 163 A 99 290 a 42 1.32 0.30 0.278 0.141 0.506 a 0.044 8.84 A 4.21 4.27 2.29 93.3 1.4 Weekly TCDD dose 0.008 *igAST 0.04 ftg/kg 573.3 A 64.2 819 A 216 130 A 33 295 A 36 529.6 A 61.4 649 A 133 90 A 61 267 a 41 0.2 iig/ke Dose- response test 473.1 A 41.6 6 409 A 139 6 82 * 25 236 A 28 P <0.01 1.43 * 0.31 0.229 0.059 0.519 a: 0.070 8.23 A 2.43 4.23 A 1.31 1.23 A 0.22 0.171 a 0.114 0.505 0.064 7J26 a 1.66 3.72 a 1.17 0.36 A 0.28 P < 0.01 0.176 A 0.059 P <0.05 0.612 0.102* P <0.05 6.14 A 2.91* P <0.01 3.63 A 2.37 P <0.05 92.5 85.2 88.2 L2 1.5 1.3 *Mean values A SD, 10 animals per group. 6 P <0.01. P <0.06. 154 Environmental Health Perspectives GENP 011665 784476 F ig u r e 3. Forty-eight hour akin reaction to 1.25 Mg tuberculin PPD in guinea pigs sensitized 17 days . earlier with 0.05 mi of an oil suspension containing killed Mycobacterium tuberculosis Ha Ra: (a) con trol animal, note the edema and diffuse cellularity of the dermis (D) and the infiltration both diffuse and around small vessels in the subcutaneous adipose tissue (A), muscle (M) and connective tissue (C); (6) guinea pig treated with 8 weekly doses of 0,2 Mg TCDD/kg. There is less edema and less cellularity in the dermis and the focal cellular infiltration is much smaller. Hematoxylin and eosin; 52 X. mber 1973 15a 784477 ON ON ON there was less cellularity and less edema in the skin sections of the 0.2 /*g/kg groups which explains the reduced thickness of the skin at the site of the tuberculination. The cellularity as found in the controls con sisted mainly of mononuclear cells, both dif fuse and around vessels, was clearly de creased in the dermis, subcutaneous adipose tissue, muscle, and connective tissue of the 0.2 fig/kg' treated animals. In contrast to the former experiment, there was a significant decrease (dose-response test) in the relative weight of the popliteal lymph node of the injected right hind leg (Table 2). Granulo mas, probably caused by the adjuvant ir jection, were seen in all these lymph node* Experiment with Rats All animals survived the experiments period. Body weights, organ weights, an; leukocyte counts of the animals killed afte 45 days are given in Table 5. Body weight and absolute and relative thymus weight were significantly reduced a t the 5.0 ^g/kj level. A significant increase in relative spleei weight is also seen at that dose level. Th effect of TCDD on adrenal weight in the ra differed from the response in the guinea pig Table 3. Serum antitoxin values as measured by single radial immunodiffusion of guinea pigs treated with TCDD for 8 weeks (tetanus toxoid stimulation)/ Weekly dose of TCDD, /igAg 0 0.008 0.04 0.2 ' Dose-response test Tetanus antitoxin concentrations, International Units/ml Primary response _______________Secondary response (day 35) Day 49 Day 56 5.55 0.69 6.42 0.64 * 6.71 * 0.61* 5.29 0.82 NS 259.7 i 99.7 225.4 98.5 203.2 57.7 149.2 32.8 c P <0.01 246.0 92.8 265.4 79.9 230.0 51.6 175.8 * 47.8 P =0.05 1Mean values SD, 10 animals per group, except at the 0.2 tg/kg level (7 animals). The animals were injected with tetanus toxoid at days 28 and 42. Antibody concentrations were measured at days 35, 49, and 56 (7, 21, and 28 days after the first tetanus toxoid injection). bP <0.05. ' P <0.01. Table 4. Skin reaction (delayed hypersensitivity to tuberculin) of guinea pigs treated with TCDD for 8 weeks/ Weekly _______________Diameter skin reaction, mm_______________ Thickness skin reaction, mm TCDD, ;*g/kg Tuberculineation on day 47 24 hr 48 hr Tuberculination on day 54 24 hr 48 hr (tuberculination on day 54) 24 hr 48 hr 0 0.008 0.04 0.2 Doseresponse test 18.20 1.87 15.15 1.72 19.45 1.23 -15.20 1.90 2.23 0.49 17.50 1.35 13.85 1.67 18.45 * 1.23 13.85 1.73 2.10 * 0.60 16.80 1.83 12.90 1.31* 17.65 0.91* 13.40 1.13* 1.93 0.46 12.90 2.56* 8.80 2.98* 15.20 2.91* 8.95 3.25* 1.13 0.43* 1.73 2.00 1.38 0.63 0.43 0.54 0.34 0.41* P < 0.01 P < 0.01. P < 0.01. P <0.01. P < 0.01. P < 0.01 *Mean values SD, 10 animals per group. The animals were sensitized on day 35 with 0.05 ml of an oil suspension containing killed Mycobacterium tuberculosis. Intradermal tuberculination (1.25 Mg tuber culin PPD) was performed on days 47 and 54. Skin reactions were measured 24 and 48 hr after tu berculination. bP <0.05. ' P < 0.01. 156 E nvironm ental H ealth Perspectives 784478 GENP 011667 Absolute adrenal weights decreased significantiy in the ra t at the 1.0 and 5.0 f t g / k g (Dse levels. The dose-response test for the screase in relative adrenal weights was highly significant. Also, the total leukocyte :md lymphocyte counts in the ra t differed Yom the response in the guinea pig. There xus no lymphopenia, but some increase of both cell counts, showing a slight but not significant dose-response pattern. Microscopically, the only effect seen in the lymphoid organs and adrenals, was a slight to moderate cortical atrophy in the thymuses of the 5 ^g/kg group. Also, the cellularity of the cortex was somewhat less dense. The results of the measurements of the skin reactions are given in Table 6. Also in contrast with the guinea pig, there was no effect on the Table 5. Body weights, organ weights, and leukocyte counts of rats treated with TCDD for 6 weeks (skin test).* Weekly TCDD dose 0 0.2 a g / k g 1.0 u g / k g 5.0 f i g / k g 7 inal body weight, g 261.6 28.2 264.0 17.4 263.6 25.6 225.0 A 17.1 * Organ weights, mg Thymus Spleen Cervical lymph nodes Adrenals 318 112 507 85 43.0 16.1 80.6 10.6 285 a 50 498 95 40.0 10.3 71.2 7.2 289 A 16 536 53 ' 37.6 * 7.6 61.2 A 11.2" 132 39" 564 79 37.4 7.3 55.0 13.8' Organ/body weight ratios X 10* Thymus Spleen Cervical lymph nodes ^ Adrenals \ 1.20 0.32 1.94 A 0.24 0.163 A 0.049 1.09 A 0.27 1.89 A 0.35 0.153 A 0.043 0.312 0.059 0.269 A 0.014 1.10 A 0.13 2.04 0.15 0.142 * 0.0X6 0.231 0.028 0.59 A 0.16' 2.53 0.48' 0.167 0.032 0.242 A 0.044 Total leukocytes X 10*, per mm1 5.88 A 2.35 7.98 A 4.41 7.68 2.32 10.65 A 5.00 Lymphocytes X 10*, per mm* 4.13 A 1.63 * 5.94 A 2.80 *Mean values SD, five animals per group. bP <0.05. *F < 0.01. 5.53 A 1.89 7.65 A 3.11 Dosetest P <0.05 P <0.05 P <0.05 P < 0.01 NS NS Table 6. Skin reaction (delayed hypersensitivity to tuberculin) of rats treated with TCDD for 6 weeks.* Weekly dose of TCDD, Mg/kg 0 0.2 1.0 5.0 Diameter skin reaction, mm 24 hr 48 hr 14.65 2.94 15.15 2.49 15.40 a 1.93 14.75 A 1.83 10.95 A 2.53 9.80 A 2.74 10.30 A 2.52 10.45 2.99 Thickness skin reaction, nun 24 hr 2.10 A 0.45 1.90 A 0.59 2.23 A 0.68 1.88 A 0.70 48 hr 1.46 A 0.30 1.42 A 0.44 1.58 A 0.64 1.65 0.54 Mean values A SD, 10 animals per group. The animals were sensitized on da7 28 with 0.05 ml of an oil suspension containing killed Mycobacterium tuberculosis. Intradermal tuberculination (5 fig tuber culin PFD) was performed on day 42. Skin reactions were measured 24 and 48 hr after tuberculination. September 1973 157 99X10 <3NslO 784479 diameter the thickness, or histological ap pearance of the tuberculin reactions in the rats. Experiment with Mice '" One animal of the 25 /ig/kg group died after 24 days. Body and thymus weights of the donor mice are given in Table 7. There was no difference in the final body weights of controls and TCDD-treated animals due to lower initial weights of the controls, but there was a significant effect on weight gain at the 25 ug/kg level. Absolute and relative thymus weights were signficantly reduced at the 5.0 and 25 /*g/kg dose levels. Mean thymus weight in the high dose group was only 13fa of the mean weight of the con trols. Microscopically (Figs. 4 and 5), there Figure 4, Thymes oi a control mouse with the cortex at C and the medulla at 31. Hematoxylin and ecsiu; X 61. 1BO Environmental Health Perspectives 784480 GENP 011669 a nearly complete loss of the thymic The cellularity of the remaining cor- is less dense and, there was destruction of lymphocytes. Spleens of the animals receiv ing 25 pg/kg TCDD were very small. The rieid of spleen cells was too small to inject a sufficient number of recipients. As shown '"i* *i*> A-SfT" GENP 'icurs 5. Thymus of a moose that was-killed after receiving four weekly oral doses of 25 Mg TCDD/kg. Severe cortex atrophy is present; the cellularity in the remaining cortex is leas dense making a distinction between cortex and medulla difficult at this magnification. Hematoxylin and eosin; 61 x . nber 1973 159 784481 a\ Table 7. Body and thymus weights of donor mice treated with TCDD for 4 weeks.* Weekly dose of TCDD, Mg/kgf 0 0.2 1.0 5.0 25 Dose-response test Final body weight, g 26.50 0.29 25.84 2.80 26.34 2.38 25.76 2.10 22.02 2.44 NS Weight change, g +2.24 0.42 +1.26 2.09 +1.78 1.27 +1.04 2.13 -1.92 2.69 b ` Mean values SD, five or six animals per group. bP <0.01. Thymus, mg 51.8 4.1 41.2 14.6 40.8 10.0 29.0 5.8* 6.7 9,1b Thymus/body weight ratio X 10* 1.95 * 1.57 0.46 1.56 Q.3g 1.12 Q.i5 fc 0.31 0.09* P < 0.01 Table 8. Graft versus host activity of spleen cells from TCDD-treated donor mice injected into recipients.* Donor TCDD dose, pgfag No. 0 0.2 , 10 9 1.0 12 5.0 14 Dose-response test Recipients Right/left ratio Popliteal lymph nodes 5.83 4.32 5.82 2.47 3.62 1.47 1.98 0.72b P <0.01 *Mean values SD. Recipients (C57B1/6 x DBA-2, P -1) were injected in the right hind foot pad with 1 x 10Tviable nucleated donor (C57B1/6) spleen cells. hP <0.01. /xg/kg donor group was 38% lower than the control value. There was a highly significant monotonic dose-response relationship. Discussion As shown in Table 4 and Figure 3, it is clear that TCDD suppressed the cell-medi ated immunity in guinea pigs at the 0.2 and 0.04 ig/kg levels (measuring the delayed hypersensitivity to tuberculin). The vulner ability of the lymphopoietic system, having a high mitotic activity, is clearly demon strated by the thymus atrophy and lympho penia (Tables 1 and 2, Figs. 1 and 2). In this context, it is worth mentioning that in hibition of mitosis has been observed in dividing endosperm cells of the African blood lily when exposed to TCDD {15). Also, there was a significant reduction in the relative weights of the stimulated popliteal Iympl nodes in the skin test experiment (Table 2) Humoral immunity, measuring the antibod) production against tetanus toxoid, was slight ly depressed in the guinea pig at the 0.2 ig/kg level (Table 3). There was no effect on the weights of the stimulated popliteal lymph nodes in this experiment (Table 1). In the second guinea pig experiment, there was no difference in serum cortisol and corticosteron concentrations between the dif ferent groups (Table 2). Microscopically, there was no effect on the adrenals in the 0.2 /ig/kg groups. Therefore, it is likely that the increased relative adrenal weights (Tables 1 and 2) are only due to a decrease in body weight gain and not due to adreno cortical hyperfunction. Thus, indirect im munosuppression by stimulation of adreno cortical activity can be excluded. Besides the effect on the lymphoid system, there was only minor pathology in guinea pigs receiving a lethal dose of TCDD, i.e., mild liver injury, hyperplasia of the bladder epithelium, hemor rhages, atrophy of the zona glomerulosa of the adrenal cortex (5) and thrombocytope nia (P). Analogous to the situation in the guinea pig, there was suppression of the cellmediated immunity in the mouse. The graft versus host activity of donor spleen cells was significantly suppressed at the 5.0 pgfag level. Donor cells of the 1 pgfag group gave a g raft versus host response that was 62% oi the control value (Table 8). Thymus atrophy (Table 7, Figs.. 4 and 5) and lym pliop^1- 160 Environmental Health Perspective 784482 GENP 011671 ) were sensitive indices for TCDD ex. \.ure. As in the- case of the guinea pig, pathology was mild (5). " in contrast to the guinea pig and mouse there is quite a different situation in ',\ie Vat. Cell-mediated immunity (delayed :',vpL.[-sensitivity to tuberculin) was not sup pressed (Table 6). Possibly, this test is not ;?nsitive enough to detect an immunosup pressive effect in rats. Also, except for thy mic atrophy at sublethal dose levels, there was no lymphopenia (Table 5). In addition, - there was a dose-related decrease in both ;lfisi>Iute and relative adrenal weight (Table >cheating an adrenal hypofunction. Howt-wr. in view of the marked induction by TCDD of glucuronyl transferase in rat liver Ift), an important .enzyme involved in the metabolism of corticosteroids, adrenocortical hyperfunction regulated through the hypothalamus-pituitary-adrenocortical axis would lie expected. An inhibitory effect on adrenal hypertrophy, induced by surgical trauma, was so- :i in rats treated with FCB (unpublished data) and with p ^ -D D T ; in this latter study ' verterence with the feedback mechanism y glucocorticoid hormones homeostasis was N proposed (17). In rats, exposed to lethal concentrations of TCDD, there was severe liver injury (de generative and necrotic changes),, thyroid pathology, hemorrhages (3), platelet depres sion, increased serum bilirubin values, and increased SGOT and SGPT activities (0). Liver damage can be considered to be the major cause of death in rats exposed to TCDD. In both guinea pigs and mice, there was only mild liver injury (3). The most significant findings are seen in the lymphoid system. Suppression of the cell-mediated immunity might be the m ajor cause of death v. these species. To test this hypothesis, a >tudy is indicated to determine whether iymphoid (thymus) ceil grafts are capable to protect mice from dying when exposed to concentrations of TCDD that are lethal for the control' mice. Also, it would be worth while determining whether there is a dif ference in lethal TCDD levels in mice kept inder sterile and under conventional or itember 1973 SPF conditions. Experiments will be con ducted to determine the effect of in utero exposure of TCDD upon the cell-mediated immune rponse of the offspring. Summary In two experiments, groups of 10 female guinea pigs were dosed weekly for S weeks with 0, 0.008, 0.04, 0.2, or 1.0 j*g/kff TCDD/ kg body weight to test the cell-mediated and humoral immunity. All animals at the 1 /ig/kg level died or were killed when mori bund, they showed severe weight loss, lymphopenia,, and atrophy of the lymphoid organs. Weight gain was depressed at the 0.2 /ig/kg level. Cell-mediated immunity was assessed by vaccination of the remaining animals with an oil suspension of killed Myco bacterium tuberculosis. Thymus atrophy and lymphopenia was observed. The diameters of the skin reactions, measured 24 and 48 h r after tuberculination, were significantly re duced at the 0.2 and 0.04 pg/kg levels. In direct immunosuppression by stimulation of adrenocortical activity was excluded. The humoral immune system was stimulated in a second experiment by tetanus toxoid injec tions at days 28 and 42. Serum tetanus anti toxin concentrations were slightly depressed in the 0.2 ig/kg level at days 49 and 56. A skin test, similar to th at in guinea pigs, was done in rats treated weekly for 6 weeks with 0, 0.2, 1.0, and 5.0 /*g TCDD/kg body weight. Animals at the 5 ig/kg level had sig nificant lower body, thymus and adrenal weights. No effect was found on the skin reactions. Cell-mediated immunity was tested in mice in a graft versus host assay. Groups of donor mice (C57B1/6) were treated weekly for 4 weeks with 0, 0.2, 1.0, 5.0, and 25 /ig TCDD/ kg body weight. Weight gain was depressed at the 25 ftg/kg level. There was a remark able thymic atrophy. Parental strain spleen cells (up to the 5.0 tg/k g group) were in jected into the feet of hybrid recipients (C57B1/6 x DBA-2 F -l). The weights of the draining popliteal lymph node, as a para meter for the graft versus host activity, were significantly lower in the animals injected 161 784483 with spleen cells from the 5 /ig/kg donor group. It is concluded that TCDD -at -sublethal dose levels suppresses the cell-mediated im munity in both guinea pigs and mice. Hu moral immunity was slightly suppressed in the guinea pig. The possible role of immune suppression in the death of TCDD-treated guinea pigs and mice is discussed in view of the absence of major pathologic effects ex cept in the lymphoid system. Acknowledgements The authors are grateful to Mrs. L. D. Lawson for valuable technical assistance and to Dr. J. K. Haseman for statistical analyses. REFERENCES 1. Kirnmig, J.( und Schulz, &, H. Berufliche Akne (Sog. Chlorakne) durch chlorierte aromatische 2. zyklische ther. Dermatologies 115: 540 (1957). Bauer, H., Schulz, K. H., und Spiegelberg, TJ. Berufliche Vergiftungen bei der Herstellung von Chlorphenol--Verbindungen. Arch. Gewerbepath. Gewerbehyg. 18: 538 (1961). 3. Higginbotham, G. R., et al. Chemical and toxi cological evaluations of isolated and synthetic chloro derivatives of dibenzo-p-dioxin. Nature 220: 702 (1968). 4. Courtney, K. D.( and Moore, J. A. Teratology studies with 2,4,5-trichlorophenoxyacetic acid and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl- Pharmacol. 20: 396 (1971). 5. Buu-Ho, N. P., et al. Organisms as targets of "dioxin" (2,3,7,8-tetrachIorodibenzo-p-dioxin) in 6. toxication. Naturwias. 59: 174 (1972). Harris, M,, Moore, J. A., and Vos, J. G. General biological effects of 2f3,7f8-tetrachlotodibQIW^ _ dioxin in laboratory animals. Environ Perspect. No. 5: 101 (1973). * 7. Vos, J. G., and van Genderen, H. Toxic,, aspects of immune suppression. In: ^ 2 * and the Environment: A Continuing Conw vewy. Symposia Specialists, Miami, 1373 8. Gupta, B., et al. Pathologic effects cf 2A?#* tetrachlorodibenzo-p-dioxln in laboratory Li" mals. Environ. Health Perspect. Nc, VZ (1973) m 9. Zinkl, J. G., et al. Hematologic and cUalesi chemical effects of 2,3,7,8-tetrachlorodibento-m. dioxin in laboratory animals. Environ. Health Perspect, No. 5: 111 (1973). 10. Vallejo-Freire, A. A simple technique for r*. peated collection of blood samples from guim* 11. pigs. Science 114? 524 (1951). Mancini, G-, Carbonara, A, O., and Hererasnt, J. F. Immunochemical quantitation of antigens by single radial immunodiffusion. Immunochem. 12. 2: 235 (1965). Kliman, B.f and Peterson, R. E. Double isotope derivative assay of aldosterone in biological . tracts. J. Biol. Chem. 235: 1639 (1960). 13. Miller, R. G. Simultaneous Statistical Inference, McGraw-Hill, New York, 1966. 14. Jonckheere, A. R. A distribution-free K-aampta teat against ordered alternatives. Biometrika 41: 133 (1954). 15. Jackson, W. T. Regulations of mitosis. HI. On tological effects of 2,4,5-trichlorophenoxyacetle acid and of dioxin contaminants in 2,4,5-T for mulations. J. Cell Sci. 10: 15 (1972). 16. Lucier, G. W,, et al. Studies on TCDD-induced changes in rat liver microsomal and mito chondrial enzymes. Environ. Health Perspect. No. 5: 199 (1973). 17. Wassermann, M., et al. Effects of organochlcrine insecticides on body defense systems. Ann. N. Y. Acad. Sci. 160: 393 (1969). GENP 011673 162 Environmental Health Perspectives 784484 Carcinogenesis Bioassay of Chlorinated Oibenzodioxins and Related Chemicals by Maurice E . King,* Alan M . Shefner,' and Richard R. B a te s T Introduction Chlorinated dibenzo-p-dioxins have been identified as trace contaminants of herbicides mid pesticides. They may be formed as by products in the manufacture of chlorinated phenols and, for this reason, were impli cated in outbreaks of chick edema disease i ; . Fat trimmings obtained from animal hi-.es preserved with dioxin-containing chlor^'i'^nol products and subsequently used in en feed contained the chick edema facJhloracne had been earlier observed in workers who had handled technical quality 2.4.5-trichlorophenol (2 ). Occasionally the condition was complicated by liver involve ment and psychopathological changes. Pure 2. ;.5-trichlorophenol had no effect when testc'; in an animal bioassay system. Tetrachlororiibenzodioxin (TCDD), which was isolated from the technical mixture, produced an ef fect at a concentration of 0.005%, however (13). In addition, TDCC is the contaminant of the herbicide 2,4,5-T that produced tera togenic effects in rats (4). Subsequent work with pure TCDD has indeed shown terato genic and embryotoxic effects in rats and nice (5-7). The fact that the chlorinated dioxins are highly toxic, teratogenic, and acnegenic does not of necessity indicate they are carcinogens "Life Sciences Research Division, IIT Research Institute, Chicago, Illinois 6Q616. ^Experimental Pathology Branch, National Can cer Institute, National Institute of Health, Beth-- . esda, Maryland 20014. as well. However in view of the widespread use of products that might contain dioxins as contaminants and their extreme stability under environmental conditions, examination of their possible long-term effects is impera tive. This program was initiated to determine the chronic toxicity and potential carcino genicity of a series of chlorinated dibenzodioxins by oral administration and skin application. The unsubstituted, 2, 7-dichloroand octachlorodibenzodioxins are relatively innocuous with regard to toxicity. The 2,3,7trichloro-, 2,3,7,8-tetrachloro- and hexachlorodibenzodioxins, however, are quite toxic and require the use of specialized facil ities to minimize the possibility of human exposure. This report describes our results to date with the nontoxic dioxins. Work is being initiated with the toxic dioxins as they become available. Experim e ntal The octachlorodibenzodioxin used in these studies was prepared by Spectratec Inc., Washington, D. C. The other dioxins were synthesized by the Chemistry Division of IIT Research Institute. Swiss-Webster and B6C3F1 mice and Osborne-Mendel rats were obtained from contract supported colonies at Charles River Breeding Laboratories through arrangement with the Mammalian Genetics and Animal Production Section, Na tional Cancer Institute. Skin carcinogenesis studies will be contin ued for a total of 78 weeks. Each treatm ent am ber 1973 163 784485 GENP 011674 group consists of 30 male and 30 female Swiss-Webster mice th at are shaved weekly. In testing for complete carcinogenicity 0.2 ml of a solution of the test compound dis solved in acetone is applied three times week ly to the backs of mice. The octachloro-, dichloro-, and unsubstituted dibenzodioxin so lutions in acetone contain 0.2, 3.0, and 80 mg/ml, respectively. For the study of pro motion activity, each mouse was initially treated with 50 ig dimethylbenzanthracene (DMBA) 1 week prior to initiation of test compound application. Osbome-Mendel rats and B6C3F1 mice are being used in the oral *administration studies. Groups of 50 male and 50 female mice and 35 male and 35 female rats are receiving the dioxins at levels of 1 % and 0.5% of the diet, and similar groups are receiving 1 % and 0.5% dioxane in their drinking water. In the case of the supposedly nontoxic octachlorodibenzodioxin, in which numerous animal deaths occured after 20 weeks of feeding, additional animals were placed on test at loWer dietary levels. Results The results for the skin carcinogenesis study to date are shown in Table 1. None of the dioxin-treated mice in the complete carcinogenesis study exhibited skin tumors, although subcutaneous tumors are present in two mice treated with octachlorodibenzodioxin. Histopathological results are not yet available for all mice that have died. Of those th at have been examined, however, a malign ant lymphoma of the lymphocyte type was found in a mouse treated with unsubstituted dioxin. No other significant lesions were found in any of the mice from the complete study. The promotion study has yielded apparent skin tumors in male mice treated with un substituted dibenzodioxin or dichlorodibenzodioxin. As the animals showing these skin lesions are still alive, histologic confirma tions of this observation is not yet available. A plasmacytoma has been found in one mouse from the dichlorodibenzodioxin promotion group. Acetone was originally included in complete study as a solvent control, sine* dioxins are dissolved in it for skin an it tion. No skin tumors have been found?' either surviving mice or in tissue fmZ those that have died. The acetone promoti groups, however, were started 3 months U? er, and skin tumors have been confirmed br microscopic examination of skin of n ju mouse that died. A reticulum cell m a w ! ant lymphoma was found in one of th6 ^ males that was examined, but there were no other significant lesions. The dioxins can be considered as deriva tives of 1,4-dioxane, and this compound was included in the studies because of the struc tural similarity. One carcinoma not micro scopically confirmed and a subcutaneous tumor are apparent in the surviving mice of the complete study. Tissues from those that died revealed a reticulum cell malignant lymphoma but no evidence of skin lesions. The dioxane promotion groups are of par ticular interest since their response unex pectedly rivaled that of the croton oil posi tive controls in both mortality and mice exhibiting papillomas. The activity of croton oil and dioxane as promoting agents fol lowing DMBA initiation is shown in Figures 1 and 2. Weekly counts of papillomas and suspect carcinomas were made by gross ex amination. The fraction of mice bearing skin tumors is-nearly identical for both materials, as is the time course for tumor development (Fig. 1). However, croton oil treatment led to a much higher multiplicity of skin tumors per mouse than did treatm ent with dioxane (Fig. 2). Carcinomas were produced in both treatment groups in direct proportion to the number of papillomas present. Histopathological results from the two treatments differed in th at effects observed with croton oil were primarily neoplastic. A majority of the dioxane-treated mice had liver lesions of a mild nature (megalocyto- sis, occasional distended bile canaliculi, occas ional necrotic centrolobular necrosis, cuffed * triad vessels, general mononuclear periportal infiltration, and mild peripherolobular fibro sis). The distribution of preneoplastic and 164 Environmental Health Perspecrives 784486 GENP 011675 r i "si 03 0-N0( er Compound Octachloro- dibenzodioxin Unsubstituted dibenzodioxin Croton oil Dichlorodibenzodioxin Acetone Dioxane Treatment Complete Promotion Complete Promotion Promotion Complete Promotion Completo Promotion Complete Promotion Table 1. Summary of dioxin akin carcinogenesis data. Week of teat 60 69 69 68 66 64 63 62 60 60 . 69 Sex Number M 20 F 28 M 24 F 29 M 24 F 24 M 26 F 29 M0 F1 M 17 F 27 M 23 F 26 M 24 F 28 M 29 F 26 M 22 F 26 M4 F6 Survivors Tumor response Suspected Subcutaneous Papillomas carcinomas tumors 0 0 1\ 0 0 1J 0 0 3l _0 0 1 J 0 0 0\ 0 0 0; 1 3 2\ 0 0 0; ___ -- --\ 0 1 1} 0 0 0 0 0 0 \ I 6 2 2\ 0 0 0 0 0 0 0 0 0 \ i 8 1 1\ 0 0 0j 0 0 1\ 0 1 0j 2 3 2\ 2 3 0/ i Pathological results No papilloma or significant pathology No papilloma or significant pathology No papilloma, 1 malignant lymphoma No papilloma or significant pathology Neoplastic lesions of skin and lungs No papilloma or significant pathology No papilloma, 1 plasmacytoma No papilloma or > significant pathology 1 papilloma, 1 malignant lymphoma No papilloma, 1 malignant lymphoma Neoplastic lesions of skin, lungs, and kidney 8 99ir0dM3O I L-it Figure l. Comparative rate of promotion activity for croton oil and dioxane on male mice. neoplastic lesions in the dioxane-treated group is shown in Table 2. In treated skin, conditions ranged from hyperplasia to der mal fibrosarcoma. Squamous cell carcinoma of the nasal septum was observed in one animal which had skin papilloma. Of nine mice with lung tumors, seven had malig nant lung lesions, of which three were con sidered metastatic from other sites. The average weights of controls and test animals receiving the dioxins in their diet and dioxane in w ater are shown in Figures F ig u r e 2. Promotion activity of 1% croton oil and dioxane on Swiss-Webster mice. 166 F ig u r e 3. A v e ra g e w e ig h t o f treated fem ale and male rata. 3 and 4 for rats and Figures 5 and 6 for mice. The stimulatory effect of dioxane on weight gain in male mice and rats is most interesting, especially in view of the skin carcinogenesis results. The effect is not sz marked in females; however, after 10 weeks the weight gain of female mice at 0.5% ex ceeds th at of the controls. Generally, the dioxins cause a decrease in growth rate al though dichlorodibenzodioxin, which has not been on test as long as the other compounds, does not appear to affect growth rate at the levels tested. A summary of the results to date in the feeding studies is shown in Table 3. Despite the stimulation of growth in surviving ani mals, dioxane has caused an appreciable mortality in rats but has had little effecton mice. Chronic bronchopneumonia and chronic murine pneumonia above the back ground level were the major changes observ- Environmental Health' Perspectives 784488 GENP 011677 Table 2. Distribution of preneoplastic and neoplastic lesions in 15 mice treated wun. dioxane following DMBA initiation. Number of mice with specific'lesions ^-vpe pf lesion_______ Skin ^ r o p h y and/or hyperplasia papilloma Carcinoma tn situ. ^uamous cell carcinoma bronchial adeno- matiod lesion Uveolar adenoma ilronchiolar or alv'olar carcinoma -br ;arcoma .'ndirentiated sarcoma \lnlignant lym phom a Lymphocyte type Reticulum cell sarcoma 6 2 3 1 Nasal septum Trachea 1 1 Lu-ng 1 2 3 4 3 1 1 1 1 Spleen Kidney Liver 1 11 11 rcoBB 4. Average weight of treated female and male rats. iber 1973 ed in the rats that have been examined. Ap preciable mortality has also been observed in female mice th at are receiving 1 % unsub stituted dioxin. Hepatotoxicity was the m aj or effect noted, but one mouse exhibited broncheolar mucosal hyperplasia. The rats th at received this compound exhibited hepato toxicity and chronic murine pneumonia. Toxicity in animals fed octachlorodioxin was evidenced by both growth depression and mortality. A distinct difference in the susceptibility of the sexes to the compound was observed for mice and rats. Thus all male mice died by 10 and S weeks, respective ly, in the 1% and 0.5% levels while, a t 37 weeks, there were 5 female 1% survivors and 45 female survivors at 0.5%. On the other hand, all female rats died by 22 and 25 weeks, respectively, in the 1% and 0.5% groups while the last males died a t 32 and 37 weeks. New test groups were .set up at 0.25% for all animals and a t 0.125% for fe male rats and male mice since these groups appeared to be more drug-sensitive. Even a t these lower dose levels, all of the mice have died. ^ and mice fed Octacillorodibenzodioxin had changes ranging from early hepatotoxic 167 784489 i' G E N P 011678 Compound Controls 1% Dioxane 0.5% Dioxane 1% Unsubstituted dibenzodioxin 0.5% Unsubstituted dibenzodioxin 1% Dichlorodibenzodioxin 0.5% Dichlorodibenzodioxin 1% Octachlorodibenzodioxin 0.5% Octachlorodibenzodioxin 0.25% Octachlorodibenzodioxin 0.125% Octachlorodiobenzodioxin Table 3. Summary of dioxin oral administration data. -..-Rats (35 per group) (50 fcrcup) Week of Survivors Significant pathology Week of Signified ' pathology Sex test Lung Liver test Survivors Lung TTT M 34 35 31 60 F 34 35 31 50 M 42 24 8/11 1/11 40 50 F 42 20 3/5 2/5 43 49 M 42 26 6/6 1/6 40 49 F 42 32 3/3 1/3 43 49 M 42 33 2/2 2/2 39 48 F 42 32 1/2 M 42 31 3/4 2/2 1/4 39 34 29 50 3/14 7/14 F 42 35 39 49 M 17 35 F 17 35 M 17 35 17 49 0/1 17 48 1/2 17 50 0/1 1/2 F 17 35. 17 49 M 32 0 1/5 5/5 10 0 0/5 5/5 F 22 0 0/5 5/5 37 5 0/5 6/6 M 37 0 1/5 5/5 8 0 0/5 5/5 F 25 0 37 45 M 17 28 1/2 2/2 17 1 F 17 15 0/3 3/3 15 0 2/2 1/2 M-- ---- -- 90 F 17 30 -- -, r lesions (diffuse vacuolar degeneration, fatty metamorphosis, megalocytosis, necrosis, and cholangioiar epithelial hyperplasia) to circhosis with disrupted lobule architecture, necrosis, fibrosis, parenchymal regeneration and cholangioiar proliferation. There were areas of bronchiolar mucosal hyperplasia in the lungs of two rats, and one of these rats also had bronchiolar adenomatoid lesions. In one mouse, there were bronchiolar mucosal hyperplasia and metaplasia and one bron chiolar carcinoma in which signs of early squamous change could be seen. Discussion The preliminary results reported here are by no means conclusive with regard to the carcinogenicity of the chlorinated dioxins. The oral studies indicate primarily hepatotoxicity, especially in the case of octachloro dibenzodioxin. A supplementary analysis report from Midwest Research Institute pro vides a possible explanation for some of 168 the toxicity observed. The batch of octach* lorodibenzodioxin used in preparation of the diets was found to contain hexachlorodibenzodioxin a t a level of less than 0.1%. A dietary level of 1 % octachlorodibenzodioxin which contained this contaminant would pro vide a hexachlorodibenzodioxin intake of approximately 150 ^g/day for rats and 50 /xg/day for mice. Chronic toxicity values for this compound are not available. However, in preliminary experiments at Dow Chemi cal Co. (5), hexachlorodibenzodioxin at 100 ig/kg/day fo r 10 days caused growth depres sion, liver pathology, and embryotoxicity in rats. Tests with animals receiving the first batch of octachlorodibenzodioxin will continue as long as there are survivors. Addi tional groups of animals at dietary levels of 0.5% and 0.25% octachlorodibenzodioxin shown on analysis not to contain the hexa chlorodibenzodioxin contaminant have been started. A fter 6 weeks, many of the male mice at 0.5% have died, thereby indicating Environmental Health Perspectives 784490 GENP 011679 Time (weeks) Time (weeks) [c u k e 5. A v e ra g e w e ig h t o f tre a te d fe m a le and male mice. F ig u r e 6. A verage w eight of treated fem ale and m ale mice. possible toxicity of the octachlorodibenzodioxin itself. A comparative feeding and in tubation study is currently underway to determine if the octachloro causes a decrease in dietary intake. The skin carcinogenesis assay for the ontoxic compounds is nearing completion, and no skin tumors have been observed in the complete carcinogenesis studies. The na ture of the internal growths must be de termined at necropsy. The fact th at only male mice have exhibited skin tumors in the pro motion studies can probably be attributed to the additional factor of wounding. As the ember 1973 mice are housed in groups of 10, there is considerable fighting among the males, result ing in scars on the backs of the less aggres sive animals. Boutwell (5) has shown that wound healing plays an integral role in the promotion of skin carcinogenesis, and fight ing among males treated with unsubstituted and dichlorodibenzodioxin as well as ace tone may have contributed to the promotion of skin tumors. No explanation can be given at this time for the lack of tumors in the octachlorodioxin group, since this compound was also dissolved in acetone. The positive control group treated with 169 784491 DMBA and croton oil responded as expected in terms of skin tumor promotion. The extent and variety of the lesions produced by DMBA and dioxane were unexpected, how ever, previous studies have shown dioxane to be a hepatocarcinogen (9) and to induce carcinomas in the nasal cavity of rats re ceiving dioxane in drinking water (10). No previous reports of dioxane as a promot ing agent in skin carcinogenesis could be found. Because of the use of dioxane as a solvent in industry and in histology labora tories for tissue processing, the results ob tained are of potential significance in rela tion to possible carcinogenic hazards to humans. Acknowledgement This work was supported by contract No. N IH -71-2338 from the National Cancer In stitute of the National Institutes of Health. Dr. Bruce Christie performed the histopathological examinations. REFERENCES \ 1. Higginbotham, G. R., Huang, A., Firestone, D. Verrett, JMRoss, J., and Campbell, A. Chemical and toxicological evaluations of isolated and avnthetic chloro derivatives of dibenzo Nature 220: 702 (1968). 2. Bauer, H., Schulz, K .,and Spiegelberg, U. Bern, fliehe ' Vergtungen bei der Herstellung chlornhenol-verbindungen. Arch. Gerwetbepafc Gerwerbehyg. 18: 538 (1961). 3. Kimmig, J., and Schulz, K. Berufliche Akne (og. Chlorakne) durch chlorierte aromatische tykllsehe ther. Dermatologiea 115: 540 (1S57). 4. Courtney, H.( Gaylor, D., Hogan, M., Falk, R_, Bates, R., and Mitchell, I. Teratogenic evaluation of 2,4,5,-T. Science 168: 864 (1970). 5. Rowe, V. K., Toxicology of the chlorinated di* benzo-p-dioxins. Paper presented at American Chemical Society Meeting, Washington, D. C, Sept. 12-17, 1971. 6. Sparschu, G., Dunn, F., and Rowe, V., Study of teratogenicity of 2,3,7,8,-tetrachlorodibenzo-p. dioxin. Food Cosmet. Toxicol. 9: 405 (1971). 7. Courtney, K., and Moore, J., Teratology studle with 2,4,5-trichlorophenoxyacetic acid and 2,3,73Tetrachlorodibenzo-p-dioxin. Toxicol. Appl Pharmacol. 20: 396 (1971). 8. Boutwell, R. K. Some biological aspects of skin carcinogenesis. Progr. Exp. Tumor Res. 4: 207 (1964). 9. Argus, M., Arcos, J. and Hoch-Ligeti, C. Studies on the carcinogenic activity of protein-denatur ing agents: hepatocarcinogenicity of dioxane. J. N at Cancer In st 35: 949. 10. Hoch-Legeti, C., Argus, M., and Arcos, J. Induc tion of carcinomas in nasal cavity of rats by dioxane. B rit J. Cancer 24: 164 (1970). 170 Environmental Health Perspectives Chlorinated Dibenzodioxins and pentachlorophenol by R .L. Johnson,' P .J . Behring/ R .J. Kociba/ and B .A . S c h e tzf Introduction Pentachlorophenol is a registered anti microbial agent whose principal use is for the preservation of wood. Typical commer cial pentachlorophenol contains a variety of substances which are considered to be "in active" from the aspect of antimicrobial ef ficacy. Consequently, pentachlorophenol is snid as an antimicrobial agent with 95 fo a. rive ingredients and 5 % "inert" ingredi ents. Analysis of .acceptable commercial rtachlorophenol is shown in Table 1. he "caustic insolubles," sometimes re ferred to as the "nonphenolic or neutral im purities," include chlorinated dibenzo-p-dioxins and chlorinated dibenzofurans (1). Recently developed analytical technology has allowed quantitation of hexachiorodibenzo.-dioxins and octachlorodibenzo-p-dioxin in pentachlorophenol. Portrayed in Table 2 are concentration ranges for these two chlorodibenzo-p-dioxins in samples of currently available commercial grade pentachlorophe nol. Techniques capable of detecting 0.05 ppm showed no 2,3,7,8-tetrachlorodibenzo-p-dioxin in any sample of pentachlorophenol examined by us. The absence of this compound in pentachlorophenol is not surprising, because the appropriate precursors for its formation are not present. ` Designed Products Department, The Dow Chemi cal Company, Midland, Michigan 48640. tChemical Biology Research Laboratory, The Dow Chemical Company, Midland, Michigan 48640. Table 1. Commercial pentachlorophenol composition. Pentachlorophenol Tetrachlorophenol Trichlorophenol Higher chlorophenols Caustic insolubles (maximum) Content % 85-90 4-8 <0.1 2-6 1 Table 2. Concentration ranges of some chlorinated dioxins in commercial pentachlorophenol. Chlorinated dibenzo-p-dioxin 2,3,7,8-TetrachIorodibenzop-dioxin Hexachloradibenzo-p-dioxins Octachlorodibenzo-p-dioxin Concentration range, ppm None 9-27 57S-2510 Heptachlorodibenzo-p-dioxin and hexa-, hepta- and octachlorodibenzofurans have been qualitatively detected in commercial pentachlorophenol. However, the lack of ap propriate standards for these materials does not allow their quantitation. Severe toxicological responses have been attributed to certain chlorodibenzo-p-dioxins (2). For example, the LD30 of 2,3,7,3tetrachlorodibenzo-p-dioxin ranges from 0.6 Mff/kg in male guinea pigs to 115 fig/k g in rabbits of mixed sexes. A benzene solution of this agent containing as little as 0.04 fig/ml produces acne in the rabbit ear bio assay. Very high embryotoxicity and the production of edema in chicks are other properties of this material. The no-effect tember 1973 171 784493 oo dose levels for these latter activities are 0.03 and 0.1 fig/kg-day, respectively, ...Although pertinent for perspective, it is re-emphasized that this agent is not a contaminant of pentachlorophenol. Limited lethality data are available on hexachlorodibenzo-p-dioxins and octachlorodibenzo-p-dioxin. The former material killed 1 of 2 and 0 of 2 male rats given overdoses of 100 and 10 mg/kg, respectively. Doses of 1 g /k g and 4 g/kg of octachlorodibenzo-pdioxin failed to kill female rats and male mice, respectively. With regard to acnegenic activity, chloro form solutions containing 10 to 50 ^g/ml hexachlorodibenzo-p-dioxin are active while preparations of octaehlorodibenzo-p-dioxin are inactive. Hexachlorodibenzo-p-dioxin administered to pregnant rats at a dose of 100 /xg/kg-day has been found teratogenic, while doses of 1 or 10 /xg/kg-day produced only subcutane ous edema. A dose of 0.1 fig/kg/day was not associated with untoward effects in em bryos or fetuses. Studies to date have re vealed that octachlorodibenzo-p-dioxin is es sentially devoid of untoward activity in the embryo and fetus. Daily doses of 10 and 100 f i g / kg hexachlorodibenzo-p-dioxin produce a positive re sponse in the chick edema bioassay, while doses of 0.1 and 1.0 /*g/kg are negative. As with the other untoward effects previously referred to, octachlorodibenzo-p-dioxin ap pears to be devoid of this activity. Although the principal chlorodibenzo-p- dioxin contaminant of pentachloropenol, octachlorodibenzo-p-dioxin, in the amounts normally present does not appear to present a significant hazard, the presence of hexa- chlorodibenzo-p-dioxin as well as the other contaminants in pentachlorophenol previ ously mentioned give rise to concern. In at least one instance, the product literature warns that frequent skin contact may result in an acneforra dermatitis (3). However, documentation of such occurrences is un available. In addition, it has been stated that contamination of fat used in the diet of chickens with pentachlorophenol mav be sponsible for the "toxic fat" syndrome. ** In response to the concern about the tori cological significance of the nonphenolics b pentachlorophenol, we conducted studies determine whether their presence in th product may contribute to its tasrfcdogiSi properties. After finding that the toxicolo*. ical properties of the nonphenolic3 could] indeed, be detected by toxicological eval uations, acnegenic response, chick edema assay, and 90-day dietary feeding studies in rats, we set out to develop a product which was a "toxicological mimic" of pure pentv chlorophenol. Before presenting the results of our toxicity studies, it should be empha sized that pentachlorophenol is an economic poison, and although we believe we have de veloped the capability to produce a product in which the contaminants do not contribute to its potential hazard, it remains an eco nomic poison. Experimental The rabbit ear bioassay test was con ducted according to published procedures (2, 4 ). The chick edema bioassay test was conducted according to the procedure de scribed in the official methods of analysis of the Association of Official Agricultural Chemists (5). The feeding studies utilized the Spartan strain of Sprague Dawley rats which were maintained on diets formulated to supply the various dose levels for 90 days. Parameters monitored in these feeding stud ies included the following: body weights, food consumption, appearance and demeanor of the rats, routine hematologic and urinary, parameters, routine serum enzymes, ter minal organ weights and gross and histo pathologic examination of tissues. Results The concentration, of hexa- and octachlorodibenzo-p-dioxin in the commercial pentachlorophenol sample utilized in the toxicological studies reported herein a-'-'e given in Table 3. Table 4 is a compii-tion .of the toxicological data on this sample 01 172 Environmental Health Perspeedvc-s 784494 G EN P 011683 nercial pentachlorophenol. A positive re0IlSe was noted in both chick edema and ear bioassays. In the 90-day ra t feed`;rr <tudy, untoward effects were noted in a of the parameters monitored. Hemitoi,>(rical examination revealed a depres sion of erythrocytes, hemoglobin, and packed ceil volumes a t a dose level of 30 m? kg-day pentachlorophenol. Clinical chemistry alterations included an elevation of serum alkaline phosphatase at 30, 10, or :>mg/kg-day and a depression of serum al bumin at 30 or 10 mg/kg-day. The weights ,,t liver and kidneys were increased at 30, in. ov 3 mg/kg-day. Pathologic examination revealed minimal focal hepatocellular de generation and necrosis at 30 mg/kg-day. Thus, it is evident that commercial pentacklorophenol induced untoward effects in each of the three toxicological tests. Table 3. Concentrations of chlorinated dioxin "indi:itors" in commercial pentachlorophenol utilized in toxicological evaluations. J Chlorinated dioxin Octachlorodibenzo-p-dioxin Hexachlorodibenzo-p-dioxins Concentration, ppm 1980 19 A chemically pure pentachlorophenol havng no detectable concentrations of any chlorinated dioxins was subjected to the same toxicological tests. The toxicological data on this chemically pure pentachlorophenoi, summarized in Table 5, include neg ative responses in both the chick edema and rabbit ear bioassays. In the 90-day rat feeding study, the only changes noted were increased liver weights a t 30 or 10 mg/kgday and increased kidney weights at 30 m g/ kg-day. However, in contrast to commercial pentachlorophenol, gross and histopathological alterations did not accompany these in creases in organ weights. Thus, by utilizing the results of these three tests, it may be concluded that the presence of the contami nants in commercial pentachlorophenol may be detected by toxicological evaluation. An analysis of a sample of pentachloro- Table 4. Toxicological data on sample of commercial pentachlorophenol. Study Chick edema bioasaay Rabbit ear bioassay Rat feeding study Food consumption Body weight Hematology 30 mg/kg-day 10 mg/kg-day 3 mg/kg-day Urinalysis Clinical chemistry 30 mg/kg-day 10 mg/kg-day 3 mg/kg-day Liver weight 30 mgAg-day 10 mg/kg-day 3 mgAg-day Kidney weight 30 mgAg-day 10 mg/kg-day 3 mg/kg-day Pathology 30 mg/kg-day 10 mg/kg-day 3 mgAg-day * 4- denotes effect; --denotes no effect. Result * + + -- -- + -- -- -- + 4+ + 44- 4* 44- 4-- -- phenol representative of that which we are capable of producing is shown in Table 6. The toxicological data on this sample of pentachlorophenol are summarized in Table 7. Both the chick edema and rabbit ear bio assays gave negative responses. In the 90-day rat feeding study, the only unequivocal changes were increased liver weights a t 30 or 10 mg/kg-day and increased kidney weights at 30 mg/kg-day. There were no gross or histopathological alterations noted. To reiterate the toxicological findings on these three samples of pentachlorophenol, a comparison is provided in Table 8. Commercial pentachlorophenol gave posi tive responses in both the chick edema and rabbit ear bioassays; in contrast, the chem ically pure pentachlorophenol and the im proved pentachlorophenol both gave nega tive responses in these bioassays. In the rat feeding studies, commercial pen tachlorophenol was associated with hema- oeptember 1973 173 784495 G E N P 011684 J Table S. Toxicological data on sample of chemically pare pentachlorophenol. Study Results1 Chick edema bloassay Babbit ear biaassay Rat feeding study Food consumption Body weights Hematology Urinalysis Clinical chemistry Liver weight 30 mg/kg-day 10 mg/kg-day 3 mg/kg-day Kidney weight 30 mg/kg-day 10 mg/kg-day 3 mg/kg-day Pathology -- -- -- -- -- -- -- + + -- + -- -- -- * 4- denotes effect; -- denotes no effect. Table 6. Concentrations of chlorinated dioxin "indicators'* in improved pentachlorophenol utilized in toxicological evaluations. Chlorinated dioxin Octachlorodibenzo-p-dioxin Hexachlorodibenzo-p-dioxins Concentration, ppm 26 10.1 Table 7. Toxicological data on sample of improved pentachlorophenol. Study Results * Chick edema bioassay Babbit ear bioassay Bat feeding study Food consumption Body weights Hematology Urinalysis Clinical chemistry Liver weights 30 mg/kg-day 10 mg/kg-day 3 mg/kg-day 1 mgAg-day Kidney weights 30 mg/kg-day 10 mg/kg-day 3 mg/kg-day 1 mg/kg-day Pathology -- -- -- -- -- -- 44* -- -- 4-- -- -- -- + denotes effect; -- denotes no effect. tologic changes, clinical chemistry alter tions, liver damage, plus liver and kidn^ weight increases at dose levels of 30 in 7 3 m z/k fr-d ay. ' ,or The results of the feeding studies with improved pentachlorophenol gave results which closely paralleled the results obtained with the chemically pure pentachlorophenol, wherein changes were limited to increased liver and kidney weights at the higher W levels. Table 8. Comparative evaluation of toxicological data obtained on pentachlorophenol (PCP) samples.* Study ChemCommer- ically cial pure Improved PCP PCP PCP Chick edema bioassay 4* Babbit ear bioassay 4- Rat feeding study Hematologic depression 4- Clinical chemistry alterations 4- Liver damage (histo- pathology) + Liver weight increase 30 mg/kg-day 4- 10 mg/kg-day 4- 3 mg/kg-day 4- Kidney weight increase 30 mg/kg-day 4- 10 mg/kg-day 4- 3 mg/kg-day 4- -- ---- ---- ---- ---- 4- 44- 4---- 4- 4---- ---- *4- denotes effect; -- denotes no effect. Discussion and Summary The toxicological data have enabled us to conclude that a commercial pentachlorophe nol conforming to the "improved" penta chlorophenol would not elicit chick edema and chloracne. In addition, the histopathological effects resulting from the impurities in commercial pentachlorophenol would be eliminated. The toxicological findings discussed in this paper have been utilized as p art of an application for registration of new penta chlorophenol. This application has been ap proved by the Environmental Protection Agency in accordance with the requirements of the Federal Insecticide, Fungicide and 174 Environmental Health Perspectives 784496 Rodenticide Act. The composition specifica/ for this new commercial pentachloro- l are cited in Table 9. conclusion, it is `feasible to produce a pen tachlorophenol in commercial quantities which by comparative evaluations mimics pi.re pentachlorophenol in toxicological re sponses. table 9. Composition and specifications of improved pentachloropfaenoL Content Pentachlorophenol 8 8 -9 3 % Tetrachlorophenol 12- 7% Trichlorophenol < 0.1% Higher chlorophenols 0.1% ( ''.iorinated dioxins Octach lor odib enzo -p -dio xin 30 ppm H e x a c h lo ro d ib e n z o -p -d io x in s 1.0 ppm (m ax.) (m ax.) REFERENCES 1. Plimmer, J. R., Ruth, J. M., and Woolson, E. A. Mass spectrometric identification of the heptaand octachlorinated dibenzo-p-dioxins and dibenzofurans in technical pentachlorophenol. J. Agr. Food Chem. 21: 90 (1973). 2. Schwetz, 6. A., et al. Toxicology of chlorinated dibenzo-p-dioxins. Adv. in Chemistry, in press. 3. The Dow Chemical Company. Antimicrobial agents product literature, Section IV-7, Dow Chemical Co., Midland, Mich., 1969. 4. Adams, E. M., et aL The response of rabbit skin to compounds reported to have caused Acneform Dermatitis. Ind. Med. Ind. Hyg. Sec. 10: (2)1 (1941). 5. Official Methods of Analysis, 10th ed. Associa tion of Official Agriculture Chemists, Washing ton, D. C., 1965, Sections 26.087-26.091. ) G E N P 011686 ntember 1973 175 784497 v_ ^ Toxicity of 2,3,7,8-Tetrachlorodibenzo- p-dioxin [TCDD] in Aquatic Organisms* by Richard A . M ille r; Logan A . N o rris ,* and Clifford L . Hawkes* Herbicides are particularly important in modern forest management as foresters at tempt to make fullest use of a constantly shrinking production base. In forestry, (2,4,5,-T) 2,4,5-trichloro-phenoxyacetic acid is used to control undesirable woody species that compete with more desirable timber-producing conifers for light, space, '.oisture, and nutrients. Herbicide applica tions can result in markedly increased con- ~growth, but such applications must not Jilt in degradation of environmental qual ity (1 ). In evaluating hazards, scientists have focused on the herbicide, but biolog ically significant contaminants like 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) must also be considered (2). Herbicides may enter streams by several processes. Direct application or drift of spray materials to surface waters will occur only briefly during the application, but they may cause high concentrations of pollutant in streams. Herbicides could also move to streams in mass overland flow during peri ods of intense precipitation, but this seldom `Oregon Agricultural Experiment Station, Tech nical Paper No. 3624. Supported in part by Supple ment No. 72 to the Master Memorandum of Under standing between the U.S. Forest Service and Ore gon State University. tDepartment of Fisheries and Wildlife, Oregon State University, Corvallis, Oregon 97331. tU. S. Department of Agriculture, Forest Service, Pacific Northwest Forest and Range Experiment Station, Forestry Sciences Laboratory, Corvallis, ,O*^gon 97331. occurs on forest lands because the infiltra tion capacity of the forest floor is much greater than most rates of precipitation. Leaching through the soil profile is a slow process capable of only moving small amounts of herbicides short distances and offers little potential for serious stream pol lution (3). Studies in the Northwest indicate most contamination of forest streams by herbi cides results from drift or direct application of chemical to the water surface. Detectable quantities of 2,4-D and 2,4,5-T have not been found in western streams during fall and winter months after spray applications to nearby forest lands the previous spring (4 ). In spray operations involving the use of 2,4,5-T, we expect small amounts of TCDD will enter the water with the herbicide dur ing application. While some information is available on the toxicity of 2,4,5-T to aquatic organisms, little is known about the toxicity of TCDD. We conducted chronic toxicity tests to assess the hazard to aquatic organisms which may be exposed to TCDD in w ater or food after the use of 2,4,5-T in forestry (5). Some of the toxic characteristics of TCDD in food and water to several major classes of aquatic organisms are reported here. Materials and Methods The organisms tested were three species of fish: guppies (Poecilia, reticuiutas) , coho or silver salmon (Oncorkynchua hisutch), ptember 1973 177 784498 a o a\ oo Table 1. Summary of test procedures to determine the toxicity of TCDD in aquatic organisms. ExpL no.* 1' Organism Guppies (10-40 mm) 2 Snails (adult and juvenile) Container * 1-gal WMJ Water Water vol temperume, aturc, 1. C 3 20 1-gal WMJ 3 28-27 Exposure regime Level Ppt in water 0 100 1,000 10,000 0 200 ng/g Bw ` Dura tion, hr 120 1152 8 Worms (40 mm) 8-ln. culture 1 dish 4 Mosquito larvae 8-in. culture 1 dish 6 Salmon (7.25 g 5-gal WMJ 17 wet weight) Environmental Health Perspective 6 Salmon (1.83 g 6-gal WMJ 17 wet weight) 7 Salmon (3.51 g 6-gal WMJ 17 wet weight) 8 Salmon, (TCDD 5-gal WMJ recovery from water, 2.9 g wet weight) 0 Rainbow trout 6-gal 8 8 9 T T O rrN T iin 17 18 23-27 23-27 12-18 12-18 12-18 12-16 11-13 0 200 0 200 66 100 6G0 1,000 6.6 11.5 28.0 56.0 0.056 0.56 6.6 56.0 0 60 2.3* 1176 408 13.1 23.4 131.3 234.0 7.1 14.1 85.7 71.0 0.054 0.54 6.4 64.0 0.0063 ' 24 48 96 24 48 96 24 48 96 24 48 96 672 Total ob servation period, days 87 48 65 39 76 Feeding regime Experi Number mental of repli design * cations * TubiAx worms. ad lib. post exposure period CRD 3 n=20 Elodea, OMP ' ad lib. during exposure period CRD 4 n=7 * II to o OMP * 1/week, CRD j during exposure period 4 Yeast 2 /week, CRD during exposure period 4 n=20 OMP 3/week, SPF post exposure period 4 n=20 33 OMP 3/week, SPF 4 post exposure n=10 period 59 OMP *8/week, SPF 4 post exposure n=6 period -- None CRD 4 28 CDH 12/day CRD 6 (1.6 g oven dry weight)1 aquaria 2,if ^ t;.:i luring expo sure period Bo* *Expta. 1-8 are static water teats. Salmon experiments (5-8) were all combi nations of levels and durations of exposure listed. 4WMJ = wide-mouth jar. 1 *Nanograms TCDD per gram wet body weight. 4CRD = completely randomized; SPP = split plot factorial. toq it -- beginning number of organisms per treatment per replication. CO ' Data of Norris and Miller (10). 1OMP = Oregon moist pellet. *Concentration in food (dry weight). ' Weight TCDD tank per week, in nanogramB. 1Casein, dextrose, herring oil, fish ration (0). 4Dry weight derived from size-weight relationship for young coho salmon (/). I rainbow trout {Salmo gairdineri) ; and three aquatic invertebrates: a snail (Physa s p .) ; a worm (Paranais s p .); and mosquito larvae (Aedes aegypti). Guppies and mosquitoes were obtained from Oregon State University laboratory cultures, salmon and trout from State of Oregon fish hatcheries, and worms and snails from local streams. The TCDD (98.7% 2,3,7,8-tetrachlorodibenzo-p-dioxin) was obtained from the Dow Chemical Company. Treatment regimes are summarized in Table 1 for each experiment. I n . static w ater test 1 with fish, we expressed exposure levels as nanograms (10-9 g) TCDD per gram total body weight of organism as well as in parts TCDD per 10parts water. We do not imply the former are specific body bur dens of TCDD but rather the amount of chemical in the container relative to fish biomass at the beginning of the experiment. In some cases, the initial TCDD concentra tion in the water is also given for reference, but these are of limited value in interpreting the static water toxicity test results because the TCDD concentration did not remain con stant and cannot be related to organisms exposed in large bodies of water. We found dose-response relationships in fish were more easily expressed in terms of weights of toxicant and organism biomass. Static Water Toxicity Tests of the observation period. D uring the oh*, vation period, after TCDD exposure, th fresh w ater flow rate through the confcafo with salmon was 3 l./hr. W ater for gupJf was exchanged once each 14 days. Salmo were weighed at the beginning of priment and a t death. Guppy body leapt] was measured at death. Oven-dry weight of worms were made a t the end of the ob servation period. Table 2. Characteristics of test water. Concn level, ppm Constituent Calcium Silica Magnesium Sodium Potassium Bicarbonate Carbonate Sulfate Chloride Nitrate Iron Dissolved solids Static tests 12.0 6.9 7.8 7.7 1.45 79.0 0.0 5.1 4.3 6.1 0.03 132.0 Feeding teats 14.0 27.0 5.1 5.6 0.6 77.0 0.0 4.0 0.3 0.32 95 Hardness Specific conductance, tmko pH 64.0 164.0 6.9 56.0 132 7.7 TCDD Recovery in Static W ater Toxicity Tests For static water toxicity tests, animals were acclimatized for at least 48 h r before they were exposed to TCDD in well w ater in glass containers (Table 2). TCDD in acetone (maximum 0.3 ml acetone/1.) was added slowly and mixed by stirrin g and vig orous aeration of water. Control organ isms were exposed to an equivalent amount of acetone. At the end of the exposure pe riod, the animals were placed in fresh well water containing no dioxin for the duration Slightly modified from "A, tentative method for analysis of 2,3,7fS-tetrachlorodibenzo-p~dioxin in pond water." Personal communication, 3/12/71, Dow Chemical Company, Midland, Michigan. Details of analytical procedure available on request. To determine TCDD recovery from water containing salmon, TCDD, 0 or 900 ng in 3.6 ml acetone, was added to 17 1. of well w ater containing 10 coho salmon averaging 2.9 g wet weight each. All conditions were as in experiments 5-8 (Table 1 ). A total of 12 containers were spiked with TCDD. Each container was sampled only once. W ater samples (1.8 1.) were collected 24, 48, and 96 h r a fte r addition of chemical, and TCDD was determined by gas chromatog raphy*. Each sampling time was replicated four times. Samples of water from contain ers with salmon but no dioxin were also analyzed to verify adequacy of the cleanup procedure. ISO Environmental Health Perspectives 784501 G E N P 011690 / .gflic Oral Toxicity Testa V. ' .or determination-of chronic oral toxicity, exposed young- rainbow trout daily to various levels of TCDD in their food. TCDD, .0 .[`>fr saturation in acetone, was added to ![je ul base of a casein-dextrose-herring .it! 1M1 tdf slightly modified from Lee et al. '(*>). Acetone was removed from the oil by vacuum evaporation, leaving an average of (>3fr acetone in the dry fish food. The fish food contained 2.3 ppm, 2.3 ppb, or 2.3 ppt T C D D ; exposure levels are in Table 1. Two hundred young rainbow trout se lected for uniformity of size were randomly aligned among 20 aquaria which received tn it water at the rate of 9 l./hr. The 20 aquaria were assigned at random among one control and three treatments in five replica tions. Fish were acclimatized to the aquariaflowing water systems and TCDD-free food for 3 weeks before beginning the experi ment. Preweighed food given daily at 0900 hr contained the daily dose of TCDD. At l.'OO h r daily, food without TCDD was given ad libitum, and total daily consump- ii was recorded. To determine growth, were photographed once each week, and fish size index (the product of fish length and depth) obtained from a sideview photograph. In other experiments with coho salmon, sideview area was highly correlated with dry weight (7). We will establish a similar relationship for rainbow trout and express ;:.sh size in dry weight in later reports. Results and Discussion TCDD Recovery in Static W ater Toxicity Tests The TCDD level in w ater with young salmon declined significantly with time (Fig. 1). Regression analysis indicated re covery between 24 and 96 h r was linear ' with time (5): F = 63.1-13.5.X where Y is percentage recovery of TCDD, X is time in hours after addition of TCDD to w ater containing coho salmon; r 3 - 0.86. TCDD concentration decreased more rap idly between 0 and 24 hr than between 24 F ig u r e 1. Average recovery of TCDD from water spiked with 50 ppt TCDD and containing 10 young coho salmon (four replications). and 96 hr, which may suggest more than one mechanism of loss was operative. The rapid loss of TCDD during the first 24 hr may largely be the result of adsorption phe nomena which rapidly attain equilibrium. This hypothesis is supported by results from a similar test in which fish were not in cluded and TCDD recovery was 60.0% 4 hr after addition of the chemical. The fate of TCDD in the system is not known, but we suspect a combination of up take by fish, adsorption on glass and sus pended organic matter, and possibly loss due to aeration. Organisms in our static water toxicity tests were exposed to rapidly declining levels of TCDD because exposure solutions were not replenished. The exposure levels in Table 1 are the initial exposure levels, no adjustment being made for possi ble changes in TCDD concentration with time. Fish Symptoms after TCDD Exposure A difficulty in studying the toxicity of TCDD to fish iq that the response to the chemical is not immediate. In most static water test procedures, observations would have been terminated after 96 hr (9). In our tests, initial response to the chemical did not occur for 5 to 10 days after the beginning of the exposure period, and mortality often extended over the next 2 months. jteniber 1973 181 l mm 784502 Fish exposed to toxic levels of TCDD in w ater or food showed a declining interest in feeding. Salmon reduced feeding 8 days after TCDD exposure, while guppies responded in 5 days. Affected animals often spit food out shortly after taking it in. Growth of sal mon exposed to TCDD in water was mark edly inhibited (Fig. 2 ). I quently did not occur for 10 days after the beginning of the exposure period, regardless of exposure level (Fig. 3). In experiment 5 the effects of exposure to more than 23 TCDD/fish wet weight (23 ng/g) for 24 hr was irreversible, and most fish died within 60 days. The effects of level of exposure were quite marked while the effects of duration of exposure were less prominent. F ig u r e 2. TCDD--exposed (13.1 n g /g , 96 h r ) and con tro l yotong coho salmon 80 days a f t e r b eg in n in g of exposure period. F ig u r e 3. Survival of young coho salmon (experi ment 5) after exposure to TCDD in water. Values Skin discoloration and fin necrosis began are means for 24, 48, and 96 hr exposure (four replications). J to appear 15 and 30 days after initial ex posure of guppies and salmon, respectively (Fig. 2). Complete loss of the caudal fin oc In experiment 7, smaller salmon were used curred in both guppies and salmon. Areas as we attempted to identify the minimum showing skin discoloration often became the threshold response level for TCDD in water sit of attack for disease organisms. In sal (Table 1). The pattern of delayed mortality mon, large fungal growths completely en observed in experiment 5 was also promi circled some animals and inhibited swim nent in experiment 7 (Fig. 4). Exposure to ming. Erosion of the upper jaw was seen in TCDD levels of 54 ng /g for 24 h r or longer guppies surviving 1 to 2 months after ex was irreversible and killed all fish within 40 posure but not in salmon. Prior to death, days. Exposure to 5.4 n g /g resulted in 55% fish often remained close to the bottom of mortality during the 60-day observation the test containers and showed very little period. Levels of TCDD as low as 0.054 ng/g movement. There was no definite pattern caused 12% mortality in the 60-day exposure prior to death; some fish that appeared per period compared to 2 % mortality on controls. fectly healthy one day were dead the next It appears these lower levels may be ap day while other apparently diseased indi proaching the minimum threshold-response viduals remained alive for weeks. We de level. The duration of exposure appears less tected no differences in behavior between important than levels of exposure in deter treated and control invertebrate organisms. mining mean survival time (Fig. 5). For statistical analysis, data were ex Toxicity of TCDD in W ater to Young Coho pressed as days to death and subjected to Salmon multivariate analysis of variance. Mean sur Effect of level and duration of exposure-- vival time was significantly reduced with The deaths among exposed salmon fre- increasing TCDD exposure levels in experi- 182 Environmental Health Perspectives 784503 GENP m i Figure 4. Survival of young coho salmon (experi ment 7) after exposure to TCDD in water. Values are means for 24, 48, and 96-hr exposure (four replications). but the levels of exposure were 2 -2 0 times as great as.w ith coho salmon in experiment 5. Effect of size of fish on survival time--In both salmon and guppies, larger fish sur vived for longer periods than smaller fish after TCDD exposure. In some earlier work (10), mean survival time was plotted as a function of body length for TCDD exposed guppies ranging from 10 to 40 mm in length (Fig. 6). The regression equation was lin ear and highly significant (P <0.01). Body length accounted for 93% of the variation of the dependent variable. A similar effect was observed in salmon when data from ex periments 5, 6, and 7 were combined (Fig. 7). Time to 50% mortality for salmon ex posed to 10 ng for 96 hr was determined graphically for each experiment. Regression analysis showed that the effect of body weight on survival time was linear and sig nificant (P < 0.01): 7=13.8+7.7 X where 7 is time, in days, to 50% mortality, X is body wet weight, in grains; r^O .8 7 . Similar responses have been reported for other toxicants (11). The ability to tolerate environmental stresses increases with in creasing body mass and age, up to a point, in many organisms. Toxicant uptake, storage, and detoxification probably change with fish age, lipid levels, and gill surface area-body mass ratios (12). F ig u r e 5. In flu e n c e o f d u ra tio n of exposure to TCDD on m ean s u rv iv a l tim e o f y ou ng coho salmon (experim ent 7 ) (fo u r replications). merits 5, 6, and 7 (P < 0.01) Figs. 3*and 4). The duration of exposure effect was less marked, but was significant in experiments 5, 6 , and 7 (P < 0.05). The duration of exposure-concentration interaction was not significant in any experiment. We feel the duration of exposure effect in salmon may be more pronounced as the minimum thresholdresponse level is approached and as the dura tion of exposure is reduced. A duration of v posure effect was not observed in guppies, ^--ptember 1973 F ig ure 6. E ffect of body length on m ean survival tim e o f guppies exposed to 100, 1,000, and 10,000 ppt TCDD f o r 120 h r (JO) (th re e re p lic a tio n s ). 