Document Vj5EbBo0YMkBmZKKy0xy8NB8g

Adenofibrosis in the Rat Liver With Persistence of Polychlorinated Biphenyls in Adipose Tissue Rcnatc I). Kimbrough, AID; Ralph E. Linder; Virlyn W, Burse; Ralph W. Jennings, Chamblee, Ga RvV. J- vy* < Fifty mate Sherman strain rats were fed 500 ppm of a polychlorinated biphenyl (PCB) (Aroclor 1254) for six months. Five each were hilled zero, one. two, three, four, six, eight, and ten months after exposure to Aroclor had ceased. The livers of these rats were examined by light and electron mi croscopy. Liver lesions persisted although exposure to PCBs ceased. Ten months after exposure ceased, 1,192 ppm PCBs were still present in the rats' adipose tissue and 22.65 ppm in the rat livers. Aroclor patterns found in the tissues by electron capture gas chromatography dif fered from patterns of dietary Aroclors. Mass spectral analysis of liver and adipose tissue revealed three major Aroclor compo nents with masses of 324, 358, and 392. These contained isotopic clusters indica tive of the presence of Cl,, Cl,, and Cl,, respectively. In a previously reported 6tudy' Sherman strain rats were fed poly chlorinated biphenyls (PCBs) (Aroclor 1254 and Aroclor 1260). The rats developed characteristic mor phological changes in the liver. These changes consisted of hypertrophy of the liver cells, a brown pigment in Kupffer cells, lipid accumulation in the cytoplasm of hepatocytes and, at the higher dietary levels, adenofi brosis. Because of the wide distribu tion of PCBs in the environment* and Submitted for publication March 8* 1973; ac cepted April )9. From the Environmental Protection Agency, Bio-Eftccta Branch, Chamblee, Ga. Reprint requests to ]65 Avery Dr NE, Atlanta 30309 (Dr. Kimbrough!. their persistence, the hepatic lesion wa6 studied further. The present )l study was undertaken to see whether, once exposure to PCBs was stopped, )' . v' the morphological changes produced fig 3.-Sechin in the liver might disappear. nagnification 5 Materials and Methods A total of 60 male weanling, specificpathogen-free, random-bred Sherman etrnin rats were tagged individually and group-caged, ten rats per cage. They were given Aroclor 1254 in their diet for six months. The dose of 500 ppm, which ii equivalent to an average doily intake of 36.4 mg/kg body weight,' was chosen be cause in the previous study all ten exposed male rats developed adenofibrosis within * period of eight months. The Aroclor wax mixed into ground laboratory chow as pre viously described.' After six months Of fig 4. - Port'.'n :i1rate-uranyl ci ?V ' fig l.-Area of adenofibrosis in liver section consisting primarily of duels surrounded by flattened epithelium. Ducts are surrounded by little fibrosis (hematoxylin-eosin, original magnification x 150). fig 2. - Rat liver section showing cluster of small columnar cells resem bling pancreatic tissue adjacent to blood vessel. Hepatocytes surround ing lesion are vacuolated (hematoxylin-eosin, original magnification x 175). ft L- 390 Arch Environ Health/Vol 27, Dec 1973 Adenofibrosis/Kimbrough et al Arch Environ DSW 025591 STLCOPCB4009546 hepatic lesion Tin' present (o see whether, s was stopped, npes produced ipear. Methods Fig 4. - Portion of hepatocytc with tubules of endoplasmic reticulum (ER) arranged in disorderly fashion. N represents nucleus; M, mitochondria (lead I[inline, specific- ' rale-uranyl acetate, original magnification x 20,235). tired ,Sherman ndiviiliiMlly and I'ujje. They were icir diet, for ix ' ppm. which in daily intake of was el io.se n bev all ten exposed fibrosis within * i'lie AMielor was ory eli nv as pre air months of nnnor cells resem.locytc i. surroundnal m. tj;r iif ic a (too imbrough et al 'JArch Environ Health/Vol 27, Dec 1973 1 v. I i l nnupaM) Adenofibrosis/Kimbrough et al 391 OSW 025592 vwctWffo1 wi!11, 91 ntyi wu'tfnjn STLCOPCB4009547 501I 40 30 2C *o ,,L 0 TO&0 Tig 5.