Document OJ13OXgbwX2BpD2pkXQQoZNvL

-- / mg in a keeping ter. The insuring it a dis'sis. The io buffer I, taking > AE/Am ted from of man- des was occur in terniinacthod is Ml, Chief tent. O-to '/I - LSVBOKMKNTAL MWAKI 4, 481--495 (1971) Polychlorinated Biphenyls: Distribution and Storage in Body Fluids and Tissues of Sherman Rats1 August Curley, Virlyn W. Burse, Mary E. Grim, Ralph W. Jennincs, . a.vo Ralph E. Livder Chamblec Toxicology Laboratory, Environmental Protection Agency, 4770 Buford Highway, Chamblec, Georgia 30341 Received July 16, 1971 An analytical merited fur determination of PCB (Amdnr 1254) in blood, urine, fece*, and tissue* of Sherman strain rats after prolonged dietary intake (100, 500, and 1000 ppnt) is presented. The method involves hontogenixation and extraction with hexane, cleanup on a silica gel nticrocolumn and analysts by electron-capture gas- liquid chromatography. Recovery' data are shown at 0.1, 5, 14. and 100 ppnt. Muss spectroscopic characteristics of PCB in tissues are given. Distribution, storage, and excretion rates during feeding and after discontinuance of PCR in the diet arc presented. Polychlorinated biphenyls (PCBs) were first described in 18S1, by Schmidt and Shultz (1). Commercial production was begun in 1930 ( 2). These compounds have been and still arc widely used as plasticizers and resins and in chlorinated rubber (3, 4). They are known to enhance the insecticidal properties of lindane (5), DDT, and dieldrin (6). They may be used as stationary phases in gas--Iicjuitl chromatography (7). Their widespread use may have resulted in contamination of the cco-svstem as suggested by many recent reports. Aroclor1 and DDT were both used during World War II. In 1948, total organic chlorine analysis provided data for the first determination of DDT storage in man (8). Since this method has no specificity for DDT, the total organic chlorine content would not preclude the presence of other organic chlorine-containing moieties. In terms of halide specificity, the . Dohrmunn microcoulometrie gas chromatographic method for detecting pesticides was well established by 1960 ( 9), and electron-capture was in use by 1961 (10). Roburn (11) first observed extraneous peaks in 1965; he used a potentionu tric mode of analysis for total chlorine and found disagreement between this method and total chlorine content calculated from gas chromatographic results in which Argon ionization and electron-capture were used for the analysis of tissues and ' The Monsanto Company manufactures PCBs chiefly in the United State* under the trade name of Aroclor. They arc Manufactured in France by Prodelce, and in Germany by Buyer, with trade names, Phcnochlor and Colphcn, respectively, i These compounds are designated hy numbers The first two digits represent tire molecular type: 12, chlorinated biphenyls; 25 and 41, bleed* of chlorinated biplienyls and chlorinated tcrpiienyls {75*5 biphenyl and 60V biphenyl, respectively); 54, chlorinated tcrphenyls. The last two digits give the weight- percent of chlorine. Thus, Aroclor 1254 and 12(30, with which this paper is concerned, are chlorinated biphenyls containing 51 and 60V chlorine, respectively. y' 1972 hy Academic Prc^s, Inc. 431 " . 0S\N 332457 - STLCOPCB4079033 .u. 482 CURLEY ET AL. eggs of wild birds. The samples showed significant amounts of unknown elec tron-capturing compounds. Roburn postulated that these were metabolic inter mediates of one or mote of the organochlorinc pesticides. The differences could also have been indicative of PCB contamination. It was not until 1966 that S. Jen sen (12) using electron-capture and mass spectrometry was able to identify as PCBs unknown peaks found in the analysis of fish and a bird carcass. Un doubtedly, the time lag between initiation use and detection of PCB in the en vironment is a matter of accumulative concentration and/or instrumentation development. The presence of PCBs in environmental samples has been well established (13. 14). Acute and subacute feeding studies to determine toxicities and some residue levels have been done with rats, birds, and guinea pigs (15-18, 22). Tol erance studies have been done with fish and other forms of marine life (19, 21, 22). Pathological changes in the liver and other organs have been shown in guinea pigs, rats, and rabbits at high concentrations (22). Hydropericardium and growth depression proportional to dietary levels of the toxicant have been pro duced in the chicken (23). Polychlorinated biphenyls arc known to affect estrogen levels in birds which in turn reducts calcium reserves and results in thin eggshells. Chemical porphyria from oral administration has also been shown in chicks (18). This paper presents distribution, storage, and excretion rates of Arodor in cer tain tissues, body fluids, and excreta of rats following (1) a