Document 1QYyaNJXoZwxpGqYKbOzxxYYX

TOXIC Ol.OGY AND API'UtD PHARMACOLOGY 19, 617-633 (1971) Dermal Toxicity Studies of Technical Polychlorinated Biphenyls and Fractions Thereof in Rabbits J. G. Vos AND R. B. Beems Institute of I eferinary Pathology and Institute of Veterinary Pharmacology and Toxicology, University of Utrecht, Biltslraat 172, Utrecht, The Netherlands Received October 5, 1970 Dermal Toxicity Studies of Technical Polychlorinated Biphenyls and Tractions Thereof in Rabbits. Vos, J. G., and Beems, R. B. (1971). Toxicol. AppL Pharmacol. 19, 617-6.33. A significant difference in toxicity between 3 polychlorinated biphenyl (PCB) preparations was found in a prior study: Clophcn A 60 and Phenoclor DP6 showing the highest, Aroclor 1260 the lowest, toxicity (Vos and Kocman, 1970). A subsequent study revealed the presence of tetra- and pentachlorodibenzofuran in Phenoclor and Clophen tVose/ al., 1970). In the present study, application of 118 mg of the 3 PCB's (5 times per wk. for 38 days) on the back skin of rabbits also resulted in differences in toxicity. PCB-induccd skin lesions were hyperplasia and hyperkeratosis of the epidermal and follicular epithelium. Hislopalhology of the liver included centrolobular degeneration, centrolobular liver ceil atrophy, focal necrosis, and cytoplasmic hyalin degeneration. Definite hyperplasia and hyperkera tosis of the follicular epithelium of the ear skin were seen after the topical application of fractions of Phenoclor and Clophen eluted from chromato graphic columns with 25% dicthylether in hexane. The fraction from Aro clor caused a minimal hyperplasia and hyperkeratosis of the follicular epithelium. PCB-induced kidney lesions were hydropic degeneration of the convoluted tubules and tubular dilatation with the presence of casts. This . r ..... dilatation was demonstrated also by a significantly increased relative percentage of the diameter which corresponds to the space of Bowman. Moreover, thymus atrophy and lymphopenia were found. Fecal coproporpbyrin and protoporphyrin excretion was increased. Polychlorinated biphenyls (PCB) have been identified in tissues of fish and wildlife in many countries (Jensen, 1966; Holmes el ah, 1967; Holden and Marsden, 1967; Kocnian ct al., 1967; Risebrough eta!., 1968; Koeman el a!., 1969; Jensen efal., 1969; Duke et al., 1970; Preslt er al., 1970). PCB preparations are extremely stable, oily fluids with very low aqueous solubility. They arc used as lubricants, as heat transfer media, in protective coatings for wood, metal, and concrete, and for many other applications. Which of these applications has contributed to the present environmental contamination has not yet been established. Inhalation and feeding experiments in rats with 65% chlorinated PCB resulted in liver injury (Drinker el al., 1937; Bennett el al., 1938). Liver damage and skin lesions have been described after inhalation, ingestion, and application to the skin using rabbits, guinea pigs, and mice (von Wede! et al., 1943). Hydropericardium occasionally accom panied by abdominal edema was found in chicks fed PCB (McCune et al., 1962; Flick 617 >v,wTiu^i-U vat^*V '/r \s V v- %: Vy;'' ... I"? >: ; [if S' la 1' &*$** ifeSl f Fj \. & 025597 . 'i'c ' V^^v**'*. . Vyl "j7rV,'r,n4 ^ .. f-Pi It*' r ; * V'. . */ !' . . .7 STLCOPCB4009552 fjp el al., 1965; Vos and Koeinan, 1970). Epithelial and follicular hyperplasia and hyper keratosis were found after application of PCB on the skin of rabbits (Adams et al. 1941). Subcutaneous injection of42% chlorinated PCB in rats, guinea pigs, and rabbits resulted in liver injury and skin lesions, which were essentially those of chloracne (Miller, 1944). Follicular hyperkeratosis is an important feature of the occupational disease known as chloracne, which is characterized by the appearance of papules, come dones, and cysts. It may develop after exposure to some highly chlorinated aromatic compounds. Examples are chlorinated naphthalenes (Grcenburg et al., 1939; Hambriek, 1957; Shelley and Kligman, 1957) and chlorinated phenols (Hofmann, 1957; Hauer et al., 1961; Bleiberg et al., 1964). The active compounds in chlorinated phenols have been identified as chlorinated dibenzodioxins (chlorinated diphenylene dioxides) which may appear as contaminants in the synthesis of chlorophenols (Bauer et al., 1961). Liver injury (acute yellow atrophy), whether accompanied by chloracne lesions or not, is also described in people engaged in the manufacture of the chlorinated hydrocarbons (Strauss, 1944; Braun, 1955; Behrbohm, 1959). Also for the PCB's there are early