Document bBeq1LB3zXeoR73eXY9oMOp11

IV. ENVIRONMENTAL FATE AND EFFECTS IV.Persistence. Metabolism and Fate IV.1.1. Persistence PCB have a long Ufa In tha environment (1-3). Tha more chlorlnatad compounda In particular ara raslatant to metabolism (Appendix E) and parslat in aoila (305), water (41, 42)i sediments (50, 99, 105) and biological tlsauaa (111, 130, 141, 155, 170, 208, 35) for weeks, months or even pears. Indirect evidence for their persistence in the anvlronnant is provided by their wide distribution, even In remote parts of the earth where PCBs are unlikely to have been used (1-3, 56, 225, Appendix C). - Precise data on the life-time of PCBs after release Into the environment are difficult to obtain, because natural systems ara open: If a decline In concentrations Is observed it Is difficult to distin guish degradation from net export. Horn at al. (329) provided.almost the only unequivocal data: they found PCBs In dated (varved) sedi ments In the Santa Barbara baaln off southern California. PCBs ware first detectable In sediment layers dated to the mid-1940's and their concentration in the overlying sediments Increased steadily In para llel with the known Increase In PCB usage. This shows that PCBs can last for at least 30 years In these clrcusstances (anaerobic sediments with no burrowing organisms). More recently Risebrough jt al. (56) have Identified PCBs In Antarctic snow at depths up to 6 metres be low the surface. . 0073003 WATER PCB-00043084 277 Figure IT.X.l shows concentrations of PCBs measured In oysters In Escambia Bay, Florida, alnee 1969 (330). PCBa In Escambia Bay ara S. . ' baliavad to hava originated largely from a alngle point aource. Iden tified and cloaad In 1969 (99, 330). The residues In oysters In tha bay show seasonal fluctuations (related to spawning) but otherwise have declined only slowly over tha years since 1969. Ull these data Indicate that the more chlorinated PCBs have a life-time of years, if not decades. In at least certain compartments of the environment (2, 331). IV.1.2. Differential Persistence of Lower and Higher Chlorinated Blnhenvls. Nlsbet and SaroSm (332, 2) pointed out an anomaly, viz.. that most PCBs found In environmental samples consisted of penta- and higher CBs, whereas a substantial fraction of the PCBs released Into the environment In the past must have consisted of tatra- and lower CBs. They suggested that the discrepancy was probably due to differ ential degradation of the lower CBs, ruling out the main alternative explanation (differential mobility) primarily because'samples of PCBs from remote areas also consisted primarily of higher PCBs. A critical set of observations was that of Teith and Lee (333, 334), who showed from samples of water and flah from the Milwaukee River that the' proportion of lower CBa declined as tha water and sediments moved downstream away from the major Industrial sources. However, this decline could be explained either as differential volatilization or differential degradation (333). 0073004 WATER PCB-00043085 > RESIDUES OF AROCLOR 1254 IN OYSTERS , -' . .--v i ' ILZ \ WATER PCB-00043086 - %7? The differential persistence and/or bloaccunulatlon of lower and higher CBs Is Important in establishing criteria, because It 1m .v piles differential exposure of human and other target organisms to the various components of the various commercial mixtures. According ly the following sections of this document will pay particular atten tion to differences In environmental behavior between lower and higher CBs. The potential for hisnan exposure to tetra- and loirar CBs Is of especial Importance, because the main difference hetween Aroclors 1016 and 1242 Is the lower proportion of the more persistent penta- CBs In the former (Table IX.3.1). Xn 1972 It was believed that most human exposure was to penta- and higher CBs, but recently more evi dence for human exposure to tetra-CBs has' been presented (331, 332). IV.1.3. Metabolism Studies of the metabolic transformation and degradation of PCBs are summarized In Appendix E. Bacteria are able to metabolize biphenyl and mono- and dl-CBs fairly rapidly, but tri- and tetra-CBs are degraded more slowly and penta-CBs hardly at all (Section E.l). Accordingly there Is considerable differentiation of TCB mixtures during microbial degradation, as Illustrated In Figures E.4 to E.7. Figure XV.1.2 (from ref. 13) shows that even 14 days' Incubation with activated sludge had little effect on the higher components (mostly tetra-CBs) In Aroclor 1016. Aquatic Invertebrates and fish are able to metabolise tri-CBs to a substantial'degree but have very little ability to metabolize tetra- or pentawCBp to non-polar products (Sections E.2 and E.j)^---- "ooio0b WATER PCB-00043087 AROCLOR 1016 BIODEGRADATION 100U00 TIME (MIN) Figure IV.1.2 WATER PCB-00043088 U( Birds and mumals can metabolise tetra- and penta-CBs to hydroxy darlvatlvas at varying speeds, but they have very little . ability to metabolize haxa- or higher CBs (Sections E.4 and B.5). The principal mechanism of metabolism appears to be via oxida tion to arena oxide (epoxide) Intermediates, followed by rearrangement ('DIB shift') to form hydroxy-CB*. Hydroxy1stIon at the 3- or 4- posltions Is favored (Section E.6). lass Important mechanisms of metabolism are direct hydroxylatlon, dechlorination, and Isomeriza tion. There Is evidence for metabolic formation of CDFs In chickens and rats (Section B.7). Unidentified non-polar metabolites are stored In Invertebrates and fish (Section B.3). Metabolism does not always represent detoxification, since CDFs are much more toxic and at least one hydroxy-CB Is substantially more toxic than the parent compound. The formation of arena oxide Intermediates Is of much concern because this reactive class of com pounds Is Implicated as causative dgents In toxic, carcinogenic, and mutagenic effects (Section E.8). . '" IV.1.4. Photodegradation ' The photochemical properties of PCBs were summarized la Section II.8. Photolysis of PCBs has bean studied only under laboratory conditions and It Is difficult to use the data to predict --- rates of degradation In the environment. In the experiment which most nearly simulated environmental condltlooa (61), there was little overall degradation after 3 weeks In sunlight and the principal net effact .was a shift-from higher to lower CBs In the mixture (Table II.8.3). 0073008 WATER PCB-00043089 "'V ' ( . 2 Photodegradation doaa not necessarily represent detoxification, alnce lover CBa are foraad by dechlorination and email quantities of CDFs Sl. are:sometimes formed (Sections H.8.2, II.8.3). . IV. 1.5. Transport .. PCBs are mobile In the environment and may be transported In solution, by motion of suspended sediments, as vapors, on airborne particulates, or in the tissues of mobile animals (2, 12, 332, 333, - 47, 50, 52, 53, 54, 56, ate.). They are concentrated at tha alr/vater Interface (52, 55) and may be transported across It by several mach- anlsms. Including volatilization, solution,-dry fallout of particulates, \ precipitation In rain and snow, or ejection In spray (Section II.8). Critical measurements are scanty and conflicting (56) and It Is still not possible to construct satisfactory models of transport to and from tha aquatic environment (56, 331). Within bodies.of water, there la evidence that the transport of PCBs Is controlled by the presence and transport of sediments (Section II.6.5). Tha presence' of sediments or particulates Inhibits . of PCBs volatilization/(41, 42) and Is strongly associated with their trana-___ port (50, 56, 12). In natural waters with high sediment loads It may be reasonable to treat PCBs as passively transported on particu lates , but It Is not known whether this would be valid for the ocean or for clear lakes. However, It la questionable whether conditions in aquaria without sediments provide good models for natural environ ments. . 0073009 0 WATER PCB-00043090 . 2ZJ XV.1.6. Fate Theoretically, than are three ultimata fatas for FCBs ralaasad Into the environment: metabolic degradation, photolytic degradation, and deposition In aedlmanta In lakes or tha deep ocean (2). All three processes era known to occur, but their relative importance In the natural environment remains to be determined (2, 56, 331). IV.2. B1n-eeetmilatlon and Blo-magnlflcatlon IV.2.1. Mechanisms of uptake and accumulation One of the most Important environmental properties of FCBs Is their tendency to be "blo-accumulated" by aquatic organisms -i.e., to be concentrated Into their tissues to levels much higher than those in tha ambient water (1-3). This property results from the high solubility of FCBs In lipids and their low solubility In water: Hetcalf at el. (45) have shown that tha degree of bio-accumu lation of e chemical In aquatic animals Is closely related to Its partition coefficient between organic solvents and water (Figure XI. 6.6). There Is a further tendency for FCBs to be concentrated Into the tissues of anlsmls to levels higher than those In their food (1-3): this phenomenon Is sometimes referred to as "blo-magnificatlon". Within organisms, FCBs are further concentrated Into certain organs, especially the fat (1-3). The kinetics of organochlorIna 'compounds In animals have recently bean reviewed by Moriarty (335). It Is necessary to con sider blo-accumulation as a multi-stage process: the chemicals are v # taken Into the organism via food, water, or air, circulated through 0073010 WATER PCB-00043091 28k the organism la the blood, transferred Into and out of various organa, and finally metabolized and/or excretad. Each stage of transfar la a dynamic process whose rate depends on the concentrations of the chemicals In the various media and body organa. Under conditions of constant exposure, an organism may eventually reach a quasl-equlllb- rlum. In which the concentrations of the chemical.In the tissues are constant. It la then possible to define a storage factor (ppm In organism/ppm In ambient medium). In principle, storage factors should be defined separately for each organ In the body, but In practice. It Is cannon to define a single "blo-accumulation factor" (average ppm In whole body/ppm In ambient water). However, where the time re quired to reach equilibrium la long, as It often la for PCBa, the organism's physiological state may change and a true equilibrium . may never be reached (335). ' Aquatic plants taka up PCBs directly from water, and their - blo-accumulation factors reflect partitioning across the cell' mem branes. Animals may be exposed to PCBs In food. In air, or In water. Aquatic Invertebrates and fish have to process so much water In order to breathe that they usually reach quasi-equllibrlum with PCB con centrations In water fairly quickly; Intake via food Is usually only a minor route of Intake (336). Accordingly the storage factor (ppm In whole body/ppm In asiblsnt water) Is the most useful measure of blo-accumulation. However, for air-breathing animals such as birds and mammals the food Is usually the most Important route of Intake and the ^storage factor (ppm In whole body/ppm In food) la 0073011 WATER PCB-00043092 Iff the most useful measure: this factor Is sometimes referred to as tha "blo-magniflcatlon factor1' (336). s. : Laboratory experiments on blo-accumulatlon and bio-magnifica tion are usually highly simplified. Involving constant exposure via one route only. The situation in the real world Is more complex, since animals are exposed simultaneously via several routes, levels of exposure fluctuate greatly In tliae and space, end the animals' physiological state changes In response to aging and periodic en vironmental stresses. Accordingly, while laboratory experiments provide useful Information on mechanisms of uptake and excretion, and comparative data for the various components of PCBs, they do not necessarily provide quantitative measures of the degree of blo- accumulatlon and blo-magniflcatlon to be expected in the field. Measurements on wild animals in the field are needed to gauge the extent of blo-accumulatlon and bio-magnificatIon, and to specify their variability' in time and space. ' IV.2.2. Bio-accumulation In Aquatic Invertebrates In laboratory experiments, Sanders and Chandler (98) found that uptake and blo-accumulatlon of Aroc lor 1254 by some aquatic Inverte brates was very rapid. Blo-accissulatlon factors for various organisms ranged from 2,800 in stone-flies to at least 47,000 In Daphnla barns (Table IV.2.1). In scud (gssmarus psaudollmnaeus). the trl- and tetra-CBs in Aroclor 1254 were differentially accumulated by factors 2-8 times higher than the hexa- and hspta-CBs (Table IT.2.2). Hebeker and Pugllsl (93) found blo-magniflcatlon factors of 16,000-36,000 for Ofi73012 WATER PCB-00043093 i1 i Table IV.2.1 (fro* ref. 98) '' '(T llol^tgal magnification of toocjgy HS4 by agnatic tiWerteferates Organian * I, 1 ,1.1. * Organian - . Orsehlsar ' Watar ' concentration v . concentration (4 day exposure) ppb x SS4' ppn x SB . 