single oral dose, (2) repeated dietary intake, and (3) after discontinuance of PCB in the diet. METHODS Arodor 1254 and 1260 were supplied by the Monsanto Chemical Company with' respective lot numbers, AK-38 and AK-3. Dietary formulations were prepared according to the methods described by Caines and Kimbrough (24). '. Experiment A - Arodor 1254 and ,Arodor 1260 were administered at 1600 mg/kg and 3200 mg/kg, respectively, as a single oral dose, by stomach tube to three female, 173- to 183-day-old, white Sherman rats weighing 290-318 g. Four control rats, two in each group, were given peanut oil only. Twenty-four hours later the animals were killed, and blood (oxalatcd), fat, muscle (ahdominal), liveT, kidney, lung, and brain were taken for analysis. - . . Experiment B * Arodor 1254 at 1000 ppm was fed for 9S days to 29- to 34-day-old Sherman strain male and female weanling rats, in groups of 10. Controls were fed only plain chow. Food consumption was measured at intervals as described previously by Kimbrough and Caines (4). All rats were weighed on a w'eckly basis from experiment initiation until time of sacrifice or death. The rats consumed an aver age of 72.7 mg/kg/Aroclor/day. Blood (oxalatcd), fat, muscle (abdominal), liver, kidney, long, and brain were taken from each rat at time of sacrifice. Tissues were preserved in 10? formalin and held for analysis. DSW 332^58 STLCOPCB4079034 ,vn elecic intcrcs could it S. JenMitify as iss. Un i tlic cn* Kfitation tablished nd some >2). Tol(19,21, hcnvn in lium and icen procstrogen rggshclls. cks (18). or in cerdose, (2) t. oany with prepared and 3200 ,nak*, 173- rats, two ic animals ncy, lung. 1 Sherman o fed only previously basis from d an avernal), liver, issues were Urir+i POLYCHLOatNATIO B1PHZXYU t ,' 483 Experiment C Sherman strain male weanling rats (61- to 68-days-old) were fed a diet of plain chow or ch6w fortified with Aroclor 1254 at a concentration of 100 ppm for 58 days. Fortified chow was discontinued, and samples were collected during the recovery period. Food consumption was measured at 2-day intervals for a period of 8 days; after 42 days food consumption was measured daily for 4 days. Sur viving rats were weighed once a week. At various times during the experiment rats were ind.vidually housed in mctabol;sm cages. Urine and feces were collected over a 16-hour period during which time plain chow was fed. Following excreta collections, rats were sacrificed. Blood (oxalated), fat, muscle (abdominal), liver, kidney, and bruin were taken from each rat at time of sacrifice. Tissues, urine, and feces were frozen at time of collection. Whole blood was centrifuged and the plasma refrigerated. The outline of the experiment as aforementioned is sum marized as follows: These animals (two on the diet and one control) were sacri ficed on Day 4, 9, 13, 17, 25, 34, 42, and 58. After discontinuance of PCB in the diet, two animals were sacrificed on Day 8, 16, 24, and 71. Experiments D and E Sherman strain male weanlings (36- to 41-days-old) were f*d Aroclor 1254 at dietary levels of 100 ppm or 500 ppm for 252 days. Body weights were determined initially and weekly thereafter for 121 days. On the basis of food consumption measurements, the average rates for intake of Aroclor 1254 were 6.8 mg/kg/day and 36.4 mg/kg/day for 100-ppm and 500-ppm dietary levels, respectively. Sam ples were collected in the same manner as Experiment C. Sample Preparation Polychlorinated biphenyls like DDT, are not water soluble but arc, in fact, highly lipid soluble. Solvents for tissues and plasma extractions were selected for this property. . Plasma (at least 2 ml) was extracted by the method of Dale el al. (25). Later in the study, the Bryant-Thompson modification of this method was used (26). No significant difference has been found between these methods in this laboratory with the exception that convenience favors the Bryant-Thompson modification. Samples of fat (250 mg), liver, muscle, kidney, and lungs (500 mg each) were homogenized two times in 4 ml of hexane, and the extracts were combined. Brain (500 mg) was homogenized two times in 4 ml of acetone, and the ex tracts were comb:n?d (27). - Urine (mean volume of 10 ml) at pH < 7 (using pHydrion paper) was diluted to a volume of 25 ml by the addition of water and 1 ml isopropanol. These sam ples were extracted in 125-ml separatory funnels three times with 10 ml hexane. The hexane extracts were combined, washed once with 3 ml of water, and dried with 1.5 g of anhydrous sodium sulfate. Feces were dcsiccatyd over calcium chloride for 1 week. One gram was then pulverized in a mortar and pestle and extracted for 1 hour with 25 nil of acetone on an automatic shaker. The acetone extract was decanted and placed in a 50-ml DSW 332459 STLCOPCB4079035 J 484 CURLEY rr AL. centrifuge tube. Particulate matter