re ports of occupational chloracne (Jones and Alden, 1936; Schwartz, 1936). The presentday use of PCB does not seem to have led to this kind of difficulty. A significant difference in toxicity between 3 samples of commercial PCB preparations was found in a previous study (Vos and Kocman, 1970), despite the marked resemblance of the gas chromatograms and the mass spectra (Koeman et al., 1969). One hundred per cent mortality, centrolobular liver necrosis, and sc and abdominal edema were found only in chicks fed the samples of Clophen A60 and Phcnoclor DP6. Hydropericardium was recorded in nearly all chicks fed these PCB's, and was only occasionally seen in chicks fed with Aroclor 1260. Porphyria was found as a general PCB effect (Vos and Koeman, 1970). The hypothesis that in some technical PCB mixtures toxic factor(s) responsible Tor the chick edemalike lesions and the liver necrosis were present has been the subject of further work. By means of column fractionation, gas chromatography, and mass spcctrometric techniques, tetra- and pentachlorodibenzofuran were identified as the toxic factors in the samples of Clophen A60 and Phenoclor DP6 (Vos et al., 1970). These results suggest that PCB's contaminated with polychlorodibenzofurans may also produce chloracne and associated lesions when applied to the skin of experimental animals. To test this possibility the present study was undertaken. METHODS The PCB samples, which contain an average of 60% chlorine, were obtained from Prodelccin France (Phenoclor DP6), Bayer in Germany (Clophen A60 Lot No. 912434) and Monsanto in the United States (Aroclor 1260 Lot No. AK-3). Adult female New Zealand rabbits, 5 mo old and weighing 2500-3050 g, were used in this study. They received pelleted food (Cunicon 1, Trouw and Co., Amsterdam) and water ad libitum. The animals were distributed at random into 4 groups of 4 animals, and were weighed weekly. An area of approximately 10x15 cm on the backs was clipped, the remaining hair was shaved with a shaving powder (90 g barium sulfide, 105 g zinc oxide, and ISO g starch), and the resulting small skin abrasions were allowed to heal for 3 days. DSW 025598 STLCOPCB4009553 DERMAL TOXICITY OF TECHNICAL PCB MIXTURES 619 The 3 PCB's were dissolved in isopropanol (118 mg PCB/mi). One ml of PCB solution or of isopropanol as a control was dropped daily, 5 times a wk, on an area of 5 x 10 cm by means of a calibrated syringe. In order to minimize eventual ingestion of the PCB, the animals were held for 7 hr in restraining boxes, after which time they were returned to their wire cages (2 animals per cage). In order to study the porphyrogenic action of the PCB preparations in the same rabbits, pooled samples of the feces of the 4 groups were analyzed weekly for copro porphyria and protoporphyrin contents (Rimington, 1961), using a Beckman spectro photometer. The animals which survived the 38-day test period (27 applications of 118 ing PCB) were (tilled. At necropsy, performed on all animals that died or were killed, a gross pathologic examination was carried out, including macroscopic examination in ultra violet light using red fluorescence as an indication of porphyria. In order to get infor mation on the fecal excretion of coproporphyrin and protoporphyrin of the individual animal, feces collected from the cecum were analyzed. Hematologic examination included hemoglobin and hematocrit determinations, leukocyte and differential leukocyte counts. Clinical chemical determinations were made of serum glutamic-oxaloacetic transaminase and glutamic-pyruvic transaminase (Rcitman and Frankel. 1957) using a Bcckman/Spinco 151 spectrocolorimeter. The weights of body, liver, kidneys, spleen, heart, and adrenals were recorded. The skin, liver, spleen, kidney, pancreas, ovary, heart, lung, thymus, mesenteric lymph node, stomach, small intestine, cecum, large intestine, and skeletal muscle were fixed in 10% buffered formalin and embedded in Paraplast. Sections (7 p. and 2 p) were stained with hematoxylin-eosin. For detailed histology, selected sections were stained with Peris, Azan, PAS, Zidil-Neelsen, Congo red, and thioflavine. Moreover, cryostat sections of formalin-fixed livers and kidneys were stained with Sudan black. Thiofiavinestained sections, unstained Paraplast-embedded sections, and unmounted cryostat sections of liver, kidney and small intestine were examined in a fluorescence microscope. Prompted by the observation of tubular dilatation in the kidneys in previous studies after PCB ingestion (McCune el al., 1962; Vos and Koeman, 1970), measurements were made on