1-day 3/ Magnification factor 4-day 7-day 14-day. 21-day QtpbU Daohnia Mtnt . ' 60 . 1.1 * 0.2 .52 2.0 28,700 .47,000 _. ' .* fhantoa nidge * ' - Chaoborua poaetlNunli s . . 1.3 2 0.1 30 1.8 22,000 23,000 23,800 .* w *- Scud . CoMurui DScudollMuee ` ... 1,6 2 0.1 ' 39 3.0 17,0001^24,000 28,000 27,500 27,000 KoflqnltP larvae Culex tarsalle. . 10 1.3 * 0.? ... 27 2.0- 12,*oo 18,000 20,000 ' ' . _ _ ' Glaaa ebrlap . Palaenonataa kadiakenala s' . l.S 2 0.1 . Stonafly Pteronarcrt deraaca s ' 2.8 0,8 Dobaonfly. . Corrdalua eornutua . . " '. ; *' 1,1 0.1 Crayflab . . Orconectea nale ' ::; ' * [ ; i.2 o.i . 1* 2.6 V 10,300 t7.0 0.30 2,100 5.1 0.23 1,400 0.2 0.20 370 12,300 13,700 14,200 2,500 2,800 2,900 4,800 1,700 5,700 8,800 f 3,400 4,500 18,600 2,800 6,800 3,100 ^Sanplea were taken tn triplicate.. j " ~ . ~~~ ' . ^Staple. w.r. taken In triplicate and aaproaawi M Mm Tlue 2 .tender* trwr (.03). ; -Concentration in otiaaltn/concentretlon tn water. ' . .. % WATER PCB-00043094 2X, Table IV.2.2 (from ref. 98) Changes la isomer ratios of Aroelor 1254 residues In scud * %* Peak Ko. Tlme./ 2/ Bo. Cl'*r Concentration Pactona' . Aroelor^ Sample . Average 1254 12 R.P.A. \ 2 3 _- ** ' ITS . ` . '-'*6 - 7 '8 '" 9 ' , 10 n: . 12 .. . 13 U' is : 16 - 17 0.27 0.37 0.45 0.50 0.63 0.77 0.98 1.07 1.19 1.31 1.59 1*67 1.79 2.0* 1.61 3.22 3.68' 3 2.00 4 ' * . 1.69 4 1.81 4 1.80 5 1.31 . s: 1.00 5 1.13 ' 5 : 1.07 6 ,, 0.33 .6 0.84 6 0.55 6 0.80 ' 7 . 0.43 6 0.56 6 0.50 .7 0.25 ` 7 0.25 2.00 1.72 1.81 2.00 1.31 1.00 1.16 1.16 0.66 0.85 0.55 0.80 0.57 0.60 0.62 0.50 6.25 2.00 1.71 1.81 1.90 1.31 1.00 1.15 1.12 0.49 0.85 0.55 0.80 0.50 0.58 .0.56 0.37 0.25 0.03 0.29 0.11 0.05 0.51 1.00 0.38 0.57 0.03 0.65 0.33 0.30 0.07 0.45 0.08 0.04 0.04 . --'Retention ties relative to ,* DDE on CLC^coluan of 2 an i.d. x .. 1.8 m 0.3X v/w OV-7 on 80-100 mesh Coming 110 glass heads; 15. nl/min. flow of nitrogen carrier gas; colon operated at 155 C. --2/Number of chlorine atoms: `substituted on biphenyl ring. de. termined . by gas chroaatography-mass spectrometry (9). - --^Concentration factor s Relative Peak area of sample Relative Peak areas as Arodo* 1254. ' - 00-.*^ WATER PCB-00043095 2S? the ssm peciai of scud whan axposad Co Aroclor 1242, and of 28,000- 1(^8,000 whan exposed to Aroclor 1248. In aatuarina shrimps, blo-accuom- latlon factors rangad up to 26,000 during chronic axpoaura to Aroclor 1254 (104: Tabla 17.2.3). In short-term taata, shrimp accumulated Aroclor 1016 to levels 2-4,000 times higher than those In tha ambient water within 96 hours (100). Bio-accumulation factors reported In oysters ranged up to 101,000 (108) and to 165,000 (337). In a comparative study In a model ecosystem, Hetcalf at al. (45) found higher' blo-accumulatlon factors for representative tetra- and penta-CBs than for a representative trl-CB (Tables E.4 and E.5). In snails the blo-accumulatlon factors (E.M. In Table E.5) recorded after 33 days were 5,800 for the trl-CB, 39,400 for the .tetra-CB, and 59,600 for the penta-CB. In mosquito larvae the corresponding factors ware 815, 10,600, and 17,300, but these represented only 6 days' exposure. It should be noted, however, that som of the unidentified non-polar metabolites were blo-accumulated even more than the parent CBa (Table E.4). "Unknown I" from the trl-CB, for example, was bio-accumulated 12,000 times In the snail and 8,000 times In the mosquito. . IV.2.3. Blo-accumulatlon In Tlsh . In laboratory experiments In flowing sea water, juvenile plnflsh (Lagodon rhomboldes) and spot (Leloetomua xantharua), both accumulated Increasing amounts of Aroclor 1254, up to quasi-equilib rium levels, as duration of exposure Increased (111). Spot exposed to 1 ppb Arocior *1254 for 56 days rMched maximum residue levels of 0073015 WATER PCB-00043096 Table IV.2.3 ' / Length of Exposure (hr/ days) 1 >2 3 4 8 12 16 24 / 1 . . 36 / 1.5. 48/2 72/3 96/4 154 / 6.5 336 / 14 504/21 672 / 28 840 /35 1176 / 49 1512 / 63 ACCDMOLATION OF AROCLOR 1254 IN Palaanonetea puglo WITH TIME AFTER EXPOSURES TO THE CHEMICAL IN HATER AT THREE CONCENTRATIONS (pg/l) (Each value represents a.composite sample of 10 anlaals) ' ' Control . Body Cone. Cone. Factor (g/kg) i 0.04 . Body Cone. Cone. Factor (g/kg) . 0 09 Body Cone. . Cone. Factor ., (g/kg) /V 0 62 Body Cone. Cone. Factor (g/kg) 0.1 -- -- -- _ .-- '. -- ft \ * .* ---- . --. --` --- ' --. . * 0.1 * ----- . * 0.1 0.14 0.10 * 0.1 r 0.15 ' * 0.1 * ' 0.1 * 0.1 * 0.1 . * 0.1 * 6.1 * 0,1 * 0.1 * 0.1 * 0.1 * 0.1 * . 0.1 * 0.1 * 0.1 1590 . 0.13 , 3250 ' 0.15 3750 0.17 4250 0.21 5250 PCB - STOPPED 0.1 * 0.1 * 0:1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.10 0.14 Q.15 0.33 0.43 0.45 1.J7 0.75 0.12 0.13 * 0.1 0.1 ,* * 0.1 0.1 *, 0.1 0.12 . ' 0.14 0.26 0.20 0.20 1100 0.37 1560 . 0.58 3670 0.40 3670' 1.28 4780 7/40 5000 6.67 17400 10.82 8330 16.48 * ' 3.24 * * 1.64 - 190 230 420 320 470 600' 930. 650 2060 11930 10900 17450 26580 . 1 * * * Magnification fad or not calculated. ! rCJO WATER PCB-00043097 37,000 tins* the water concentration In 14-28 days (Figure XV.2.1), whereas plnflsh expo*ad to Aroclor 1254 accumulated It to 22,000 times the water concentration (111). Thereafter, the concentration of PCB (ppm In tissues) remained roughly constant while the total amount continued to Increase as the fish grew. After placing the fish in clean water the PCBs were gradually eliminated, falling by half in about 4 weeks (Figure IV.2.1). In a parallel experiment with plnflsh exposed to Aroclor 1016, a bio-accumulation factor of about 20,000 was measured after 42 days (Table IV.2.4, Figure IV.2.1, ref. 100). Thus the bio-accumulation of Aroclor* 1016 and 1254 was similar in this fish. However, there was some differential uptake of the com ponents of Aroclor 1016, since the peaks 1-3 In the chromatogram (mono- and dl-CBs) were proportionately reduced In the PCBs stored In the fish tissue, whereas the'tri- and tetra-CBs were proportionate ly Increased In the fish tissues (Figure IV.2.2, and Table XV.2.5; cf. Figure II.3.3). DeFoe t al. (113) found that the bio-accumulation factor for adult fathead minnows at 25C Is approximately 120,000 for Aroclor 1248 and 270,000 for Aroclor 1260. Female fechead minnows accumu lated about twice as much PCBs as the males but the difference Is largely due to the greater amount of lipids in the fmnales. The residues of PCBs In the fish tissues were directly proportional to the water concentration (Figure IV.2.3). The storage factor for fish lipids (ppm in lipids/ppm In water) was between 1 and 2 million (Ibid.). Babaker et al. (112) found that the blo-accumulatlon factor 0073017 WATER PCB-00043098 ..... _ 271 BIOACCUMULATION AND DEPURATION ( OF PCB's BY FISHES CONCENTRATION IN WHOLE FISH, n g /g w et w eight Figure IV.2.1 (fvrsm reef 0073018 WATER PCB-00043099 Table IV.2.4 (from ref. 100) Toxicity and Uptake or Amoclon 1016 nr Pisnan (Lagodon rbemboidn) Exposed ' pom 42 Days in There Hcr.vtt.vTE Experiments Tost concentration (pg/liter) Nominal Measured Mortality <*. Concentration m fish (ag-g, wet,weight) Flesh Flesh and akin Whole fish Control 0.1 1.0 10.0 Control 1.0 3.2 10.0 32.0 Control 10.0 32 O' 100 0* NDND 0.8 3.0 ND 0.0 25 7.0 13 ND 68 21 50 ` 36 16 48 38 12 16 16 2S 44* 6 6 50* 30* ND 0.7 8.1 60 0.5 4.0 34 03 140 ND 23 30 38 ND 0.8 6.3 00 * 0.6 6.D 30 76 180 NI) 40 48 72 ND 2.4 ll 166 0.5 *17 65 t70 620 ND 111 106 205 * ND, nob detectable: <0.2 liter in water: <0.2 eg/g in ti*me. * Mortality significantly greater than in control fish, a - 0.01. * Exposure terminated and twsueM analysed when 50** of the fish died: 33 days at 32 jig/liter end 18 days at 100 a*,'liter. 00Ol<> WATER PCB-00043100 y Figure IV.2.2 (from ref. 100) 4 vti Chromatogram of 0 peaha of Arodor 1016 reference standard. OperatfngTOo* dittoes; gas flow nitrogen 25 ml/min; infection and detector temperature 210*C; oven tem perature 190*C; *H electron ooptme detector; 152.4 X 0.32 cm gta* column pecked with 2S OV-101 on 100-120 Cas Chrom Q. - Table IV.2.5 (from ref. 100) PanoaMTaoaa or ten 9 Measpesd Puu viom CnouAToeaasta or Anocio* 1016 Bvsumci Stooain amd rtoM Ttsstncs or Putmx Exposed to 1 eo/umm or Anocuom 1016 roa 56 Days amd tsxm Hsld n PCB-rar* Wats* roa 06 Days ^ Fereentage of peak* Chro- mato- 1 2 3 4 5 6 T 8 9 grams Befnrence standard All times through 14 days of exposure AH times, days 21-66 of exposure \ All tissues during 56 day depuration 12.4 11.3 10.0 28.5 13.3 6.4 8.1 2.4 5.3 1.6 80.6 7.7 10.3 12.1 2.5 5.0 2.1 38.2 7.8 11.9 11.3 1.0 1.7 0.3 36.2 2.3 15.0 12.4 8.8 9.4 10.8 11.9 6.2 11.4 15.9 16.7 8 7 12 7 D^nni^dM ***&' Xl00. 0073020 WATER PCB-00043101 LIPID CONCENTRATION (pG/CK) "HZ Source: Defoe, D.L., G.D. Veith, ami R.W. Carlson, 1975. Effects of Aroclor 1248 and 1260 on the fathead minnow (Plnephales promelaa), tJ.S. 1 Environmental Protection Agency, National Hater Quality Laboratory, . | Duluth, Minnesota. - j 1' ' Ir _ * __ '. . j ____ !_JFigure XV.2.3 (ref. 113) , __ Aroclor 1248 residue in fathead minnows, normalised to lipid . I . concentration In fish* : \ ' : ; I 1 ' WATER PCB-00043102 for Aroclor 1254 In fathead minnows was batwaan 110,000 and 240,000, and that for Aroclor 1242 was batwaan 32,000 and 274,000. Although the results varied somewhat between replications of these experiments there is no evidence that the bio-accumulation factors in this fish increase significantly with the degree of chlorination of the mixture. Velth and Klwus (338) conducted an experiment with fathead minnows exposed to Aroclor 1016 and found a bio-accumulation factor of about 50,000, but the concentrations in the fish were still rising after 32 days (Figure XV.2.4). This result for Aroclor 1016 falls well within the range observed for the seme species for Aroclor 1242, and within a factor of 3 of that observed for Aroclors 1254 and 1248 (112, 113). . Bio-accumulation factors for bluegllls exposed to Aroclors 1248 and 1254 were in the range 26,300 to 71,400. No major modifica tion .of the PCB isomer ratios were observed in the tissue residues, and no new compounds ware Identified (95). Several studies have shown that fish also store PCBs when fed contaminated food while kept in clean water. In these cases the bio signification factors were quite modest, because the fish ware able to exchange PCBs across their gills with the ambient water; in natural environments both water and food would contain PCBa and fish would take in PCBs via both routes. Coho salmon (Oncorhvnchus klsutch) fed Aroclor 1254-for-240-------------------------- days at dietary concentrations of 0.4 to 580 ppm (14.5 to 14,500 v . ^ig/kg body weight per day) accumulated whole body residues which ' 007*0iZ WATER PCB-00043103 Figure IV .2.4 ( r e f . 