in the extract was removed by oentrifuging and/or decanting. The extracts were evaporated with a stream of dry nitrogen to a volume of 0.5 ml in a water bath (40* ). The acetone extracts were evaporated fust to dryness and made to a volume of 0.5 ml in hexane. The samples were dried with anhydrous sodium sulfate and held for cleanup and analysis by electron capture gas-liquid chromatography. // Preparation of Siuca Gel and Micro-Column Silica gel (27) from Woelm (Activity grade 1) was activated at 130* for 4 hours and desiccated over calcium chloride for 16 hours before deactivation at 3S (v/w) by the addition of water. The silica gel was allowed to equilibrate at least 2 hours before use. After comparing the weight of the silica gel before and after activation, it was found to be activity grade I os specified by the manu facturer and fijrther activation was unnecessary; however, each new batch was routinely activated. One gram of the deactivated silica gel was placed in a disposable pipet, which had been loosely plugged with silanized glass wool, and gently seated. Columns were prewashed with 10 ml of hexane. Each sample was placed on the column in a volume of 0.5 ml. At least 1 ml of hexane was used to quantitatively transfer the sample to the head of the column. The column was eluted with 10 ml of a 1:1 benzene:hexane mixture (27). Samples were chromatographed in sets of eight with duplicate column blanks. Aroclor 1254 standards were eluted simul taneously in order to ascertain the variance in silica gel activity and to detect any possible contamination from solvents, glassware, or silica gel. Cas Chromatochapiiy Samples from experiments, A, C, D, and E as treated above were analyzed by electron-capture gas-liquid chromatography under one of the following condi tions using a MicroTek MT-220 gas chromatograph: pulse Mode (using dual columns and dual detectors); cell voltage, 36.5 volts; pulse width, 5.5 < scc; pulse rate, 120 /rscc; detector, TI, parallel plate design. No. 1 205, No. 2 198; column, X )' o.d. U-shaped pyrex glass packed with 57 OV 210 on 80/100 mesh Chromosorb VV, high performance, 170; inlet 23S; carrier gas, argon-methane (95-53), 50 psi; flow-through Detector No. 1, 50 cc/minute; Detector No. 2, 60 cc/minutc. The DC inode was a follows: cell voltage, 15 volts; detector, 'II, par allel plate design, 210*; Column, & X o.d. U-shaped pyrex glass packed with 1.5* OV-17/1.95S QF-1 on 100/200 mesh Chromosorb XV, HP, 200; inlet, 230; carrier gas, nitrogen, 40 psi; flow-through detector, 75 cc/minutc. Samples from Experiment B were injected directly as hexane extracts and analyzed by electron-capture gas-liquid chromatography under the following conditions using a MicroTek NIT 220 gas chromatograph: cell voltage, -40 volts DC; detector, B,Ni, 250; column, 6' X !" o.d. U-shaped glass column packed with 15* OV-17/1.95S QF-1 on 100/200 mesh Chromosorb XV, IIP 200^; inlet, 210*; carrier gas,'nitrogen, 40 psi: flow-through detector, SO cc/minutc, 20 cc/minute scavenger plus 60 cc/minute carrier gas flow. DSW 332460 STLCOPCB4079036 ifuging jgcn to wratcd c dried lectron ' for 4 tion at irate at >re and manuch was wliicli olnmns column ransfcr nl of a sets of simulRct any zed by condi>g dual ; pulse olumn, ) mesh icthanc ). 2, 60 II, par 'd with t, 230; :ts and Mowing !0 vults packed inlet, c/min- m | A .1 polychlorinated biphenyls , 485. r Recovery Studies ----- - Recovery studies were completed using the following methods for control tis sues, plasma, and 48-hour urine and fcccs collections. The fortification level for each type of sample was the mean concentration, with the exception of fat, of Aroclor 1254 found for that sample in Experiment C. Fat samples were fortified with 100 ppm. Tissues (fat, liver, kidney, brain, and muscle): Aroclor was added directly to the hexane extract prior to liquid chromatography. Plasma: Aroclor was added directly to fresh plasma, thoroughly mixed using a mini-mixer, and allowed to equilibrate overnight under refrigeration. Samples were then treated as mentioned in methods. Feces: Aroclor was added directly to pulverized dry feces and treated ns men tioned in methods. Urine (three slightly different approaches were used): No. 1--Aroclor was added to 1 ml of isopropanol, then mixed with a solution of urine and water (10 ml + 15 ml) and analyzed as mentioned above; No. 2-- Aroclor was added to a hexane extract of 10 ml of urine which had been prepared as mentioned above; No. 3--Urine (5 ml) was added to a known quantity of Aroclor in a 15-ml centrifuge tube, mixed on a minimixer, and allowed to equili brate at room temperature for 16 hours. Urine vvas extracted three times with 3 ml of hexane; the hexane extracts were combined and evaporated to 0.5 ml. RESULTS AND DISCUSSION Routinely, this laboratory operates dual channel solid