- Border of area of adenofibrosis in liver. Note round clusters of dark granules intermixed with web like clusters of vacuoles. These formations illustrate appearance o( brown pigment (BP). C represents collagen; LC, liver cell (lead citrate-uranyl acetate, original magnification x 20,235). receiving the experimental diet, the rats received no more Aroclor and were fed plain Inhornlory chow. At this lime five nils were killed. Additional killings of five randomly selected ruts were made one, two, three, tour, six, eight, and ten months after ex|xixure to PCH had been discontin ued. At autopsy the livers were removed, weighed, and the tissue fixed in pbosphale-huifered 5% glutaraldchyde for three hours for electron microscopic exam ination, post-fixed in 1% osmium tetroxide, embedded in mnraglass, sectioned with n glass knife, and stained with lead citrate and urnnyl acetate. Tissue for the light microscope was fixed in 4% buffered formaldehyde solution and stained with hematoxylin and eosin. Frozen tissue of selected formalin-fixed hepatic tissue was cut on the cryostat and stnined with oil red 0 Ui demonstrate lipids. At the time of the last sacrifice, ten months after exposure to I'CU-contfiining diets had been discontin ued, adipose tissue and liver from seven previously exposed rots and the adipose tissue and liver from two rats of the same nge that hud been fed only plain chow were analyzed for the presence of PCBs by electron capture (EC) gas chromatography according to methods described by Curley et al.5 In order lo determine which Aroclor constituents were of such chemical stabili ty that they remuined at a measurable residual level ten months after dietary in take had ceased, adipose and hepatic com posites were analyzed by gas chromatog raphy-mass spectrometry. The conditions for mass spectral analysis were as follows: the mass spectrometer was interfaced with a Rns chromatograph and equipped with n muss mnrkcr accurate within 0.3 mass units. The gas chromatographic column temperature was 220 C; flash healer, 235 C; glass-coiled column, 3.05 m x 0.64 cm; 1.5% OV-17/1.95% QF-1 on 60/80 mesh chromosorb "W" (high performance, acid washed, dimethyldichlorosilane treated), helium pressure; 0.84 kg/sq cm; flow rate (rotameter setting 3), 45 ml/min; separa tor temperature, 225 C; source energy, 70 ev; accelerating voltage 3.6 kv; trap cur rent 60/xamp; box current 50p.amp, and leak current 8fiamp. Results During the course of the experi ment three rats died. The livers of the rats that were killed at the time expo sure to Aroclor 1254 was discontin ued were enlarged, with average weights of 6.32% of the body weight. When the experimental diets had been discontinued for four months, the liver made up only 3.6% of the body weight, and after eight months recovery the average weight amounted to 3.34% of the average body weight. The fraction of the body weight for the liver in control adult male rats ranges from 2.44% to 2.59%.' Forty livers were studied micros copically. The hepatocytes were en larged in 39 livers and all 40 livers showed either vacuolated or foamy cytoplasm because of increased lipid accumulation. Cytoplasmic inclu sions similar to those described previously4 were observed in the liv ers of 24 rats. Adenofibrosis, which in many instances was very extensive, was present in 36 of the 40 livers, and a brown pigment was noted in the Kuplfer cells and other macrophages in 25 livers. Of the four rats that did not develop adenofibrosis, one each was killed two and six months after the poisoned diets were discontinued. The other two rats without adenofi brosis were killed ten months after onset of exposure. The adenofibrosis ranged in size from grayish-white le sions measuring 0.5 lo 1 cm to areas of 5 cm with a pitted surface. Quite often the lesion was multicentric. The adenofibrosis* is illustrated in Fig 1 and 2. In this study the appearance of 50 UJ GO 40z o a JOtf) uj 20 cc )o4 0 7T * Fig6.-G-i . extract ol . j i>' the lesit'i i extensiv . lesions v glandulm cells lh:> rounded larger 1 and con:. ducts v.li.1 ' thelial 392 Arch Environ Health/Vol 27, Dec 1973 Adenofibrosis/Kimbrough et el Arch En-'. DSW 025593 STLCOPCB4009548 U average 1, ni > ll 1 be boB\ I control adult Ini 2.4 4tt to jidiril micros- lies were en- nil -10 livers led nr foamy iicvensed lipid ismic inclu de described ,.d in the livisis. which in ,y extensive, l() livers,and noted in ttu' nncrophagcs iiitS that did Fig 6. - Gas chromatographic traces of Aroctor 1254 standard (A), hepatic tissue extract (B), and is, one each wtract of adipose tissue (C). Response expressed as percentages, with 100% "1.0 mv. nontlis alter liacontinued. the lesion, particularly when it was out ndenofi- extensive, varied a great deal. Small umths niter lesions consisted predominantly of deno fibrosis glandular pale-staining epithelial sh-whitc le- cells that formed ducts1 and were surciTi to areas 