random samples of capsules of Bowman and glomeruli in 2-/i kidney sections. Measurements of one axis of the greatest diameter of both capsule and glomerulus were made of 25 renal corpuscles in a straight traverse across (he cortex of each kidney. The relative percentage of the diameter which corresponds to the space of Bowman, as an index of dilatation of the nephron, was calculated from the measure ments of each renal corpuscle. An additional experiment was made with the chromatographic fraction (column eluted with 25% dicthylclher in hexane) of the 3 PCB's, containing tetra- and pentachiorodibenzofuran in the case of Phenoclor DP6 and Clophen A60 (Vos ei al., 1970), to determine the dermal toxicity of this fraction. Aliquots of 600 mg of the 3 PCB mix tures were fractionated {Vos et al., 1970). The 25% diethylethcr-hexane fractions were evaporated and dissolved in 0.6 ml ofethanol. Four female albino rabbits (3 test animals and 1 control). 2 mo of age and weighing 1550-1750 g, were used as the experimental animals. A 0.2-ml portion of the ethanol solution was applied weekly, for 3 wk, to an area of 4 cm2 on the inside of the rabbit ear. The opposite ear served as a control. The iteftit DSW 025599 TABLE 1 COPROPORPHYRIN AND PROTOPORPHYRIN CONTENTS (f*G/G DRY WEIGHT) OF POOLED SAMPLES OF FECES from R.otvts Treated with PCB for 38 Days, and of Control Animals O Time of sampie (days) Phenoclor* Cop.'op orphyrin Clophen* Aroclor* Control Phenoclor1 Protoporphyrin Clophen* Aroclor* Control 0 7 14 22 29 , 36 0.6 2.8 2.0 3.3 3.1 8.7 0.5 0.3 0.4 11.3 5.4 5.0 7.7 4.2 2.9 0.8 24.3 28.5 17.0 6.7 1.6 1.1 1.0 26.3 10.2 13.4 7.1 17.2 11.2 1.0 13.5 18.3 29.4 4.7 4.5 3.2 1.4 10.1 19.2 21.2 7.5 9.8 10.0 1.7 24.8 19.5 27.7 8.8 * Significantly different from controls. P <. 0.05. according to the sign test. TABLE 2 COPROPORPHYRIN AND PROTOPORPHYRIN CONTENTS (ftG/G DRY WEIGHT) OF FECES OF KILLED RABBITS Treated with PCB for 38 Days, and of Control Animals* d \r. > V. Coproporphyrin Protoporphyrin c ln X Phenoclor Clophen Aroclor Control Phenoclor Clophen Aroclor Control o N Ul Oo' o Mean: 8.1 144.1 44.5* 58.0 63.7* 74.0* 33.2 30.5* 90.0* 56.9" 4.7 31.5 33.2 21.7 22.8* 2.9 2.5 2.9 4.9 3.3 28.8 126.9 38.2* 38.2 58.0" 39.0* 30.4 49.5* 94.0* 53.2" 25.0 23.3 12.5 32.4 23.3* 21.4 7.5 9.7 11.7 12.6 * Figures arc the contents of feces, collected from the cecum, of the individual animals. b Animals that died during the experiment, c Significantly different from controls, P < 0.05. 4 Significantly different from controls, P <. 0.025. v_ IlSlPS mm &&&&& suns v. Cjp.'. Ot, C..-.V-i Bvtf>Ls5Sfc 5*^ *'-`V >r'* DERMAL TOXICITY OF TECHNICAL PCD MIXTURES 621 animals were killed after 3 \vk. Body and liver weights were recorded. Sections of the car and the liver were stained with hematoxylin-eosin. Cryostat sections of the livers were also stained with Sudan black. Statistical significance, on a one-tail significance level, was determined using the Wilcoxon test for 2 unrelated samples (van der Waerden, 1957). RESULTS Experiment with Commercial PCB Preparations in Rabbits The skin, especially of the Clophen- and Phenoclor-treated animals, became reddened after 2 days of PCB treatment. This redness was more pronounced after 1 wk of tieatmcnt, together with clear desquamation of the external epidermal layers and reduced regrowth of hair. Thickening of the skin and prominent transverse wrinkles of the treated skin developed during the experiment. ,i *>.> '.., . S1? F?-?V,!. mi _ '.wy'f.tw Vi Fig. 1. Fluorescence of porphyrins under ultraviolet light in livers from PCB-Uealcd killed rabbits. /, Arcelor; 2. Clophen: Phenoclor; and 4, control liver. The results of the spcctrophotometric determinations of fecal coproporphyrin and protoporphyrin are given in Table 1. The test animals showed, in general, a gradual loss of weight. One Phenoclor- and 3 Clophen-lreated animals died, with severe weight loss, after 10, 31, 32, and 32 days, respectively. At necropsy, carried out on all animals, it was found that 1 animal in each of the 3 test groups showed definite edema formation, wilh fluid in the abdominal and thoracic cavity, subcutis, and pericardium. All test animals, except for 1 rabbit of the Clophen group that died after 32 days of treatment, showed fluorescence in ultraviolet light. Liver (Fig. 1) and bone fluorescence IS `a Wkw'S mm? ip ip mm. ?,i i ?>:-: v, ' . ' ~ -Z'Znr. i -j?+ |PM ' 5CSS'5'^^ S<#ii :c*iA fSP^i :'f ,v ',**!