183 G E N P 011693 784504 Days to 5 0 % Death F ig u r e 7. Effect of body weight on time to 50% mortality in young coho salmon exposed to 10 ng TCDD per gram wet body weight. Values are means for 24, 48, and 96 hr exposure (four repli cations). Toxicity of TCDD in W ater to Invertebrate Aquatic Organisms In these tests, we exposed representatives from the class Insecta, a mosquito larvae; the class Oligochaeta, a worm; and the class Gastropoda, a pulmonate snail to TCDD in static water toxicity tests. Toxicity to snails--Adult pulmonate snail deposited numerous egg cases in container of well w ater which originally contained 0 or 0.2 ppb TCDD during a 36-day exposure period. Snail eggs completed development in the original exposure solution, and live juvenile snails and empty juvenile snail sheik were counted 48 days after the beginning of the experiment. There was no significant dif ference between the survival of treated and control adult snails (Fig. 9). Differences in the total snail hatch be tween treated and control organisms were observed in each replication, but variation among replications reduced the statistical sensitivity of these tests ( P = 0.056). Dif ferences in the percentage survival of young snails were not significant. TCDD appeared to have its m ajor impact on the reproductive success of snails rather than on survival of either adult or juvenile forms, in that the major effect was on total number of juvenile snail shells (Fig. 9). Toxicity to mosquito larvae--In tests with mosquitoes, we observed the maturation of larvae from the second instar through pupa tion during and after 17-day exposure in water which originally contained 0 or 0.2 ppb TCDD. There were no significant differ ences in total pupation or the rate of pupa tion among treated and control mosquitoes during the 30-day test period (Fig. 8). F ig ure 9. Total hatch and survival o f juvenile snails from egg masses deposited in 0 or 200 ppt TCDD in water during a 36-day adult snail exposure pe riod. Counts were made 48 days after the begin ning of the exposure period. F ig u r e 8. Pupation of mosquitoes exposed to 0 or 200 ppt TCDD for 17 days in water (four repli cations) . 184 Toxicity to aquatic worms--Adult Oligocnaete worms were exposed to 0 or 0.2 ppb TCDD in water for 55 days. Animals were counted at 30, 48, and 55 days a fte r the be ginning of the exposure period. A t 55 days total and mean dry weights were deter mined. Environmental Health Perspectives 784505 OENP OU694 _,______ TCDD, ppt Table 3. Toxicity of TCDD in water to Oligochaete worms. Initial Number of worms 30 days 48 days 55 days Biomass, mg dry wt Total at 55 days Mean individual at 55 days ~~ 0 80 233 409 414 374 200 80 195 310 266 193 0.90 0.73 Exposure of worms to TCDD resulted in a decrease in the total number of worms present at the end of the 55-day exposure period (P <0.05) (Table 3). Reductions in total worm biomass between treated and contr i organisms occurred in each replication, hi;: variation among replications reduced the statistical sensitivity of this test (P=0.057). TCDD exerted its principal ef fect on reproduction rather than growth of individual worms. Toxicity of TCDD in Food to Young Rain bow Trout Young rainbow trout (10/aquaria) were exposed to 0, 6.3 pgf 6.3 ng, or 6.3 fig TCDD tank per week in food (Table 1). The J D-containing ration was offered each morning, and TCDD-free food was offered each afternoon. Survival was tallied daily, and growth was measured weekly. There were no deaths among fish ex posed to TCDD in the first 28 days of the xperiment, but deaths began to occur in xsh exposed to 6.3 fig TCDD per tank per week after 33 days of exposure. The appetite of fish receiving this dose began to decline after 10 days, and by 14 days fin necrosis was apparent. No loss of appetite or fin ero sion occurred in fish exposed to lower levels of TCDD. We observed no differences in the growth if fish receiving 0, 6.3 pg, or 6.3 ng TCDD per tank per week during the first 28 days of the experiment (Fig. 10). The growth of fish receiving 6.3 fig TCDD per week de parted markedly from the others after 7 days, and they lost weight for the remaining 21 days of the exposure period. The data, expressed as the fish size index (product of fish length and body depth) were subjected to lysis of variance to test for differences F ig u r e 10. Average fish size index (length x body depth) of rainbow trout receiving TCDD in food daily (five replications). among fish size after 28 days of exposure. There were no differences in the size of fish receiving 0, 6.3 pg, or 6.3 ng TCDD per tank per week. The difference between this group of fish and fish receiving 6.3 ig TCDD per tank per week was highly significant (P<0.01). Our data indicate TCDD in food can cause growth reduction and mortality in fish. The oral threshold response level for exposure periods up to 28 days is greater than 6.3 ng TCDD per tank per week. Addtional experimentation is necessary to define more precisely the oral threshold response levels and to determine the impact of long term chronic exposure in food to fish. TCDD Residues in the Forest TCDD residues have not been reported in either terrestrial or aquatic components of the forest, but we are not aware of any serious sampling efforts. 2,4,5-T and other herbicides have been reported in Northwest forest streams (4 ). Existing stocks of 2,4,5T may contain up to 0.5 ppm TCDD, but new formulations must contain less than 0.1 ppm TCDD. We calculated the levels of TC- M em b er 1973 185 G EN P 011695 784506 DD which might be in forest streams after the aerial application of 2,4,5-T assuming the principal route of entry was drift or di rect application of spray materials to the stream surface. Levels of 2,4,5-T are not ex pected to exceed 0.1 ppm if applications are carefully controlled (Table 4). A more com plete determination of threshold response levels will be required, however, before an adequate assessment of TCDD hazard to stream organisms can be made. Conclusions TCDD in water or food is toxic to fish. The effects of exposure for 24-96 hr of young salmon to TCDD in water at levels greater than 23 n g /g is irreversible, and death results in 10-80 days. Duration of exposure is less important than level of ex posure except as threshold response levels are approached. The critical exposure period may be somewhat less than 24 hrs in static w ater toxicity tests in which TCDD concen tration may change markedly with time. Small fish are more sensitive than large fish on an equivalent exposure level basis. TCDD in food at 2.3 ppm markedly reduced growth of young rainbow trout (10/aquaria) ex posed to 6.3 /ig TCDD per tank per week for 4 weeks. TCDD at 0.2 ppb had no effect on pupation of mosquito larvae, but reduced the reproductive success of a pulmonate snail and an Oligochaete worm. Table 4. TCDD in streamwater after aerial application of 2^,5-T to forest land. streamwater, ppm 1.0 0.1 0.05 0.01 0.005 Anticipated TCDD in streamwater, ppt Level 1 * Level 2* 0.5 0.05 0.025 0.005 0.0025 0.1 0.01 0.005 0.001 0.0006 *Level 1: 2,4,5-T contains 0.5 ppm' TCDD. "Level 2: 2,4,5-T contains 0.1 ppm TCDD. Our research has established some import ant toxicity characteristics of TCDD in fish, but considerable work remains to be done. Establishment of minimum threshold r t sponse levels during long- and short-term exposure are important. The impact of previous and current TCDD exposure on long-term growth and reproduction of needs attention. Information on its move ment, persistence, and fate of TCDD in aquatic systems will be required to adequate ly assess the impact of TCDD in streams. Serious attempts to determine TCDD resi dues in various parts of the natural aquatic ecosystem are badly needed. The most sen sitive analytical techniques and positive means of residue identification will be neces sary. REFERENCES 1. Norris, L. A. Chemical brush control--assessing the hazard. J. For. 69: 715 (1971). 2. Courtney, K. D., et al. Teratogenic evaluation of 2,4,5-T. Science 168: 864 (1970). 3. Norris, L. A., and Moore, D. J. The entry and fate of forest chemicals in streams. In: Forest Land Uses and Stream Environment J. T. Krygler and J. D. Hall, Eds., Oregon State University, Corvallis, 1570, p. 138. 4. Norris, L. A. Chemical brush control and herbi cide residues in the forest environment In: Herbicides and Vegetation Management in For ests, Ranges, and Noncrop Lands. M. Newton, Ed., Oregon State University, Corvallis, 1967, p. 103. 5. Sprague, J. B. Measurement of pollutant toxicity to fish. I. Bioassay methods for acute toxicity. Water Research 3: 793 (1969). 6. Lee, D. J. et ai. Effect of 3 fatty acids on the growth rate of rainbow trout Salma gairdnerL J. Nutr. 92: 93 (1967). 7. Hawkes, C. L. Precise growth measurements of live, unanesthetized fish by photography. Pro ceedings of the 23rd Annual Northwest Fish Cultural Conference, 23: 93 (1973). 8. Draper, N. R., and Smith, H. Applied Regression Analysis. Wiley, New York, 1966, 407 pp. 9. American Public Health Association. Standard methods for the examination of water and wastewater. 13th ed. American Public Health As sociation, Inc., Washington, D.C., 1971, 874 pp- 10. Norris, L. A., and Miller, R. A. The toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in guppies. Bull. Environ. Contain. ToxicoL in press. 11. Post, G., and Schraeder, T. R. The toxicity or fonr insecticides to four salmanid species. Bull. Environ. Contain. Toxic. 6: 144 (1971). 12. Cope, Oliver B. Interactions between pesticides and wildlife. Ann. Rev. Entomol. 16: 325 (1971). 186 Environmental Health Perspectives 784507 G E N P 011696 5 Acute and Chronic Oral Toxicity of Chlorinated Dibenzofurans to Salmonid Fishes by V . Z it k o ,' D .J. W ildish,' C. Hutzinger/ and P.M .K. Choi* A median mortality of 120 30 days oc curred among juvenile Atlantic `salmon Salmo salar), fed dry fish food containing ..7, 5.7, 2.8, and 9.1 *ig/g wet weight of 2,8' id-, tetra-, and octachlorodibenzofuran, ctively (2 ). Only octachlorodibenzofur an. was detected in the- dead fish, and the level was 0.03 /*g/g in the muscle and 0.21 <g/g in the gut (both values on wet weight oasis). The fate of the lower chlorinated 'libenzofurans was not known and addition al experiments, described in this paper, were aim ed out with immature brook trout iSalvelinus fontinalia), fed relatively high levels of 2,8-dichlorodibenzofuran. Experimental The preparation of 2,8-dichIorodibenzofuran was carried out as described (2). Three crystallizations were necessary to remove tri- and tetrachlorodibenzofuran from the product. ` Environment Canada, Fisheries Research Board, Biological Station, St. Andrews, New Brunswick, Canada. tAtlantic Regional Laboratory, National Research Council of Canada, Halifax, Nova Scotia, Canada. ritTi-iiaster University, Hamilton, Ontario, Canada. I*__ nber 1973 Ten immature brook trout, tagged with colored tags were kept in running water (2 l./min) in a 200-1. Fiberglas tank and fed twice a day commercial fish food (Trout Chow, Ralston Purina Company) in the form of floatable pellets and in clear No. 3 gelatin capsules (E. Lilly & Company). Both pellets and capsules contained approxi mately 100 mg of food. Three fish, each fed on the average 700-800 mg of food per day, also received gelatin capsules containing cry stalline 2,8-dichlorodibenzofuran. Once crys tallized product, still containing a trichlorodibenzofuran and a small amount of a tetra chlorodibenzofuran was fed to one fish. By accident, one fish from the former group re ceived one capsule containing the mixture of chlorinated dibenzofurans. The feeding schedule of 2,3-dichloridibenzofuran and the initial and final weights of the fish are pre sented in Figure 1. The fish were sacrificed as indicated and kept frozen ( --14C) un til analysis. Fish R was isolated after the final capsule (10.3 mg) in a small tank containing 4 1. of water. W ater was changed after 24, 48, 72, and 144 h r and processed as described (5) to detect possible hydroxylated metabolities of 2,8-dichlorodibenzofu ran. 187 G E N P 011697 784508 oe e- -- R 178 --' -- evi t--ir t - ti -< 4 166 Y 178 o h-- I 185 W 133 ^v C(*M: CM *T --0 11l 1I I1*A1A? CM d All* BY 192 tn r n m iO 200 i 133 1 1111 0 10 20 30 40 5 Days Results and Discussion The levels of PCBs and lipid in the tis sues, liver weight, and sex of the fish are summarized in Table 1. PCBs were of the Aroclor 1254 variety and had to be separ ated before the quantitation of chlorinated dibenzofurans. As an example, the GLC patterns of the Y liver extract are presented in Figure 2. The tracing A was obtained on the hexane eluate from the silicia col umn. A fter the removal of PCB3 in the first fraction from the alumina column {2% methylene chloride in hexane, tracing B), FIGURE l. Feeding schedule. Initial and final weights (g) of fish are on the left and right margin,' re spectively. Administered amounts of 2,8-dichlorodibenzofuran (mg) are indicated by vertical lines. The asterisks (*) indicate administration of oncecrystallized 2,8-dichlorodibenzofuran containing tri- and tetrachlorodibenzofuran. Samples of the white lateral muscle, taken between the dorsal fin and the lateral line, and whole livers were extracted with hexane and the extracts were cleaned up by column chromatography on alumina and silica (4 ). Polychlorinated biphenyls (PCBs) were sep arated from chlorinated dibenzofurans by column chromatography on alumina (5) Gas chromatographic analysis was per formed as described {4). A DuPont CEC 2111QB mass spectrometer was used to record mass spectra. Lipid concentrations refer to hexane-extractable lipid. F ig u r e 2. GLC p a tte rn s o f Y liv e r e x tr a c t: ( A ) hexane eluate fro m the silica column, containing a m ix tu r e o f PCBs and chlo rin ated dibenzofurans and separated b y chrom atography on alum ina tc y ie ld the PCB fra c tio n (B ) and the dibenzofuran fra c tio n (C ); (D) GLC p a tte rn of once-crystal lized 2,8-dichlorodibenzofuran (first peak from the le ft ) , containing t r i- and tetrachlorodibenzo fu ra n (second and th ird peaks, respectively). Table 1. Chemical and biological characteristics of fish. PCBs, ftg/g wet weight ` Fish Muscle Liver R 0.05 0.25 Y 0.05 0.34 W 0.15 0.35 BY 0.08 0.12 T 0.06 0.33 Liver weight, % of body weight 0.75 3.12 1.86 1.77 1.96 Lipid, % Muscle Liver 0.32 3.05 0.40 14.45 2.98 8.01 0.42 2.71 0.67 8.60 *Aroclor 1254. Sex (immature) Male Male Male ' Female Female 188 Environmental Health Perspectives 784509 GENP 011 6 9 8 Table 2. Relative accumulation of 2,3-di-, tri-, and tetrachlorodibenzofurans. J Sample -2,3-Dichlorodibenzofuran Peak height ratios Trichlorodibenzofuran Tetrachlorodibenzofuran d preparation muscle 'iver ; muscle ; liver liver 1 1 1 1 1 1 0.92 0.18 0.40 0 2.06 1.87 0.87 0.28 1.04 0.36 1.11 1.16 chlorinated dibenzofurans appeared in the second fraction (2 0 % methylene chloride in hexane, tracing C). The tracing D of the nnce-crystallized 2,8-dichlorodibenzofuran is :resented for comparison. Table 3. A dm inistered, and tissue levels of 2,8-dichlorodibenzofuran. 2,8-Dichlorodibenzofuran, pg/g wet weight Administered Found Fish Muscle Liver R 361 0.052 ( 0.307 Y 107 0.048 ' 0.150 W 254 0.340 0.400 BY 260 ` 0.230 1.04 -- -- 0.146 ce-crystallized preparation It can be seen th at the relative concen tration of the tri- and, particularly, of the tetrachlorodibenzofuran is higher in the iso lated mixture than in the administered pre paration. The relatively higher accumula tion of these compounds was observed in both fish Y and BY, and also in the liver of fish T, which was not actually fed chlorin ated dibenzofurans, but was present in the tank during the feeding of all fish (Table 2). The administered doses of 2,8-dichIorodibenzofuran and tire detected tissue residues are presented in Table 3. In all cases the tissue residues are very low, 0.01-0.13 % of the administered dose in the muscle and 0.08-0.4% in the liver. The feeding history affects the residual levels. Thus fish W, sacri ficed 1 day after dosing contains higher levels of 2,8-dichlorodibenzofuran than fish R , which received a sim ila r last dose a n d a higher total dose, but was sacrificed only 15 days after the last dose. Mass spectra of the crude mixtures of or ganic compounds isolated from the water, containing excreta of fish R, indicate the presence of a dichlorodibenzofuran (M*236) and `of a hydroxydichlorodibenzofuran (M* 252). According to the feeding data, 2,8-dichlo rodibenzofuran has a low acute toxicity to immature brook trout, since no mortality re sulted even after administering a single dose at a level as high as 122 mg/kg. The low acute toxicity and residual levels of 2,8dichlorodibenzofuran may be due to the poor absorption of the compound in the gut and to its excretion in the form of a conjugated hydroxy derivative. Acknowledgement We thank Mr. D. J. Embree for recording the mass spectra, Mrs. Madelyn M. Irwin for typing the manuscript, and Messrs. P.W.G. McMullon and F. B. Cunningham for drawing the figures. REFERENCES 1. Zitko, V., and Choi, F, M. K. Oral toxicity of chlorinated dibenzofurans to juvenile Atlantic salmon, Bull Environ. Contam. Toxicol. 10: in press (1973). 2. Gilman, H. G., et al., Dibenzofuran. III. Nuclear substitutions, J. Amer. Chem. Soc., 56: 2473 (1934). 3. Hutzinger, O., et al., Polychlorinated biphenyls: Metabolic behavior of pure isomers in pigeons, rats, and brook trout. Science 178: 312 (1972). 4. Zitko, V., Problems in the determination of poly chlorinated biphenyls. Intern. J. Environ. Anal. Chem., 1: 221 (1972). 5. Zitko, V., Absence of chlorinated'dibenzodioxins and dibenzofurans from aquatic animals. Bull. Environ. Contam. Toxicol. 7: 106 (1972). ,'.mber 1973 189 I G E N P 011699 784510 The Search for Chlorinated Dibenzofurans and Chlorinated Dibenzodioxins in Wildlife Populations Showing Elevated Levels of Embryonic Death by Gerald W. Bowes,* Bernd R. S im o n eit/ A.L. Burlingame,7 Brock W. de Lappe,* and Robert W. Risebrough* Introduction arious kinds of reproductive abnormali sms have been documented among wildlife jecies in North America. All are associ ated with areas of high agricultural or in dustrial activity and have been shown, or are assumed to be, pollutant-induced. Se lected wildlife populations may therefore serve as the best indicators of the' presence ; ' the environment of compounds which ave deleterious effects on organisms at very low concentrations, such as several of the chlorinated dibenzo-p-dioxins and chlorin ated dibenzofurans. The best documented of the abnormali ties observed in wildlife is the shell thinning of eggs of raptorial and fish-eating species of birds. Contemporary samples show a sig nificant increase in the variance of such arameters as shell weight, shell thickness and an index of shell thickness when com- `Toxic Chemicals Section, Canadian Wildlife Service, Ottawa, Ontario K1A OH3, Canada. tSpace Sciences Laboratory, University of Cali fornia at Berkeley, Berkeley, California 94720. tBodega Marine Laboratory, University of Cali fornia, Bodega Bay, California 94923. ptember 1973 pared with samples obtained prior to 1940 (1). Most if not all of the increased variance can be explained, however, by a function of the concentration of the DDT compound 1,1dichloro-2,2-bis(p-chlorophenyl)-e t h y 1e n e (p/p'-DDE) (2). Embryonic mortality is reducing the re productive success of several bird species, including the merlin, Falco columbarius, and prairie falcon, Falco mezicanus, of western Canada (R. Fyfe, personal communication), the osprey, Pandion haliaetus, of the north eastern United States (P. Spitzer, personal communication), the herring gull, Larus argentatus, of the Great Lakes region (3), and the common terns, Sterna himndo, of Lake Ontario (M. Gilbertson, personal communi cation). In 1972 virtually no young hatched in the colonies of the latter two species on Lake Ontario fM. Gilbertson, personal com munication). The embryotoxicity of com mercial PCB has been attributed to chlorin ated dibenzofuran contaminants in the pre parations U -d). These compounds, associ ated with or derived from PCB, are there fore a possible cause of the embryonic death observed in wild bird populations. Other pollutants which could contribute to the known to be toxic to either rats ( i n mortality, including1 the organochlorine in chicks (5). examiningsecticides and their derivatives, are invar In the current study we are 1 iably present in these samples; some of the the hypothesis that several kinds of abnor mortality could also be an indirect result malities, as yet unexplained, may be caused of the DDE-induced shell thinning. by chlorinated dioxins or benzofurans that A low incidence of birth defects has have accumulated in the aquatic food webs been documented in a colony of common of polluted areas; the initial results are re terms in Long Island Sound (7). A propor ported in the present paper. tionately larger number were found in 1972 in a common tern colony in Lake Ontario Materials and Methods (M. Gilbertson, personal communication). The same kinds of birth defects--primarily those of the beak, eyes and feet--have been produced experimentally by the chlorinated dibenzodioxins (8). The sea lions, Zalophus califomiantis, inhabitating the coastal wafers of southern California, have shown an incidence of pre mature births in recent years which is judged to be substantially higher than the expected normal. A group of females giving birth to premature pups in 1970 had higher levels of both the DDT and PCB compounds than did a group of full term parturient fe males (9). There is no evidence of fetal death in this -population; rather, the ab normality consists of a premature onset of The initial material selected for examin ation consisted of a pooled collection of 45 herring gull eggs and pooled samples of sea lion blubber and liver. The herring gull eggs were obtained in the spring of 1972 from Scotch Bonnet island in Lake Ontario. Very few of the eggs laid in this colony have hatched in recent years (M. Gilbertson, per sonal communication). The sea lion samples had been obtained from nine females that had just given birth to premature pups on San Miguel Island, California, in the spring of 1972, during a study undertaken by the U.S. National Marine Fisheries Service, Se attle, and the Naval Underseas Center, San Diego. partus. It is therefore unlike the fetal mor tality and resorption observed among do mestic mink; Mustela vison, that are fed either coho salmon, Oncorhynchtts kisutch, obtained from Lake Michigan, or a diet sup plemented with the commercial PCB pre paration Arocior 1254 (1 0 ). The pattern of fetal mortality and resorption observed among the mink is very similar, however, to that found in female rats on a diet con taining 2,3,7,8-tetrachlorodibenzo-p-dioxin ( I I ) . PCB levels in the sea lion population of southern California are among the highest found in North American mammals (5): they therefore provide favorable material in which to look for the associated chlorin ated dibenzofurans. Chlorinated dibenzodioxins and dibenzo furans were looked for in tissues of marine Preparation of Standards Clophen A-60, a German PCB prepara tion containing 60Jo chlorine, was selected as a source of chlorinated dibenzofurans on the basis of the earlier work by Vos et al. (5). A subsample of the Clophen A-60 in which the chlorinated dibenzofurans had been detected (Lot No. 912434) was pro vided to us by J. G. Vos. The first phase of the isolation procedure followed is similar to that outlined by Vos et al. (5). A 550-mg portion of the Clophen was dissolved in 400 ml of hexane and added to a Florisil column in 50-mi portions. The Florisil was activat ed and freed of contaminants by methods previously described (13). Each portion was allowed to sink to the surface at an elution rate of 5-10 ml per minute. The Florisil (180 g) was packed in a glass column with fish and birds from the Bay of Fundy and a Teflon stopcock and an outside diameter were not detected (12). The detection lim . of 44 mm. As the last portion of the FCL its, however, were substantially above those solution eluted to the Florisil surface, a 4GG- 192 Environmental Health Perspectives 78451? G EN P 011701 v } portion of hexane was added to the col umn- The first 50-ml portion was used to rinse the walls and was allowed to elute to the surface as before. The hexane was fol lowed successively by 400-ml portions each 0I- o% diethyl ether-hexane, 25% diethyl ether-hexane, and acetone. A cleaner pre paration of the chlorinated dibenzofurans isolated from Clophen has been found in an acetone fraction after elution with 25% diethyl ether-hexane (J.G. Vos, personal communication). The diethyl ether and acetone fractions uvre evaporated just to dryness under a stream of dry nitrogen, and the residue taken up in hexane. This step was repeated twice. Each of the three fractions was then placed on a micro activated alumina column (Fisher alumina, Cat. No. A-540), to fur ther separate the suspected contaminants from FCB (14), In order to remove all of the PCB, the volume of the first eluting sol vent system, 1 % methylene chloride-hex;.::e, was increased to 15 ml, and that of second, 2 0 % methylene chloride-hexane, 10 mi. Chlorinated dibenzo-dioxins ap, ' j r in the latter fraction, and it is prob able that the chlorinated dibenzofurans do as well (M.L. Porter and J.A. Burke, per sonal communication). This fraction was then evaporated just *o dryness, and the residue taken up in hex ane. This step was repeated twice. Gas chro matographic analysis and fraction collec tion for subsequent mass spectrometric iden tification of components is described below. Preparation of Environmental Samples The herring gull eggs were individually weighed, the length and breadth measured for subsequent shell thickness index deter mination, and the contents pooled and freeze'iried for 5 days. Total wet weight was 3121 S', and total freeze-dried weight 567 g. The dried eggs were ground with sodium sul fate (2 :1) in a heavy glass m ortar, and an appropriate amount was placed in a glass thimble for extraction in a Soxhiet appar atus (extra-large size, Corning Glass Works 415070) fitted to a 2000-mi flask con- taining 1400 ml hexane-acetone azeotrope. Each portion' was extracted for at least 8 hr. This procedure was repeated until the entire sample was extracted. The original solvent was left in the flask and was used for the extraction of all portions. The total extraction period was 119 hr. The solvent was then concentrated with the use of a rotary evaporator, transferred to a 1000-ml graduated cylinder, and the volume adjusted to 1000 ml with hexane. Lipid determinations were then made. Three 3-mi aliquots were transferred from the cyl inder to each of three preweighed aluminum pans (3.0 ml volume in a 5.0-ml disposable pipet), allowed to air-dry, then placed in a 100C oven fo r.30 min. A fter cooling, the dishes containing the lipid were weighed. The total lipid recovered was 250 g. The clean-up procedure used on the re maining sample extract was adapted from that detailed in Stanley and LeFavoure (15). The extract was taken from the graduatedcylinder in four 200-ml portions with a 50ml volumetric pipet. Each of the four 200-ml portions was passed over 350 g of Celite-- sulfuric acid in a Buchner funnel, followed by three 500-ml elutions with hexane. The cleaned extract was concentrated for subse quent fractioning into various chlorinated hydrocarbon components (see below). The maximum amount of lipid placed on each Davidow column of unit 15 gCelite-9 ml sulfuric acid-9 ml fuming sulfuric acid was kept to 4 g. Sea lion liver samples from nine females were weighed, pooled, and freeze-dried for 5 days. The total wet weight was 425.1 g, and the total freeze-dried weight was 121.4 g. The dried samples were ground with sodi um sulfate (1 :8 ), and a portion placed in a Soxhiet apparatus for extraction. Condi tions were similar to those outlined above for herring gull eggs. The total extraction time for all portions was 72 hours. Solvent was concentrated as above, and aliquots taken for lipid determination. The lipid re covery was 16.6 g. The remainder was di vided into two portions. Each was passed over 60 g of Celite-sulfuric acid, followed Ctfptember 1973 193 G E N P 011702 784513 by three 150 ml elutions with hexane. The ture, 225 C ; detector temperature, 285* r cleaned-up extract was concentrated for fur column flow rate 60 cc/min; purge flow J&} ther fractionation and analysis. 50 cc/m in; carrier and purge gas, w tro J !' Sea lion blubber samples were from eight Several of the extracts were further ft / of the nine females sampled for liver. The tionated by collection of individual total wet weight was 493 g. Small slices components with the use of a 10:1 s p in were ground with sodium sulfate (1:6 ), and (Western Scientific, Danville, California) the entire mixture was transferred to a Sox- Glass capillary tubes, bent to a V-shapa* hlet apparatus (giant size, Ace Glass, Inc., were immersed in a beaker of liquid uitro- Vineland, N.J., Cat No. 6810, size H; thim gen during collection of the peak compon ble, Cat. No. 6812, size H ). The sample was ents through a septum. extracted for 4 days with 6300 ml hexaneacetone azeotrope placed in the 12000-ml Mass Spectrometry flask. The extract was concentrated, aliqu The samples were dissolved in 5 pi of meth ots taken for lipid determination, and one ylene chloride and transferred onto the di quarter of the total was cleaned up and pre rect inlet probe of the mass spectrometer, pared for subsequent fractionation and ana where the solvent was allowed to evaporate. lysis. The total lipid recovery was 420 g. The probe was rapidly introduced into the Fractionation of Biological Extracts Each of the three extracts (herring gull eggs, sea' lion liver, and sea lion blubber) was divided in two, and each portion was placed on a large Florisil column (180 g) and fractionated by elution with the four solvent systems described for the Clophen, except that the volume of hexane was in creased to 800 ml. Subsequent fractiona tion on the alumina column did not remove enough PCB interference in the 20 % methyl ene chloride eluate. Chlorinated residues from this eluate (dissolved in hexane) were then placed on a smaller activated Florisil column (8 g, 10 mm ID), and eluted with smaller volumes of the same solvent se quence: 150 ml hexane, 200 ml 5% diethyl ether-hexane, 200 ml 25% diethyl etherhexane, 200 ml acetone. These fractions, ion source and multiple scans were recorded. Both low and high resolution mass spectrometric analyses were carried out on a GECAEI MS-902 mass spectrometer. For high resolution the instrument was used on line to an XDS Sigma 7 computer as described by Burlingame (16, 17) and Burlingame et al. (15). The ion source operating condi tions were: resolution, 10,0 0 0 ; ionizing cur rent, 500 jtiA; ionizing voltage, 50 eV; tem perature, 200-220C. The scan rate was 16 sec per decade with a dock rate of 24 kHz. Multiple scans were taken during each anal ysis and then sum-averaged together during data reduction. For low resolution the in strum ent was scanned either on line to an XDS Sigma 2 computer (19) or manually with an ultraviolet-sensitive strip-chart re corder. evaporated to dryness and taken up in hex ane, were then placed on alumina columns and chlorinated hydrocarbon components separated as before. Gas Chromatography Results and Discussion The pooled sample of herring gull eggs was found by GC-electron capture analysis to contain 35 ppm DDE and 300 ppm FCB on a wet weight basis or 440 ppm and 3700 Extracts were analyzed with the use of an electron-capture (EC) flaNi Tracor MT220 gas chromatograph. The stationary phase was 3% OV-1 on 100-120 mesh Supelcoport, placed in a 6 ft x 4 mm. I D Pyrex ppm, respectively, on a lipid basis. The PCB content was therefore approximately 0.9 * The sea lion liver sample contained 12 PPm DDE and 3 ppm PCB on a wet weight basis or 300 ppm and 80 ppm, respectively, on a column. Other operating parameters includ - lipid basis. The'PCB content of the ed : oven temperature, 200 C ; inlet tempera was therefore approximately 1 mg. 194 Environmental Health Pei'spu^1-vc- - 784514 GENP 011703 pCB concentrations in the blubber were r^.i jmd 62 ppm on a wet weight basis, 630 nd ^ PPm on a ^ P ^ basis. The PCB cont 0f the entire sample was therefore 30 ^ ,f which 7.5 mg was used for the ex- -rac'.'.un process. The electron capture chromatogram of the \\rr methylene chloride-hexane extract of f'lophen A-60 is shown in Figure 1. The components of peaks A -G were collected for individual analysis by low resolution mass spectrometry. Retention times of the peaks relative to dieldrin, and the identity of the compounds identified are listed in Table 1. All of the peaks were identified as chlorin ated dibenzofurans; no other significant com ponents were detected. Compounds having the same retention time relative to dieldrin as peaks A and C have previously been iden tified by Vos et al. (5) as tetrachloro- and pentachlorodibenzofuran, respectively. In the present study, peak B, consisting of two components on the EC gas chromatogram, was identified as tetrachlorodibenzofuran, Table 1. Relative retention times and identification of chromatograph peaks shown in Figure 1. Peak A B C D E F G Retention time, relative to dieldrin 1.49 1.60 2J2S 2.64 2.95 4.73 5.24 Composition CuHm,crci4 CnHiOmuCl m CuHiO"CL * Mass 3m04 33*8 m 37m2 Tetrac*hlorodibenzmofuran Pentac0hlorodibenz0ofuran 0m Hexachlorodibenzofuran m0 September 1973 195 784515 probably isomers of peak A. In addition, peaks D and E were identified as pntachlo- rodibenzofurans and F and G as hexachlorodibenzofurans. Representative mass spectra of a tetra- (B), penta- (E ), and hexa hi rodibenzofuran (F) are shown in Fioni 9 Tetrachlorodibenzofuran fragments w et&^ follows (cf. Fig. 2a): the molecular ior^ J 150 200 250 300 350 b. 130 2 0 0 250 3 0 0 3 3 0 <0C Figure 2. Low-resolution mass spectra of compounds isolated from Clophen A-6G: (a) tetrachlorodibcu^ofuran; (6) pentachlorodibenzofuran; (c) hexacblorodibenzofuran. 196 Environmental Health Perepecilves 784516 G E N p oi 17n< fV \ ;u n d -at m /e 304r-312, loss of Cl yields the peaks at m /e 269-275, subsequent elim ination of CO results in m /e 241-245 and double Cl loss yields the minor peaks at m /e 03 '-238, which on loss of CO and CHO yield tlu* group at m /e 205-210. The doubly charged species of the molecular ion are found at m /e 152-156, and the strong loss of Cl followed by CO is confirmed by the doubly charged peaks at m /e 120-122. The penta- chlorodibenzofuran and the hexachlorodi- henzofuran fragm ent were identified in an analogous fashion as described above (cf. F:<rs. 2b and c, respectively). The elemental composition of these com- pminds was confirmed by high-resolution mass spectrometric analysis (20, 21) of the total mixture. Peaks of the homologous ser ies C,:H_nOClnwere only detected for n=*4-6, and other halogenated compounds (e.g., chlo- 'onaphthalenes) were present as trace con stituents. > Sixteen major peaks were present on the V.C chromatogram of the second 5% diethyl ' ((r-hexane fraction of the gull eggs, de- d from the first 5% diethyl ether-hex- liiic fraction eluted from the large Florisil column. Heights of these peaks were com parable to those of the Clophen A-60 peaks shown in Figure 1. E xtract volumes and vol umes injected into the gas chromatograph were also comparable. This extract was herefore chosen for analysis by low resolu tion mass spectrometry. The components of all 16 peaks were isolated. Retention times relative to dieldrin on the OV-1 column were: A, 1.38; B, 1.56 and 1.73; C, 2.14; D, 2.32; E, 2.52; F, 2.72; G, 3.14; H, 3.66; I, 4.21; J, 4.65; K, 4.98; L, 5.20; M, 5.95; 6.75; and O, 8.20. No chlorinated diben- zofurans were detected. The majority of the peaks represented chlorine-containing com pounds with mass numbers from 362 to 508 that have not yet been identified. Peak G was subjected to high-resolution mass spectrometric analysis (HRMS) and was found to be mainly a hexachloronaphtha- lene (Cl0Ha33CIa, m /e 331.8252; CioH^CV7 Cl, m/ e 333.8275; C10H,33CVTa 2, m /e ' ~~.S2Q2) Fragm ent ions due to loss of two Cl where also present in these data. Peak N was"also subjected to HRMS anal ysis. A group of intense peaks was ob served at m /e 436.8744, 438.8743, 440.8758, 442.8733, and 444.8622. The compound struc ture has not yet been identified. The entire 5% diethyl ether-hexane frac tion of the gull eggs that had been derived from the initial acetone fraction was ex amined by HRMS. No chlorinated dibenzofuran or dibenzodioxin compounds were de tected. None of the appropriate sea lion liver extracts produced EC peaks of sufficient height to merit analysis, by mass spectro metry. Components of individual peaks of the 5% diethyl ether-hexane fraction of the initial acetone fraction, and of the 25% diethyl ether-hexane fraction derived from the initial 25% diethyl ether-hexane frac tion of the sea lion blubber, were isolated for low resolution mass spectrometry; a por tion of the latter extract was also analyzed by HRMS. No chlorinated compounds of in terest were detected, in part because of the high hydrocarbon content of the samples. A more exhaustive examination of these and other wildlife samples is evidently re quired to permit a conclusion about the pres ence of either chlorinated dibenzodioxins or chlorinated dibenzofurans in the North American environment. Acknowledgement We thank M. Gilbertson for assistance in obtaining herring gull eggs, R. L. DeLong and W. G. Gilmartin for obtaining the sea lion samples, J. G. Vos for providing the Clophen A-60, J. A. Burke for helpful dis cussions on analytical methodology, M. J. T. Mulvihill and J. T. Mendola for technical laboratory assistance, and F. C. Walls for assistance with the mass spectrometry. Financial assistance from the National Science Foundation (Grant GB-11649), the National Aeronautics and Space Administra tion (G rant NGL-05-003-003) and the Can adian Wildlife Service is gratefully acknowl edged. ooptember 1973 197 784517 G E N P 0 1170 6 i REFERENCES 1. Anderson, D. W., and Hickey, Eggshell changes in certain North American birds. Proc. XV Int. Omith. Cong.: 433 (1972). 2. Risebrough, R. W. Effects of environmental pol lutants upon animals other than man. In Pro ceedings Sixth Berkeley Symposium Mathemati cal Statistics and Probability, L. M. Lecam, J. Neyman, and E. L. Scott, Eds., University of California Press, Berkeley-and Los Angeles, 1972, p. 443. 3. Keith, J. A. Reproduction in a population of Herring Gulls ( L o t u s a r g e n i a t u s ) contaminated by DDT. J. Appl. Ecol. (Suppl.) 3: 57 (1966). 4. Vos, J. G., and Koeman, J. H. Comparative toxi cologic study with polychlorinated biphenyls in chickens with special reference to porphyria, edema formation, liver necrosis and tissue resi dues. Toxicol. Appl. Pharmacol. 17: 656 (1970). 5. Vos, J. G., et al. Identification and toxicological evaluation of chlorinated dibenzofuran and chlorinated naphthalene in two commercial poly chlorinated biphenyls. Food Cosmet. Toxicol. 8: 625 (1970). 6. Vos, J. G. Toxicology of PCBs for mammals and for birds. Environmental Health Perspect. No. 1: 105 (1972). 7. Hays, H., and Risebrough, R. W. Pollutant con centrations in abnormal young terns from Long Island Sound. Auk 89: 19 (1972). 3. Verrett, J. Statement before the Subcommittee on Energy, Natural Resources, and the Environ ment of the Committee on Commerce, United States Senate, Ninety-first Congress. Second Session on Effects of 2,4,5-T on Man and the Environment. Serial 91-60, U.S. Government Printing Office, 1970, p. 190. 9. DeLong, R. L., Gilmartin, W. G., and Simpson, J. G. Premature births in California sea lions associated with high organochlorine pollutant residue levels. Science, in press. 10. Ringer, R. K., Aulerich, R. J., and Zabik, M. Effect of dietary polychlorinated biphenyls on growth and reproduction of mink. Paper pre sented at 164th National Meeting, American Chemical Society, Division of Water, Air and Waste Chemistry, New York City, August 28- September 1, 1972. 11. Sparschu, G. L., Dunn, F, L,, and Rowe, V. K. Study of the teratogenicity of 2,3,7,8-Tetrachlo- rodibenzo-p-dioxin in the rat. Food Cosmet. Toxicol. 9: 405 (1971). 12. Zitko, V., Hutzinger, 0., and Choi, p. j Contamination of the Bay of Fundy-Gulf Maine area with polychlorinated biphenyls t i chlorinated terphenyls, chlorinated diberutod!" oxins, and dibenzofurana. Environ. Health P* " spect 1: 47 (1972). r' 13. Bowes, G. W., and Lewis, J. A. Extraction of polychlorinated biphenyls (P C B ): evaluation or a column technique applied to polar bear and seal tissue. J. Assoc. Offlc. Anal. Chem., in. preg*, 14. Porter, M. L., and Burke, J. A. Industrial them- icals: separation of three chlorodibenzo-p-dioxina from some polychlorinated biphenyls by chroma tography on an aluminum oxide column. 54* 1426 (1971). 15. Stanley, R. L., and LeFavoure, H. T. Rapid digestion and cleanup of animal tissues for pes ticide analysis. J. Assoc. Offic. Anal. Chem. 48: 666 (1965). 16. Burlingame, A. L. Data acquisition, processing and interpretation via coupled high-speed realtime digital computer and high resolution mass spectrometer systems. In: Advances in Maas Spectrometry, Vol. 4, E. Kendrick, Ed., The Institute of Petroleum, London, 1968, p. 15. 17. Burlingame, A. L. Developments and applica tions of real-time high resolution mass spec trometry. In: Recent Developments in Mass Spectroscopy, K. Ogata and T. Hayakaw, Eds., University of Tokyo Press, Tokyo, 1970, p. 104. 18. Burlingame, A. L., et al. Real-time high resolu tion mass spectrometry. In: Computers in An alytical Chemistry: Progress in Analytical Chemistry, Vol. 4, C. H. Orr and J. A. Norris, Eds., Plenum Press, New York, 1970, p. 17. 19. Smith, D. H., et al. Real-time organic mass spectrometry: LOGOS--a general laboratory system for high and low resolution GC-M'S and closed-loop applications. Anal. Chem. 43: 1796 (1971). 20. Simoneit, B. R., et ai. Application of real-time mass spectrometric techniques to environmental organic geochemistry. I: General considerations. Arch. Environ. Contamin. Toxicol, in press. 21. Simoneit, B. R., et al. Application of real-time mass spectrometric techniques to environmental organic geochemistry. II: San Francisco Bay area water. Arch. Environ. Contain. Toxicol, in press. 198 Environmental Health Pei-sptcuves 784518 ^ "N TCDD-Induced Changes in Rat Liver Microsomal Enzymes by G.W . Lu c ie r,* O .S . M cDaniel,* B .E .R . H ook,* B.A. Fow ler,' B.R. Sonaw ane,' and E. Faeder in tr o d u c tio n 2,3,7,8-TetrachIorodibenzo-p-dioxin (TC DD), a contam inant of the herbicide 2,4,5trichlorophenoxyacetic acid (2,4,5-T), is ex tremely toxic (I), although the mechanism of toxicity is not known. Other papers pre sented at this conference cover the spectrum of environmental and health hazards of jhlorinated dibenzodioxins and dibenzofu,-ans. It should suffice to say here that these ~ iipounds are teratogens {2-U) in rodents, ) the extensive use of 2,4,5-T, especially in Vietnam, has focused concern on their potential health hazards. Recently TCDD was shown to be an inducer of S-aminoIevulinic acid synthetase in the chick embryo (5) and also to decrease hexobarbital sleep ing times in rats (<?). These reports prompted us to investigate the effects of sublethal doses of TCDD on activities of hepatic mi crosomal and mitochondrial enzymes. The microsomal enzymes include components that are involved in the detoxication of for eign compounds and the regulation of many endogenous compounds such as the steroid hormones (7). Microsomal constituents and activities investigated in this study were: cytochrome P-450, cytochrome bs, *National Institute of Environmental Health Sci ences, National Institutes of Health, F.O. Box 12233, Research Triangle Park, North Carolina 27709. tHuman Studies Laboratory, Environmental Pro tection Agency, Research Triangle Park, North Carolina 27709. benzpyrene hydroxylation, aniline hydroxyla tion, aminopyrine dmthylation, benzphetamine dmthylation, ethylmorphine dm thylation, NADPH cytochrome c reductase, /3-glucuronidase, and TJDP glucuronyltransferase. We also monitored possible changes in oxidative phosphorylation rates in rat liver mitochondria to determine if the toxic action of TCDD could be related to disrup tions in bioenergetic pathways. Materials and Methods Animals Male and female rats (Charles River, CD strain) were used in these experiments. On the day of treatment rats weighed ap proximately 200 g (males 6 weeks old, fe males 8 weeks old). TCDD was adminis tered as a single oral dose in 0.5 ml acetonecorn oil, and controls received 0.5 ml ace tone-corn oil (5). Preparation of Sabceilular Fractions Rats were killed by cervical dislocation, and approximately 4.0 ml blood was imme diately drawn from the dorsal aorta. Livers were removed, minced, and homogenized in 1.15% KC1 buffered with 0.02M N-2-hydroxyethyipiperazine-N'-2-ethanesulfonic acid (H EPES), pH 7.5, at 5C to make a 2 0 % (w /v) mixture. Homogenization was accom plished by using 6 strokes in a motor-driven Potter-EIvehjem homogenizer. Nuclei and cell debris were removed by centrifugation a t 670g for 10 min and mitochondria re- ptember 1973 199 784519 GENP 011708 rJ moved by centrifugation of the 67Op super natant at 10000g for 15 min. Microsomes were pelleted by centrifugation of the postmitochondrial supernatant at 105000 g for 70 min, washed once with HEPES-KCl buf fer, and finally resuspended in HEPES so th at 1.0 ml of microsomal suspension con tained material from 0.5g liver (wet weight). Smooth- and rough-surfaced endo plasmic reticulum (SER and RER) frac tions were prepared by homogenizing- chop ped liver sections in 0.25AT sucrose (pH 7.0) and preparing the microsomal subfractions on discontinuous sucrose gradients by the procedure of Gram et al. (9), 10 ml of postmitochondrial supernatant and 12 ml of 1.3M sucrose being used. Liver mitochondria were prepared by the procedure of Nelson et al. (1 0 ) and resuspended in 0.25A/ sucrose so that 2.0 ml of suspension contained mito chondria from 3.0g liver (wet weight). Assay Methods Cytochrome P-450 was measured by its carbon monoxide difference spectra in an ACTA III spectrophotometer following re duction with dithionite, and cytochrome b* was measured by its difference spectra fol lowing reduction with NADH (1 1 ). For the determination of in vitro microsomal hydroxylation of aniline and dmthylation of aminopyrine and ethylmorphine, the pre viously described incubation medium (12) was employed, with the exception th at HE PES buffer was used instead of Tris buffer. Concentrations of substrates were: 3.5mA/ aniline, 2.5mM ethylmorphine, or 2.5mM am inopyrine in 3.0 ml incubation medium. En zyme reactions were started by the addi tion of 1.5-2.0 mg microsomal protein. Benz pyrene (BP) hydroxylation and benzphetamine dmthylation rates were determined by using the incubation medium of Hook et al. {13). There was essentially no difference when an NAD PH regenerating system {12) or saturating concentrations of NADFH (3.1 mM) were used in the incubation medium. Aniline hydroxylation was quanti fied by the method of Kato and Gillette (14). Formaldehyde released by the dm thylation of benzphetamine, aminopyrine, and ethylmorphine was measured by the Nash reaction (15). BP hydroxylation was measured by the fluorescence method of Wattenberg et ai. {16). NADPH cytochrome c reductase was measured by the reduced cyto chrome c peak at 550 nm (17). ^-Glucuroni dase was determined by the modified meth od (18) of Talalay et al. {19), phenoiphthalein /3-D-glucuronide being used as the sub strate. p-Nitrophenol glucuronyltransferase was determined spectrophotometrically {20) by using 0.9mM p-nitrophenol, 0.8mA/ UDPGA, 10mM MgCL, and Triton X-100-treated microsomes {21). A fter 3 min incubation, the reaction was stopped by the addition of 5.0 ml glycine buffer, pH 10.4 {21). Experi mental data for glucuronyltransferase were similar in all cases whether activity was measured by p-nitrophenol disappearance {20) or p-nitrophenyl /3-D-glucuronide ap pearance at 312 nm (22). Oxidative phos phorylation rates in isolated liver mitochon dria were measured polarographically with a Clark oxygen electrode. The reaction mix ture contained 120mM KC1, 12mM sub strate (succinic acid), 8mM MgCL, 5mM K-HP04, 10.0mA/ ADP and 20mM glyclgly cine buffer (pH 7.4). The total volume was 1.6 mi and the tem perature was maintained at 30C. Oxygen content in the vessel was cal ibrated by using NADH. Microsomal and mi tochondrial protein contents were determined by the method of Lowry et ai. (23). Results and Discussion Time-Course Studies Male rates were administered TCDD as a single oral dose at 5 or 25 ng/kg, and hepa tic microsomal enzyme activities and cyto chrome contents were measured 1, 3, 9, 16, and 28 days after treatment. The purposes of this study were to determine whether TCDD affected microsomal enzyme activi ties and, if so, to determine the time-course alterations in enzyme activities. The LDao value for TCDD is approximately 100 jug/kg (John Moore, personal communication), ar.a no lethality of TCDD to male rats was ob served at 5 or 25 /ig/kg in test animals. 200 Environmental Health Perspectives 784520 GENPoil709 I-- jniline hydroxylation--Time-course ef fects of TCDD on aniline hydroxylation are presented in Figure 1. Aniline hydroxyla tion is expressed as nanamoles p-aminopheol formed per minute per milligram pro tein. Enzyme activity was slightly but not significantly enhanced at day 1. By day 3, hydroxylation rates were increased over 100%, and the same level of induction was observed through day 16. A fter day 16, en zym e activities began to return to control values, although aniline hydroxylation was still significantly elevated 38 days after Treatment with 25 /ig/kg TCDD. F igure 2. Time-course effects of a single oral dose of TCDD on liver microsomal aminopyrine d mthylation. An asterisk indicates that values are significantly different from controls at P < 0.05. N = 3 male rats. Figure 1. Time-course effects of a single oral dose of TCDD on liver microsomal aniline hydroxyla tion. An asterisk indicates that values are signifi cantly different from controls at P < 0.05. JV = 3 male rats. Figure 3. Time-course effects of a single oral dose of TCDD on liver microsomal cytochrome P-450. An asterisk indicates th at values are significantly different from controls a t P < 0.05. N -- 3 male rats. Aminopyrine dmthylation-- The effects on aminopyrine dmthylation were oppos ite those observed for aniline hydroxylation (Fig. 2 ). Specific enzyme activity was de creased approximately 30% a t days 3, 9, and 16 by the 25 /ig/kg dose. Values were essentially unchanged at the lower dose. Cytochrome P-450 and 65--Cytochrome P-450 was increased by 40% at day 1 and cy tochrome b was unchanged at day 1, while at day 3 contents of both microsomal cyto chromes were elevated although P-450 was increased more than b9 (Figs. 3 and 4). Nine days after treatm ent P-450 and' b5 were both increased by 60%. The lag period b5 effects compared to P-450 might be September 1973 F igure 4. Time-course effects of a shingle oral dose of TCDD on liver microsomal cytochrome b*. An asterisk indicates that values are significantly different from controls at P < 0.05, N = 3 male rats. related to the slower turnover-rate of cyto chrome b5 (24). Thirty-eight days after TC DD treatm ent (25 ig/kg), b3" content was increased by 60% and P-450 was increased by 40%. Increased Pr450 content was.asso- 201 784521 GENP011710 dated with increased oxidative hydroxyla tion and decreased oxidative dmthylation. This induction pattern is similar 'to-"that observed for 3-methylcholanthrene (25, 26). However, TCDD appears not to shift the peak in the carbon monoxide difference spectra from 450 nm to 448 nm such as oc curs with 3-methylcholanthrene induction (26). TCDD obviously cannot be considered a phenobarbital-type inducer, which is char-, acterized by increased P-450 content, in creased hydroxylation activity, and in creased oxidative ^-dmthylation (27). UDP glucuronyitransferase--Effects on glucuronyltransferase were the most strik ing observed in the time-course study (Fig. 5). Following TCDD treatment at 25 #ig/kg enzyme activity was enhanced by 51% on day 1, 162% on day 3, 565% on day 9, 636% on day 16, 154% on day 28, and 162% on day 38. Levels of increases were slightly less at the 5 /*g/kg dose compared to the 25 fig/k g dose, and time-course effects were similar. Increased glucuronlytransferase ac tivity was not associated with changes in km values for substrate (p-nitrophenol, 0.26mM) or c-factor (TJDPGA, 0.5SmM) (28). Vans in control animals was 126 nmole pnitrophenol conjugated/min-mg protein com pared to 539 in microsomal preparations from TCDD-treated rats (28). UDP glucuronyltransferase activity is phospholipid-dependent (29, 30), and microsomal cholesterol has been theorized to function in the main tenance of endoplasmic reticulum structure (31). However, TCDD elevation of glucuronyltransferase activity does not appear to be related to alterations in total microsomal phospholipid and cholesterol levels (28), al though individual microsomal phospholipids have not been quantified following TCDD treatment. Divalent cations and detergents are in vitro activators of microsomal glucuronyltransferase (2 1 ) and the effects of TCDD might be related to detergentlike ac tions on the endoplasmic reticulum or to mobilization of endogenous magnesium or other stimulatory divalent cations. However, the magnitude of the effect on glucuronyltransferase was the same whether glucuron- 300 PlGUBE 5. Time-course effects of a single oral dose of TCDD on liver microsomal UDP glucuronyltransferase. An asterisk indicates that values are significantly different from controls at P < 0.05. .V = 3 male rats. yltransferase was measured in the presence or absence of Mg*2 or Triton X-100 (28). These data suggest that TCDD effects on glucuronyitransferase are not related to mor phological alterations in endoplasmic reti culum structure, although this possibility has not been excluded. However, at this stage it appears that the possibilities that best fit the experimental data are related to in creased enzyme synthesis or decreased de gradation rates. TCDD induction of ALA synthetase in the chick embyro was blocked by cycloheximide (5) but data from proteinsynthesis inhibition experiments would be difficult to obtain in rats due to the lag per iod in TCDD induction and the rapid toxi city of most antimetabolities. Elevation of glucuronyitransferase occurred in kidney microsomes as well as liver microsomes (28), although distribution studies have demon strated that liver accumulates OCDD-Cr8 equivalents at much higher levels than kid neys (32). TCDD did not affect glucuronyltransferase when added directly to the in cubation medium at 10*8 M. Microsomal protein--Since enzyme activi ties and cytochrome contents were measured on a per-milligram protein basis, it was of importance to measure time-course effects on microsomal protein contents. Previous 202 Environmental Health Perspectives 784522 GENP 0 1 1 7 1 1 S .,,.n/iyts show that chronic exposures to chlocd triphenyl compounds markedly inid microsomal "protein contents (33). Hepatic ultrastructural studies of TCDD:rented rats revealed general proliferation f RER and increases in SER in specific t'patic areas (34). In our studies, micro somal protein contents were not signifi cantly changed till 28 days after a single TCDD treatm ent when levels were enhanced by 60fo (Table 1). Therefore, enzyme ac tivities per gram liver had essentially the sam e relative values between control and TCDD-treated rats as when enzyme activi ties were calculated per milligram protein. Other enzymes--NADPH cytochrome c reductase and /3-glucuronidase were not af fected at either dose or a t any period during the time-course experiment. Liver function--Possible hepatotoxicity was monitored by serum orinthine transcarbamylase activities (35). Results show that after rats received 5 or 25 ^g TCDD/kg there vere no indications of liver damage at any time period. on oxidative hydroxylations and dmthyla tions. Single doses of TCDD at 0.2, 1.0, 5.0, and 25 /ig/kg were used, and microsomal en zymes assayed 3 days after treatment. Table 2. Changes in activities of male rat liver micro somal enzymes following a single oral dose of TCDD.1 Enzyme Cytochrome P-450 Cytochrome b* Aminopyrine demethylation Benzphetamine demethylation Ethylmorphine demethylation Aniline hydroxylation Benzpyrene hydroxylation Glucuronyl- transferase Protein Relative change from control values (100)" TCDD, TCDD, TCDD, TCDD, 0.2 1.0 5.0 25.0 Mg/kg Mg/kg Mg/kg Mg/kg 119 162* 184* 193* 101 139* 167* 195* 103 90 86* 72" 101 68* 70* 59* 101 79 77 124* 160* 202* 198* 102 163* 163* 467* 138 * 167* 112* 113 385 1 471* 105 126* -Response Studies Dose-response relationships were exam ined in both male and female rats to obtain information concerning sex differences of hepatic microsomal responses to TCDD and to determine what is the lowest dose that results in induction of microsomal en zymes. Enzymes investigated in these stud ies included those used in the time-course experiment plus BP hydroxylation, benzphetamine demethylation, and ethylmorphine demethylation. These enzymes were added so th at a more extensive comparison could be made on the different effect of TCDD ' Rats were killed 3 days after TCDD treatment at various TCDD dose levels. Each value is derived from four animals. b Control values S.D. were: cytochrome P--450, 0.66 0.08 nmole/mg protein; cytochrome bs, 0.40 * 0.01 nmole/mg protein; aminopyrine de methylation, 9.1 0.4 nmole formaldehyde released/min-mg protein ; benzphetamine de methylation, 8.1 1.5 nmole formaldehyde released/min-mg protein, ethylmorphine de methylation, 15.9 i 3.0 nmole formaldehyde released/min-mg protein; aniline hydroxylation, 3.1 0.5 nmoles p-aminophenol foimed/minmg protein; gluouronyltransferase, 49.1 0.3 nmoles p-nitrophenol conjugated/min-mg pro tein ; microsomal protein, 25.3 --2.7 mg/g liver. * Significantly different from controls at P<0.05. Table 1. Time-course effects of a single oral dose of TCDD on microsmal protein.1 Dose, TCDD, Mg/kg 0 5 25 1 day 16.8 1.7 18.4 i 3.3 17.8 A 3.6 Microsomal protein after TCDD treatment, m g/g liver 3 days 9 days 16 days 28 days 16.8 0.5 19.0 - 5.4 23.1 3.2 b 18.1 3.0 21.8 2.2 19.1 1.2 19.3 1.8 21.2 i 0.9 20.6 3.5 17.6 2.6 28.2 3.1 29.4 3.5* 1Values at various times after TCDD treatment; N = 3 male rats. b Significantly different from controls at P <0.05. ember 1973 203 784523 Males--Data on the effects of TCDD on male rat liver microsomal enzymes'are pre sented in Tables 2 and 3. Increases in cyto chrome P-450 and bs contents were not evi dent until animals received 1.0 ig/kg and the level of induction was dose-dependent up to 25 ig/kg. Activities of oxidative d mthylation enzymes for all three substrates were decreased in a dose-dependent manner. Aniline hydroxylation was enhanced by 0.2 tgTCDD/kg (24%) and maximum increases occurred after 5.0 fig TCDD/kg (102%). In duction of BP hydroxylation was similar in magnitude to aniline hydroxylation at the three lower doses but after male rats received 25 fig TCDD/kg, BP hydroxylation was increased by 300% and aniline hydro xylation by 100%. Glucuronyltransferase was increased by 38, 67, 285, and 371% at the four doses from the lowest to highest, respectively. When enzyme activities per nmole cytochrome P-450 were calculated, oxi dative dmthylation values of the three sub strates tested were significantly decreased (50-70%) at TCDD doses of 1.0 /xg/kg or greater (Table 3). Hydroxylation values per P-450 unit were essentially unchanged, with the exception that BP hydroxylation per unit of P-450 increased by approximate ly 100% following a dose of 25 fig TCDD/kg. Table 4. Changes in activities of female rat liver aicrosomal enzymes following a single oral dose of TCDD,* Enzyme Cytochrome P--450 Cytochrome b Aminopyrine dmthylation Benzphetamine dmthylation Benzpyrene hydroxylation Clucuronyltransf erase Protein Relative change from control values {10G) TCDD, 0.2 rg / k g 126 e 122 * TCDD, 1.0 fig/ kg 153* 131* TCDD, 5.0 t g / kg 196* 158 * 120 e 131' 120' 115 112 118 783 * 257' 94 1225' 506 108 ' 1403' 487' 115 * *Rats were killed 3 days after TCDD treatment at various TCDD dose levels. Each value de rived from four animals. " Enzyme activities were expressed as indicated in Table 2. Control values S.D. were; cyto chrome P-450, 0.45 0.06 nmole/mg protein; cytochrome b,, 0.36 0.03 nmole/mg protein; aminopyrine dmthylation, 4.4 0.3 nmole formaldehyde/min-mg protein; benzphetamine dmthylation, 2.34 0.45 nmole formaldehyde/min-mg protein; benzpyrene hydroxyla tion, 0.06 0.01 nmole/mg-min protein; glu curonyltransferase 23.9 5.4 nmole/min-mg protein; 'and microsomal protein (21.0 5.0 mg/g liver). ' Significantly different from controls at P<0.05. Table 3. Effect of TCDD on mixed function oxidase activity per cytochrome P--450 unit in male rat liver microsotnes Enzyme Aminopyrine dmthylation Benzphetamine dmthylation Ethylmorphine dmthylation Aniline dmthylation Benzpyrene hydroxylation Activity, nmole substrate metabolized/nmole P--450 TCDD, TCDD, TCDD. 0 0.2 Mg/kg 1.0 Mg/kg 5.0 Mg/kg 13.7 <4a 1.0 12.4 1.5 7.6 0.8 " 6.4 0.7" 12.2 1.8 10.9- 2.3 5.1 0.9 4.6 0.4 * 23.3 1.5 21.3 3.8 11.7 -* 2.2 " 10.0 4 - 1.6 " 3.4 0.2 3.2 0.7 3.8 0 . 