state electrometers at a sensitivity of 4 X 10 10 amps full scale which permits the detection of 10 to 30 picograms of the more common chlorinated hydrocarbons. There was a significant difference between the response characteristics of the Aroclors as compared to die chlorinated hydrocarbons. To achieve a representative gas chromatographic trace, at least 350 and 490 pg of Aroclor 1254 and Aroclor 1260 were required. A gas chromatogram illustrating the difference between the electron-capture re sponse characteristics of these compounds as compared to chlorinated hydrocar bons is shown in Fig. 1. Experiment A Data resulting from electron-capture analysis of plasma and tissues in Experi ment A are presented in Tabic 1. Quantitation is an approximation. In all cases, the peak heights (in mm) of all PCB components were summed for the respec tive standard at a given picograin quantity injected. A response factor was deter mined for each standard in pg/mm. For each sample, a summation was made of the total response (in mm), and the product of this times the response factor cal culated to give picograms of PCB or PCB-dcrivcd material in the sample. When the gas chromatograms for the samples were compared with those of the Aroclor Standard, there were no recurrences of the general Aroclor pattern. Some peaks had completely disappeared. This w as also true for each of the succeeding studies. Although the dosage level of Aroclor 1260 was twice that of Aroclor 1254, the ' i\ . DSW 332461 STLCOPCB4079037 m i a i r p u 11 mi iy ; * kn Fic. 1. Cos chromatograms for Arodor 1254 and chlorinated hydrocarbon standards. (A plot of detector response vs tirrie.) - TABLE I Distribution or Akocloka 24-hours aftkk Oral Ingestion by Stomach Tube Aroclor 1254 (dosage level-- 1600 mg/lcg) Aroclor 1260 (dosage level--3200 mg/kg) - Plasma 'Mean 24 03 SE 8 0S Brain . .. Mean 13S 00 SE 36 41 Fat Mean 11469 . BE 574.5 Liver Mean 141 20 SE 47 80 Kidney Mean 274 ai 8E 30.47 Lung Mean 65 91 SE 20 4S Muscle Mean 80 2M SK Cl -V ----- 1----------------------- ---------- 15,70 0.99 14517 23 60 030.0 426.6 236 1 116 3 328.5 114.9 105 17 21.13 37.00 22 37 * DSW 332462 l STLCOPCB4079038 (lards. (A ) *VBE POLYCHLORINATED BIPHENYLS ; I' 487 mean storage levels after 24 hours were essentially the same. Considerable vari ation in the storage levels was found in each group. Differing absorption rates from the gut might account for the variation in levels found; variation was highest \ in the fat and lowest in the plasma. According to Hayes (35), acute dosages, such as the one given here, do not result in storage, but rather in a flooding of * the organism. Experiment B Only surviving rats were analyzed in Experiment B. Five of the ten male rats died--one on the last day (day 9S), and one each on days 85 and 56 and two on day 50. Eight of the ten female rats died--one each on days 89, SO, 47, 41, and 35, and three on day 55. Tucker and Crabtree (15) fed male albino rats 1000 ppm Aroclor 1254 and reported lethal effects to none of 6 by 14 days, 1 of 5 by 28 days, 3 of 4 by 43 days and 4 of 4 by 53 days including a 21Z reduction, in food con sumption. All survivors in Experiment B appeared to be normal; however no in crease in weight was observed in surviving males after 84 days of dietary intake and after 77 days for surviving females. Food consumption for dosed males and females was 66 and 647, respectively, of consumption by controls. Data from electron-capture analysis of tissues and plasma are presented in Table II. The variation in the concentrations of PCB in this experiment, as in Ex periment A, is least in the serum and brain and highest in the muscle and fat. A significant difference between males and females was not found for storage of TABLE II Dibtkihction of PCB-Dkkivkd Matkhial Foli-owing OS-Pay F.xposukk to a Dietaht Level or 1000 PPM Akoclok 1254 . Male* (ppm) Female* (ppm) Male* v Females ' Plaaiua Mean 17 t SE 6 Fat Mean SE 1I27N 5742 Muade Mean 155 SE 117 Lung Mean 7N SE Zi Brain Mean 111 S SE + u Kidnev 56 4' SE ' H Liver Mean 155 SE 30 IS 5 S431 579 753 - 721 52 20 V4 21 54 ` 24 210 13 p > 0.50 p > 0.50 p > 0.20 . p > 0.50 p > 0.50 p > 0.50 p > 0.20 DSW 332463 STLCOPCB4079039 488 CURLEY ET AL. Aroclor 1254. Hayes (35) and Dale et al. (36) report that storage of DDT in female rats exceeds that of male rats wlten both ore maintained on the same diet and that the enhancement in storage exceeds the greater food intake by females. The absence of certain peaks in the Aroclor standard when compared to PCBderived material in the samples became more dramatic in this experiment as compared to Experiment A. Generally, the percentage of matching Aroclor peaks found in each tissue was as follows: plasma SIS, fat 74S, brain 66S, liver 64?, lung 63S, muscle 62S, and kidney 58?. These differences could be attributed to: (1) selectivity in storage for each tissue; (2) enzyme activity resulting in some form of metabolism; (3) excretion favoring some isomeric forms as opposed to others. Experiment C The data from the analysis of tissue, plasma, and excrement obtained according to the schedule outlined in Experiment C arc shown in Figs. 2 and 3. There is a steady buildup of Aroclor in all tissues, while the excretion trend is quite erratic. 