*-.33 |SS1 %3 mmm D Co ac o Nj \J) a Mo TABLE 3 Results of the Hematologic Examinations Made in Killed Rabbits Treated with PCB for 38 Days, and in Control Animals" Treated animals Controls Hemoglobin X {g/100mi) Hematocrit (%) Leukocytes (lO-Vmm3) SGOT (U) SGPT (U) 8 9.2 (6.1-12.7) 27.0 (20-44) 7.2 (3.3-1 LI)* 10.7 (7.0-14.5) 47.1 (18-98) 4 9.0 (8.2-10.0) 33.3 (27-38) 15.0 (9.5-26.2) 8.2 (6.5-10.0) 23.8 07-27) " Mean values; the range is given in parentheses. * Significantly difTcrent from controls, P < 0.025. TABLE 4 Organ Weights and Organ: Body Weight Ratios oe Killed RAimns Treated with PC 13 eok 38 Days, and oe Controls" Organ Weight (g) Compound N Body (g) Heart Liver Spleen Kidneys Adrenals Phcnoclor Clophen Aroclor Control 3 2300 (1900-2900) 4.74 (4.16-5.78) 133 (85-169) 0.72 (0.21-1.55) 14.0 (13.1-14.6) 0.31 (0.24-0.34) 1 2900 5.95 137 1.18 14.5 0.27 4 2312 (1950-2600)* 4.31 (3.88-5.00) 138 (106-715) 0.61 (0.33-0.92) 14.3 (12.7-16.5) 0.27 (0.23-0.37) 4 2912 (2700-3100) 5.09 (4.60-5.95) 124 (112-152) 0.64 (0.52-0.76) 13.5 (12.0-15.1) 0.37 (0.28-0.43) Organ:body weight ratio (g/1000 g body weight) Phcnoclor Clophen Aroclor Control 3 1 4 4 2.07 (1.98-2.25) 2.05 LS8 (1.63-2.22) 1.74 (1.58-2.02) 58.5 (44.8-80.5) 47.1 59.0 (45.0-82.6) 43.0 (37.8-56.2) 0.28 (0.10-0.53) 0.41 0.26 (0.17-0.42) 0.22 (0.18-0.25) ' Mean values; the range is shown in parentheses. ' Significantly different from controls, P < 0.025. 6.2 (5.0-6.9) 5.0 6.2 (5.2-6.7)* 4.7 (3.8-5.1) 0.13 (0.11-0.16; 0.09 0.12 (0.09-0.15) 0.13 (0.09-0.15) > v: o wrn 6/3 >'~A r.->r fm ;.r ; ' -J '.%.. >.:?>*. ' &. '<, ;' ' 41-. . VY^'i.'v^ M.v.\t.. DERMAL TOXICITY OF TECHNICAL PCB MIXTURES 623 was observed most frequently; fluorescence of incisors,' kidneys, small intestine, and subcutis was found in about half of the test animals. There was no fluorescence in the bone marrow. Fluorescent microscopic examination of cryostat sections of the livers revealed that porphyric animals had porphyrin diffusely distributed throughout the whole lobule. Small foci inside parenchymal cells, probably nuclei, revealed an even more intense red fluorescence. Kidney sections showed, when positive, a weak red fluorescence, which was found in the cytoplasm of the tubules in marrow and cortex. Fluorescence of granules (perhaps nuclei) in glomeruli was also noted. of the epidermal and follicular epithelium. Follicular plugging is prominent. Note the hyalin foci in the cytoplasm ofepidermal cells (/), and ihe resting (te)ogen) phase of the hair follicles (7). Hematoxylin and cosin. >46. Flo. 3. Norma! skin of back of a control animal treated with the solvent isopropanol showing epidci mis at / and dermis at 7. Several groups of hair follicles, cut somewhat tangentially to their long a xes, arc seen at groups of hair bulbs are seen at 4. Hematoxylin and eosin. x46. Coproporphvrin and protoporphyrin contents of feces collected from the cecum at necropsy are given in Table 2, The degree of excretion was not related to the in tensity of porphyrin fluorescence; for example, the Clophcn- and Aroclor-treatcd rabbits with a coproporphyrin excretion of 30.5 and 4.7 jug/g (Table 2), showed, respectively, no fluo rescence at all, and fluorescence of liver, bone, incisors, kidneys, small intestine, and subcutis. The results of the hematologic determinations are given in Table 3. The PCB-trcated animals of the 3 groups have been taken together since there were no differences be tween the groups. The loial number of leukocytes was significantly reduced in the test DSW 025603 624 VOS AND BERMS animals. Mean SGOT and SGPT values were increased, but the large variation in individual values and the small number of animals made the apparent differences statistically indistinguishable. The differential count of leukocytes was not significant!) altered. Organ weights and organrbody weight ratios, arc given in Table 4. A significant decrease in body weight and a significant increase in the relative weight of the kidneys was found in the Arocfor-trcated animals. Mean absolute and relative liver weights of the test animals were increased, but the variation was also very high. At microscopic examination of the skin, thickening of the skin due to hyperplasia and hyperkeratosis of the epidermal epithelium was apparent (Fig. 2). A pronounced hyperplasia and hyperkeratosis of the follicular epithelium was also present, with the formation of comedolike structures (hair follicles which become dilated and filled with a keratinous plug) (Fig. 2). These changes were found, in general, in all treated animals, but the Phenoclor-trcated animal which died after 10 days showed less advanced altera tions. Follicular changes in the Aroclor group were generally less