1 3.7 Am 0.3 0.80 * 0.14 0.72 0.17 0.80 -4 0.03 0.71 0.07 *Rats sacrificed 3 days after TCDD treatment at various dose levels. iV = 4 male rats. bSignificantly different from controls at P <0.05. TCDD. 25.0 Mg/kg 4.6 0.3" _ 3.5 0.6 1.95 0.2 204 Environmental Health Perspective ' 784524 GENP 011713 Table 5. Effect of TCDD on mixed function oxidase activity per cytochrome P-450 unit in female rat liver microsomes.* ---- Enzyme Aminopyrine dmthylation Benzphetamine dmthylation Ethylmorphine dmthylation Benzpyrene hydroxylation 0 9.9 -+* 1.5 Activity, nmole substrate metabolized/nmole P-450 TCDD, 0.2 f i g / k g TCDD, 1.0 M g / k g TCDD, 5.0 t ig / kg 9.3 0.2 8.4 0.6 6.1 1.0 " 5.1 1. 4.7 0.6 3.7 0.7 3.1 0.4B 7.3 1.4 6.4 0.5 4.4 0.2 * 3.4 0.76 0.13 0.05 0.8 0.1h 1.0 0.21 " 0.96 0.16 h 1Rats sacrificed 3 days after TCDD treatment. N = 4 male rats. " Significantly different from controls at P <0.05. Females-- Data on the effects of TCDD on female microsomal- enzymes are presented in Tables 4 and 5. Female rats were more susceptible to TCDD induction of BP hydrox ylation and glucurony ltransferase than male rats (Table 4). This sex difference was quite evident following a dose of .2 fig/kg; male liver microsomal glucuronyltransfer ase increased 38% and female liver microf mal glucuronyltransferase increased 157 %f BP hydroxylation increased 2% and remale BP hydroxylation increased 683%. In control animals, BP hydroxylation rates were eight times greater in liver microsomes from males compared to females, but in TCDD-treated rats activities of liver micro somal BP hydroxylase were- approximately the same in both sexes. Glucuronyltransfer ase activity of liver microsomes from males was twice that of females in controls, but in TCDD-treated rats (0.2 fig/kg) activities were higher in microsomes from females than males. Oxidative dmthylation acti vits were two to four times as high in mi crosomes from control males compared to females. Female hepatic microsomal N-demethylations were slightly increased by TCDD, whereas corresponding enzyme activ ities in male hepatic microsomes were de creased although V-demethylation rates were still higher in TCDD-treated males com pared to TCDD-treated females at all dose levels. Since maximum elevation of P-450 in females, as in males, was approximately 0 %, iV-demethylations per unit P-450 were decreased, but in general the observed decrease in V-demethylations per P-450 unit in females was not as much as that in males. These data demonstrate that TCDD mark edly increases activity of some microsomal enzymes, particularly glucuronyltransferase and BP hydroxylase, and that female rats are more susceptible to action of TCDD than males. Increases in activity of microsomal enzymes after a dose of 0.2 fig TCDD/kg is quite significant in comparison to doses re quired for effects by other inducing agents. TCDD is approximately 100,000 times more potent an inducing agent than phnobarbi tal or 3-methylcholanthrene on a ig/kg basis in rats. In addition to our studies. Hook et al. (3fi) have shown that a single TCDD dose of 0.2 fig/kg to female rats increased biphenyl 2-hydroxylation by approximately 900% and biphenyl 4-hydroxylation by ap proximately 100%. Therefore, an oral dose of 40 ng TCDD to 200 g rats markedly in creases activity of microsomal enzymes. Norback et al. (32) report that 95% of labeled TCDD is excreted in the feces following oral administration. Therefore, it appears that only a small portion of the adminis tered TCDD reaches the liver, although it is possible that much of the fecal radioactivity has been added via biliary excretion and that intestinal absorption rates might vary with dose. ALA synthetase activity was in creased in chick embryos by extremely low concentrations of TCDD (5), but in the same group of rats used in our studies, hepa- September 1973 205 G EN P 011714 784525 tic ALA synthetase was not changed (37) by doses 50 times that needed to increase UDP glucuronyltransferase, benzpyrene hy droxylation, and biphenyl 2-hydroxyIation. The extreme sensitivity of microsomal en zymes to TCDD body burdens suggest that alterations in activities of these enzymes could be related to the toxic action of TCDD and its teratogenic effects by disrupting nor mal steriod regulation. Adrenal size is not enlarged in TCDD-treated rats (38), indi cating that if steriod excretion is enhanced the compensatory feedback mechanism con trolling steroid synthesis may not be oper ative. Effects of TCDD on SER and RER Effects of TCDD on the distribution of microsomal components in SER and RER of male rats are summarized in Table 6. SER to RER ratios were decreased in all parameters tested following TCDD treat ment (25 ig/kg). AT-Demethylation ratios (SER:RER) were approximately 2.4 in con trols compared to 0.7 in treated rats. Speci fic dmthylation activities were decreased by 75% in SER and were essentially un changed in RER. BP hydroxylation was elevated in both SER and RER, but induc tion was greater in RER resulting in de creased SER:RER from 1.77 to 1.19. q ju curonyltransferase was also markedly jn] creased in both subfractions (SER, 200' " RER, 300%) and SER?RER decreased from 0.56 to 0.37. Microsomal protein also exhi bited decreased SER.'RER following TCDD treatm ent although the change was not sig nificant. These changes in SER:RER ra tios are similar to those seen after 3-methvlcholanthrene treatment (26). Ultrastructur al studies revealed overall RER proliferation and SER proliferation in isolated regions of rat liver hepatocytes (34). Effects on Oxidative Phosphorylation The gradual wasting of animals that pre cedes death in TCDD-exposed animals sug gested that toxicity might be an expression of bioenergetic disturbances. However, oxi dative phosphorylation rates in isolated rat liver mitochondria from TCDD-treated rats (5 or 25 itg TCDD/kg) did not significantly differ from those in controls (Table 7). Parameters investigated were state 3 respir ation, state 4 respiration, respiratory control (R.C.) (state 3/state 4) and A D P:0. The only difference observed between control and treated rats was greater uncoupling rates of the treated group on storage at 4 C. Table 6. Sabmicrosomal distribution of male rat liver microsomal enzymes following a single oral dose of TCDD (25 fig/kg).* EnzymeB Aminopyrine dmthylation, nmole HCHO/min-mg Benzphetamine dmthylation, nmole HCHO/min-mg Benzpyrene hydroxylation, nmole/min-mg Glucuronyltransferase, nmole/min-mg Protein, mg/g liver SER 9.3 1.0 Control RER 4.0 0.6 SER: RER 2.34 9.1 1.7 3.S 0.4 2.38 0.46 0.09 0.26 0.08 1.77 69.8 17.2 124.5 31.9 0,56 5.9 0.5 12.8 0.6 0.46 SER 2.6 0.1 TCDD RER 3.6 0.7 SER: RER 0.73* 2.1 0.1s 3.5 0.4 0.61* 2.4 0.3* 2.0 0.2* 1.19* 205.8 :22.8 * 556.7 87.9 " 0.37* 6.2 0.4 16.8 * 1.8* 0.37* *Rats were killed six days after TCDD treatment. jV = 4 mala rats. " Enzyme activities expressed as outlined in footnotes b to Table 2. *Significantly different from controls at P<0.05. 206 Environmental Health Perspectives T / T I A JTK T'nro . 784526 Table 7. Effects of a single oral dose (25 Mg/kg) of TCDD on oxidative phosphorylation rates in rat liver mitochondria.* Time after treatment, days Controls 1 3 6 9 16 28 State 3 respiration, ng-atom O/min-mg protein 13S 14 133 7 159 22 145 4 127 6 140 13 146 5 State 4 respiration, ng-atom 0 / min-mg protein 34 5 31 3 37 7 37 2 31 2 36 1 37 3 R.C. 4.06 4.33 4.27 3.92 4.10 3.89 3.95 ADPrO 1.85 0.06 1.88 0.04 1.71 0.09 1.73 0.08 1.73 0.14 1.78 0.21 1.75 0.07 Succinate used as the substrate; each value (mean S.D.) derived from an average of four male rats. Summary vealed that TCDD markedly decreased SEE Male or female rats' were administered a single oral dose of TCDD at 0.2, 1.0, or 5.0 ug/kg, and activities of hepatic microsomal enzymes were monitored three days after treatment. Our data demonstrate thiat TCDD has an extremely potent effect on some microsomal enzymes, particularly glucuron- to RER ratios in all parameters tested. There was no biochemical or histologic evidence of hepatotoxicity nor any effects on oxida tive phosphorylation rates in liver mito chondria. These studies indicate that hepa tic microsomal enzymes are extremely sensi tive to TCDD body burdens. vltransferase and benzpyrene hydroxylase, hat female rats may be more susceptible Needs for Further Research ODD actions than males. Marked in Our studies to date have served only to creases in hepatic enzyme activity in female characterize the effects of TCDD on hepatic rats was observed following a single oral microsomal enzymes following animal ex dose of 0.2 fig TCDD/kg (LDso-100 Mg/kff) posures. The need for future research re (glucuronyltransferase + 157%, benzpyrene lated to TCDD-microsomal interactions are hydroxylation + 683 % ), and the levels of in many, and the following represents a list of duction increased with dose so that after 5.0 some of the most urgent research needs: Mg TCDD/kg glucuronyltransferase was in (1 ) study microsomal effects of in vitro duced 500% and benzpyrene hydroxylation addition of TCDD and related compounds 1400%. TCDD also increased cytochrome P- directly to the incubation medium; (2 ) com 450, cytochrome bs, and aniline hydroxyla-" pare inductive properties of TCDD with 3- tion, whereas oxidative demethylations of methylchoianthrene and other inducers; aminopyrine, ethylmorphine, and benzphet- (3) determine structural requirements for amine were decreased. NAD PH cytochrome induction by using structural analogs of c reductase and /3-glucuronidase were unaf TCDD as effectors of microsomal enzymes; fected by any TCDD dose. Time-course stud (4) determine if TCDD induces extrahepa- ies revealed that increases reached a plateau tic microsomal enzymes and determine levels 3 days after TCDD treatm ent following an of induction in species other than the ra t; initial lag period. Increased levels were (5 ) determine if induction of microsomal maintained at the day 3 values through day enzymes is related to increased synthesis, 16 a fte r which time activities began to re decreased degradation, or membrane effects; turn to normal although effects were still (G) determine rates of in vivo metabolism evident 38 days after treatment. Subfrac and excretion of test compounds, including tion of microsomes into SEE and BEE re steriods, following TCDD treatm ent; (7) ember 1973 207 GENP011716 784527 determine if maternal exposure of TCDD induces fetal enzymes; and (8 ) study pos sible relationships between TCDD induction of microsomal enzymes and teratogenic ef fects. REFERENCES 1. Kimbrough, R. D. Toxicity of chlorinated hydro carbons and related compounds. Arch. Pathol. 94: 125 (1972). 2. Sparschu, G. L.f Dunn, F. L., and Rowe, V. K. Teratogenic study of 2 ,3,7,8-tetrachlorodibenzop-dioxin in the rat. Toxicol. Appl. Pharmacol. 17: 317 (1970). 3. Clegg, D. J. Embryotoxicity of chemical con taminants of foods. Food Cosmet. Toxicol. 9: 195 (1971). 4. Sparschu, G. L., Dunn, F. L., and Rowe, V. K. Study of the teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Food Cosmet. Toxicol. 9: 405 (1971). 5. Poland, A., and Glover, E. 2,3,7,8-Tetrachiorodibenzo-p-dioxin: A potent inducer of -amino levulinic acid synthetase. Science 179: 476 (1973). 6. Greig, J. Effect of 2,3,7,8-tetrachlorodibenzo-1,4dioxin on drug metabolism in the rat. Biochem. Pharmacol. 21: 3196 (1972). 7. Parke, D. V. Biochemistry of Foreign Com pounds, Pergamon Press, New York, 1968, pp. 34-35. 8. Harris, M., Moore, J., and Vos, J. General bio logical effects of TCDD in laboratory animals. Environ. Health Perspect. No. 5: 101 (1973). 9. Gram, T. E., Hansen, A. R., and Fouts, J. R. The submicrosomal distribution of hepatic UDP glucuronyltransferase in the rabbit. Biochem. J. 106: 587 (1968). 10. Nelson, B. D., Drake, R., and McDaniel, O. Effects in vitro and in vivo of methylenedioxphenyl compounds on oxidative phosphorylation in rat liver mitochondria. Biochem. Pharmacol. 20: 1139 (1971). 11. Omura, T,, and Sato, R. The carbon monoxide binding pigment of liver in microsomes. J. Biol. Chem. 239: 2370 (1964). 12. Lucier, G. W., et al. Effects of methylmercory hydroxide on rat liver microsomal enzymes. Chem. Biol. Interact. 4: 265 (1972). 13. Hook, G. E. R., Bend, J. R., and Fouts, J. R. Mixed-function oxidases and the alveolar macro phage. Biochem, Pharmacol. 21: 3267 (1972). 14. Kato,.R,, and Gillette, J. R. Effect of starvation on NADPH-dependent enzymes in liver micro somes of male and female rats. J. Pharmacol. Exptl. Therap. 150: 279 (1965). 15. Nash, T. The colorimetric estimation of formaldehyde by means of the Hantzsch reaction. Bio chem. J. 55: 416 (1953). 16. Wattenberg, L. W., Leong, J. L., and Strand, P. J. Benzpyrene hydroxylase activity in the gastrointestinal tract. Cancer Res. 22: noQ (1962). 17. Masters, B. S., Williams, C. H., and Kamin, H. The preparation and properties of microsomal TPNH cytochrome c reductase from pig liver. In: Methods of Enzymology, R. W. Estabrook and M. E. Pullman (Eds.), Academic Press, New York, Vol. X, 1967, p. 565. 18. Lucier, G. W., and McDaniel, O. S. Alterations in rat liver microsomal and lysosomal j. glucuronidase by compounds that induce hepatic drug-metabolizing enzymes. Biochim. Biophys. Acta 261: 168 (1972). 19. Talalay, P., Fishman, W. H., and Huggins. C. Chromogenic substrates II. Phenolphthalein glu curonic acid as a substrate for the assay of glucuronidase activity. J. Biol. Chem. 166: 757 (1946). 20. Hollman, S. and Touster, O. Alterations in tissue levels of UDP glucose, UDP glucuronic acid pyrophosphatase, and glucuronyltransferase in duced by substances influencing the production of ascorbic acid. Biochim. Biophys. Acta 26: 33S (1962). 21. Lucier, G. W., McDaniel, 0. S,, and Matthews, H. B. Microsomal rat liver UDP glucuronyltransferase: Effects of piperonyl butoxide and other factors on enzyme activity. Arch. Biochem. Biophys. 145: 520 (1971). 22. Temple, A. R., Done, A. K., and Clement, M. S. Studies of glucuronidation III. Measurement of p-nitrophenyl glucuronide. J. Lab. Clin. Med. 77: 1015 (1971). 23. Lowry, O. H., et al. Protein measurement with Folin Phenol reagent. J. Biol. Chem. 193: 265 (1951). 24. Bock, K. W., and Siekevitz, P. Turnover of heme and protein moieties of rat liver microsomal cytochrome bs. Biochem. Biophys. Res. Commun. 41: 374 (1970). 25. Sladek, N. E.( and Mannering, G. J. Evidence for a new P-450 hemoprotein in hepatic micro somes from methylcholanthrne-treated- rats. Biochem. Biophys. Res. Commun. 24: 668 (1966). 26. Gram, T. E,, Rogers, L. A., and Fouts, J. REffects of pretreatment of rabbits with phno barbital or 3-methylcholanthrene on the distri bution of drug-metabolizing enzyme* activity in subfractions of hepatic microsomes. J. Pharma col. Exptl. Therap. 157: 435 (1967). 27. Cram, R. L., Juchau, M. R., and Fouts, J. R* Differences in hepatic drug-metabolism irt vanous rabbit strains before and after treatment with phnobarbital. Proc. Sac. Exptl. Biol. Med. 118: 872 (1965). ^ C 208 Environmental Health Perspectives i / . T T n _tk 784528 Oj 23, Lucier, G. W., Hook, G. E. R., and McDaniel, " ~ ''i>. S. Mechanism of induction of UDP glu- / jronyltransferase by 2,3,7,3-tetrachlorodibenzop-dioxin. Submitted to Biochem. J. 29. Vessey, D. A. and Zakim, D. Regulation of microsomal enzymes by phospholipids II. Acti vation of hepatic uridine diphosphate glucuronyltransferase. J. Biol. Chem. 246: 4649 (1971). 30. Vessey, D. A. and Zakim, D. Regulation of microsomal enzymes by phospholipids IV, Spe cies differences in the properties of microsomal UDP-glucuronyltransferase. Biochim. Biophys. Acta 268: 61 (1972). 31. Deenen, L. L. M. van. Phospholipids and bio membranes. Prog. Chem. Fats. Lipids 8: 1 (1966). :i2. Norback, D. H., Engblom, J. F., and Allen, J. R. Chlorinated dibenzo-p-dioxin distribution within rat tissues and subfractions of the liver. En viron. Health Perspect. No. 5: 233 (1973). 33. Norback, D. H. and Allen, J. R. Chlorinated aro matic hydrocarbon induced modifications of the hepatic endoplasmic reticulum: Concentric mem brane arrays-.. Environ. Health Perspect. No. 1: 137 (1972). 34. Fowler, B. A., et al. Ultrastructural changes in rat liver cells following a single injection of TCDD. Environ. Health Perspect. No. 5: 141 (1973). 35. Faeder, E. Hepatotoxicity evaluation by serum ornithine transcarbamyiase. In preparation. 36. Hook, G. E. R., and Lucier, G. W. Induction of biphenyl 2- and 4-hydroxylation by 2,3,7,8-tetrachlorodibenzo-p-dioxin. In preparation. 37. Woods, J. S. Studies on the effects of 2,3,7,8tetrachlorodibenzo-p-dioxin on mammalian hepa tic aminolevulinic acid synthetase. Environ. Health Perspect No. 5: 221 (1973). 38. Vos, J. G., Moore, J. A., and Zinkl, J. Effect of TCDD on the immune system of laboratory ani mals. Environ. Health Perspect. No. 5: 149 (1973). iptem ber 1973 209 784529 Oo Effects of 2,3,7,8-Tetrachlorodibenzop-dioxin on Drug Metabolism and Hepatic Microsomes of Rats and Mice by J.B . Greig* and F . De Mattel's* Introduction It has been reported (I) that one effect of a single, intra-peritoneal dose of 2,3,7,8tetrachlorodibenzo-p-dioxin (dioxin) is to cause a reduction in the duration of action . of the drug 2-amino-5-chIorobenzoxazoIe [ (zoxazolamine) in the rat. This effect is 'maximal with doses of dioxin above 100 ig/ kg and significant even at 5 pg/kg. Our own work showed that oral dosage had a similar effect but also that the duration of action of hexobarbitone was considerably pro longed (2). The results described here ex tend these observations. Materials and Methods Animals Rats of the albino Porton strain bred in these Laboratories and weighing 180-200 g (male 6 -8 weeks and female 7-10 weeks old) had free access to water and diet 41B un less otherwise indicated. Mice of the C57BL/6 and DBA/2 strains were bred in these laboratories and kept on Sterolit bedding (Mineral & Chemical Cor poration of America, Menlo Park, New Jer sey, U.S.A.) for at least 3 weeks before use. They had free access to diet and water. MRC Toxicology Unit, Medical Research Council Laboratories, Woodmansterne Road, Carahalton, Surrey, England. September 1973 Dosages Dioxin, prepared as described elsewhere (5), was administered as a solution (100 pg/ ml) in Arachis oil. Control animals received an equivalent volume of oil. Zoxazolamine (McNeil Laboratories, Inc.) was dissolved in I N HC1 (1.2 ml/1 0 0 mg), diluted with 0.9% NaCl to 10 mg/ml and administered IP at 100 mg/kg. 5-Cyclo-hex-r-enyi-l,5-dimethylbarbituric acid (hexobarbitone) sod ium (May & Baker Ltd.) was dissolved in water at 50 or 25 mg/ml and administered IP at 150 m g/kg (male rats), 75 m g/kg (fe male rats), 100 m g/kg (male and female mice). dZ-Ethionine (Koch-Light) was dis solved in water (20 mg/mi) and administer ed as indicated in the text. Microsomal Preparations Liver microsomes were isolated either as described by Bond and De Matteis (4 ), ex cept that the microsomes were sedimented at 105,000#, or by the calcium/sucrose meth od 05) adapted to the extent of sediment ing the calcium-treated microsomes at 19000 and washing them once in a 0.0125M sucrose solution containing 8mM CaCL. The final preparations were suspended in O.likf phos phate buffer (pH 7.4) containing XmM Na2EDTA (1.4 -7 .0 mg protein/m l). Protein was estimated by the biuret method (5) and cytochrome P-450 (or P-448) by using an 211 784530 G E N P 011719 * .extinction coefficient of 91/m3f-cm _ (7) for the absorbancy change between 450 (448) and 490 nm of the CO difference spec trum of a Na.SaO^reduced suspension. Assays Plasma barbiturate levels following hexobarbitone administration were measured by the method of Chromy and Babjuk (5). Hexobarbitone oxidase was measured in vitro by the modified (9) method of Brodie et al. (10). Zoxazolamine hydroxylase was measured by the methods of Burns and his colleagues (11, 12) with minor modifications. Difference Spectra ` Difference spectra were recorded on a Unicam SP 1800 spectrophotometer with oxi dized microsomal suspensions (2.7 ml, 1.42.2 mg protein /m l) in each 1-cm cuvet and the addition of ^either aniline (5-200 jJ, 0.22M in 0.1M phosphate buffer, pH 7.4) or hex obarbitone sodium (5-100 pi, 0.054AT in wa ter) to the sample cuvet and of an equal volume of the appropriate solvent to the re ference cuvet. Pyridine difference spectra were recorded with Na;S50.v-reduced micro somal suspensions (2.7 ml) and 0.3 ml of either 0.5M aqueous pyridine or water added to the cuvets. The wavelength of cytochrome P-450 (or P-448) maximum absorption was measured on a Cary 14 spectrophotometer calibrated with a holmium filter. Statistics Results are quoted as the mean S. E. M. and were analyzed by Student's i-test except where the nature of the results necessitated the use of a ranking test. Results and Discussion Effect of Dioxin Administration on the In Vivo Action, and In Vitro Metabolism of Drugs in the Rat One or 3 days after being given a single oral dose of dioxin rats show a shortening of the zoxazolamine paralysis time but an increase in the duration of action of hexo barbitone (2 ). The effect on the hexobarbi tone sleeping time becomes progressively more marked with time until, 2 weeks after dosing, rats sleep over 4 hr and some of them die without waking (Table 1). Table I. Effect of dioxin on the hexobarbitone sleeping'time of male ra ts/ Sleeping time, min 6 Time a f te r ------------------------------------------------------------------------ ------------- dosing Dioxin (200 Mg/kg, PO) Solvent 12 hr 24 hr 72 hr 1 week 2 weeks 32.9 3.5 40.0 * 3.1 75.5 5.9 125.6 * 14.1 > 244 (6) (6) (6) (6) (6) 37.7 3.3 (5) 27.4 1.8 (6) 33.6 3.3 (6) 21.5 2.9 (5) 25.2 2.7 (6) . Increase over controls, P % - NS 46 <0.01 125 <0.001 480 <0.001 870 0.0022 e *Hexobarbitone sodium administered as described in Materials and Methods section. bNumbers of animals in parentheses. e By ranking test. In view of the prolonged reduction of food intake of rats dosed with dioxin (3) it is possible that a starvation effect (13) might contribute significantly to the increased hexo barbitone sleeping times at later stages in the intoxication (1 or 2 weeks after dos ing). However earlier experiments (2) had suggested that, 1 day after dosing with di- 212 oxin, the reduced food intake was not the sole cause of the prolongation of the hexo barbitone sleeping time. This interpretation is open to criticism, since in these experi ments the animals had not been prefasted before dioxin administration and the dioxir might have slowed down the absorption oi the residual food in the stomach and intes Environmental Health Perspective GENP 011720 784531 ; Accordingly groups of eight female rats t*vife starved for 39 hr and then dosed with dioxin (200 .g/kg, PO) or oil. Following 24 hr further starvation the hexobarbitone sleeping time was measured; the value for the dosed group, 89,0 3.8 min, was signific antly higher than that of the control group, 63.2 5.3 min, (F <0.005). Thus the early ef fect of dioxin on the sleeping time cannnot be entirely due to differences in food con sumption or absorption between control and treated animals. Therefore in all follow ing experiments we have used rats which had been given a single dose of dioxin (200 ig/kg) 1 or 3 days previously. It was considered possible that dioxin might prolong the hexobarbitone sleeping time by altering either the sensitivity of the nervous system or the distribution of the barbiturate within the body. These possibili ties were ruled out by an experiment in which the sleeping times of groups of dioxintreated and control rats were measured and, immediately after their waking, blood was refected for the analysis of plasma barbi- te levels (8). Table 2 indicates that, ' -Jugh there was a significant increase in the sleeping time of the dosed animals, the waking plasma barbiturate levels of the two groups were not significantly different. Table 2. Effect of dioxin on the sleeping time and waking plasma barbiturate level of female rats.* Treatment N Plasma Sleeping time, barbiturate, min n g /m l Dioxin Controls P 6 99.3 19.8 67.9 2.2 6 45.9 4.2 64.5 2.9 <0.025 NS *Rats were dosed with dioxin (200 Mg/kg, PO) or oil and 3 days later the sleeping time in duced by hexobarbitone sodium (75 mg/kg, IP) was measured. Immediately after waking the animals were anesthetized (ether) and blood collected from the heart. Plasma bar biturate was measured by the method of Chromy and Babjuk (5). Three days after a single oral dose of either dioxin or oil to male or female rats the liver microsomes were isolated and in' ` ated with an NADPH-generating system and either hexobarbitone or zoxazolamine (Table 3). With*either sex, following dioxin treatment, there was a significant increase in the amount of zoxazolamine metabolized, and with males there was a significant de crease in the quantity of hexobarbitone oxi dized. With the microsomes from female rats the control level of hexobarbitone metabol ism was lower than the males and, although in the treated livers it was lower still, the difference was not significant. A part from this, the results agree with the in vivo ex periments and indicate that dioxin modi fies the duration of the pharmacological ac tion of both zoxazolamine and hexobarbi tone by changing the rate of metabolism of these drugs by the liver microsomes. Table 3, Effect of dioxin treatment on the metabolism of hexobarbitone and zoxazolamine by rat liver microsomes.* Treatment Sex Zoxazolamine metabolized, nmole/mg microsomal protein/hr6 Hexobarbitone metabolized, nmoles/mc microsomal protein/hra Dioxin Controls Dioxin Controls M 32.2 4.2 (10) 31.0 23.9.(5) M 15.7 * 2.0 (10) 192.0 18.7 (5) P <0.005 <0.001 F 46.9 * 7.9 (5) . 29.4 47.8 (5) F 6.3 i 5.4 (5) 109.2 49.7 (5) P <0.005 NS 'R ats (180-200 g) received dioxin (200 /ig/kg) or oil PO and were killed 3 days later. Micro somes were isolated by the calcium/sucrose method (see Materials and Methods) and the assays performed essentially as described else where (0-12). bNumbers of animals in parentheses. Effect of Dioxin Administration on the Liver Content and Spectral Properties of Micro somal Cytochrome P--450 Besides oxygen and NADPH, it is known that three other components are required for the reconstitution of a system capable of metabolizing drugs in vitro. These are a lipid, phosphatidyl choline; a flavoprotein, NADPH/cytochrome P-450 reductase, and a hemoprotein, cytochrome P-450 (or P448) (14-Iff). We have investigated the ef fect of dioxin treatment on cytochrome P450 of rat liver microsomes. Groups of male o^titember 1973 213 784532 Table 4. Changes in liver weight and microsomal protein and cytochrome P-450 induced by dioxin treatment of male rats.1 Treatment Dioxin Controls Weight wet liver. g/100 g BW 4.34 0.12 3.21 0.06 P <0.001 Microsomal protein, mg/g liver 26.5 0.9 27.3 0.6 NS Microsomal cytochrome P-450 nmoles/g liver nmoles/mg protein (nm) 55.1 1.5 28,1 0.7 <0.001 2.08 = 0.07 1.03 0.02 < 0.001 447.6 0.04 449.6 O.OS < 0.001 *Rats received dioxin (200 /xgAg. PO) or oil and 3 days later were killed and liver microsomes isolated at 105,000g (see Materials and Methods). Each value is the mean of five observations. rats received an oral dose of either dioxin or oil and three days later were killed and their liver microsomes isolated. Table 4 in dicates that there was a significant in crease in the weight of wet liver from the treated animals. There was no increase in the microsomal protein content, expressed per gram of tissue, but the cytochrome con tent doubled. Further, the wavelength of the peak maximum of the cytochrome spec trum was shifted from 450 nm to 448 nm. Such a shift is a feature of the microsomal enzyme induction brought about by com pounds such as 20-methylchoianthrene (17). Since this inducer stimulates the metabolism of aromatic compounds, including zoxazolamine (18) this change correlates well with the observed stimulation of zoxazolamine metabolism both in vivo and in vitro. How ever it should be remembered that, if the CO/cytochrome P-448 complex has an extinc tion coefficient more than four times that of P-450, as has been reported (19), then there has in fact been a decrease in hemo-protein concentration. The interaction of the reduced cytochrome P-450 of liver microsomes with the ligands ethyl isocyanide or pyridine is known to result in the formation of a difference spec trum with two peaks in the region 400-460 nm (20). The relative intensity of these two peaks, related to the absorption at 500 nm, is dependent on the pH of the suspend ing medium (2 0 ). In the case of liver microsomes prepared from animals pre treated with 20-mcthylchoIanthrene and therefore containing P-448 rather than P450, the curve of pH dependence of the peak height ratio is so shifted that the peaks are of equal intensity at a lower pH (21). The pH dependence of the pyridine difference spectra of microsomes from methylcholanthrene-, dioxin- or oil-treated rats was measured and is shown in Figure 1. It can be seen that the curves due to dioxin or methylcholanthrene treatment are similar to each other but distinct from that of the control. F ig u r e X. Rats (male, 180-200 g) received dioxin (200 iig/kg, PO) or Arachis oil (2 ral/kg, FO) 3 days before killing. 20-Methylcholanthrene was injected (20 mg/kg, 10 mg/ml in oil, IP) 3 and 2 days before killing. All were starved for 24 hr before decapitation. Portions of liver microsomes equivalent to 0.4 g of wet liver were isolated by the calcium/sucrose method (5) and suspended in 0.1M PO, buffer (10 ml, containing lmM Na,EDTA) of the appropriate pH. Pyridine differ ence spectra were measured as stated in Materials and Methods. Environmental Health Perspective GETST 011722 J-. is known that, in the oxidized state, ^trie cytochrome P-450 of liver microsomes will interact with various substrates of the drug metabolizing system to produce charactn'istic difference spectra {22). We have tudied the effect of dioxin pretreatment of nits on the interaction of aniline and hexoharbitone with ra t liver microsomes. With uiiline the control microsomes showed a normal Type II spectral change; this was intensified in the case of the microsomes from treated rats (Fig. 2 ). Such a change is consistent with the increased cytochrome P-448 in these preparations. However, a ouble reciprocal plot of the spectral change/ nmole of P-450 (P-448) against aniline con centration indicated that there was an in crease in K s (binding affinity constant) and in the maximal Type II spectral change (Fig. 3). A similar effect has been reported following methylcholanthrene induction {23). Figure 2. Rats (male, 180-200 g) received dioxin (200 /ig/kg) or oil PO 3 days before killing. They were starved for 24 hr before isolation of liver microsomes at 105,000g. Preparations from two animals were combined. Difference spectra (aniline 0.4l7iiAf final concentration) measured as de scribed in Materials and Methods on suspensions containing microsomes equivalent to 0.073g wet ^ ' ver/ml. dp tem ber 1973 F igure 3. Rats (male, 170-190 g) received dioxin (200 fig/kg) or oil PO 3 days before killing. Liver microsomes, equivalent to 2.8 g wet liver, were isolated by the calcium/sucrose method (5) and were suspended in 0.1M PO buffer (25 nil, pH 7.4) containing 1mM EDTA. Aniline concentra tions in the cuvet (see Materials and Methods) were varied from 2.0 to loJZmM. Each point is the mean of observations on five animals. On each axis the mean intercepts for groups of dosed and control animals were significantly different (P < 0.05). An extinction coefficient of 91/imVf-cm (7) was used in the estimation of cytochrome P--450 (-448). The spectral changes observed are related to the total cytochrome content of the cuvet. The interaction of hexobarbitone with mi crosomes from control animals produced the expected Type I difference spectra {22) as illustrated in Figure 4. However the micro somal preparations from rats which had re ceived dioxin 3 days previously consistently gave a difference spectrum with a peak at 412 nm and a trough at 380 nm. This type of spectrum has been termed a modified Type II spectral change {22) and has been ob served following methylcholanthrene pre treatm ent {24), but only with microsomal preparations from female rats. A decrease in the intensity of the Type I difference spec trum due to hexobarbitone has been reported for methylcholanthrene-treated male and female rats (23). 215 784534 Comparison of the Effects of 20-MethyIchol- anthrene and Dioxin on Zoxazolamine Action in Mice of the DBA/ 2 Strain F ig u r e 4. All details as in the legend to Figure 2 except that hexobarbitone was added to the sample cuvetts to a concentration of 0.79mili. All the changes of the properties of the liver microsomes from dioxin-treated rats are consistent with dioxin being a powerful inducer of the methylcholanthrene type. It has been reported th a t in mice of the DBA/2 strain the levels of hepatic aryl hy drocarbon hydroxylase are unaffected by methylcholanthrerie (25) or benz[a]anthra cene (26) treatment. Indeed when DBA/2 mice (male, 20-30 g) were injected with methylcholanthrene or oil as described by Nebert et al. (25) we found, 24 hr after injection, no significant difference between the zoxazolamine paralysis times of the treated and control groups (58.0 5.1 min, V = 6 and 59.3 3.9 min, N = 6, re spectively). This contrasts with C57BL/6 mice (male, 18-26 g) in which methylchol anthrene pretreatm ent significantly reduced the paralysis time (dosed: 15.0 1.0 min, IV = 9, controls : 79.5 5.3 min, N = 9 ; P < 0.0001). However in both strains a single oral dose of dioxin (200 /g/kg) given 3 days before zoxazolamine significantly lowered the paralysis time (DBA/2, male, dosed: 10.5 0.8 min, N = 6, controls: 65.7 4.2 min, N = 5, P < 0.0 0 1 ; for C57BL/6 see Table 5). n o rlHHO Table 5. Effect of dioxin on sleeping and paralysis times of C57 BL/S mice.* Treatment Sex Dioxin Oil M M Dioxin Oil F F After 3 days 52.6 6.2 (7) 71.3 3.8 (6) P <0.05 52.6 * 6.5 (7) 65.1 8.4 (7) P NS Sleeping time, m inb After 10 days 60.6 4.7 (7) 60.2 7.7 (4) NS -- -- After 20 days >159.7 (7) 53.4 7.3 (6) <0.01e -- -- Paralysis time after 3 days, min 7.0 0.7 (4) 46.8 i: 3.5 (5) <0.001 5.9 0.6 (5) 37.7 * 4.0 (7) <0.001 *Mice (11-27 g) received dioxin (200 Mg/kg) or oil PO and hexobarbitone sodium or zoxazolamine aiter the stated intervals. 6Numbers of animals in parentheses. ' By ranking test. -f 216 Environmental Health Perspectives 784535 Assuming that, under these conditions, the duration of zoxazolamine action is en tirely governed by the rate of metabolism in the liver it would appear that the appar ent genetic noninducibility of hepatic aryl iiydrocarbon hydroxylase in certain mouse trains is not an absolute tra it but depend ent on the inducer used. This is in agree ment with the observations (26) that this enzyme, although present in smaller amounts in extrahepatic tissues, is inducible in such organs of strains in which the liver enzyme is unaffected. investigation of the Relationship between Stimulation of Zoxazolamine Metabolism and Inhibition of Hexobarbitone Metabolism Although in the ra t dioxin produces sim ultaneous and divergent effects on the met abolism of zoxazolamine and hexobarbitone by liver microsomes, it is not known whether these two effects are related to each other. An alternative would be that dioxin has two separate and distinct effects: the induc tion of zoxazolamine hydroxylase and the depression of hexobarbitone oxidase. The following two experiments indicate that, after dioxin treatment, a stimulation of zo xazolamine metabolism can be observed in the absence of any inhibition of hexobarbi tone metabolism. In mice of the C57BL/6 strain, as in the rat, a marked reduction of the zoxazola mine paralysis time was observed at 3 days after dioxin treatment. However, at this time the hexobarbitone sleeping time was either unchanged (in female mice) or short ened (in male m ice); a prolongation was seen in males only at 20 days (Table 5). Table 6. Effect of ethionine on dioxin-induced alterations in sleeping time and paralysis time in rats.* Route of administration - Sex Dioxin Oil Ethionine M IP _ -- -- IP -- M IP -- PO -- IP PO M PO -- IP --. PO IP P IP -- PO -- IP PO Sleeping time, min " 75.4 13.9 (5)e 36.0 2.2 (5) 41.3 3.6 (5) 45.0 3.1 (5) 44.7 3.1 (6) 42.0 3.4 (6) 83.0 9.4 (6 )r 136.4 7-2 (6) Paralysis time, min0 -- -- 84.0 6.5 (S)"** >383 (5) -- -- 119.2 9.4 ( 6 ) - ' >480 (6) * Animals received di-ethione (200, 100, and 100 mg/kg) at t = 0, 4, and 8 hr, respectively; dioxin (200 jtg/kg) or oil at = 0.5 hr and sleeping or paralysis times were measured at t = 24.5 hr. All were starved during the experiment. b Numbers of animals in parentheses. * Significantly different from controls at P <0.05. 11Significantly different from controls at P <0.01. *By ranking test. ' Significantly different from controls at P <0.005. The effect of dioxin on hexobarbitone met abolism can also be suppressed in rats by administering dZ-ethionine together with the dioxin. Along with a single oral dose of dioxin male rats were given a series of in jections of dZ-ethionine, an inhibitor of pro' synthesis which has been reported as .enting the induction of drug metabol- September 1973 Ising enzymes by either 20-methylcholanthrene (27) or phenobarbitone (2S). It was found (Table 6) that, whereas the ethionine was ineffective in preventing the reduction of paralysis time in dioxin-treated animals, it did abolish the prolongation of hexobar bitone sleeping time due to dioxin treat ment. When the routes of administration 217 784536 of dioxin and ethionine were interchanged there waa still no difference in hexobarbitone metabolism. When the experiment was carried out with female rats, in which ethionine is a more effective inhibitor of protein synthesis (29), the duration of both paralysis and sleep of the dioxin-treated animals was shorter than that of the controls (Table 6 ). Thus although these doses of ethionine are not effective in blocking the induced metabolism of zoxazolamine (27) in either sex, they are capable of preventing or even reversing the effect of dioxin on hexobarbitone metabolism. C onclusions 1. The divergent effects of dioxin on the duration of action of hexobarbitone and zo xazolamine in the rat in vivo (2) are a consequence of changes in hepatic metabol ism of these drhgs. 2. Dioxin causes alterations in the proper ties of cytochrome P-450 of ra t liver microsomes which are similar to those produced by methylcholanthrene. 3. It is the most effective stimulator of aromatic hydroxylation known, see (X), and can apparently overcome a genetic resistance to hepatic microsomal enzyme induction in mice of the DBA/2 strain, 4. The effects of dioxin on hexobarbitone and zoxazolamine metabolism can be separ ated by the use of C57BL/6 mice or ethionine-treated rats and might be due to two separate modes of action. Acknowledgement We thank McNeil Laboratories, Inc., Camp Hill Road, F o rt Washington, Pa. and May & Baker Ltd., Dagenham, Essex for gifts of zoxazolamine and hexobarbitone sodium, re spectively, and Mr. C. M. Puah and Mr. B. S. Sood for technical assistance. REFERENCES 1. Buu-Hoi, N. P., et al. Froprietes canceromimetiques de la tetrachloro-2,3,7,S-dibenzo-p-dioxine ("dioxine"). Compt. Rend., Ser. D 272: 1447 (1971). 2. Greig, J. B. Effect of 2,3,7,3-tetrachlorodibenzoI, 4-dioxin on drug metabolism in the rat. Biochem. Pharmacol. 21: 3196 (1972). 3. Greig, J. B., et al. Toxic effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Food Cosmet. Toxicol 11: 585 (1973). 4. Bond, E. J., and De Matteia, F. Biochemical changes in ra t liver after administration of car bon dianlphide, with particular reference to microsomal changes, Biochem. Pharmacol. IS: 2531 (1969). 5. Hamath, S. A., and Rubin, E. Interaction of cal cium with microsomes: a modified method for the rapid isolation of rat liver microsomes. Bio chem. Biophys. Res. Commun. 49: 52 (1972). 6. Aldridge, W. N. Adenosine triphosphatase in the microsomal fraction from ra t brain. Biochem. J. 83: 527 (1962). 7. Omura, T., and Sato, R. Fractional solubiliza tion of haemoproteins and partial purification of carbon monoxide-binding cytochrome from liver, microsomes. Biochem. Biophys. Acta. 71: 224 (1963). 8. Chromy, V., and Babjuk, J. S. Determination of barbiturates in biological fluids. Clin. Chim. Acta 37: 547 (1972). 9. Cooper, J. R., and Brodie, B. B. The enzymatic metabolism of hexobarbital (Evipal). J. Phar macol. Exp. Therap. 114: 409 (1955). 10. Brodie, B. B. et al. The fate of pentobarbital in man and dog and a method for its estimation in biological material. J. Pharmacol. Exp. Therap. 109: 26 (1953). 11. Burns, J. J. et al. Zoxazolamine. Physiological disposition, uricosuric properties. Amer. J. Med. 25: 401 (1958). 12. Conney, A. H., Trousof, N., and Burns,. J. J. The metabolic fate of zoxazolamine (Flexin) in man. J. Pharmacol. Exp, Therap. 128: 333 (1960). 13. Dixon, R. L., Shultice, R. W., and Fouts, J. R. Factors affecting drug metabolism by liver mi crosomes. IV. Starvation. Proc. Soc. Exp. Biol. Med. 103: 333 (1960). 14. Lu, A. Y. H., and Coon, M. J. Role of hemoprotein P-450 in fatty acid -hydoxylation in a soluble enzyme system from liver microsomes. J. Biol. Chem. 243: 1331 (1968). 15. Lu, A. Y. H., Junk, K. W., and Coon, M J* Resolution of the cytochrome P-450-containing -hydroxylation system of liver microsomes into three components. J. Biol. Chem. 244: 3714 (1969). 16. Lu, A. Y. H., et al. Reconstituted liver micro somal enzyme system that hydroxylates drugs, other foreign compounds, and endogenous sub strates. IV. Hydroxylation of aniline. Arch. Biochem. Biophys. 153: 294 (1972). 17. Alvarea, A. P., et al. Studies on the induction of CO-binding pigments in liver microsomes by 218 Environmental Health Perspectives ' 784537 GENP 011726 V J .enobarbital and-3-methyicholanthrene. Bio- ~~ chem. Biophys. Res. Commun. 29: 521 (1967). L8. Conney, A. H., et al. Adaptive increases in drug- metabolising enzymes induced by phnobarbital and other drugs. J. Pharmacol. Exp. Therap. 130: 1 (I960). 19. Hildebrandt, A., Remmer, H.f and Estabrook, R. W. Cytochrome P-450 of liver microsomes-- one pigment or many. Biochem. Biophys. Res. Commun. 30: 607 (1968). 20. Imai, Y., and Sato, R. Anomalous spectral in teractions of reduced P-450 with ethyl isocya nide and some other lipophilic ligands. J. Bio chem. (Tokyo) 62: 464 (1967). 21. Sladek, N. E., and Mannering, G. J. Evidence for a new P-450 hemoprotein in hepatic micro somes from methylcholanthrene treated rats. Biochem. Biophys. Res. Commun. 24: 668 (1966). 22. Schenkman, J. B.( Remmer, H.t and Estabrook, R. W. Spectral studies of drug interaction with hepatic microsomal cytochrome. Mol. Pharmacol. 3: 113 (1967). 23. Kato, R., Takanaka, A., and Takayanaghi, M. Substrate-induced spectral change of liver mi crosomes in phnobarbital and methylcholan threne-treated male and female rats. J.'Biochem. (Tokyo) 68: 395 (1970). 24. Schenkman fcJ. B,, et al. On the problem of possi ble other forms of cytochrome Poo in liver mi crosomes. Biochem. Biophys. Acta 171: 23 (1969). 25. Nebert, D. W., Goujon, F. M. and Gielen, J. E. Aryl hydrocarbon hydroxylase induction by polycyclic hydrocarbons: simple autosomal domi nant trait in the mouse. Nature New Biol. 236: 107 (1972). 26. Wiebel, F. J.f Leutz, J. C., and Gelboin, H. V, Aryl hydrocarbon (benzo[a]pyrene) hydroxy lase: inducible in extrahepatic tissues of mouse strains not inducible in liver. Arch. Biochem. Biophys. 154: 292 (1973). 27. Conney, A. H., Miller, E. C., and Miller, J. A. The metabolism of methylated aminoazodyes. V. Evidence for induction of enzyme synthesis in the rat by 3-methylcholanthrene. Cancer Res. 16: 450 (1956). ` 28. Kato, R., Chiesara. E., and Vassanelli, P. Factors influencing induction of hepatic microsomal drug-metabolising enzymes. Biochem. Phar macol. 11: 211 (1962). 29. Farber, E. Etbionine carcinogenesis. Advan. Cancer Res. 7: 383 (1963). GENP oi 1727 y September 1973 219 784538 Studies of the Effects of 2,3,7,8-Tetrachlorodibenzo-p-dioxin on Mammalian Hepatic s-Aminolevulinic Acid Synthetase by James S. Woods' Introduction The toxicity of 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD), a contaminant formed ng the manufacture of the herbicide - i-trichlorophenoxyacetic acid (2,4,5-T) is well known (1-U). The widespread utiliza tion of this compound has caused increased concern about the potential health hazards created by the presence of TCDD in the en vironment. The articles which accompany this account, indeed, attest to the highly toxic nature of TCDD in both laboratory animals and man. Interest in TCDD as a potential porphyrogenic agent arose when porphyria cutanea tarda, a form of hepatic porphyria, occurred in industrial workers ' associated with the manufacture of 2,4,5-T (5). -Hepatic porphy ria is a syndrome characterized by a variety of symptoms including the overproduction and excretion of porphyrins, pigmentation of the skin, photosensitivity, and intestinal and neurological disorders. The disease is characterized biochemically by an increase in the activity of the mitochondrial enzyme Pathologic Physiology Branch, National Institute of Environmental Health Sciences, National Insti tutes of Health, P.O. Box 12233, Research Triangle North Carolina 27709. September 1973 S-aminoIevulinic acid (ALA) synthetase, which is the first and rate-limiting enzyme in the heme biosynthetic pathway {6) (Fig. 1). The possibility that TCDD has porphyrogenic properties has been indicated by the recent observation that TCDD is a potent inducer of hepatic ALA synthetase in chick embryos (7). There is, however, no evidence to indicate that TCDD produces similar ef fects in mammalian species. The chick em bryo system is exquisitely sensitive to the effects of agents which induce ALA synthe tase (5). Previous studies from our labora tory (9, 10), on the other hand, have shown that mammalian species demonstrate a striking variability in their response to porphyric agents, especially at different stages of development. These studies were undertaken, therefore, to determine the possible porphyrogenicity cf TCDD in mammalian species and to as sess fu rth er the utility of the chick embryo liver system as an indicator of the potential porphyrogenic effects of environmental agents in mammals. Materials and Methods ALA synthetase activity was determined by two procedures. In liver homogenates 221 784539 GENP 011728 Glycine + Succinyl-CoA ALA S y n th etase 6-A m inolevulinic Acid (ALA) ALA D ehydratase Porphobilinogen URO S y n th etase v Uroporphyrinogen (URO) i Coproporphyrinogen i Protoporphyrin \ Heme Heme S y n th etase r ++ F ig u r e 1. Heme biosynthetic pathway. ALA synthetase activity was assayed using the ion exchange chromatography technique described by Mauzerall and Granick {11), with liver homogenates prepared for incuba tion as described by Marver et al. (12). ALA synthetase activity in subcellular frac tions and at various stages of enzyme puri fication was measured using the procedure described by Scholnick et al. (13). In the latter case the incubation medium was modi fied to include lO-tftf GTP in addition to the prescribed substrates. The method of Scholnick was also used for the isolation and 50-fold purification of ALA synthetase from porphyric rat liver. Hepatic subcellular fractions were prepared as described by Hayaski et al. (1-4). Protein concentrations were assayed by the method of Lowry etal. (15). All test animals were treated orally with a standard solution of TCDD (10 ^g/ml) dissolved in a com oil/acetone mixture (6 :1) 24 h r prior to sacrifice unless otherwise in dicated. Two groups of control animals were used in all in vivo experiments. The first group received only the corn oil/acetone mix ture in an amount equivalent to that in which TCDD was administered to test ani mals. The second group was treated subcu taneously with allylisopropylacetamide (ALA) (400 mgA&), which is a well known and potent inducer of ALA synthetase in mammals (12, 16). TCDD was obtained from Dow Chemical Company, Midland, Michigan. AIA was a gift from Hoffmann-LaRoche, Nutley, New Jersey. All animals and other chemicals were obtained from standard sources. Results and Discussion Initial studies were designed to determine the potential porphyrogenic effects in rats of TCDD when administered in doses up to 25 ig/kg, the reported LD5(J for this species (17). Male rats were treated with a single 5 or 25 /ig/kg dose of TCDD, and hepatic ALA synthetase activity was assayed at pe riods up to 28 days thereafter. In animals receiving a single dose of AIA, ALA syn thetase activity increased to approximately seven times the control level after 24 hr and returned to control levels by the third day after treatment. On the other hand, TCDD did not significantly alter ALA syn thetase activity, as measured in whole liver homogenates, during any part of the test period in any of the animals. All measure ments of enzyme activity were within the range observed in controls. It has been recently determined (14) that ALA synthetase is a mitochondrial enzyme but is synthesized extramitochondrially on the cytoplasmic ribosomes. The enzyme is subsequently incorporated into the mito chondria, where it becomes active. ALA synthetase activity may be altered by agents which interfer with any aspect of this pro cess. It was, therefore, of interest to deter mine if TCDD might influence the subcellu lar localization of ALA synthetase and thereby alter the regulation of hepatic hem- 222 Environmental Health Perspectives . 784540 GENP 011729 Table 1. Subcellular distribution of hepatic ALA synthetase in norm al and AIA- and TC D D -treated rats.* V. ALA, nmole/mg protein-hr S.E. Group 1 2 3 4 Treatment Corn Oil AIA TCDD AIA + TCDD Mitochondrial fraction 0.52 0.09 2.26 0.80 0.58 0.10 2.91 0.87 9,0000 supernatant 0.27 s: 0.08 1.44 0.40 0.34 0.06 1.37 0.51 105,0000 supernatant 0.24 * 0.10 1.47 0.30 0.15 0.43 1.17 0.51 Microsomal fraction 0.03 0.01 0.28 0.10 0.01 s 0.01 0.27 0.08 Rats were treated with TCDD (25 Mg/kff) and/or AIA (400 mg/kg) 24 hr prior to sacrifice. atopoiesis in a manner which could not be detected when ALA synthetase activity was measured in whole liver homogenates. The results of these experiments are seen in Ta ble 1. Analysis of the subcellular distribu tion of ALA synthetase in adult rat liver reveals a 2:1 distribution of activity between mitochondrial and postmitochondrial frac tions. Most of the postmitochondrial activity is retained in the 105,000fir supernatant fraction with very little found in the microsomes. Control rats in the group treated AIA showed substantial increases in ILA synthetase activity in all fractions iiu;' relatively little alteration in the distri bution of activity between fractions after 24 hours. TCDD administered alone at 25 ng/k g doses or together with AIA caused no significant alteration of ALA synthetase activity from that observed in control groups. In addition, TCDD did not affect the induction of ALA synthetase by AIA nor alter the subcellular distribution of the enzyme during induction. Similar observa tions were made when TCDD was adminis tered in 100 m g/kg doses or when these ex periments were conducted with female rats or mice. In vitro tests of the possible effects of TCDD or ALA synthetase activity were conducted in liver homogenates, isolated mi tochondria and on ALA synthetase purified 50-fold from porphyric rat liver. In enzyme incubation mixtures containing TCDD in concentrations ranging from 10-9 to 10-Hf no discernable effects on the enzyme activity could be observed. Finally, the potential porphyrogenic ef fects of TCDD during the perinatal period were investigated. These studies were con ducted in fetal rats which were 3 days from delivery, ( --3 day), and on two groups of newborn rats, 4 and 12 days after delivery on the day of sacrifice. Newborn rats were treated orally with 25 ig/kg doses of TCDD, whereas fetal rats were treated by way of the mother. Livers from mother rats served as adult samples. The results are shown in Table 2. In no case did ALA synthetase ac tivity in TCDD-treated rats differ from that observed in untreated animals. Fetal ALA synthetase is typically five to eight times that of the adult (5) and declines to adult levels shortly after birth. Refractoriness to induction of ALA synthetase is observed un til the activity approaches that of the adult {10', 18). At no stage of development, how ever, did the ALA synthetase activity in TCDD-treated rats significantly differ from that observed in controls. These results are of particular interest in view of the potent induction of ALA syn thetase activity produced by TCDD in the chick embryos. The lack of a significant ef fect of TCDD on ALA synthetase in mam mals suggests that major differences exist among these species in the biological mech anisms which determine the ultimate phar macological disposition of chemicals such of TCDD. Species variations in drug distribu tion, biotransformation, and excretion are well known (15). On the other hand, m ajor differences in the mode of action of TCDD may also reflect alterations in the regulation tember 1973 223 784541 T able 2. S tudy o f p o ten tial effects o f TCDD on A LA sy n th e ta se d u rin g developm ent in rats.* ' " ALA, nmole/hr-g liver S.E. Group 1 2 3 Treatment Com oil TCDD AIA --3 day 241 31 253 47 248 28 + 4 day 94 7 76 21 118 12 +12 day 50 1 6 60 8 113 13 Adult ~ 45 5 44 13 347 17 ' 1Pregnant and newborn rata were treated 24 h r prior to sacrifice with TCDD (25 *g/kg) or AIA (00 mg/kg). of hepatic heme synthesis in these species. Striking variations in the developmental aspects of ALA synthetase regulation have already been described. The chick embryo liver is one of the most sensitive systems available in which to study the induction of ALA synthetase (8). In contrast, our previ ous studies (10) have shown that fetal mammals are totally insensitive to the ef fects of drugs which induce or repress hepatic ALA synthetase in chick embryo or in adult animals. Moreover, fetal mamma lian ALA synthetase activity is significantly elevated in comparison with th at of the adult (9), whereas the control level of ALA synthetase in 17-day old chick embryo liver is only one-third that measured in the adult chicken (20). Differences in porphyrin metabolism in these two species is also suggested by the failure of a wide variety of drugs which in duce hepatic porphyrin accumulation in chick embryos to do so in mammalian liver (21). Among the drugs capable of causing significant increases in porphyrin levels in chick embryo liver are 3,5-dicarbethoxy-l,4dihydrocollidine (DDC), glutethimide, methsuximide, secobarbital, methylprylon, and mephenytoin. Of this group, only DDC is capable of producing porphyrin accumula tion in mice. In addition, it has been recently indicated that uroporphyrinogen synthe tase, another enzyme in the heme biosyn thetic pathway, may play a rate-limiting role in hepatic heme synthesis in certain strains of mice (22). It therefore appears that significant dif ferences exist among these species with re gard to various steps in the regulation of hepatic heme synthesis and porphyrin meta- bolism. These differences, along with varia tions in the capacity to metabolize and dis tribute drugs such as TCDD, may account for the differences in susceptibility of these species to the porphyrogenic effects of TCDD and perhaps to other environmental contaminants. The foregoing considerations attest to the increasingly prevalent observation that many environmental agents are hazardous not only by virtue of their inherent toxicity but also because of their specificity of action in different species. While utilization of nonmammaiian test systems may provide, in some cases, a sensitive indication of the po tentially toxic effects of certain drugs and chemicals in mammals, this study demon strates the necessity for developing test pro cedures which will more clearly predict the deleterious effects of environmental con taminants in mammalian species, especially in man. REFERENCES 1. Courtney, K. D., and Moore, J. A. Teratology studies with 2,4,5-trichlorophenoxyacetic add and 2,3,7,8-tetrachlorodibenzo-p-dloxin. Toxicol. Appl. Pharmacol. 20: 396 (1971). 2. Buu-Hoi, N. P., et al. Organs as targets of "dioxin" (2,3,7,8-tetrachlorodibenzo-p-dioxin). Naturwiss. 59: 174 (1972). 3. Fishbein, L., and Flamm, W. G. Potential en vironmental chemical hazards. Part I: Drugs. Sci. Total Environ. 1: 15 (1972). 4. Sparschu, G. L., Dunn, F. L., and Rowe, V. K. Study of the teratogenicity of 2,3,7,8-tetrachIorodibenzo-p-dioxin in the rat. Food Cosmet. Toxi col. 9: 405 (1971). 5. Poland, A. P., et al. A health survey of workers in a 2,4-D and 2,4,5-T plant. Arch. Environ. Health 22: 361 (1971). 224 Environmental Health Perspectives 784542 GENP 011731 6. Granick, S., and Urata, J. Increase in the ac tivity of -aminolevulinic acid synthetase in liver mitochondria induced by feeding 3,5-dicarboxyethyl-l,4-dihydrocollidine. J. Biol. Chem. 238: 821 (1963). 7. Poland, A., and Glover, E. 2,3,7,8-Tetrachlorodibenzo-p-dioxin: A potent inducer of J-aminolevulinic acid synthetase. Science 179: 476 (1973). 8. Granick, S. The induction in vitro of the syn thesis of -aminolevulinic acid synthetase in chemical porphyria. J. BioL Chem. 241: 1356 (1966). 9. Woods, J. S., and Dixon, R. L. Perinatal differ ences in delta-aminolevulinic acid synthetase activity. Life Sci. 9: 711 (1970). 10. Woods, J. S., and Dixon, R. L. Studies of the perinatal differences in the activity of hepatic -aminolevulinic acid synthetase. Biochem. Phar macol. 21: 1735 (1972). 11. Mauzerall, D., and Granick, S. The occurrence and determination of -aminolevulinic acid and prophobilinogen in urine. J. Biol. Chem. 219: 435 (1956). 12. Marver, H. S., et al. -Aminolevulinic acid syn thetase. I. Studies in liver homogenates. J. Biol. Chem. 241: 2803 (1966). 13. Scholnick, P. L., Hammaker, L. E. and Marver, H. S. Soluble -aminolevulinic acid synthetase of ra t liver. I. Some properties of the partially purified enzyme. J. Biol. Chem. 247: 4126 (1972). A. Hayashi, N., Kurashima, Y., and Kikuchi, G. Mechanisms of allylisopropylacetamide-.induced increase of -aminolevulineate synthetase in liver mitochondria. Arch. Biochem. Biophys. 148: 10 (1972) . 15. Lowry, 0. H., et al. Protein measurements with the Folin Phenol reagent. J. Biol. Chem. 193: 265 (1951). 16. Song, C. S., Lee, W.f and Kappas, A. 3-Aminolevulinate synthetase and drug-induced disease in microsomal cytochrome P-450 of the liver. Clin. Res. 18: 389 (1970). 17. Schwetz, B., et al. Chlorodibenzo-p-dioxin toxi cology. Environ. Health Perspect. No. 5: 87 (1973) . 18. Song, C. S., et al. The influence of postnatal development on drug-induced hepatic porphyria and the synthesis of cytochrome P-450. J. Exptl. Med. 134: 1349 (1971). 19. Goldstein, A., Aronow, L.( and Kalman, S. M. Drug toxicity in lower animals and man. In: Principles of Drug Action, Harper and Row, New York, 1968. Chapt. 5. 20. Creighton, J. M., and Marks, G. S. Drug-induced porphyrin biosynthesis VTI. Species, sex, and developmental differences in the generation of experimental porphria. Can. J. Physiol. Phar macol. 50: 485 (1972). 21. Racz, W. J., and Marks, G. S. Drug-induced porphyrin biosynthesis II. Simple procedure for screening drugs for pophyria-inducing activity. Biochem. Pharmacol. 18: 2009 (1969). 22. Hutton, J. J., and Gross, S. R. Chemical induc tion of hepatic porphyria in inbred strains of mice. Arch. Biochem. Biophys. 