4i s;o FlC. 2. Distribution of PCB-dsrrived material in tissues and plasma from rats on a dlr-tary Ifvd 109 ppm Aroclor 12A1. sw 332464 STLCOPCB4079040 DT in re diet inolcs. i PCBent as peaks >r 64?, ed to: i some sed to ording re is a :rratic. i ------- -- f 't--T 1 'rii-i V*--rl *- POLYCHLORINATED BIPHENYLS .! }' 490 A 5 dietary Fie. 3. Concentration of I*CB and PCB-derived material in feces and urine from rats on a dietary level of 100 ppm Aroclor 1234. - The excretion trend may be due in part to the discontinuous collections; there fore, these results reflect only the excretion pattern at specific times. The residue levels observed in Experiment B created some doubt that levels observed after the 58-day dietary intake were representative of steady state values for Aroclor 1254. Hayes (35) states that DDT fed to rats at a constant rate is increasingly stored in their fat until it reaches a plateau, and on a diet containing 200 ppm or less this plateau-is reached within 90 to 140 days. Dale et al. (36) found mean concentrations of DDT and DDE in the fat of male rats fed 200 ppm in the diet for 90 days to be 525 and 51 ppm, respectively, while male rats on the same diet for 140 days had mean concentrations of 506 and 42 ppm. A maximum concentra tion in liver, kidney, and brain is achieved within a few days for a dietary level of 1000 ppin (35). Experiment D and E Results of Experiment D showed that rats on the same dietary levels stored more Aroclor in their tissues nftcr 240 days than at the end of 58 days. The 240day group showed no significant difference in the quantity of the PCB-derived material found in the urine but showed significantly more PCB excreted in the feces. Analysis of tissue, plasma, and excreta from Experiment E indicates that DSW 332465 STLCOPCB4079041 400 CURLEY ET Al_ Aroclor stomge, like DDT, is directly related to the daily dosage. However, the points at which equilibrium storage is achieved arc yet to be determined. At dietary levels of 100, 400, and 800 ppm for 2 years (37), the mean values of - DDT stored in the fat of rats at equilibrium were reported as 95, 1028, and 4200 ppm, respectively. The concentrations of Aroclor 1254 found in the fat after 240 days at dietary levels of 100 and 500 ppm were 1101 and 10,021, respectively. Generally, the levels of DDT stored in the fat of male iats after 180 days are equivalent to levels after 2 years (37), although there is some reduction in DDT storage beyond 2 years. These levels in the fat illustrate the vast differences in storage at the steady states for these two compounds, and the vast differences in their relative toxicitics and/or lipid solubilities. (The brain concentration, rather than fat concentration is a better indication of toxicity) (38). DDT storage can be reduced if exposure is reduced or discontinued. Lang and 1 Fitzhugh (37) found retention of 50-75? and 25? of DDT in the fat, 30 and 90 ! ' days after discontinuing diets containing 5 to 50 ppm. After 58 days on a dietary ! level of 100 ppm Aroclor and a recovery time of 71 days, 80? of the concentration in the fat remained. Elimination seems to parallel that of DDT. These data indicated the concentration of Aroclor found at time of sacrifice to be in the following order: fat > liver > feces > kidney > brain > plasma > urine. The order of storage for tissues and plasma seems to parallel lipid content. Fecal excretion exceeds urine excretion with Aroclors as it docs with DDT and other chlorinated hydrocarbons. Evaluation or Method - The efficiency of the silica gel microcolumn and the extraction techniques em ployed were evaluated by determining the recovery of Aroclor 1254 from in vitro fortification. Recovery data for all samples eluted from a silica gel microcolumn , are presented in Table 3. All in vitro recoveries, whether from liquid chromntography or a combination of extraction and liquid chromatography, were at an acceptable level with the exception of urine. Recoveries of DDT from human urine using liquid-liquid extraction techniques have been reported by Cueto and , Biros (28) and Cranmcr et al. (29) with recoveries of 72 and 96?, respectively. In both instances, their extracting solvent or combination of solvents was more polar than hexane. The recovery in approach No. 1 and possibly 3 might be . attributed to the partition coefficient of hexane in an aqueous system. In order to evaluate this, reciprocal