deep; the hyperplasia and hyperkeratosis were more restricted to the upper part of the hair follicles. Thickening of the surface and follicular epithelium was accompanied by the presence of large quantities of keratohyalin granules in the stratum granulosum. In general, PAS-positive, diastase-resistant hyalin foci were seen in the cytoplasm of epidermal cells, located above the cutis papilla (Fig. 2). Dedifferentiation of the sebaceous gland tissue was found in primarily the Phenoclor and the Ciophcn groups. Hair follicles in the resting phase (telogen phase) were predominantly seen in the same 2 experimental groups (Fig. 2). The differences in response between Aroclor-treated rabbits on the one hand and the Clophen- and Phenoclor-treated rabbits on the other arc quantitative, not qualitative. No differences in response were found between the two latter groups. In control rabbits treated with the solvent isopropanol, these changes did not occur (Fig. 3). Hislopathology of the livers of the treated animals showed a considerable diversity in lesions (Figs. 4 and 5), among them centrolobular degeneration (degenerated cells with nuclear pycnosis, rhexis, and lysis), focal hydropic degeneration (foci of swollen cells with a light cytoplasm that did not stain with Sudan black and PAS, suggesting a pronounced cellular edema), focal necrosis (foci of necrotic cells that were found in various portions of the lobules), cenirolobular liver cell atrophy (atrophy of the cyto plasm ofliver parenchymal cells in the centrolobular area), cytoplasmic hyalin degenera tion (foci of intracellular eosinophilic hyalin material), and pigment in Kupfferand to a lesser degree in parenchyma! cells. The pigment stained positive with Zichl-Neelsen. PAS and Peris and showed a yellow-brownish fluorescence; these arc strong indications for ceroid (Porta and Hartroft, 1959J. The results of the microscopic examination are given in Table 5. In general, PCB-induced liver lesions were most pronounced in the Clophen group and least in the Aroclor group. Periportal fibrosis was a common finding in the treated animals, but it was not found in the controls. Since some coccidia were found in a bile duct in 1 liver, it is quite possible that the fibrosis has to be considered as a healed coccidiosis. Material positive to Sudan black was seen in some fibrotic portal triades. Kidney damage was found in all PCB-trcated animals, though with a wide individual variation. Hydropic degeneration (negative to PAS and Sudan black) of the convoluted DSW 025604 *. -v .. ' ! \ .,,/> .jfS| i 'i,'- . ,. STLCOPCB4009559 V \ i ` ' .......... ................................................................................................. DERMAL TOXICITY OF TECHNICAL PCH MIXTURES 625 r. .j 1 k -*,-V*-i:n. 'V?vVrffcV-', KpSSPunifVl *VI* ' lV* ?< Rv-.VV-' -rvV,,'1'> 1'. V: * *,<!*; +'xfjgV'Trr Tv J-r*-.:- `-`V Ie * -i ?*j ,- ^ #. y*?-* -v .. -. l Fio. 4. Liver of a Clophen-trcated rabbit that died after 31 days of treatmenl showing (he portal triadc at /, and degeneration and atrophy of the cytoplasm of the hepalocyles in the centrolobular area at 2. Hematoxylin and eosin. xl50. Fto, 5. Higher magnification of the liver shown in Fig. 4. /: central vein; 2: hydropic cell; 3: focal cytoplasmic hyalin degeneration. Note also the enlarged nuclei and the karyopycnosis. Hematoxylin and eosin, *375. STLCOPCB4009560 STLCOPCB4009561 TABLE 5 Pathologic Findings in Livers of Rabbits Treated with PCB for 38 Days, and of Control Animals4 Phenocior Clophen Arocior Control Fatty degeneration Ccntrolobular loss of glycogen Ccntrolobular degeneration Focal hydropic degeneration Focal necrosis Centrolobular liver cell atrophy Enlargement of nuclei D tf) Cytoplasmic hyalin degeneration Ceroid pigment in Kupffer eel s O Periportal fibrosis N tft O' * Severity of lesion: l -- slight, 2 = marked. o * Animals that died during Ihc experiment. O' b 1 2 I-- 2 2 1-- !-- --1 b 1 II 21 I2 12 2 _22 2222 21 21 2222 ? 2 I iI1 f2 - -- 1 2---------------- -111 -- ---- ---- I 2 -- ---- 1--12 -- ---- -212 I-- 1 I -- ---- -- ---- olA > '/ c cnm 2v. HI- - ?''-'-3-O^v" Sy-'VW-- kf-*%? "fi-A * Vrv V, - E vvv--.