338) Hhh 33 2H 0O**Oj u^ tSJ W **., ..... ,.'> V ..V.-S--5-. -I WATER PCB-00043104 wera 0.9 to 0.5 times the dietary levels (95). The highest residue val^ie was 300 ppm. Accumulation end retention of Aroclor 1254 by rainbow trout (Salmo galrdnerll) from the diet followed the same pattern as that iM for coho salmon (339). The relative concentration of PCB In the lipid fraction of trout Increased rapidly during 8 weeks' exposure to 15 ppm Aroclor 1254 In the food, then tended to equilibrate at about 95 ppm (Figure IV.2.5). While the relative aatount In the fish reached quasl-equlllbrlum, the absolute quantities continued to In crease as the trout grew (Figure IT.2.6). The distribution of PCB among the tissues was dependent on the lipid content of the various tissues (Table IV.2.6). The uptake of pure PCB Isomers by fish has been studied In three experiments. Gruger et al. (121) fed 3,4,3',4' tatra-CB and two hexa-CBs to juvenile coho salmon at a dietary concentration of 3.3 ppm each (total 10 ppm) for 165 days. The material fed during the first 24 days was retained well in the tissues, but tissue con centrations tended to level off and aftar 28 days the concentration of the tetra-CB was only about half that of either of the two haxa-----------CBs (Tables IV.2.7, IV.2.8). The experiment Is complicated by the fact that the treated flah lost weight and depleted their lipid reserves, and that equlllbrlisn concentrations in the tissues had not been peached (Table XV.2.7). As In the cate of the trout, most of the PCBs were concentrated In the.llpld-rlch tissues, end the average concentration f PCBs In lipids was similar (about 70 ppm) in all 0073024 WATER PCB-00043105 '.... ( ,y ; '' _ ___ ; __ i __ ___ __ 1 i Source: Lieb, A.J., D.D. Bills and R.O. Slnnhuber, 1974. Accumulation j of dietary polychlorinated biphenyls (Aroclor 1254) by rainbow : trout (Salmo galrdneri), Journal of Agricultural and Food . Chemistry. 22(4): 638-642. 0073025 Figure XV.2.5 (ref. 359) PCB in lipid fraction of rainbow trout after exposure to 15 ppm Aroclor 1254. " 6-76 WATER PCB-00043106 \ .I 1 i?n h I \ Source: \" K. Lleb, A.J., D.fi. Bills, and R.O. Sinnhuber, 1974. Accumulation of dietary polychlorinated biphenyls (Arodor 1254) by rainbow' trout (Salmo galrdnerl), Journal of Agricultural and Food Chemistry. 22(4): 638-642. .. i . Figure IV.4.6 (ref. 339) Total amount of PCB per fish: (A) fish on diet Contain,lng 15 ppm of PCBj (B) fish removed from diet containing 15 ppm of . . PCB at end of 16 weeks. 00l*oib I WATER PCB-00043107 1- :-( " : Cf TY'lE-- . " ' viable IV.2.6 ......... I ' DISTRIBUTION OF AROCLOR 1254 IN TISSUE OF RAINBOW TROUT ' 'j 1 ' Tissue -: X lipid in tissue Copen of PCB in lipid. ppo Concn . of PCB In tissue ppd Visceral adipose Gill . Muscle Stomach Liver Whole fish 92.8 9.7 2.7 6.5 3.5 8.5 ; .' ' 1U 113 104 104 57 96 . 103 11,3 2.8 6.8 2.3 8.2 ' ' - Source: Lieb, A.J. , D.D. Bills, and R.D. Slnnhuber, 1974 Accumulation of dietary polychlorinated biphenyls (Aroclor 1254) by. rainbow trout (Salmo gairdneri) Journal of Agricultural and Food Chemistry 22(4): 638-642. '. ( .; ____ _____ _____________ ... 3o i I i f il l i ( TE.83 _ 0073027 WATER PCB-00043108 Table XV.2.7 (ref. 121) Group Days , ANALYSES OF WHOLE JUVENILE COHO SALMON WHEN FED FOR 24, 53, AND 108 DATS1 .:vr ' ! ; i' l Body wt> g 3,4,3',4', . 2,4,5,2\4\5' 2,4,6,2',4\6* Lipid content. chloroblphenyl chloroblphenyl chloroblphenyl Total ; VtZ Wet tissue^ Kg/g Jest 24 53 108 Control 24 53 108 4.24 5.26 3.0 3.79 5.42 7.87 - ; 7.60 6.48 3.8 5.83 . 7.38 6.8 0.49 (472) 0.59 (30Z) 0.65 (162) 0.059 0.044 0.107 0.73 (702) 0.93 (472) 1.41 . (352) 0.090 0.047 0.023 0.66 1.88 (632) (602) 0.90 2.42 (452) (412) 1.42 3.48 (352) (292) 0.050 0.199 0.053 0.144 . 0.030 0.160 1. Three flah per group. 2. Percent of chloroblphenyl aaount found in fish relative to aaount fed<la given In parentheses. { Source: : : . . ': Gruger, B.H., Jr., N.L. Karrlck, A.I. Davidson and T. Hruby. 1975. Accumulation of ' 3,4,3',4'-tetrochlorobiphenyl and 2,4,5,2',4',5'-and 2,4,6,2',4,,6'-hexachloroblphenyl ( In juvenile coho saloon. Envlronnental Science and Technology 9(2): 121-127. .! . ' .. .' . *1 l i ezottoo WATER PCB-00043109 Table IV.2.8 (ref. 121 ) COHCENTRATIONS OF CHLOR0BIPHENTLS AND LIPID CONTENT IN TISSUES OF JUVENILE COHO SAIMON FROM TEST AND CONTROL GROUPS Tissue specimens st 117 days Lipid content. vtl Test group iiraln . 7.1 Xlver . 2.9 White suscle 2.6 Intestines 4.3 Stoaach and pyloric caneca . 3.3 Spinal column 6.6 Heart . - Lateral line suscle Spleen > 6.1 . . . - .. Adipose 73.0 Control group - Brain . 6.6 ' Liver ' 3.9 White suscle 5.5 Intestines , 6.6 Stomach and pyloric caneca 4.2 Spinal column Heart 23.7 - Lateral line suscle Spleen , 11.4 Adipose ' 73.5 ' 3,4,3',4' chlorobiphenyl 0.15 0.25 0.29 . 0.30 0.43 0.92 0.98 0.77 . 0>85 10.6 ' 0.020 . 0.034 0.028 01098 0.032 . 0.096 0.85 0.10 (2.0). 0.30 2,4,5,2\4\5* 2,4,6,2*,4',6' 1 chlorobiphenyl chlorobiphenyl Total 1 Wat tissue, jig/g i 1 0.31 0.35 0.63 0.79 0.76 . 1.4 1.2 1.8 2.0 19.3 0.38 0.84 0.50 1.1 0.64 1.6 0.82 1.9 0.95 2.i j . 1.4 3.7 1.5 3.7 1 1.9 4.5 ] ' 2.1 . 4.9 18.8 48.7 ; 0.009 0.014 , 0.015 0.032 0.014 0.026 . 0;068 0.058 0.13 0.19 0.017 0.037 0.025 . 0.069 0.022 0.032 0.18 0.077 0.12 " 0.36 0.046 0.085 0.068 0.20 0.069 0.15 1.1 0.24' (2.3) 0.85 l I I at 165 days - Test Group White suscle ' Brain 0.086 0.26 0.20 0.37 0.22 0.56 . 0.51 1.2 0073029 (a) 0 WATER PCB-00043110 2*2 the tissue* tested after 117 days (Table IV.2.8). After 48 days' starvation the concentrations of PCBs in the tissues were depleted .V. . somewhat (to a total of 33 ppm in adipose), but those In the brain were Increased slightly (Table IV.2.8). The tatra-CB was excreted only slightly faster than the hexa-CBs (121). Sanborn at al. (307) exposed green sunfish to three represen tative chloroblphenyls In two pulsed exposures (24 hours' exposure to 1 or 3 ppb 9 days apart). Although the tri-CB was rapidly excreted and was almost eliminated after IS days, the penta-CB was well re tained and there was no evidence of elimination in the 15-day period of study (Figure IV. 2.7). The tetra-CB was taken up and retained slightly less well than the penta-CB, but the concentrations in the fish at the end of the 15-day experiment were only 2-3 times smaller than those of the penta-CB, despite the limited exposure time (Figure IV.2.7). Thus the smjor difference in uptake and retention appeared to be between the tri- and tetra-CBs. Metcalf at al. (45) exposed mosquito fish to the same three chloroblphenyls under model ecosystem conditions (exposure via both food and water). Although the exposure to the fish was for only 3 days, they nevertheless accumulated the tri-CB to 6,400 times the water concentration, and the tetra- and penta-CBs to about 12,000 tlmas the water concentration (Table B.5). Thuk under these circum stances there was no difference in bio-accumulation batwean the tetra- and penta-CBs. To susnarise, under laboratory conditions fish blo-accumulate A. # PCBs to concantratlons 30,000 to 300,000 times higher than thosa In 00^03 WATER PCB-00043111 3^ tMTMl ' Unborni J.L, V.P. ChlMoro, on* E.L. Kotcolf, 1973. Dptoho f throo polychlorinated biphonylo, DDT, on* DDE by tho (toon ouofloh, Lononto eronolluo. EAT., bulletin of Enylronnnntnl Contamination and Toxlcolorr. 13(2)1*209-217. Dptoks it trl-, totro-, and pontaohloroblphoayl by tho groon oanfinh. Figure IV.2.7 0f303l WATER PCB-00043112 the ambient water (up to 2 million times higher In lipids). There Is evidence that mono-, dl-, end trl-CBs are poorly retained In fish tissues, but the differences In retention between tetra- and higher CBs are small, no more than a factor of 2-3 In. the circumstances of these experiments. IV.2.4. Blo-accumulatlon In Model Aquatic Ecosystems , Two studies have been published of the behavior Of PCBs in model aquatic ecosystems, designed to simulate the transport of PCBs from water through simple food chains. In one study by Metcalf et al. (45), radlolabelled chloroblphenyls were applied to a mixed terrestrial-aquatic system and subsequently traced In water and various aquatic organisms. The results have already been sumarlzed In the previous sections and In Appendix E (Tables S.3-E.5). The study showed how PCBs and their metabolites move from the terrestrial to the aquatic environsiant and are taken up there by aquatic organisms. Including plants, Insects, fish, and snails. Because the fish were only Introduced Into tha syatem for the laat few days of the experi ment, the results do not indicate the full potential for blo-accumulation In food-webs: the figures of 6,000-60,000 for blo-accumulatlon factora in snails (Table E.5) give the best awaaure of this potential.- While Metcalf's system Is primarily designed to Investigate storage and blodegradablllty (45), a system designed by SUdergren was designed to trace the movement of PCBs through a simple food-web (340). Plgure IV.2.8 shows the basic design of the system and v Plgure IV.2.9 shows the results of a test with Clophen A50. Most of '0073032 WATER PCB-00043113 3a< y. Figure IV.2.8 (from ref. 340) i Principle design of ' tin model aquatic eco- j system. The or|anisms not j mentioned in the text con* j stituie alternative food* J chains viable in the systeaJ but not included in this 1 study. I , 0073033 WATER PCB-00043114 .y. lMil fMM* Dlttribuiion and budget of Clophen A 50 within the model aquatic ecosystem (tig), n.d. * no substance detected. Figure IT.2.9 (from ref. 340) 0073034 WATER PCB-00043115 3oS the transport of the material to the first consumer organism (the cyprlnld fish Leucasolus dellneatus) took place via the alga Chlorella . . pvrenoldosa. There were some changes In the relative proportions of the components of the Clophen mixture as they passed through the system (Figure IV.2.10). However, these changes were very minor ex cept In the two predatory fish (the pike Beox Indus and the perch Perea fluvlatllls). Even the first component In the Clophen mixture (2,5,2',5'-tetraCB) was passed essentially unchanged through the alga to the first consumer (Figure IT.2.10). IV.2.5. Storage and Blo-magnlflcatlon In Birds Storage and blo-magnlflcatlon In birds are of Interest for two reasons: (a), to Indicate the range of residues likely to be in gested by humans eating wild or domestic birds; (b) to Indicate the degree to which passage of residues up the food chain can lead to toxic effects In the birds themselves or In their predators. . In a study with chickens (149, 150), laying females fed Aroclor 1248 at 0.5 ppm accumulated 3.1 ppm In their adipose tissues and 0.22 ppm In their eggs after 8 weeks; correspondingly higher levels In tissues and eggs were found at higher feeding levels (Table IV.2.9). Similar results were obtained with Aroclor 1242: hens fed at 5 ppm laid eggs containing 1.7 ppm after 6 weeks (152). However, laying hens may be a poor model for other birds because their contin uous egg-laying provides them with an Important route of excretion (341) not evallable to wild birds which lay few eggs. Pheasants given 12.5 mg Aroclor 1254 weekly (equivelent to about 50 ppm In the "0073035 WATER PCB-00043116 iwctvs 307 Qit chromatograms obtained from various com a. ponents in a model aquatic ecosystem when testinf Clophen A 50. Figure XV.2.10 (from ref. 340) _ o0l33 WATER PCB-00043117 Table IV.2.9 (from ref. 150) --Transfer of PCBs' from the toying diet to the adlpote tissues and eggs of hens Diet PCBs PCBs in adipose tissues no. in diet 1 wk. 4 wks. Swks. 1 (Diet A) 3 4 5 PP-m. 0 V 0.5 1.54 i.O 2.21 10.0 10.4 20.0 t6.3 p.g.m. 2.16 4.41 22.7 54.7 pr3.10 6.62 37.1 82.7 ' Aroclor 1248, Monsanto. Inc., St. Louis, MoT 1 wlc. Tm- 