141: 284 (1970). GENP 011732 September 1973 1 225 784543 Effect of 2,3,7,8-Tetrachlorodibenzo-p-dioxin on the Biliary Excretion of Indocyanine Green in Rat by Shang W. Hwang* Chlorinated dibenzodioxins have been found as contaminants of various technical chlorinated compounds such as 2,4,5-trichlorophenoxyacetic acid (2,4,5-T), which are widely used in agriculture. The dioxin contaminants may be involved in various pathologic states resulting from exposure to "clinical chlorinated compounds (I, 2). Liver crosis has been observed in the animals .vhich were treated with derivatives of chlorophenol (3) or other chlorinated com pounds (-4), and it was suggested that liver necrosis-causing factors were present in these compounds. In view of the above ob servation, the present study was undertaken to investigate whether 2,3,7,8-tetrachloro-pdibenzodioxin (TCDD) has any effect on the hepatobiliary function of rat. Biliary excre tion of indocyanine green (ICG) was used as the index of function. The dye was chosen because it is completely and rapidly ex creted by normal liver into the bile by an active process (5). Materials and Methods Male random bred CD rats weighing 350400 g were treated PO with a single dose of 25 fig/k g or 5 *g/kg TCDD in acetone and "Pharmacology and Toxicology Branch, National Institute of Environmental Health Sciences, Na tional Institutes of Health, P.O. Box 12233, Research Triangle Park, N.C. 27709. corn oil. The controls received an equivalent volume of the vehicle. At day I, 7, and 16 after treatment, the effect of TCDD on the bile flow and the biliary excretion of ICG was examined. Animals were first anesthe tized with penobarbital Na (50 m g/kg) IP. Through an abdominal incision, renal pedi cles were ligated, and the common bile duct was cannulated with a blunt 23-gauge hypo dermic needle shaft attached to an 8-in. piece of PE 50 tubing. Bile was collected for a 20-min period, and the amount collected was measured by weighing. At the end of 20 min, freshly prepared indocyanine green in aqueous solvent (Hynson, Westcott and Dun ning, Inc.) was injected at a dose of 6.25 mg/ kg into the femoral vein, and the bile was collected for another 20 min. The body tem perature was monitored with a telether mometer and was maintained at 37 C by warming with an incandescent lamp. At the end of the experiment, blood was withdrawn by heart puncture and the liver was excised. The ICG concentration in bile, plasma, and liver was determined by measuring its ab sorption at 805 nm. The rate of ICG disap pearance from plasma was also determined by measuring the dye concentrations in a series of plasma samples which were with drawn by heart puncture 1, 2, 5, 8, 12, and 20 min after ICG injection from groups of control and TCDD-treated rats. iptember 1973 227 784544 Results and Discussion Bile flow during the first 20 min of the experiment increased after TCDD treatment as shown in Figure 1. Th initial flow rate continued to increase through the 16th day, and rats receiving 25 ^g/kg TCDD had higher flow rates than those receiving 5 nS/k g TCDD. decreased significantly as shown in Figure 3. Less hepatic excretion of the ICG was ob served with the larger dose of TCDD than with the smaller dose. The rate of excretion of this dye was still markedly suppressed by the 16th day after treatment. F ig u r e 1. Effect of TCDD on bile flow. Each point ia the mean A SEM of four animals. Difference from control is significant (P <0.005) for the7th and 16th day after 25 Mg/kg TCDD treatment and 16th day after 5 Mg/kg TCDD treatment. F igure 2. Liver weight of ra t after TCDD treat ment. Each point is the mean SEM of four animals. The difference is significant (P<0.05) for the 7th and 16th day after treatment of either 5 mg/kg or 25 Mg/kg TCDD. The increase in bile flow could be due to the increasing secretion of water by the hepatic cell or due to the decreasing reab sorption of w ater along the bile duct. Both mechanisms were possible, but further in vestigation will be needed for clarification. TCDD also caused an increase in liver weight as shown in Figure 2. The increase in liver weight as expressed in grams liver weight per 100 g body weight was also doserelated since the higher dose caused a great er weight increase. Whether this increase in liver weight has any effect on the bile flow is not known. Similar results of bile flow increase and liver weight increase were observed in ra t after phnobarbital treatm ent (d), and both TCDD and phno barbital are potent microsomal enzyme in ducers. ICG excretion was also affected by TCDD treatment. The total amount of ICG excreted during 20 min after injection of the dye was 228 F igure 3. Effect of TCDD on ICG biliary excretion. The dose of ICG was 6.25 mg/kg, and the bile was collected for 20 min after ICG injection. Each point is the mean s SEM of four animals. The difference from control is significant (P <0.05) fo r the 7th and 16th day after treatm ent by either 5 Mg/kg or 25 Mg/kg of TCDD. Environmental Health Perspectives 784545 } T ab je 1. E ffect o f TCDD on liv e r u p ta k e and biliary excretion o f ICG. Time after TCDD treatment, days Control 1 day 7 days 16 days ICG concentration in plasma, fig/m l1 TCDD 5 pg/kg 1.0 0.09 2.3 0.10 3.0 0.23 3.1 0.46 TCDD 25 pg/kg 1.9 0.09 2.75 0.29 3.45 0.15 3.90 0.58 ICG concentration in liver, fig/gm TCDD 5 pg/kg TCDD 25 pg/kg 28.5 0.9 29.5 2.3 37.5 3.6 41.0 5.4 28.5 0.9 29.0 1.8 31.0 2.1 35.0 3.6 ICG concentration in bile, Mg/ml * TCDD 5 pg/kg 1130 83 1015 79 869 66 860 59 TCDD 25 pg/kg 1130 83 999 64 710 45 624 46 *Dose of ICG was 6.25 mg/kg. Concentrations were determined 20 min after ICG injection. Each value is the mean SEM from four animals. The concentrations of ICG in plasma, liver, and bile were also analyzed separately for each animal 20 min after dye injection. The results (Table 1) showed that concentration of ICG in bile of TCDD-treated ra t was lower than that of the control, and the ICG levels in plasma and liver of TCDD-treated rats were higher than that of the control. A greater depression of ICG concentration in bile and a greater retention of IQG in blood were caused by 25 fig/k g dose of TCDD. In contrast, animals treated with 5 ng/kg TCDD accumulated more ICG in liver than e animals treated with 25 pg/kg TCDD. / ) sign of recovery was shown by the 16th day after treatment. Bile-to-plasma, bile-to-liver and liver-toplasma concentration ratios of ICG were also separately calculated and compared, as shown in Figure 4. Bile-to-plasma and bileto-liver ratios decreased after TCDD treat ment, and the higher dose decreased these ratios even further. Liver-to-plasma ratio decreased only after 25 pg/kg but not after 5 kg TCDD pretreatm ent. TCDD ap peared to inhibit both ICG uptake by the hepatic cell and the active secretion of ICG by the hepatic cell. Inhibition of both steps could result in the decreased total ICG ex cretion and lowered bile-to-plasma and bileto-liver ICG concentration ratios as found. The excretion of ICG from hepatic cell to bile was probably inhibited to a similar ex tent by both 25 pg/kg and 5 pg/kg TCDD, However, the inhibition of ICG uptake by the hepatic cell might be more dose-depend ent; the inhibition was greater with 25 pg/kg than with 5 p g/kg TCDD, so the liver cumulated less ICG after 25 pg/kg TCDD F ig u r e 4. Change of ICG concentration ratios after TCDD treatment. Each bar represents the mean of four animals. The difference from control is significant (P <0.05) for all ratios a t 7th and I6th day after TCDD treatment except the liver/ plasma (L/P) ratios after 5 tig/kg TCDD treat ment. treatment, and liver-to-plasma ICG concen tration ratios decreased in animals treated with 25 pg/kg but not 5 pg/kg TCDD. Con trol animals should take up ICG into liver faster than TCDD-treated animals, but the secretion from liver to bile was even faster compared to the treated animals, so the con trol animals accumulated less ICG in. both plasma and liver. oeptember 1973 229 784546 The rate of disappearance of ICG in plasma decreased after TCDD treatment-, as shown in Figure 5. The 25 /ig/kg dose gave a greater reduction of disappearance rate than did the 5 ig/kg dose, and the rate was lower at the 16th day than at the 7th day after, treatm en t The decreasing rate of ICG dis appearance in plasma gave further evi dence th at ICG excretion was damaged by' the TCDD treatm ent. F ig u r e 5. Effect of TCDD on the ICG disappearance rate in plasma. Each point is the mean of three rats. Summary From the above observations, it was con cluded that TCDD inhibited hepatobiliary excretion of ICG, and the inhibitory effect appeared to be a long-lasting one. Many anionic compounds, including endogenous substances such as bilirubin, are actively se 230 creted through the sim ilar mechanism by the hepatobiliary system. The decreased ICG excretory ability might also apply to other anionic compounds, and the etiology of re ported cases of jaundice and porphyria after exposure to TCDD might be partly ac counted for by the decreased biliary excre tory ability. Environmental Health Perspectives 784547 GENP 011736 ;know ledgem en i . The author is indebted to Dr. John A. Moore, Chief, Animal Science and Technol ogy Branch, NIEHS, for supplying the con trol and TCDD-treated rats. The author also wishes to acknowledge the assistance of Dr.James R. Fouts and Dr. Larry G. H art in the preparation of the manuscript. REFERENCES X. Higginbotham, G. R., et al. Chemical and tox icological evaluation of isolated and synthetic chloro derivatives of dibenzo-p-dioxin. Nature 220: 702 (1969). 2. Kimbrough, R. D. Toxicity of chlorinated hydro carbons and related compounds. Arch. Environ. Health; 25:-125 (1972). Bauer, H.t Schulz, K. H., and Spiegelberg, U. Berufliche Vergiftungen bei der Herstellung von Chlophenol-Verbindungen. Arch. Gewerbepathol. Gewerbehyg. 18: 538 (1961). 4. Vos, J. G., and Koeman, J. H. Comparative tox icologic study with polychlorinated biphenyls in chicks with special reference to porphyria, edema formation, liver necrosis and tissue resi dues. Toxicol. Appl Pharmacol. 17: 656 (1970). 5. Cherrick, G. R., et al. Indocyanine green: ob servations on it3 physical properties, plasma decay and hepatic excretion. J. Clin. Invest. 39: 592 (1960). 6. Klaassen, C. D. Biliary flow after microsomal enzyme induction. J. Pharmacol. Exptl. Therap. 168: 218 (1969). r ~\ GENP 011737 September 1973 231 784548 i Effects of 2,3,7,8-Tetrachlorodibenzop-dioxin on Drug Metabolism and Hepatic Microsomes of Rats and Mice by l.B. Greig* and F. De Matteis* Introduction It has been reported (1) that one effect of a single, intra-peritoneal dos,e of 2,3,7,8tetrachlorodibenzo-p-dioxin (dioxin) is to cause a reduction in the duration of action of the drug 2-amino-o-chIorobenzoxazole (zoxazolamine) in the rat. This effect is maximal with doses of dioxin above 100 f i g / kg and significant even at 5 fig/kg. Our own work showed that oral dosage had a similar effect but also that the duration of action of hexobarbitone was considerably pro longed (). The results described here ex tend these observations. Materials and Methods Animals Eats of the albino Porton strain bred in these Laboratories and weighing 180-200 g (male 6-8 weeks and female 7-10 weeks old) had free access to water and diet 41B un less otherwise indicated. Mice of the C57BL/6 and DBA/2 strains were bred in these laboratories and kept on Sterolit bedding (Mineral & Chemical Cor poration of America, Menlo Park, New Jer sey, U.S.A.) for a t least 3 weeks before use. They had free access to diet and water. *MBC Toxicology Unit, Medical Research Council Laboratories, Woodmansterne Road, Carahalton, Surrey, England. September 1973 Dosages Dioxin, prepared as described elsewhere (3), was administered as a solution (100 fig/ ml) in Arachis oil. Control animals received an equivalent volume of oil. Zoxazolamine (McNeil Laboratories, Inc.) was dissolved in IN HC1 (1.2 ml/100 mg), diluted with 0.9% NaCI to 10 mg/ml and administered IP at 100 mg/kg. 5-Cyclo-hex-T-enyl-l,5-dimethylbarbituric acid (hexobarbitone) sod ium (May & Baker Ltd.) was dissolved in water at 50 or 25 mg/ml and administered IP at 150 mg/kg (male rats), 75 m g/kg (fe male rats), 100 m g/kg (male and female mice). eZZ-Ethionine (Koch-Light) was dis solved in water (20 mg/ml) and administer ed as indicated in the text. Microsomal Preparations Liver microsomes were isolated either as described by Bond and De Matteis (-4), ex cept that the microsomes were- sedimented at 105,000g, or by the calcium/sucrose meth od (J) adapted to the extent of sediment ing the calcium-treated microsomes at 1900ff and washing them once in a 0.0125ilf sucrose solution containing 8mM CaCL. The final preparations were suspended in 0.1M phos phate buffer (pH 7.4) containing 1mM Na2EDTA (1.4 -7 .0 mg protein/m l). Protein was estimated by the biuret method (3) and cytochrome P-450 (or P-448) by using an 211 784549 GENP011738 extinction coefficient of 91/mikf-cm ___(7) for the absorbancy change between 450 (448) and 490 nm of the CO difference spec trum of a Na..S30 4-reduced suspension. Assays Plasma barbiturate levels following hexobarbitone administration were measured by the method of Chromy and Babjuk (8). Hexobarbitone oxidase was measured in vitro by the modified (9) method of Brodie etal. (10). Zoxazolamine hydroxylase was measured by the methods of Burns and his colleagues (11, 12) with minor modifications. Difference Spectra * Difference spectra were recorded on a Unicam SP 1800 spectrophotometer with oxi dized microsomal suspensions (2.7 ml, 1.42.2 mg protein /m l) in each 1-cm cuvet and the addition of either aniline (5-200 pi, 0.22M in 0.1M phosphate buffer, pH 7.4) or hex obarbitone sodium (5-100 p\t 0.054M in wa ter) to the sample cuvet and of an equal volume of the appropriate solvent to the re ference cuvet. Pyridine difference spectra were recorded with Na^SiO.,-reduced micro somal suspensions (2.7 ml) and 0.3 ml of either 0.5A/ aqueous pyridine or water added to the cuvets. The wavelength of cytochrome P-450 (or P-448) maximum absorption was measured on a Cary 14 spectrophotometer calibrated with a holmium filter* Statistics Results are quoted as the mean S. E. M. and were analyzed by Student's t-test except where the nature of the results necessitated the use of a ranking test. Results and Discussion Effect of Dioxin Administration on the In Vivo Action, and In Vitro Metabolism of Drugs in the R at One or 3 days after being given a single oral dose of dioxin rats show a shortening of the zoxazolamine paralysis time but an increase in the duration of action of hexo barbitone (2). The effect on the hexobarbi tone sleeping time becomes progressively more marked with time until, 2 weeks after dosing, rats sleep over 4 hr and some of them die without waking (Table 1). Table 1. Effect of dioxin on the hexobarbitone sleeping time of male rats.* Sleeping time, min6 Time a f te r ---------- ------------------ ;--------- ;----------------------------------------- dosing Dioxin (200 A*g/kg, ___ PO) Solvent 12 hr 24 hr 72 hr 1 week 2 weeks 32.9 3.5 40.0 3.1 75.5 5.9 125.6 14.1 > 244 (6) (6) (6) (6) (6) 37.7 3.3 27.4 1.8 33.6 3.3 21.5 2.9 25.2 2.7 (5) (6) (6) (5) (6) Increase over controls, P % - - NS 46 < 0.01 125 < 0.001 480 < 0.001 870 0.0022e ` Hexobarbitone sodium administered as described in Materials and Methods section. *Numbers of animals in parentheses. eBy ranking test. In view of the prolonged reduction of food intake of rats dosed with dioxin (3) it is possible that a starvation effect (13) might contribute significantly to the increased hexo barbitone sleeping times at later stages in the intoxication (1 or 2 weeks after dos ing). However earlier experiments (2) had suggested that, 1 day after dosing with di 212 oxin, the reduced food intake was not the sole cause of the prolongation of the hexo barbitone sleeping time. This interpretation is open to criticism, since in these experi ments the animals had not been prefasted before dioxin administration and the dioxin might have slowed down the absorption of the'residual food in the stomach and intes- Environmental Health Perspectives CCI TTf . k f in 784550 ( ; Accordingly groups of eight female rats and either hexobarbitone or zoxazolamine starved for 39 hr and then dosed with (Table 3). With either sex, following dioxin dioxin (200 ig/kg, PO) or oil. Following 24 treatment, there was a significant increase hr further starvation the hexobarbitone in the amount of zoxazolamine metabolized, sleeping time was measured; the value for and with males there was a significant de the dosed group, 89.0 3.8 min, was signific crease in the quantity of hexobarbitone oxi antly higher than that of the control group, dized. With the microsomes from female rats 63.2 5.3 min, (P <0.005). Thus the early ef the control level of hexobarbitone metabol fect of dioxin on the sleeping time cannnot ism was lower than the males and, although be entirely due to differences in food con in the treated livers it was lower still, the sumption or absorption between control difference was not significant. Apart from and treated animals. Therefore in all follow this, the results agree with the in vivo ex ing experiments we have used rats which periments and indicate that dioxin modi had been given a single dose of dioxin (200 fies the duration of the pharmacological ac ng/kg) 1 or 3 days previously. tion of both zoxazolamine and hexobarbi It was considered possible that dioxin tone by changing the rate of metabolism of might prolong the hexobarbitone sleeping these drugs by the liver microsomes. time by altering either the sensitivity of the nervous system or the distribution of the barbiturate within the body. These possibili ties were ruled out by an experiment in which the sleeping times of groups of dioxintreated and control rats were measured and, immediately after their waking, blood was r\o4?ected for the analysis of plasma barbi- <.te levels (5). Table 2 indicates that, ough there was a significant increase in the sleeping time of the dosed animals, the waking plasma barbiturate levels of the Table 3. Effect of dioxin treatment on the metabolism of hexobarbitone and zoxazolamine by rat liver microsomes.1 Treatment Sex Zoxazolamine metabolized, nmole/mg microsomal protein/hr " Hexobarbitone metabolized, nmoles/mg microsomal protein/hr ' Dioxin Controls Dioxin Controls M 32.2 4.2 (10) M 15.7 2.0 (10) P <0.005 F 46.9 7.9 (5) F 6.3 5.4 (5) 31.0 23.9 (5) 192.0 18.7 (5) < 0.001 29.4 47.8 (5) 109.2 49.7 (5) two groups were not significantly different. P <0.005* NS Table 2. Effect of dioxin on the sleeping time and waking plasma barbiturate level of female rats.1 Treatment Dioxin Controls P 'N 6 6 Sleeping time, min 99.3 19.8 45.9 4.2 <0.025 Plasma barbiturate, Mg/ml 67.9 2.2 64.5 2.9 NS *Hats were dosed with dioxin (200 Mg/kff. PO) or oil and 3 days later the sleeping time in duced by hexobarbitone sodium (75 mg/kg, IP) was measured. Immediately after waking the animals were anesthetized (ether) and blood collected from the heart. Plasma bar biturate was measured by the method of Chromy and Babjuk (5). Three days after a single oral dose of either dioxin or oil to male or female rats the liver microsomes -were isolated and in- ated with an NADPH-generating system *Rats (180-200 g) received dioxin (200 g/kg) or oil PO and were killed 3 days later. Micro somes were isolated by the calcium/sucrose method (see Materials and Methods) and the assays performed essentially as described else where (9~12), bNumbers of animals in parentheses. Effect of Dioxin Administration on the Diver Content and Spectral Properties of Micro somal Cytochrome P-450 Besides oxygen and NADPH, it is known that three other components are required for the reconstitution of a system capable of metabolizing drugs in vitro. These are a lipid, phosphatidyl choline; a flavoprotein, NADPH/cytochrome P-450 reductase, and a hemoprotein, cytochrome P-450 (or P448) (14-16). We have investigated the ef fect of dioxin treatment on cytochrome P450 of rat liver microsomes. Groups of male September 1973 213 G ENP 011740 784551 Table 4. Changes in liver weight and microsomal protein and cytochrome P-450 induced by dioxin treatment of male rats.* Treatment Dioxin Controls Weight wet liver. g/100 g BW 4.34 i 0.12 3.21 0.06 P < 0.001 Microsomal protein, mg/g liver 26.5 0.9 27.3 i 0.6 NS Microsomal cytochrome P-450 nmoles/g nmoles/mg liver__________ protein________ Xm,, (nm) 55.1 1.5 28.1 & 0.7 <0.001 2.08 0.07 1.03 0.02 < 0.001 447.6 0.04 449.6 Q.0S < 0.001 *Rats received dioxin (200 ng/kg, PO) or oil and 3 days later were killed and liver microaomes isolated at 105,000g (see Materials and Methods). Each value is the mean of five observations. rats received an oral dose of either dioxin or oil and three days later were killed and their liver microsomes isolated. Table 4 in dicates that there was a significant in crease in the weight of wet liver from the treated animals. There was no increase in the microsomal protein content, expressed per gram of tissue, but the cytochrome con tent doubled. Further, the wavelength of the peak maximum of the cytochrome spec trum was shifted from 450 nm to 448 nm. Such a shift is a feature of the microsomal enzyme induction brought about by com pounds such as 20-methylcholanthrene (17). Since this inducer stimulates the metabolism of aromatic compounds, including zoxazolamine (18) this change correlates well with the observed stimulation of zoxazolamine metabolism both in vivo and in vitro-. How ever it should be remembered that, if the CO/cytochrome P-448 complex has an extinc tion coefficient more than four times that of P-450, as has been reported (19), then there has in fact been a decrease in hexno-protein concentration. The interaction of the reduced cytochrome P-450 of liver microsomes with the ligands ethyl isocyanide or pyridine is known to result in the formation of a difference spec trum with two peaks in the region 400-460 nm (20), The relative intensity of these two peaks, related to the absorption at 500 nm, is dependent on the pH of the suspend ing medium (20). In the case of liver microsomes prepared from animals pre treated with 20-mnthyicholanthrene and therefore containing P-448 rather than P450, the curve of pH dependence of the peak height ratio is so shifted that the peaks are of equal intensity at a lower pH (21). The pH dependence of the pyridine difference spectra of microsomes from methylcholanthrene-, dioxin- or oil-treated rats was measured and is shown in Figure 1. It can be seen that the curves due to dioxin or methylcholanthrene treatm ent are similar to each other but distinct from that of the control. Figure 1. Rats (male, 180-200 g) received dioxin (200 Mg/kg, PO) or Arachis oil (2 ml/kg, PO) 3 days before killing. 20-Methylcholanthrene was injected (20 mg/kg, 10 mg/ml in oil, IP) 3 anc 2 days before killing. All were starved for 24 hr before decapitation. Portions of liver microsomes equivalent to 0.4 g of wet liver were isolated by the calcium/sucrose method (5) and suspended in 0.1M PO, buffer (10 ml, containing l " ** NajEDTA) of the appropriate pH. Pyridine differ ence spectra were measured as stated in Maieria-s and Methods. 214 Environmental Health Perspectives 784552 GENP 011741 . Lis known thatL in the oxidized state, the cytochrome P-450 of liver microsomes will interact with various substrates of the ,Ii-ug metabolizing system to produce characeristic difference spectra (22). We have tudied the effect of dioxin pretreatm ent of "its on the interaction of aniline and hexoljarbitone with ra t liver microsomes. With aniline the control microsomes showed a normal Type II spectral change; this was intensified in the case of the microsomes t'rom treated rats (Fig. 2). Such a change is consistent with the increased cytochrome P-448 in these preparations. However, a iouble reciprocal plot of the spectral change/ nmole of P-450 (P-448) against aniline con centration indicated that there was an in crease in Ka (binding affinity constant) and in the maximal Type II spectral change (Fig, 3). A similar effect has been reported following methylcholanthrene induction (23). Figure 2. Rats (male, 180-200 g) received dioxin (200 Mg/kg) or oil PO 3 days before killing. They were starved for 24 hr before isolation of liver microsomes at l05,000g. Preparations from two animals were combined. Difference spectra (aniline 0.41nuV/ final concentration) measured as de scribed in Materials and Methods on suspensions containing microsomes equivalent to 0.073g wet 'isr/mi. / St^teraber 1973 Figure 3. Rats (male, 170-190 g) received dioxin (200 ng/kg) or oil PO 3 days before killing. Liver microsomes, equivalent to 2.8 g wet liver, were isolated by the calcium/sucrose method (5) and were suspended in 0.1M PO buffer (25 ml, pH 7.4) containing 1mM EDTA. Aniline concentra tions in the cuvet (see Materials and Methods) were varied from 2.0 to 15.2mM. Each point is the mean of observations on five animals. On each axis the mean intercepts for groups of dosed and control animals were significantly different (P < 0.05). An extinction coefficient of 91/mjV/-cm (7) was used in the estimation of cytochrome P-450 (-448), The spectral changes observed are related to the total cytochrome content of the cuvet. The interaction of hexobarbitone with mi crosomes from control animals produced the expected Type I difference spectra (22) as illustrated in Figure 4. However the micro somal preparations from rats which had re ceived dioxin 3 days previously consistently gave a difference spectrum with a peak at 412 nm and a trough at 380 nm. This type of spectrum has been termed a modified Type II spectral change (22) and has been ob served following methylcholanthrene pre treatm ent (24), but only with microsomal preparations from female rats. A decrease in the intensity of the Type I difference spec trum due to hexobarbitone has been reported for methylcholanthrene-treated male and female rats (23). 215 784553 Comparison of the Effects of 20-Methylcholanthrene and Dioxin on Zoxazolamine Action in Mice of the DBA/2 Strain Freuse 4. All details as in the legend to Figure 2 except that hexobarbitone was added to the sample cuvetts to a concentration of Q.79mAf. All the changes of the properties of the liver microsomes from dioxin-treated rats are consistent with dioxin being a powerful inducer of the methylcholanthrene type. It has been reported that in mice of the DBA/2 strain the levels of hepatic aryl hy drocarbon hydroxylase are unaffected by me thylcholanthrene {25) or benz [a] anthra cene (26) treatment. Indeed when DBA/2 mice (male, 20-30 g) were injected with methylcholanthrene or oil as described by Nebei't et al. (25) we found, 24 hr after injection, no significant difference between the zoxazolamine paralysis times of the treated and control groups (58.0 5.1 min, V - 6 and 59.3 3.9 min, N = 6, re spectively). This contrasts with C57BL/6 mice (male, 18-26 g) in which methylchol anthrene pretreatment significantly reduced the paralysis time (dosed: 15.0 1.0 min, iV = 9, controls: 79.5 5.3 min, N - 9; P <0.0001). However in both strains a single oral dose of dioxin (200 fig/kg) given 3 days before zoxazolamine significantly lowered the paralysis time (DBA/2, male, dosed: 10.5 0.8 min, N = 6, controls : 65.7 4.2 min, iV = 5, P < 0.001; for C57BL/6 see Table 5). 011743 Table 5. Effect of dioxin on sleeping and paralysis times of C57 BL/6 mice.* Treatment Sex Dioxin Oil M M Dioxin Oil F F After 3 days 52.6 6.2 (7) 71.3 3.8 (6) P <0.05 52.6 6.5 (7) 65.1 8.4 (7) P NS Sleeping time, min6 After 10 days 60.6 4.7 (7) 60.2 7.7 (4) NS -- -- After 20 days >159.7 (7) 53.4 7.3 (6) < 0.01 9 -- -- Paralysis time after 3 days, min 7.0 0.7 (4) 46.8 3.5 (5) < 0.001 5.9 0.6 (5) 37.7 4.0 (7) < 0.001 *Mice (11-27 g) received dioxin (200 fig/kg) or oil PO and hexobarbitone sodium or zoxazolamine af ter the stated intervals. *Numbers of animals in parentheses. ' By ranking test. 2 216 Environmental Health Perspectives 784554 ijsumingr that, under these conditions, the duration of zoxazolamine action is en tirely governed by the rate of metabolism in the liver it would appear that the appar ent genetic noninducibility of hepatic aryl hydrocarbon hydroxylase in certain mouse trains is not an absolute tra it but depend ent on the inducer used. This is in agree ment with the observations (26) that this enzyme, although present in smaller amounts in extrahepatic tissues, is inducible in such organs of strains in which the liver enzyme is unaffected. investigation of the Relationship between Stimulation of Zoxazolamine Metabolism and Inhibition of Hexobarbitone Metabolism Although in the ra t dioxin produces sim ultaneous and divergent effects on the met abolism of zoxazolamine and hexobarbitone by liver microsomes, it is not known whether these two effects are related to each other. An alternative would be that dioxin has two separate and distinct effects: the induc tion of zoxazolamine hydroxylase and the depression of hexobarbitone oxidase. The following two experiments indicate that, after dioxin treatment, a stimulation of zo xazolamine metabolism can be observed in the absence of any inhibition of hexobarbi tone metabolism. In mice of the C57BL/6 strain, as in the rat, a marked reduction of the zoxazola mine paralysis time was observed at 3 days after dioxin treatment. However, at this time the hexobarbitone sleeping time was either unchanged (in female mice) or short ened (in male m ice); a prolongation was seen in males only at 20 days (Table 5). Table 6. Effect of ethionine on dioxin-induced alterations in sleeping time and paralysis time in rats.* i Route of administration iex ---------------------------------- ---------- Dioxin Oil Ethionine M IP _ -- -- IP -- M IP -- PO -- IP PO M PO --. IP -- PO IP- P IP -- PO -- IP PO Sleeping time, min * 75.4 13.9 (5)e 36.0 2.2 (5) 41.3 3.6 (5) 45.0 . 3.1 (5) 44.7 3.1 (6) 42.0 3.4 (6) 83.0 9.4 (6)f 136.4 7.2 (6) Paralysis time, min _ -- 84.0 6.5 (5)d * >383 (5) -- -- 119.2 9.4 (6) *,f >480 (6) 1Animals received di-ethione (200, 100, and 100 mg/kg) at t = 0, 4, and 8 hr, respectively; dioxin (200 Mg/kg) or oil at t = 0.5 hr and sleeping or paralysis times were measured at t s 24.5 hr. All were starved during the experiment. bNumbers of animals in parentheses. *Significantly different from controls at P <0.05. a Significantly different from controls at P < 0.01. *By ranking test. f Significantly different from controls at P <0.005. The effect of dioxin on hexobarbitone met abolism can also be suppressed in rats by administering dZ-ethionine together with the dioxin. Along with a single oral dose of dioxin male rats were given a series of in jections of dZ-ethionine, an inhibitor of pro,, synthesis which has been reported as ' enting the induction of drug metabol September 1973 ising enzymes by either 20-methylcholanthrene (27) or phenobarbitone (28). It was found (Table 6) that, whereas the ethionine was ineffective in preventing the reduction of paralysis time in dioxin-treated animals, it did abolish the prolongation of hexobar bitone sleeping time due to dioxin treat ment. When the routes of administration 217 784555 of dioxin and ethionine were interchanged there was still no difference in hexobarbitone metabolism. When the experiment was carried out with female rats, in which ethionine is a more effective inhibitor of protein synthesis (29), the duration of both paralysis and sleep of the dioxin-treated animals was shorter than that of the controls (Table 6). Thus although these doses of ethionine are not effective in blocking the induced metabolism of zoxazolamine (27) in either sex, they are capable of preventing or even reversing the effect of dioxin on hexobarbitone metabolism. . Conclusions 1. The divergent effects of dioxin on the duration of action of hexobarbitone and zo xazolamine in the rat m vivo (2) are a consequence of changes in hepatic metabol ism of these drugs. 2. Dioxin causes alterations in the proper ties of cytochrome P-450 of rat liver microsomes which are similar to those produced by methylcholanthrene. 3. It is the most effective stim ulator of aromatic hydroxylation known, see (1), and can apparently overcome a genetic resistance to hepatic microsomal enzyme induction in mice of the DBA/2 strain. 4. The effects of dioxin on hexobarbitone and zoxazolamine metabolism can be separ ated by the use of C57BL/6 mice or ethionine-treated rats and might be due to two separate modes of action. Acknowledgement We thank McNeil Laboratories, Inc., Camp Hill Road, Fort Washington, Pa. and May & Baker Ltd., Dagenham, Essex for gifts of zoxazolamine and hexobarbitone sodium, re spectively, and Mr. C. M. Puah and Mr. B. S. Sood for technical assistance. REFERENCES 1. Buu-Ho, N. P., et al. Proprits cancromimtiques de la ttrachloro-2,3,7,8-dibenzo-p-dioxiiie ("dioxine7'). CompL Rend., Ser. D 272: 1447 (1971). 2. Greig, J. B. Effect of 2,3,7,8-tetrachlorodibenzoI, 4-dioxin on drug metabolism in the rat. Biochem. Pharmacol. 21: 3196 (1972). 3. Greig, J. B., et al. Toxic effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Food Cosmet, Toxicol. 11: 585 (1973). 4. Bond, E. J., and De Matteis, F. Biochemical changes in rat liver after administration of car bon disulphide, with particular reference to' microsomal changes. Biochem. Pharmacol. IS: 2531 (1969). 5. Kamath, S. A., and Rubin, E. Interaction of cal cium with microsomes: a modified method for the rapid isolation of rat liver microsomes. Bio chem. Biophys. Res. Commun. 49: 52 (1972). 6. Aldridge, W. N. Adenosine triphosphatase in the microsomal fraction from rat brain. Biochem. J. 83: 527 (1962). 7. Omura, T., and Sato, R. Fractional solubiliza tion of haemoproteins and partial purification of carbon monoxide-binding cytochrome from liver, microsomes. Biochem. Biophys. Acta. 71: 224 (1963). 8. Chromy, V., and Babjuk, J. S. Determination of barbiturates in biological fluids. Clin. Chim. Acta 37: 547 (1972). 9. Cooper, J. R., and Brodie, B. B. The enzymatic metabolism of hexobarbital (Evipal). J. Phar macol. Exp. Therap. 114: 409 (1955). 10. Brodie, B. B. et al. The fate of pentobarbital in man and dog and a method for its estimation in biological material. J. Pharmacol. Exp. Therap. 109: 26 (1953). 11. Bums. J. J. et al. Zoxazolamine. Physiological disposition, uricosuric properties. Amer. J. Med. 25: 401 (1958). 12. Conney, A. H., Trousof, N., and Burns,. J. J. The metabolic fate of zoxazolamine (Flexin) in man. J. Pharmacol. Exp. Therap. 128: 333 (1960). 13. Dixon, R. L., Shultice, R, W., and Fouts, J. R. Factors affecting drug metabolism by liver mi crosomes. IV. Starvation. Proc. Soc. Exp. Biol. Med. 103: 333 (1960). 14. Lu, A. Y. H., and Coon, M. J. Role of hemoprotein P-450 in fatty acid w-hydoxylation in a soluble enzyme system from liver microsomes. J. Biol. Chem. 243: 1331 (1968). 15. Lu, A. Y. H., Junk, K. W., and Coon, M- J. Resolution of the cytochrome P-450-containing w-hydroxylation system of liver microsomes into three components. J. Biol. Chem. 244: 3714 (1969). 16. Lu, A. Y. H., et al. Reconstituted liver micro somal enzyme system that hydroxylatea drugs, other foreign compounds, and endogenous sub strates. IV. Hydroxylation of aniline. Arch. Biochem. Biophys. 153: 294 (1972). 17. Alvares, A. P., et. al. Studies on the induction of CO-binding pigments in liver microsomes by 218 Environm ental H ealth Perspectives 784556 GENP 011745 p. , enobarbitai and _3-methylcholanthrene. Bioehem. Biophys. Res. Commun. 29: 521 (1967). 18. Conney, A. H., et al. Adaptive increases in drugmetabolising enzymes induced by phnobarbital and other drugs. J. Pharmacol. Exp. Therap. 130: 1 (1960). 19. Hildebrandt, A., Remmer, H., and Estabrook, R. W. Cytochrome P-450 of liver microsomes-- one pigment or many. Biochem. Biophys. Res. Commun. 30: 607 (1968). 20. Imai, Y., and Sato, R. Anomalous spectral in teractions of reduced P-450 with ethyl isocya nide and some other lipophilic ligands, j, Bio chem. (Tokyo) 62: 464 (1967). 21. Sladek, N. E., and Mannering, G. J. Evidence for a new P-450 hemoprotein in hepatic micro somes from methyicholanthrene treated rats. Biochem. Biophys. Res. Commun. 24: 668 (1966). 22. Schenkman, J. B,, Remmer, H., and Estabrook, R. W. Spectral studies of drug interaction with hepatic microsomal cytochrome. Mol. Pharmacol. 3: 113 (1967). 23. Kato, R., Takanaka, A., and Takayanaghi, M. Substrate-induced spectral change of liver mi crosomes in phnobarbital and methylcholanthrene-treated male and female rats. J. 'Biochem. (Tokyo) 68: 395 (1970). 24. Schenkman, J. B., et al. On the problem of possi ble other forms of cytochrome P in liver mi crosomes. Biochem. Biophys. Acta 171: 23 (1969). 25. Nebert, D. W., Goujon, F. M. and Gielen, J. E. Aryl hydrocarbon hydroxylase induction by polycyclic hydrocarbons: simple autosomal domi nant trait in the mouse; Nature New Biol. 236: 107 (1972). 26. Wiebel, F. J., Leutz, J. C., and Gelboin, H. V. Aryl hydrocarbon (benzo[a]pyrene) hydroxy lase: inducible in extrahepatic tissues of mouse strains not inducible in liver. Arch. Biochem. Biophy3. 154: 292 (1973). 27. Conney, A. H., Miller, E. C., and Miller, J. A. The metabolism of methylated aminoazodyes. V. Evidence for induction of enzyme synthesis in the rat by 3-methylcholanthrene. Cancer Res. 16: 450 (1956). * 28. Kato, R., Chiesara, E., and Vassanelli, P. Factors influencing induction of hepatic microsomal drug-metabolising enzymes. Biochem. Phar macol. 11: 211 (1962). 29. Farber, E. Ethionine carcinogenesis. Advan. Cancer Res. 7: 383 (1963). GENP 011746 September 1973 219 784557 Studies of the Effects of 2,3,7,8-TetrachlorodibenzorP-dioxin on Mammalian Hepatic s-Aminolevulinic Acid Synthetase by James S. Woods* Introduction The toxicity of 2,3,7,8-tetrachlorodibenzo-dioxin (TCDD), a contaminant formed ing the manufacture of the herbicide '5-trichlorophenoxyacetic acid (2,4,5-T) is well known (1-4). The widespread utiliza tion of this compound has caused increased concern about the potential health hazards created by the presence of TCDD in the en vironment. The articles which accompany this account, indeed, attest to the highly toxic nature of TCDD in both laboratory animals and man. Interest in TCDD as a potential porphyrogenic agent arose when porphyria cutanea tarda, a form of hepatic porphyria, occurred in industrial workers associated with the manufacture of 2,4,5-T (5). Hepatic porphy ria is a syndrome characterized by a variety of symptoms including the overproduction and excretion of porphyrins, pigmentation of the skin, photosensitivity, and intestinal and neurological disorders. The disease is characterized biochemically by an increase in the activity of the mitochondrial enzyme Pathologic Physiology Branch, National Institute of Environmental Health Sciences, National Inatitptf* of Health, P.O. Box 12233, Research Triangle c. North Carolina 27709. _ September 1973 S-aminolevulinic acid (ALA) synthetase, which is the first and rate-limiting enzyme in the heme biosynthetic pathway (6) (Fig. 1). The possibility that TCDD has porphyrogenic properties has been indicated by the recent observation that TCDD is a potent inducer of hepatic ALA synthetase in chick embryos (7). There is, however, no evidence to indicate that TCDD produces similar ef fects in mammalian species. The chick em bryo system is exquisitely sensitive to the effects of agents which induce ALA synthe tase (5). Previous studies from our labora tory (9, 10), on the other hand, have shown that mammalian species demonstrate a striking variability in their response to porphyric agents, especially at different stages of development. These studies were undertaken, therefore, to determine the possible porphyrogenicity of TCDD in mammalian species and to as sess further the utility of the chick embryo liver system as an indicator of the potential porphyrogenic effects of environmental agents in mammals. Materials and Methods ALA synthetase activity was determined by two procedures. In liver homogenates* 221 784558 li4 t. ; Glycine + Succinyl-CoA ALA Synthetase 5-Aminolevulinic Acid (ALA) ALA Dehydratase Porphobilinogen URO Synthetase Uroporphyrinogen (URO) 1 Coproporphyrinogen 1 Protoporphyrin Heme Heme Synthetase -Fe+ + Figure 1. Heme biosynthetic pathway. ALA synthetase activity was assayed using the ion exchange chromatography technique described by Mauzerall and Granick {11), with liver homogenates prepared for incuba tion as described by Marver et al. (12). ALA synthetase activity in subcellular frac tions and at various stages of enzyme puri fication was measured using the procedure described by Scholnick et al. (13). In the latter case the incubation medium was modi fied to include 1 0 GTP in addition to the prescribed substrates. The method of Scholnick was also used for the isolation and 50-fold purification of ALA synthetase from porphyric ra t liver. Hepatic subcellular fractions were prepared as described by Hayaski et al. ( H ) . Protein concentrations were assayed by the method of Lowry etal. (15). All test animals were treated orally with a standard solution of TCDD (10 /g/ml) dissolved in a corn oil/acetone mixture (6:1) 24 h r prior to sacrifice unless otherwise in dicated. Two groups of control animals were used in all in vivo experiments. The first group received only the corn oil/acetone mix ture in an amount equivalent to that in which TCDD was administered to test ani mals. The second group was treated subcu taneously with allylisopropylacetamide (AIA) (400 m g/kg), which is a well known and potent inducer of ALA synthetase in mammals (12, 16). TCDD was obtained from Dow Chemical Company, Midland, Michigan. AIA was a gift from Hoffmann-LaRoche, Nutley, New Jersey. All animals and other chemicals were obtained from standard sources. Results and Discussion Initial studies were designed to determine the potential porphyro genic effects in rats of TCDD when administered in doses up to 25 fig/kg, the reported LDS0 for this species (17). Male rats were treated with a single 5 or 25 /tg/kg dose of TCDD, and hepatic ALA synthetase activity was assayed at pe riods up to 28 days thereafter. In animals receiving a single dose of AIA, ALA syn thetase activity increased to approximately seven times the control level after 24 hr and returned to control levels by the third day after treatment. On the other hand, TCDD did not significantly alter ALA syn thetase activity, as measured in whole liver homogenates, during any p art of the test period in any of the animals. All measure ments of enzyme activity were within the range observed in controls. It has been recently determined (14) that ALA synthetase is a mitochondrial enzyme but is synthesized extramitochondrially on the cytoplasmic ribosomes. The enzyme is subsequently incorporated into th e . mito chondria, where it becomes active. ALA synthetase activity may be altered by agents which interfer with any aspect of this pro cess. It was, therefore, of interest to deter mine if TCDD might influence the subcellu lar localization of ALA synthetase and thereby alter the regulation of hepatic hem- 222 Environmental Health Perspectives 784559 GENP011748 Table X. Subceilular distribution of hepatic ALA synthetase in normal and AIA- and TCDD-treated rats/ ALA, nmole/mg protein-hr S.E. Group 1 2 3 4 Treatment Com Oil AIA TCDD AIA + TCDD Mitochondrial fraction 0.52 0.09 2.26 0.80 0.58 0.10 2.91 0.87 9,000(7 supernatant 0.27 0.08 1.44 0.40 0.34 0.06 1.37 0.51 105,0000 supernatant 0.24 0.10 1.47 0.30 0.15 0.43 1.17 0.51 Microsomal fraction 0.03 0.01 0.28 0.10 0.01 0.01 0.27 0.08 4Rats were treated with TCDD (25 jig/kg) and/or AIA (400 mg/kg) 24 hr prior to sacrifice. atopoiesis in a manner which could not be Finally, the potential porphyrogenic ef detected when ALA synthetase activity was fects of TCDD during the perinatal period measured in whole liver homogenates. The were investigated. These studies were con results of these experiments are seen in Ta ducted in fetal rats which were 3 days from ble 1. Analysis of the subceilular distribu delivery, ( - 3 day), and on two groups of tion of ALA synthetase in adult ra t liver newborn rats, 4 and 12 days after delivery reveals a 2:1 distribution of activity between on the day of sacrifice. Newborn rats were mitochondrial and postmitochondrial frac treated orally with 25 /ig/kg doses of TCDD, tions. Most of the postmitochondrial activity whereas fetal rats were treated by way of is retained in the 105,000sr supernatant the mother. Livers from mother rats served fraction with very little found in the micro- as adult samples. The results are shown in somes. Control rats in the group treated Table 2. In no case did ALA synthetase ac AIA showed substantial increases in tivity in TCDD-treated rats differ from that VLA synthetase activity in all fractions observed in untreated animals. Fetal ALA uui relatively little alteration in the distri synthetase is typically five to eight times bution of activity between fractions after that of the adult (9 ) and declines to adult 24 hours. TCDD administered alone at 25 levels shortly after birth. Refractoriness to Mg/kg doses or together with AIA caused no induction of ALA synthetase is observed un significant alteration of ALA synthetase til the activity approaches that of the adult activity from that observed in control (10, 18). At no stage of development, how groups. In addition, TCDD did not affect ever, did the ALA synthetase activity in the induction of ALA synthetase by AIA TCDD-treated rats significantly differ from nor alter the subceilular distribution of the that observed in controls. enzyme during induction. Similar observa These results are of particular interest in tions were made when TCDD was adminis v view of the potent induction of ALA syn tered in 100 m g/kg doses or when these ex thetase activity produced by TCDD in the periments were conducted with female rats chick embryos. The lack of a significant ef or mice. fect of TCDD on ALA synthetase in mam In vitro tests of the possible effects of mals suggests that major differences exist TCDD or ALA synthetase activity were among these species in the biological mech conducted in liver homogenates, isolated mi anisms which determine the ultimate phar tochondria and on ALA synthetase purified macological disposition of chemicals such of 50-fold from porphyric ra t liver. In enzyme TCDD. Species variations in drug distribu incubation mixtures containing TCDD in tion, biotransformation, and excretion are concentrations ranging from 10'9 to IQ~*M well known (19). On the other hand, major no discernable effects on the enzyme activity differences in the mode of action of TCDD could be observed. may also reflect alterations in the regulation ember 1973 223 784560 Table 2. Study of potential effects of TCDD on ALA synthetase during development in rats.* ' ALA, nmole/hr-g liver S.E. Group 1 2 3 Treatment Corn oil TCDD AIA --3 day 241 31 253 47 248 2 8 +4 day 94 7 76 21 118 12 +12 day ' 50 16 60 8 113 13 Adult 45 5 44 * i 3 347 *Pregnant and newborn rats were treated 24 hr prior to sacrifice with TCDD (25 Mg/kg) or AIA (400 mg/kg). of hepatic heme synthesis in these species. Striking variations in the developmental aspects of ALA synthetase regulation have already been described. The chick embryo liver is one of the most sensitive systems available in which to study the induction of ALA synthetase (5). In contrast, our previ ous studies (10) have shown that fetal mammals are totally insensitive to the ef fects of drugs which induce or repress hepatic ALA synthetase in chick embryo or in adult animals. Moreover, fetal mamma lian ALA synthetase activity is significantly elevated in comparison with that of the adult (3), whereas the control level of ALA synthetase in 17-day old chick embryo liver is only one-third that measured in the adult chicken (20). Differences in porphyrin metabolism in these two species is also suggested by the failure of a wide variety of drugs which in duce hepatic porphyrin accumulation in chick embryos to do so in mammalian liver (21). Among the drugs capable of causing significant increases in porphyrin levels in chick embryo liver are 3,5-dicarbethoxy-l,4dihydrocollidine (DDC), glutethimide, methsuximide, secobarbital, methylprylon, and mephenytoin. Of this group, only DDC is capable of producing porphyrin accumula tion in mice. In addition, it has been recently indicated that uroporphyrinogen synthe tase, another enzyme in the heme biosyn thetic pathway, may play a rate-limiting role in hepatic heme synthesis in certain strains of mice (22). It therefore appears that significant dif ferences exist among these species with re gard to various steps in the regulation of hepatic heme synthesis and porphyrin meta- bolism. These differences, along with varia tions in the capacity to metabolize and dis tribute drugs such as TCDD, may account, for the differences in susceptibility of these species to the porphyrogenic effects of TCDD and perhaps to other environmental contaminants. The foregoing considerations attest to the increasingly prevalent observation that many environmental agents are hazardous not only by virtue of their inherent toxicity but also because of their specificity of action in different species. While utilization of non mammalian test systems may provide, in some cases, a sensitive indication of the po tentially toxic effects of certain drugs and chemicals in mammals, this study demon strates the necessity for developing test pro cedures which will more clearly predict the deleterious effects of environmental con taminants in mammalian species, especially in man. REFERENCES 1. Courtney, K. D., and Moore, J. A. Teratology studies with 2,4,5-trichIorophenoxyacetie acid and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 20: 396 (1971). 2. Buu-Hoi, N. P., et al. Organs as targets of "dioxin" (2,3,7,8-tetrachlorodibenzo-p-aioxin). Naturwiss. 59: 174 (1972). 3. Fishbein, L., and Flamm, W. G. Potential en vironmental chemical hazards. Part I: Drugs. Sci. Total Environ. 1: 15 (1972). 4. Sparschu, G. L., Dunn, F. L., and Rowe, V. K. Study of the teratogenicity of 2,3,7,8-tetrachiorodibenzo-p-dioxin in the rat. Food Cosmet Toxi col. 9: 405 (1971). 5. Poland, A. P., et al. A health survey of workers in a 2,4-D and 2,4,5-T plant. Arch. Environ. Health 22: 361 (1971). 224 Environmental Health Perspectives 784561 GENP0H750 6. Granick, S,t and Urata, J. Increase in the ac tivity of a-aminplevulinic acid synthetase in liver mitochondria induced by feeding 3,5-dicarboxyethyl-l,4-dihydrocoIlidine. J. Biol. Chem. 238: 821 (1963). 7. Poland, A., and Glover, E. 2,3,7,8-Tetrachlorodibenzo-p-dioxin: A potent inducer of 5-aminolevulinic acid synthetase. Science 179: 476 (1973). 8. Granick, S. The induction in vitro of the syn thesis of 3-aminolevulinic acid synthetase in chemical porphyria. J. Biol. Chem. 241: 1366 (1966). 9. Woods, J. S., and Dixon, R. L. Perinatal differ ences in delta-aminolevulinic acid synthetase activity. Life Sci. 9: 711 (1970). 10. Woods, J. S., and Dixon, R. L. Studies of the perinatal differences in the activity of hepatic 5-aminoIevulinic acid synthetase. Biochem. Phar macol. 21: 1735 (1972). 11. Mauzerall, D., and Granick, S. The occurrence and determination of 3-aminolevulinic acid and prophobilinogen in urine. J. Biol. Chem. 219: 435 (1956). 12. Marver, H. S., et aL 3-Aminolevulinic acid syn thetase. I. Studies in liver homogenates. J. Biol. Chem. 241: 2803 (1966). 13. Scholnick, P. L., Hammaker, L. E. and Marver, H. S. Soluble 3-aminolevulinic acid synthetase of rat liver. I. Some properties of the partially purified enzyme. J. Biol. Chem. 247: 4126 (1972). '4. Hayashi, N., Kurashima, Y., and Kikuchi, G. Mechanisms of allylisopropylacetamide-induced increase of -aminolevulineate synthetase in liver mitochondria. Arch. Biochem. Biophys. 148: 10 (1972) '. 15. Lowry, 0. H., et al. Protein measurements with the Folin Phenol reagent. J. Biol. Chem. 193: 265 (1951). 16. Song, C. S., Lee, W., and Kappas, A. J-Aminolevulinate synthetase and drug-induced disease in microsomal cytochrome P-450 of the liver. Clin. Res. 18: 389 (1970). 17. Schwetz, B.f et al. Chlorodibenzo-p-dioxin toxi cology. Environ. Health Perspect. No. 5: 87 (1973) . 18. Song, C. S., et al. The influence of postnatal development on drug-induced hepatic porphyria and the synthesis of cytochrome P-450. J. Exptl. Med. 134: 1349 (1971). 19. Goldstein, A., Aronow, L., and Kalman, S. M. Drug toxicity in lower animals and man. In: Principles of Drug Action, Harper and Row, New York, 1968, Chapt. 5. 20. Creighton, J. M., and Marks, G. S. Drug-induced porphyrin biosynthesis VII. Species, sex, and developmental differences in the generation of experimental porphria. Can. J. Physiol. Phar macol. 50: 485 (1972). 21. Racz, W. J., and Marks, G. S. Drug-induced porphyrin biosynthesis II. Simple procedure for screening drugs for pophyria-inducing activity. Biochem. Pharmacol. 18: 2009 (1969). 22. Hutton, J. J., and Gross, S. R. Chemical induc tion of hepatic porphyria in inbred strains of mice. Arch. Biochem, Biophys. 141: 284 (1970). jeptember 1973 225 784562 Effect of 2,3,7,8-Tetrachlorodibenzo-p-dioxin on the Biliary Excretion of Indocyanine Green in Rat by Shang W. Hwang* Chlorinated dibenzodioxins have been found as contaminants of various technical chlorinated compounds such as 2,4,5-trichlorophenoxyacetic acid (2,4,5-T), which are widely used in agriculture. The dioxin contaminants may be involved in various pathologic states resulting from exposure to -^ .h n ic a l chlorinated compounds (1,2). liv e r f xosis has been observed in the animals .inch were treated with derivatives of chlorophenol (5) or other chlorinated com pounds (4), and it was suggested that liver necrosis-causing factors were present in these compounds. In view of the above ob servation, the present study was undertaken to investigate whether 2,3,7,8-tetrachloro-pdibenzodioxin (TCDD) has any effect on the hepatobiliary function of rat. Biliary excre tion of indocyanine green (ICG) was used as the index of function. The dye was chosen because it is completely and rapidly ex creted by norinal liver into the bile by an active process (5). Materials and Methods Male random bred CD rats weighing 350400 g were treated PO with a single dose of 25 fig/k g or 5 /ig/kg TCDD in acetone and Pharmacology and Toxicology Branch, National Institute of Environmental Health Sciences, Na tional Institutes of Health, P.O. Box 12233, Research Triangle Park, N.C. 27709. corn oil. The controls received an equivalent volume of the vehicle. At day 1, 7, and 16 after treatment, the effect of TCDD on the bile flow and the biliary excretion of ICG was examined. Animals were first anesthe tized with penobarbital Na (50 mg/kg) IP. Through an abdominal incision, renal pedi cles were ligated, and the common bile duct was cannulated with a blunt 23-gauge hypo dermic needle shaft attached to an 8-in. piece of PE 50 tubing. Bile was collected for a 20-min period, and the amount collected was measured by weighing. At the end of 20 min, freshly prepared indocyanine green in aqueous solvent (Hynson, Westcott and Dun ning, Inc.) was injected at a dose of 6.25 m g/ kg into the femoral vein, and the bile was collected for another 20 min. The body tem perature was monitored with a telether mometer and was maintained at 37 C by warming with an incandescent lamp. At the end of the experiment, blood was withdrawn by heart puncture and the liver was excised. The ICG concentration in bile, plasma, and liver was determined by measuring its ab sorption at 805 nm. The rate of ICG disap pearance from plasma was also determined by measuring the dye concentrations in a series of plasma samples which were with drawn by heart puncture 1, 2, 5, 3, 12, and 20 min after ICG injection from groups of control and TCDD-treated rats. v ptember 1973 227 784563 Results and Discussion Bile flow during the first 20 min of the experiment increased after TCDD treatment as shown in Figure 1. The initial flow rate continued to increase through the 16th day, and rats receiving 25 /tg/kg TCDD had higher flow rates than those receiving 5 MS/kg TCDD. decreased significantly as shown in Figure 3. Less hepatic excretion of the ICG was ob served with the larger dose of TCDD than with the smaller dose. The rate of excretion of this dye was still markedly suppressed by the 16th day after treatment. Figure 1. Effect of TCDD on bile flow. Each point is the mean SEII of four animals. Difference , from control is significant (P <0.005) for the 7th and 16th day after 25 Mg/kg TCDD treatment and 16th day after 5 Mg/kg TCDD treatment. Figure 2. Liver weight of rat after TCDD treat ment. Each point is the mean SEII of four animals. The difference is significant (P<0.05) for the 7th and 16th day after treatment of either 5 Mg/kg or 25 Mg/kg TCDD. The increase in bile flow could be due to the increasing secretion of water by the hepatic cell or due to the decreasing reab sorption of w ater along the bile duct. Both mechanisms were possible, but further in vestigation will be needed for clarification. TCDD also caused an increase in liver weight as shown in Figure 2. The increase in liver weight as expressed in grams liver weight per 100 g body weight was also doserelated since the higher dose caused a great er weight increase. Whether this increase in liver weight has any effect on the bile flow is not known. Similar results of bile flow increase and liver weight increase were observed in ra t after phnobarbital treatm ent (6), and both TCDD and phno barbital are potent microsomal enzyme in ducers. ICG excretion was also affected by TCDD treatment. The total amount of ICG excreted during 20 min after injection of the dye was 228 Figure 3. Effect of TCDD on ICG biliary excretion. The dose of ICG waa 6.25 mg/kg, and the bile was collected for 20 min after ICG injection. Each point is the mean S SEM of four animals. The difference from control is significant (P <0,C5) for the 7th and 16th day after treatment by 'either 5 Mg/kg or 25 Mg/kg of TCDD. Environmental Health Perspective 784564 r r / T T O _ T K T 'T T r\ Table 1. Effect of TCDD on liver uptake and biliary excretion of ICG. Time after TCDD treatment, days Control 1 day 7 days 16 days ICG concentration in plasma, /g/ml * TCDD 5 /ig/kg 1.0 2: 0.09 2.3 0.10 3.0 * 0.23 3.1 * 0.46 TCDD 25 /g/kg 1.9 0.09 2.75 0.29 3.45 0.15 3.90 0.58 ICG concentration in liver, /ig/g * TCDD 5 /ig/kg 28.5 0.9 29.5 2.3 37.5 a 3,6 41.0 5.4 TCDD 25 /ig/kg 28.5 0.9 29.0 1.8 31.0 2.1 35.0 3.6 ICG concentration in bile, Mg/mi ` TCDD 5 /igAg 1130 83 1015 79 869 66 860 59 TCDD 25 ag/kg 1130 83 999 64 710 * 45 624 46 1Dose of ICG was 6.25 mg/kg. Concentrations were determined 20 min after ICG injection. Each value is the mean SEM from four animals. The concentrations of ICG in plasma, liver, and bile were also analyzed separately for each animal 20 min after dye injection. The results (Table I) showed that concentration of ICG in bile of TCDD-treated ra t was lower than that of the control, and the ICG levels in plasma and liver of TCDD-treated rats were higher than that of the control. A greater depression of ICG concentration in bile and a greater retention of IQG in blood were caused by 25 /g/kg dose of TCDD. In contrast, animals treated with 5 /g/kg TCDD accumulated more ICG in liver than /e animals treated with 25 /ig/kg TCDD. o sign of recovery was shown by the 16th day after treatment. Bile-to-plasma, bile-to-liver and liver-toplasma concentration ratios of ICG were also separately calculated and compared, as shown in Figure 4. Bile-to-plasma and bileto-liver ratios decreased after TCDD treat ment, and the higher dose decreased these ratios even further. Liver-to-plasma ratio decreased only after 25 /g/kg but not after 5 p.g/k g TCDD pretreatm ent. TCDD ap peared to inhibit both ICG uptake by the hepatic ceil and the active secretion of ICG by the hepatic cell. Inhibition of both steps could result in the decreased total ICG ex cretion and lowered bile-to-plasma and bileto-liver ICG concentration ratios as found. The excretion of ICG from hepatic cell to bile was probably inhibited to a similar ex tent by both 25 /g/kg and 5 /g/kg TCDD. However, the inhibition of ICG uptake by the hepatic cell might be more dose-depend ent; the inhibition was greater with 25 Mg/kg than with 5 /g/kg TCDD, so the liver cumulated less ICG after 25 /g/kg TCDD Figure 4. Change of ICG concentration ratios after TCDD treatment Each bar represents the mean of four animals. The difference from control is significant (P <0.05) for all ratios at 7th and 16th day after TCDD treatment except the liver/ plasma (L/P) ratios after 5 ng/kg TCDD treat ment treatment, and liver-to-plasma ICG concen tration ratios decreased in animals treated with 25 /ig/kg but not 5 /g/kg TCDD. Con trol animals should take up ICG into liver faster than TCDD-treated animals, but the secretion from liver to bile was even faster compared to the treated animals, so the con trol animals accumulated less ICG in. both plasma and liver. September 1973 229 784565 V The rate of disappearance of ICG in plasma decreased after TCDD treatment, as shown in Figure 5. The 25 /ig/kg dose gave a greater reduction of disappearance rate than did the 5 tg/kg dose, and the rate was lower at the 16th day than at the 7th day after treatment. The decreasing rate of ICG dis appearance in plasma gave further evi dence that ICG excretion was damaged by the TCDD treatm ent. Figure 5. Effect of TCDD on the ICG disappearance rate in plasma. Each point is the mean of three rats. Sum m ary From the above observations, it was con cluded that TCDD inhibited hepatobiliary excretion of ICG, and the inhibitory effect appeared to be a long-lasting one. Many anionic compounds, including endogenous substances such as bilirubin, are actively se- 230 creted through the similar mechanism by the hepatobiliary system. The decreased ICG excretory ability might also apply to other anionic compounds, and the etiology of re ported cases of jaundice and porphyria after exposure to TCDD might be partly ac counted for by the decreased biliary excre tory ability. Environmental Health Perspective 784566 GENP 011755 ik n o w led g em en t The author is indebted to Dr. John A. Moore, Chief, Animal Science and Technol ogy Branch, NIEHS, for supplying the con trol and TCDD-treated rats. The author also wishes to acknowledge the assistance of Dr. James R. Fouts and Dr. Larry G. H art in the preparation of the manuscript. REFERENCES 1. Higginbotham, G. R., et al. Chemical and tox icological evaluation of isolated and synthetic chloro derivatives of dibenzo-p-dioxin. Nature 220: 702 (1969). 2. Kimbrough, R. D. Toxicity of chlorinated hydro carbons and related compounds. Arch. Environ. Health. 25^425 (1972). 3. Bauer, H., Schulz, K. H., and Spiegelberg, U. Berufliche Vergiftungen bei der Herstellung von Chlophenol-Verbindungen. Arch. Gewerbepathol. Gewerbehyg. 18: 538 (1961). 4. Vos, J. G., and Koeman, J. H. Comparative tox icologic study with polychlorinated biphenyls in chicks with special reference to porphyria, edema formation, liver necrosis and tissue resi dues. Toxicol. Appl PharmacoL 17: 656 (1970). 5. Cherrick, G. R., et al. Indocyanine green: ob servations on its physical properties, plasma decay and hepatic excretion. J. Clin. Invest. 39: 592 (1960). 6. Klaassen, C. D. Biliary flow after microsomal enzyme induction. J. Pharmacol. Exptl. Therap. 168: 213 (1969), September 1973 231 784567 0 w 1 *-0 Biological Responses of the Nonhuman Primate, Chicken, and Rat to Chlorinated Dibenzo-p-dioxin Ingestion by D.H. NorbackT and J.R. Allenf In 1958 Schmittle et al. (1) reported the development of hydropericardium and as cites in poultry following ingestion of feeds containing industrially contaminated fats. The toxic component was demonstrated to have the chlorinated dibenzo-p-dioxin /ODD) structure by Cantrell et al. in 1969 ). Experimental animal studies in our moratory have shown that CDD adminis tration causes varying responses in different animal species. Gastric hyperplasia and ul ceration, hydropericardium, ascites, reduced spermatogenesis, focal liver necrosis, de creased hematopoiesis, skin lesions, and -eventual mortality have been demonstrated in nonhuman primates (3). Chickens suc cumbed very rapidly to the same dietary concentration with hydropericardium, hy drothorax, and ascites. They also developed liver necrosis, hypoplastic testes, and al tered capillary permeability and decreased hematopoiesis (4-6). The rat was more re sistant to the morbid effects of CDD but de veloped a hypertrophied liver composed of enlarged hepatocytes with a proliferated in- *This investigation was supported in part by U.S. Public Health Service grants ES-00472 and RR00167 from the National Institutes of Health. Pri mate Center Publication No. 13-010. tDepartment of Pathology, University of Wiscon sin Medical School, and Regional Primate Research Center, University of Wisconsin, Madison, Wisconsin "T06. tracellular membrane system (7). The re sults of these investigations are reviewed and the variable responses and pathogenesis of lesions are discussed. The material used in the investigations was crude industrial fat capable of produc ing hydropericardium, ascites, and death in the chicken. Gas-liquid chromatographic and nuclear magnetic resonance analysis of the materials demonstrated bi-, tri-, tetra-, penta-, hexa-, and heptachlorodibenzo-pdioxin present in the material, with the tetrachlorinated compound comprising 64 % (mass) of the total dioxins present. In ad dition, rats were given radioactive octachloro- or radioactive tetrachlorodibenzo-pdipxin in separate experiments (8, 9). Macaco, rmilatta monkeys were given a diet that contained varying quantities of the crude industrial fat. TTie percentage of fat that allowed survival of the nonhuman pri mate for 100 days produced 50% mortality in chickens within 15 days. The survival time of the monkeys was inversely related to the percentage of the CDD-containing fat given to the animals. Clinical and patho logic changes occurring at their demise were similar regardless of concentration of ma terial in their diet. At death all had devel oped ascites, hydropericardium, and anasar ca. Prior to their demise the monkeys de veloped a decrease in total serum protein oeptember 1973 233 784568 U -\ <1 from 7.5 to 5.4 g/100 ml and a decrease in th percentage albumin from 61% to -35%. There:,was a decrease in hematocrit from 41% fo'hnd in the control animals to 16% in the experimental group, a decrease in the white blood count from 6.8 x 103 per mm3 to 3.0 x 103i and1a decrease in the red cell count from 6.5 x 10s per mm3 to 2.5 x 109. The hemoglobin values were correspondingly reduced. Analysis of the sternal bone mar row showed a hypoplastic bone marrow with diminished myeloid and erythroid cells being replaced by fatty tissue. The lymphoid tissue of the spleen and lymph nodes was hypoplastic. The skin changes of the monkeys included alopecia and subcutaneous edema which pro gressed from the eyelids to the remainder of the face, eventually involving the sub cutaneous tissue of the trunk, extremities, and scrotum. Microscopically, there was edema of the dermal layer with disarray of the collagen fitters. Hair follicles, particularly of the face, contained numerous keratin cysts with hyperplasia of the epithelium (Fig. 1). cells. However, there was a decreased num ber of primary and secondary spermato cytes, and spermatids were inapparent in most instances (Fig. 2). The interstitial cells of Leydig appeared normal. Figure 2. In a monkey fed CDD, seminiferous tub ules of testes contained decreased numbers of pri mary and secondary spermatocytes without sper matids. Spermatogonia, Sertoli cells, and inter stitial cells were normal in appearance. Light micrograph of testicular tissue fixed with formalin and stained with hematoxylin and eosin. x llS . Figure 1. Hair follicles, particularly of the face and eyelids, of monkeys fed CDD contained numerous keratin cysts. Light micrograph of skin fixed with formalin and stained with hematoxylin and eosin. X15. The seminiferous tubules of the testes con tained abundant spermatogonia and Sertoli 234 Figure 3. Cardiac fibers in the hearts of monkeys fed CDD were hypertrophied, and the myofila ments were widely separated by the increased intracellular fluid. Light micrograph of heart fixed with formalin and stained with hematoxylin and eosin. X115. Environmental Health Perspectives 784569 GE1stP 011758 The heart was dilated, particularly in e right chamber-, and an increase in the circumference of both the tricuspid and mi tral valves was noted. Microscopically, hy pertrophic muscle fibers were separated by fluid (Fig. 3). Electron microscopic exam ination demonstrated separation of the myo fibrils and swelling of the mitochondria with widely separated cristae (Fig. 4). In over 60 % of the experimental monkeys, marked hypertrophy of the gastric mucosa occurred in the fundic and pyloric regions. The hypertrophied mucosal layer penetrated the muscularis mucosae to form crypts and mucin-containing cysts in the submucosa (Fig. 5). In the same areas, gastric ulcera tions of the mucosal layer were present (Fig. 6). The livers were moderately yellow. On microscopic examination, enlarged multinucleated hepatocytes and fat vacuoles were apparent. Terminally the animals developed centrilobular necrosis and bile duct hyper plasia. Changes in the biliary tree were ob served as proliferation and stratification of the bile duct epithelial cells of the small ducts within the portal area and in the larg er ducts, including the common bile duct running through the head of the pancreas (Fig. 7). Electron microscopic examination of the hepatocytes demonstrated hypertro phied cells with numerous autophagosomes and fat droplets, an increase in the smooth endoplasmic reticulum with a decrease in the rough endoplasmic reticulum, and swol len mitochondria (Fig. 8). In the morbid animals the parenchymal cells showed num erous degenerative changes. Many of the cells i r s *, a > 0 mm. -T*-' .A- .