* - a:rT v * A-.; STLCOPCB4009562 628 VOS AM) BEEMS TABLE 6 Diameter Measurements (/i) of Renal Corpuscles or Killed Raudits Treated with PCB for 38 Days, and of Control Animals" Capsule of N Bowman Glomerulus Relative % of the diameter corresponding to the space of Bowman Treated animals 8 Controls 4 94.5 (76.4-105.8) 82.2 (69.4-96.2) 91.5 (86.0- 96.5) 85.2 (79.4-90.4) 12.5 (8.1-22.1)'' 6.5 (5.0- 7.9) * Mean values; the range is shown in parentheses. * Significantly different from controls, P < 0.005. The tubular lesions explain the results of the diameter measurements of the renal cor puscles (Table 6). Hemosiderin was found in the splenic macrophages of the test animals in large quantities than in the controls. Moreover, a reduction in the number of germinal cen ters in the spleens as well as in the lymph nodes was found in the PCB groups. Atroph; of the cortex of the thymus was a common finding in the PCB-treated animals. No evi dence of PCB-induced damage was found in the sections of the other organs. Experiment with the Dierhyleilier-Hexane Fractions on the Ear Skin of Rabbits The four animals did not differ in weight gain during the test period. After two weekl topical applications of the fraction from the 3 PCB samples, follicular keratosis wa seen in both rabbits treated with Ihe fraction of Phenoclor and Clophen. After 3 wk when the animals were killed, the changes were more apparent (Fig. 8). The polar frac lion of Aroclor did not give rise to gross changes. Fig. 8. Chloracne-like lesions on the inside of a rabbii's cur skin treated weekly, for 3 weeks, with tl 25" ,, dieihylelher in hexane fraction from Clophen. Note the follicular keratosis in the treated ear ( as compared with the control (2). DSW 025608 STLCOPCB4009563  S-K+. ..y. '' ' '^' *'''11 * T--' '-'^` '.nAA,Mttw-^fcfa`wwViiV''Yk*> tiViVi i*-..vy *5f'.iVsi'V4;4 * A' ~ ... iy,'i. .i ^-v-,- DERMAL TOXIC !! Y OF TECHNICAL PCD MIXTURES 629 Fig. 9. Response of the inside of the rabbit's ear after topical application of the 2S% dicthylctherbesanc fractions from technical polychlorinated biphenyls as seen in hematoxylin and eosin sections, '60. (a) Skin of control animal treated with ethanol. Note the hair follicle at 1, sebaceous gland tissue at 2, and cartilage at 3. (b) Ear skin of the animal treated with the fraction from Aroclor. Some hyperplasia and hyperkeratosis of the follicular epilhelium can be seen. (Cl Ear skin of the rabbit treated with the fraction from Clophen, Considerable hyperplasia and hyperkeratosis of the follicular epilhelium. (d) Ear skin of the rabbit treated with the fraction from Phenoclor, Part of a section that shows themost severe lesion. The gravity of the response was in general the same as seen in Fig. 9c. Note the cystic di lated hair follicle with prominent hyperplasia and hyperkeratosis of the follicular and epidermal epithelium. i , r. . STLCOPCB4009564 630 VOS AND BKEMS Histologically, hyperplasia and hyperkeratosis of the follicular epithelium with 1 presence of many keratohyalin granules (Fig. 10) were evident in the ear skin ofanim; treated with the Phenoclor and Clophen fractions (Fig. 9c and d). This comedoli formation was topically most pronounced in the rabbit treated with the Phenoclor fra tion (Fig. 9d). Sebaceous gland tissue was decreased by treatment with both fractio (Fig. 10). Only mild hyperplasia and hyperkeratosis of the follicular epithelium w noted in the animal treated with the fraction of Aroclor (Fig. 9b). Hyperplasia ai hyperkeratosis of the epidermal epithelium were also seen in the rabbits treated with t Phenoclor (Fig. 9d) and Clophen fractions but not in the animal treated with t Aroclor fraction. Fjg. 10. Higher magnification of the ear skin shown in Fig. 9d. Hyperplasia and hyperkeratosis of l! follicular epithelium with the presence of many keratohy alin grannies (/). Note also a diminution (d-. differentiation) of the sebaceous gland tissue (2). Hematoxylin and eosin. *375. No red fluorescence was recorded at examination of the total body and of liver set tions in ultraviolet light. Liver sections of the test animals showed centrolobula vacuolisation, which stained positive with Sudan black. ' DISCUSSION An increase in fecal excretion of coproporphyrin and protoporphyrin induced b; PCB is as readily established in the rabbit (Table 1) as in the chicken (Vos and Kocman 1970). Absence of porphyria in animals treated with the toxic fractions indicates tha PCB itself is porphyrogenic. Liver and kidney lesions, except the loss of glycogen, the enlargement of nuclei, tin ceroid deposits, the periportal fibrosis, and the dilatation of the space of Bowman DSW 025610 STLCOPCB4009565 DERMAL TOXICITY OF TECHNICAL PCB MIXTURES 631 confirm the observations of Miller (1944). The lymphopenia (Table 3), the reduced number of germinal centers in the spleens and lymph nodes, the atrophy of the cortex of the thymus, and the atrophy of the white pulp and lymphoid foci found in the spleens of chicks fed PCB (Vos and Koeman, 