0.10 0.19 1.05 2.19 PCBs in eggs 4 wks. 6 wks. 7 wks. p.p.m. 0 pr 0.16 0.21 0.20 0.33 0.42 0.45 2.21 2.83 3.11 4.51 5.37 5.72 8 wks. p.p.m. 0 0.22 0.41 3.06 7.04 0073037 WATER PCB-00043118 3i) diet) for only 16 weeks accumulated an average of 24 ppm In their whole bodies; this did not decline significantly even after 6 months on a clean diet (141). Mallards fed 25 ppm Aroclor 1254 laid eggs containing 33-56 ppm PCBs (wet weight) (131). Ring doves fad 10 ppm Aroclor 1254 In the diet accumulated 8 ppm In the muscle, 5.5 ppm in the brain, 15.3 ppm In the liver, 736 ppm In the fat, and 4.8 ppm In the eggs (342, 161-163); after starvation the PCBs were mobilized from the fat and the birds died with 293 ppm In the brain (Table IV.2.10). American Kestrels fed 10 ppm Aroclor 1254 In the diet laid eggs with average residues of 225 ppm (dry weight, corresponding to about 30 ppm, wet weight) (164). A number of workers studying the storage of PCBs In birds have reported that the lower chlorobiphenyls In Aroclor 1254 (tatra-CBs) are retained less well than the higher components (141, 131, 342, etc.). Call at el. (343) have provided quantitative data showing that tetraand penta-CBs are stored at relatively lower levels and hexa- and hepta-CBs at relatively higher levels In Japanese quail than In the Aroclor 1260 with which they were fed. N However, the differential storage of PCBs In birds Is not re lated simply to the degree of chlorination. DeFreitas And Norstrom (27), studying the elisiinatlon of components of Aroclor 1254 from pigeons, found that the components which were excreted and/or metab olized most rapidly were those with 2,3,- 3,4-, or 2,3,6- substitutions In at least one ring. Components with 2,4,5-, 2,3,4-, and 2,3,4,5substltutlbn patterns were not eliminated. Components with 2,5- 0073038 WATER PCB-00043119 Table IV.2.10 (from raf. 161) Organ lavda, ppm* w*t weight hnii. Muaele Brain Liver Fre-etrees (n-4) Poet-stress (n-5) 8.1*2.3 (4.1--9.9) 172.0*38.3 (120.0-227.0) 5.5*1.ft (4.8-8.0) . 293.0*27.0 (254.0-340.2) 15.3*12.0 (3.0-27.5) 1118*207 (937.3-1088.0) Figures art means, standard deviations, and range. Ne fat present. Fat 730.1*211.0 (481.2-1009.4) mm* . 0073039 WATER PCB-00043120 Table IV.2.11 (from rtf, 344) Mna Total PCS Raaiduas t'p.p.m.), found *n Quail Tl**ue after leading 'Arodor 1242' and 'Arodor t 254* at 250pp.rn.for 20 Day* Tissue Liver Heart Brain - Omental fat 'Arodor 1242* I7.0Z6* 1.2 0.7 7.1 1.7 12217 .'Arodor 1254* 21.1 7.9 7.0 1.1 7.7* 2.3 204*31 . AS figures are mean of six birds t s.t. 0073040 WATER_PCB-00043121 3^ 36 Gas-liquid chromatographic "spectra", normalized to a common peak (stippled column) of 'Arocior 1242* standard and mean pigeon liver residues following feedirg at 900 p.p.m. in N diet for 28 days. ' Figure IV.2.11 (from ref. 344) 0073041 WATER PCB-00043122 Sis' substitutions were eliminated, but much more slowly than those with 2,3-, or 3,4- substitutions. Bush et al. (155) obtained similar results for differential elimination of components of Aroclor 1254 from laying hens, and suggested that 4,4' substitution Is more 1m- o portent than number of chorine atoms In determining the degree of persistence of chloroblphenyls In hens. This agrees with the find ings of Matthews and Anderson (328) In rats. " Finally, Bailey and Buoyan (344) found that although Aroclor 1242 was not accumulated so strongly from the diet by Japanese quail as Aroclor 1254, the difference was only by a factor of 2-2.5 (Table IV.2.11). Although the lower components In Aroclor 1242 (dl- and some trl-CBs) were not well retained In-pigeon tissues, some tri- and most tetra-CBs were well retained and were only slowly excreted over a 2-6 month period (Figure XV.2.11), Of the components well retained In the pigeon tissues, only one (probably 2,5,3',4'-tetra-CB) Is not present In Aroclor 1016 (Figure II.3.3).. Thus, although there Is substantial differentiation of the components of PCB mixtures as they are taken up, stored, and ellminatad by birds, some tetra- and even 'S trl-CBs are reasonably well retained In tissues and secreted Into eggs. Very little Information Is available on blo-magnification of PCBs by any flsh-aatlng birds, which are of Interest as being the species at greatest risk In Che environment. However, cormorants (Phalacrocorax carbo) dosed with an average of about 300 ppm Clophen A60 (about 25 mg/kg/day) accumulated 850-2,750 ppm PCB residues In 0071042 WATER PCB-00043123 their whole bodies in 55-125 days', including residues of 10-20,000 ppm in their fat (142). This indicates a somewhat higher degree of .V. blo-magniflcatlon than for any of the other birds listed above. 00/3043 WATER PCB-00043124 ?7 IV.2.6. Storage and Blo-magnlflcatlon In Manual* V The storage and kinetics of FCBs In manmals are of Interest (a) to Indicate the range of residues likely to be Ingested by humans eating domestic mammals; (b) as models for the storage and kinetics of PCBs In humans themselves. There Is an extensive literature on the uptake, .kinetics, and elimination of FCBs In manuals, but for this document primary Interest Is attached to the results of long term feeding at low levels. Absorption of PCBs from the gastrointestinal tract appears to be very efficient. Albro and Flshbeln .(345) fed nine Individual chloroblphenyl Isomers ranging from mono- to hexa-CBs to rats and found that 91-99Z of the dose was retained In the body. After absorp tion PCBs are. distributed throughout the body: their distribution tends to parallel the lipid content of the.tissues, so that PCB storage In the tissues occurs in the following order: fat liver feces 7 kidney z* brain y plasms urine (346) . The concentration-of PCBs In adipose tissue is 10-100 times the con centration found In other tissues, both early after Bingle doses (182) and after prolonged Intake (346). - ' Allen et el. (196) fed six adult female rhesus monkeys 23 ppm Of Aroclor 1248 for two months. PCB concentrations In samples of adipose tissue averaged 127 ^ig/g fat for all animals. At that time, the experimental diet was discontinued. Eight months later the PCB content was 34 /ig/g fat. PCBs were transferred across the placenta of one female whlph gave birth and concentrated at high levels In the fat and adrenals of the Infant (Table IV.2.12). In an Infant born WATER PCB-00043125 3(74 to a female 29 months following the discontinuation of PCBs, the levels within the adipose tissue were 3.38 ppm at 4 months of age (69). Female monkeys given 5.0 ppm PCB In their diets attained maxi mum levels of PCBs within thalr adipose tissue at 6 months (141 to ' 177 ppm adipose tissue). However, It required approximately 14 months on the 2.5 ppm diet for the monkeys to reach similar maximum PCB levels In their adipose tissue (126 to 144 ppm). Hales which re ceived 5.0 ppm PCBs attained levels ranging from 128 to 200 ^g par g adipose tissue at 14 months. Infants born of mothers exposed to 2.5 and 5.0 ppm PCBs within the diet contained concentrations of PCBs ranging from 1.0 to 4.8 ppm within the skin at birth. While nursing from mothers consuming PCB diets, the Infants continued to accumulate the compound. At 3 months the levels within the tissues ranged from 86 to 136 ppm. The concentration of PCBs within the milk ranged from 0.15 to 0.40 ppm. The tissues of the Infants which died while nursing PCB-fed mothers contained high levels of PCBs within the thymus, ovaries, brain, kidneys, adrenal glands and pancreas (20-48 ppm tissue). Lower levels were found in the liver, lymph nodes and bone marrow (8-16 ppm) (69). - 0073045 WATER PCB-00043126 Tablt IV.2.12 (from rnf. 196) POLYCHLORINATED BlPIIENVL CONTENT OF TISSUES OBTAINED from Mother and Intant Following Consumption of AROCLOR 1248 BY TUB MOTHER PRIOR TO PREGNANCY Organ Weight (x) PCB content Wg) Mother liver Fat Placenta Infant Fat Adreijats Liver Muscle Brain Kidney* Large intestine Small intestine. Stomach Lung Skin -- -- -- -- 0.2 11.9 -- . 52.9 2 -- -- -- 6-1 -- . . 56.3 50.0 0.9 27.70 24.40 0.01 0.98 029 0.10 0.08 0.52 OJJ 0.21 051 0013^ WATER PCB-00043127 317 Curley at el. (346) fed weanling Sherman rata a diet contain ing 100 ppm Aroclor 1254 for 58 days and for 240 days. Animals were sacrificed at various Intervals. A steady buildup of PCB In all tissues was observed over the 58 day period. The rats stored more PCB In their tissues efter 240 days than at the end of 58 days. In experiments at two dose levels (100 ppm and 500 ppm) for 240 days, Curley et al. (346) observed that PCB In fat after 240 days at diet ary levels of 100 and 500 ppm were 1,101 and 10,021 ppm, respectively. They did not reach a point of equilibrium storage. Grant at al. (347) fed male rats a diet containing either 0, 2, 20 or 100 ppm of Aroclor 1254 for 246 days. The residues In the blood, brain, heart, liver end fat were-found to be dose related. All components of Residues were not metabolized at the same rate. Burse et al. (208) fed rats diets containing 100 ppm of Aro clor 1242 or Aroclor 1016 for up to ten months. Rats were sacrificed at Intervals and tissues analyzed. A roughly steady state concentra tion of PCBs In adipose tissue was approached In two months and reached In 4-8 months (115 ppm of Aroclor 1242 and 214 ppm of Aroclor 1016) (Figure IV.2.12). Equilibrium levels of Aroclor 1016 end 1242 seem to be reached sooner than equlllbrlun levels of Aroclor 1254 and the levels of Aroclors 1016 and 1242 ware 5-10 times smaller than those reached after continuous feeding with Aroclor 1254 (208, 346).. This Indicates considerably less blo-magnlflcation of trl- and tetra- CBs than of penta- and hexa-CBs In rats. After the'rats were placed on a PCB-free diet the residues of Aroclor 1016 were eliminated some- v , . what faster than those of Aroclor 1242 (Figure IV.2.12). Comparison . ooi50*'1 WATER PCB-00043128 -Residues resulting from a con tinuous diet of a 100 ppm concentration of Arodor 1016 (a) or Arodor 1242 () for parted of tan month*. Residue* following a continuous diet for only aix months and re covery for ftys month* for ArocJor 1016 () and for ialx months for Arodor 1242 (o) ara Illustrated by tha broken Unas. Each point represents tha mean and stan dard error (D of four rats. Figure IV.2.12 (from ref. 208) ool ^ WATER PCB-00043129 32-1 of chromatograms (Figures IV.2.13, IV.2.14) shows that only the later eluting components (penta-, tetra-, and some trl-CBs) of .V ' Aroclors 1242 and 1016 were stored In the adipose tissue. The lower components were largely metabolized, as hydroxylated metabolites were found In the urine (348). Storage and dynamics of FCBs have also bean studied In domes tic animals. Hansen et al. (216) fed sows with 20 ppm Aroclor 1242 throughout gestation and nursing and found PCBs In many tissues of the sows and their offspring. PCB levels In the fat and blood of tha sows averaged 14 ppm and 0.25 ppm respectively (Table III.9.9). The offspring appeared to have somewhat lower concentrations of FCBs In their fat but hlgjher concentrations in the blood (Figure IV.2.15). As with other mammals, there was considerable differentiation of the components of the Aroclor 1242 mixture (Figure IV.2.15): later elut ing components (primarily penta- and tatra-CBs) were retained much better In the fat than the earlier eluting components. One-third to one-half of the FCBs stored In the fat of the sows consisted of com ponents that would not be present' In Aroclor 1016. Fries et al. (349) fed 200 mg of Aroclor 1254 dally (0.4 mg/kg/ day, equivalent to 12 ppm in the diet) to nine cows. Residues In the body fat and milk fat built up to 40 and 60 ppm, respectively during a 60-day feeding period, but did not reach equilibrium levels In that time (Figure IV.2.16). Following discontinuation of dosing, levels of FCBs in fat and milk fell off steadily, but those In body fat had fallen only to 30 ppm after 60 days. Thq cows excreted only 0073049 WATER PCB-00043130 322- .V. Figure IV.2.13 (from ref. 208) 0073050 WATER PCB-00043131 It3 S. Figure IV.2.14 (from ref. 208) 0073051 WATER PCB-00043132 D *t.ctor Rmpomm --Bectron capture g chromatographs of PCB mixture in stan dard and tfoue extracts from sows given PCS mixture in the feed and from their offspring. Back-fat samples were diluted 100-fold. Figure IV.2.15 (from ref. 208) 007**Z WATER PCB-00043133 !