--. -- ga>s V f, a te. 2VS Figure 4. Myofibrils of dilated cardiac fibers within the heart of a monkey fed CDD were separated, and the mitochondria were moderately swollen. Electron micrograph of heart fixed with Veronal acetatebuffered osmium tetroxide solution and stained with uranyl acetate. X9,700. September 1973 235 784570 GENP 011759 09IIO rTNmr% Figure 7. Following CDD ingestion by the monkeys, bile duct hyperplasia occurred within the liver. Tall columnar epithelial cells, many of which ap peared stratified, replaced cuboidal cells found in the normal bile duct and epithelial folds extended into' the'lumen.-Light micrographs of liver tissue fixed in formalin and stained with hematoxylin and eosin. X115. September 1973 Figure 9. Light and dark staining ceils were present in the livers of chickens fed CDD. Numerous fat droplets infiltrated the hepatocytes. Light micro graph of liver tissue fixed in Veronal acetatebuffered osmium tetroxide solution and stained with toluidine blue. X610. 237 784572 r: G E N P O im i niferous tubules within the testes were nor mal; however, there was a reduction' ftrthe number of primary and secondary sperma tocytes, and no spermatozoa were present. Gastrointestinal changes were not observed. The perfusion of the mesenteric vessels with ferritin, thorium dioxide, iron oxide, and car bon black demonstrated a decided alteration in capillary permeability of these experi mental animals. Following administration of approximate ly five times the concentration of the CDDcontaining fat in the diet sufficient to cause hydropericardium, ascites, and focal necrosis of the liver in chickens and in nonhuman primates, the rat developed liver alterations with a 50% mortality at 80 days. At 6 weeks, enlarged livers contained hypertrophied hp atocytes with an increase in droplets and a higher quantity of extractable lipid. The histologic pattern of the livers consisting of sinusoids separated by single sheets of hepatocytes radiating from the portal areas to the central veins was maintained. Within the large hepatocytes a prolifera tion of the smooth endoplasmic reticulum and reorganization of the parallel cisternae of the rough endoplasmic reticulum to form large agranular concentric membrane arrays was demonstrated electron microscopically (Fig. 10). F igure 10. Hepatocytes of rats which ingested CDD developed a proliferated smooth endoplasmic retic ulum consisting of numerous vesicles and concentric arrays of agranular membranes. The number of lipid droplets was increased. Electron micrograph of liver tissue fixed in Veronal acetate-bnfferec os mium tetroxide solution and stained with uranyl acetate, x-20,400. 238 Environm ental Health Perspectives 784573 c t / Administration- of tetrachlorodibenzop-dioxin (1 /iff/day) resulted in a 50% mor tality of the rats at 21 days. The morpho logic appearance of the livers was similar to that produced by ingestion of the crude CDD-containing lipid material by rats. The smooth endoplasmic reticulum was prolifer ated, and large concentric membrane arrays were present. Over a 21-day period of administration of labeled octachlorodibenzo-p-dioxin to rats (100 /ig/rat/day, approximately 12.4 mg/kg administered over 21 days), 93% of the com pound passed unabsorbed through the gas trointestinal tract. An additional 5% was excreted in a lipid-soluble form in the urine. The administration of the octachloro compound produced few morbid alterations in the rats. The animals continued to gain weight and maintained normal activities and gross appearance. Approximately 50% of the material present within the body tis sues was located in the liver. Other reser voirs containing radioactivity at lesser lev e l s were the adipose tissue, skeletal muscula t u r e , and skin. Over 95% of the CDD pres ent within the liver was located in the mi crosomal fraction with equal distribution within the rough and smooth fractions. D iscu ssio n The data from these experimental animal studies have shown that CDD administra tion causes varying responses in the chicken, monkey, and rat. The chicken develops ex treme morbidity and mortality at dietary concentrations that are only mildly toxic to rats while the monkey is intermediate in its response to the CDD. The chicken and monkey developed ascites, hydrothorax, hydropericardium, and anas arca; however, the ra t failed to develop in creased extracellular fluid. These modifica tions in the fluid content of the tissues and body cavities were attributed in part to hep atic degenerative changes and altered cap illary permeability of the chicken and monkey. The low serum albumin, a direct result of hepatic dysfunction, was associ ated with decreased osmolarity of the blood and. subsequent extravasation of the fluid. In addition, the capillaries were dem onstrated to be more permeable to colloidal particles before the decline in serum protein was sufficiently severe to produce an accu mulation of fluid in the tissues. Gastric hyperplasia and ulceration were limited to the nonhuman primate. Hyper plastic changes are thought to be related to the chronic irritation following ingestion of the compounds and other closely related chlorinated aromatic hydrocarbons (10). The dysplastic histologic and cytologic pat tern as demonstrated by the invasion of the mucosal cells' through the muscularis muc osae and the stratification of the epithelial cells within the cysts are changes suggestive of an eventual neoplastic transformation. Monkeys developed widespread alopecia, moderate hyperkeratosis, follicular keratin cysts, and hyperplasia of the epithelium of the hair follicles, particularly of the face. However, the skin of rats or chickens was not altered appreciably. Hypoplasia of the lymph tissue and bone marrow was present in ail three animal spe cies. However, the blood-forming tissues of the chicken and the monkey were affected earlier and more severely than were those of the rat. As a result of these changes in the lymph tissue and bone marrow, the ani mals became anemic and displayed a pro gressive leukopenia. Due to the reduced re sistance of these animals they became prime hosts for opportune pathogens which in many instances were responsible for their death. Hypoactivity of the seminiferous tubules of the testes was associated with chronic intoxication of the monkeys and chickens. Young chickens exposed to low levels of thedioxins experienced retardation in the de velopment of the testes and at maturity were of normal size with hypoplastic gon ads. There were no other alterations in growth, blood elements, or histologic ap pearance of the tissues. Enlargement of the liver occurred in all animals used in these investigations. In creased size was related to the cell hyper- September 1973 239 784574 GENp oi 1763 I trophy resulting from proliferation of the smooth endoplasmic reticulum and accumu lation of lipid within the cytoplasm of the hepatocytes. The chicken rapidly developed widespread liver necrosis; similar degenera tive changes occurred at a less rapid rate in the liver of the monkey, and the ra t was very resistant to hepatic necrosis. Radioactive studies which determined tis sue and cell fraction levels of the CDD in the ra t and possibly in other animal species demonstrated the proliferated hepatic en doplasmic reticulum present in animals fol lowing exposure to the CDD may serve not only as a source of enzymes to enhance the metabolism of foreign substances but may "also function as an area of localization for these toxic compounds. The presence of a large portion of the ingested CDD within the microsomal fraction of the hepatic tis sue may be one explanation as to why the ra t is able to tolerate larger doses of the dioxins. The localization of the dioxins in these membranes may prevent their move ment to other tissues of the body that are more susceptible to the toxic effects of these compounds. Although the specific reason for the difference in response of various animal species to the dioxins has not been estab lished, further studies on absorption, meta bolism, body distribution, excretion, or sen sitivity of the tissues to the toxic effects of the dioxins are avenues of research that will likely clarify these questions. REFERENCES 1. Schmittle, S. C., Edwards, H. M., and Morris, D. A disorder of chickens probably due to a toxic feed--preliminary report. J. Amer. Vet Med Assn. 132*, 216 (1958). 2. Cantrell, J. S., Webb, N. C., and Mabis, A. J. The identification and crystal structure of a hydropericardium producing factor: 1,2,3,7,8,9-hexachlorodibenzo-p-dioxin. Acta Cryst. B25: 150 (1969). 3. Allen, J. R., and Carstens, L. A. Light and elec tron microscopic observations in Macaco, mulatto monkeys fed toxic fat. Amer. J. Vet. Rea. 28; 1513 (1967). 4. Allen, J. R. The role of toxic fat in the produc tion of hydropericardium and ascites in chickens. Amer. J. Vet. Res. 25; 1210 (1964). 5. Allen, J. R., and Carstens, L. A. Electron micro scopic alterations in the liver of chickens fed toxic fat. Lab. Invest. 15; 970 (1966). 6. Allen, J. R., and Lalich, J. J. The effects of "toxic fat" on spermatogenesis. Froc. Soc. Exp. Biol. Med. 109; 48 (1962). 7. Norback, D. H., and Allen, J. R. Morphogenesis of the toxic fat-induced concentric membrane arrays in rat hepatocytes. Lab. Invest. 20; 33S (1969). 8. Norback, D. H. Morphological and biochemical responses of the rat hepatic endoplasmic re ticulum to polychlorinated triphenyls and to chlorinated bibenzo-p-dioxins. Ph.D. disserta tion, University of Wisconsin, Madison, Wise., August 1973; Dissertation Abstr., in press. 9. Norback, D. H., and Engblom, J. F. Chlorinated dibenzo-p-dioxin distribution within rat tissue and subfractions of the liver. Fed. Froc. 32; 236 (1973). 10. Allen, J. R., and Norback, D. H. Polychlorinated biphenyl- and triphenyl-induced gastric mucosal hyperplasia in primates. Science 179; 498 (1973). GENP 011764 240 Environmental Health Perspective 784575 L Excretion and Tissue Distribution of 2,3,7,8-Tetrachiorodibenzo-p-dioxin in the Rat by Walter N. Piper," James Q. Rose/ and Perry J. Gehring7 The compound, 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD), is highly toxic. The LDS0 for male and female rats given a single oral dose is 23 and 45 /xg/kg, respectively (1). Adverse effects have been observed in a ratology study in which pregnant rats were iiven oral doses of 0.125-2.0 /ig/kg-day TCDD from day 6 through day 15 of gestation (2). The adverse effects were increased fetal mortality, early and late resorptions and intestinal hemorrhage in the fetuses. No ad verse effects were noted at the 0.03 /*g/kgday level. In humans and rabbits, contamination of the skin with TCDD produces chloracnelike lesions (3, 4). This disease is character ized by the appearance of hyperkeratosis, papules, comedones and cysts. There is no available information on the absorption, excretion or tissue distribution of TCDD in animals. Therefore, this study was done to determine the excretion and tissue distribution of radioactivity derived from TCDD-UC following a single oral dose of the labeled compound. "Oakdale Toxicology Center, Department of Phar macology, University of Iowa, Iowa City, Iowa 12240. tToxicology Unit, Chemical Biology Research, The low Chemical Company, Midland, Michigan 48640. M eth od s Animals Male Spartan strain Sprague-Dawley rats weighing 165-210 g were used. The rats were acclimated to the environment of the metabolism cages 5 days prior to dosage. Food and water were provided ad libitum throughout the experiment. 1`C-Tetrachlorodibenzo-p-dioxin Uniformly labeled TCDD-l4C was synthe sized at the Radiochemistry Research Lab oratory of The Dow Chemical Company. The specific activity was 2.8 /iCi/mg. Mass spectrometric and gs-liquid chromatogra phic analyses of the TCDD" C sample indi cated a purity of 93.3 and 95.0 respec tively. Dosage TCDD was dissolved in acetone. Subse quently, one part of the acetone solution was added to and mixed with nine parts of USP corn oil. The acetone-corn oil solu tion of TCDD was given to rats in 5 m l/kg amounts by intubation. This volume of the solution provided a dose of 50 p.%/kg and 0.14 /tCi/kg TCDD-14C. Sample Collection A fter administering the solution contain ing TCDD-^C, the rats were placed in all- September 1973 241 784576 S91U0 dNSO ..g la ss Roth metabolism chambers which were equipped for separate collection of urine, feces, and expired air. The C02 in the exiting air stream was trapped by bub bling it through a 3:7 ethanolamine-2methoxyethanol mixture. Radioactivity Analysis Samples of urine, feces, and tissues were oxidized to " CO,, by combustion and analyzed for radioactivity (5). By using this method, the recovery of radioactivity from samples spiked with 14C was 95 5%. To determine the radioactivity expired as CQ-, 5-ml ali quots of the solution used to trap the C 02 were added to 15 ml of a scintillation count ing solution containing 4 g 2,5-diphenyloxazole (PPO) and 0.1 g l,4-bis-2-(5-phenyloxazolyl) benzene (POPOP) per liter of 1:1 toluene-2-methoxyethanol. Samples were counted for radioactivity in a Nuclear Chi cago Mark II liquid scintillation counter, and quench corrections were made by use of the internal standard technique. R esu lts The percentage of the total dose of radi oactivity excreted daily in the feces, urine, or expired air over a 21-day period follow ing a single oral dose of TCDD-l4C is shown in Figure 1. Approximately 30% of the " C activity was excreted in the feces dur ing the first 48 hr. Most of this probably represents unabsorbed TCDD-" C. Over the remaining 19 days, 1-2% per day of the " C activity was excreted in the feces. A total of 53.2 3.8% of the administered dose was excreted via the feces over the 21-day period. The total cumulative amount excreted in the urine and expired air was 13.2 1.3 % and 3.2 0.1%, respectively. To determine the overall rate of clear ance of 14C administered as TCDD from the body, the total cumulative amount of 14C excreted in feces, urine, and expired air at the end of each day was subtracted from the total dose administered to the ani mal. These values, representing the percen tage of the total dose remaining in the ani mal at the end of each day, were then plotted 242 F igure 1. Excretion of "C activity by rata following a single oral dose of 50 ng/kg (0.14 ,uCi/kg) 2,3,7,8tetrachlorodibenzo-p-dioxin. Each point represents the mean SE for three rats. F igure 2. Clearance of ,4C activity from the body of rats given a single oral dose of 50 jig/kg (0.14 fiCi/kg) 2,3,7,8-tetrachIorodibenzo-p-dioxin. Each value represents the mean SE for three rats. semilogarithmically as a function of time (Fig. 2). Except for the first 2 days fol lowing administration, the clearance of l4C activity from the body followed a p p a r e n t first-order rate kinetics. The half-life for Environmental Health Perspectives 0 0 / T TO JK T T T r\ 784577 learance, tu., was 17.4 5.6 days. As pre^ - riously indicated; it was assumed that the relatively large amount excreted during the first 2 days had not been absorbed. There fore, these values were not used in calculat ing the clearance rate. Analyses of tissues indicated that the 14C activity derived from TCDD and/or its breakdown products was located chiefly in the liver and fat (Table 1). The percentage of the dose per gram of liver 3, 7, and 21 days following administration was 3.18, 4.49 and 1.33 % /g, respectively. Comparable values for fa t were 2.60, 3.22, and 0.43 %/g. Smaller concentrations of 14C activity were found in other tissues: muscle, testes, lungs, heart, skin, spleen, stomach, pancreas, brain, bone, kidneys, and adrenals (Table 2). Standard errors as large as the mean suggest that some of the values presented in Table 2 may be a result of experimental error: adrenals, 3 days; bone, 7 days; spleen, 21 days; pancreas, 21 days. Radioactivity ex ceeding background in these tissues at the . .. indicated time was detected in only one of three rats. The value given in Table 2 for adrenals 21 days following administration suggests that this tissue may concentrate TCDD-14C and/or a metabolite. This observation is very likely due to experimental error. The disintegrations per minute (dpm) above background for this tissue were only 17, 29, and 90. Since the total amount of tissue avail able for analysis was less than 20 mg, the multiplication factor may have magnified the error manyfold. Total recovery of the administered 14C activity was determined for those rats used in the 21-day experiment. The I4C activity remaining in the unused carcass was de termined by analyzing an aliquot of a homo genate of the remaining carcass. The recov ery was 96.8 3.0%. Discussion In the study reported herein, the tissue distribution and excretion of 14C activity have been evaluated in rats following a single oral dose of TCDD-14C. Almost 30% of the Table 1. "C activity expressed as percentage of dose per gram in the liver and fat of rats 3, 7, and 21 days following-a single oral dose of TCDD-*C.8 Activity, % /g 1> Tissue Liver Fat 3 days 3.180.21 (47% ) 8 2.600.48 7 days 4.490.62 (45%)" 3.220.63 21 days 1.33 0.70 (11%) 0.43 d ` Dose: 50 ,ugAg (0.14 f t d /k g ) ; 3 rata/observation. bMean standard error. *Percentage of the total dose found in the entire liver. dMean for 2 rats. Table 2. "C activity expressed as percentage of dose per gram in various tissues of rats 3, 7, and 21 days following a single oral dose of TCDD--``C.* Tissue Muscle Testes Lungs Heart Skin Spleen Stomach Pancreas Brain Bone Kidneys Adrenals Activity, 9c/g" 3 days 7 days 21 days 0.380.01 0.380.03 0.270.02 0.200.03 0.19 0.10 0.150.02 0.160.05 0.110.06 0.06 0.00 0.090.03 0.00 d 0.790.79 0.210.05 0.360.10 0.39 0.14 0.40 0.16 0.190.10 0.950.53 0.10 0.00 0.16 8 0.13 0.09 0.420.42 0.340.17 0.02 0.02 0.20 0.12 0.110.09 0.06 0.05 0.09 0.05 0.090.04 0.22 0.22 0.02 0.02 0.160.16 0.01 0.01 0.080.08 0.00 d 3.69 1.77* ` Dose: 50 tgAs (0.14 nCi/kg) ; 3 rats/observa tion. bMean standard error. *Mean of 2 rats. dNo activity above background in all 3 rats. `This large value may be an experimental error. The activity above background for the adrenals of the three rats was 17, 29, and 90 dpm. Since the total amount of tissue was less than 20 mg, the multiplication factor may have magnified the error manyfold. dose administered was eliminated via the feces during the first 48 h r following treat ment. The excretion of 14C activity via the feces after the first 48 h r ranged from 1 to 2% /day. It appears th at TCDD is incom pletely absorbed from the gastrointestinal tract. The 14C activity derived from the ab- September 1973 243 GENP 011767 784578 4 V sorbed TCDD-l4C also is excreted mainly (1). With doses that do not induce untoward -via the feces. effects, the compound may be excreted at a Once absorbed in the body, most of the different rate. UC activity derived from TCDD-l4C is local The results do not differentiate between ized in the liver and fat. The data suggest 14C activity derived from TCDD and that of th a t the level in these tissues is approxi possible metabolites. However, small amounts mately 10-fold that in other tissues. The of 14C activity were detected in the expired 14C level in liver and fat seemed to increase air and urine within the first 10 days follow between 3 and 7 days following administra ing administration. This is evidence that tion; however, the 14G activity in liver and some metabolic alteration or breakdown of fa t decreased more between 7 and 21 days TCDD occurs. than would have been predicted on assum ing that the rate of clearance from these tissues would be equal to the rate of clear ance from the body. Between days 7 and 21, the I4C level in muscle remained essen tially unchanged. Therefore, redistribution Acknowledgement The authors thank Mr. W. W. Muelder for preparation of 2,3,7,8-tetrachlorodibenzo-pdioxin-u C. of TCDD or metabolites of TCDD may have REFERENCES been occurring. The apparently high level in the adrenals 21 days after administra tion results probably from experimental er ror. The dose of TODD given to the male rats 1. Schwetz, B. A., et al. Toxicology of chlorinated dibenzo-p-dioxins. In: Advances in Chemistry Series 120, American Chemical Society, Wash ington, D.C., 1973, p. 55. 2. Sparschu, G. L., Dunn, F. L., and Rowe, V. K. Study of the teratogenicity of 2,3,7,3-tetracnloro- used in this study, 50 pg/kg, was approxi mately twice the LDS0 (23 pg/kg). This large dose was necessitated because of the specific activity of the TCDD-14C used. Rats lost weight, and their physical condition was dibenzo-p-dioxin in the rat. Food Cosmet. Tox icol. 9: 405 (1971). 3. Kimmig, J., and Schulz, K. H. Occupational chloracne caused by aromatic cyclic ethers. Dermatologia 115: 540 (1957). 4. Jones, E. L., and Krizek, H. A technic for testing poor; this typifies the insidious response to acnegenic potency in rabbits, applied to the TCDD (i). Survival of the rats for 21 days was not totally unexpected, because in pre vious studies on the lethality of TCDD deaths frequently occurred 20 days or more follow ing a single oral dose of similar magnitude potent acnegen, 2,3,7,8-tetrachlorodibenzo-pdioxin. J. Invest. Dermatol. 39: 511 (1962). 5. Smith, G. N., et al. Simple apparatus for com bustion of samples containing (TMabeled pesti cides for residue analyses. Agr. Food Chem. 12: 172 (1964). G E N P011768 244 Environmental Health Perspectives 784579 Studies on the Mechanism of Toxicity of the Chlorinated Dibenzo-p-dioxins* by Alan Poland1, and Edward Glover1 In trod u ction Concern about the potential health haz ards resulting from environmental pollution by the chlorinated dibenzo-p-dioxins and dibenzofurans arises from our recognition of the extraordinary potency of these com pounds as toxins and teratogens and their inadvertant dispersion in the environment as contaminants of chlorinated phenolic prod u c t s . Questions concerning the extent of en} ronment contamination and those concern i n g the mechanism of toxicity produced by these compounds are at present unanswer able. Several papers in this symposium have presented the historical background which led to our current understanding and concern about this problem: (1) the "chick-edema" outbreaks caused by "toxic fats" in poultry feed and the eventual isolation and identi fication of a hexachlorinated dibenzo-p-dioxin; (2) the occurrence of acne among work ers in several 2,4,5-T factories and recogni tion of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) as the tiologie agent; (3) the NCIcommissioned study on the potential terato genicity, carcinogenicity and mutagenicity of 2,4,5-T; (4) and the widespread use of Agent ^Supported by: NIH Special Postdoctoral Fellow ship, 5 F03ES 46196; Ford Motor Company Grant for Toxicology; and NIH Center Grant for Toxicology Research and Training, 2P11-GM 15190-06AT. ^Department of Pharmacology and Toxicology, University of Rochester School of Medicine and v Dentistry, Rochester, New York 14642. Orange as a defoliant in Viet Nam and Cam bodia. The starting point of our studies was a re port by Bleiberg and colleagues (1) th at of 29 workers in a 2,4,5-T factory, all of whom had industrially acquired acne, 11 had por phyrinuria and several had overt clinical porphyria cutanea tarda. TCDD has been shown to be the causative agent of the acne; however, the cause of the porphyria was un certain. Porphyria cutanea tarda is an ac quired defect of hepatic porphyrin metabo lism characterized by an overproduction of porphyrins by the liver, increased urinary ex cretion of porphyrins, mechanical fragility and photosensitivity of the skin (blistering in areas exposed to sunlight), hyperpigmen tation, and hirusitism. We restudied the fac tory 5 years later and found no evidence of porphyria in the employees (2). The fact that this syndrome abated following meas ures to reduce the formation TCDD and mini mize employee exposure to this contaminant, suggested that TCDD might have been the causative agent of the industrial outbreak originally reported by Bleiberg et al. (1). Methods and Materials The halogenated dibenzo-p-dioxins and dibenzofurans and analyses of their purity were generously provided by Dr. A. Fohland, Food and Drug Administration, Washington, D. C. and Mr. George Lynn, Dow Chemical Company, Midland, Michigan. In addition, Drs. J. Wade and A. Kende synthesized .September 1973 245 784580 and analyzed a number of dibenzo-p-dioxins which were tested and reported elsewhere in this symposium. Animals Our experiments were performed in chick embryos which were routinely 15-20 days of age. The various dibenzo-p-dioxins were dissolved in p-dioxane, and 25 pi of the solu tion was injected into the egg through a small hole punched in the air sac. Male Sprague-Dawley rats, 70-100 g, were used in some experiments. Analysis S-Aminolevulinic acid synthetase acti vity was measured as previously reported (3). Aryl hydrocarbon hydroxylase was as-' sayed essentially by the method of Gielen, Goujon, and Nebert (4). One unit of hydroxy lase activity is defined as that amount of enzyme catalyzing the formation per minute at 37C of hydroxylated product causing fluorescence equivalent to that of 1 pmole of 3-hydroxybenzo [a] pyrene. The assay was performed on the 10,000p supernatant, and results are expressed as units per milligram wet weight of liver. 300 F igure 1. Logarithmic dose-response curve for the induction of ALA synthetase by TCDD. Chicken eggs of 17 days' gestation were injected with 25 /il of solvent (control) or solvent containing vari ous doses of TCDD, and hepatic enzyme activity was assayed 48 hr later. The points represent the mean standard error of three or four groups of pooled livers. reflection of the long biological half-life of TCDD. We next screened a series of 15 halogenated dibenzo-p-dioxins for their abil ity to induce ALA synthetase. As seen in Fig ure 2, all the isomers which were inducers A lA STNIKETASE ACTIVITY (<itwaatiALA/qabr/ha*t R esu lts A variety of xenobiotics produce experi mental hepatic porphyria and all have in com mon the ability to induce the initial and rate limiting enzyme in the heme biosynthesis pathway, 3-aminolevulinic acid synthetase (ALA synthetase). To test whether TCDD was in fact porphyrigenic, we administered the compound dissolved in 25pi p-dioxane to chick embryos. The embryos were sacrificed 48 h r later, and the ALA synthetase activity assayed in their livers. As seen in Figure 1, TCDD produced a dose-related Increase in enzyme activity. As little as 4.66 x 10-12mole egg (1.5 ng) produced a doubling enzyme activity and the highest level tested 1.55 x 10*" mole/egg (0.5 pg) produced a 35-fold induction. TCDD is more potent than any other inducer of ALA synthetase yet report ed by at least three orders of magnitude, and, unlike most other porphyrigenic chemicals, induction is very prolonged, most likely a 246 F igure 2. Structure-activity relationships of the halogenated dibsnzo-p-dioxins: induction of ALA synthetase. Seventeen-day embryos were injected with the solvent (p-dioxane) or solvent containing the dioxin tested and enzyme activity wa3 assayed 48 hr later. Uninjected control values (n = 9), do not differ appreciably from solvent injected controls (n = 12). Each bar for the test groups represents the value for a single group of three to five pooled livers. Environmental Health Perspectives 0LL110rTNF>Tr> 784581 ;iad two common properties: (1) halogen *toms occupy a t least three of the four lateral ring positions (2, 3, 7, and 8) and (2) there is at least one free, nonhalogenated carbon atom. Note that octachlorodibenzo-p-dioxin is inactive. The compounds that were not inducers were tested at 200-400 times the molar concentration of TCDD that produced a significant response. To the extent th at toxicologic data are available (5), all those dioxins which are a t low doses lethal, teratogenic, or produce acne, also induce ALA synthetase, and those dioxins which are not potent toxins do not induce ALA synthetase. We have also test ed a limited series of dibenzofurans; the unsubstituted compound and 2,8-dichloro and octachloro derivatives all fail to induce, and a mixture of di- tri- and tetrachlorodibenzofurans is potent as an inducer of ALA synthetase. While the data are very limited, it appears the structure-activity relation ship is similar in the dibenzofiiran series. As reported elsewhere in this symposium (d), TCDD does not induce ALA synthetase \in several laboratory mammals. We have Ubund it to be a poor inducer in the rat. This should not be interpreted to mean that the results obtained in the avian embryo have no relevance to man. For instance, many sex steroids appear to play a role in precipitating acute interm ittent prophyria and porphyria cutanea tarda in man; how ever, while induction of ALA synthetase by these compounds can be shown in the chick embryo, it does not occur in the rat. There is an empiric relationship observed by numerous investigators that many com pounds which induce ALA synthetase also induce microsomal mixed-function oxygen ase activity in the liver (also called the "drug metabolizing enzymes" ). Two points are of note about this correlation: not all drugs in duce both enzyme activities; also, the rela tionship many have a theoretical basis, in that heme is the prosthetic group of the ter minal component of microsomal oxygenase, cytochrome P-450. The high concentration of cytochrome P-450 and rapid turnover relative to all other hepatic hemoproteins, accounts for a large fraction of the total heme synthesized in the control liver. Some investigators have suggested th at coordinate induction of ALA synthetase, the rate- lim iting step in heme synthesis, and cytochrome P-450 and microsomal oxygenase activity may have a basis in providing the extra heme necessary for forming the new cyto chrome P-450. Despite this not very satisfy ing teleologic explanation, there is a sizable literature reporting the concommitant induc tion of ALA synthetase and microsomal ox ygenase activity. We next studied the effect of TCDD on microsomal oxygenase. As a measure of this enzyme complex we choose to investigate aryl hydrocarbon hydroxylase activity, because aromatic hydroxylation is induced primarily by aromatic compounds, which in our view chemically resemble TCDD. As seen in Figure 3, TCDD produces a doserelated induction of aryl hydrocarbon hydro xylase in chick embryo liver. At the lowest dose tested, 1.55 x 10_l mole/egg (0.5 ng) there is a nearly twofold increase in enzyme activity, and maximal induction is produced TCDD (moles/egg) F ig u h e 3. Logarithmic dose-response curve for the induction of aryl hydrocarbon hydroxylase. Eighteen-day embryos were injected with TCDD dissolved in p-dioxane or p-dioxane alone (control), and hepatic enzyme activity was assayed 24 hr later. Each point represents the mean standard error of four groups of pooled livers. September 1973 247 784582 by 1.55 X 10~l mole/egg (50 ng). Following - the administration of TCDD to an egg, hepat ic aryl hydrocarbon hydroxylase activity rises to reach a maximum at about IS hr, and then the elevated hydroxylase activity presists for at'least 5 days. We screened 15 halogenated dioxins for their ability to induce aryl hydrocarbon hy droxylase at three dose levels: 4.7, 47, and 470 x 10" mole/egg (Fig. 4). The structureactivity relationship is identical to that seen with the induction of ALA synthetase: (1) the compounds which are potent inducers (induction at 4.7 or 47 x 10-11 mole/egg) have halogen atoms at least three of the four lateral ring positions and (2) they have at least one nonhalogenated ring position. There i one exception, the 1,2,4,6,7,9-hexachloro- dioxin, which at high doses (470 x 10'11 mole/egg) produced a modest induction of aryl hydrocarbon hydroxylase. This com pound was only 90 % pure by gas-liquid chro matography, and the induction observed at the highest dose could be produced by con tamination with as little as 0.1% (w/w) TCDD or an equipotent dioxin. Clarification must await the availability of a purer pre paration of this hexachloro isomer. There are certain advantages to investigat ing the induction of hydroxylase activity produced by TCDD in the rat, namely: one can more fully examine the duration of in duction, and also the spectral changes in cytochrome P--450 accompanying aryl hydro carbon hydroxylase induction are more fully documented in the rat. The administration ' DOSE AHH ACTIVITY (M O U -IO ^ EG G ) (UMTS/mq LIVES) 024 <70 3 DOSE AHHACTIVITY (M O L E > IO % 1 (U N |Ty*g UVGl) 0 2 44 q ` 470 81 10 12 470 470 47 47 47 47 47 47 47 47 3- <JjN jP 011772 Figure 4. Structure-activity relationship of halogenated dibenzo-p-dioxins: induction of aryl hydrocarbon hydroxylase. Eighteen-day embryos were injected with 25 m) of p-dioxane or p-dioxane containing the test dioxin, and enzyme activity was assayed 24 hr later. Each bar represents the mean standard error of four groups of pooled livers, except the control and TCDD groups where n = 12. Environmental Health Perspectives 784583 ' TCDD a t a dose of 3.11 x 10-10 mole/kg duces nearly a* fivefold induction in rat liver enzyme activity (Fig. 5). Maximal in duction of aryl hydrocarbon hydroxylase was produced by a dose of 3.11 x 10-3 mole/kg (10 /ig/kg). It is estimated that the half maximal response is elicited by a dose of 8.5 x 10-10 mole/kg (0.265 /ig/kg), roughly one hundredth the dose th at kills 50% of rats (5). 3-MethylchoIanthrene, a polycylic hydro carbon carcinogen is perhaps the most widely used compound as an inducer of aryl hydro carbon hydroxylase. TCDD was found to be nearly 3 X 104 times as potent as 3methylcholanthrene at inducing hydroxylase activity. Both drugs produce the same maxi mal degree of enzyme induction in ra t liver, and the administration of both drugs to gether, each at a dose which produces maxi mal induction, elicits a response that is no greater than that produced by either drug alone. In Figure 6 we have plotted the results of the time, course of induction produced by 3-methylcholanthrene (20 m g/kg in corn oil), TCDD (10 /ig/kg in p-dioxine) and control rats (receiving either corn oil or pdioxane). Aryl hydrocarbon hydroxylase is induced to the same extent at 1 and 4 days by maximally inducing doses of each drug. However, by 8 days, hydroxylase activity in the 3-methyichoIanthrene-treated rats re turned to control levels, while induction per sisted in the TCDD-treated rats for over 1 month. Accompanying the induction of aryl hydrocarbon hydroxylase by both drugs the following changes were noted: an increase in the total CO-binding microsomal cytochrome (cytochrome P-450 and P-448), a shift in the CO-maximum peak by difference spectro scopy from 450 nm to 448 nm, and a shift in the ratios of the peaks observed with ethyl isocyanide as a ligand by difference spectroscopy. These changes are interpreted to mean both 3-methylcholantrene and TCDD Figure 5. Logarithmic dose-response curve for the induction of aryl hydrocarbon hydroxylase in rat liver. Male Sprague-Dawley rats weighing about 80 g were injected intraperitoneally with TCDD dissolved in p-dioxane or p-dioxane alone (0.3 ml/kg), and hepatic aryl hydrocarbon hydroxylase activity was assayed 24 hr later. Each point is the mean standard error of five animals. itember 1973 Figure 6. Time course of aryl hydrocarbon hydroxy lase induction in ra t liver following TCDD and 3methylcholanthrene administration. Male SpragueDawley rats, weighing about 80 g were given a single injection of TCDD in p-dioxane (10 Mg/kg), . 3-methylcholanthrene in corn oiL (20 mg/kg), or the solvents alone. Each point represents the mean standard error of four rats. 249 784584 induce the formation of a new type of cyto chrome P-450 which contains its heme iron in a high-spin state. We have also found that both TCDD and methylcholanthrene pro duce induction of aryl hydrocarbon hydroxy lase in a number of extrahepatic tissues. In summary, TCDD appears to be simi lar to 3-methylcholanthrene as an inducer of microsomal oxygenase differing primarily in potency (by four orders of magnitude) and duration of action. The persistent induc tion following TCDD administration is prob ably a result of its long biological half-life. Discussion The results of our investigation demon strate TCDD is a potent inducer of ALA synthetase and aryl hydrocarbon hydroxy lase in the chick embryo liver. There is a perfect correlation between those dioxins which induce both enzymes and the toxicity lata, to the extent the data are available on the various dioxins. The structure-activity relationship reveals that all dioxins which ire potent inducers have halogens at three 3f the four lateral ring positions and at least me nonhalogenated carbon atom. The sen sitivity of induction of aryl hydrocarbon hydroxylase by TCDD and other toxic dioxins suggests this response might be a very valu able screening bioassay to detect the pres ence of the toxic dioxins in commercial pro ducts or environmental samples. It should be emphasized that the nonspecificity of the response makes it imperative that one ex tract the samples tested to remove polycylic hydrocarbons, and the test is only collabora tive, not definitive for TCDD and related dibenzo-p-dioxins. Now, I should like to tu rn to the broader question of the mechanism of toxic action produced by TCDD. Any proposed mecha nism of toxicity must account for several observations about TCDD (1). TCDD is a nearly planar, highly lipophilic, and a rather chemically unreactive molecule, which pos sesses remarkable biologic potency, and hence specificity (2). There are very large differ ences in susceptibility of different species to TCDD, as presented by Schwetz (5). The oral LD,n in the guinea pig is one .thousandth that of the dog (3). There is a very sharp struct ure-activity relationship among the dioxins. The oral LD,<, values of the 2,7-dichloro and octachloro derivatives are greater than 10s times th at of TCDD in the rat (4). TCDD seems to produce hepatic cell necrosis, and liver insufficency is the presumed cause of death in the rat. However, multiorgan in volvement in the rat has been reported at this symposium and elsewhere. Furthermore, as reported by Vos and Moore and colleagues (7, 5), hepatic necrosis is minimal in the mouse and guinea pig and perhaps insuffic ient to account for death. Thus we must ac count for the different pattern of histologic damage in different species (5). TCDD is remarkably slow in its toxic action leading to death. Regardless of dosage, animals die weeks after a single administration of TCDD (6). TCDD is an extraordinarily potent tera togen in a number of species (7). Finally, our investigations suggest that all dioxins which are potent toxins, as acneogens, tera togens, producing mortality or chick edema, also are potent inducers of aryl hydrocarbon hydroxylase activity. This enzyme complex is present and inducible in a number of tissues and is responsible for the aromatic hydroxylation of many xenobiotics. It is useful to examine the proposed me chanism of toxicity for other aromatic or halogenated aromatic compounds that, like TCDD, are chemically relatively unreactive and highly lipophilic. The two most exten sively investigated models are the liver necro sis produced by halogenated benzenes and the carcinogenesis produced by polycyclic hydrocarbons (5). Briefly the literature can be summarized as follows. The parent com pound is metabolized to a very reactive arene oxide intermediate. This intermediate may then chemically rearrange to a phenol, be further metabolized to a dihydrodiol or glu tathione conjugate, or, react chemically to covalently bind to various cellular macromo lecules which act as nucleophiles. In the case of bromobenzene centrolobular liver necrosis, the epoxide is believed to attach to proteins, and in the case of the polycyclic 250 Environmental H ealth Perspectives 784585 GENP 011774 hydrocarbon carcenogenesis the critical event is believed to be the binding- of the "K. region" epoxide to DNA. We propose a similar model for the toxi city of TCDD. The parent compound enters the cell and binds to some induction-recept or site which initiates the events which ulti mately lead to the formation of more aryl hydrocarbon hydroxylase activity. TCDD is recognized by a second site in the cell, the enzyme-active center of microsomal oxygen ase, and converted to a reactive metabolite, possibly an epoxide. The conversion of TCDD to its reactive metabolite is the rate-limit ing step in dioxin metabolism, and this step is increased by the induction of aryl hydro carbon hydroxylase. Some of these reactive metabolite molecules bind to cellular macro molecules producing some impairment of function which gradually produces cell death. It is useful to examine this hypothesis; in light of the known facts concerning TCDD toxicity. The large difference in species susceptibility to TCDD might be explained by the differences in the rate of metabolism \ o f TCDD. The multiple organ damage produc e d by TCDD and variable pattern of histo logic damage in different species might be explained by the relative rate of formation and further inactivation of the reactive metabolite-in different organs. (3) As pointed out by Gehring {10), there is some evid ence, far from unequivocal, that the admin istration of C14-TCDD results in unextractable radioactivity in ra t liver-. (4) The tera togenic effect of TCDD may be a result of mutagenesis by intercalation of the parent compound into DNA {11) or by intercala tion and covalent binding of the metabolite, analogous to the acridine and aminofluorene compounds {12,13). This hypotesis is highly speculative and presented only to encourage further investi gation. The major assumptions remain'un supported: (1) TCDD is metabolized, and (2) the metabolite covelently binds to some cellular constituent. The demonstration that both these events do or do not occur must await the synthesis of radioactive TCDD of high specific activity. The chlorinated dibenzo-p-dioxins are worthy "of much greater investigation, not only because the potential public health haz ard they pose, but also the remarkable poten cy and sharply defined structure-activity re lationship they demonstrate suggests an un common specificity of action. Ultimately, TCDD, like other potent toxins, (i.e., botulinus toxin, tetrodotoxin, organic phosphates) may become a useful biologic tool. REFERENCES 1. Bleiberg, J., et al., Industrially acquired por phyria. Arch..Dermatology 89: 793 (1964). 2. Poland, A., et al., A health survey of workers in a 2,4-D and 2,4,5-T plant. Arch. Environ. Health, 22: 316 (1971). 3. Poland, A:, and Glover, E. 2,3,7,8-Tetrachlorodibenzo-p-dioxin: potent inducer of 5-aminolevulinic acid synthetase. Science 179: 476 (1973). 4. Gielen, J., Goujon, F., and Nebert, D. Genetic regulation of aryl hydrocarbon hydroxylase in duction. II. Simple mendelian expression in mouse tissues in vivo. J. Biol. Chem. 247: (1972). 5. Schwetz, B. Chlorodibenzo-p-dioxin toxicology. Environ. Health Perspect. No. 5: 87 1973. 6. Woods, J. Studies of the effects of TCDD on mammalian hepatic 5-aminolevulinic acid syn thetase. Environ. Health Perspect. No. 5: 221 (1973). 7. Harris, M., Moore, J., and Vos, J. General bio logical effects of TCDD in laboratory animals. Environ. Health Perspect. No. 5: 101 1973. 8. Vos, J. G., Moore, J. A., and Zinkl, I. J. Effects of TCDD on the immune system of laboratory animals. Environ. Health Perspect. No. 5: 149 (1973). 9. Daly, J., Jerina, D., and Witkop, B. Arene oxides and the NIH shift: the metabolism, toxicity and carcinogenicity of aromatic compounds. Experientia 28: 1129 (1972). 10. Piper, W. N., Rose, J. Q., and Gehring, P. J. Excretion and tissue distribution of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Environ. Health Perspect. No. 5: 241 (1973). 11. Hussain, S.. et al. Mutagenic effects of TCDD on bacterial systems. Ambio 1: 32 (1972). 12. Ames, B., et al.. Derivatives of 2-acetylaminofluorene and other aromatic amine carcinogens. Proc. Nat. Acad. Sci., 69: 3128 (1972). 13. Creech, H., et al.. Antitumor and' mutagenic properties of a variety of heterocyclic nitrogen and sulfur mustards. J. Med. Chem., 15: 739 (1972). .Septem ber 1973 251 G E N P 011775 784586 Studies on the Bioaccumulation and Microbial Degradation of 2,3,7,8-Tetrachlorodibenzo-p-dioxin by Fumio Matsumura* and Herman I. Benezet* While the problem of pesticidal contamina tion of the environment is far from being solved, considerable useful information has emerged from the research efforts made by many scientists in recent years. First, we now know by experience that the chemicals that cause environmental problems are the ones which are extremely persistent nature, biologically active, and easily yicentrated in biological systems. Com m un ds which lack any of the above qualifi cations usually do not play any significant role in pesticidal pollution no m atter how acutely toxic they are. The above analysis becomes more important, when one considers other aspects of pesticidal pollution. For instance, we are concerned about only bio logical effects in considering pollution, with particular emphasis on the effects on non target organisms. In the case of polychlorinated dibenzo-pdioxins, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), the question of bioactivity is in disputable, as it is one of the most toxic compounds known to occur as a pesticidal impurity (1-3). Its chemical stability is also questionable. Thus the central question of its hazard to the environment must be stud Department of Entomology, University of Wis consin, Madison, Wisconsin' 53706. ite m b e r 1973 ied from the viewpoint of bioconcentration in various ecosystems. Published data on environmental fate of chlorodibenzo-p-dioxins are scarce at pres ent. Zitko U), for instance, could not find residues of dioxins in several aquatic ani mals at detection limits of 0.01-0.04 jug/g. Isensee and Jones (5), studied absorption and translocation of TCDD and its 2,7-dichIoro analog, in root and foliage, and con cluded that dioxins are neither readily picked up from the soil residues nor translocated into foliage, at least in the case of soy beans and oats. Studies by Kearney et al. () indicate th at the average TCDD remaining after weathering in soil for 1 y r is of the order of 50-60% at all concentrations tested (1 to 100 ppm), while the actual field survey (7) on the test areas with Agent Orange (19621970) revealed that much less TCDD resi dues remaining in soil than is expected from the above figure. In the study reported herein we have made efforts to measure the degree of bio accumulation of TCDD in relation to well established pesticides by using several model ecosystems. The data are still preliminary, in th at several model ecosystems are still be ing compared for their relative merits in assessing the actual impact of pesticides in 253 GENP 011776 784587 --nature. The data obtained have been,- how deposited on the inner surface of the glass ever, useful in assessing the relative ten container by evaporating the solvent to form dency of a pesticide in comparison with a thin film. The prim ary food organism were other pesticides. grown in the container for 24 hr and then transferred along with the culture media to Materials and Methods the aquarium containing the test inverte brate organism. Microbial Degradation of TCDD In model III, the pesticides (5 and 10 pmole) Microbes--Approximately 100 microbial strains which have previously shown the ability to degrade persistent pesticides were screened for their ability to degrade TCDD. were deposited on 1 g of sand and the solvent evaporated to form a thin film on the sur face of the sand particles. The sand was added to the test aquarium containing in vertebrates an d /o r fish. Procedure--Screening was carried out In all cases the test organisms were main and the metabolic products were examined tained in the aquarium at room temperature by thin-layer chromatography (TLC) by (24 C), except for the fish cultures which the method of Matsumura and Boush.(8). were maintained at 12 C. Test organisms Translocation of Pesticides were either homogenized in counting solu tion or carbonized (Model 300 Packard Tri- Procedure--The pesticides (0.1 mole each) were deposited on 1 g of clean sea sand, which was placed on a column of sandy loam type soil. Water was then slowly dropped onto the surface of the sand at a rate of approximately 2 ml/hr. The water and sections of soil were extracted with Carb Oxidizer), and the amount of 14COa measured. Measurements of the amount of labeled material in the water, prim ary food organism, and on sand and glass surfaces were made by extracting with chloroform. All studies were short-term (4-7 days), in small volume containers (200 ml). chloroform. Results and Discussion r Bioaccumulation of Pesticides ) As shown in Table 1, the extent of trans Animals--Three groups of invertebrates location of TCDD from the sand to the or were used for the pesticide accumulation ganic soil layer is extremely small. Virtually study : Ostracoda species, Artemia saiina, and no TCDD was found to leach out from the Aedes aegypti larvae, and one fish species, column. The mobility of TCDD in soil, there northern brook silverside, Laludesthes sic- fore, must be considered much less than cuius siccuius. that of DDT. Thus, the mode of transloca Pesticides--Four pesticides were selected from representative groups of important compounds: dioxin (TCDD), DDT, y-BHC, and zectran. All compounds were uC-labeled in the benzene rings. tion of TCDD in the environment would be limited to movement of soil particles or dust-carried dispersion and biological trans fer (but not plant-mediated transfer), par ticularly in aquatic environments. As for the microbially mediated degrada Procedure--Three model ecosystems were tion of TCDD, our current survey indicates used to study bioaccumulation. that such capabilities are rather rare ii U i In model I, the pesticides (5 and 10 pmole) nature. Approximately 100 microbial strains XA dissolved in a solvent were added directly in which the ability to degrade persistent ^ to w ater along with the prim ary food or pesticides has been previously demonstrated ganism, such as algae and yeast, and this were screened for this purpose. Among them, ^ mixture was then added to the aquarium only five strains showed some ability to de- v- containing the invertebrate test organisms. grade this compound (Fig. 1). We have net ^ In model II, the pesticides (20 pmole) were been able to manipulate cultural conditions 1 254 Environmental Health Perspectives 784588 Table l._ Vertical translocation of pesticides from sand to organic soil.* m m A v ir id e 21A # 96 uao | T P. p u tid a Pesticide content, % Top sand 0-0.5 cm .5-1.0 cm l.Q-1.5 cm 1.5-2.0 cm 2.Q-2.5 cm Water eluate * 1st 50 ml 2nd 50 ml 3rd 50 ml Dioxin 6 90.41 7.32 1.04 0.50 0.26 0.18 0.12 0.08 0.09 DDT 6 65.01 30.75 3.51 0.55 0.26 0.19 Zectran6 0.07 0.06 0.08 0.05 0.06 0.06 0.06 49.4 0.04 17.6 0.02 29.1 *10 x 1.5 cm glass column. 6 Pesticide introduced: 0.1 ^mole each (33.8 jig for dioxin, 35.5 fig for DDT, and 22.2 fig for Zectran). *W ater eluted per day, 50 ml. F igure l. Autoradiograph of microbial degradation of TCDD. The results are shown in terms of thinlayer chromatograms of extraits from micro organisms that were incubated with TCDD. The top spots at R( 6 are TCDD. The TLC system was: Silica gel C with carbon tetrachloride and methyl ene chloride (1:1) as a mobile phase. The names or the identification numbers are indicated below each chromatogram. to increase the rate of degradation of TCDD in any of the microorganisms so far. In studyingthe extent of biological trans fer of TCDD, three different model systems were devised. In model system I, pesticides in acetone were introduced directly into water along with the prim ary food orga nisms. In model system II, pesticides were applied to the inner surface of a glass con tainer, and the prim ary food organisms were grown in the container for 24 hr and were transferred to the aquarium. In model III, pesticide-coated sands were placed di rectly in the aquarium containing the test organisms. In the model I experiment (Table 2), DDT behaved quite differently from other pesticides, showing high degrees of affinity to each test organism, in close agreement with the phenomenon actually observed in nature. Although this model system is sim ple and appears to offer a quick straight forward answer to the general tendency of pesticidal accumulation by biological sys tems, it has one weakness, i.e., that one is forced to work above the limit of water solubility of some of the compounds. TCDD for instance was measured at a level 100 times its w ater solubility. Also the extent of direct pick-up due to partitioning and food intake is uncertain. In the model II experi ment, where only the portion of pesticide picked up by the primary food organisms and the media were introduced into the*test aquarium, the levels of total pick-up were further reduced in the case of TCDD (but not DDT) (Table 3). To circumvent the problem of solubility, the model III system was devised. In this way, only th at portion of pesticide th at is soluble should be present in water at any time. The results shown in Table 4 indicate that the rate of TCDD pick-up is extremely low in brine shrimp and fish under the ex perimental conditions. Mosquito larvae, which are bottom feeders, showed a surpris ing rate of TCDD pick up. The reaction is not at its maximal rate, since further in crease in the level of the pesticide apparent ly increases the pick up by the larvae. Also noted is the difference between the biocon centration pattern in fish as compared to other invertebrates. y-BHC, -in particular, September 1973 255 784589 shows high degree of concentration in fish. To study the effects of food consumption,J:he same test was repeated in the presence of mosquito larvae. As expected, the level of TCDD (Table 5) in the fish increased in the presence of mosquito larvae, which are the best concen trators of TCDD among the organisms tested. On the other hand, the levels of other pesticides did not significantly change, in dicating that the route through ingestion of mosquito larvae does not represent the m ajor source of uptake in these pesticides. It is apparent from these data that the reaction of biological concentration is great ly influenced by the external conditions and the design of the experiment, the physical and biological nature of the organisms, and by chemical characteristics of the pesticides. To facilitate understanding of the role of chemical nature of pesticides in determin ing the rate of bioconcentration, a compre hensive list has been prepared to illustrate their important properties (Table 6). It can be seen here that general tenden cies of bioaccumulation in invertebrate spe cies follow closely the trend of the parti tion coefficients. In model II experiments, however, the values for TCDD come much lower than expected from this rule. Thus it is likely that water solubility (and solv ent solubility) must play an important role where the initial pick-up is the rate-limiting factor. It is apparent that species-specific factors play a much more important role than once Table 2. Bioaccumulation of pesticides by aquatic invertebrates for model I (pesticides introduced directly into ambient water with the primary food organisms). Test organisms (primary food) Dapknia (algae) Oatracod (algae) Brine shrimp (yeast) Pesticide Dioxin DDT Zectran Dioxin DDT Zectran Dioxin DDT 7-BHC Zectran Original concentration in water, ppb 32.4 35.3 22.2 32.4 35.8 22.2 16.2 17.9 14.7 11.1 Final concentra tion found in test organisms, ppb 1,592 44,164 1,969 7,069 50,771 7,265 1,956 12,336 2,688 155 Concentration factor 49 1234 89 218 1418 327 121 689 183 14 Table 3. Bioconcentration of pesticides by aquatic invertebrates for model II (primary food organisms allowed to pick up pesticide from glass surface and then given to the test organisms). Test organism (primary food) Daphnia (algae) Ostracod (algae) Pesticide Dioxin DDT Zectran Dioxin DDT Zectran Original amount, concentration, ppb) 6.48 (162) 3.58 (179) 2.22 (111) 6.48 (162) 3.58 (179) 2.22 (111) Final concentration found, ppb Water aquarium 0.4 22.9 15.1 2.6 50.8 43.5 Test organisms 879 43,123 37,499 279 36,391 6,177 Concentration factor * 2,198 1,883 2,483 107 716 142 1Measured against the final pesticide concentrations actually found at the end of the test. 256 Environmental Health Perspectives i\ Table 4. Bioconcentration of pesticides by aquatic organisms for model III (pesticides introduced into system in the form of residues on sand). Test organism Brine shrimp Mosquito larvae Pesticide Dioxin DDT 7-BHC Zectran Dioxin DDT 7-BHC Zectran Fish (silverside) - Dioxin DDT 7-BHC Zectran Amount of pesticide, MS 1.62 1.79 1.47 1.11 1.62 3.24 1.79 3.58 1.47 2.94 - 1.11 2.22 1.62 1.79 1.47 1.11 Concentration found, ppb W ater (including food) 0.1 0.5 5.2 5.0 0.45 2.40 0.85 1.40 6.6 13.1 5.45 10.8 0 2.1 1.8 4.7 Test organisms 157 3,092 495 89 4,150 12,000 14,250 30,200 1,450 2,900 0 89 2 458 2,904 213 Concentration factor 1,570 6,184 95 18 9,222 5,000 16,765 21,571 220 221 0 8 -- 218 1,613 45 Table 5. Two-step bioconcentration of pesticide by mosquito larvae, and northern brook silverside (model 111). Pesticide Dioxin DDT 7-B H C Zectran Amount of pesticide, MS 1.62 1.79 1.47 1.11 Concentration found, ppb W ater (including food) 1.3 1.1 1.8 5 Mosauito 1larvae 3,700 17,900 690 0 Fish 708 337 1080 76 Concentration factor Mosauito Fish larvae 2,846 16,273 383 0 54 306 600 15 Table 6. Physicochemical characteristics of dioxin in comparison with other insecticides. Dioxin DDT Zectran 7-B H C *Estimates. Water solubility 0.2 ppb 1.2 ppb >100 ppm 10 ppm Solvent solubility Water solubility 10* 10U 10* 10* Partition coefficient (vs. hexane) 1,000 100,000 . 100* 1,700 Benzene solubility, g/100 g 0.047 30 -- so suspected. For instance, the pattern of bio accumulation and concentration in fish is quite different from those in other orga nisms studied, in that both y-BHC and zectran show higher degrees of affinity than September 1973 DDT and TCDD, respectively. Although the data are not sufficient to permit a definite conclusion, they suggest the possibility that water-soluble pesticides tend to accumulate in fish. 257 784591 The data indicate that TCDD is not likely to accumulate in as many biological sys tems as DDT. This is likely because of TCDD's low solubility in w ater and lipids as well as its low partition coefficient in lipids. Since microbial degradation is not expected to be a major factor, the predomi nant mode of elimination of this compound in the environment is photodecomposition by sunlight, (9). REFERENCES 1. Higginbotham, G. R., et al. Chemical and toxi cological evaluations of isolated and synthetic chloro derivatives of dibenzo-p-dioxin. Nature 220: 702 (1968). .2. Courtney, K. D., et al. Teratogenic evaluation of 2,4,5-T. Science 168: 864 (1970). 3. Sparschu, G. L., Dunn, F. L., and Rowe, V. EC. Study on the teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Food Cosmet. Toxicol. 9: 405 (1971). 4. Zitko, V, Absence of chlorinated dibenzodioxins and dibenzofurans from aquatic animals. Bull. Environ. Contain. Toxicol. 7: 105 (1972). 5. Isensee, A. R.( and Jones, G. E. Absorption and translocation of root and foliage applied 2,4,dichlorophenol, 2,7-dichlorodibenzo-p-dioxin, and 2,3,7,8-tetrachlorodibenzo-p-dioxin. J. Agr. Food. Chem. 19: 1210 (1971). 6. Kearney, P. C., Woolson, E. A., and Ellington, C. P. Jr. Persistence and metabolism of chlorodioxins in soils. In preparation. 7. Woolson, E. A., Young. A. L., and Hunter, J. H. Chemical analysis for dioxin and defoliant resi dues in soil from tested area C-52A, Eglin Air Force Base, Florida, Paper No. 173, Presented at meeting of the Weed Science Society of America, St. Louis, Mo., Feb. 8-10, 1972. 8. Matsumura, F., and Boush, G. M. Dieldrin: degradation by soil microorganisms. Science 156: 959 (1967). 9. Crosby, D. G., et al. Photodecomposition of chlo rinated dibenzo-p-dioxins. Science 173: 748 U971). GENP 011781 258 Environmental Health Perspective 784592 Environmental Generation and Degradation of Dibenzodioxins and Dibenzofurans by D.6. Crosby,* K.W. Moiianen,* and A.S. Wong* Introduction almost all of which contains low but detect Chlorinated dibenzo-p-dioxins and dibenzo furans have been known to chemists for many years. 