'rounded by very little fibrosis. The markedly dilated. The dilated ducts contained mucus or necrotic debris, and at times formed small cysts mea suring up to 0.5 cm in diameter. Small dilated ducts were lined by co lumnar epithelium, and larger ones rlhce. Qui`e ' larger lesions had extensive fibrosis centric. The and contained collagen, and often the by flattened epithelium (Fig 1). In areas of extensive fibrosis, and i-d in Fig 1 ' ducts which were formed by the epi i|H!nutceof | thelial palc-stnining cells were also adjacent to blood vessels in areas of normal hepatic parenchymu (Fig2), clusters of small glandular cells were seen in 15 of the livers which resembled the epithelium observed normally in the pancreas. These small clusiersofpanCrealic-like tissue have not previously been described in the livers of rats or of other species. Special stains for esterase, esterase with sodium taurocholate, and tryp tophan showed a positive reaction indicative of salivary gland tissue. The significance of these cells is pres ently not understood. Normally tis sue of a pancreatic type is not ob served in tile livers of our rats. Clus ters of small columnar epithelial cells resembling pancreatic tissue also occurred in portal triads without be ing surrounded by typical adenofihrosis. In the present study, one to two rats at each killing showed these small columnar cells with highly aci dophilic cytoplasm and a small, round nucleus. In the older animals, the fibrosis was quite extensive in some instances and the glandular proliferation was present mainly at the periphery of the lesion. Signs of regression of the lesion were not noted, except for the disappearance or epithelial cells from the center of the lesion. Furthermore the lipid accumulation and the hyper trophy of the hepatic cells remained constant throughout the study. Ex tensive areas of necrosis were ob served in three of the 40 livers. In addition to the change described previously,' electron microscopic examinations of sections of liver tis sue showed numerous small lipid vacuoles in the cytoplasm of many hepatocytes, particularly in the latter part of the recovery phase (Fig 3!. The smooth and rough endoplasmic retic ulum was arranged in a disorderly fashion (Fig 4). Many of the mito chondria were atypical and surround ed by double membranes. The electron microscopic appear ance of the brown pigment present in the macrophages and Kupfier cells is illustrated in Fig 5. Large clusters of dark granular material, intermixed with lipid vacuoles, were observed, as well as round clusters of granular material about the size of mitochon dria. In our previous studies it was found thnt at least some of this pig- ibrough et si Arch Environ Health/Voi 27, Dec 1973 Adenoiibrosis/Kimbrough et al 393 DSW 025594 STLCOPCB4009549 menl contained iron1 but in addition high concentrations of uroporphyrin <75% 8-carboxyprophyrins and 25% 7-carboxyprophyrins) were found in the livers,1' which could account for some of the pigment. In addition to these findings, oil red O stain demon strated granules of red-staining lipid material in the area of the brown pigment. Since the morphological changes in the liver did not disappear, the re maining seven rats were killed after a ten-month recovery phase and their adipose tissue and livers analyzed for Aroclor. In adipose tissue, the concen trations of PCBs ranged from 924 ppm to 1,688 ppm, with an arithmetic moan of 1,192 ppm. The concentra tions in the liver ranged from 17.30 ppm to 26.24 ppm, with an arithmetic mean of 22.65 ppm. The PCB levels in control adipose and liver tissues were <1.0 ppm. The gas chromatogram obtained from the EC analysis of adipose tissue did not differ from that of liver, but was significantly different from standard Aroclor 1254. An average of seven peaks was observed as a resi due in the tissues. Under these gas chromatographic conditions, at least 18 peaks have been observed in the standard material (Fig 6). The differ ence is gas chromatographic patterns between stored and ingested Aroclor lias been observed before in this labo ratory and by other researchers in the field."