/ %-------- -------- ------------------------ -------- ft MTS Concentration of PCB in milk (at and body fat of cow* fad 200 mg of PCB par day. Each point It an average of nine cows standard deviation. The curve for body fat is extrapolat ed before 30 days. ' Figure IV.2.16 (from ref. 349) - Typical gas chromatograms of an Arooior 1254 stan dard. a mkk fat residue sample, and a body fat residue sample. Retention times of peaks 1. 2. 4. 6. and 9 were equal to the re tention times of the pure compounds 2.5,2',5'-tetrachlorobi- phenyi, .2,3.2\5'-tetrachiorobiphenyl, 2.5,3',4'-tetrachlorobi- phenyl, 2,3.4.2',5'-pentachJorobtphenyl. and 2,4.5.2',4'.5%hexe- chiorobiphenyl, respectively. . Figure IV.2.17 (from ref. 349) 0073053 WATER PCB-00043134 20-251 of the dally dose la their milk, Implying that even under con ditions of continuous milk production they were still far from equill- .V ' . ' brim after 60 days. There was differential accumulation of the higher components of the mixture In both body fat and milk (Figure XV.2.17), but peaks 2 and 3 (both tetra-CBs) were as well represented In the milk fat as In the original Aroclor mixture. Jan et al. (404) found that cows excreted substantial amounts of 3,5,3',5'-tetraCB In milk (up to 2 ppm In milk fat after a single oral dose of 0.8 mg/kg). Frank et al. (405) reported data on body burdens of PCBs In captive harp seals (Pagophllua groenlandlcua), as shown In the follow ing table: " Age at Death and time In Captivity (months) 12-11 15-14 24-23 27-26 38-35 52-50 Estimate PCB Wt. at Death Intake last 2-4 (kg) . Months (mg/day) 52.0 0.75 68.0 1.14 ` 70.0 . 1.07 63.8 . 1.01 65.0 0.75 54.0 0.69 . PCB PCB (mg) (mg/kg) 163 3.13 143 2^15 613 8.75 659 10.32 785 12.10 697 12:90 The diet of the seals contained an averaga concentration of about 0.1-0.3 ppm PCBs, but over 4 years' captivity the average whole-body concentration built up to 12.9 ppm (aaaq|x=S0 ppm In fat). 0^ 0 WATER PCB-00043135 IV.2.7. Storage and Blo-maanlfleatlon In Humana . Although (xtauslvs surveys have been made of the occurrence of PCBa In human tissues, there Is little systematic Information on rates of uptake or elimination. PCBs are efficiently absorbed through the gastrointestinal tract: Price et al. (350) measured PCBs In diets, urine and feces of eight preadolescent girls In Virginia, and esti mated absorption of the Ingested PCBs as 88Z. Humans are also ex posed via air, water, and dermal contact, but It Is generally believ ed that the major exposure la via the diet (1-3, 331, 351). . The most extensive data on distribution of PCBs in human tissues derive from Japan (35, 25). Fujiwara cited mean levels of 3.1 ppb In blood plasma, 4.;7 ppm In body fat, and 50 ppb In milk of persons surveyed in.the Kyoto area (25). He also cited a national survey which showed a mean level of about 34 ppb In 1116 samples of hvtnan milk. Kuratsune (35) cited a national survey which reported PCB levels In the range 0.04-1.7 ppm In skin, 0.3-6.4 ppm In fat,- and 0.01 to 0.6 In liver (Table III.16.9). . Although, these figures do not all refer to the same individuals, they suggest a distribution In the body tissues paralleling that of lipids, as in other animals. 0073055 WATER PCB-00043136 Surveys In Canada show that the majority of Canadians have residues of 1-2 ppm In their adipose tissue. Human milk of Ontario . residents was found to have average residues of about 1.1 ppm FCBs on a fat basis (342). . Extensive surveys In the United States have shown that the median level of PCBs In human adipose tissue Is slightly less than 1 ppm (Table IV.2.13) (353). Less Information Is available on the occurrence of PCBs In other human tissues. Finklea et al. (354) found mean levels of 2-3 ppb In blood plasma In South Carolina. Meaaurements of residues In human milk have ranged from not detect able up to 100 ppb; 30 ppb has been suggested as a typical level (331). The Total Diet Survey of the Food and Drug Administration pro vides an estimate of the dietary Intake of an average person In the United States: In recent years this estimate has been close to 9 jig/person/day (351: Table 17.2.14). The Total Diet Survey Is based upon a high consunptlon diet which Includes 4 kg food/day, almost twice that consumed by the "average" man (355). Hence the average concentration of PCBs In the food sampled In the Total Diet Survey Is about 2.2 ppb. Comparing this with the median level In adipose tissue from the Human Monitoring Program (353), the storage factor for human fat (ppm In fat/ppm In diet) appears to be at least 400. This Is far higher than that recorded for any experimental animal: see Section IV.2.6, where high figures for storage factora In mammal ian fat range from 120-180 in seals, through 35-60 In monkeys, 11-20 In rats, and 5^ln cows, to 0.7 In pigs. Several explanations are possible for this discrepancy: .013^ WATER PCB-00043137 Table IV.2.13 (from ref. 353) Levels of polychlorinated biphenyls In human adipose tissue Data Sample Percent . Percent ' 'Percent Percent source size nondetected < 1 ppm 1-2 ppm > 2 ppm Yobs. 1972 FY 1.973 Survey FY 1974 Survey 688 1277 1047 34.2 24.5 9.1 33.3 40.2 50.6 27.3 29.6 35.4 5.2 5.5 4.9 Table IV.2.14 (from ref. 351) Fiscal year Estimates of dally PCB Intakes . (total diet study--teenage male) Average dally Intake of PCB's* Total diet (ug/day) Meat-fish-poultry food class (pg/day) 1971 1972 1973 1974 1975 (1st half) 15.0 12.6 13.1 8.8 8.7 9.5 . 9.1 8.7 8.8 8.7 aLower limit of quantitative reporting 0.05 ppm with analytlcalNmethod employed. 00730*1 m WATER PCB-00043138 Zll (1) the dietary Intake estimated by the Total Diet Survey may be too low, because of a high frequency of residues near the level of detectability; ' (11) routes other than the diet may give significant exposures; (111) the residues stored In fat may reflect much higher exposures In the past; or (lv) humans may store PCBs much more efficiently than other animals. Jellnek and Cornellussen considered (1) and attempted to mini mize errors by assigning finite residue values to "trece" findings (3S1). Although other routes may be locally significant, no quantita tive data have been provided that suggest Intakes greater than a few micrograms per day for a typical person (2, 8, 331). The rapid de cline of PCBs observed In the Tusho victims (Table III.16.9) makes It unlikely that present storage levels reflect exposures more than a few years ago. Hence it seems probable that humans do In fact store PCBs more efficiently than other animals. Humans are known to store other chlorinated hydrocarbons, such as dleldrln (335, 356) much more efficiently than other animals. This Is of considerable importance for establishing "safe" levels for human intake, because It Implies that for a given dietary concentration, humans will have tissue levels up to 10 or more times higher than'those of experimental animals. ' -' There Is no precise Information on the molecular or isomeric constitution of the PCBs stored In human adipose, tissue in the 1* *6 00 WATER PCB-00043139 United States, although Kutz and Strassman state that most consist ofpenta-, hexa- and hepta-CBs (353). A detailed study of a pooled sample of human adipose tissue from Sweden Indicated that It contain ed 4-67. tetra-CBs, 23-25Z penta-CBa, 44-4651 hexa-CBs, 21-2251 heptaCBs, 451 octa-CBs, and 0.651 deca-CB (Table II.4.1). Muslal et al. (357) reported that most samples of human milk analyzed In Canada had PCBs similar In constitution to Aroclor 1254, but that peaks corresponding to Aroclor 1242 (l.e, tetra-CBs) were detected In all samples. Thus It seems likely that tetra-CBs, although minor, are not negligible constituents of the PCBs In human tissues. IV.2,8. Blo-accumulatlon and Blo-magnlflcatlon In Natural Ecosystems . Several writers have consented that blo-accumulatlon and blo- magnlflcatlon factors observed In wild animals In the field are often much greater than those measured under controlled conditions In the laboratory. Some examples, of this phenomenon as It relates to PCBs are given below. Escambia Bay. Florida. Hansen (74) and Nimmo (330) have pointed out that blo-accumulatlon factors estimated for several wild species In Escambia Bay are much greater than those measured In the laboratory (99, 105; see Table IV.2.15). The discrepancies are especially great for crabs (>230,000 in the bay, versus a maximum of 4,600 In the laboratory) and fish ( >670,000 in the bay, versus a maximum of 37,000 In the laboratory). O WATER PCB-00043140 ( >J Table IV.2.15 (from ref. 330) - HQACOMJLAnCN OF AROCLOR 1254 IN ESTUARINE ORGANISMS ORGANISM GETTER SHRIMP CRABS FISHES LABORATORY 165,000 26,000 4,6do 37,000 OONUNTRATICN FACTORS ESCAMBIA BAY > 100,000 > 470,000 > 230,000 > 670,000 r' kr J* , I 007306 WATER PCB-00043141 Hudson River. New York. In a field experiment, Telth et al. (358) held a number of fish of five species In a live-car (In contact with' river water but Isolated from sediments) for 14 days In the Hud son River below Fort Edward, N.Y. Whole-body residues of Aroclor 1016 Increased to 1.8-3.8 ppm during the 14-day period of exposure. Velth et al. calculated the average concentration of Aroclor 1016 In the river to be 100-170 ppt, and hence estimated blo-accumulation factors In their test fish to be 10-40,000, comparable to those ob- ' served In the laboratory (358). However: (a) FOB levels In wild fish In the Hudson River have been up to 100 or more times higher than those observed In the test fish (Appendix C, 359); (b) Telth et al. calculated the FOB concentration In the river from the known dis charges and the river flow: adsorption to the sediments Is likely to have reduced the actual cgncentratlon considerably. ' The Great Lakes. Extensive data are available on PCBs In water, sediments, and biota In the Great Lakes (Appendix C, pp. 31-54; 93, 360-368). Estimates of the degree of bio-accumulatlon are hampered by the difficulty of measuring concentrations of PCBs In water In the low parts per trillion characteristic of the open water of the lakes. The best set of data obtained by modern sampling techniques (use of polyurethane foam plugs for extraction from 20 liter samples see ref. 56) Is from Lake Superior, where the mean PCB level at Duluth was estimated at 0.8 parts per trillion (363). Corresponding mean levels In biota were estimated as 0.1 ppm In zooplankton, 0.25 ppm In sculplns, and 0.3-3.3 ppm In several other fish species (363). ooiobl WATER PCB-00043142 One of Che fish species with the highest residues in this survey was the^ lake trout (Salvellnus namaycush), with mean residues of 0.7 to 1.9 ppm at various sampling stations in western Lake Superior (363). Another survey of lake trout in western and northern Lake Superior reported residues In the range 2.7-13.8 ppm (mean 7.0 ppm) in the whole fish, and 7.6-62.8 ppm (mean 25.6 ppm) In the fat (extracted oil) (367). Canadian data (360) fall Into the same ranges, with the lake trout showing generally the highest residue levels (Table IV.2. 