2,8-Dichlorodibenzo-p-dioxin able levels of TCDD. The dioxins are formed most directly by heating an o-chlorophenol or its salts above was first reported in 1941 (1) ; the 2,3,7,8- about 200 C in either the presence or ab tetrachloro- (TCDD) and * octachloro- sence of a solvent. For example, heating pen- (OCDD) analogs followed in 1957 (2). Ex tensive research by Gilman (5) and by Poh- tachlorophenol (PCP) at 300C for 24 h r provided a low yield of OCDD [eq. (1)], land and Yang (4) provided data on many others. Likewise, simple chlorinated dibenzo J furans have been reported since the early 1930's (5), while experience with the more highly chlorinated ones is comparatively re while heating the sodium salt under the same conditions provided a much higher yield (2, 14, 15). Reaction of the dipotassium salt of catechol with 1, 2, 4-trichlorobenzene in di methyl sulfoxide at 175C produced a 70% cent (6, 7). yield of 2-chlorodibenzo-p-dioxin (4); the Although the unusual toxicity of certain of these compounds was recognized at an early date through occupational illness and nitro group can serve instead of chlorine as the displaced group (4, 15). Although diben zofurans may be prepared by dehydration of toxic effects on domestic animals (5), major 2,2'-dihydroxybiphenyls (16,17,18), 2-chloro- concern arose only when the toxic and tera 2-hydroxy- and 2,2'dichlorobiphenyls also togenic properties of TCDD became apparent provide them when treated with aqueous in widely-distributed pesticides such as alkali (19, 20), analogous .to dioxin forma 2.4.5- T (9, 10). Subsequent analysis (7, 11, tion. 12) showed that a variety of chlorinated Although related reactions have been sug dibenzo-p-dioxins and dibenzofurans can gested to operate by a free-radical mecha occur as impurities from the manufacture of nism (21), the preponderance of evidence many industrial and agricultural chemicals indicates that the above processes actually based on chlorophenols and certain chlori nated aromatic hydrocarbons. To cite but a single example, production statistics (13) suggest that at least 50 million pounds of represent nucleophilic substitution (22, 23). However, the C-Cl bond of the aromatic ring is much more stable than th at in correspond ing aliphatic compounds, and high tempera 2.4.5- trichlorophenol and its derivatives are tures are required to achieve reasonable manufactured in the United States each year, reaction rates unless activating ring substi tuents such as nitro groups are present. Department oi Environmental Toxicology, Uni Other means can be used to activate the versity of California, Davis, California 95616. ring, for example the intermediacy of a September 1973 259 784593 GENP 01.1782 copper complex (the Ullmann reaction) (23). In the presence of powered copper or cop per salts, a few hours a t 160C converts normally unreactive o-chlorophenols to dibenzo-p-dioxins (3, 4, 13). Ultraviolet light also can provide ring activation for both nucleophilic displacements and radical reac tion (24) ; thus, either heat or light theo retically could transform almost ubiquitous chlorinated aromatic compounds into dioxins and dibenzofurans under environmental con ditions. Thermal Generation Despite the suggestion (25) that phenol derivatives such as 2,4,5-T might be de composed to dibenzodioxins by heat, further investigation has failed to demonstrate the conversion (26, 27). Actually, few circum stances are apparent in which such compounds would be subjected to both the concentration and pyrolytic conditions nec essary for thermal condensation to occur; the accidental or intentional burning of herbicide-treated rangeland, for example, provides one such theoretical possibility. However, one very real possibility does ex ist. A major proportion of currently produc ed exterior plywood and other millwork is treated with PCP, as are a number of other wood products. The burning of scrap plywood and mill wastes should provide sufficient heat and concentration to convert the fungicide into dioxins or phenoxyphenols (predioxins) whose high degree of chlorination would cause them to volatize or move with smoke rather than burn. Predioxins would be ex pected to form dioxins and polyphenyl ethers upon pyrolysis (28). To test this concept, small chips of com mercial plywood containing 53 pg /g of PCP by analysis were completely charred. The smoke and volatiles were trapped and their dioxin content compared with that of simi lar, solvent-extracted chips. The OCDD level in the wood was only 1 ng/g, while that in the pyrolyzate was estimated at about twice this amount, despite a persistent but nonchlorinated interference. Hepta- and hexa- 260 (l) chlorodibenzo-p-dioxin but no TCDD were present in each sample. Identities were con firmed by gas chromatography with a massspectrometer detector (GC/MS). Photochemical Generation The ultraviolet component of sunlight is sufficiently energetic to generate free radicals by homolytic dissociation of both phenols and chlorinated aromatic compounds. The shortwavelength cutoff of ultraviolet radiation by the atmosphere is approxim ately-.290 nm (Fig. 1), and spectral energy increases sharp ly above this. The homolytic dissociation energies of the Ar-Cl bond (about 80 kcal/ einstein) and the ArO-H bond (about 90 k cal/ einstein) correspond to wavelengths near 360 and 320 nm, respectively--clearly within the sunlight region--and free-radical mecha nisms have been proposed for reactions of PCP (20) and other chlorophenols (30) in sunlight. Blessed indeed (at least by the environ mental photochemists) are those parts of the world which experience reliable, intense sunlight during most of the year. Because it varies so drastically with latitude, eleva tion, climate, and even the degree of a ir pollution, not to mention its short diurnal availability, the ultraviolet portion of solar radiation often is simulated for laboratory investigation; most of the work described here was conducted in a "sunlight-simulator" equipped with F40BL fluorescent lamps (General Electric Co.) exhibiting the spectral energy distribution also shown in Figure 1 (31). It is a fundamental law of photochemistry Environmental Health Perspectives 784594 GENP01178 F ig u r e l. Spectral energy distribution of an F40BL fluorescent ultraviolet lamp compared with sum mer sunlight. that radiation must be absorbed before reac tion can occur. However, it is not necessary that the light energy absorbed by aromatic rings bring about bond scission directly. As in the previous examples, the absorption of energy can actviate the ring and facilitate nucleophilic displacements. In typical photonucleophilic reactions (24), chloride may be displaced from 4-CPA (4-chlorophenoxyacetic acid), 2,4-D (2,4-dichlorophenoxyacetic acid), and 2,4,5-T (2,4,5-trichIorophenoxyacetic acid) by hydroxide ion to generate phenols, and, in accordance with earlier work of Munakata (29, 32) we now find that PCP forms chlorinated catechols and resorcinols in aqueous media irradiated at sunlight wave lengths. In polychlorinated compounds, dis placement of the chlorine ortho to the oxygen predominates. Under these circumstances, it might be expected that a chlorophenate anion could attack the light-activated ring of another chlorophenol to form a diphenyl ether, and the reaction could be repeated intramolecularly to generate a chlorinated dibenzo-pdioxin. Indeed, the irradiation of aqueous solutions of dioxin-free sodium PCP was found to generate OCDD, although only very small amounts could be detected by gas chromatography (33). Repeated attempts to detect TCDD after the irradiation of 2,4,5-T, 2,4J3-trichIorophenol, or sodium 2,4,5trichlorophenate solution were unsuccessful. The expected photonucleophilic formation of phenols from other aromatic halides such as chlorobiphenyls was demonstrated both with simple models (5-4-37) and with highly complex polychlorinated biphenyl (PCB) mixtures (Arochlors) (55). For example, ir radiation of an aqueous suspension of 4,4'dichlorobiphenyl provided 4-chloro-4'-hydroxybiphenyl and 4-chlorobiphenyl. How ever, as in the classical examples of syn thesis cited previously (19,20), appropriately substituted chlorinated biphenyls also might be expected to form chlorodibenzofurans. Irradiation experiments with five pure 2chlorinated biphenyls as 5 mg/1. aqueous suspensions, followed by resolution and ex amination of the products by GC/MS, showed that traces of 2-chlorodibenzofuran were de tectable (eq. (2)], although only the 2,5dichloro- and 2,5,2',5'-tetrachlorobiphenyls provided identifiable amounts (a roughly steady 0.2% yield during a 7-day irradia tion). ci ci ci (2) Photochemical Degradation Our failure to detect TCDD as a prod uct of 2,4,5-trichlorophenol photolysis can be explained on the basis of the extreme in stability -of the lower chlorinated dioxins to light (35). In either sunlight or the sun light-simulator, 2,7-dichloro-, 2,3,7-trichloroand 2,3,7,8-tetrachlorodibenzo-p-dioxins (5 mg/1. in methanol) were entirely decomposed within a few hours (Fig. 2). However, OCDD was much more stable under these conditions, which explains why traces of it could be September 1973 261 784595 c i ci uv ci .0 ci ci ci c h 3oh ci 0 (3) F ig u h e 2. Photodecomposition rates of (O) OCDD (2.2 mg/1.) and ( ) TCDD (5 mg/1.) in purified methanol undpr simulated sunlight (F40BL lamp). isolated after the irradiation of PCP in water. The rapid photodecomposition of TCDD in alcohols is accomplished by reductive de chlorination [eq. (5)], and an effective hy drogen donor appears absolutely necessary. Consequently, photoreduction in organic media proceeds efficiently to replace halogens one at a time with hydrogen atoms; the reduction is very sluggish in w ater and fails entirely in thin solid films of pure TCDD or on dry, irradiated soil surfaces (39, 40). The mechanism has not been determined, al though both the well-known free-radical hy drogen abstraction from solvent (30, 41) and hydride transfer (24) are possible. OCDD photoreduction provided detectable traces of hepta- and hexachlorodioxins, and chlori nated pesticides including 4-CPA, 2,4,5-T, PCP, PCNB (pentachloronitrobenzene), and hexachlorobenzene behaved similarly, as did the chlorinated biphenyls. On the basis of our previous experience with dioxins, biphenyls, and the other chlori nated compounds, it came as no particular surprise that model chlorinated dibenzofurans also were photoreduced in methanol 262 or even in water. For example, when a 5 m g/liter methanol solution of pure 2,8dichlorodibenzofuran was irradiated in th sunlight-simulator, 2-chIorodibenzofuran was the product detected by GC/MS, and more than 95% of the starting material was photolyzed within 48 hr (Fig. 3). However, the inadvertent use of highly purified meth anol in a similar experiment revealed only very slow photolysis within the same period of irradiation. The irradiation of a more concentrated solution (10 mg/1.) in the puri fied solvent also indicated slow decomposi tion; after 90 hr, the addition of 10 mg/1. of 4,4'-dichlorobenzophenone as a photosen sitizer resulted in a sharp increase in the photolysis rate, although acetone did not appear to sensitize the photolysis in this instance. It appears, then, that impurities in common solvents can drastically alter photo decomposition rates. Discussion A remarkable level of scientific attention recently has been devoted to the chlorinated dibenzodioxins and dibenzofurans. From the smattering of articles representing the four decades preceding 1970, two recent con ferences (42, 43) alone have provided more than 60 research papers. While earlier work dealt largely with basic chemistry and oc cupational health, more recent concerns have been primarily associated with the possible environmental occurrence and environmen tal effects. Indeed, much of the current effort to understand and control these com pounds arose because of unanswered ques tions about their environmental impact on humans, domestic animals, and wildlife. Although improved manufacturing and Environmental Health Perspectives 784596 GENP 011785 F ig u r e 3. Photodecomposition rates of 2,8-dichlorodibenzofuran (DCDBF): ( ,A ) in acetone-free methanol (10 mg/1.); (O) in acetone-free meth anol containing added acetone (320 mg/1.); ( ) in laboratory-grade methanol. The arrow indicates addition of 10 mg/1. of 4,4'-dichlorobenzophenone. Straight lines drawn for comparison only. \ surveillance methods have dramatically re duced dioxin and dibenzofuran levels among industrial chemicals (12), there is evidence that under idealized conditions, at least, both types of compounds can be generated from commonly used chlorophenols and chlorinated biphenyls under environmental conditions. The driving force can be either the ultra violet component of sunlight impinging upon thin dims, surfaces, or solutions or heat such as might be encountered in burning wood. Conceivably but not demonstrably, phenol derivatives such as the phenoxy herbicides also might provide the raw material, but, in any case, the necessary generative reac tions clearly have been demonstrated. On the other hand, environmental persis tence or accumulation theoretically could be balanced by the simultaneous destructive action of light or heat. Photolysis can be rather rapid under idealized conditions, al though a hydrogen donor appears to be required, and OCDD is rather stable com pared to the highly toxic TCDD. The char ring temperature of wood (>350C) is well above that frequently reported to convert PCP into OCDD, but the thermal stability of the dioxins has not been reported. Just how probable are the environmental form ation-and decomposition of dibenzodioxins and dibenzofurans ? The ultraviolet component of sunlight undoubtedly is capable of energizing their degradation in the pres ence of organic hydrogen donors, and those donors can be expected to be present at and after the time of application as either the usual formulation solvents (such as xylene or petroleum hydrocarbons), as active constituents of the formulation (for example, the alkyl esters of 2,4-D and 2,4,5-T), or as natural organic films. Although TCDD has been reported to be stable to irradiation on soil surfaces, it was not accompanied in those experiments by a nonvolatile organic source of hydrogens as it would be in the field. It was quite stable, too, when suspended in water but rapidly degraded when benzene also was present (35). Fhotosensitization also must be an im portant consideration, as demonstrated by the effect of dichlorobenzophenone on the photoreduction of chlorinated dibenzofurans. Such sensitizers have been shown to be ex pected under field conditions (44--46) and could drastically increase the photolysis rates of xenobiotics. At the least, photochemical investigations in the laboratory should recog nize the possibility of sensitization, for ex ample by impurities commonly present in even purified. grades of methanol. All to gether, degradation appears to outrace generation in most of our laboratory studies. However, the same solvents and active ingredients which could act as hydrogen donors and sensitizers also could protect dioxins and dibenzofurans from photolysis under practical conditions. F or example, while light absorption by TCDD might be only partially obscured by the absorption of a 2,4,5-T formulation (Fig. 4) and chlori nated dibenzofurans would be little shielded by PCBs (Fig. 5), the low level of OCDD in a technical formulation of PCP or sodium PCP should remain well-protected from light (Fig. 6), although an improved, low-dioxin product (Dowicide EC-7) was somewhat less opaque. Clearly, too, OCDD is environmental- September 1973 263 784597 F igure 5. Ultraviolet absorption spectra of PCB (Arochlor 1254) and 2,5-dichIorobiphenyl in meth anol compared with that of 2,3-dichlorodibenzo- furan (2,8-DCDBF). F igure 4. Ultraviolet absorption spectra of two concentrations (A,B) of commercial brush-killer (25.1% 2,4-0 isooctyl ester and 12.0% 2,4,5-T isooctyl ester in aromatic hydrocarbon solvent) in methanol compared with that of TCOD. ly generated--especially by heat--albeit in efficiently. Considering the probable photolysis of TCDD to nontoxic products and the in significant rate of dibenzofuran formation from biphenyls, it is hardly surprising that they remain undetected in environmental samples. In fact, OCDD and its immediate toxic reduction products--the hexa- and heptachlorodioxins--emerge to be of even more environmental health interest. For example, they appear to be the best candi dates for both photochemical formation and stability; obviously, they are released by 264 heat from common materials of construction; and the relatively innocuous OCDD readily undergoes reduction to form less chlorinated homologs which are both rather stable and toxic W ) . But are these compounds released, in fact? Are chlorinated dibenzo-p-dioxins and dibenzofurans actually generated and destroyed out there in the real world? Regrettably, no one seems to know. Experiments such as we have described generally are conducted under idealized conditions: pure compounds and solvents, artificial substrates, and even arti ficial sunlight. W hat of the sensitizers and sunscreens, the hydrogen donors and com peting nucleophiles which must be present in actual environmental applications? It is apparent that rather large stakes in terms of biological research investments, manufac turing, and public health now are riding on what previously may have seemed to be only exotic details of xenobiotic chemistry. Environmental Health Perspectives 784598 GENP 011787 Absorbance F igure 6. Ultraviolet absorption spectra of com mercial PCP (35.3% PCP and 4.1% other chlori nated phenols in petroleum distillate) and Dowicide EC-7 (88% PCP and 12% 2,3,4,6-tetrachIoro- v phenol) in methanol and commercial NaPCP (80' 85% sodium pentachlorophenate) in water com pared with th at of OCDD in methanol. A ck now ledgm ent We thank C. J. Soderquist and C. A. Reese for technical assistance in some aspects of this research and Carolyn Roeske for results of PCP analysis. The work was funded, in part, by U.S. Public Health Research Grant ES-00054 and by Regional Research Project W-45 of the U.S. Department of Agriculture. REFERENCES 1. Ueo, S. 2,6-Dichlorodiphenylene dioxide. Bull. Chem. Soc. Japan 15: 177 (1941). 2. Sanderman, W.t Stockman, H.f and Caston, K. Pyrolysis of pentachlorophenol. Chem. Ber. 90: 960 (1957). 3. Gilman, H., and* Dietrich, J. J, Halogen deriva tives of dibenzo-p-dioxin. J. Amer. Chem. Soc. 79: 1439 (1957). 4. Pohland, A. E., and Yang, ,G. C. Preparation and characterization of chlorinated dibenzo-pdioxins. J. Agr. Food Chem. 20: 1093 (1972). 5. Gilman, H., Brown, G. E., Bywater, W. G., and Kirkpatrick, W. H. Dibenzofurans III. Nuclear September 1973 substitution. J. Amer. Chem. Soc. 56: 2473 (1934). 6. Kimmig, J., and Schulz, K. H. Berufliche Akne (sog. Chlorakne) durch chlorierte aromatische zyklische ther. Dermatologica 115: 540 (1957). 7. Vos, J. G., et al. Identification and toxicological evaluation of chlorinated dibenzofuran and chlorinated naphthalene in two commercial poly chlorinated biphenyls. Food Cosmet. ToxicoL 8: 625 (1970). 8. Kimbrough, R. D. Toxicity of chlorinated hy drocarbons and related compounds. A review including chlorinated dibenzodioxins and chlori nated dibenzofurans. Arch. Environ. Health 25: 125 (1972). 9. Courtney, K. D., et al. Teratogenic evaluation of 2,4,5-T. Science 168: 864 (1970). 10. Sparschu, G. L., Dunn, F. L. and Rowe, V. K. Study of teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Food Cosmet. Toxicol. 9: 405 (1971) . 11. Firestone, D., et al. Determination of polychlorodibenzo-p-dioxins and related compounds in commercial chlorophenols. J. Assoc. Offic. Anal. Chem. 55: 85 (1972). 12. Woolsen, E. A., Thomas, R. F., and Ensor, P. D. Survey of *polychlorodibenzo-p-dioxin content of selected pesticides. J. Agr. Food Chem. 20: 351 (1972) . 13. U.S. Tariff Commission. Synthetic organic chemicals: production and-sales. TC Publ. 479, Washington, D.C., 1970. 14. Denivelle, L., Fort, R., and Hai, P. V. Octachloroand octabromodibenzo-p-dioxins and decachloroand decabromodiphenyi ether. Bull. Soc. Chim. France 1960: 1538. 15. Tomita, M., Ueda, S., and Narisada, M. Dibenzop-dioxin derivatives. XX.V1I. Synthesis of poiyhalodibenzo-p-dioxins. Yakugaku Zasshi 79: 186 (1959). 16. Tauber, E., and Halberstadt, E. Ber. 25: 2745 (1892). 17. Cullinane, N. M., Davey, H. G., and Padfleld, H. J. H. Diphenylene oxide series (IV). J. Chem. Soc. 1934: 716. IS. Cullinane, N. M., and Davies, C. G. Synthesis of some heterocyclic compounds, Rec. trav. chim. 55: 881 (1936). 19. Zahn, K., and Schimmelschmidt, K. Hydroxybiphenylene oxides. U.S. Pat. 2,172,572 (Sept. 12, 1940): Chem. Abstr. 34: 1032 (1940). 20. Case, F. H,, and Schock, R. U., Jr. Nitration of halobiphenyls (II). Di- and tetranitro deriva tives of -2.2'-dichlorobiphenyl3. J. Amer. Chem. Soc. 65: 2086 (1943). 21. Fanta, P. E. Ullmann Synthesis of biaryls. Chem. Rev. 64: 613 (1964). 265 784599 .:3 & l 22. Bacon, R. G. R., and Hill, H. A. 0. Copper complexes in organic chemistry. Quart. Rev, 19: . 95 (1965). 23; Weingarten, H. Mechanism of the Ullmann con,, densation. J, Org. Chem. 29 : 3624 (1964). 24. Crosby, D. G., et al. Photonucleophilic reactions of pesticides. In: Environmental Toxicology of Pesticides. F. Matsumura, G. M. Boush, and T. Misato, Eds., Academic Press, New York, 1972. .25. Buu-Hoi, N. P., Saint-Ruf, G,, Bigot, P., and Mangane, M. Preparation, properties, and iden tification of "dioxin" (2,3,7,8-tetrachIorodibenzodioxin) in the pyrolyzate of defoliants based on 2,4,5-trichlorophenoxyacetic acid and its esters and contaminated vegetation. Compt. Rend. Acad. Sci. Paris 273D: 708 (1971). 26. Langer, H. G. The formation of dibenzodioxins and other condensation products from chlori nated phenols and derivatives. Environ. Health Perspect. No. 5: 3 (1973). 27. Johnson, J. E. Safety in the development of herbicides. Proc. Calif. Weed Conf. 23: 43 (1971). 28. Rappe, C., and Nilsson, C.-A. An artifact in the gas chromatographic determination of impuri ties in pentachlorophenol. J. Chromatog. 67: 247 (1972). 29. Munakata, K., and Kuwahara, M. Photochemical degradation products of pentachlorophenol. Resi due Rev. 25: 13 (1969). 30. Plimmer, J. R. The photochemistry of halogenated herbicides. Residue Rev. 33: 47 (1970). 31. Crosby, D. G. Experimental approaches to pesti cide photodecomposition. Residue Rev. 25: l (1969). 32. Kuwahara, SI., Kato, N., and Munakata, K. The photochemical reaction of pentachlorophenol. I. The structure of the yellow compound. Agr. Biol. Chem. 30: 232 (1966). 33. Crosby, D. G., and Wong, A. S. The effects of light on phenoxy herbicides. Paper presented at 160th Meeting, American Chemical Society, Chi cago, September 17, 1970. 34. Crosby, D. G., and Moilanen, K. W. Annual Report, Food Protection and Toxicology Center, Univ. of Calif., Davis, Calif., 1971. 35. Safe, S., and Hutzinger, O. Polychlorinated bi- phenyls: photolysis of 2,4,6,2',4',6',-hexachlorobiphenyl. Nature 232: 641 (1971). 36. Ruzo, L. O., Zabik, M. J., and Schuetz, R. D. Polychlorinated biphenyls: photolysis of 3,4,3', 4'-tetrachlorobiphenyl and 4,4'-dichlorobiphenyl in solution. Bull. Environ. Contain. Toxicol. 8: 217 (1972). 37. Hutzinger, 0., Safe, S., and Zitko, V. Photochem ical degradation of chlorobiphenyls (PCBs). Environ. Health Perspect. No. 1: 15 (1972). 38. Crosby, D. G., et al. Photodecomposition of chlorinated dibenzo-p-dioxins. Science 173: 748 (1971) . 39. Plimmer, J. R., et al. Photochemistry of dibenzop-dioxins. In: Advances in Chemistry Series, 120 American Chemical Society, Washington, D.C., 1973, p. 44. 40. Kearney, P. C., Woolson, E. A., and Ellington, G. P., Jr. Persistence and metabolism of chlorodioxins in soils. Environ. Sci. Technol. 6: 1017 (1972) . 41. Crosby, D. G., and Hamadmad, N. The photo reduction of pentachlorobenzenes. J. Agr. Food Chem. 19: 1171 (1971). 42. Symposium on chlorinated dibenzo-p-dioxins in the environment. 162nd Meeting, American Chemical Society, Washington, D.C., Sept. 1971. 43. Conference on chlorinated dibenzodioxins and dibenzofurans. NIEHS, Raleigh, N.C., April 2-3, 1973. 44. Plimmer, J. R., and Klingebiel, U. Riboflavin photosensitized oxidation of 2,4-dichlorophenol: assessment of possible chlorinated dioxin forma tion. Science 174: 407 (1971). 45. Ivie, G. W., and Casida, J. E. Sensitized photo decomposition and photosenaitizer activity of pesticide chemicals exposed to sunlight on silica gel chromatoplates. J. Agr. Food Chem. 19: 405 (1971). 46. Ross, R. D., and Crosby, D. G. The photolysis of ethylenethiourea. J. Agr. Food Chem., 21: 335 (1972). 47. Higginbotham, G. R,, et al. Chemical and toxi cological evaluations of isolated and synthetic chloro derivatives of dibenzo-p-dioxin. Nature 220: 702 (1968). GENP 011789 266 Environmental Health Perspectives 784600 Photochemical Degradation of Di- and Octachlorodibenzofuran by 0. Hutzinger,"S. Safe,' B.R. Wentzell/ and V. Zitko* In tr o d u ctio n In the last few years, it has been shown - that polychlorinated biphenyls undergo de composition on irradiation with artificial ultraviolet light sources and sunlight (1-7). Reductive dehalogenation of one or more chlorine atoms was a major reaction, par,,jl ticularly in hydrocarbon solvents. In fluoro carbon solvents or thin films, small quanti ties of chlorobiphenyls with increased chlo rine content could also be detected. Irradia tion in hydroxylic solvents gave the dechlorinated species as well as photoproducts containing oxygen. In addition, chlorinated terphenyls and quaterphenyls were also de tected in some instances, and on prolonged irradiation polymeric products were formed. I t has been speculated for some time that chlorodibenzofurans may be formed from chlorobiphenyls under photochemical condi tions which lead to oxygenated products. Al though no chlorodibenzofurans could be de tected in a number of chlorobiphenyl sam ples which had been exposed to sunlight for over 2 months (-4), preliminary results in dicated the formation of chlorodibenzofur- *Atlantic Regional Laboratory, National Research Council of Canada, Halifax, Nova Scotia, Canada. tDepartm ent of Chemistry, Acadia University, Wolfville, Nova Scotia, Canada. Environment Canada, Fisheries and Marine Serv ice, SL Andrews, New Brunswick, Canada. ans from 2,2/,4,4',6,6'-hexachlorobiphenyl in model experiments (irradiation in methanol) (8) - The toxicity of pure chlorodibenzofurans has not been thoroughly investigated (9) but appears to be higher by several orders of magnitude than that of the chlorobi phenyls, which gives this reaction particular significance. In this regard, the photochemical behav ior of chlorodibenzofurans themselves be comes of interest from the point of view of products formed (decomposition to less tox ic products) and stability. As an approxima tion [eq. ( l) L it appears that if chloro dibenzofurans are formed from chlorobi phenyls in the environment by photochem ical reaction, accumulation will be a problem if > kt. Chlorobiphenyl -- chlorodibenzofuran - ka > decomposition products (1) Since a good selection of chlorodibenzo furans is not available, only preliminary investigations with two representatives, one of low (2,8-dichlorodibenzofuran) and one of high chlorine content (octachlorodiben zofuran) are reported. Equipment and Methods Chemicals 2,8-Dichlorodibenzofuran (10) and octa chlorodibenzofuran (11) were prepared by Septem ber 1973 267 GENP 011790 784601 literature methods. Samples for the initial experiments were provided by Dr. -At E. Pohland, (FDA, Washington, D.C.) Irradiations For the irradiations in solution (450 ml hexane or methanol), a Rayonet (the South ern New England Ultraviolet Co.) photo chemical reactor equipped with 16 RFR3100 lamps (310 nm) was used. Aliquots (25 ml) were taken at times specified in Figure 1. For mass spectrometric analysis of the photolysis mixture, 2,8-dichlorodibenzofuran and octachlorodibenzofuran were irradiated in methanol for 20 min. Thin films (50 mg) of the chlorodibenzofurans coated on the inside of quartz tubes (length: 25 cm; diameter, 4 cm) were ex*posed to sunlight for 10 weeks in the pres ence of w ater (2 ml) during July-September 1972. Total duration of bright sunshine was ca. 580 hr (Meteorological Services, Canadian Forces Base, Shearwater, N. S., private communication). Instrum ents Quantitative data were obtained with a Packard model A7901 instrument equipped with 6 f t x 4 mm columns packed with either 4% SE-30 (for the 2,8-dichlorodibenzofuran) or 3% OV-210 (for the octach lorodibenzofuran) on Chromosorb W. A DuPont/CEC 21-110B double focussing mass spectrometer was used for obtaining spectra by direct introduction. The probe was heated carefully, and spectra were re corded as the temperature was raised from 20C to ca. 180C. Analysis of Samples Quantitative analyses (photochemical stability of chlorodibenzofurans) were car ried out by gas chromatography (GC) with electron capture detection on the aliquots taken from the photochemical reactor. For the characterization of products formed, the solvent was removed from the samples and the residue chromatographed on Merck silica thin-layer plates (F-254; 0.25 mm thickness). The solvent used was hexane. For the mass spectroscopic analy sis, small fractions of the total sample were transferred to a mass spectrometer sample tube. Samples from the quartz tubes were dissolved in benzene-acetone and treated as described above. Results and Discussion The correct numbering of the dibenzofuran ring system (Chemical Abstracts and Ring Index) is shown in Figure 2. 9I 82 3 F igure 1. Photochemical degradation of chlorodibenzofurans in solution. Irradiation wavelength 310 nm. 268 F ig u r e 2. Numbering of the dibenzofuran nucleus. Environmental Health Perspectives 784602 f Some older systems are still in use occasion ally. Calculated molecular weights (C l = 35) for dibenzofuran and its chlorine substitu tion products are given in Table 1. Table 1. Molecular weights for chlorodibenzofurans (monoisotopic formula; Cl = 35) Formula CuHiO C,,H:C10 CiiHoCUO CtiHiCliO CuHtCUO CnHiCUO Ci:H-.CUO C,iH CbO CnCl*0 Molecular weight 168 202 236 270 304 338 .372 406 440 Irradiation of 2,8-Di- and Octach loro dibenzo furan in Solution The relative rate of decomposition of 2, 8-di- and octachlorodibenzofuran is shown in Figure 1. From these results and from the mass spectra of samples which were ir radiated for 20 min it is evident that de composition is faster in methanol than in hexane. In contrast to the chlorinated dibenzo-p-dioxins, where the degradation of the octachloro derivative on irradiation in methanol solution is much slower than that of the 2,7-dibenzo-p-dioxin (12), the di- and octachlorodlbenzofurans show similar rates of decomposition. For the analysis of products formed, sam ples which were irradiated for 20 min in methanol (short exposure) and for ca. 20 hr in hexane (long exposure) were chosen. Thin-layer chromatography of the sam ples exposed for 20 min showed only two spots, a relatively weak one on the origin and a large spot with an R f similar to those starting materials (Rj ca. 0.55 for 2,8-dichIorodibenzofuran; R/ = ca. 0.75 for octachlorodibenzofuran). Since organochlorine compounds which give very strong molecular ions can be analyzed by mass spectrometry in mixtures (13-15), samples of the irradiated prod ucts were carefully heated in the mass spectrometer probe and spectra recorded at different temperatures. Typical spectra are shown in Figures 3 and 4. No other products but those formed by dechlorination of the corresponding chlorodibenzofuran were found to be present No useful mass spectra could be obtained for the yellow gum which resulted from the 20-hr irradiation experiments. A large num ber of peaks was observed and no recogniz able chlorine isotope pattern was apparent. 1 i* A GENP 011792 F ig u r e 3. Maas spectrum (70 eV) of 2,8-dichlorodibenzofuran photolysis mixture (310 nm; 20 min; solvent methanol). Probe temperature: 40* C. September 1973 269 784603 Figure 4. Mass spectrum (70 eV) of octachlorodibenzofuran photolysis mixture (310 mn; 20 min; solvent; methanol). Probe temperature: 80 C. Exposure of 2,8-Di- and Octachlorodibenzo furan as Thin Film to Sunlight [eiudiation of cCTCHLoOD[RE!tgniiua The analysis by mass spectrometry was carried out as described above. Reductive dechlorination of octachlorodibenzofuran was observed to a much lesser degree than in solution. 2,8-Dichlorodibenzofuran gave, C Clt. C l.l (C l.) J in addition to a monochloro derivative, a IR1AO. > :0 S r -- W tTM R trichlo rodibenzofuran (M*-270). A summary of the results of the photoly sis experiments in solution and thin films is shown in Figures 5 and 6. F ig u r e 6. Summary of results from octachlorodi benzofuran irradiation experiments. Summary and Conclusions Photolysis of 2,8-di- and octachlorodiben zofuran in methanol and hexane solutions results in rapid dechlorination of the sub strates with the eventual accumulation of mup i Tip*or ;.i-fliQiiowim8Etonm<t t m o . . lS h r -- WITHER F igure 5. Summary of results from 2,8-diehlorodibenzofuran irradiation experiments. unidentified resinous polymeric products. Dechlorination is also observed to a certain extent when thin films of these compounds were exposed to sunlight. These preliminary data do not allow dir ect comparison of dibenzofuran degradation rates with rates of photochemical formation from corresponding chlorobiphenyls (5). In view of the photochemical lability of chlorodibenzofurans, however, it seems unlikely that accumulation of these compounds formed from chlorobiphenyls by photochem ical reaction in the environment will occur. REFERENCES 1. Safe, S,, and Hutzinger, O. Polychlorinated bi phenyls: photolysis of 2,4,6,2',4'6'-hexachlorobiphenyl. Nature 232: 641 (1971). 2. Hustert, K., and Krte, F. Synthese polychlo rierter Biphenyle und ihre Reaktion bei UVBestrahlung. .Chemosphere 1: 7 (1972). 3. Hutzinger, 0., Safe, S.f and Zitko, V. Photo chemical degradation of chlorobiphenyls. En viron. Health Perspect. 1: 15 (1972). 4. Hutzinger, 0., et aL Photochemical degradation of isomerically pure di-, tetra-, hexa-, octa-, and decachlorobiphenyls. Paper presented at 164th Meeting, American Chemical Society, Division of Water, Air and Waste Chemistry 1972; Ab stracts: 74 (1972). 5. Herring, J. L., Hannan, E. J., and Bills, D. D. UV irradiation of Aroclor 1254. Bull. Environ. Contain. Toxicol. 8: 153 (1972). 6. Ruzo, L. 0., Zabik, M. J., and Schuetz, R. D. Polychlorinated biphenyls: photolysis of 3,4,3',4'tetrachlorobiphenyl and 4,4'-dichlorobiphenyl in solution. Bull. Environ. Contain. Toxicol. 8: 217 (1972). 7. Nishiwaki, T., et al. Dechlorination of poly chlorinated biphenyls by UV irradiation (in Jap anese). Nippon Kagaku Kaishi: 2225 (1972); Chem. Abstr. 78: 29339 (1973). 8. Andersson, K., et al. Photochemical degradation of polyhalogenated biphenyls. Paper presented at PCB Conference II, Stockholm, 1972. 9. Vos, J. G,, et al. Identification and toxicological evaluation of chlorinated dibenzofuran and chlorinated naphthalene in two commercial poly chlorinated biphenyls. Food Cosmet. Toxicol. 8: 625 (1970). 10. Gilman, H., et al. Dibenzofuran. III. Nuclear substitutions. J. Amer. Chem. Soc. 56: 2473 (1934). 11. Hutzinger, 0., Safe, S., and Zitko, V. Analysis of chlorinated aromatic hydrocarbons by exhaus tive chlorination. Int. J. Environ. Anal. Chem. 2: 95 (1972). 12. Crosby, D. G., et al. Photodecomposition of chlorinated dibenzo-p-dioxins. Science 173: 748 (1971). 13. Hutzinger, 0., Jamieson, W. D., and Zitko, V. Identification of polychlorinated biphenyls and DDT in mixtures by mass spectrometry. Nature 226: 664 (1970). 14. Hutzinger, 0., and Jamieson, W. D. Identifica tion of organochlorine pesticides in crude ex tracts by -mass spectrometry. Bull. Environ. Contain. Toxicol. 55: 587 (1971). 15. Hutzinger, 0., and Jamieson, W. D. Application of high resolution mass spectrometry to residue analysis: identification of organochlorine and organometallic pesticides and pollutants in crude extracts. Pesticide Chemistry, (Proc. 2nd In ternational IUPAC Congress), A. S. Tahori, Ed., Vol. 4, 1971, p. 7. eptember 1973 271 784605 'I Tetrachlorodibenzodioxin in the Environment Sources, Fate, and Decontamination by P.C. Kearney,* E.A. Wooison,* A.R. Isensee,* and C.S. Helling* Introduction Research on the behavior and fate of pesticides in the environment has pro vided a number of valuable techniques for assessing the impact of any organic substance intentionally or unintentionally released in the environment. These tech niques can be applied to the 2,3,7,8-tetrachIorodibenzo-p-dioxin (TCDD) to determine its behavior under comparable conditions. " Important parameters affecting the environi ,-mental life history of any compound are movement, persistence, metabolism, plant uptake, translocation, photodecomposition, volatilization, and bioconcentration. Once these environmental properties are known about a compound, its anticipated behavior can be estimated, based on a comparison of information collected on registered pesti cides. The modern registration process for pesticides entails a detailed accounting of the above mentioned parameters, plus in formation on efficacy and toxicology. A second m ajor consideration in assessing the impact of any organic substance in the environment is the total amount th at may be present from current and past usage. For large-scale manufactured substances, produc tion information is available if there are at least three m anufacturers. Estimates on the occurrence of an impurity in any manufac- *Pesticide Degradation Laboratory, Agricultural Environmental Quality Institute, ARS, United States Department of Agriculture, Beitsville, Maryland 20705. September 1973 tured product may be extremely difficult to determine, unless this impurity has been recognized as a major contaminant and sur veillance is implemented at an early stage of the production process. Finally, once a substance has been iden tified as a potentially hazardous material, a safe, practical, and economically feasible decontamination process must be devised to use or destroy any known stocks. The Agricultural Research Service be came interested iri the dioxin problem soon after the disclosure that it was associated with one of the older and widely used herbicides, 2,4,5-T [(2,4,5-trichlorophenoxy) acetic acid]. Our preliminary studies were directed toward identifying other sources of dioxins in pesticidal chemicals and learning something about the fate or life his tory of the 2,3,7,8-tetrachiorodibenzo-p-dioxin (TCDD) in the environment. More re cently time has been devoted toward helping in the assessment of decontamination meth ods that will successfully destroy TCDD. The present paper considers the sources, fate, and decontamination of TCDD in the environ ment. Sources Historically, TCDD has been associated with any process that uses 2,4,5-trichlorophenol as a starting material. A number of the currently used synthetic organic pesti cides are derived from 2,4,5-trichlorophenol. In a survey of 17 pesticides examined for 273 !t '!! Mii . I i-lj '! I i: > ! )ti i !! ? GENP 011795 TCDD contamination, Woolson et al. (1) could detect TCDD only in older samples of 2,4,5-T and in one sample of silvex' [2^(2,4, 5-trichlorophenoxy) propionic acid]. The level of contamination in the single sample of silvex was 1.4 ppm. Of 42 samples of 2,4, 5-T, 20 contained TCDD measured at a limit of sensitivity of 0.1 ppm. Of these 20 posi tive samples, 7 contained less than 10 ppm TCDD and 13 contained between 10 and 100 ppm TCDD. It appears, then, that TCDD contamination occurred primarily in sam ples of 2,4,5-T. The United States production and domes tic disappearance of 2,4,5-T for the years 1960-1970 are shown in Figure 1. The total production for this ten-year period was 106,310.000 lb of 2,4,5-T expressed as the acid equivalent. The domestic disappearance fig ures for 2,4,5-T include military shipments abroad, primarily as defoliants for use in Vietnam. The peak year of production was 1968, when 17-5 million lb (about 16% of the ten-year total) was manufactured. Production sharply declined in 1969, to 5 million lb. The production figures for individual manufactur ers are unavailable on a yearly basis. Al though current data are lacking, older avail able data give some indication of the extent of use of 2,4,5-T. The total use of 2,4,5-T on farms, rights-of-way, and other nonfarm uses in 1964 was about 8,912,000 lb on 7.939.000 acres in the United States. The average rate of 2,4,5-T application on all domestic acreage in 1964 was about 1 lb / acre with a range from 0.25 to 2 lb/acre. The best estimates on dioxin content in past samples of 2,4,5-T come from an ex tensive survey of approximately 15 million pounds of Herbicide Orange (200 samples) conducted by the U.S. Air Force (unpub lished observations, 1973). Herbicide Orange is a defoilant containing about a 50/50 mixture of the butyl esters of 2,4,5-T and 2,4-D [(2,4-dichlorophenoxy) acetic acid]. The average dioxin content of the mixture was 1.91 ppm by weight (arithmetic mean) by use of a technique sensitive to 0.05 ppm. A frequency profile for these samples is shown in Figure 2. Of the 200 samples of Herbicide Orange, 136 or 68%, contained 0.5 ppm or less of TCDD. The highest sample contained 47 ppm TCDD. In the pesticide survey conducted by Woolson et al. (1) a 90% decline in TCDD was noted between 1968 and 1969 from one manufacturer. However, the level was still at 2-3 ppm in the 2,4,5-T acid. It is difficult and perhaps dangerous to extrapol ate the total amount of dioxin added to the environment from previous use of 2,4,5-T for the following reasons: (1) from pre liminary evidence available to us, it appears that different manufacturers produced 2,45-T with different TCDD contents over the TO* 90 F igure 1. Production and domestic disappearance (includes military shipments) of 2,4,5-T acid in the U.S. (1960-1970). 274 TCDD F igure 2. Distribution of TCDD content in 200 sam ples of Herbicide Orange. All samples containing more than 10 ppm appears as the last bar to the right of the figure. Environmental Health Perspectives 784607 GENP 0 1 1 7 9 6 10-year period, 1960-1969; (2) total pro duction of each m anufacturer and conse quently the total dioxin content on a yearly basis is unknown, (3) it is difficult, if not impossible, in certain instances to attribute specific lots of 2,4,5-T to certain manufac turers; (4) the statistical significance of the inferences drawn from 200 samples is prob ably too small to reach any valid conclusion on a realistic input of TCDD from herbicide applications. In 1971 when the dioxin issue first became known, we obtained and analyzed current production samples of 2,4,5-T acid then in production by the three principal manufac turers. They contained <0.1, 2.3, and 0.1 ppm. By 1971, industry could provide com mercial samples" of 2,4,5-T that met the suggested limits of less than 0.5 ppm TCDD and could probably routinely produce 2,4,5-T with a TCDD content of about 0.1 ppm. Although we can make no statement about the past input of TCDD into the en vironment for the reasons stated above, some predictions can be advanced as to future input if the level of TCDD is held at 0.1 ppm in 2,4,5-T. If one assumes the 1964 usage of 2,4,5-T indicates the normal peace time usage, i.e., about 8 million lb applied at 1 lb pound per acre, then the amount of TCDD reaching the soil would be equal to 0.05 m g/acre. Assuming the surface 3 in. of soil weighs 1 million lb, the concentration would be 0.1 parts per tril lion. Fate in the Environment A number of review articles covering var ious aspects of dioxin in the environment (2, 3) and in various environmental com ponents including soils (2, 4, 5) sunlight (5), plants (7), and wildlife (J) have appeared from the USD A Pesticide Degradation Lab oratory in Beltsville. A summary of these findings was presented in the review by Helling et al. (5) as follows. TCDD was not photodecomposed on wet or dry soil surfaces nor was it produced phctolytically from 2,4,5-trichlorophenol in water. TCDD is slowly lost in aqueous sus pensions and in methanol. VerticaI~movement of TCDD did not oc cur in a wide range of soil types. Contamina tion of underground water supplies seems very unlikely. Approximately half of the TCDD applied at concentrations of 1, 10, and 100 ppm per sisted in two moist soils after 1 yr under laboratory conditions. Field applications of high rates (up to 942 lb/A ) of 2,4,5-T on Lakeland sand pro duced no detectable (< 1 ppb) TCDD residue when sampled 6 yr later to a depth of 1 m. Soil metabolism of 2,4,5-trichlorophenol does not lead to the formation of TCDD as a condensation product. Small quantities of TCDD (<40 ppb) were accumulated by young oats and soybeans grown on a sandy loam contaminated with TCDD (60 ppb). No TCDD was detected (< 1 ppb) in mature plants or seeds grown on these same soils. TCDD was not translocated from the point of application on the leaf surface to other plant parts. Some wash-off and pos sibly volatilization did occur. Analyses of 19 eagle carcasses revealed no detectable (<50 ppb) dioxins. Admittedly more sensitive analytical techniques are needed before a clear picture of low TCDD wildlife levels can be assessed. Because of its low solubility (3-5 ppb), re latively long persistence, lack of vertical mobility in soils, and inability to translo cate in higher plants, TCDD more nearly resembles the shorter-lived chlorinated hy drocarbon insecticides in behavior than it does the more biodegradable phenoxyalkanoic acid herbicides, e.g., 2,4-D and 2,4,5-T. Decontamination Suspension of the use of Herbicide Orange in Vietnam has created large military sur pluses th at now must be used or destroyed. The U.S. Air Force has about 24 million lb of Herbicide Orange. Approximately 15 million lb contains an average of 1.91 ppm dioxin. Of the remaining material, which can not be identified by manufacturer, approxi- September 1973 275 784608 mately 80% contains less than 0.4 ppm dioxin. The political implications associated ,,with Herbicide Orange may preclude its use as an herbicide, even though a substantial portion of the excess stock meets currently establish ed EPA policy. Several conventional disposal methods, in addition to some new techniques, have been investigated to determine the safest, most economical, and most practical method for disposal of this material. Three options under consideration are incineration, soil biodegradation, and chlorinolysis. Since an environmental impact statement must be filed on the ultimate method or methods used for dealing with these surplus herbi cides and their dioxin contaminate, several scientific investigations have addressed their efforts toward determining the parameters needed for successful destruction of TCDD. Incineration appears to offer one of the safest methods for complete destruction of TCDD. In the present context incineration is defined as 'the high temperature reaction of TCDD with oxygen conducted in a unit equipped with proper emission control de vices. By the use of differential thermal analysis, Kennedy and Stojanovic (Missis sippi Agricultural and Forestry Experi ment Station, personal communication, 1973) have estimated that temperatures of 800-1000C will cause complete destruc tion of TCDD. These estimates are in good agreement with the value of 800C .pub lished by Langer etal. (9). Additional information is needed, how ever, on the composition of the stack gases and particulates, as well as w ater emanating from any industrial incineration process in volving dioxins. A second feasible method of dioxin dis posal is soil biodegradation. Our work (4) indicates that approximately 50% of the added TCDD had been destroyed after 1 y r in soils at reasonably high concentrations. Woolson et al. (5) were unable to detect any TCDD in a Lakeland sand receiving approximately 1,000 lb of 2,4,5-T per acre over a 7-yr period. No TCDD could be detected at a minimum detection limit of <1.0 ppb in core samples to a depth of 6 ft. Although TCDD is fairly recalcitrant in soils, it is immobile and thus would offer no ground w ater contamination problems. Soil incorporation to a depth of about 6 in. seems to offer an additional advantage, i.e., it would prevent aerial movement on soil borne par ticles away from the site of application. A third method of disposal of TCDD in volves a new technology called chlorinolysis. This process involves the high temperature conversion of carbon compounds in pres surized atmosphere of chlorine to form car bon tetrachloride. In the United States, chlorinolysis is still in an experimental phase. We have cooperated with the Diamond Shamrock Company in pilot studies to in vestigate the stability of TCDD in samples of Herbicide Orange. A detailed examina tion of carbon tetrachloride produced from Herbicide Orange revealed no TCDD at a level of sensitivity of 10 parts per trillion, as measured by electron-capture gas chro matography. Chlorinolysis represents a ma jor departure from conventional disposal systems in that it converts one product into .another useful resource. In addition to car bon tetrachloride, phosgene and hydro chloric acids are products of chlorinolysis. Carbon tetrachloride has a major market in the production of freon, which has wide application as a coolant and propellant. Summary The major source of TCDD input in the past has been from use of the herbicide 2,4,5-T. Future inputs will be minimal if the dioxin content is held at low levels (0.1 ppm) during the manufacturing process. In the environment, TCDD behavior is similar to some of the shorter-lived chlorin ated hydrocarbon insecticides. It is fairly persistent and immobile in soils, not taken up into the economic portion of plants, and slowly decomposed in water in sunlight. Several disposal options are available for the safe decontamination of surplus stocks of 2,4,5-T containing TCDD. These include incineration, soil disposal, and chlorinolysis. 276 Environmental Health Perspectives 784609 GENP 011798 REFERENCES Woolson, E. A.,-Thomas, R. F., and Ensor, P. D. J. Survey of polychlorodibenzo-p-dioxin con tent in selected pesticides. 2. Helling, C. S. Pesticide mobility in soils. II. Ap plications of soil thin-layer chromatography. Soil Sci. Soc. Amer. Proc. 35: 737 (1970). 3. Kearney, P. C., et al. Environmental signifi cance of chlorodioxins. In: Advances in Chemi stry Series, No. 120. American Chemical So ciety, Washington, D.C., 1973, Chap. 11. 4. Kearney, P. C., Woolson, E. A., and Ellington, C. P., Jr. Persistence and metabolism of chlorodioxins in soils. Environ. Sci. Technol. 6: 1017 (1972). 5. Woolson, E. A., et al. Dioxin residues in Lake land sand and bald eagle samples. In: Advances in Chemistry Series, No. 120. American Chemi cal Society, Washington, D.C., 1973, Chap. 12. 6. Crosby, D. G-, Wong, A. S.( Plimmer, J. R., and Woolson, E. A. Photodecomposition of chlor inated dibenzo-p-dioxins. Science 173: 748 (1971). - ~ 7. Isensee, A. R., and Jones, G. E. Absorption and translocation of root and foliage applied 2,4dichlorophenol, 2,7-dichlorodibenzo-p-dioxin, and 2,3,7,8-tetrachlorodibenzo-p-dioxin. J. Agr. Food Chem. 19: 1210 (1971). 8. Helling, C. S., et al. Chlorodioxins in pesticides, soils, and plants. J. Environ. Qual. 2: 171 (1973). 9. Langer, H. G., Brady, T. P., Dalton, L. A., Shannon, T. W., and Briggs, P. R. Thermal chemistry of chlorinated phenols. Paper pre sented at 162nd Mtg. American Chemical So ciety, Washington, 1971; Abstracts, PEST Sect., No. 83. 10. Fowler, D. L., and Mahan, J. N. The Pesticide Review 1971. Agricultural Stabilization and Conservation Service, U.S. Department of Ag riculture, Washington, D.C. 20250, 1972. f\ V ._ ,' "Jeptember 1973 1 277 784610 jmmary: Conference on Dibenzodioxins and Dibenzofurans, National Institute of Environmental Health Services, April 2 - 3 , 1 9 7 3 by Edward J. Burger, Jr.* To attem pt to summarize the proceedings of a meeting such as this one in any rigor ous sense is clearly an unreasonably ambiti ous task. It is made more challenging be cause of the tentative character of much of the work commented upon here. This meeting, one of a series (and I hope a growing series) for the National Institute v Invironmental Health Sciences, reprean extraordinarily useful concept. It is designed to bring together a variety of scientific bedfellows who have contemplated (and indeed investigated) a subject for its academic interest and for its v ery .timely topical interest. These same scientists would have learned about each other eventually but, in the best tradition of science, it would have taken a long time. By design, we are witnessing the cutting edge of scientific research for this area. This meeting has, as its avowed and very virtuous purpose, the calling out of the walls a good deal of unmatured and not totally interpreted or confirmed research. It is im portant to keep this tentative and unconfirmed character in mind. One is struck, too, by the character of the research which has been reported at this meeting. There are still many gaps in our knowledge about dibenzofurans and dioxins. Office of Science and Technology, Executive Office >f th President, Washington D.C. 20606. mber 1973 Yet, the thought that has gone into the design of the research reported here re presents a striking degree of sophistication in many cases for which the participants should be very proud. As late as 1970 or 1971, there existed only the crudest hint of a ranking of biological activity of members of the family of chlorinated dioxins. (1, 2) The reports at this conference contained descriptions of dose-response information, some beginning insight into mechanisms, and the first probings toward structureactivity relationships. I am struck by the fact that the early observations (from the occupational environment) and the early approximations of rankings have been sus tained and essentially confirmed by the data reviewed at this meeting. Chemistry, Analysis, and Chemical and Physical Properties Dr. Langer's paper on the formation of dioxins from precursors through condensa tion reactions represents a good example of what we should do more of. I speak here of the attem pt to predict probable and im probable behavior in the environment from knowledge of physical and chemical pro perties. Dr. Laager's nuptial analogy is apt in more than one way. He reminded us that chemical courtship leading to marriage (in this case, condensation) was family-specific 279 784611 and adhered to some orthodox rules. Family character, and how this is perceived by the other party to the arrangement, seems to be important. It seems to me that Dr. Danger has thrown down the challenge of confirm ation for dioxin formation, reported on by some as feasible in the environment, I am sure that this challenge will be taken up. One other point was raised by Dr. Langer and was echoed independently by others was the spurious formation of condensation products within the chambers of the very instruments used to detect them (e.g., the gas chromatograph). What is the evidence of "weathering"-- the formation in nature of dioxins or furans from chlorinated materials through the add ition of energy from somewhere? Crosby et al. and Hutzinger et al. suggested that condensation reactions to form dioxins or dibenzofurans might be promoted by expos ure to sunlight and presented the results of a few preliminary experiments to examine this subject. What they properly reminded us of, however, was the fact that the story does not end there. The presence of detect able condensation products depends on the dynamics of both production and ensu ing decomposition. Decomposition, they re minded us, occurs typically through reduc tion and here depends on an available hydrogen source. (They demonstrated the point in their laboratory experiments by us ing a hydrocarbon medium.) Tetrachlorodibenzo-p-dioxin was found to be more la bile than the octochlorinated member of the family. It was speculated that there was sufficient organic material in most en vironmental situations to assure hydrogen donors. In brief, environmental persistence seems unlikely. The evidence seems to sug gest that those impurities which are found are of the less toxic varieties. However, we need more samples because of the large va riety of commercial products. A. E. Pohland et al. revealed some of the potential and the limitations of two analytic methods, electron spin resonance and visible light spectroscopy. As I heard this paper, these sounded particularly useful as con firmatory techniques. The authors cited the need for pure standards to realize the po tential or their methods. Dr. Crummett's paper is perhaps the latest in a growing series of examples of how the power of analytic methods tends to "drive" manufacturing procedures to be more rigo rous and produce greater degrees of purity. His point is well made. We are all better off as a result. It is heartening to note the similarity in degrees of resolution reported both by Crummett, et al. and by Drs. Baughman and Meselson for measurement of dioxin through mass spectroscopy. Understandably, these results rested, apparently, on an exten sive clean-up procedure. (I think some may still be bothered by what I understand is a wide intrasample variability among some of the measurements.) Since they are "push ing" their a rt to a point near its limits, perhaps these results deserve as much con firmation as possible. Nevertheless, the power of sensitivity and resolution are indeed im pressive. Biological Effects I think that it is extremely important to acknowledge the fact that the original bio logical insight into dioxins came from a series of observations made by Dr. Suskind of accidental occupational exposures in the late 1940's. The exposure was the result of accidental release of chemical intermediates in a 2,4,5-T plant in 1949 resulting in ex posure of a number of workers to manifest chloracne. In 1957, Kimmig and Schulz re ported chloracne among workers in 2,4,5-T plant in Germany. A third occupational in cident occurred in a 2,4,5-T plant in the United States in 1964. In a way, it is some what disappointing that there is not more human experience reported at this meeting. I think that there is still a clouded issue, an unclear distinction between the effects of PCB, 2,4.5-T and of dioxins and furans. For example, we should somehow ascertain whether Yusho disease in -Japan was a re flection of exposure to polychlorinated bi phenyls or to furan impurities. Dr. Firs- 280 Environmental Health Perspectives 784612 GENP 011801 vne reviewed the history of the contribuof the chick edema factor to our under standing of dioxins. The sleuthing done by the FDA pieced together the story of the large-scale loss of poultry (which happened first in 1957), related it to the use of tallow in poultry feed and, eventually, to the presence of dioxin impurities. Higgin botham et al. first offered a rough approxi mation of ranking of biological activity of dioxins which has turned out to be strikingly accurate. What has emerged from this meet ing is the very wide range of toxicity for the several members of the dioxin family (perhaps as much as 10T). T h is' meeting revealed some interesting (and perhaps, ultimately successful) at tempts to relate chemical structure to bio logical activity. Several participants re minded us that to be a successful toxic dioxin, a candidate needs two halogens at the 2 and 3 positions and one at the 7 position; it also needs halogens' on both henezene rings. Bromine confers more bio logical activity than chlorine, and chlorine ire than fluorine. 711 say very little about teratogenesis. It seems to me that a strong case can be made for clearing the air about the mechanism of teratogenesis. Is this an example of acute (embryo) toxicity with a steep dose-re sponse curve and a demonstrable threshold? It's not clear that everyone who reports birth defects is talking about the same pheno menon. Dr. Moore, at this meeting, described some fascinating, postnatal effects of ma ternal exposure to TCDD through a series of cross-fostering and reciprocal cross-fost ering studies of mice. These deserve further attempts at interpretation. The effects on experimental animals are difficult to summarize completely. However, there are some underlying currents showing through; (1) there is a variation in susceptibility among species--guinea pigs versus rats and mice; (2) there are striking sex differences; (3) delayed effects are very prominent (liver changes 2 weeks after exposure); (4) the Uiajor sites of toxic action appear to be the liver (seen as a variety of changes in liver function),.the_hematopoietic system (plate let depression, altered platelet function, leucocytosis and hemoconcentration) and the lymphatic system (spleen, thymus and lymph nodes). The atrophy of the thymus and the general lymphoid depletion reported at this meeting were very striking. As for morphological alterations, the changes in the ultrastructure under the elec tron microscope are of course most interest ing. Perhaps the most significant point is that the morphological alterations tend to follow and confirm the functional changes which were described independently. There seems to be a delay (perhaps on the order of 3 clays) between exposure and manifest structural changes. The magnitude of the change is dose-related, and the changes are reversible with time. A particularly fascinat ing finding was that of multinucleated hep atic cells. Do these represent a stage of attempted regeneration and repair? Alter natively, are they possible precursors of neoplastic change ? This conference pre sented little evidence that dioxins would induce or promote neoplastic changes in tissues. Patterns of absorption into the organism and of distribution among organs once ab sorbed are beginning to emerge. Not unex pectedly. water and lipid solubility seems to emerge as a major influence, although clearly not the only one. For tetrachlorodibenzop-dioxin (high doses in male rats), the amount absorbed via the intestine from an ingested dose appears to be about 70%. The majority of the absorbed dose appears in the feces at a rate of 1-2% day and in the urine at a rate of 0.5% day. A small amount can be detected in the expired air (<0.1% day). The material resident in the organ ism is characteristically found in the adi pose tissue and the liver. It is notably ab sent from certain other fatty tissues such as those of the central nervous system. By contrast, for octachlorodibenzo-p-di- oxin, only 5% achieves absorption. Again, a large share is found in the liver (50%) and in the adipose tissues (127/-). Once in 1 ,, y ptember 1973 281 784613 the liver, this material apparently tends to remain resident for long periods in thTiver microsomes. This meeting served to bring together a remarkable amount of work on the effect of dioxins on cellular enzymes. This was all the more remarkable, as none of this work had even been conceived of two years ago. A number of hepatic enzymes were found to be induced and a few depressed as a result of dioxin exposure. (A general caveat was voiced over what appeared to be unu sually high doses of dioxin used in some of the experiments.) The degree of induction was at times striking. The experiments re vealed a dose-response relationship. Again, there was an unequivocal sex difference and a characteristic latent period between ex posure and induction. Effects were often long-lasting (for example, a persistent threefold increase 38 days after exposure in one experiment). Again, lipid solubility may play a large role. The meaning of enzyme changes is as yet unclear. There are some striking dif ferences among species. ALA synthetase, whose activity is related to the disease, por phyria, can be'induced by dioxin adminis tration in the chick embryo but apparently not in mammals. One has the impression of being very close to some insight into mechanisms yet not close enough. The com bination of enzyme induction studies and changes in cellular ultrastructure could prove very helpful. Where do we stand on our knowledge of biological activity? For furans, we know very little. It seems to me that we still must determine whether Yusho disease was a reflection of PCB exposure or a result of exposure to dibenzofuran or other impurity. For dioxins, we now are better equipped. However, we must now reconcile a number of somewhat paradoxical observations: (i) extraordinarily high degree of biological activity, especially for certain chemical form:; (tetrachlorodibenzo-p-dioxin was pointed out to be the most potent small molecule toxin known); (2) striking species differences in activity; (3) sex differences; (4) biological activity falls off rapidly with changes in chemical structure; (5) latent period before toxic manifestations; (6) doserelated effects; (7) long-lasting but ulti mately reversible effects. E ffects on W ildlife Understandably there is less work here, than one would like. The preliminary work reported by Bowes concerning survey of wildlife is a good model and should be con tinued. Preliminary results seemed to sug gest a very wide variety of chemical species found in the animals examined with an un certain role for dioxins and furans. REFERENCES 1. Higginbotham. G. R., Ress, J., and Firestone, D. Chick edema factor in fats and fatty acids. Chem. Eng. News 44: 53 (1966). 