-7 ' The total ion current (TIC) chro matogram for the liver tissue com posite is shown in Fig 7. Mass spec tral analysis revealed the presence of only three major Aroclor components with masses of 324, 358, and 392. These components contained isotopic clusters indicative or the presence of Cl,, Cl,,, and ClT, respectively. Adi pose tissue analysis by mass spec trometry revealed the same patterns. Although only three different mo lecular components were present, examination of the EC and TIC chromatograms coupled with mea sured intensities from multiple scan ning of eluting peaks indicates that stereoisomers of these three mole cules are obviously present. Peaks 1 to 3 (Fig 7) did not contain any ppm eight tine bee -. sue a Tli" t alt'-: san . ppm nunit pi-nth adi-iv ail-ni Fig 7.-Total ion current chromatogram of sample of composite liver. Response expressed as percentages, with 100%= 10.0 mv. chlorinated moieties, while other peaks observed in the liver TIC con tained the following molecular ions: peak 4 (324); peaks 5 to*-7 (392 and 358); and peaks 8 to 9 (392). Comment The adenofibrosis of the liver pro duced in this study and the one pre viously reported1 is the result of Aro clor consumption. Whether the Aro clor itself, or a contaminant such as a chlorinated dibenzofuran, is responsi ble still needs to be resolved. A study of the lesion Bhows that epithelial proliferation occurs in the periphery, while the central portion of the lesions contains primarily fibrous tissue. The present study does not completely resolve the question of reversibility of this lesion since the storage levels of PCBs in adipose tis sue and liver may have been suffi ciently high to stimulate continued growth of the altered hepatic tissue. The levels in the liver suggest that a constant low-level turnover of PCBs takes place in the rats. In a previous study, Curley et aP found levels of 10,000 ppm Aroclor 1254 in adipose tissue of rats that had been fed 500 ol ''<> en o h dele: M. pl-.- : o..-l. SCI VI Sll 111 o!'!!l In ph liydi u rin by f re! i til-,: In -X; 125. [-1 til ici ( as a P proc Wh ile.chit rat. n.-.h mn J-..b ml !' i, I V, . I: H' 394 Arch Environ Health/Vol 27, Dec 1973 Adenofibrosis/Kimbrough et ai DSW 025595 STLCOPCB4009550 i>m or the mixture in their diet for 1ight months. This finding indicates hat a great deal of the Aroclors had een mobilized from the adipose tis- ue and excreted in the present study. The storage levels found ten months liter onset of recovery are in the lame range of those of rats fed 100 Jppm PCBs in their diet for eight months/' The dietary level of 100 ppm produces a much lower incidence of idenofibrosis,1 and it is possible that sdenofibrosis represents a permanent 'lesion, particularly since the percent of body weight of the livers had de creased. The Aroclors found in the tissues 'differed from those fed in the diet. Metabolism of lower chlorinated bi phenyls is considered as a possible explanation for the differences ob served analytically between con sumed and stored PCB. Earlier work efBlock and Cornish* has shown that biphenyls and chlorobiphenyls can be hydroxylated and excreted in the urine of rabbits. A more recent study by Hutzinger et al* indicates that the rat is capable of hydroxylating mono-, di-, and tetrachlorobiphenyls, but not hexachlorobiplienyls. Since Aroclor 1254 averages five chlorines per mol ecule, it is conceivable that the heav ier chlorinated moieties could remain as a residue. P-dimethylaminoazobenzene also "I--------r 13 14 produces adenofibrosis. Edwards and White"' and Bennett et a!" described the lesion in rats when a mixture of . chlorinated naphthalenes and chlori- nnted diphenyls were fed. Unfortu- e evptessed as nately, the finding was not well de- , scribed in Bennett et al's report, but occurs in the Edwards and White mentioned Ben nfr.il portion nett et al's finding as another exam ns primarily ' pie of the induction of adenofibrosis nl .-i ucl.v does by chemicals in an excellent descrip " ; lestionof .oil since the i) mlipose tisc l oon suffiite .ontinued 'pn'ic tissue, tion of the lesion. A critical review of the histopathogenesis of this lesion was made by Stewart and Snell12. The presence of the brown pigment in Kupffer cells and macrophages is of interest and was to a lesser extent iggcsl thala ''. IT of PCBb ; ii a previous - aid levels of 1 in udipoBe ' on fed 500 also observed in the earlier study.1 Electron microscopic examination of the pigment shows that particularly the large clusters of it resemble very closely illustrations of ceroid pig ment.13 Experimentally ceroid pig ment can be produced by hemorrhage into fatty tissues rich in