p-valucs were determined using tire technique of Bow man and Bcroza (30). The average of triplicate analyses gave a reciprocal p-value of 0.761, or 76.47 of the Aroclor was partitioned into the hexane layer from the aqueous system. This value agrees favorably with approach No. 3 but docs not account for the 17? difference in approach No. 1. Zitko (20) states that he found fractionation taking place when emulsions resulting from mixing Aroclor 1254 with water were broken by centrifugation. Gas chromatographic analysis of a hexane extract of the supernatant showed it to be richer in the lower chlorinated ` biphenyls than the original preparation. This varied with batches of Aroclor. . When a peak-for-peak comparison was made, whether in vitro fortification or p-value stud.es were being considered, recovery was generally low for all cojn- ' ' os**314* STLCOPCB4079042 I lUllfrf 44b. >ver, the alucs of nd 4200 Iter 240 ectively. lays arc in DDT miccs in ences in t, rather - uig and i and 90 dietary ntration rifice to asma > content. DT and ues em in vitro column nnatoge at an human cto and ctively. is more ight be irder to if Bowp-value roni the Iocs not c found or 1234 sis of a rinated Aroclor. ition or all con- i FOLYCULORINATED BIPHENYLS 491. TABLE III Recovery or Im Vitro Fortification or Control Samfles with Aroclor 1254 Tagus Weight Fortified (ppm) AR 1254 Percent recovery* Plasma Fat liver Muscle Brain Kidney Feees Urine (Approach 1,2,3) . 2.0 .350 .500 .500 .500 .500 1.000 0.5 100 6.0 4.0 3.0 2.0 14.0 0.1 0.1 0.1 08.5 82.2 94.8 103.1 100.4 100.6 00.8 62.0 114.0 77.0 Plasma, fat, liver, muscle, brain, kidney, feces--mean of 4 determinations, and Urine X mean of duplicate denominations. . stituents, and the average recovery was reported. Low recovery from the urine . was probably due to the extracting solvent. The values reported were not cor rected for in vitro recoveries. Although p,p'-DDT und p,p'-DDE were determined at the 0.01 ppm sensitivity level in liver from a control rat, the degree to which silica gel removes interfering moieties from tissue or excreta extracts cannot be determined at this time. Before and after elution from silica gel there usually was a marked difference in the ap pearance of the extracts depending on the weights of tissue and excreta used. (Law and Goerlitz (39; report the success of pigment removal with silica gel from water samples.) Muscle and plasma were the least affected by liquid-solid chromatography and could have been analyzed very easily without this treatment. In fact, for the con centrations of Aroclor found, most of the samples could have been analyzed without any cleanup, taking into consideration the problems associated with column and detector contamination. After elution patterns of Aroclor 1254 were established, 10 ml of benzene: hexane were not required for quantitative elution of PCB, in fact, only 4 nil were required. This saved time in subsequent analysis. The elution pattern of Aroclor 1254 was unchanged at the fortification levels mentioned in Table III. The Aroclor 1254 standard that was chromatographed simultaneously with each batch of samples analyz.cd had a mean recovery of 104% with a SE of 3.1 for 23 samples. Interference of DDT with PCB Analysis The presence of PCBs, DDT, and DDT-like moieties in samples presents a problem both qualitatively and quantitatively during analysis. Initially, samples of liver and fat from control rats for each sampling period were analyzed for DDT and its metaliolites; samples were taken from the oldest control rat in the study. These tissues usually showed measurable amounts of DDT, DDD, and DDE. The values for liver and fat, respectively, were as follows: p.p'-DDT, 0.046 and 1.366 ppm; p,p'-DDD, 0.03G and 0.5-16 ppm; p.p'-DPE, 0.016 and 1.09S ppm. Since preliminary evaluations showed liver and fat storage of PCBs to be a i DSW 332467 STLCOPCB4079043 i ---------- ----- A --------- -- 492 CURLEY ET AL. minimum of 10 to 100 times the above mentioned values, interference was con sidered to be nonexistent. However, three possible approaches to separation of PCB and DDT-like materials without any chemical degradation of the pesticides ooncerncd were evaluated. Thin-layer chromatography was considered first. Using silica gel G and essentially the techniques as outlined by Walker and Beroza (31), both Aroclor 1234 and Aroclor 1260 in the presence of the chlorinated hydro carbons, p,p'-DDE, p,p'-DDT, o,p'-DDE, o,p'-DDT, dieldrin, and heptachlor epoxide, moved with the solvent front. Using polar and nonpolar mobile phases, the Rf values for PCBs and p,p'-Dt>E were identical and not resolved. This can possibly be used in confirmation and identification of Aroclors. The two other approaches utilized column chromatography. Florisil was used, is proposed by Reynolds (32). PCBs are eluted from