2. Higginbotham, G. R., Huong, A., Firestone, D., Verrett, J., Ress, J,, and Campbell, A. D. Chem ical and toxological evaluations of isolated de rivatives of dibenzo-p-dioxin. Nature 220: (1968). GENP011803 282 Environmental Health Perspectives 784614 Chlorinated Dibenzodioxins and Dibenzofurans* by Jam es Edward H uff1 and John S. W a sso ir The problems and universal concern about chlorinated dibenzodioxin and dibenzofuran compounds were brought to the forefront by :he scientific community during the National institute of Environmental Health Sciences' (NIEHS) Conference on this subject which :is held at Research Triangle Park, North irolina, April 2 and 3, 1973. This idea for this literature collection emanated from that scientific gathering and resulted in this an notated bibliography of 242 references. These references are categorized by year and ar ranged alphabetically by author. The number of references per year are: 56/1973; 67/ 1972; 66/1971; 24/1970; 29/1969-1934. Sources searched are summarized in Table 1. Due to the time available to complete this collection, some errors and omissions were inevitable; we apologize for these and hope those using this literature survey will supply us with past, present, and future topical Work supported by Toxicology Information Pro gram, National Library of Medicine; National Insti tute of Environmental Health Sciences; and the National Cancer Institute under contract with the Union Carbide Corporation. t Biomedical Studies and Toxicology Information Response Center. X Environmental Mutagen Information Center, P.O. Box Y, Bldg. 9224, Environmental Information System Office, Oak Ridge National Laboratory, Oak v ige, Tennessee 37830. reprints or citation information. We plan to maintain and up-date this file continually. Most of the nomenclature or terms searched are listed in Table 2. Each particular author, journal, secondary abstracting serv ice, and news copy unfortunately utilizes separate and distinct terminology when re porting on the chlorinated dibenzodioxins and dibenzofurans. As can be seen from Table 2, it Is vitally important, therefore, to become thoroughly familiar with the sources being utilized before mounting a massive effort to collate all that is written or reported about a particular compound, series, or class of compounds, or subject. The magnitude of the search effort for this report is selfevident when noting all the necessary terms used. The original papers were annotated when ever possible; some were gleaned from ab stract journals. We did not alter author's remarks or conclusions ; the facts are pre sented in these annotations as they appeared in the literature. As many salient points as possible, due to space limitations, were taken from each paper or report. Many papers and reports were referred to, after the fact, by more recent authors as having dealt intimate ly with the chlorinated dibenzodioxins and dibenzofurans; these were not included in this bibliography unless the dioxins or furans were mentioned specifically : most of the September 1973 283 784615 t708TT0rTKm r> articles were consulted and read however to determine if these compounds were pres ent. For example, numerous articles reporting on the adverse effects of 2,4,5-T were screen ed but not included because direct mention of TCDD and other dioxin derivatives was missing. Table 1. Sources and time periods searched. Multidisciplinary Information Resources Bibliography of Agriculture Biological Abstracts Biological and Agricultural Index Bioresearch Index Chemical Abstracts Chemical-Biological Activities Food Chemical News Healtk Aspects of Pesticides Abstract Bulletin Health Effects of Environmental Pollutants Index Medicus Pesticide Chemical News Science Citation Index Teratology Lookout Toxicology Bibliography Specialized Information Centers and Libraries Environmental Mutagen Information Center (EMIC) Environmental Information System Office (EISO) Oak Ridge National Laboratory (ORNL) Toxicology Information Response Center (TIRC) On-Line Computer Data Bases MEDLine TOXLine Journals Ambio Archives of Environmental Health Bulletin of Environmental Contamination and Toxicology Clinical Toxicology Environment Environmental Health Perspectives Federation Proceedings Food and Cosmetic Toxicology Journal of Agriculture and Food Chemistry Journal of the Association of Official Analytical Chemists Journal of Chromatography Mutation Research Residue Reviews Science Teratology Toxicology and Applied Pharmacology Period Vol. 21 (1957) to Vol. 37(1) (1973) VoL 31 (1957) to Vol. 55(8) (1973) Vol. 19 (1964) to Vol. 24 (1970) Vol. 1 (1965) to Vol. 9(4) (1973) Vol. 1 (1907) to Vol. 78(20) (1973) 1965 to 1971 Vol. 13 (42) 1972 to Vol. 15(9) (1973) Vol. 1 (1966) to Vol. 6(4) (1973) Vol. 1 (1972) to Vol. 2(3) (1973) Vol. 60 (1956) to Vol. 14(5) (1973) Vol. 1 (1-24) (1973) Vol. 1 (1961) to Vol. 6 (1965); 1966 to 1972 Vol. 3 (1972) Vol. 1 (1968) to Vol. 6(1) (1973) Vol. 1 (1972) to Vol. 2 (1973) Vol. 7 (1963) to Vol. 26(2) (1973) Vot. 1 (1966) to Vol. 7 (1973) Vol. 1 (1) (1971) to Vol. 6(1) (1973) Vol. 12 (1970) to Vot. 15(3) (1973) No. 1 (1972) to No. 5 (1973) Vol. 16(1) (1957) to Vol. 32(4) (1973) Vol. 1 (1969) to Vol. 10(6) (1972) Vol. 1 (1953) to Vol. 2 1(2) (1973) Vol. 40 (1957) to Vol. 56 (1973) Vot. 1 (1958) to Vol. 73 (1973) Vol. 1 (1964) to Vol. 10(6) (1972) Vol. 1 (1962) to Vol. 41 (1972) Vol. 157 (1967) to Vol. 179 (1972) Vol. 1 (1968) to Vol. 6(2) (1972) Vol. 1 (1959) to Vol. 24(3) (1973) G E N P 011805 284 Environmental Health Perspectives 784616 Table 2. Terms searched. Nomenclature Chemical Abstracts registry number 84 Dibenzo-p-dioxin (diphenylene dioxide) (phendioxin) 2,3-Dichloro2.7- Dichloro2.3.7- Trichloro1.2.3.4- Tetrachloro1,3,6,8-Tetrachloro2.3.6.7- Tetrachloro2.3.7.8- Tetrachloro PentachloroHexachloro1.2.3.7.8.9- Hexachloro Heptachloro- 1.2.3.4.6.7.3.9- OctachloroBenzoBiphenylChlorinated Chlorinated dibenzodioxin(s) Chlorinated dibenzo-p- dioxin (s) DibenzoDibenzo Chlorodibenzo dioxinfs) Chlorodibenzo-p-dioxinfs) Dibenzodioxin(s) Dibenzo-p-dioxins Dihenzo-p-dioxins, tetra- chloroDibenzo-p-dioxin, other chloroDioxin Halogenated dibenzo dioxin(s) Halogenated dibenzo-p- dioxin (s) OCDD Polychlorinated dibenzodioxin Poiychlorodihenzodioxin TCDD Dihenzofuran (biphenylene oxide) 3-Chloro2.4- Dichloro1.2.4- TrichloroTetrachloro1.2.3.4- TetrachloroPentachloroBenzoBipheyl- 262-12-4 29446-15-9 33857-26-0 33857-28-2 30746-58-8 33423-92-6 1746-01-6 36088-22-9 34465-46-8 19408-74-3 3268-87-9 132-64-9 25074-67-3 24478-74-8 24478-73-7 23076-57-6 24478-72-6 32076-58-7 September 1973 Chlorinated Chlorinated dibenzofuran(s) ChlorodibenzoDi benzo DibenzoDibenzofuran Puran Halogenated dihenzofuran(s) Polychlorinated dibenzofurans 1973 Anonymous. Herbicides better for birds than bacteria. Food Cosmet. Toxicol. 11(1): 149-150 (1973). Teratogenic effects of the 2,4,5-T contaminant, 2,3,7,8-tetrachlorodibenzo-p-dioxin (dioxin), are well known and the effect of this compound on liver enzymes resembles that of certain carcinogens. Di oxin has also been associated with mutagenic prop erties through its possible intercalation with DNA; the observed effects resemble those of acridine. Anonymous. TCDD residue disappears. Down To Earth 28(4): 18 (Spring 1973). TCDD was not detected (<1 ppm) in 3-ft soil core samples in a sandy area where a total of 947 lb of 2.4.5-T per acre was applied over a 3-yr period. No TCDD residues were found f<0.05 ppm) in hald eagle tissue gathered from 15 states. Anonymous. 2,4,5-T comes to public atten tion again. Pest. Chem. News 1(19): 6-8 April 11, 1973). EPA was asked to suspend the remaining uses of 2,4,5-T until the extent, if any, of TCDD-contaminated food chains is determined. TCDD was labeled a cumulative poison and one of the most potent agents of birth defects in animals ever dis covered. Anonymous. Health hazard of dioxins still uncertain. Chem. Eng. News 51(16): 12 (April 16, 1973). A selected summary report of the National In stitute of Environmental Health Sciences meeting on chlorinated dibenzodioxins and dibenzofurans held at Research Triangle Park in North Carolina on April 2-3, 1973. The conference as a whole, however, seemed to raise as many questions as it answered. From the data discussed there is no doubt that these contaminants are highly toxic and teratogenic, but there is still some doubt as to how much of an actual hazard they represent to human health. 285 784617 GENP OO O as Anonymous. New look may be taken at trichlorophenol compounds. Pest. Chem. News 1(21): 9-10 April 25, 1973)" Work showing "surprisingly high" levels of di oxin (ppt) in fish caught in Vietnam has placed doubt on earlier government consideration about dioxin residues. It had been previously thought that the dioxin content of 2,4,5-T was so low that there was little opportunity of residues appearing. In ad dition to 2,4,5-T and silver, any compound using trichlorophenol intermediates "may be suspect". Anonymous. Correction. P e st Chem. News 1(22): 2 (May 2,1973). The National Academy of Science's Advisory Committee on 2,4,5-T recommended that the regis tration of 2,4,5-T be restored with the following exceptions: (a) a permissible residue of not more than 0.1 ppm 2,4,5-T on edible parts of food products and in water for human consumption and (b ).a limit of 0-5 ppm of contamination with TCDD, ex cept that in all formulations to be used around the home and recreational areas, TCDD contamination should be limited to 0.1 ppm. Anonymous. Recall program, another look at trichlo'rophenols urged on EPA. Pest Chem. News 1(23): 3-5 (May 9, 1973). The General Accounting Office urged the En vironmental Protection Agency (EPA) to imple ment full-scale recall procedures for suspended pesticides and raised questions on the dioxin content of trichlorophenol herbicides. GAO stated that "be cause silvex, ronnel, erbon. and hexachlorophene can contain the same level of dioxin as 2,4,5-T and because a safe level of dioxin has not been de termined, we believe EPA should establish a stand ard for dioxin content and prohibit the use of all pesticides containing dioxin in excess of the es tablished standard." Baughman, R. W., and Meselson, M. S. An analytical method for detecting TCDD (dioxin) : levels of TCDD in samples from Vietnam. Environ. Health Persped. (No. 5): 27 (1973). An analytic procedure involving extensive clean up and mass spectroscopy detects approximately one picogram of TCDD. The method separates TCDD from DDE, PCB's, and other chlorinated hydrocar bon residues. Bowes, G. W., Simoneit, B. R., Burlin game, A. L., de Lafpe, B. W., Peakall, D. B., and Risebrough, R. W. The search for chlorinated dibenzofurans and chlorinated dibenzodioxins in wildlife populations showing elevated levels of embryonic death. Environ. Health Persped. (No. 5): 191 (1973). Embryonic deaths have been recorded in the lab oratory among birds treated with PCB. These deaths have been attributed to chlorinated dibenzofuran contaminants. Also, birth defects in wild populations of birds and sea lions are believed caused by chlorinated dibenzodioxins. High-resolu tion mass spectrometry was used to examine pre pared from aborted sea lions and dead embryos of the herring gull and osprey. Crosby, D. G., Moilanen, K. W.f and Wong, A. S. Environmental generation and de gradation of dibenzodioxins and diben zofurans. Environ. Health Persped. (No. 5) : 259 (1973). Both the chlorinated dibenzodioxins and dibenzo furans are unstable to light in the presence of or ganic substrates. Even if generated under environ mental conditions, light provides a mechanism for rapid destruction. Crummett, W. B., and Stehl. R. H. Determi nation of chlorinated dibenzodioxins and dibenzofurans in various materials. En viron. Health Persped. (No. 5): 15 (1973). Chlorinated dibenzo-p-dioxins and chlorinated di benzofurans can be determined in chlorinated phen ols, chlorinated phenoxv herbicides, ronnel. fat, and conhustion products by such analytical techniques as gas chromatography, liquid chromatography, thin-layer chromatography, and gas chromatogra phy-mass spectrometry. Dougherty, W. H .; Coulston. F . ; Golberg, L. Non-teratogenicity of 2,4,5-trichIorophenoxyacetic acid in monkeys (Macaco, mulatta). Twelfth Annual Meeting, So ciety of Toxicology, New York, NY (March 18-22, 1973), Abstract 9, p. 7. Technical grade 2,4,5-T which contained less than 0.05 ppm 2,3,7,8-tetrachlorodibenzo-p-dioxin was ad ministered orally to forty pregnant Rhesus monkeys daily from day 22 through day 38 of gestation. Dose levels used in the experiment were 0.05, 1.0, and 10.0 mg/kg. Hematology, clinical. chemistry, ana urinalysis data were recorded for all females before and at various times following treatment until par turition: no toxicity was observed. Examination of live born infants revealed no terata. E nvironmental Protection Agency, Pub lications and Information Section. Toxicology and pharmacology of 2,4,5-T 286 Environmental Health Perspectives 784618 GENP 011807 {includes dioxins). Bibliography Numt ber 73-04 (February 1973). A bibliographic listing of 62 references pertaining :o 2,4,5-T and dioxins. ENVIRONMENTAL PROTECTION AGENCY, PUB LICATIONS and Information Section. Chemistry and residues of 2,4,5-T (in cludes dioxins). Bibliography Number 73-05 (February 1973). A bibliographic collection of 58 references on 2.4.5-T containing a limited number of citations on dioxins. E nvironmental Protection Agency and Institute of Rural E nvironmental Health. Environmental chemicals: hu man and animal health. Fort Collins, CO (July 23-27, 1973). The Institute of Rural Environmental Health. Colorado State University, and the Office of Pes ticide Programs, U. S. Environmental Protection Agency, conducted a one-week course on environ mental problems, contaminants, toxicants,t and chem icals (including dioxins); and human and animal health problems; and poisoning. JF irestone. D. Etiology of chick edema disease. Environ. Health Perspect. (No. 5) : ` 59 (1973). Early work indicated that chick edema factors (CEF) were chlorinated aromatic compounds; later, he compounds were shown to belong to a family of chlorodibenzo-p-dioxins. Further investigation snowed that (a) chlorophenols were precursors of the chlorodioxins and (b) chlorodioxins and related compounds are commonly present as minor com ponents in commercial chlorophenols. Characteristic chick edema disease symptoms include excessive fluid in the heart sac and abdominal cavity fol lowed by high mortality starting in the third week. Fowler, B-, Lucier. G., Brown. H., and Mc Daniel, O. Ultrastructural changes in rat liver cells following a single injec tion of TCDD. Environ. Health Perspect. (No. 5) : 141 (1973). Ultrastructure changes in ra t liver microsomes and mitochondria were examined at various inter vals from 1 to 30 days following a single TCDD in jection of 0. 5, or 25 ug/kg. No histologic difference was noted between groups. Observed changes in cluded: proliferation of smooth endoplasmic reticu lum (SE R ), mild increase in rough endoplasmic reticulum (RER), moderate swelling of mitochon dria: at 1 days, large aggregates of SER, massive amounts of RER, and small numbers of moderately swollen mitochondria were seen from the 25 mg/kg dosed rats. Greig, J. B. Biochemical toxicity of TCDD in rat liver. Environ. Health Perspect (No. 5): 211 (1973). The persistent toxic effect of 2,3,7,8-tetrachlorodibenzodioxin in rat3 is evidenced by death as long as 15 weeks after a single oral dose. Alterations, how ever, in liver constitution (microsomes and cyto chrome P-450) and drug metabolism (zoxazolamine and hexobarbital) occurred within 24 hr after dosing. Gupta, B., Vos, J., Moore, J., Zinkl, J., and Bullock, B. C. Pathologic effects of TCDD in laboratory animals. Environ. Health Perspect. (No. 5): 125 (1973). Gross and microscopic examinations were per formed on rats, guinea pigs, and mice treated with TCDD. A spectrum of dose ranges and schedules were used. Lymphoid organs (thymus, spleen, and lymph nodes) were affected consistently. Thymus atrophy (dose related decrease in weight) was found to be a sensitive index of TCDD exposure. The most severe hepatic effects were seen in rats that received a lethal dose of TCDD. The magnitude of the degenerative and necrotic liver changes were diminished in guinea pigs and mice. Harris. M.. Moore. J., and Vos, J. General biological effects of TCDD in laboratory animals. Enviy'on. Health Perspect. (No. 5) : 101 (1973). Albino rats were grouped and. treated with single oral doses of 0. 5, 25, 50, or 100 .ug/kg TCDD in an acetone-corn oil mixture. Animals that eventually died continued to lose weight until death while sur vivors exhibited a depressed weight gain. Ruffled hair coat, hunched posture, inactivity, and jaundice were the overt signs seen in the high dose group. Daily oral administration of 10 M?/kg caused death in 15/16 rats with a mean time of 21.8 days. Death resulted in 9 / 10 female guinea pigs after receiving an oral dose of 3 ^g/kg; the mean survival time was 18.1 days. A single oral dose of 1, 10, or 50 ug/kg to adult mice had no effect on appearance or body weight. H utzinger. O., Safe, S., Wentzell, B. R.. and Zitko, V. Photochemical degradation of di- and octachlorodibenzofurans. En viron. Health Perspect. (No. 5) : 267 (1973). Irradiation of 2,3-dichlorodihenzofuran (low chlo rine content) and octachlorddibenzofuran (high chlorine content) in hexane and methanol caused jtember 1973 287 784619 decomposition to compounds which are formed by reductive dechlorination as well as polar substances. Hwang, S. W. Effect of TCDD on Biliaryexcretion of indocyanine green. Environ. Health Perspect (No. 5): 227 (1973). Bile flow and biliary excretion of indocyanine green (ICG) in male rats 1, 7, and 16 days after receiving a single oral dose of 5 or 25 ig/kg TCDD. Bile flow rate increased at day 1 through day 16. During a 20-min. collection period, both the concen tration and total ICG excreted in bile decreased. ICG disappearance rate decreased with time. These effects were dose related. In contrast, the 5 pg/kg dosed rats accumulated more ICG in liver. J ensen, S., and Renberg, L. Various chlori nated dimers present in several technical chlorophenols used as fungicides. En viron, Health Perspect. (No. 5): 37 (1973). The presence of 2,3,7,8-tetrachIorodibenzo-p-dioxin in 2,4,5-trichIorophenoxy acid esters originates from 2,4,5-trichlorophenol during the manufacturing process. Dimerization occurs when the phenol is produced by the action of alkali on tetrachlorobenzene. All products originating from alkali-treated chlorinated benzenes logically may contain chlori nated dibenzo-p-dioxins. Dimers in pentachlorophenol and in 2.4,6-tri- and 2,3,4,6-tetrachIorophenols from direct chlorination of phenol were presented. J ohnson, R. L., Gehring, P. J., and Kociba, R.J. Chlorinated dibenzodioxins and pen tachlorophenol. Environ. Health Per spect (No. 5): 171 (1973). Pentachlorophenol enjoys widespread use as a wood preservative. Commercial grades have been found to contain up to 2500 ppm chlorinated dibenzop-dioxins. The predominant dioxin is octachlorodibenzo-p-dioxin, one of the least toxic members. Eval uating pentachlorophenol toxicity in animals re vealed that some untoward effects (chloracne, chick edema disease, and histopathologic alterations) were caused by chlorinated dibenzo-p-dioxin content. Purified pentachlorophenol did not produce these effects. A new procedure was capable of producing pentachlorophenol containing lowered concentra tions of chlorinated dibenzo-p-dioxin and devoid of dioxin-like toxic effects. Kearney, P. C., Woolson, E. A. Isensee, A. R., and Helling, C. S. Tetrachlorodibenzodioxin in the environment: sources, fate, and decontamination. En viron. Health Perspect. (No. 5): 273 (1973). TCDD does not leach in soils, does not reside in the economic portion of plants growing in contam inated soil, degrades to about 509e after 1 yr in soils, and does not result from microbial or chemical condensation of 2,4,5-trichlorophenol in soil. Kende, A. S., and Wade, J. J., Synthesis of new steric and electronic analogs of 2, 3,7,8-tetrachlorodibenzo-p-dioxin. EnvU ron. Health Perspect. (No. 5): 49 (1973). Structural-activity relationships for a series of TCDD analogs were accomplished emphasizing chemical studies in an attempt to distinguish steric from electronic requirements for toxicity. Catechol condensation was used to explore the scope and lim itations of polyhalobenzene electrophiles. King, M. E., and Shefner, A. M., Carcino genesis bioassay of chlorinated dibenzo dioxins and related chemicals. Environ. Health Perspect. (No. 5): 163 (1973). Chlorinated dibenzodioxins were dissolved in ace tone and applied to the backs of mice three times a week to assess the activity of dioxins as complete carcinogens and/or promoting agents. Octachlorodioxin caused skin tumor formation in only one female mouse. No other dioxin produced papillomas. Langer, H. G. Formation of dibenzodioxins and other condensation products from chlorinated phenols and derivatives. E 71viron. Health Perspect. (No. 5): 3 (1973). Chlorodioxins are formed in a two-step condensa tion reaction from ortho-substituted halophenoxy radicals or anions. Reaction of chlorine with penta chlorophenol at elevated temperature proceeds by radicals; anionic condensation products result from strongly exothermic reactions of alkali metal salts of chlorinated phenols above 300 C. Reaction pro duct distribution depends on the total number of halogen substituents, the crystal lattice arrangement of the molecule, steric effects, and an electronic ef fect. Dioxin formation was the major condensation product only for sodium pentachlorophenate. Lucier, G. W,,- McDaniel, 0. S., F owler. B. A., F aeder, E., Hook, G.f and Scnawane, B. R. Studies on TCDD-induced changes in rat liver microsomal and mitochondrial enzymes. Environ. Healih Perspect. (No. 5): 199 (1973). A single oral dose of 5 or 25 pg/'kg was adminis tered to male rats and time-course measurements were made on some hepatic microsomal and mito chondrial enzymes. Cytochrome P-450 and b-a contents were increased, hydroxyiation of aniline 288 Environmental Health Perspectives 784620 GENP 011809 ms induced, microsomal protein contents were in cased; aminopyrine-dmthylation rates were de creased, and most strikingly UDP glucuronyltransferase was increased about 8-fold. Martin, R. L,, P orter, M. L., Pomerantz, I. H. Studies on the formation potential and presence of chlorinated dibenzofurans in chlorinated biphenyls. NIEHS conference, April 2-3, 1973. Chlorinated dibenzofurans may arise by photo chemical alteration of chlorinated biphenyls under appropriate conditions; this did not occur with pentachlorobiphenyl or 2,2'-dichlorobiphenyi. Matsumura, F., and Benezet, H. J. Studies on the bioaccumulation and microbial degradation of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Environ. Health Perspect. (No. 5) ; 253 (1973). Most organisms capable of degrading other chlo rinated hydrocarbons showed no ability to metabolize TCDD; a few exhibited a limited degree of TCDDmetabolizing activity, TCDD leached from sand :o organic soil much less than did DDT. With pescicide-coated sand in aquaria containing various or ganisms, TCDD had the lowest biologic accumulation and affinity. " eselson, S. Vietnam dioxin contamination. Center for Short-Lived Phenomena, Event 51-73, No. 1611, Smithsonian In stitute, Cambridge, Mass. (April 19, 1973). Various fish and shellfish collected in 1970 from the Dong Nai and Saigon Rivers and along the Can Guo Coast contained dioxin; catfish had the highest concentration.' Meselson, S. Vietnam dioxin contamination. Center for Short-Lived Phenomena, Event 51-73, No. 1627, Smithsonian In stitute, Cambridge, Mass, (8 May 1973). Samples of fish and crustaceans caught In Viet nam in September 1970 were analyzed for dioxin content by using mass spectrometry. Dioxin con centrations ranged from 18 to 814 ppt; Dong Nai river carp averaged 540 ppt dioxin. Miller, R. A., Norris, L. A., and Hawkes, C. L. Acute and chronic toxicity .of 2,3,7,8-tetrachlorodibenzo-p-dioxin (dioxin) in aquatic organisms. Environ. Health Perspect. (No. 5) : 177 (1973). Guppies and coho salmon fingerlings were exposed to dioxin concentrations ranging from 0.056 ppt to than 0.2 ppb for 24, 48, and 96 hrs. The initial concentration was found to be more important in causing death ..than duration of exposure. The coho salmon fingerlings threshold response level for all exposure periods was between 0.056 and 0.56 ppt dioxin. Mosquito larvae, oligochaete worms, and puiminator snails were maintained in water initially dosed with 0.2 ppb dioxin. These aquatic organisms were less sensitive than fish. Moore, J .; Gupta, B.; Vos, J .; Zinkl, J. Postnatal effects of maternal exposure to TCDD. Environ. Health Perspect. (No. 5): 81 (1973). Maternal exposure of C5731/6 mice to TCDD caused dose-related variations in fetal kidney matu ration and development. Thymuses were reduced in size; cystic kidneys developed. Mean body weights and thymus and spleen weights (absolute and rela tive) were reduced in litters whose mothers received 10 Mg'kg TCDD;. the 3 ng/ kg dose group exhibited no weight deviations. Kidney effects were seen at both dose levels. Neubert, D.pZens, P., and Rothenwallner, A. Survey of the teratogenic effects of 2.3,7,8-tetrachlorodibenzo-p-dioxin in mammalian species. Environ. Health Perspect. (No. 5); 67 (1973). The frequency of cleft palate induction was used as a criterion of *the teratogenic effects of TCDD. Dose-response relationships and potentiating effects of TCDD with other agents were presented. NIEHS Conference on Chlorinated Dibenzodioxins and Dibenzofurans. National In stitute of Environmental Health Scien ces, Research Triangle Park, N.C. (April 2-3, 1973). The two-way conference reviewed critically and summarized the world literature and research ac tivities on the chlorinated derivatives of dibenzodioxin and dihenzofuran. Approximately 35 papers were presented on all aspects ranging from chemical nature to ultimate biologic effects. More than 110 scientists were in attendance. The proceedings are given in full in Environmental Health Perspectives, Experimental No. 5, (1973). (this issue). Ncrback, D. H., and E ngblom, J. F. Chlori nated dibenzo-p-dioxin distribution with in rat tissues and subfractions of the liver. Fed. Proc. 32(3): 236 (1973) Ab stract 138. Radioactivity from orally intubated " Cl-labeled octachlorodibenzo-/j-dioxin in rats was confined to the liver, adipose tissue, and skin after 7 weeks on a control diet; concentrations were about 20Tr that in tissues of rats after 21 days. .. Member 1973 289 G EN P011810 784621 NORBACK, D. H., Engblom, J. F., and ALLEN, J. R. Chlorinated dibenzo-p-dioxin dis tribution within rat tissues and subtrac tions of the liver. Environ. Health Per spect. (No. 5): 233 (1973). Male rats received daily for 21 days 100 ng "Cllabeled actachlorodibenzo-p-dioxin by gastric intuba tion. Feces contained 95% of the total dose and urine 4%. Significant levels were found in the kid neys, heart, and serum. The liver contained the highest concentration per unit weight; adipose tis sue had 1/3 that of the liver. Microsomes (rough and smooth fractions) had 95% of the liver radio activity. Urine radioactivity resided in the lipid fraction. NoVICK, S. Dioxin. Environment 15(4) ; 2324 (May 1973). . A news item reports the detection of dioxin m fish and shellfish used for food in Vietnam. Plimmer, J. R., Ruth, J. M., and Woolson, E. A. Mass spectrometric identification of the hepta- and octa-chlorinated difaenzo-p-dioxins and dibenzofurans in technical pentachlorophenol. J. Agr. Food Client. 21(1): 90-93 (1973). The presence of contaminant dioxins and dibenzo furans in some samples of technical pentachloro phenol were confirmed by using mass spectrometry. Three samples, collected in 1970 contained hexachlorodibenzo-p-dioxin (0.5 to 37 ppmw) and heptachlorodibenzo-p-dioxin (90 to 135 ppmw). It was reemphasized that high-resolution spectra do not provide chlorine orientation information. P limmer, J. R. Technical pentachlorophenols--origin and analysis of base-insol uble contaminants. Environ. Health Persped. (No. 5): 41 (1973). F ats used as feed additives fiom hides preserved with technical pentachlorophenol have been impli cated as a source of chick edema factor. Polychlori nated dibenzodioxins and dibenzofurans have been identified in the neutral fractions of pentachloro phenol by gas chromatography, mass spectrometry, and a combination of these two. P ohland, a . E., Yang, G. C., and Brown, N. Analytical and confirmative techniques for dibenzo-p-dioxins based upon their cation radicals. Environ. Health Perspect (No. 5): 9 (1973). Chlorinated dibenzo-p-dicxins form cation radi cals when dissolved in strong acids such as trifluoromethane sulfonic acid, in the presence of ultraviolet light, or an oxidizing agent like potassium nitrate. These cation radicals are quantified by using elec tron spin resonance and visible spectroscopic tech niques. A general bathochromic shift was observed with increasing chlorine content. These shifts were dependent upon the position of the chlorine atoms. Poland, A. P., and Glover, E. 2,3,7,8-Tetrachlorodibenzo-p-dioxin: A potent inducer of S-aminoIevulinic acid synthetase. Science 179(4072): 476-677 (Feb. 2, 1973). As little as 4.66 X 10'" mole (1.5 ng) of TCDD per egg induced hepatic-aminolvulinic acid synthe tase (ALAS) activity in the chick embryo. Enzyme induction was dose-related and prolonged in time: 70% of the maximum induced activity was present 5 days after a single 150 ng dose. TCDD was linked to an outbreak of porphyria cutanea tarda where 2.4,o-T was synthesized and manufactured. At least three of the 2.3,7, and 3 positions on the ring must be occupied to induce ALAS. Poland, A. P., and Glover, E. Studies on the mechanism of action of the halogenated dibenzo-p-dioxins. Environ. Health Perspect. (No. 5 ): 245 (1973). Aminolevulinic acid synthetase (ALAS) was stimulated by TCDD in the chick embryo liver; 4.66 X 10'" mole/egg (1.5 ng) caused doubling of ALAS activity and 1.55 X 10"' mole/egg (0.5 ng) caused a 35-fold stimulation of enzyme activity. A single dose of TCDD stimulates hepatic aryl hydro carbon hydroxylase (AHH) and cytochrome P-450 for 35 days and more in the rat. AHH activity was induced in chick embryo liver. It is suggested that the chemically inert parent compound is not the toxic moiety, but that a highly reactive intermed iate causes cell damage. Schwetz, B. A., Norris, J. M., Sfarsceu, G. L., Rowe, V. K., Gehring, P. J., Emerson, J. L .; Gerbig, C. G. Chlorodibenzo-p-dioxin toxicology. Environ. Health Perspect. (No. 5): 87 (1973). 2,7-Dichlorodiben2o-( DCDD), 2,3.7,8-tetrachlorodibenzo-(TCDD), hexachlorodibenzo-(HCDD), and octachlorodibenzo-p-dioxin-(OCDD) were evaluated toxicology. TCDD and HCDD were acnegenic. embryo toxic (TCDD markedly so), teratogenic, and positive for chick edema factor (CEF). DCDD and OCDD were negative for acnegenicity, teratogen icity, and CEF; OCDD was embryotoxic, while DCDD was not. The lethal dose range for DCDD, TCDD, HCDD, and OCDD was g/kg, ,ug/kg, mg/kg, and g/kg, respectively. S'TAPLEY, D . Herbicides: AAAS study finds dioxin in Vietnamese fish. Science ISC ` (4083): 235-286 (April 20, 1973). 290 Environmental Health Perspectives 784622 GENP 011811 ish and shellfish f-rora areas of South Vietnam "that were heavily sprayed during the U. S. defolia tion campaign contained significant quantities of dioxin. This is a news report on the data R. Baugh man and M. Meselson presented to the NIEHS Con ference on chlorinated dibenzodioxins and dibenzorfifurans. Shapley, D. Herbicides: Agent Orange stockpile may go to the South Americans. Science 180(4081): 43--45 (April o, 1973). The U. S. Air Force has a surplus stockpile of 2,338,900 gal of Agent Orange (50 9c 2,4,5-T and O'r 2.4-D); some of these mixtures contain as much as 28 times the maximum acceptable safety limit of dioxin. Presently, dioxin concentrations per missible for new herbicides ar 0.1 ppm and 0.5 ppm for stocks already manufactured. Vos, J. G,, Moore, J7 A., and Zinkl, J. Effect of TCDD on the immune system of labo ratory animals. Environ. Health Per spect. (No. 5) : 149 (1973). TCDD at subiethal dose levels caused atrophy of the thymus, suppressed the cell-mediated immunity in both guinea pigs and mice, but did not affect the humoral immunity in guinea pigs. ^^N ldbott, G. L. Effects of Environmental . Pollutants. C. V. Mosby Co., St. Louis, Mo., 1973. Dioxin compounds are listed with other environ mental pollutants as examples of airborne pollu tants. These substances are contaminants of the popular weed killer, 2,4,5-T. Dioxins caused death and gastrointestinal hemorrhage in ra t fetuses when mochers were treated with doses of 0.125 to .8 ig. The mutagenic effects are due to its intercalation with DNA. Dibenzofurans were implicated as being responsible for some of the toxic effects attributed to PCBs. WEBBER, T. J. N,, and Box, D. G. The exam ination of tetrachlorvinphos and its formulations for the presence of tetrachlorodibenzo-p-dioxins by a gas-liquid chromatographic method. Analyst 98: 181-189 (1973). A gas-liquid chromatographic analytical method was developed for use in detecting tetrachlorodibenzo-p-dioxin impurities in samples of the insecti cide tetrachlorvinphos and its formulations. Tetra chlorvinphos is the Z- or trans-styrene isomer of 2chloro-l-(2.4,5-trichlorophenyi)vinyl dimethyl phos phate. Sequential use of silica gel and aluminum oxide column chromatography followed by concen- ed sulfuric acid treatment of the resultant eluate made it possible to detect TCDD levels down to a limit of 0.025 ppm with electron-capture detec tion. Tetrachlorodibenzo-p-dioxins were not detected in any of the 21 tetrachlorvinphos samples or any of its formulations selected for analysis either be fore or after accelerated storage at 55 9C for 2 weeks. Weissberg, J., and Zinkl, J. Effects of TCDD upon hemostasis and hematologic func tion in the rat. Environ. Health Persped. (No. 5) : 119 (1973). Daily oral doses of 10 ig/kg TCDD to female rats for 10 and 14 days caused nonspecific altera tions of hematopoietic function, thrombocytopenia, thromhocytoasthenia, and derangements in blood co agulation. Platelet-aggregation, bleeding time, and platelet factor III activity were normal, but clot re traction was abnormal. Wilson, J. G. Teratologieal potential of 2,4,5-T. Down to Earth 28(4) : 14-17 (Spring 1973). A chronology of the hazards of 2,4,5-T is pre sented. The report issued by the President's Science Advisory Committee appointed to study the 2,4,5-T question served as the basis for this review. The dioxin contaminant of 2.4,5-T, 2,3,7,8-tetrachlorodibenzo-/)-dioxin. is discussed briefly as a toxicogen. W oods. -J. S. Studies of the effects of 2,3,7,8tetrachlorodibenzo-p-dioxin on mamma lian hepatic 5-aminoIevulinic acid synthe tase. Enrirnn. Health Perspect. (No. 5) : 221 (1973). No differences in hepatic -aminolevulinic acid syn thetase activity were seen in control rats and those receiving orally 5, 25, or 100 Mg.'kg TCDD for up to 30 days. Mice and guinea pigs were also nonreactive. Thus, TCDD is not porphyrogenic in mam mals, even at several times the LD dose levels. Zinkl, J,, Moore, J. A., Vos, J. G.( and Gupta, B. N. Hematologic and clinical chemical effects of 2,3,7,8-tetrachlorodibenzodioxin in laboratory animals. En viron. Health Perspect. (No. 5 ): 111 (1973). TCDD-induced changes observed in female rats after 10 days treatment with 10 ^g/kg and 17 days at 1 Mg.'kg were increases in serum glutamic oxalo acetic transaminase. After 13 days treatment with 10 ,(ig, k g serum glutamic pyruvate transaminase was increased. Platelet depression was observed after 10 days at all dose levels. Zitko, V., Wildish, D. J . ; Hutzinger, O., and Choi, P. M. K. Acute and chronic oral beptember 1973 291 GENP 011812 784623 toxicity of chlorinated dibenzofurans to salmonid fishes. Environ. Health' Perspect. (No. 5): 187 (1973). Dry fish food was fed to juvenile Atlantic salmon contaminated with a mixture of 2.7 tg/g di-, 5,7 tri-, 2.8 tetra-, and 9.1 octachlorodibenzofuran. Med ian mortality was 12 30 days. Only octachlorodibenzofuran was found in tissues of dead fish (0.03 (tg/g in muscle and 0.2 ftg/g in the gut). Fish sur viving 140 days feeding contained corresponding values of 0.01 and 0.02 ftg/g. 1972 Abelson, P. H. Pollution by organic chemi cals. Our Chemical Environment, J. C. Giddings and M. D. Monroe, Eds., Canfield Press, San Francisco, 1972, Chap. 28, pp. 183-184. The most toxic chlorine-containing compound known is 2,3,7,8-tetrachlorodibenzodioxine (CuK^OCI.), often called dioxin. The acute oral LD*> dose in male guinea pigs is about 10-* g/kg. In spite of its toxicity, the behavior of dioxin in the food chain has not been worked out. Anonymous. TCDD residues disappear. Apr. Res. 21(4) : 8 (1972). The 2,4,5-T contaminant, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), can be formed during syn thesis of some chlorinated phenols if high tempera tures are used. Anonymous. Research heightens concern over PCB*s. Chem. Eng. News 50(10) : 27-28 (1972). Certain polychlorinated biphenyls may contain traces of tetrachlorodibenzofuran. A contaminant found in PCB's, not conclusively identified, has the 3ame mass spectrum as tetrachloro-p-dibenzofuran. Anonymous. Dioxin with a bang l Food Cosmet. Toxicol. 10(1): 110-111 (1972). During a manufacturing plant explosion, dioxin was formed by the interaction of sodium 2,4,5-trichlorophenate molecules under the influence of the exothermic decomposition of sodium 2-hydroxyethoxide. Anonymous. . . . B u t 2,4,5-T is in the dock again. Food Cosmet. Toxicol. 10(5) : 722 (1972). Negative teratogenic results have been reported for rats and rabbits treated with 2,4,5-T samples with very minute levels of dioxin impurity. Terato genic effects have been detected, however, in three mouse strains treated with 2,4,5-T containing as little as 0.05 or 0.5 ppm dioxin. Anonymous. 2,4,5-T and dioxins accused of teratogenicity. Food Chem. News 13 (43) : 24-27 (Jan. 17, 1972). Responding to challenges on 2,4,5-T cancellation, EPA concentrated its replies on lack of proof that 2,4,5-T and contaminants are not teratogens. The dose-response curves for 2,4,5-T and dioxin (TCDD) have not been determined, and the possibility of no effect levels is only a m atter of conjecture. Anonymous. FDA annual report shows in crease in enforcement. Food Chem. News 13(47): 20 (Feb. 14, 1972). In the herbicide section of its 1971 annual report, FDA said the toxicity of 2,4,5-T is chiefly due to dioxins. Anonymous. Senate settlement expected on pesticides bill. Food Chem. News 14 (27) : 53-54 (Sept. 25, 1972). A wide range of teratogenic dioxins can be pro duced both in the manufacture of 2,4,5-T and during pyrolysis (incomplete combustion). Boer, F. P., Neuman, M. A., and Aniline, O. 2,8-Dichlorodibenzo-p-dioxin. Acta Crystallogr. B28(9) : 2878-2880 (1972). Crystals of 2,8-dichlorodibenzo-p-dioxin are or thorhombic; molecuies are slightly nonplanar with an unusual packing arrangement. Boer, F. P., and North, P. P. Crystal and molecular structure of 2,7-dichlorodiben zo-p-dioxin. Acta Crystallogr. B2S(5) : 1613-1616 (1972). Three-dimensional single-crystal x-ray diffraction data revealed the crystal and molecular structure of 2,7-dichlorodibenzo-p-dioxin. The C-Cl bond dis tance is 1.742 A, the C-0 distances are 1.380 and 1.382, and the 6 C-C distances range between 1.370 and 1.397. The C-O-C angle in the heterocyclic ring is 116.3*. Boer, F. P., Van Remoortere, F. F,, and Muelder, W. W. Preparation and struc ture of 2,3,7,8-tetrachlorodibenzo-p-dioxin and 2,7-dichlorodibenzo-p-dioxin. J . Amer. Chem. Soc. 94(3) : 1006-1007 (1972). The preparation, isolation, and isometric struc tures of 2,3,7,8-tetrachloro and 2,7-dichlorodibenzop-dioxin are described. Boer, F. P., Van Remoortere, F. P., North, P. P., and Neuman, M. A. Crystal and *molecular structure of 2,3,7,8-tetrach- 292 Environmental Health Perspectives 784624 GENP 0 1 1 8 1 3 ( ' lorodibenzo-p-dioxin. Acta Crystallogr. ^ B28 (4) : 1023-1029 (1972). 2,3,7,8-Tetrach lorodibenzo-p-dioxin was studied by using three-dimensional single-crystal x-ray diffrac tion. The four unique C-CI distances range from 1.726 to 1.730 A, the 4 C -0 distances from 1.377 to 1,379, and the 12 C-C bonds are all between 1.374 and 1.388. The C-O-C angles are 115.6 and 115.8. Brenner, K. S., Muller, K., and Sattel, P. Detection and determination of 2,3,7,8tetrachloro dibenzo-p-dioxin in chlorosubstituted phenoxyalkane acids. J. Chromatogr. 64: 39-48 (1972). Dioxin separation from the herbicide phenoxyaikane acids was accomplished by extractive distil lation of the potassium salts with n-hexane in the Bleidner apparatus. Quantitative determination of the hexane extracts for dioxin was done by gas chromatography. Sensitivity was about 0.1 ppm. Brooks, G. T. Pesticides in Britain. In: En vironmental Toxicology of Pesticides, F. Matsumura, G. M. Boush, and T. Misato, Eds., Academic Press, New Ydrk, 1972, pp. 61-114. The reported teratogenic effects of 2,4,5-T in ,,^ ^ a m a ls and the isolation of the highly toxic Cott le jmant tetrachlorodibenzo-p-dioxin are cautionary, 2,4,5-T has been used in Europe for nearly 15 years without: evidence of ill effects. More studies are needed in comparative detoxication between mammals, birds, fishes, and insects. Brownrigg, J. T., Eastwood, D., and Hornig, A. W. Identification of polychlori nated biphenyls in the presence of DDTtype compounds. Office of Research and Monitoring, U.S. Environmental Protec tion Agency, Washington, DC (Oct. 1972), EPA-R2-72-004. Low temperature (77K) luminescence tech niques could in principle be applied to a wide variety of compounds including the highly toxic chlorinated dibenzofurans and dibenzo-p-dioxins. Buu-Hoi, N. P., Chanh, P-H., Sesque, G.. ? zum-Gelade, M. C., and Saint-Ruf, G. Enzymatic functions as targets of the toxicity of "dioxin" (2,3,7,8-tetrachlorodibenzo-p-dioxin). Naturwiss. 59(4): 173-174 (1972). Following a single I.P dose of 10 mg/kg dioxin to rats, deep pertubations in several enzymatic sys tems were observed, i.e., a decrease in cholinesterase ' n increase in serum glutamic oxaloacetic acid transaminase. Effects in homeostasis indicate the liver is one of the main targets for dioxin intoxica tion. Buu-Hoi, N. P., Chanh, P-H., Sesque, G., Azum-Gelade, M.C.; Saint-Ruf, G. Or gans as targets of "dioxin" (2,3,7,8-tetrachlorodibenzo-p-dioxin). Naturwiss. 59(4) : 174-175 (1972). Organ damage was found along with weight loss and hematologic effects following IP injection of 1 and 10 mg/kg dioxin to rats. Ten days after treat ment, damage was observed to the liver, thymus, heart. Less damage was observed in the lungs and blood cells. Buu-Hor, N. P., Saint-Ruf, G., and Mangane. M. Fragmentation of dibenzo-pclioxin and its derivatives under electron impact. J. HeterocycL Chem. 9(3): 691693 (1972). Mass spectra were reported and discussed on five chloro derivatives of dibenzo-p-dioxin: 2,7-dichloro-, 1,6-dichloro-, 1,3,6-trichloro-, 2,3,7,8-tetrachloro-, and 1,2,3,4,6,7,8,9-octachlorodibenzo-p-dioxin. Chen, J. T. Infrared studies of chlorinated dibenzo-p-dioxins and structurally re lated compounds. Division of Pesticide Chemistry, 164th National Meeting of the American Chemical Society, New York. Aug. 27-Sept. 1, 1972, Abstract No. 12. Data are reported on the reference infrared spectra of 24 chlorinated dibenzo-p-dioxins and the observed characteristic frequencies are tabulated. Courtney, K. D. The teratogenic evaluation of the herbicide 2,4.5-T and dioxin. Item 4.0785, Page 1-632. Part 4. Pesticides, Environmental Protection Research Catalog, Research Information Division, U.S. EPA, Washington, D.C., Jan. 1972. Phenoxyacetic acid herbicides and dioxins are be ing evaluated for teratogenic and perinatal toxic effects. Crosby, D. G., Moilanen, K. W., Nakagawa, M., and Wong, A. S. Photonucleophilic reactions of pesticides. Environmental Toxicology of Pesticides, F. Matsumura. G. M. Bousch, and T. Misato, Eds., Aca demic Press, New York, 1972, pp. 423433. Photonucleophilic displacement of chloride by chlorophenoxide ion in o-chlorophenols introduced .^tem ber 1973 293 784625 the possibility of the photochemical generation of chlorinated dibenzo-p-dioxins. Sunlight wavelengths exposure to sodium pentachlorophenate yielded 1,2,3,4,6,7,8,9-oetachlorodibenzo-p-dioxin; no 2,3,7,8TCDD resulted from irradiated 2,4,5,-T or 2,4,5-trichlorophenol. Cunningham, H. M., and Willu m s, D. T. Effect of tetrachlorodibenzo-p-dioxin on growth rate and the synthesis of lipids and proteins in rats. Bull. Environ. Con tain. Toxicol. 7 (1 ): 45-51 (1972). Protein concentration in the rat liver was slightly reduced 3 days after dioxin treatment. This de crease was accompanied by a significant increase in "C-leucine incorporation into liver proteins. The lowest single dose of dioxin that caused an increase in rat liver weight was 0.1 /ig/kg Curley, A., Burse, V. W., and J ennings, R. Metabolite or contaminant of Aroclor 1254 found in rat urine. Division of Pesticide Chemistry, 163rd National Meeting of the American Chemical So ciety, Boston, April 9-14, 1972, Abstract No. 5. Mass spectra of urine collected from rats on die tary levels of 100 or 500 ppm Aroclor 1254 for intervals up to eight months indicated the presence of a molecular ion at m/e 304 and the characteristic isotopic cluster of 4 chlorine atoms. Others have shown the presence of tetra and pentad ibenzofurans (mass'numbers 304 and 338) as contaminants in the foreign products. Environmental Protection Agency, P ub lications and Information Section. Chemistry and residues of 2,4,5-T (in cludes dioxins). Bibliography Number 72-59 (1972). A reference list of 43 references is given. E nvironmental Protection Agency, Pub lications and Information Section. Toxicology and pharmacology of 2,4,5-T (includes dioxins). Bibliography Num ber 72-58 (1972). A list of 75 references is given. Epstein, S. S. Environmental pathology. A review. Amer. J. Pathol. 66(2) : 352-373 (1972). Toxicity testing must not be confined to the test agent per se, but should be extended to its chemical and metabolic derivatives, its pyrolytic and degrada tion products and its contaminants and reaction products, especially when various derivatives or de- gradation products are of toxicologic or environ mental consequence. Dioxin pyrolytic products in phenoxy herbicides are illustrative. F irestone, D., Ress, J., Brown, N. L,, Bar ron, R. P., and Damico, J. N. Determi nation of polychlorodibenzo-p-dioxins and related compounds in commercial chlorophenols. J. Assoc. Ofiic. Anal. Chem. 55(1) : 85-92 (1972). Twenty-one commercial chlorophenols were dis solved separately in aqueous alkali, extracted with petroleum ether, fractionated on an alumina col umn, and examined by electron capture gas chrom atography and combined gas chromatography-mass spectrometry. The 2,3,7,8-tetrachlorodioxin was found in 3 of 6 samples of 2,4,5-trichlorophenol but not in any of the I t samples of tetra- and pentachlorophenol. Hexachlorodioxin, present in all 8 pentachlorophenols tested, ranged from 0.17 to 39 ppm. Hexa-, hepta-, and octachlorodioxins and chlorofurans were present in most of the tetra- and pentachlorophenols. Fishbein, L. Human directed aspects of PCBs. In: Polychlorinated Biphenyls and the Environment, Report No. IT F PCB-72-1, (COM-72-10419), Inter departmental Task Force on PCBs, Washington, D.C., May, 1972, pp. 122151. Work reported previously in the literature was summarized. Emphasis was placed on the tetra- and pentachlorodibenzofuran impurities in commercial PCB products; the tri- and tetrachlorodibenzofuran single oral liver necrotic dose (0.5 to 1.0 mg/kg) inrabbits; and the 2,3,7,8-tetrachlorodibenzo-p-dioxin contaminant in 2,4,5-T and 2,4,5-trichlorophenol caused lethal liver necrosis and chloracne in rabbits at a dose range of 0.05 to 0.1 mg/kg. FISHBEIN, L. Chromatographic and biological aspects of polychlorinated biphenyls. J. Chromatog. 68(1) : 345-426 (1972). Chromatographic methods of analysis for chlori nated dibenzofurans were reviewed as part of a more extensive review of polychlorinated biphenyls. F ishbein, L., and F lamm, W. G. Potential environmental chemical hazards. P art II. Feed additives and pesticides. Sd. Total Environ. 1: 31-64 (1972). The action of alkali on 1,2,4,5-tetrachlorobenzone, a by-product from lindane synthesis, produces 2,4,5trichlorophenol which, when interacted with sodium monochloracetate, yields 2,4,5-T. The chloracnegen. 2,3,7,3-tetrachlorodibenzo-p-dioxin, is an impurity produced in the manufacture of 2,4,5-T; 1,2,4,5-tetrs- 294 Environmental Health Perspectives 784626 G E N P 011815 orobenzene under the influence of high pressure ' i'i temperature, methanol, and sodium hydroxide, is converted to sodium trichlorophenate which re acts with another molecule of sodium trichlorophen ate which reacts with another molecule of sodium trichlorophenate with high temperatures to form TCDD; or two molecules of trichlorophenol combine to form TCDD. Goldmann, P. J. Severe acute chlorine acne caused by trichlorophenol decomposition products: A contribution to the pema problem. Arbeitsmed. Sozialmed. Arbeitshyg. 7 (1 ): 12-18 (1972) (Ger.). Occupational and case histories are reported de scribing 2,3,6,7-tetrachlorodibenzodioxin as the causative agent in 42 cases of serious skin changes, 14 cases of internal organ damage, and 7 cases of nervous system disturbances. Greig, J. B. Effect of 2,3,7,8-tetrachlorodibenzo-1 ,4-dioxin on drug' metabolism in the rat. Biockem. Pharmacol. 21(23): 3196-3198 (1972). Rats given a single oral 200 /ig/kg dose of dioxin exhibited a decreased duration of zoxazolamine (100 mg/kg IP) induced paralysis by 54%. After 200 g/kg oral dioxin, sleeping time induced by hexa- irbitat, 150 mg/kg (male rats) or 75 mg/kg (fe- aie rats), was prolonged--more than double at 3 ,Lys. These results indicate that dioxin has simul taneous stimulatory and inhibitory effects on dif ferent pathways of oxidative drug metabolism in the rat liver. Hammond, A. L. Chemical pollution: poly chlorinated biphenyls. Science 175 (4018) : 155-156 (Jan. 14, 1972). Some commercial PCB mixtures, especially those manufactured in Europe or Japan, may contain trace amounts of dibenzofurans or other toxic im purities. These may be the cause of some toxic effects of the PCBs. H ussain, S., E hrenberg, L., Lofroth, G., and Gejvall, T. Mutagenic effects of TCDD on bacterial systems. Ambio 1(1) : 32-33 (1972). Results from three distinct bacterial assay sys tems showed TCDD to be mutagenic: (1) reversion to streptomycin independency in E. coli SD--1, (2) reversion to histidine prototrophy in Salmonella typhimurium strains, and (3) prophage induction in E. coli K-39. The results indicated that an acridine like behavior of DNA intercalation may have caused these genetic effects. Huston, B. L. Identification of three neutral contaminants in production grade 2,4-D. J. Agr. Food Chem. 20(3) : 724-727 (1972)-.- Three chemical impurities of 2,4-D are profiled. These contaminants interfere with the gas-liquid chromatographic analysis of 2,4-D for 2,3,7,8-tetrachlorodibenzo-p-dioxin. J ackson, W. T. Regulation of mitosis. III. Cytological effects of 2,4,5-trichlorophenoxyacetic acid and of dioxin contami nants in 2,4,5-T formulations. J. Cell Sci. 10: 15-25 (1972). Inhibition of mitosis and development of cytolog ical abnormalities observed in dividing endosperm cells of the African blood lily were believed caused by 2,3,7,8-tetrachlorodibenzo-p-dioxin, a contaminant of 2,4,5-T, rather than the herbicide itself. In con trast to 2,4,5-T, which has no effect, dramatic in hibition of mitosis was observed in cells subjected to 0.2 to 1.0 iig/1 dioxin, 0.2 /ig/l dioxin plus 10~*M 2,4,5-T, or lO'Wf 2,4,5-T containing dioxin as a contaminant. These preparations also induced for mation of dicentric bridges and chromatin fusion with formation of multinuclei or a single large nu cleus. J ensen, S. The PCB story. Ambio 1(4) :123131 (1972). The work of Vos et. al. is reported. Tetra- and pentachlorodibenzofuran impurities in commercial PCBs may have been formed from a phenolic con taminant in PCB in similar manner as 2,3,7,8-tetrachlorodibenzo-p-dioxin is formed from 2,4,5-T. J ensen, S., and Renberg, L. Contaminants in pentachlorophenol : chlorinated diox ins and predioxins (chlorinated hydroxydiphenylethers). Ambio 1 (2 ): 62-65 (1972). Octachlorodioxin and 2-hydroxynonachlorodiphenyl ether (labeled predoxin) were identified in technical pentachlorophenate using ion exchange, diazome thane treatment, gas chromatography-mass spec trometry, and thinlayer chromatography. Predioxins have not been discovered or detected previously be cause; (1) clean-up procedures remove the sub stance, e.g., alumina column chromatography or concentrated sulfuric acid treatment of extract, (2) predoxin spontaneously forms dioxin in a gas chromatograph. Kearney, P. C., Woolson, E. A., and El lington, C. P. Persistence and metabo lism of chlorodioxins in soils. Environ. Sci. Tecknol. 6(12) : 1017-1019 (1972). Persistence of 1,10, or 100 ppm 2,3,7,8-tetrachIorodibenzo-p-dioxin (TCDD) was determined by elec tron-capture gas chromatography in soils after 20, dept ember 1973 295 784627 40, 80, 160, and 350 days. After 1 yr, 56-63% TCDD was recovered. Neither 2,7-dichlorodibenzo-pdioxin (DCDD) nor TCDD was detected in soils~receiving 10, 100, or 1000 ppm 2,4-D or 2,4,5-T after 70 days. TCDD is degraded slowly in soils. TCDD and DCDD are not biosyntheaized by microbial con densation reactions. Khera, K. S., and McKinley, W. P. Preand postnatal studies on 2,4,5-trichiorophenoxyacetic acid, 2,4-dichlorophenoxyacetic acid and their derivatives in rats. Toxicol. Ajrpl. Pharmacol. 22: 14-28 (1972). 2,4,5-T (containing less than 0.5 mg/kg 2,3,7,8tetrachlorodibenzo-p-dioxin) induced fetopathy, and skeletal anomalies in progeny from females treated with a single daily oral dose of 100 to 150 mg/kg on gestation days 6 to 15. Number of conceptions and numbers of viable and dead fetuses per litter gave .no indication that in utero treatment of offspring with up to 100 mg/kg 2,4,5-T had impaired fertility. Kimbrough, R. D. Toxicity of of chlorinated hydrocarbons and related compounds. A review including chlorinated dibenzodioxins and chlorinated dibenzofurans. Arch. Environ. Health 25(2) : 125-131 (1972). Trace amounts of chlorinated dibenzofurans and dibenzodioxins were identified as contaminants in many chlorinated technical compounds, e.g., 2,4,5trichlorophenol, 2.4,5-trichorophenoxyacetic acid (2,4.5-T), and European chlorinated biphenyls (Phenoclor DP6 and Clophen A60). Toxic fat con taining chick edema factor also contained chlorinated dibenzodioxins. Kins, N,, and Barandy, J. Short method for the detection of chick edema factor in fats, oils, and fatty acids by electron capture gas chromatography. J. Amer. Oil Chem. Soc. 49(2): 115-117 (1972). A modification of the official electron capture-gas chromatographic AOAC assay for chick edema fac tor. e.g., hexa-, hepta-, and octachlorodibenzo-p-dioxins. was presented to shorten the assay time and eliminate problems in the alumina column frac tionation. In place of the alumina fractionation following 2,2,4-trimethyIpentane extraction, a sec ond sulfuric acid treatment and a caustic wash were done before the final sulfuric acid wash. Kraybill, H. F. Chemical and physical prop erties of FCBs. InP o ly ch lo rin ated Bi llhenyIs and the Environment, Report No. ITF-PCB-72-1, (COM-72-10419), Interdepartmental Task Force on PCBs, Washington, D.C. (May 1972). pp. 2240. Tetrachlorodibenzofuran (mass number 304) and pentachlorodibenzofuran (mass number 338) were identified from a fractionated PCB sample. Maier-Bode, H. Contribution to 2,4,5-T ques tion. A m . Schaedlingskd. Pflanzenschuiz 45(1) : 2-6 (1972) (Ger.). The teratogenic effect of 2,4,5-T in an earlier U.S. experiment is attributable to the content (30 ppm) of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the test sample. Martin, H. (E d.). Pesticide Manual. Basic Information on the Chemicals used as Active Components of Pesticides, 3rd. ed. British Crop Protection Council, Worcester, England (1972). A contaminant of 2,4,5-T (2,3,7,8-tetrachlorodibenzo-p-dioxin) caused serious acne in man and pro duced fetal death in hamsters at 9.1 #ig/kg. Modern methods of synthesis now limit the dioxin concen tration in 2,4,5-T to less than 0.5 ppm. Matsumura, F. Biological effects of toxic pesticidal contaminants and terminal residues. In: Environmental Toxicology of Pesticides, F. Matsumura, G. M. Boush, and T. Misato, Eds., Academic Press, New York, 1972, pp. 525-548. Chlorinated dibenzo-p-dioxins have been classified as terminal residues-chemicals which accumulate in biologic material in the environment as a result of pesticide introduction. The presence of terminal residues in the environment is due to stable pesti cides, conversion products, and chemical impurities (dioxins) that remain in the environment longer than the principal pesticides. The importance of such terminal residues in relation to the final mag nitude of environmental hazardousness needs fur ther definition. Neubert, D., and Dillmann, I. Embryotoxic effects in mice treated with 2,4,5-tri- chlorophenoxyacetic acid and 2,3,7,8-tet- rachlorodibenzo-p-dioxin. Naunyn- Schmiedeberg's Arch. Pharmacol. 272: 243-264 (1972). Oral doses of 1 ng/kg 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDBD) in NMRI mice on days 6 to 15 of gestation produced cleft palate; embryo mortal ity was significant for doses of 4.5 jig/kg or more. Cleft palate without pronounced embryo lethality was produced in mice with high single doses (2050 fig/kg) TCDBD given between days 7 and IS of pregnancy. A maximum teratogenic effect was 296 Environmental Health Perspectives 784628 G ]yp U 817 vjserved on day 11 of gestation, but another peak cleft palate frequency occurred when TCDBD -i?as given on day 8 of pregnancy. N e u m a n , M. A., N o r th , P. P,, and Boer, F. P. Crystal and molecular structure of octachlorodibenzo-p-dioxin. Acta Crystallogr. B28(8) : 2313-2317 (1972). Three-dimensional single-crystal x-ray diffraction was used to elucidate the crystal and molecular structure of octachlorodibenzo-p-dioxin. The four unique C-Cl distances ranged from 1.714 to 1.718A, the two C -0 distances were 1.373 and 1.374, and the six C-C bonds ranged between 1.382 and 1.396. The C-O-C angle was 115.8a. N orback, D. H., and A l l e n , J. R. Chlorinated aromatic hydrocarbon induced modifica tions of the hepatic endoplasmic retic ulum : concentric membrane arrays. En viron. Health Perspect. 1: 137-143 (1972). Separate groups of rats were fed diets containing 17c highly chlorinated triphenyls (PCTs, Arador 5460), 0.027c polychlorinated biphenyls (PCBs, Arochlor 1254), and 0.0027'c chlorinated diphenyi-p-dioxin. The dioxin and PCT groups ate readily and attained 80?c of control weight in 3 weeks. Liver hypertrophy varied from moderate enlargement (di- group) to an increase in relative liver weight f 3,TOO body weight for the PCB and PCT groups. ' - o h la n d , A. E., and Y an g , G. C. Prepara tion and characterization of chlorinated dibenzo-/>-dioxins. /. Agr. Food Chem. 20(6) : 1093-1099 (1972). A series of 13 chlorinated dibenzo-p-dioxins (nine of these were new and previously unreported) was prepared containing from 1 to 8 chlorine atoms to provide pure standards of the various chlorinated . derivatives for use in methods development and tox icology studies. Synthesis, utility, yield, purity, and physical and chemical properties were presented and discussed. Stability, color reactions, and in frared, ultraviolet, nuclear magnetic resonance, and phosphorescence spectra were also reported. Phos phorescence and triplet state lifetime wavelengths were dependent on the number of chlorine atoms and their positions on the dibenzo-p-dioxin nucleus. R a p p e , C., and N ilsso n , C.-A. An artifact in the gas chromatographic determina tion of impurities in pentachlorophenol. -/. Chromatogr. 67: 247-253 (1972). The main impurity of commercial samples of pen tachlorophenol was 3,4,5,6-tetrachIoro-2-(2,3,4,5,pentachlorophenoxy) phenol ; this compound under went ring closure during gas chromatography to 1,2,3,4,6,7,8,9-octachIorodibenzo-p-dioxin, Saint--Ruf, G. Formation of "dioxin" in the pyrolysis of sodium a-(2,4,5-trichlorophenoxy) -propionate. Naturiuiss. 59 (12) : 648 (1972). A residue of 2,3,7,8-tetrachlorodibenzo-p-dioxin re sulted from the pyrolysis (500'C for 5 hr) of sodium-a-(2,4,5-trichlorophenoxy)-propionate, a com pound used in the synthesis of Sylvex. The dioxin impurity yield was much less than th at observed during 2,4,5-T synthesis. Task F orce. Polychlorinated biphenyls and the environment. Report No. IT F -F C B 72-1, (COM-72-10419), Interdepart mental Task Force on PCBs, Washing ton, DC (May, 1972). This report emanated from a six-month review on the polychlorinated biphenyls by five Federal agencies. The main emphasis was PCBs; however, numerous statements and discussions centered on chlorinated dibenzodioxins and dibenzofurans. For example, fractionated samples of some PCBs of foreign manufacture contained as contaminants tetra- and pentachlorodibenzofurans. The task force recommended: (1) toxicological evaluation of a se lect number of representative, purified PCB isomers as well as purified trace contaminants such as the chlorinated dibenzofurans; (2) definitive mammalian elaboration of the kinetics, absorption, distribution, metabolism, and excretion of the technical PCBs as well as a number of key isomers and the chlorinated dibenzofurans; and (3) elaboration of the subcellular and intracellular actions of the technical PCBs as well as a number of representative isomers and chlorinated dibenzofurans. Vasiliu, G., and Baciu, I. Acetylation of some 2,7-dihalodibenzo-p-dioxins. Rev. Chim. (Bucharest) 23(9) : 523-525 (1972) (Rom.). Reaction of 2,7-dichlorodibenzo-p-dioxin with chloroacetic acid in the presence of aluminum chlo ride and carbon disulfide gave the corresponding 3monoacetyl derivative. Confirmation was by infrared spectra. Vos, J. G. Toxicology of FCB's for mammals and for birds. Environ. Health Perspect. 1: 105-117 (1972). Polychlorinated dibenzofurans or other toxic im purities in crude PCB preparations caused difficul ty in interpreting toxicity studies. Vos, J. G. Toxicology of polychlorinated bi phenyls (PCB's) and impurities. Tijdschr. Diergeneesk. 97(22) : 13781385 (1972) (Neth). ; ptember 1973 297 GENP 011818 784629 The moat toxic impurities in polychlorinated bi phenyls were found to be chlorinated benzofurans. Vos, J. G., and Notenboom-Ram, E. Com parative toxicity study of 2,4,5,2',4',5',hexachlorobiphenyi and a polychlorinat ed biphenyl mixture in rabbits. Toxicol. Ayyl. Pharmacol. 23: 563-578 (1972). The major acnegenic action of crude PCB mix tures originates from chlorinated dibenzofurans. The probable contribution of chlorinated dibenzofuran and pure PCB to the toxicity of technical PCB preparations in rabbits is discussed: chloracne ( --furan + PCB + ), edema formation ( + -r, --), liver damage ( + -r, -p), and hepatic porphyria ( - . - h - r ). Warmbrunn, K. Considerations regarding the ban or limitation on the use of some pesticides imposed July 23, 1971. Gesunde Pffanz. 24(1) : 6-8 (1972) (Ger). The dioxin content of 2,4,5-T produced in Ger many is approximately 1 ppm, a level at which mal formation hazards are practically eliminated. Williams, C. S. The current status of phenoxy herbicides. Down to Earth 27(4) : 20-24 (Spring1 1972). A chronology of events from April 13, 1966 to December 6, 1971 pertaining to phenoxy herbicides is presented, with particular attention devoted to TCDD. Williams, D. T., and Blanchfield, B. J. Screening method for the detection of chlorodibenzo-p-dioxins in the presence of chlorobiphenyls, chloronaphthalenes, and chlorodibenzofurans. J. Assoc. Offic. Anal. Chem. 55(1): 93-95 (1972). The chlorodibenzo-p-dioxins were chlorinated to octachlorodibenzo-p-dioxin and identified by electron capture-gas-liquid chromatography. 2,3,7,8-Tetrachlorodibenzo-p-dioxin was determined at the 1 ppm level in corn oil. This method should be considered as a screen for dioxins, since it does not distinguish among the large number of isomers. Williams, D.' T.( Cunningham, H. M,, and Blanchfield^B. J. Distribution and ex cretion studies of octachlorodibenzo-pdioxin in the rat. Bull. Environ. Contain. Toxicol. 7(1). 57-62 (1972). Absorption, distribution, and excretion of octachlorodibenzo-p-dioxin in the ra t were examined. The AOAC method was used to analyze organs and tissues for dioxin content. The only gross pathology observed was congestion of the liver. Octachlorodi- benzo-p-dioxin was absorbed by the rat mainly in the liver with small amounts in the adipose tissue and bile. Williams, D. T., and Blanchfield, B. j . Improved screening method for chloro dibenzo-p-dioxins. J. Assoc. Offic. Anal. Chem. 55(6): 1358-1359 (1972). Conversion of 2,3,7,3-tetrachlorodibenzo-p-dioxin to octachlorodibenzo-p-dioxin allowed detection by electron capture-gas-liquid chromatography of as low as 0.05 ppm in corn oil. The author's previous method was modified to include different chlorination conditions and a later alumina column cleanup. Wilson, J. G. Teratological potential of 2,4,5-T. Proceedings Twenty-Fifth An nual Meeting Southern Weed Science Society, Dallas, TX, Jan. 18-20, 1972. A chronology of the hazards of 2,4.5-T was pre sented. The report issued by the President's Science Advisory Committee appointed to study the 2,4,5-T question served as the basis for this review. The dioxin contaminant of 2,4,5-T, 2,3,7,8-tetrachlorodibenzo-p-dioxin, was discussed briefly as a toxicogen. Woolson, E. A., Thomas, R. F., and E nsor, P. D. J. Survey of polychlorodibenzo-pdioxin content in selected pesticides. J. Agr. Food. Chem. 20(2): 351-354 (1972). Electron capture-gas chromatography was used to examine 129 samples of 17 different pesticides de rived from chlorophenols for polychlorinated dibenzo-p-dioxins. Clean-up entailed concentrated sul furic acid extraction of impurities from hexane and mild nitration of the chlorophenol, extracts. 