unsaturated fatty acids, particularly in the ab sence of an antioxidant (vitamin E deficiency). Ceroid pigment is com monly referred to as lipofuscin. These lipid pigments are at times associated with hemosiderin and may contain iron-positive granules as in the pres ent case.1 Ceroid granules were also described by Edwards and White in the livers of rats fed butter yellow. Bennett et al also observed a brown pigment , in rats fed a mixture of chlorinated naphthalenes and chlori nated biphenyls. These authors claimed that it failed to stain in a manner characteristic of either hemo siderin or hemofuscin. The porphy rins demonstrated in the livers of our rats fed Aroclor may also contribute to the overall pigment deposits. Cer oid pigment increases in people with age and has therefore been called "Abnutzungspigment" by the Ger mans. Attempts have been made to relate it to senility. It has been associ ated with neuronal ceroid-lipofusci nosis (Batten disease) in man. In minks, a disease called yellow fat disease11 is associated with the accu mulation of excessive amounts of cer oid in adipose tissue. The accumula tion of ceroid in mink can be prevent ed by addition of vitamin E to the diet. It occurs when the food rations cunlain high percentages of fish scrap. However, Gorham et al suggest that other factors may also be in volved, since high dietary percent ages of fish scrap did not promote growth as well as the stock ration, even when vitamin E was added to the diet. These observations raise the question whether high doses of vita min E, and possibly the addition of choline to the diet, could prevent or minimize the observed effects of PCBs on the liver of our rats. Richard L. Moore aesUtod with the animat tests, Linda W. Anderson assisted with preparation and interpretation of the eloelromicrogniphs. Annie R, Alford performed the tissue prepara tions. II. L. Stewart, MD, and the tissue lalx>rntory of the National Cancer Institute, Bethesda, Md, performed evaluations of liver slides. References 1. Kimbrough RD, Linder RE, Guinea TH: Morphological changes in liver of rots led poly chlorinated biphenyl*. Arch Environ Health 25: 354-364. 1972. 2. Risebrough RW, et al: Polychlorinated bi phenyls in the global ecosystem Nature 220: 1078-1102, 1968. 3. Curley A, et al: Polychlorinated biphenyls: Distribution and storage in body fluids and tis sues of shurmun rats. Environ Res 4:481-495, 1971. 4. Kimbrough RD, Gaines TH, Linder HE: The ultiusti ucture of livers of rats fed DDT und dieldrin. Arch Environ Health 22:469-467, 197). 5. Goldstein JA, Hickman P, Jwc D: Hepatic porphyria induced by polychlorinnted biphenyls (PCB's): Relationship between porphyria, 6aminolevulinic acid synthetase and cytochrome P-450, abstracted. Fed Froc 32:702, 1973. 6. Grant DL, Phillips WEJ, Villoneuvc DC: Metabolism of a (Kilychlovimttcd biphenyl (Aroclor 12.54) mixture in the rat. Bull Environ Confarn Toxicol 6:102-112, 1971. 7. Vos JG, Koumnn JH: Comparative toxicol ogic study with polychlorinated biphenyls in chickens with special reference to porphyria, edema formation, liver necrosis and tissue resi dues. Toxicol Appl Pharmacol 1 7:656-658, 1970. 8. Ufock WI), Cornish HH: Metabolism of 1mphenyl and 4-chlorobiphenyl in the rabbit. J Bw! Chem 234:3301-3302, 1959. 9. Hutzinger O, et al: Polychlorinnted biphen yls: metabolic behavior of pure isomers in pi geons, rats, und brook trout. Science 178 312 313,1972. 10. Edwards JE, While JrPnlhologic changes with special reference to pigmentation and clas sification of hepatic tumors in rats led p dinielhylaminoazobemeenc (butter yellow), J Hut/ Cancer Inst 2:157-183, 194). 11. Bennett GA, Drinker CK, Warren ME: Morphological changes in the livers of nils re sulting from exposure to certain chlorinalcd hy drocarbons. J hid Hyp Toxicol 20.97-123, 1938 12. Stewart HL.SncII KC: The histoputhnlogy of experimental tumors ol the liver of the rat: A critical review of the histopathogenesis. Acta Uniolnt Contra Cuncntni 13:770-803, 1957. 13. Harlroft WS. Observation and interpreta tion of lipid pigments (LipofuHCi ns) in the pathol ogy of laboratory unimals. Crit Bee Toxicol 1:379-411, 1972. 14. Gorham JR, Baker GA, Hoe N: Observa tions on the etiology of yellow fat disease in mink: Preliminary report, Vet Med 46:100-102, 1951. 'through et al Arch Environ Health/Vol 27, Dec 1973 Adenofibrosis/Kimbrough et al 395 D$W 025596 STLCOPCB4009551