1970) are strong indications for an immunosup pressive ciTect. This may need further study, particularly with regard to the extent to which these observations arc due to stress (release of glucocorticoids). It can be concluded that dermal application of PCB mixtures causes, besides the effect on the skin, systemic lesions of liver, kidneys, and lymphoid tissue. From the response of the back skin and the liver of the rabbit to the 3 PCB mixtures, and from the response of the ear to the 25% diethylether-hcxane fractions, it can be concluded that there are definite quantitative differences in toxicity, at least between the samples that were used in the present and the prior studies. The extent to which these samples are representative of the normal commercial output has not been studied. In a prior study (Vos el al., 1970), it appeared that the fraction of Clophcn A60 was respon sible for the toxicity of the PCB mixture, which also can be assumed to be true in the case of our samples of Phenoclor and Aroclor. From the response of the ear skin to the I fractions it can be concluded that the presence of tetra- and pentachlorodibenzofuran I in Clophen A60 and Phenoclor DP6, found by mass spectrometry (Vos et al, 1970), is definitely proved. The presence of a smaller quantity of toxic impurity in Aroclor, when ( compared with the Clophen and Phenoclor samples, is likely, considering the hydro pericardium found in chicks fed with Aroclor (Flick et al., 1965) and also considering the present results. Preparation of pure PCB is therefore necessary for a comparative study. UiKontaminated PCB should be available. These findings suggest that chloracne lesions induced by chlorinated naphthalenes and chick-cdema-like lesions in chickens (Pudelkiewicz et al, 1959) may also be due to toxic impurities, as in the case of PCB's and chlorinated phenols (Bauer et al., 1961). Coal tar, which is an important source for naphthalene, may contain traces of dibenzofuran (Fiescr and Fieser, 1956). Chlorination of technical naphthalene could then also produce chlorinated dibenzofuran. Since PCB is a porphyrogenic chemical, it is possible that the skin lesions in man due to PCB are a combination of chloracne and acquired porphyria cutanea tarda. Bleiberg ft al. (1964) found an etiologic relationship between the chloracne in workers engaged in the manufacture of dichlorophenol and trichlorophenol and acquired porphyria cutanea tarda. From (he present and prior studies (Vos and Koeman, 1970; Vos el al., 1970) it is concluded that it is important to take into account, in the evaluation of toxicity data of PCB, the possibility that PCB samples may differ in an important respect: the pre sence of toxic impurities. The interpretation of residue data in wildlife is very compli cated for this reason. ACKNOWLEDGMENTS The authors arc grateful to Mr. H. Baart de la Faille, dermatologist (Department of Dermatology, Hospital of the State University, Utrecht) and Dr. P. Kanaar, dermatologist (Department of Dermatology, Hospital of the State University, Leiden) for their help in the evaluation of the skin lesions. DSW 025611 STLCOPCB4009566  *'' C \ r."cSi../: .,,.:vV > ,v.V' ^ '.` i K-^v.C \,.V *v-v- r $ V'L 'j P@W| SSS-Ml :%. , *{*>:. $/ *J` r *--. 'iV,; ,,*. >V. - *Vj&-Y *' `i: '> * .*V>> , > ' <r.*4 V- %; Vv vY ' ` 'TL >. 632 VOS AND BHF.MS REFERENCES Adams, E. M., Irish, D. D., Spencer, H, C., and Ronve, V. K. (1941). The response of rabbi skin to compounds reported to have caused acncform dermatitis. hid. Med., hid. Hyg. Serf 2, 1-4. Bauer, H., Schulz, K. H., and Spiegelderg, U. (1961). Bcrufiichc Vcrgiftungen bei de Herslellung von Chlorphenol-Verbindungen. Arch. Gewerbepathoi. Gewerbehyg. 18, 538 555. Behrbohm, P. (1959). Ober Gefahren beim Umgang mil chlorierlen Phenolen. Dent. Gcsund heitsw. 14, 614-619. Bennett, G. A., Drinker, C. K., and Warren, M. F. (1938). Morphologic changes in tin livers of rats resulting from exposure to certain chlorinated hydrocarbons. J. Ind. Hyg Toxicol. 20, 97-123, Bleiberg, J., Wallen, M,, Brodkin, R., and Applebaum, I. L. (1964). Industrially ncquirtt porphyria. Arch. Dermatol. 89, 793-797. Braun, W, (1955). Chlorakne, Monographien zur Zeitschrift "BcrujsdemiatosenBand I. Editio Canter, Aulendorff, Wiirtemberg, Germany. Drinker, C. K., Warren, M. F., and Bennett, G. A. (1937). The problem of possible systemii effects from certain chlorinated hydrocarbons. J. hid. Hyg. Toxicol. 19, 283-311. Duke, T. W., Lowe, J. I., and Wilson, A. J. (1970), A polychlorinated biphenyl (Aroclor 1254 in the water, sediment, and biota of Escambia Bay, Florida. Bid!. Environ. Contamiii. Toxicol. 