the column with hexane, while other chlorinated hydrocarbons arc eluted with a diethylether: hexane mixture. Reynolds evaluated Florisil with a mixed standard of Aroclor 1254 and a chlorinated hydrocarbon mixture containing lindane, heptachlor, aldrin, heptachlor epoxide, p,p'-DDE, dieldrin, p,p'-DDD, and p,p'-DDT. Aldrin, pp'-DDE, and heptachlor are eluted in the PCB fraction. Although these pes ticides were not separated, this approach might still be acceptable since aldrin and heptachlor are seldom found in environmental samples in their unepoxydated forms. However, when an attempt was made to reproduce the work of Reynolds using PR grade Florisil, a mixture of Aroclor 1254 plus the chlorinated hydro carbons, lindane, heptachlor epoxide, dieldrin, ptp'-DDD, p,p'-DDE and p,p'- DDT, measurable quantities of p,p'-DDT, p,p'-DDD and lindane were found in Fraction 1, i.c., 66, 10, and 321 of the total, respectively. Reynolds docs not men tion the quality of the. Florisil although he docs mention how it was handled. The problem, however, seems not to be related to the grade of Florisil. Bcvenue and Ogata (33) have presented data using both qualities of Florisil and were unable to reproduce the work of Reynolds. There is some disagreement between the fractioning pattern of Bcvenue and Ogata (33) and the one observed in this work. This could probably be due to the quality and treatment of the Florisil. These problems suggest that the elution characteristics of Florisil should be evaluated Irofoie use for separating PCB and DDT. Armour and Burke (34) used a column of silicic acid (treated) and Cclitc 545. The column is eluted first with petroleum ether for PCBs and then with a mixed solvent system of dichlorpmcthanc-acctonitrilc-hcxanc (80:1:19) for the chlo rinated hydrocarbons. This system w;is evaluated using standards at levels below those that would be expected in environmental samples, i.e., 0.5 ppm Aroclor 1254 and Aroclor 1260 and 0.1 ppm y- and /3-isonier of BI1C, heptachlor epoxide, oj)'-DDE, p.p'-DDE, dieldrin, p.p'-DDD, and p,p'-DDT. Separation has been n problem with the p,p'-DDE; Fraction 1 contained 207 p,p'-DDE with the remain ing 805 in Fraction 2. Although some recoveries were less than 855, all other pesticides evaluated were only found in Fraction 2. According to Armour and Burke (34) the elution pattern can be altered by the water content of the silicic acid. The first pesticide to be affected in the group of compounds that were used to show the changes was p.p'-DDE. . The fact that there was not a complete peak for pcal^ match between stored and I DSW 332468 STLCOPCB4079044 vas oon-ation of esticidcs ;t'. Using za (31), 1 hydroptachlor phases, rhis can xs used, n with vlethcr: Aroclor tachlor, Aldrin, sc pcst aldrin cydatcd cynolds hydro- d P.P'* tund in >t mend. TV uc and unable cn the in this ``kjrisil. uld be tc 5-45. mixed .* chlobclow r 1254 oxide, x-cn a mainother ir and silicic ; used d and * DSW 332469 STLCOPCB4079045 i. -- .. UL..AU iMk. 494 cuinxr rr al. r excrctcd PCB and the dietary standard is not new (Fig. 4). Whether this is evidence of metabolism is yet to be determined. These differences were success fully demonstrated by electron capture gas-liquid chromatography and gas chro matography-mass spectrometry in that there were changes in the relative abun dances of some of the components of Aroclor 1254 in fat and the urine samples. This has not yet been demonstrated in any other tissue. The mass spectra showed molecular ions at m/e 288, 323, 358, and 392 con taining CU, CU, Cl*, and CI; isotopic clusters in the fat and 254, 2S8, 324, and 358 in the urine. Two additional ions which were observed in the urine only, were m/e 304 and 405. These ions contained the Cl4 isotopic cluster The presence of unaltered Aroclor in the feces may be indicative of lack of absorpt'on; this was the only sample that showed no alteration when compared to standard Aroclor 1254. The neutral urinary excretion pattern was quite the contrary to feces. This may in part be due to a solubility phenomena which Zitko (20) encountered when working with PCB and water. Tire situation of Zitko with either urine or alcoholwater mixtures under in vitro conditions was not reproduced. Under in vivo con ditions, there may be selective solvation of certain constituents of Aroclor. Selected samples of urine were analyzed by gas chromatography using the elec trolytic conductivity detector in tiro reductive mode. The presence of chloride ion was further substantiated. The retention times of these components compared favorably with peaks in the Aroclor standard. CONCLUSIONS It is hoped that these data will elucidate some of the problems inherent in studying the polychlorinated biphenyls: and that some correlations can be found between the residue levels reported and the LDM determinations