16). These data indicate bio-accumulation factors in fish In the range 3 x 10 to 9 x 10 (3 x 10^ when expressed on a fat basis). In Lake Michigan the average concentration of PCBs in water has not been measured precisely but appears to be substantially less than 10 ppt (362, 368): levels measured in the intake water of munici palities along the Michigan shore were all, with one exception, below 10 ppt (368), and even the concentrations reported in major tributary rivers to the lake have been only in the range 10-65 ppt (Appendix C, p. 46). FCB levels measured in fish in Lake Michigan have heen con sistently in the range 2-20 ppm, wet weight (16-200 gpm, lipid weight) (361, 368; Appendix C, pp. 40-45). These figures indicate blo-accumulatlon factors ranging up to at least 2 x 10 in whole fish (2 x 10^ in lipids). In Lake Ontario concentrations of PCBs in water in the range 35-56 ppt were reported in 1972 (Appendix C, pp. 31-32), but the siethod of extraction and analysis was not stated and the reportedly high levels in water are only partly reflected in high levels in .0073062 WATER PCB-00043143 Table IV.2.16 (from ref. 360) Polychlorinated biphenyls in commercial fish and fishery products, (area 3) Location. Lake Superior (A) Major species chub lake herring lake trout whlteflsh smelt yellow perch ' Landings8 PCB (pounds) (mean ppm) 306,000 1,611,000 195,000 328,000 788,000 67,000 3,295,000. 0.96 ( 8) 1.17 (10) 2.02 (37) 0.68 (11) 0.35 ( 6) 0.31 ( 5) Lake Huron (B) . chub 671,000 whtteflsh 950,000 carp 56,000 suckers - 176,000 yellow pickerel 269,000 sheepshead (drum) - . 29,000 coho salmon noncommercial 2,151,000 2.09 (17) 0.95 ( 8) 1.55 ( 4) 1.33 (12) 0.61 ( 9) 0.75 ( 2 5.11 (6) Lake Erie . alewlfe . 332,000 1.22 (14) rock bass 49,000 0.27 ( 2) carp 41,000 1.27 ( 7) yellow perch 12,190,000 0.19 (10) smelt 15,760,000 0.42 (21) yellow pickerel 234,000 1.16 ( 9) white bass 2,346,000 1.26 (28) catfish 88,000 2.04 ( 8) bullhead 109,000 0.26 ( 4) sheepshead (drum) 354,000 0.74 (15) coho salmon noncommercial 3.14 ( 8) 31,503,000 Lake Ontario (D) '.V bullhead yellow perch smelt white perch sunflsh carp rock bass eel coho salmon 248,000 699,000 103,000 290,000 203,000 395,000 42,000 222,000 noncommercial 2,205,000 0.73 (12) 1.23 (10) 4.16 (17) 1.84 (22) 0.74 ( 6) 1.69' (10) 1.76 (18) 17.14 (49) 4.97 ( 3) landings less than 25,000 pounds per species are not Included. * 0OllQb3 WATER PCB-00043144 biota. Average PCB levels In fish are similar or slightly higher th^n those In Lake Michigan, In the range 4-30 ppm (360). Taking the water data at face value, these Indicate blo-accumulatlon factors In the range 10 -10, somewhat lower than those In the same fish In Lakes Michigan and Superior. ' Data on eggs of flsh-eatlng birds from the Great Lakes Indi cate further blo-magnlflcation of PCB residues by the birds. Table IV.2.17 summarizes PCB residues In eggs of Herring Gulls (Larus argentatus) from the Great Lakes (366). The geographical pattern of contamination generally parallels that found In the fish (Table 1T.2.17), Taking Lake Superior as an example, the mean PCB level of 60 ppm In .gull eggs (roughly 1000 ppm on a lipid basis) Is much great er than that of 0.3-7 ppm In fish (5-40 ppm In lipids) (363). This suggests a degree of blo-magnlflcatlon by the-birds much greater than anything reported In the laboratory (Section IV.2.5). The California Current. Young et al. (12, 369-370) have pre sented extensive data on PCB residues In the marine environment off southern California. Besldues In Inshore areas are variable because of localized sources, and hence are difficult to use to derive field estimates of blo-accumulatlon. However, the maximum water concentra tion measured at the surface above one sewage outfall was 4 ppt (370, p. 332); corresponding tissue levels In fish, crabs, and mussels ranged up to 4.3, 3.8 and 1.7 ppm respectively. Indicating bio accumulation factors pf the order of 10. ' Young jj. al. also presented measurements of the background concentrations of PCBs In the California Current which flows Into V.1V*' WATER PCB-00043145 Table IV.2.17 (from ref. 366) > Organochlorin. residue levels** In Herring Gull egggln the Great Lakeib DCS ODD * DOT ' Dieldrin Heptachlor Epoxide Minx 7. Hexachlorobenzene PC9c ' "i Lake Ontario Lake Erie Lake Huron Lake Superior Lake Michigan 39 22.0 (9.3-35.1) 42 7.04 (3.9-14.3) 40 13.9 (5.4-41.9) 39 19.6 (9.6-47,1) 10 31.9 ' (15.9-145) 0.09 (trace-0. S3) 0.09 (trace-0.24) 0.10 * (trace-0.39) 0.15 (trace-0.4) 0.01 / (0.01-0.07) 0.09 (0.02-1,04) 0.37 (0.09-1.09) 0.04 (0.01-0.15) 0.30 (0.10-0.69) 0.09 0.41 (0.01-0.32> (0.13-0.97) 0.12 (0.02-0.59) 0.13 ' (0.07-0.39) 0.39 (0<13-1.35) 0.49 (0.3-0.92) 0.12 (0.01-0.36) 0.14 (0.04-0.29) 0.12 (0.04-0.26) 0.14 (0.07-0.39) 0.16 (0.11-0.60) S.06 (1.9S-19.6) 0.31 (0.14-2.19) 0.S6 (0.06-6.92) 0.66 (0.2-5.17) <0.01 0.19 142 0.51 (0.01-0.72) (73.9-261) (0t29-1.471 0.11 65.9 0.22 (0.06-0.31) (41.2-110) (0.11-0.3$) 0.14 Sl.S 0.23 (0.05-0.42) (IS. 4-III) (0.11-0.50. 0.11 60 0.3i (0.02-0.33) (33.4-149) (0.16-0.13) 0.04 91.3 (0.02-0.14) (55.1-395)* n.d. 0073065 sHedlsn Values and ranges (In brackets) In parts per nllllon, vet weight. . bEggs were collected fron two colonies in etch lake In both 1974 and 197S except fro* lake Michigan where they were fron a single-colony in 197S. cPolychlorinatcd biphenyl values are based on a 1:1 nixturc of Aroclor 1260:1254. n.d.Not detected. UJ j WATER PCB-00043146 the area from the northwest: these fell Into the range 0.27-0.49 ppt with a mean of 0.40 ppt (369, p. 29). Away from the Influence of s. local sources of FCBs, mean residues In mussels and dover sole were In the ranges 14-26 ppb and 40-50 ppb respectively. Referred to the reported background level In the water these Indicate blo-accumulatlon factors of the order of 5 x 10^ and 10^ respectively. However, Risebrough et al. (56, pp. 12-20) have presented evidence that the FOB levels reported In water by Young et al. are at least 10 times too high; If so, these estimates of blo-accumulatlon factors would be at least 10 times too low. Reported levels In birds and mammals In the California current and the outer Islands are much higher than those reported In mussels and fish: 0.15-2 ppm In whole bodies of various flsh-eatlng birds (371); 10 ppm-In the whole bodies (200 ppm In lipids) In plankton eating birds (225, 371); 20-400 ppm In the egg-llplds of flsh-eatlng birds (372, 373); 12-145 ppm In the fat and 0.5-9.7 ppm in the livers of sea-lions (374). These figures indicate a substantial degree of bio-magniflcation by the vertebrates and an overall storage ratio (ppm In fat/ppm In ocean water) exceeding 10 In some cases. Despite the variability In residue levels near to major sources of FCBs, the Southern California studies provide useful in formation about differential blo-accumulatlon of PCB components. At one of the largest sources, a sewage effluent at Falos Terdes, about 667. of the FCBs discharged during 1974-75 consisted of components characteristic of Aroclor 1242 (tetra- and trl-CBs): the remainder ' ooi*066 WATER PCB-00043147 23? A consisted of components characteristic of Aroclor 1254 (penta- and he*a-CBs) (369: Table 1). The corresponding ratio in the tissues of fish (dover sole) sampled at the same place was reversed (67J Aroclor 1254) (370: Table 3). This suggests that in these circumstances penta- and hexa-CBs were accumulated roughly 4 times more efficiently by fish than tetra- and tri-CBs. . The North Atlantic Ocean. Several authors have published measurements of the order of 1 ppt or higher for concentrations of 0073067 WATER PCB-00043148 PCBs in seawater in Che open Atlantic Ocean (52, 53, 375, 376), but the^ problems of sampling without shipboard contamination are very severe and Rlsebrough et al. (56) have given evidence that all the figures are much too high. Certainly 1 ppt la an upper limit for levels to be expected in ocean water.* Discounting data from locally polluted Inshore waters, PCB residues reported from biota samples from the open Atlantic Ocean include the following: Organism PCB residues (ppm) Whole body or muscle Lipid Ref. Zooplankton It Fish Deep-sea fish Oceanic birds Seabird eggs (Iceland) Seals 0.01 - 0.12 0.1 - 0.45" 0.002 - 2.1 0.008 - 0.10.07 - 2.65 0.001 - 0.17 0.2 - 21 0.4 - 27 0.1 - 9.8 0.3 - 260 0.4 - 31 0.2 - 3 4 - 12.5 0.04 - 38 0.25 - 74 4 - 320 2 - 75 378 379 376 380 360 56 376 387 380 386 Despite the wide variability (and some problems in sampling plank ton -- see ref. 56) these data indicate that PCB residues in plank ton and fish often reach 0.1 ppm in the whole body and 1 or even 10 . ppm in lipids. Even accepting the 1 ppt levels reported in seawater, these figures indicate bio-accumulation factors of the order of at least 10s (10- 10? in lipids),. Plankton-eating and fish-eating birds and mammals often haveresidues an order of magnitude higher* 1 The first report of levels up to 150 ppt (375) has been questioned (377) and is inconsistent even with subsequent reports of around 1 ppt, which are themselves questionable. 0o7308 WATER PCB-00043149 till. Levels were highest o all In predators (up to 88 ppm In muscle and 535 ppm In fat In Glaucous Gulls (Larus hyperboreus) at Beat Island -- ref. 38t). The data cited above show that In natural environments animals generally blo-accumulate and blo-magnlfy PCBs by factors substantially higher than those usually measured In laboratory experiments. Despite the difficulties In measuring PCBs In water at levels near or below 1 ppt, there Is now evidence from several areas that aquatic Inverte brates and fish can blo-accumulate them by factors as high as several million fold from ambient water, and that fish-eating birds and mam mals can further blo-magnlfy these concentrations by factors of 10 100. Reasons for this enhanced magnification In natural environments have been discussed by Nlsbet (336). One likely reason is that FCB residues In many organisms, increase with age (see Figure IV.2.18 for an example involving lake trout). Hence long-lived wild animals have an opportunity to accumulate PCBs to much higher levels than indivi- . 'duals exposed experimentally for limited periods in the laboratory.* Other reasons proposed for this field effect include selective ex posure in a patchily contaminated environment, and selective preda tion on contaminated prey (336). Whatever the explanation for the phenomenon, the blo-accumulatlon factors observed In the field are those that are relevant to assessment of exposure to organisms that may aat the contaminated fish. * See also ref. 405 for an example involving seal's, whose body bur dens were still ineraasing after 4 years In captivity (see p. 326 above). ,v 0073069 WATER PCB-00043150 (' ,v. ' Tie concentration of PCB's in Cayuga Lake trout as a function of age. Figure. IT.2.18 (from ref. 381) (' 0073070 m - WATER PCB-00043151 IV. 3. Presence In the Aquatic Environment A review of PCB levels In the environment Is eppended to this V; document as Appendix C and only a brief summary Is Included here. PCBa are widely distributed In the environment of North America. Data gathered from nationwide monitoring activities by the U.S. Geological Survey (48) show mean concentrations of PCBs In the range 0.01 to 0.05 /ig/1 (10-50 ppt) In unfiltered water samples and from 0.25 - 218 ppb In bottom sediments. The highest levels were found In the Southeast (South Atlantic Slope and Eastern Gulf of Mexico drainage basins) and the Northeast (North Atlantic Slope and St. Lawrence River drainage basins). (See Appendix C, pp. 3-15). In the National Pesticide Monitoring Program, PCBs have been i-f>tk detected^at all of the 100 monitoring stations, although the percen tage of composites with detectable residues fell from 1001 to 511 between 1969 and 1973 (Appendix C, p. 27). In the same period the .number of composite samples of fish with residues greater than 5 ppm fell from 611 to 40% (Ibid.). Generally, the higher concentrations were found In certain river systems with Industrial activity. Includ ing the Allegheny, Kanasha, Cumberland, Tennessee, Ohio, Mississippi and Missouri Rivers (93). High residues have also been found In fish in the Great Lakes, In the Hudson and other northeastern rivers, and In a number of estuaries and inshore waters (Appendix C, pp. 31 87; 2, 93, 359-364, 367-370, etc.). PCB residues In fish detected In monitoring programs Include chlorobiphenyls characteristic of Aroclors 1242 and 1248 (trl- and tetra-CBs) as well as those 0oi?oiv WATER PCB-00043152 characteristic of Aroclors 1254 and 1260 (penta-, hexa-, and hepta- CBs (93, 370, 367, 382). For example, in the crosscheck samples ;s. . analyzed for the National Pesticide Monitoring Program, 1970-73, 26 of 131 samples listed, contained.components characteristic of Aroclors 1232 or 1242, and 70 had components characteristic of Aroclor 1248; all had components characteristic of Aroclors 1254 and 1260 (382). In lake trout samples in Lake Superior, more than half the PCB resi due components were listed as matching Aroclors 1242 and 1248 (367). Similarly, numerous tetrachloroblphenyls and at least one trichloro- biphenyl component were identified in lake trout and coho salmon from Lake Michigan (95). Thus trl- and tetra-CBs as well as higher CBS are widely stored in fish in the United States. PCBs are also found very widely and often at high levels in fish-eating birds (1, 2, 56, 93, 225, 364-366, 371-373) and fish eating mammals (374, 383-386). Levels are particularly high in the fat (blubber) of marine mammals such as seals, dolphins and polar bears (383-386; see Table IV.3.1). PCBs are widely distributed in human tissues, ^reflecting general intake in the diet (351-356). ,. ooi'oU WATER PCB-00043153 Table IV.3.1 (from ref. 383) .