5, 171-180. Fieseh, L. F., and Fieser, M. (1956). Organic Chemistry, 3rd ed., pp. 527-531. Reinhold, New York. Flick, D. F., O'Df.ll, R. G., and Childs, V. A. (1965). Studies of the chick edema disease. 3. Similarity of symptoms produced by chlorinated biphenyl. Poultry Sci. 44, 1460-1465. Greenburg, L., Mayers, M. R., and Smith, A, R. (1939). The systemic effects resulting from exposure to certain chlorinated hydrocarbons. A hid. Hyg. Toxicol. 21, 29-38. Hambrick, G. W. (1957). The effect of substituted naphthalenes on the pilosebaceous appa ratus of rabbit and man. J. Invest. Dermatol. 28, 89-103. Hoemann. H. T. (1957). Neuerc Erfahrungen mit hochtoxischen Chlorkohlcnwassersloffen. Nuunytt-Schmiedebergs Arch. Exp. Pathol. PharnuikoL 232, 228-230. Holden, A. V., and Mausden, K. (1967). Organochlorinc pesticides in seals and porpoises. Nature (London) 216, 1274-1276. Holmes, D. C., Simmons, J. H., and Tatton, J. O'G. (1967). Chlorinated hydrocarbons in British wildlife. Nature {London) 216, 227-229. Jensen, S. (1966). Report of a new chemical hazard. New Sci. 32, 612. Jensen, S., Johnels, A. G., Olsson, M., and Otterund, G. (1969). DDT and PCB in marine animals from Swedish waters. Nature (London) 224, 247-250. Jones, J. W., and Alden, H. S. (1936). An acneform dermatergosis. Arch. Dermatol. Svphilol. 33, 1022-1034. ' Koeman, J. H., Oskamp, A. A. G., Veen, J., Brouwer, E., Rooth, J.. Zwart, P., v. d. Broek, E., and van Genderen, H. (1967). Insecticides as a factor in the mortality of the sandwich tern {Sterna sandvicensis). Meded. Rijksfac. Landhouwiselensch. Gent 32, 841-854. Koeman, J. H., ten Noever de Brauw, M. C., and de Vos, R. H. (1969). Chlorinated biphenyls in fish, mussels and birds from the river Rhine and the Netherlands coastal area. Nature {London) 221, 1126-1128. McCune, E. L., Savage, J. E., and O'Dell, B. L. (1962). Hydropericardium and ascites in chicks fed a chlorinated hydrocarbon. Poultry Sci. 41, 295-299. Miller, J. \Y. (1944). Pathologic changes in animals exposed to a commercial chlorinated diphenyl. Pub!. Health Rep. 59, 1085-1093. Porta, E. A., and Hartroit, W. S. (1969). Pigments in Pathology (M. Wolman, cd ), p. 210. Academic Press, New York. Prestt, I., Jefferies, D. J., and Moore, N. W. (1970). Polychlorinated biphenyls in wild birds in Britain and their avian toxicity. Environ. Pollution 1, 3-26. DSW 025612 STLCOPCB4009567 DERMAL TOXiCITY OF TECHNICAL PCB MIXTURES 633 I'i du.kiewicz, NY. J,, Boucher, R. V., Callexoach, E. \V,, and Miller, R. C. (1959). Some physiological responses of New Hampshire chickens to a mixture of penia- and hexachloronaphthalenes. Poultry Sci. 38, 424-430. ' Redman, S., and Frankel, S, (1957). A colorimetric method for the determination of scrum glutamic oxaloacetic and glutamic pyruvic transaminase. Amer. J. Clin. Pathol. 28, 56-63. Rimington, C. (1961). Quantitative determination of porphobilinogen and porphyrins in urine and faeces. Broadsheet No. 36 of the Association of Clinical Pathologists. Copies may be obtained from Dr. R. B. H. Tierney, Pathology Laboratory, 75, Boutport Street, Barnstupie, Devon. Risebrough, R, W., Rieche, P., Peakai l, D. B., Herman, S. G., and Kirven, M. N. (1968). Polychlorinated biphenyls in the global ecosystem. Mature (London) 220, 1098-1102. Schwartz, L. (1936). Dermatitis from synthetic resins and waxes. Amer. J. PM. Health 26, 586-592. Shelley, W. B., and Kligman, A. M. (1957). The experimental production of acne by peniaand hcxachloronaplultalcnes. A.M.A. Arch. Dermatol. 75, 689-695. Strauss, N. (1944). Hepatotoxic effects following occupational exposure to Halowax (chlori nated hydrocarbons). Rev. Gastroenterol. II, 381-396. van per NVaerden, B, L. (1957). Muthematische Statistik, pp, 269-284. Springer, Berlin, , yon Wedel. H., Holla, W. A., and Denton, J. (1943). Observations on the toxic effects result* . ing from exposure to chlorinated naphthalene and chlorinated phenyls with suggestions for i prevention. Rubber Age iMew York) 53, 419-426. ' Vos. J. G., and Koeman, J. H. (1970). Comparative toxicologic study with polychlorinated biphenyls in chickens wiih special reference to porphyria, edema formation, liver necrosis, I and tissue residues. Toxicol. Appl. Pharmacol., 17, 656-668. i Vos. J. G., Koeman, J. H., van de.r Maas, H. L,, ten Nocver de Brauw, M. C., and deVos, ( R. H. (1970). Identification and toxicological evaluation of chlorinated dibenzofuran and ) chlorinated naphthalene in two commercial polychlorinaied biphenyls. Food Cosrnct. Toxicol. 8, 625- 633. i STLCOPCB4009568