including the pathological findings to be reported in a later publication by Kimbrough et al. (40); There arc some basic generalizations that can be drawn from the data presented: (1) Following dosage with Aroclor 1254 or 1260, whether acute or chronic, residue amounts can be detected in all body tissues, fluids, nnd excrement. (2) At the same dosage level rats store more PCBs than DDT. (3) No sign:ficant difference is apparent in the storage of PCBs by male or female rats when fed the same dietary levels. (4) Polychlorinated biphenyls are stored primarily in adipose tissue. (5) Since there was no recurrence of the general Aroclor pattern in the gas chromatograms or total ion current traces of the mass spectrometer between the Aroclor standards and components observed in the fat nnd urine, pos sible metabolism or differential absoqition is suggested. ' ACKNOWLEDGMENTS ' Wt Mrs. Estelle Cray for statistical analysis, Mr. TIithiuis B. Caines and Dr. K. D. Kinilirnnch for assistance with the animal studies. . ' REFERENCES ' l. Sciimiot, H., and SiicLTA, G. (1881). Arm. Clicm. 207, 338--11. 1. PfcAKW.i., Dwid R., axd Lixceh, J. L. (1970). Bio. Sci. 20 ( 17). ' os*3'*410 ( STLCOPCB4079046 *< I r this is success es chroe abuntamples. 192 conand 358 Iv, were lack of mpared his may d when ilcoholvo conVroclor. ic elccridc ion mpaicd rent in ? found ing the h et al. ic data `hronic, rement. nalc or the gas etween it-, pos- r. R. D. i Us -- i'mW -I.. si. >> mtA POLYCHLORINATED BIPHEVYLS 495 3. Monsanto Technical Bulletinf 0/PL-311A, O/PL-306, and O-FF/1. 4. Monsanto Technical Bulletins 0/PL-311A, 20. 5. Hornstein, I., and Sullivan, \V. N. (1953).- J. Econ. Entomoi. 46, 937. 6. Lichtenstein, E. P. (1969). J. Econ. Entomoi. 62, 761. 7. Supelco, Inc. Catalog (1970). 8. Howell, D. E. (1946). Pate. Okla. Acad. Set. 29, 31. 9. Coulso.y, D. M, Cavanach, L. A., DeVries, J. E., and Wolther, B. (I960). J. Agile. Food Chem. 8, 399. 10. Cooowlx, E. S., Cokedon, R.a and Reynolds, J. G. (1961). Analyst (London) 81, 697. 11. Roburx. J. (1965). Analyst (London) 90, (1073 ) 443. 12. Author unknown (1986). New Set. 32, 612. `. 13. Risebrouch, R. W., Rieche, P., Pkamli, D. B., Herman, S. C., and Kirven, M. N. (1960). Saltire 220, 1098. 14. Jensen, S. (1889). Nature 224, 247. 15. Tucker, R. K., and Crabtree, D. C. (1970). "Handbook of Toxicity of Pesticides to Wildlife." U. S. Dept of Interior, Bureau of Sport Fisheries and Wildlife. Resources Pub lication No. 84. . 18. Freak, D. E. II. (1968). "Pesticide Handbook Entonin." College Science Publishers, State College', Pa. 16801. 17. Prestt, I., Jeffries, D. J., and Moore, N. W. Environ. Pollut. 1(1), 3-86. 18. De Vos, J. C., and Koeman, J. H. (1970). Toxicol. Apfd. Pharmacol. 17, 656-68. 19. Duke, T. W., Lowe, J. I., and Wilson, A. J., Jr. (1970). Bull. Environ. Contamin. Toxicol. 5, (2). . 20. Ziteo, V. ( 1970). Bull. Environ. Contam. Toxicol. 5 ( 3), 279-285. 21. Wildish, D. J. (1970). Bull. Environ. Contam. Toxicol. 5 ( 3), 202-204. r 22. Miller, J. W. (1944). Pub. Health Rep. 59 ( 33), 1085-93. i 23. Flick, D. F. (1965). Poultry Sci. 44, 1460. i '24. Gaines, T. B., and Kimbrough, R. D. (1964). Bull. W. II. O. 31, 737-745. : 25. Dale, Wm F , Curley, A., and Cueto. C. (1966). Life Sci. 5, 45-51. 26. Ervant, A. C., and Thompson, J. (to be published). 27. Jennings, R. W. (1968). Biological Sample Cleanup for Chlorinated Hydrocarbon and OrganopItospSnrus Analysis Using Electron-Capture Gas-Liquid Chromatography. Paper presented Natl. ACS Meeting, Atlantic City, NJ. 28. Cueto, C., and Biros, F. (1967). Toxicol. Appl. Pharmol. 10, 261-269. ' 29. Cranmer, M. F., Carroll, J. J., and Copeland, M. F. (1969). Bull. Environ. Contam. Toxicol. 4(4), 214-23. . 30. Bowman, M. C., and Behoza, M. (1965). Journal of the Association of Ofjickd Agri cultural Chemists 4S (5). 913-9-52. 31. Walker, K. C., and Beroza, M. (1963). Journal of the Association of Official Agri cultural Chemists 46 (2). 250-261. 32. Reynolds, L. M. (1969). Bull. Environ. Con/am. Toxicol. 4 (3), 28--12. 33. Bevenue, A., and Ogata, J. N. (1970). J. Chromato. 50, 142-44. 34. Armour, J., and Burke, J. (1970). Journal of the Association of Official Analytical Chemists 53 (4), 7G1-6S. 35a. Haves, W. J., Jr., (1959). "DDT," S. W. Simmons, Birkhauser Verlag, Basel, b. Hayes, W. J., Jn. (1965). Ann. Rev. Pharmol. 5, 27--52. 36. Dale. W. E., Caines, T. B.t and Hayes, W. J., Jr. (1962). Toxicol. Api>l. Pharm. 4 ID, 37. Lanc, E. P., and Frmivcii, O. C. (1916). J. Pharm. S7( 1), 18-23. 38. Dale, W. E . Caines, T. B., and Hayes, W. J,, Jr. (1963). Science 142, (3398), 1474, 78. 39. Law, L. M., and Goeki.itz, D. F. (1970). Journal of the Association of Official Ana- lytiral Chemist. 5.3 (6), 1276-S6. 40. Kimkrolch. R. D., and Gaines, T. B. Arcli. Environ. Health (in press). DSW 332471 i I I STLCOPCB4079047