< LOCATION Washington: Puget Sound California: San Miguel Island TOTAL DDT PLUS PCBs IN BLUBBER OF HARBOR SEALS COLLECTED IN 1971 NO. OF SEALS . '. MONTH COLLECTED i AGE. YEARS TOTAL PDT & PCBs (mg/kg, wet wt.) MEAN * INDIVIDUAL VALUES 2 June-August - 1 862.7 459.4; 1,620.0 5 June 9--30f 610.7 380.7; 318.5; 425.5; 518.3; 2,350.0 Oregon: Columbia River Alaska: . - Prlbllof Islands '3 i . 3 May . August 2-4 (a.) 62.5 U.3 27.7; 80.4; 109.9 . 6.-8; 7.1; 27.8 . a. canine teeth lost 0073073 O'J t-J WATER PCB-00043154 IV.4. Effects on Biota and Natural Ecosystems >, Effects of PCBs on organisms and ecosystems fall Into two categories: (a) effects observed In the laboratory under controlled conditions of exposure to PCBs; (b) effects observed In natural . systems and associated with PCB'exposure by epidemiological or equiv alent methods. . Effects of type (a) are relatively easy to extrapolate to natural ecosystems: If an effect Is observed In the laboratory when test organisms are exposed to a certain level of PCBs, It Is reasonable to assume that similar effects will take place In the field If the same or comparable species are exposed to the same level of PCBs. It Is often likely that effects will occur at lower levels In the field, for the following reasons: (1) natural systems Involve many species and It Is likely that some will be more sensitive to PCBs than the species chosen for testing; '. (11) under natural conditions organisms are exposed Inter mittently to environmental stresses (temperature, pH, oxygen depletion, food shortage, predation, etc.) which ' generally make them more vulnerable to the effects of an additional toxic stress; . (Ill) natural systems are exposed to other pollutants and It Is known that additive and and synergistic toxic effects may occur. 001^074 WATER PCB-00043155 Thus It Is reasonable to expect adverse effects to occur In the field at lower exposure levels than in the laboratory, but it is difficult to predict the lowest hazardous levels without actual field studies, Effects of type (b), however, which depend on field studies, are often difficult to interpret or to establish with reasonable certainty. When an adverse effect is observed in the field, it is often associated with exposure to more than one pollutant: it is difficult to establish whether it is caused by FCBs, by another pollutant, or by both acting together. Furthermore, FCBs are patchily distributed in the environment, sb without extensive measure ments of concentrations in water it is difficult to establish the exact degree of exposure of the affected organisms. Accordingly to judge the effects of PCBs on natural systems it is necessary to weigh laboratory and field evidence together. ^ IV.4.1. Effects Observed in the Laboratory The following effects of PCBs have been observed in control led experiments at exposure levels which occur locally in the field: (i) Change in species composition of phytoplankton communities (Section III.3). Effects on the growth and rate of cell division of sensitive species of phytoplankton have been noted at PCB concentra tions as low as 0.1 ppb (84, 86) and large effects on the species composition of mixed cultures took place at this concentration (92). The effects of lower concentration!have not been studied. The changes involve replacement of sensitive diatoms by resistant green 0073075 WATER PCB-00043156 algae and are expected to have substantial effects on consumers and hence on food-webs (83). Synergistic effects with DDE are also in volved (90). (ii) Change in species composition of marine animal communi ties. Effects on species composition and biological diversity of communities of estuarine animals developing in the laboratory from planktonic larvae in flowing seawater were noted at PCB concentra tions as low as 0.1 ppb, the lowest concentration tested (ref. 109; Section III.4.2). . The results reported in these experiments (refs. 88. 92. 109) are of particular importance because they demonstrate that PCBa affect mixed communities of plants and animals at concentrations much lower than those required to show effects on single species popula tions . (ill) Effects on aquatic invertebrates (Section III.4). Reproduction and growth is impaired in aquatic invertebrates at con centrations lower than those required to cause mortality. The water- flea Daohnla magna and the midge Tanvtarsus dlsslmllls are both affected at levels between 0.4 and 0.5 ppb (93: Table III.4.1). Synergistic effects with DDT are also involved (94). .. (iv) Sublethal effects on fish (Section III.5). Although there is little evidence that PCBs affect survival of fish at con centrations below 1 ppb, adverse effects on reproduction have been noted at concentrations as low as 0.1 - 0.4 ppb, corresponding to residues of 5 ppm in eggs (113, 114, 115). Effects were also noted , 00730/6 WATER PCB-00043157 on thyroid function In fish exposed to a dietary level of 0.48 ppm PCBs (which corresponds to roughly 0.4 ppm whole-body residue) S. (refs. 73, 95; Figure III.5.2). (v) Effects on reproduction In birds (Section III.6.4). Marked effects on reproductive success (especially embryonic mortal ity) have been noted in chickens at dietary levels as low as 8-10 ppm of PCBs, corresponding to residues of only 1-2 ppm in whole eggs. It Is doubtful whether the chicken Is a good model for wild birds, since other species have proved less sensitive In laboratory tests. However, second generation effects and synergism with DDE have been noted In other species at dietary levels of 10 ppm (161, 164). (vl)'Hepatic porphyria In birds (Section IV.6.3). PCBs in duced ALA-synthetase and caused porphyria In quail at doses as low as 1 mg/kg/day and probably at 0.1 mg/kg/day. These effects were associated with residues as low as 1 ppm In the liver (145: Figure III.6.1). (vll) Effects on mink (Section III.9.4). Mortality and total reproductive failure took place In mink fed a diet cpntalnlng only 0.64 ppm of PCB residues, the lowest dose tested (201).. In another experiment at a higher dose level, there was evidence of synergistic effects with DDT and dleldrln (200: Table III.7.5). The mink Is the only fish-eating manual whose sensitivity to PCBs has been tested. All these effects have been noted at exposure levels which occur quite frequently In natural environments. Concentrations of 0.1 ppb are reported not Infrequently from polluted fresh waters 0073077 WATER PCB-00043158 zw (48; Appendix C)., and concentrations as high as 0.4 ppb are reported occasionally. Fish throughout the United States have tissue residues those associated with thyroid malfunctions (Appendix C; 93). Fish-eating birds often are exposed to fish with PCB residues as high as 8-10 ppm (Section IT.3); many, perhaps most, fish-eating birds in the United States have residues in their eggs as high as the 1-2 ppm that affect hatchabllity in chickens (Section IT.3.; 1, 2, 56, 225, 364-366, 371-373). Many fish-eating birds, even from remote parts of the Atlantic Ocean, have PCB residues In the liver as high as 1 ppm (140, 387, 388). Fish-eating mammals In the Great Lakes and many other areas of'the United States will be exposed to diets containing 0.64 ppm of PCBs (Section IV.3; 93). Thus it is reasonable to suppose that at least some of the adverse effects reported in the Laboratory are taking place in natural environ ments . IV.4.2. Effects Observed in the Field. ' Several adverse effects- on wild animal populations have been reported which have been associated reasonably plausibly with ex posure to PCBs. '. . (i) Effects on reproduction in salmon. Swedish investigators have reported a statistically significant association between hatch ing failure in eggs of Atlantic salmon and PCB residues. Concentra tions in the range 0.4-1.9 ppm, wet weight (7.7-34 ppm, lipid weight) were associated with mortalities between 16 and 100 percent (118). This report wasipublished only as a preliminary note and complete details are not available. WATER PCB-00043159 (II) Reproductive failure In other fish. Several Species of fish In Lake Michigan have not been reproducing effectively In recent yeat-s (409-410). Despite annual stocking of lake trout since 1965, for example, no wild-raised young have been collected (409). Eggs of Atlantic salmon and northern pike have shown very low hatchabillty In the laboratory (410). These reproductive failures have been associa ted circumstantially with residues of PCBs and other chemicals, but nuclear link has yet been established. (III) Pin rot syndrome in marine fish. "Fin rot syndrome" has been Induced in the laboratory In spot exposed to 3-5 ^ig/1 of Aroclor 1254 (106). The syndrome appears Identical to that observed In wild fish (croakers and spot) which died under conditions of warm weather and oxygen depletion in Escambia Bay, Florida (106). Although the deaths of the wild fish were attributed originally to lack of oxygen In the water (99), the levels of PCBs In the dying fish (around 10 ppm) were similar to those in fish experimentally poisoned In the laboratory (99, 111). This suggests that PCBs may have'played at least a contributory, role In the Incidents. In independent observations, McDermott et aU (370) have reported that fin erosion has recently become prevalent In dover sole off Palos Verdes and Orange County, California. Diseased sole have higher PCB residues than healthy fish (median concentration 2 ppm, wet weight, versus 1 ppm), but the difference was not quite statisti cally significant (P 0.1). The syndrome was associated with con centrations of PCBs In water not greater than 4 ppt (370, p. 32). 