2,3,7,8Tetrachlorodibenzo-p-dioxin (TCDD) was detected in the samples analyzed: 76(7c contained less than 0.1 Mg/g, 79r between 0.1 to 1.0 Mg/g, and had greater than 10 Mg/g. The 20 tri-, tetra-, and pentachlorophenola examined contained no TCDD. Phenoxy herbicides from current production had less than 0.5 Mg/g TCDD. Yang, G. C., and P ohland, A. E. Electron spin resonance studies of cation radicals in trifluoromethane sulfonic acid. J. Phys. Chem. 76(10) : 1504-1505 (1972). Electron spin resonance spectra were obtained for chlorinated dibenzo-p-dioxins in trifluoromethane sulfonic acid (M SA); no oxidizing agents were used to produce the spectra in sulfuric acid. An oxidizing agent or ultraviolet light was needed for octachloro dibenzo-p-dioxin to yield a spectum in MSA; no spectrum was shown in sulfuric acid. ZlTKO, V. Absence of chlorinated dibenzodi- 298 Environmental Health Perspectives 784630 g ENP0118i 9 oxins and djbenzofurans from aquatic animals. Bull. Environ. Contain. Toxicol. 7(2/3) : 105-110 (1972). Chlorinated dibenzodioxin and dibenzofuran resi dues were not detected in any of the aquatic samples analyzed. Zitko, V., Hutzinger, 0., and CHOI, P. M. K. Contamination of the Bay of FundyGulf of Maine area with polychlorinated biphenyls, polychlorinated terphenyls, chlorinated dibenzodioxins, and dibenzofurans. Environ. Health Perspect. 1: 4750 (1972). No residues of chlorinated dibenzodioxins and di- benzofurans were found in the samples analyzed: muscle and liver of white shark, eggs of double- crested cormorants and herring gulls, commercial herring oil, and ground fish herring meal. Detection limits ranged from 0.01 octachlorodibenzofuran wet tissue to 0.04 ^g/TCDD g wet tissue. 1971 Anonymous. A close look at T'CDD. Agr. Res. 20(4) : 8-10 (1971). The toxicological properties of tetrachlorodiben>-p-dioxin were investigated following the discovery iat the mutagenic and teratogenic potential of 2,4,5-T may be due to this contaminant. Anonymous. Herbicides: More research on 2,4.5-T. Ckem. Eng. News 49(20): 11 (1971). The President's Science Advisory Committee's re port on 2,4,5-T and its dioxin impurity is reviewed. The panel made a number of recommendations, in cluding one asking for a mechanism that would temporarily restrict the use of certain registered pesticides on the basis of information that impli cates the chemical as a possible health hazard, pend ing the collection of more conclusive information. Anonymous. Herbicides: Secret 2,4,5-T re port. Chem. Eng. Neivs 49(29) : 15 (1971). An advisory committee's report submitted to the Environmental Protection Agency concerning 2.4, 5-T and its dioxin contaminant was criticized by a dissenting advisory committee member and the Com mittee for Environmental Information. Specifically stated was that a level at which TCDD is not tera togenic has not been established and that the report did not consider the consequences of the fate of TCDD in food chains and animal tissue. ^Anonymous. A taste of honey (flavoured with 2,4.5,-T). Food Cosmet. Toxicol. 9: 1 5 r(1 9 7 l). This is a commentary on the teratogenic evalua tion of 2,4,5-T by K. D. Courtney, et al.; a note added in proof indicate the 2,4,5-T contained 30 ppm dioxin. Thus, conclusions from the paper label ing 2,4,5-T teratogenic must be considered tentative. Anonymous. The FCB story unfolds. Food Cosmet. Toxicol. 9(4): 568-571 (1971). Lesions resembling those produced by chick edema factor were traced to a PCB contaminant, chlori nated dibenzofurans which is present in some com mercial PCBs. Anonymous. Working with 2.4,5-T. Food Cosmet. Toxicol. 9: 908-909 (1971). Chloracnc, characterized by inclusion cysts, come dones, and pustules, was found in 13 workers, and was correlated in severity with the presence of scarring, hyperpigmentation, hirsutism, and com plaints of eye irritation. A. P. Poland et al. com mented that chloracne was not correlated with oc cupations within plants manufacturing 2,4,5-T. du ration of employment, or coproporphyrin excretion. Anonymous. Tetrachlorodibenzodioxin--in timations of carcinogenicity? Food Cos met. Toxicol. 9: 909 (1971). Although such effects as described by Buu-Hoi. et al. of the carcinomimetic activity of TCDD exist, the relationship is by no means exclusive. For example, stimulation of hydrexyiating enzymes can be induced by many compounds, including BHT, which has been shown not to be carcinogenic in long term feeding studies. Anonymous. PSAC hiccoughs over 2,4,5-T. Nature 231(5300) : 210-211 (1971). The President's Science Advisory Committee (PSAC) report on 2,4.5-T is questioned: justifica tion by the PSAC concerning permissable levels of dioxin in 2,4,5-T was not based on scientific review; also, the need for establishing a dose-response curve for the teratogenicity of 2,4,5-T should have been cited. Anonymous. 2,4,5-T report attacked. Na ture 232(5308) : 218 (1971). The response of the Committee for Environmental Information (CEI) to the scientific advisory com mittee's recommendation that 2.4,5-T be restored was one of strong opposition. With respect to dioxins, the CEI indicated that dioxins may accumulate in the soil and that present analytical techniques were not sensitive enough to detect them. Furthermore, lack of concrete proof that the use of 2,4,5-T was correlated with the occurrence of birth defects in Vietnam did not mean -that no correlation existed. September 1973 299 784631 Ba u g h m a n , R. W., and Me se l so n , M. An improved analysis for 2.3.7.8-tetrachlorodibenzo-p-dioxin (TCDBD). Division Pesticide Chemistry, 162nd National Meeting of the American Chemical So ciety, Washington, D.C., Sept. 12-17, 1971, Abstract No. 89. Current analytical methods for the detection of potentially hazardous levels of 2,3,7,8-tetrachlorodibsnzo-p-dioxin are inadequate. A combined gas chro matography-mass spectroscopy method was utilized with modifications in extraction concentration and detection. B ev en u e, A., and K aw ano, Y. Pesticides, pesticide residues, tolerances, and the law (U.S.A.). Residue Rev. 35: 103-149 (1971). Politics and science surround the 2,4,5-T and hence dioxin controversy. Contents include: Federal In secticide, Fungicide, and Rodenticide Act (FIFRA), as amended through 1969; the DDT episode; the 2,4.5-T episode: Federal Food, Drug, and Cosmetic Act (FDCA); and food imports and international pesticide control. Mention is made of the USDA's investigation of possible excessive contamination of 17 polychlorophenolic pesticides by tetrachlorodibenzo-p-dioxin. Boer, F. P., N e u m a n , M. A., van R emoort e r e , F. P., N orth, P. P., and R i n n , H. W. X-ray diffraction studies of chlori nated dibenzo-p-dioxins. Division Pesti cide Chemistry, 162nd National Meeting of the American Chemical Society, Wash ington, D.C., Sept. 12-17, 1971 : Abstract Mo. 32.. The crystal structures of four chlorinated dibenzo-p-dioxin compounds (2,7-dichloro; 2.8-dichioro ; 2,3,7,8-tetrachloro ; and octachloro) were determined by x-ray diffraction. The structures were elucidated in order to provide absolute standards for isomeric structure assignment. Buu-Hoi, N. F., H ie n , D. P., S a in t -R u f , G., and Servoin-S idoine, J. Canceromimetic properties of tetrachloro-2,3,7,8 dibenzo-p-dioxin ("dioxin" ). C. R. Acad. Sci. (Paris) D272(10) : 1447-1450 (1971) (F r). TetrachIoro-2,3.7.8-dibenzo-p-dioxin in rats caused induction of zoxazolamine hydroxylase, marked re duction in phnobarbital sedative effect, and a de crease in hepatic arginase. Dioxin therefore, has a pronounced inhibitory effect on the enzymatic sys- terns as do the carcinogens benzo [a]-pyrene and pdimethylaminoazobenzene. Bu u -H ot, N. P .f Sa in t -R u f , G., B igot, P., and Mangane, M. Preparation, pronerties and identification of "dioxin" in pyrolysates of defoliants containing 2.4,5-trichlorophenoxyacetic acid, thenesters, and contaminated plants. C. 2. Acad. Sci. (Paris) D273(7): 708-711 (1971) (F r). Synthesis and physico-chemical properties of 2, 3,7,8-tetrachlorodibenzo-p-dioxin and its less toxic isomer l,3.6.8-tetrachlorodibenzo-p-dioxin are de scribed. Caution was urged about the burning of wooded areas treated with 2,4,5-T or its butyl ester because of possible dioxin formation. Ca m pb ell. A. D., and F ir e st o n e , D, Chick edema factor--toxic dioxins. I n : Interna tional Symposium on Identification and Measurement of Environmental Pollu tants, June 14-17, 1971, Ottawa, On tario, Canada, B. Wesley, Ed., National Research Council, Canada, 1971, pp. 195-198. The formation, occurrence, toxicity, tissue distri bution in the chick, and analytical methods for the detection of polychlorodibenzo-p-dioxins are de scribed. Dioxins were implicated in chick edema disease. Dioxins belong to one of the most toxic classes of chlorinated compounds known, being al most 10n0 times as toxic as most pesticides. Co l lin s. T. F . X., W il l ia m s , C. H ., and Gray, G. C. Teratogenic studies with 2,4,5-T and 2,4-D in the hamster. Bull. Environ. Contam. Toxicol. 6(6): 559567 (1971). 2,4,5-T commercial samples given on days 6 to 10 of organogenesis were feticidal and teratogenic in the golden Syrian hamster. The incidence of the observed effects increased with an increasing con tent of 2,3,7,8-tetrachIorodibenzo-p-dioxin. Dioxin impurity caused edema and hemorrhages in new born animals. Com m ittee for E nvironm ental I nform a tion. Critique of the report of the ad visory committee on 2,4,5-T. Environ ment 13(7): 24, 29 (1971). Criticism by the Committee for Environmental Information levied against the Advisory Commit tee's report on 2.4,5-T centered on the following areas: no effect levels: environmental accumulation, transmission and degradation; human hazards from 300 Environmental Health Perspectives 784632 GENP 011821 iioxins; 2,4,5-T teratogenicity; and the benefits vs. bit. Food Cosmet. Toxicol. 9(3): 395- j.-isks of 2,4,5-T usage. 404 (1971). Courtney, C. D., and Moore, J. A. Terat Treatment of rabbits and rats with 2,4,5-T con ology studies with 2,4,5-trichlorophenoxyacetic acid and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl. Pftarmacol. 20(3): 396-403 (1971). 2.4.5-T and 2,3,7,S-tetrachlorodibenzo-p-dioxin (TCDD) were teratogenic in three strains of mice; combinations of these compounds did not potentiate taining 0.5 ppm 2,3,7,8-tetrachlorodibenzo-p-dioxin did not cause any teratogenic or embryotoxic effects. Rats received 1, 3, 6, 12, or 24 mg 2,4,5-T/kg on days 6 to 15 of pregnancy. Rabbits received 0, 10, 20, or 40 mg 2,4,5-T/kg on days 6 to 18 of pregnancy. F irestone, D. Determination of polychlorodibenzo-p-dioxins (dioxins) in chloro- observed results. Both produced cleft palates and kidney malformations. Studies with rats gave nega tive results with 2,4,5-T; TCDD induced kidney anomalies. Crosby, D. G., Wong, A. S., Plimmer, J. R., and Woolson, E. A. Photodecomposition nf chlorinated dibenzo-/-dioxins. Science 173(3998): 748-749 (1971). phenols and lipids. Division Pesticide Chemistry, 162nd National Meeting of the American Chemical Society, Wash ington, D.C., Sept. 12-17, 1971; Ab stract No. 80. High resolution mass spectrometry was reported to otfer a sensitive method for detecting specific dioxins after adequate sample cleanup, e.g., chroma 2.1.5-T and 2,8,7,8-tetrachlorodibenzo-p-dioxin tography and gas chromatography. homologs photodecompose rapidly in an alcohol solu tion exposed to artificial light and natural sunlight. The decomposition rate was correlated with the de gree of chlorination. Negligible photodecomposition F irestone, D. F lick, D. F., Ress, J., and Higginbotham. G. R. Distribution of chick edema factors in chick tissues. J. was observed in aqueous suspensions ^nd on wet or dry soil. Assoc. Offic. Anal. Ckem. 54(6): 12931298 (1971). Davring. L.. and Sunner, M. Cytogenetic Chickens fed toxic animal fat containing 9.9 ppm effects of 2,4,5-triehlorophenoxyacetic chlorinated dihenzo-p-dioxins excreted over 9 0 the acid on oogenesis and early embrvogene- ingested hexa-, hepta-, and octachlorodioxins. De sis in Drosophila melanogaster. Heredi- creased chlorination resulted in greater body reten tas 68(1) : 115-122 (1971). tion. Residues were found in many tissues but was most prominent in the liver. 2.4.5- T with dioxin contaminants less than 0.1 ppm was evaluated in a wild-type Drosophila population. Hansen. W. H.. Quaife. M. L.. Habermann. Canton-5 109. Adult flies were exposed to 250 ppm R. T.. and F itzhugh. O. G. Chronic 2.1.5-T in their food within or later than 24 hr of toxicity of 2,4-dichIorophenoxyacetic eclosion. Results indicated that this 2,4.5-T formula tion effected early oogenesis and caused chromosome disturbances which could result in sterility. Elvidge. D. A. The gas-chromatographic de termination of 2,3,7,8-tetrachloro-p-di- acid in rats and dogs. Toxicol. Appl. Pharmacol. 20: 122-129 (1971). 2.4-D preparations, having undetectable levels of 2.7-dichlorodibenzo-p-dioxin or 2,3,7,8-tetrachlorodi benzo-p-dioxin, were tested for chronic toxicity in oxin in 2,4,5-triehlorophenoxyacetic acid (2.4,5-T), 2,4,5-T ethyl-hexyl ester, for mulations of 2.4,5-T esters and 2,4,5-tri- rats and dogs. No significant effect on growth, sur vival, organ weight, or hematologic values were noted. chlorophenol. Analyst 96(1147): 721- Hays. H., and Risebrough, R. W. The early 727 (1971). warning of the terms. Natural History A gas chromatographic method is described for the detection of 2,3,7,8-tetrachlorodibenzo-p-dioxin impurities in herbicides. Dioxin recoveries ranged from 89 to 98'> ; limit of detection was 0.05 ppm. 80(9): 39-46 (Nov. 1971). Possible health implications to humans and wild life from environmental pollution with chlorinated dibenzofurans are reviewed. E merson, J. L., T hompson, D. J., Strebing, R. .T.t Gerbig, C. G., and Robinson. V. G. Teratogenic studies of 2,4,5-trichlorophenoxyacetic acid in the rat and rab- Helling. C. S. Pesticide mobility in soils. II. Applications of soil thin-layer chro matography. Soil Sci. Soc. Amer. Proc. 35(5) : 737-743 (1971). oeptem ber 1973 301 784633 . Relative mobility by thin-layer chromatography and diffusion in moist and air-dry soils were used to test pesticide movements. The two chlorinated dibenzo-p-dioxins were immobile. Holden. C. Critics weigh EPA herbicide re port, find it wanting. Science 173(3994) : 312 (1971). The EPA's Advisory Committee on 2,4.5-T recom mended that the ban on 2,4,5-T be lifted with cer tain qualifications, i.e., no more than 0.1 ppm 2,4,5-T in drinking water and formulations containing no more than 0.1 ppm dioxin. The Committee for En vironmental Information voiced five major objec tions. Isenpee. A. R., and J ones, G. E. Absorption and translocation of root and foliage applied 2,4-dichlorophenol, 2,7-dichlorpdibenzo-p-dioxin, and 2.3,7,8-tetrachlorodibenzo-p-dioxin. J . Agr. Food Ckem. 19(6): 1210-1214 (1971). Measured uptake of "C-labeled 2,7-dichloro- and 2,3,7,8-tetrachlorodibenzo-p-dioxin from nutrient so lution, soil, and foliage by oats and soybeans indi cated that accumulation of TCDD in plants via the soil is unlikely. J ohnson, J. E. The public health implica tions of widespread use of the phenoxy herbicides and picloram. Bioscience 21 (17) : 899-905 (Sept. 1, 1971). A review presenting the myriad historical aspects of the dioxin-2.4.5-T( and other herbicides) picture. The potential adverse health effects from phenoxy herbicides were documented. One conclusion reported was that impurities--particularly the chlorodibenzop-dioxins--can be an important factor, but these can be controlled by proper manufacturing techniques. J ohnson. J. E. Safety in the development of herbicides. Proc, Ann. Calif. Weed Conf. 23: 43-67 (1971). Experiments using 2,4,5-T alone and in combina tion with dioxin allowed the conclusion that com mercially produced 2,4,5-T containing less than 1 ppm dioxin does not present a hazard to health. J ohnson, J. E. Safety in the development of herbicides. Down to Earth 27(1): 1-7 (1971). After the discovery of highly toxic dioxin resi dues in the herbicides 2,4,5-T, precautionary control measures were instituted to insure greater quality control in pesticide production. Pregnant rats were treated orally with 2,4,5-T at doses of up to 100 mgr/kg/ day. No symptoms of teratology were pro duced in the offspring. Dioxin was toxic to embryos at 0 125 ug/kg/day, whereas 0.03 .ug.'kg/day was be low the "no-effect" level. Some aspects of production and costs are discussed. Kearney, P. C., and Woolson, E. A. Per sistence and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in soils. Division Pesticide Chemistry, 161st Na tional Meeting of the American Chemical Society, Los Angeles, March 28-April 2, 1971; Abstract No. 21. TCDD applied 1, 10, and 100 ppm to two soils held at 30*C remained relatively unchanged (80-85^ of original samples were recovered) after 20, 40, 80, and 160 days. Incubation of 10. 100, and 1000 ppm, 2,4-dichlorophenol, and 2,4.5-trichlorophenol in soils for 70 days produced no detectable di- or trichlorodioxins. Kearney, P. C., Isensee, A.. Helling, C. S. Woolson. E. A., and P limmer, J. R. Environmental significance of chlorodioxins. Division Pesticide Chemistry, 162nd National Meeting of the American Chemical Society, Washington, D.C., Sept. 12-17, 1971 ; Abstract No. 90. Studies on TCDD under controlled environmental conditions revealed that TCDD was persistent and immobile in soil, not readily absorbed by plants, sub ject to photodecomposition, and slowly degraded to polar metabolites. Other studies revealed that the environmental contamination is small and not detect able in biological samples. Kh era , K. S., and R uddick, J. A. Perinatal effects of dibenzodioxins in Wistar rats. Division Pesticide Chemistry, 162nd Na tional Meeting of the American Chemi cal Society, Washington, D.C. Sept. 1217, 1971; A bstract No. 87. Oral administration of 16 to 0.125 yg/kg 2,3,7,8tetrachlorodibenzo-p-dioxin to pregnant rats on days 6-15 of gestation caused a reduction in viable litter size and a high incidence of resorptions. No effect was observed at the 0.125 level. Histological exami nation of fetuses revealed brain and intestinal hemorrhages and edema of subcutaneous tissue. Khera. K. S., H uston. B. L., and McKinney. W. P. Pre- and postnatal studies on 2, 4.5-T, 2,4-D, and derivatives in W istar rats. Toxicol. Appl. Pharmacol. 19:369370 (1971). .Results from teratogenic studies with Tats treated with 2,4.5-T, 2,4-D, and several derivatives wert 302 Environmental Health Perspectives 784634 E N P 011823 either negative o r inconclusive. The 2,3,7,S-tetrachlorodibenzbo-p-dioxin'levels in these preparations were not known in all cases, but most were suspected to be less than 0.5 ppm. Langer, h . G., Brady, T. P., Dalton, L. A. Shannon, T. w ., and Briggs, P. R. Thermal chemistry of chlorinated phe nols. Division Pesticide Chemistry, 162nd National Meeting of the American Chemical Society, Washington, D.C., Sept. 12-17, 1971; Abstract No. 83. At the decomposition temperatures of chlorinated phenols and derivatives, the following conditions are ideal for dioxin formation: (a) high concentration of the phenol, (b) high degree of chlorination of the phenol, and (c) presence of base, Loefroth, G. Environmental poisons, man and the physician. Laekartidningen 68 (18): 2113-2133 (1971) (Swed.). Possible adverse health effects from environmental poisons demands caution. Tetrachlorobenzo-p-dioxin was mentioned for its adverse effects of fetal develop ment. > Mil n e s , M. H. Formation of 2,3,7,8-tetrachlorodibenzodioxin by thermal decom position of sodjum 2,4,5-trichlorophennate. Mature (London) 232(5310) : 395-396 (1971). Hydrolysis of 1.2,4,5-tetrachlorobenzene in methyl alcohol with sodium hydroxide at high pressures or in ethylene glycol at L atm formed sodium monochlorophenate. This allowed exothermic decomposi tion of 2.4,5-trichIorophenate sodium to 2,3,7,8-tetrachlorodibenzodioxin. Muelder, W. W., and Shadoff, L. The prep aration of uniformly labeled l4C-2,7-dichlorodibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Division Pesti cide Chemistry, 162nd National Meeting of the American Chemical Society, Washington, D.C., Sept. 12-17, 1971; Abstract No. 78. 2,7-Dichlorodizenzo-p-dioxin was prepared from isotopic potassium 2.4-dichlorophenate uniformly labeled with "C. Chlorination of the dioxin in a chloroform solution containing FeCL and I: gave a mixture of tri-, tetra-, and pentachlorosubstitution products. Purification hy digestion in boiling chloro form, fractional sublimation, and reerystall ization from anisole yielded a product containing 92ri2,3.7,8-tetrachIorodibenzo-p-dioxin. Piper, W. N., and Rose. J. Q. The excretion and tissue distribution of 2,3,7,8-tetra- chlorodibenzo-p-dioxin in the rat. Divi sion Pesticide Chemistry, 162nd National Meeting of the Amercian Chemical So ciety, Washington, D.C., Sept. 12-17, 1971; Abstract No.-88. TCDD was administered to male rats in a single oral dose of 50 pg/kg. The primary excretion route was via the feces and the highest tissue concentra tion was* in the liver. Plimmer, J. R., Crosby, D. G., Wong. A. S,, and Klingebiel. U. I. Photochemistry of dibenzo-p-dioxins. Division Pesticide Chemistry, 162nd National Meeting of the American Chemical Society, Wash ington. D.C., Sept. 12-17, 1971; Ab stract No. 85. Ring closure and photochemical reaction of o-chlorophenols yield dioxin. Dibenzo-p-dioxins were decomposed in solution hy light or sunlight. Absorp tion of dioxin on soil reduced its rate of photode composition. PLIMMER, J. R,, and KLINGEBIEL, U. I. Ribo flavin photosensitized oxidation of 2,4dichlorophenol: assessment of possible chlorinated dioxin formation. Science 174(4007): 407-408 (1971). Chlorophenols in aqueous solution were exposed to the action of light in wavelengths greater than 280 nm. Products were characterized by gas chromatoeraphy and mass spectrometry. Possible chlorinated dihenzo-p-dioxins from ring closure of a 2-phenoxyphonol derivative were not detected in the photolysis products. Under environmental conditions, dioxins are unlikely products of lower chlorinated phenols or phenoxy-alkanoic acids. Pohland. A. E.( Yang, G. C., and Hansen. E. A. The preparation and characteriza tion of chlorinated dibenzo-p-dioxins. Di vision Pesticide Chemistry, 162nd Na tional Meeting of the American Chemi cal Society, Washington, D.C., Sept. 12-17, 1971: Abstract No. 77. Twelve chlorinated dibenzo-p-dioxin standard com pounds containing from one to eight chlorine atoms were prepared as standards to develop analytical techniques and to use in toxicity studies. The wave length associated with the observed phosphorescence and the triplet state lifetime were dependent upon the number of chlorine atoms and their positions on the dihenzo-p-dioxin nucleus. Poland. A. P., Smith. D., Metter, G.. and PossiCK. P. A health survey of workers September 1973 303 GENP 011824 784635 in a 2f4-D and 2,4,5-T plant. Arch; En viron. Health 22:316-327 (1971). A study and review of employee health in a plant producing 2,4,S-T and 2,4-D. Chloracne was found in 13/73 male workers and was believed caused by chlorinated dioxins. The plant has begun several programs to reduce dioxin levels in 2,4,5-trichlorophenol; the contaminant concentration has dropped from 10 to 25 ppm to 1 ppm. P orter, M. L., and Burke, J. A. Separation of three chlorodibenzo-p-dioxins from some polychlorinated biphenyls by chro matography on an aluminum oxide colum. Chlorodioxins (2,3-di-, 2,3,7-tri-, and 2,3,7,8-tetrachlorodibenzo-p-dioxin) were separated from poly chlorinated biphenyls by column chromatography o n ' aluminum oxide. Recoveries of both dioxins and the polychlorinated biphenyls were approximately 1009fc. Risebrough, R. W. Determination of poly chlorinated biphenyls in environmental samples. In International Symposium on Identification and Measurement of En vironmental Pollutants, Ottawa, Ontario, Canada, June 14-17, 1971, B. Westley, Ed., National Research Council, Canada, 1971, pp. 147-153. Even though it is not known whether dibenzofurans can be formed from environmental residues of polychloronated biphenyls (PCB), they should be considered as a part of the PCB problem. Dibenzofurans have been found in commercial PCB mixtures sampled from France, Germany, and the United States in concentrations ranging from 5 to 20 ppm. The European samples had higher levels than the U.S. sample. Roll, R. Teratogenic effect of 2,4,5-T [2,4, 5-trichlorophenoxyacetic acid] in mice. Food Cosmet. Toxicol. 9(5): 671-676 (1971) (Ger). 2,4,5-T having a 2,3,7,8-tetrachIorodibenzo-p-dioxin content of less than 0.1 ppm caused embryotoxic effects and a significant increase in the incidence of cleft palate in mice. Concentrations used in the study ranged from 35 to 130 mg/kg 2,4,5-T/dioxin which were given orally on days 6-15 of pregnancy. The teratogenic "no-effect" level of 2,4,5-T/dioxin was found to be 20 mg/kg. Rowe, V. K., Norris, J. M., Sparschu, G. L., Schw eiz, B. A., and Gehring, P. J. Toxicology of chlorinated dibenzo-p-dioxins. Division Pesticide Chemistry, 162nd National Meeting of the Ameri can Chemical Division, Washington, D.C., Sept. 12-17, 1971, Abstract No. 86. The toxicology of several chlorinated dibenzo-pdioxins were investigated in rats, guinea pigs, and rabbits. The compounds studied were 2,7,-dichlorodibenzo-p-dioxin (DCDPD); 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDPD); hexachlorodibenzo-pdioxin (HCDPD), and octachlorodibenzo-p-dioxin (OCDPD). TCDPD was found to be the most toxic of the compounds investigated. DCDPD and OCDPD were the least toxic. For instance, little if any fetotoxic effects were observed with these two com pounds; doses of 2 to 4 g/kg were not lethal to rats. On the other hand, TCDPD and QCDPD induced chloracne, fetotoxicity, and teratogenic effects. Schwetz, B. A., Sparschu, G. L., and Gehring, P. J. The effect of 2,4-dichlorophenoxyacetic acid (2,4-D) and esters of 2,4,D on rat embryonal, foetal and neo natal growth and development. Food Cosmet Toxicol. 9:801-817 (1971). The teratogenicity of 2,4-D was investigated in the rat. Formation of 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD) in the manufacture of 2,4-D is not theoretically possible. The 2.4-D sample used was analyzed for the TCDD contaminant with a method having a 0.2 ppm sensitivity and none was found. Doses of 12.5, 25, 50, 75, and 87.5 mg/kg 2,4-D were administered orally on days 6-15 of gestation. High dose levels caused signs of embryotoxicity and feto toxicity. Somers, E., and Smith, D. M. Source and occurrence of environmental contami nants. Food Cosmet Toxicol. 9(2): 185193 (1971). The sources, occurrences, and some of the toxi cological effects of polychlorinated biphenyls and chlorodioxins were discussed. Chlorophenols exposed to high temperatures may form chlorinated deriva tives of dibenzo-p-dioxin. Inadequate methodolgy prevents analysis for these contaminants in foods. Sparschu, G. L., Dunn, F. L., Lisowe, R. W., and Rowe. V. K. Study on the effects of high levels of 2,4,5-trichlorophenoxyace tic acid on foetal development in the rat. Food Cosmet Toxicol. 9(4): 527-530 (1971). Commercial-grade 2,4,5-T containing 0.5 ppm 2,3,7,8-tetrachlorodibenzo-p-dioxin did not induce a teratogenic response when administered orally tc rats on days 6 to 15 of gestation a t a dosage level of 50 m g/kg/ day. 304 Environmental Health Perspectives UtslNF 011825 784636 V ^ arschu, G. L.f'D unn, F. L., and Rowe, V. K. Study of the teratogenicity of 2.3.7.8- tetrachlorodibenzo-p-dioxin in the rat. Teratology 4: 247 (Abstract) (1971). 2,3,7,8-Tetrachloradibenzo-p-dioxin was adminis tered orally to pregnant rats during days 6 to 15 oi gestation in doses of 0, 0.03, 0.125, 0.5, 2.0, and 8.0 g/kg/day [sic]. No effect was observed on the fetus or mother a t the 0.03 dose level. However, at the 0.125 level and above, fetal mortality, early and late resorptions, and fetal intestinal hemorrhage occur red. The effects became more pronounced with the increase in dose. Maternal toxicity was noted be ginning at the 0.5 level which also increased with the dose. Sparscru, G. L., Dunn, F. L., and Rowe, V. K. Study .of the teratogenicity of 2.3.7.8- tetrachlorodibenzo-p-dioxin in the rat. Food Cosmet. Toxicol. 9(3): 405412 (1971). Investigations of the teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats revealed that this compound is capable of inducing a high degree of adverse effects in the fetus and mother. Doses of Q-, 0.03, 0.125, 0.5, 2.0, and 8.0 jig kg/day were given 'ally to pregnant rats during days 6 to 15 of gestajn. Effects such as fetal mortality, intestinal hemorrhage, and early and late resorptions began appearing at the 0.125 level and increased with dose. It is suggested that the teratogenic effects of 2.4,5-T observed in other studies may have been due to this contaminant (30 ppm). STEHL. R. H., Papenfuss, R. R., Bredweg, R. A., and Roberts, R. W. The stability of pentachlorophenol and chlorinated dioxins to sunlight, heat, and combus tion. Division Pesticide Chemistry, 162nd National Meeting of the Ameri can Chemical Society, Washington, D.C., Sept. 12-17, 1971; Abstract No. 92. Degradation of pentachlorophenol in the environ ment could theoretically lead to the formation of chlorinated dibenzo-p-dioxins. Several simulated en vironmental conditions were checked for possible dioxin formation. Pentachlorophenolf-treated wood and paper indicated no increase in dioxins when burned: likewise photolysis of sodium pentachloro phenol yielded only a trace of the compound. 2.7-Diand 2,3,7,8-tetrachlorodibenzo-p-dioxin were rapidly decomposed under artificial sunlight, whereas octachlorodibenzo-p-dioxin was not. .Stehl, R. H., Wilke, E., P apenfuss, R. R., and Matalon, R. Determination of non- phenolic impurities in chlorinated phe nols and related compounds. Division Pesticide Chemistry, 162nd National Meeting of the American Chemical Divi sion, Washington, D.C., Sept. 12-17, Abstract No. 81. Analytical techniques for the detection of neutral impurities (chlorinated dibenzo-p-dioxins) in chlori nated phenols, were ion-exchange chromatography for removal of the matrix, liquid-liquid chromato graphy for separation of the various neutral com ponents, followed by gas chromatographic examina tion of the separated components using electron capture detection. Storherr, R. W., Watts, R. R., Gardner. A. M., and Osgood, T. Steam distillation technique for the analysis of 2,3,7,8tetrachlorodibenzo-p-clioxin in technical 2,4,5-T. J. Assoc. Offic. AjioZ. Chem. 54(1): 218-219 (1971). 2,3,7,8-Tetrachlorodibenzo-p-dioxin was separated from 2,4,5-T by steam distillation and analyzed by mierocoulomecric gas-liquid chromatography. Thompson. D. J., E merson, J. L., and SparSCHU. G. L. Study of the effects of 2,4,5tvichlorophenoxyacetic acid (2,4,5-T) on rat and rabbit fetal development. Tera tology 4:243 (Abstract) (1971). 2.4,5-T was investigated for its effects on fetal development in the rat and rabbit. Doses ranging from 1 to 50 m g'kg were administered during days ti to 15 of gestation. No clinical or gross pathology was observed. Detailed examination of the visceral and skeleton revealed no teratogenic effects. How ever, rats receiving 100 mg,'kg during day 6 to 10 of gestation did produce maternal toxicity and death, early fetal resorptions, fetal toxicity, but no mal formations. Currier, W. F., Graham, C., Gratkowski, H., and Norris, L. A., U-S. Forestry Service. Report on background informa tion for the phenoxy herbicides 2,4-D-- . 2,4,5-T--2,4,5-TF. U.S. Forestry Service Report, pp. 1-165 (1971). A through review is given on the phenoxy herbi cides, 2,4-D (2,4-dichlorophenoxyacetic acid), 2.4,5-T (2,4,5-tfichlorophenoxyacetic acid), and 2,4,5-TP [2(2.4.5-trichlorophenoxy)propionic acid], and their potential contaminant, e.g., dioxin. Vinopal. J. H., Yamamoto, I., and Casida, J. E. Preparation of tritium-labeled 2.3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and structure-activity investigations of September 1973 305 784637 GENP 011826 TCDD and other related dibenzo-pr dioxins. Division Pesticide Chemistry, 162nd National Meeting of the American Chemical Society, Washington, D.C., Sept. 12-17, 1971, Abstract No. 79. Preparation of tritium-labeled 2,3,7,8-tetrachIorodibenzo-p-dioxin was investigated. One approach used was the tritiation of 2,4-dichlorophenol or 2,4,5trichlorophenol, lithiummetalation and subsequent aqueous hydrolysis of a chlorinated dibenzo-p-dioxin, and the catalytic reduction of a chlorinated dibenzop-dioxin. Westing, A. H. Herbicides as agents of chemical w arfare: Their impact in re lation to the Geneva Protocol of 1925. Environ. Affairs 1(3): 578-586 (1971). The use of herbicides as chemical warfare agents is reviewed with respect to the Geneva Protocol of 1925. The total impact of these agents on the en vironment and man is not usually known or con sidered before widespread use. For example, 47 mil lion lb of 2,4,5-T was used in Indochina over an 8yr period before, it was discovered that about 1000 lb dioxin was also being applied as an unsuspected contaminant. Williams, D. T., and Blanchfield, B. J. Thin layer chromatographic separation of two chlorodibenzo-p-dioxins from some polychlorinated biphenyls and organochlorine pesticides. J. Assoc. Offic. Anal. Chem. 54(6): 1429-1431 (1971). Thin-layer chromatography was used for the iso lation of 2,3,7,8-tetrachlorodibenzo-p-dioxin and octachlorodibenzo-p-dioxin from polychlorinated biphe nyls and organochlorine pesticides. After isolation, these dioxins were analyzed by the electron capturegas chromatography. Wilson, J. G., Boutwell, R. K., Davis, D. E., Dost, F. N., Hayes, W. J., Kalter, H., Loomis, T. A., Schulert, A., Sterl ing, T. D., and Bowen, D. L. Report of the Advisory Committee on 2,4,5-T to the Administrator of the Environmental Protection Agency. (Submitted May 7, 1971). The 2,4,5-T Scientific Advisory Committee's report to the U.S. Environmental protection Agency was stimulated after experimental results and environ mental indicators pointed to a potential health threat to humans. This nine-member committee was selected from a list of names supplied by the National Aca demy of Science. The Committee recommended by an 8 to 1 vote to restore registration of 2,4,5-T with certain conditions; those applicable to TCDD were: (1) a limit of 0.5 ppm TCDD on existing 2,4,5-T in ventories and 0.1 ppm on future 2,4,5-T production, and (2) conduct specific research on TCDD for po tential soil accumulation and food drain magnifica tion. Williams, C. S. The status of 2,4-D, 2,4,5-T, Silvex and MCPA herbicides. Down to Earth 26(4) : 12-15 (1971). 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is the causative agent in experiments reporting terotogenic effects from 2,4,5-T. TCDD was found in 2,4,5-T formulations but not in 2,4-D samples. The pre cursor for 2,4,5-T is 2,4,5-trichlorophenoI which is prepared from 1,2,4,5-tetrachlorobenzene; the high temperature and alkaline conditions necessary to produce this compound also favor dioxin formation. Since 2,4-D is formed by direct chlorination of 2,4dichlorophenol and not by alkaline hydrolysis of 1,2,4-trichlorohenzene, no dioxin impurities are theo retically possible. Woolson, E. A., Reichel, W. l ,, and Young, A. L. Dioxin residues in lakeland sand and eagle samples. Division Pesticide Chemistry, 162nd National Meeting of the American Chemical Society, Wash ington, D.C., Sept. 12-17, 1971. Abstract No. 91. Between 1962 and 1969 a lakeland region in Flori da was treated with 912 lb 2,4,5-T per acre. To assess the ecological importance of chlorinated dioxins in the environment, samples of soil to a depth of 3 ft, and tissues from eagles were analyzed for dioxins by electron capture-gas chromatography. No dioxins were detected at a minimum detection limit of 50 ppb for the eagle tissue samples and 5 ppb for the soil samples. Small residues of about 20 ppb 2,4,5-T were found in soil samples. Yang, G. C., and' Pohland, A. E. Cation radicals from chlorinated dibenzo-p-dioxins. Division Pesticide Chemistry, 162nd National Meeting of the American Chemical Society, Washington, D.C., Sept. 12-17, 1971! Abstract No. 84. Characteristic blue coloration and electron spin resonance (ESR) spectra of cation radicals were ob tained following dissolution of chlorinated dibenzop-dioxins in trifiuoromethanesulfonic acid. The monoand dichlorodibenzo-p-dioxins showed well resolved ESR by hyperfine lines; the tetrachloro analogs ex hibited single broad ESR lines. 306 Environmental H ealth Perspectives 784638 E N P 0U827 970 ABELSON, P. H. Pollution by organic chemi 2.4.5- T controversy which began after preliminary research reports indicated 2,4,5-T was teratogenic. These government studies were instituted in an at cals. Science 170(3957): 496 (30 Oct. tempt to determine whether 2,4,5-T or its dioxin 1970). contaminant was the causative agent of these effects. When manufacture of 2,4,5-T is controlled care Anonymous. The 2,4,5-T identification par fully, dioxin contamination is less than 1 ppm. Dioxin was identified in 1962 as the culprit of damage and death in 1957 of uncounted numbers of chicks. Dioxin produces neurological disturbances and is terato genic. The acute LD*> for male guinea pigs is 0.000001 g/kg. ANONYMOUS. HEW, XJSDA hold firm ; 2,4,5-T ruling postponed. Chem. Eng. News 48(7) : 11-12 (1970). Because early studies indicated that most of the adverse toxicological effects of 2,4,5-T were due to high dioxin contamination (27 ppm) of test samples, the USD A did not feel obligated to cancel registra tion of currently manufactured 2,4,5-T containing about 1 ppm dioxin. Preliminary tests have shown that 2.4.5-T containing 1 ppm dioxin is not terato genic. However, problems may still exist, since ex cessive heat exposure of the tri- or pentachloro- ade. Food Cosmet. Toxicol. 8: 596-597 (1970). Inconsistent test data on the teratogenicity of 2.4.5- T were thought resultant of the level of its chief contaminant, 2,3,7,3-tetrachlorodibenzo-p-dioxin. Levels of 27 ppm have been found in some 2,4,5-T samples. Anonymous. Defoliants, deformities: What risk? Med. World News 11(9): 15-17 (1970). Fertilized chicken eggs were injected with 2.5 pg (2.5xlO `-g) (50 ppt) of 2,3,6,7-tetrachlorodib3nzop-dioxin. The dioxin caused leg deformities, cleft palates, and beak defects in 11 of 15 survivors from 25 eggs. Unhatched birds showed tissue edema, necrotic livers, and deformities similar to hatched birds. pnenols, used as intermediates in 2,4,o-T production, liberates dioxins. Anonymous. Another herbicide on the black list. Nature 226(5243): 309-311 (1970). Anonymous. Government steps up pressure A commentary was presented to clarify the con on pesticides; recent joint action restrict ing use of 2,4,5-T herbicide portends changes in pesticide regulations and testing. Chem. Eng. News 48(18): BO61 (1970). troversy of whether 2.4,5-T or 2,3,7,8-tetrachlorodibenzo-p-dioxin was the causative teratogenic and mutagenic agent. Healings with government and industrial representatives were held before the Sub committee on Energy, Natural Resources and the Environment of the Senate Committee on Com Evidence from Congressional hearings catalyzed merce. the government's eventual suspension and cancella tion of consumer and food crop usage of 2.4,5-T and curtailment of its use in Vietnam. HEW and the USDA issued a joint statement declaring that both 2.4.5- T and its dioxin contaminant may cause birth defects. The pesticide industry reacted adversely to these actions and statements. Anonymous. Pesticide policies scored at Ag meeting. Chem. Eng. News 48(42): 7 (1971). Anonymous. The tangled tale of 2,4,5-T. PANS 16(3): 421-422 (1970). Use of 2,4,5-T has been restricted where food stuffs for human consumption might be contami nated. The possibility exists that dioxin (2,3,7,8tetrachlorodibenzo-p-dioxin) may he the contaminant in 2,4,5-T formulations or that there may be a syner gistic effect between 2,4,5-T and dioxin. Hypothetically, if dioxin contaminated the 2,4,5-T reported in food in the U.S. to the extent of 1 ppm, it would The politics and pressures brought about by the 2.4.5- T/dioxin controversy between government and industry were reviewed. The Chairman of the Sec take 1(10 years for a normal person to accumulate a dose 600 times less than that which produced toxic effects during the pregnancy of a rat. retary's Pesticide Advisory Committee (SPAC) casti gated both the federal government and the pesticides industry for failure to avoid the panicky crises which brought about an "era of chemical McCarthyism." Anonymous. The strange case of the gov ernment vs. 2,4,5-T. II. Farm Chem. 133(3) : 26 (1970). A review of the government involvement in the Courtney. K. D., Gaylor, D. W., Hogan, M. D., F alk. H. L., Bates. R. R., Mit chell. I. Teratogenic evaluation of 2,4,5-T. Science 168(3933): 864-866 (May 15, 1970). 2,4,5-T samples were evaluated in rats and mice for teratogenicity. Results indicated that 2.4.5-T s.c. or orally administered-was teratogenic and fetotoxic September 1973 307 GENP 011828 784639 to the mouse and orally to the rat. Cleft palate and cystic kidneys were among the effects observed^.A note added in proof indicated that the 2,4,5-T sam ples contained 30 ppm 2,3,7,8-tetrachlorodibenzo-pdioxin. Darsow, G., and Schnell, H. ChlorodibenzoI p-dioxins. Ger. Patent 1,930,259, Appl. 13 June 1969, Granted 17 Dec. 1970, 10 PP (1970). Chlorodibenzo-p-dioxins were prepared from tetraor hexachlorobenzene. The compounds were useful for flameproofing polymers, e.g. polyesters. Day, B. E. 2,4,5-T and government decisions. Proc. West. Soc. Weed Sci. 23: 4-6 (1970). Unpublished material which influenced the 2,4,5-T ban was held as questionable and suspect since pre parations tested contained 27 ppm dioxin. It was' stated that 2,4,5-T containing 1 ppm dioxin is not teratogenic; therefore, the alarm about the adverse etfects of 2,4,5-T was unjustified. E merson, J. L., T hompson, D. J., Gerbig, C. G,, Robinson, V. B. Teratogenic study of 2,4,5-trichlorophenoxyacetic acid in the rat. Toxicol. Appl. Pharmacol. 17: 317 (1970). The teratogenicity of 2,4,5-T containing 1 ppm of 2,3,7,8-tetrachlorodibenzo-p-dicxin was evaluated in rats. No teratogenic effects from 1,3,6,12, or 24 mg/ kg/day 2,4.5-T or other pathological signs were observed. These results do not substantiate the ad verse effects observed with 2,4,5-T containing 27 ppm dioxin. Epstein, S. S.- A family likeness. Environ ment 12(6): 16-25 (July/August, 1970). The human and environmental implications of the widespread chemical exposure problem have been accentuated by the increased use of herbicides dur ing the past ten years. For example, although re cently dioxin impurities in 2,4,5-T have been re duced, dioxin is still present; even minute concen trations are potential hazards. The analysis, forma tion. and teratogenic and other toxicological consequencies of dioxins were reviewed. E pstein, S. S, Testimony on teratogenic effects of 2,4,5-T formulations. U.S. Senate Hearing before the Subcommit tee of Energy, Natural Resources and the Environment of the Committee on Commerce (April 15, 1970). Polychlorophenol contaminants and their dioxin pyrolytic products in phenoxy herbicides are of toxi- cologic and environmental consequence. Some dioxins are highly toxic and teratogenic at the microgram per kilogram level; most dioxins are, however, toxicologically uncharacterized. Galston, A. W. Herbicide usage. Science 168 (3939) : 1607 (1970). The phenoxyacetic acid herbicides should not be used until all questions regarding the teratogenicity of 2,4,5-T have been resolved. A tetrachlorodibenzodioxin impurity found in some commercial prepara tions of 2,4,5-T was the teratogenic agent in 2,4,5-T samples tested for teratogenicity by the Bionetics Research Laboratories. The possibility that phenoxy acetic acids may be degraded in plants and soil or by fire or bright sunlight into dioxin-like teratogens should be studied. Extensive research should be ac complished on all aspects of the biomedical and en vironmental consequences of dioxins. Hearings Before the Subcommittee on Energy, Natural Resources and the En vironment of the Committee on Com merce, U. S. Senate, Serial 91-60, Gov- . ernmental Printing Office, Washington, D.C. (1970). An impurity in 2,4,5-trichlorophenol was found to be 2,3,7,8-tetrachlorodibenzo-p-dioxin, a compound with a high mammalian toxicity and teratogenic effect. Horwitz. W. (Ed.). Official Methods of Analysis, 11th Ed. Association of Official Analytical Chemists, Washington, D.C., 1970, p. 468 (28.109). The official first action gas chromatographic meth od for chick edema factor is presented in Section 28 on oils and fats. Fat, oil, fatty acid, or lipid is treat ed with HiSO., and extracted with petroleum ether. The extract is purified on an AlsOi column, further treated with H:SO,, and examined by electron capture-GLC. Peaks with retention times relative to aldrin (A.) between 8 and 45 indicate presence of chick edema factors (hexa-, hepta-, and octachlorodibenzo-p-dioxins). Kearney, P. C. Paper presented before a joint (United Kingdom, Canada, and United States) meeting on Pesticides, Washington, D.C., (November 5, 1970). The Agricultural Research Service, USDA, began a program to assess the significance of chlorinated dioxin impurities in currently registered pesticides. Theoretically, any pesticide with a chlorinated phenoxy nucleus or which is derived from a chlo rinated phenol precursor could contain chlorinated dioxin contaminants. 308 Environmental Health Perspectives 784640 Ress, J., Higginbotham, G. R.., and Fire stone, D. Methodology for chlorinated aromatics in fats, oils, and fatty acids. J. Assoc. Ofiic. Anal. Chem. 53(3): 628634 (,1970). Following review of the current status of chemi cal and biological methods for chlorophenols and chlorinated dibenzo-p-dioxins in fats, oils, and fatty acids, the modified electron capture-gas liquid chro matography method for the detection of chick edema factors was recommended for adoption as official first action to replace all existing gas-liquid chroma tography methods. Robson, J. M. Testing drugs for teratogeni city and their effects on fertility: The present position. Brit. Med Bull. 26(3): 212-216 (1970). The problems of interpreting test results involving teratogenic compounds were outlined using data gathered on the teratogenicity of 2,4,5-T and its dioxin contaminant, Sparschu, G. L., Dunn. F. Lv and Rowe, V. K. Teratogenic study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Toxi col. Appl. Pharmacol. 17(1): 317-318 (1970). 2,3,?,S-Tetrachlorodibenzo-p-dioxin was studied to determine whether this 2,4,5-T impurity could ac count for fetal abnormalities in test animals. Rats received 0. 0.03, 0.125, 0.5, 2.0, and 8.0 ^g/kg body weight /day on days 8 to 15 of gestation. At the 0.5 level, the number of fetuses was reduced, and the number of resorptions and fetal deaths was increased. Vos, J. G., and Koeman, J. H. Comparative toxicologic study with polychlorinated biphenyls in chickens with special ref erence to porphyria, edema formation, liver necrosis, and tissue residues. Toxi col. Appl. Pharmacol. 17: 656-668 (1970). Briefly stated were the concerns and evidence pertaining to dioxins, particularly to the chick edema factor and whether PCBs and/or the dioxins were the culprit(s). Vos, J. G., Koeman, J. H., Van der Maas, H. L., Ten Noever de Brauw, M. C,, and De Vos, R. H. Identification and toxicological evaluation of chlorinated dibenzofuran and chlorinated naphtha lene in two commercial polychlorinated !. .September 1973 biphenyls. Food Cosmet. Toxicol. 8(6): 625^633 (1970). Three commercial polychlorinated biphenyl com pounds (Phenoclor DP6, Clophen, and Aroclor 1260) were analyzed for impurities. In two of these (Phe noclor DPS and Clophen A60) pentachlorodibenzofuran and tetrachloradibenzofuran were among the polar compounds found. 1969 Cantrell, J. S., Webb, N. C., and Mabis, A. J. Identification and crystal structure of a hydropericardium-producing factor: 1 ,2 ,3 ,7 ,8 ,9-hexachloro-dibenzo-p-dioxin Chem. Abstr. Acta. Crystrattogr. B 2 5 (l): 150-156 (1969); 70: 51805u (0000).. Chemical identity of hexachlorodibenzo-p-dioxin, one of the toxic substances capable of producing the chick edema disease, was established by x-ray crys tallography of a sample of edema-inducing lipid material. Ried. W., and E ng, J. T. S. Reactions with cliazocarbonyl compounds. Ann. Chem. 727: 219-221 (1969) (G e r.); Chem. Abstr. 72: 12458e (0000). o-Quinone diazides added benzene with nitrogen elimination to yield chloro substituted dibenzofurans: 1,2,3,4-tetrachloro-, 1,2,3,-trichloro-, and 1,3-dichloro-. 1968 Higginbotham, G. R., Huang, A., Fire stone, D., Verrett, J., Ress, J., and Campbell, A. D. Chemical and toxico logical evaluations of isolated and syn thetic chloro derivatives of dibenzo-pdioxin. Nature 220(5168): 702-703 (1968); Chem. Abstr. 70:18509c (0000). Commercial chlorophenols Including 2,4-dichlorophenol, 2,4,5,-trichlorophenol, and 2,3,4,6-tetrachlorophenol, were pyrolyzed to chlorinated dibenzo-pdioxins. These were examined hy electron capturegas liquid chromatography and tested bibliogically by the chick embryo assay, 2,3,7,8-Tetrachlorodibenzop-dioxin was prepared by direct chlorination of dibenzo-p-dioxin. These dioxins proved to be chick edema factors (hydropericardium factors) and toxic. 1967 Anonymous. Search for chick edema factor. Chem. Eng. News 45(5): 10 (Jan. 30, 1967). 309 784641 I. - /. JS Chicks afflicted with edema, or hydropericardium, suffer an accumulation of fluid in the heart sac and gross kidney and liver damage; as little as 5 Mg can kill a chick. X-Ray crystallographers have deter* mined the molecular structure of one of the toxic compounds known as the chick edema factor. Single crystal structural analysis showed that the structure is 1,2,3,7,8,9-hexachlorodibenzo-p-dioxin. Two crys tals, each about 0.2 x 0.1x 0.1 mm and weighing about 3Mg were recrystalized from a benzene - hexane solution. The Anal data showed that the 1,2,3,'7,8,9hexachlorodibenzo-p-dioxin molecules are nearly planar, and are packed in the 044 crystallographic planes, with an interplanar separation of about 3.3 A. No unusual bond lengths or angles were apparent. 1965 Cox, J. M,, Wright, B. A., and Wright, W. W. Thermal degradation of polypheny lene oxides). J. Appl. Polymer Sci. 9(2): 513-522 (1965). The thermal degradation in vacuum of various phenylene oxide polymers and copolymers was studied by a weight-loss method. Thermal stability decreased with increasing substitution in the aro matic nuclei. This may he due to the presence of 2.3.7.8- tetrachlorodibenzo-p-dioxin, formed as an al ternate reaction product. Chloro derivatives were more stable than the corresponding bromo deriva tives, except for the tetra-substituted phenylene oxides. Octahalodibenzo-p-dioxins were prepared by heat ing pentahalophenols in the presence of catalytic amounts of halogen or halocyclohexadiene. For ex ample, 40 g pentachlorophenol and 2 g 2,3,4,4,5,6hexachloro-2,5-cyclohexadienone in 120 ml 1,2,4trichlorobenzene was refluxed at 213 C to give 30 g of octachlorodibenzo-p-dioxin. 1961 Bauer, H., Schulz, K. H., and Sfeigelberg, U. Occupational intoxications in manu facturing chlorophenol compounds. Arch. Gewerbepath. Gewerbekyg. 18: 538-555 (1961). Thirty-one workers in an industry manufacturing 2,4,5-trichlorophenol by sodium hydroxide saponifica tion of 1,2,4,5-tetrachlorobenzene exhibited acne and symptoms of psychopathologic disturbances. Solu tions of 0.01 to 0.05% of 2,3,6,7-tetrachlorodibenzo-pdioxin painted on rabbit ears produced acne; oral administration of 0.05-0.1 mg/kg caused serious liver damage. Kulka, M. Octahalodibenzo-p-dioxins. Can. J. Chem. 39: 1973-1976 (1961). Nearly 100% conversion from pentahalophenols to 1.2,3.4,6,7,3,9-octahalo-dibenzo-p-dioxins was accom plished hy using reaction initiators such as halogens or halogen-generating compounds. Kulka, M., Perhalodibenzo-p-dioxins. Can. Patent 702,144, Appl. May 3, 1961; Granted Jan. 19, 1965, 13 pp. (1965) Octahaiodibenzo-p-dioxins, useful as chemical in termediates, biocides, and flame retardants, were prepared by using a pentahalophenoi in the presence of a catalytic amount of halogen or a halogenated cyclohexadienone at 200-400*C. An 83% yield was attained for octachlorodibenzo-p-dioxin. 1962 J ones, E. L., and Krizek, H. A technique for testing acnegenic potency in rabbits, ap plied to the potent acnegen 2,3,7,8-tetrachlorodibenzo-p-dioxin. J. Invest. Derma tol. 39: 511-518 (1962). Hyperkeratinization of the inner surface of the rabbit ear was induced by topical application of 2.3.7.8- tetrachlorodibenzo-p-dioxin in acetone. The weight of keratin served as a criterion of acnegenic activity. Kulka, M., Octahalodibenzo-p-dioxins. Belg. Patent 616,197, Appl. May 2, 1961; Granted July 31,1962, 9 pp. (1962) I960 Denivelle, L.t Fort, R., and Van Hai; P. Octachloro- and octabromodibenzo-pdioxins and decachloro- and decabromodiphenyl ether. Bull. Soc. Chim. France 1960. 1538-1543 (1960). Pentachlorophenol was pyrolyzed at 300C ex tracted with carbon tetrachloride and aqueous al kali, recrystalized from phenyl nitrite, and yielded "perchlorophenylene oxide" (melting point of 326C). Steric considerations dictated ring fusion in the 1,2position leaving octachlorodibenzo-p-dioxin the only acceptable structure for perchlorophenylene oxide. Sodium 2,4-dichlorophenate gave dichlorodibenzo-pdioxin. Kaupp, J., and Klug, H. Perchlorodiphenylene dioxide. Ger. Patent 1,092,480, Appl. June 27, 1956, Granted Nov. 10, i960 (1960). Octachlorodibenzo-p-dioxin was prepared by heating_ pentachlorophenol and 2,3,4,4,5(6-hexachlarc-2.5cyclohexadien-l-one; the yield was 84%. 310 Environmental Health Perspectives 784642 GENP 011831 >1959. T om ita , M., U eda, S., and N arisada, M. Dibenzo-p-dioxin derivatives. XXVII. Synthesis of polyhalodibenzo-p-dioxin. Yakugaka Zasshi 79: 186-192 (1959). Pentachlorophenoi was heated at 300"C for 12 hr. Recrystallization of the product from benzene, yielded octachlorodibenzo-p-dioxin needles. 1958 Sa n d er m a n n , W ., Ca st e n , R., Krasting, W., and P ie p e r , J . Studies on wood-protec tion chemistry. IX. Investigations con cerning oily blue stain protection agents. Holzforsch. Holsverwert. 10(4): 57-66 (1958). Compounds tormed by heat treatment of pentachlorophenol such as octachlorodiphenylene oxide were inactive for preventing blue stain. Tetrachlorodiphenvlene oxide was especially toxic to man and beast. 1957 Gilman. H., and Dietrich, J. J. Halogen de rivatives of dibenzo-p-dioxin. J. Amer. Chem. Soc. 79: 1439-1441 (1957). Dibenzo-p-dioxin was synthesized and reacted with chlorine gas to form 2-chIorodibenzo-p-dioxin; a similar run coupled with ultraviolet light gave 2,7dichlorodibenzo-p-dioxin. Kimmig. J.. and Schulz. K. H. Chlorinatedaromatic cyclic ethers as a cause of the so-called chloracne. Naturwiss. 44: 337338 (1957) (Ger.). Trichlordibenzo furan, tetrachioradibenzofuran, and 2,3,6.7-tetrachlorodibenzodioxin were especially effective acne-inducing agents on rabbit ears. 2,3,6,7Tetrachlorodibenzodioxin proved most toxic; the characteristic changes were achieved with concentra tions o f'0.01 to 0,002rV. Higher concentrations caused liver necrosis and death. Single oral administration of 0.05 to 0.1 mg/kg body weight caused death in 1-2 weeks. Autopsy revealed necrosis and fatty de generation of the liver. Pure 2,4,5-trichlorophenol and pentachlorophenoi did not cause chloracne. Kimmig, J., and Schulz. K. H. Occupational acne (so-called chloi*acne) due to chlori nated aromatic cyclic ethers. Dermatologia 115 (4): 540-546 (1957) (Ger.). The agent causing occupational chloracne in 31 employees of chlorophenol producing factories was found to be 2.3,6.7-tetrachIorodibenzadioxtn. The toxic action in animals was due to the alkaline hydro- lysis product of 1,2,4,5-tetrachlorobenzene. Tetrachlorodibenzo-p-dioxin and tri- and tetrachioradi benzofuran were active skin irritants. 2,3,6,7-Tetrachlorodibenzodioxin formation from sodium trichlorophenolacetate was established as was isolation from the by-products. Sa n d e r m a n n , H. S., Ca st e n , R., and Stock m a n n , H. Pyrolysis of pentachlorophenol. Chem. Ber. 90: 690-692 (1957). Octachlorodiphenylene dioxide was formed after pentachlorophenoi was heated at 300"C for 24 hr, distilled at 320C, and crystallized from benzene or extracted with benzene and fractional crystalliza tion. 2,3,7,8-Tetrachlorodiphenylene dioxide was ac tive against insects and wood-destroying fungi; octa chlorodiphenylene dioxide was inactive. Schulz. K. H. Clinical and experimental stu dies on the etiology of chloracne. Arch. Klin. Exp. Derm. 206: 589-596 (1957). The acnegenic properties occurring as a refractive dermatologic condition in factory workers involved in the production of chlorinated aromatic compounds were ascrihed to TCDD, an unwanted side product in the sythesis of 2,4,5-T. 1955 Oita. K,, J ohnson, R. G., and Gilman, H. The chlorination of dibenzofuran and some of its derivatives. J. Org. Chem. ` 20: 657-667 (1955). Step-by-step reaction procedures were presented for the chemical synthesis of chlorinated dibenzofurans. i.e., 2-chloro-, 2,3-dichloro-, 3-chloro-, and a chlo r od ibe nzo furan. 1954 Ce r n ia n i, A., F a s s e r in i, R., and Righi, G. Ultraviolet spectra of some dibenzofurans. Boll. Sci. Fac. Chim. Ind. 12: 75-79 (1954). 2-Chloro* and 3-chloro-dibenzofurans were charac terized by ultraviolet spectrometry. These derivatives had little effect on spectra of the parent compound. 1953 J ulia. M., and Baillarge. M. Growth factors in plants. III. (l-Carboxymethyi-2-naphthoxy) acetic acid and (2-carboxymethyl4-chlorophenoxy) acetic acid. Bull. Soc. Chim. France. 640-643 (1953). In the preparation of 2,4-dichlorophenol. a portion of the mixture was insoluble in benzene-water; the substance was extracted with boiling ligroine and recrystallized to yield- 2,7-dichlorodibenzo-p-dioxin. September 1973 311 784643 I l1952 SHIBATA, I. S., Natori, S., and SUMI, Y rA n - tibacterial effects of lichen substances and their related compounds. J. Pharm. Soc. Japan 72: 1333-1336 (1952). Synthesis of 2,6-dichlorodiphenylene dioxide, a light yellow compound with a melting point of 207 C and a dipole moment of 0.62 was described. The di pole movement of 2,6-dichlorodiphenylene dioxide was somewhat less than the dipole moment of di phenylene dioxide which was 0.64. The chlorinated dibenzofurans synthesized and tested for antibacterial activity included: 2-chloro-, 3-chloro-, 2,8-dichloro-, and 3,8-dichloro-. The chlo rine atom, when in the 8-position of the dibenzofuran ring-, increased its antibacterial effect. 1939 British Thompson-Houston Co. Ltd. Electric insulation. Brit. Pat. 506,560 (May 31, 1939). 1951 Tomita, M., and Watanabe, W. Antibac terial activity of some organic com pounds in vitro. II. Antibacterial activi ty of dibenzo-p-dioxin, phenoxathiin, and I, 4-benzodioxan derivatives on Mycobac terium tuberculosis, Staphylococcus au reus, and Escherichia coli. J. Pharm. Soc. Japan 71: 1204-1206 (1951). 2,7-dichlorodibenzo-p-dioxin was tested for antibac terial activity and found to be inactive. 1941 Direct chlorination of diphenylene oxide in the presence of a catalyst produced solid, waxlike pro ducts of high dielectric constants; i.e., mixture of chlorinated diphenylene oxide and pentachlorobiphenyl. Clark, F. M. Chlorinated diphenylene oxide composition for treating paper in capa citors, U.S. Patent 2,198,473 (April 23, 1939). Chlorinated diphenylene oxide containing 1 to 4 atoms of chlorine per molecule imparts a constant electrical capacity when used to impregnate paper in capacitors. Higasi, K. The molecular structures of di- 1936 phenylene dioxide and phenoxathliin as Bell, F. Pyrolysis of chlorophenols. J. Chem. revealed by dipole-moment data. Sci. Soc. 1936: 1244. Papers Inst, Pkys. Chem. Research 38: o-Chlorophenol was pyrolyzed at 180-340C; di 331-340 (1941). phenylene oxide and 3,6-dichlorodiphenyl oxide were 2.6- Dichlorodiphenylene dioxide exhibited a dipole isolated from the distillate. moment of 0.62 Debyes. This indicated a folded molecule and no optical isomers. 1934 Higasi, K., and Uyeo, ,S. Polarity and mole Gilman, H., Brown, G. E., Bywater, W. G., cular structure of diphenylene dioxide. and Kirkpatrick, W. H. Dibenzofuran. J. Chem. Soc. Japan 62: 396-399 (1941). III. Nuclear substitutions. J. Amer. 2.6- Dichlorodiphenylene dioxide was found to have Chem. Soc. 56: 2473-2477 (1934). a small but finite dipole movement. This molecule and 2-CMoro- and 2,8-dichlorodibenzofuran were syn the parent probably fold along the line joining the thesized. Dibenzofuran and chlorine in carbon tetra two oxygen atoms. chloride gave 389f- of the 2,8-dichloro derivative Ueo, S. 2,6-Dichlorodiphenylene dioxide. Bull. which had a melting point of 185C. Dibenzofuran in alcohol at 60C gave the 2-chloro derivative which Chem. Soc. Japan 16: 177-179 (1941). had a melting point of 102.5C. 312 Environmental Health Perspectives 784644 Erratum The following errors were made in the article by Ulrich R. Hoegg, cigarette smoke in closed spaces, experimental issue no. 2, October 1972. (1) Figure 2, left coordinate should read co(ppm) rather than co(ptm) (2) References 40-57 were omitted and are as follows: 40. National Center for Health Statistics. 1970. Changes in Cigarette Consumption Between June 1966 and August 1968. U. S. Public Health Service. Monthly Vital Statistics Report, Health Interview Survey Data. 12:1-4. 41. Boaz. J. N. (Ed.) 1970. Architectural Graphic Standards, Wiley & Sons, Inc. New York, p. 13. 42. Am. Soc. Heat. Rad. Air. Eng. 1964. Guide and Data Book: Applications. Ashrae. New York. p. 261. 43. Am. Soc. Heat. Rad. Air. Eng. 1963. Guide and Data Book: Fundamentals and Equipment. Ashrae. New York. p. 432. 44. Ibid. p. 260. 45. Callender. J. H., (Ed.) 1966. Time-Saver Standards: A Handbook of Architectural De sign. McGraw Hill Book Company, New York, pp. 734 and 749. 46. Turk, A. 1963. Measurements of Odorous Vapors in Test Chambers: Theoretical. Am. Soc. Heat. Rad. Air. Eng. Journal. 5:55-58. 47. Am. Conf. Gov. Ind. Hyg. 1970. Industrial Ven tilation. A Manual of Recommended Practice. 11th Edition, p. 2-1. 48. Atkinson, W. O. 1970. Production of Sample Cigarettes for Tobacco and Health Research. In: Griffith. R. B. (Director) Proceedings of the Tobacco and Health Conference. Conference Report--2. University of Kentucky, Tobacco and Health Research Institute, 2:28-30. 49. Wartman. W. B., Cogbill, E. C. and Harlow, E. S. 1959. Determination of Particulate Matter in Concentrated Aerosols. Analytical Chemistry. 31:1705-1709. 50. Morse, K. M., Bumsted, H. E. and Janes, W. C. 1971. The Validity of Gravimetric Measurements of Respirable Coal Mine Dust. Amer. Indust. Hygiene Assoc. Journal. 32:104-114. 51. Benner, J. F. 1970. Tentative Summary of Leaf and Smoke Analysis of the University of Ken tucky Reference and Alkaloid Series Cigarettes. In: Grirfith, R. B. (Director) Proceedings of the Tobacco and Health Conference, Conference Re port--2. University of Kentucky, Tobacco and Health Research Institute. 2:30-34. 52. Brief, R. S. i960. Simple Way to Determine Air Contaminants. Air Engineering. 2:39-41. 53. Owens, D. F. and Rossano, A. T. 1969. Design Procedures to Control Cigarette Smoke and Other Air Pollutants. Am. Soc. Heat. Rad. Air. Eng. Transactions, p. 93-102. 54. Johansson. C. R. and Ronge. H. 1965. Acute Irri tation Effects of Tobacco Smoke in Room Air. Nordisk Hygienisk Tidskrift. 46:49-50. 55. Vernot, E. H. 1968. Contaminant Trapping and Effect of Surface Adsorption on Atmospheric Concentration of Contaminants. In: Proceedings of the 4th Annual Conference on Atmospheric Contamination in Confined Spaces. 10-12 Sep tember. (AMRL--TR--68--175). 56. Morrow, P. E. 1964. Evaluation of Inhalation Hazards Based Upon the Respirable Dust Con tent and the Philosophy and Application of Selective Sampling. Am. Indust. Hygiene Assoc. Journal. 25:213-236. 57. J. Air Pollution Control Assoc. 1971. Environ mental Protection Agency Sets National Air Quality Standards. 21:352-353. G E N P 0 1 1834 September 1973 784645 313