0073079 WATER PCB-00043160 3 5* (iv) Deaths of flsh-eatlng birds. Koeman (142) reported that a number of Cormorants (Phalacrocorax carbo) found dead in the Nether- lands In 1970 contained high residues of PCBs (mean 190 ppm In brain, 320 ppm In liver, wet weight). These concentrations were similar to those In Cormorants experimentally poisoned with PCBs (76-180 ppm In brain, 210-290 ppm In liver) and Koeman attributed the deaths of the wild birds primarily to PCB poisoning (142). However, the correspond ing levels of PCBs in the fish and water were not known. In November 1969, more than 50,000 seabirds, mostly murres (Urla aalge), were found dead In the Irish Sea. The most consistent findings at post mortem were emaciation, liver damage, and high resi dues of PCBs In the liver (2-880 ppm, wet weight) (389). Birds col lected In apparent good health at the same time and place had body burdens similar to those of the dead tilrtin\ t-f In the healthy birds' the PCBs were stored In the fat, whereas In the dead birds the PCBs were largely In the liver. It,-therefore seems likely that food shortage triggered the Incident, but since the liver residue's of PCBs In some birds were well Into the lethal range (142) tt seems likely that^ died~IPOm PCB poisoning. This Is an Important example of the interaction of environmental and toxic stresses. Residues' In fish In the same area were In the range 0.01 2.6 ppm (389). Glaucous Gulls (Larua hvnerboreus) wMch had been feeding on the eggs of seabirds at Bear Island In the Arctic Ocean had high levels of PCBs dn the liver (mean 24 ppm). One Individual found 0073080 WATER PCB-00043161 dying with failure of coordination of the limbs had 311 ppm FCBs In the liver (388) and la therefore likely to have succumbed to PCS poisoning (140, 142). The example Is Important because the PCBs were apparently acquired through the natural food chains of the North Atlantic and Arctic Oceans (387, 388). Two Bald Eagles (Hallaetus leucocenhalus) have been reported dying with high residues of PCBs in the brain. One found dead In Michigan In 1970 had 235 ppm of PCBs and 385 ppm of DDE in the brain (390). Another found dead in Michigan In 1972 had 190 ppm of PCBs and 55 ppm of DDE in the brain and 1200 ppm of PCBs In the whole body: the Immediate cause of death was drowning. As It was seen to fall off a perch Into the water (391). In each case the residues of hath PCBs in the brain were In the lethal range (142, 391) and In the first case residues of DDE Were also: It seems likely that the birds died from the combined effects of PCBs and DDE poisoning. The first bird was from the Lake Michigan area (Grand Traverse county); the second bird was probably from southeastern Michigan. The Bald Eagle is an endan gered species and is subject to the special provisions of the Endan gered Species Act of 1973. (v) Effects on reproduction In flsh-eatlng birds. Herring Gulls are experiencing low reproductive success at colonies in Lake Ontario, but appear to.be breeding normally at colonies In the other Great Lakes, Including Lake Michigan (366, 392, 393). The breeding failure Is characterized by early embryonic mortality (393) and Is associated with high levels of PCBs In the eggs (Table IV.2.17). 0073081 WATER PCB-00043162 Recent studies have shown that the embryonic mortality Is associated with porphyria and edema (subcutaneous, pericardial, and abdominal), and that these conditions ere statistically associated with high PCB levels (406)* In addition, deformities of the feet, toes, and eyes have been recorded In chicks of Herring Gulls and three other fish-eating species In Lake Ontario (407). In studies of other species of fish-eating birds, reproductive failure and eggshell-thinning have been associated primarily with residues of DDE (56, 364, 372, 373, 394) and no other study has sug gested an independent effect of PCBs. However, only a few species have been studied to date. (vi) Premature births In sea-lions. Premature pupping In sealions has been noted on San Miguel and San Nicholas Islands, Cali fornia, since 1968. Most or all of the premature pups die (374). Residues of PCBs were much higher In the females giving birth to premature pups than in those of. females giving birth at full term 00738Z WATER PCB-00043163 (means 112 ppm and 17.1 ppm PCBs In blubber, respectively). However, residues of DDT compounds, primarily DDE, were also elevated In the females giving birth prematurely (means 824 vs. 103 ppm In blubber). Fish In this area have residues generally in the range 0.01-2 ppm PCBs, ranging up to 4-6 ppm in the vicinity of sewage outfalls (370). In all cases sumnarized above, the association between PCB levels and adverse effects in wild animals Is somewhat circumstantial: this Is Inevitable In an epidemiological type of investigation. How ever, In each case the suspected adverse effect can be matched with adverse effects noted in laboratory animals with comparable residue levels. In several of the cases the adverse effects were also assoc iated with high levels of DDE, and additive or synergistic effects seem likely. The interaction of environmental and toxic stresses is also apparent in several cases. Together these examples seem suffi cient to show that current levels of environmental contamination with PCBs are having adverse effects on natural ecosystems, but the ambient concentrations at which they do so has not been established. 1V.5. Potential Effects.in the Human Population. As detailed in Section III, several different toxic effects have been observed in experimental animals exposed to PCBs in the diet at levels as low as 0.5-2.5 parts per million. Surveys by the H.S. Food and Drug Administration suggest that the average level of PCBs in the human diet in the United States is now about 2.2 parts per billion (Section IV.2.7 above). There is thus a factor of 200 1000 between tfxb average level in the human diet and levels known <*p7303 WATER PCB-00043164 to cause effects In experimental animals. ' ,V. ' In addition, residues of PCBs are sporadically distributed in the food supply and some Individuals Ingest much more than the average: persons with dietary preference for fresh-water fish end breast-fed Infants have been Identified as high-risk groups (2, 331). It is not yet possible to use epidemiological studies to estimate safe levels of intake, for the following reasons: (1) Epidemiological studies in the general population are because of the widespread exposure of the general public. ' (li) The Tusho incident, although providing some dose-response information, is confused by the high level of PCDFs in the ingested PCBs. (ill) Occupational experience in the past has provided no quantitative dose-response information and In any case may be con- ^ fused by effects of PCDFs In older formulations. - (Iv) Current studies of occupationally exposed workers are still in progress and in any case will provide information only about adults. v# ooi*08'1 WATER PCB-00043165 V. Criteria Formulation The criterion level for PCB's is 0.001 ug/1 PCB's. It should be: noted that if all PCB entry into the. environment were to stop inmediately, the environmental burden would in sane waters still Exceed the anbient water quality criterion. VSiile EPA is attempting to prevent the continuing discharge of PCB's, or at least to minimize the continuing discharge of PCB's into the aquatic environment, the responsiblity for protecting man from contaminated interstate oaimercial food products is with the Food and Drug Administration of the Department of Health, Education, and Welfare. . The chronic'effects of PCB's in man may occur at extremely low concentrations. Although it becomes virtually inpossible to state with confidence that any PCB concentration above zero provides an atiple margin of safety for man, the PCB criterion mrrfeer is believed to provide protection for the aquatic environment'. s The criterion for PCB's is derived with consideration of the bioaocunulation potential rather than solely upon the more traditional use of an application or safety factor applied to an acute toxic value. The reasons far this approach are the environmental persistence and bioaccurulation potential of PCB's as outlined below: 0073085 WATER PCB-00043166 . lfce persistence of PCB's in the environment is such that they present long-term environmental hazards. The laboratory determination of the bioaccumulation potential for PCB's in general has been shown to be around 274,000 times the PCB level in the test water. There have been reports that * ^certain animals captured from the aquatic environment have apparently bioaccunulated PCB's in their flesh to levels of between 3 and 10 million times that of the water concentration in which the animals were captured. It is important to note, however, that the level of PCB contamination in the food of the captured aquatic life, as well as the concentrations of PCB's in which these organisms resided for the predominance of their life periods, are unknown. Thus, the actual bioaocunulatian potential of PCB's in the aquatic environment is uncertain. Therefore, the bioaccunulation potential demonstrated in the laboratory of 274,000 is the ntitfcer which is used in determining a criterion level. The demonstrated physiological effects of PCB's on fish and consumers of fish and wildlife, which include reproductive failures in a broad phylogenetic group of aquatic and terrestrial animals, liver enlargement, enzyme induction and skin lesions, require the consideration of parameters beyond the usual acute toxicity and safety factors. Data which support a one part per trillion (0.001 ug/1) criterion are: 0.1 ug/1 will cause population shifts in phytoplankton. It is important to note that at this level of the food web . bioaccunulation does not play a role in the adverse effect. However, this concentration is extremely low and does have significant ecosystem consequences since shifts In population at this level will have an inpact on the food for top predators. ' 0.4 to 0.5 ug/1 lowers the reproductive potential of seme invertebrates. 0.1 ug/1 will lower the species diversity index of seme invertebrates. Bioaccunulation has been demonstrated in the laboratory at concentrations of around 274,000 times the water level. 0073086 m WATER PCB-00043167 "V . '' . . PCB contaminated fish will cause detrimental effects on pcnsuners of fish. At 0.64 ppm (milligrams per kilogram in the food), total V--reproductive failure was observed in mink. At 3.57 ppm (milligrams per kilogram in the food), mink died. At 2.5 ppm (milligrams per kilogram in the food), reproductive dysfunctions occurred in the rhesus monkey. At 3 ppm (milligrams per kilogram in the food), some mortality was observed in the rhesus monkey. ' In rats, 5 ppm (milligrams per kilogram in the food) resulted in the induction of enzymes in the liver. At levels of between 40 and 300 ppm PCB's in chicken feed, deaths of chicken occurred. Reproductive failures occurred at between 8 and 10 ppm PCB's in the food. Ulceration in the stomachs of dogs occurred after long-term feeding at 1 ppm PCB in feed. - Based upon the highest bioaccimulaticn factors demonstrated in the laboratory, a water criterion of no more than 0.001 ug/1 (1 part pier trillion) should provide protection to the aquatic ecosystem, and should provide a margin of safety to the consumers of aquatic life. ' ` ' Adverse effects in man that are due to PGB or polychlorinated dibenzofurans (PCDF) contamination have included: chloroaoene, swelling of the eyes, and liver involvement. There was no quantification of the levels of PCB's ingested by those individuals affected by PCB's. The carcinogenic threat due to PCB's has not been conclusively demonstrated, although based upon interpretation oo m67 WATER PCB-00043168 >N.f .' ' of data*, some researchers have concluded that PCB's have resulted, in cancer in the mouse and thus they represent a threat to man. teased upon the proven bioaccurulatian potential of 274,000 times the ambient eater concentration in controlled conditions, the level of 0.001 ug/1 PCB's should afford protection for consumers whose sole diet consists of aquatic organisms contaminated at the worst or maximal level predicted by the laboratory data. * 0073 88 WATER PCB-00043169