American Health Foundation, 1 Dana Road, Valhalla, New York 10595
Received February 1, 2001; accepted April 16, 2001
ABSTRACT
Polychlorinated biphenyls (PCBs) are liver-tumor promoters in rodents, but the underlying mechanisms have not been fully elucidated. Tissue sections from the PCB bioassay reported by Mayes et al. 1998, Toxicol. Sci., 4166, were evaluated by histopathological techniques that included immunohistochemistry. In females, and to a much lesser extent in males, iron accumulation in hepatocytes was found at the 26th-week sacrifice, which was pronounced in the mid- and high-dose Aroclor-1254 and -1260 groups. At 52 weeks, large accumulations of iron were also present in Kupffer cells of females, and dose-related increases in proliferating cell nuclear antigen (PCNA) hepatocyte labeling indices were found in both males and females. These changes preceded the formation of liver tumors, which were not generally found until 78 weeks. Glutathione S-transferase placental (GSTP) positive foci were present at 52 weeks in high-dose Aroclor-1254 and -1260 female groups, and small foci were found in some Aroclor 1254-exposed female rats at 26 weeks, along with centrilobular hepatocytes expressing GSTP. The results of this study suggest that PCB-induced iron accumulation in hepatocytes is an early event that may be related to tumor formation, especially in female rats. In both males and females, increases in cell proliferation at 52 weeks were statistically significantly correlated with tumor incidences at termination among the various PCB dosage groups. Consequently, iron accumulations producing oxidative damage, and enhanced cell proliferation resulting in tumor promotion may be components in the mode of action for PCB-induced hepatocarcinogenesis in rodents.
Key Words: polychlorinated biphenyls; cell proliferation; carcinogenesis; iron accumulation; Aroclor; cancer mechanism; tumor promotion; liver cancer.
Polychlorinated biphenyls (PCBs) have not been used in the U.S. since the 1970`s, but have remained a concern because of their previous widespread use in various products and their persistence in the environment. Commercial PCB mixtures differed according to their isomer composition and chlorine content. There are 209 possible isomers among 10 different PCB congeners. The U.S. commercial products were named Aroclors, which are classified by a 4-digit system beginning with "12" and ending with the percent, by weight, of chlorine, e.g., Aroclor 1260 is 60% chlorine by weight. The exception is Aroclor 1016, which contains approximately 41% chlorine.
The first report of a complete animal carcinogenicity bioassay for PCBs was published by Kimbrough et al. (1975). In this study, female Sherman rats were fed 100 ppm of Aroclor 1260 for 21 months, and a high incidence of hepatocellular carcinomas and neoplastic nodules (adenomas) were reported. Norback and Weltman (1985) and Schaeffer et al. (1984) also reported hepatocellular tumors in Sprague-Dawley rats exposed to Aroclor 1260 or Clophen A60, another commercial mixture of PCBs containing a 60% chlorine content. In the latter study, there was no significant increased tumor incidence for Clophen A30, a PCB mixture containing 42% chlorine. Also, the National Cancer Institute (1978) reported only small increases in liver tumors that were not statistically significantly different from controls in male and female F344 rats exposed to Aroclor 1254 for 24 months. Recently, a comprehensive bioassay comparing 4 different PCB mixtures at different dosage levels in Sprague-Dawley rats was reported (Mayes et al., 1998). In that study, the incidence of hepatocellular neoplasms in females showed dose-related incidences of liver tumors and differences in Aroclor potencies (Aroclor 1254 > 1260
1242 > 1016). For males, only the high-dose Aroclor 1260 group showed a statistically significant increase in tumors.
The mechanism by which PCBs produce liver cancer in rodents is not known, although the mode of action has been found to involve tumor promotion (Deml and Oesterle, 1987). PCBs are not mutagenic in bacterial systems, and most of the other tests for genotoxicity have yielded negative results (Safe, 1989
). No evidence of PCB reactivity with DNA has been found in vivo by evaluation DNA for adducts by 32P-postlabeling (Whysner et al., 1998
). Studies in the literature have reported short-term increased rates of DNA synthesis and mitotic figures (Deml and Oesterle, 1982
), development of GSTP-positive foci (Flodström and Ahlborg, 1992
) and inhibition of gap junctions (Ruch and Klaunig, 1986
; Swierenga et al., 1990
)). The time course, dose and Aroclor dependence of these parameters have not been examined in long-term studies. Also, the relationship of these cellular effects to the neoplastic process and to other mechanistic parameters has not been established.
The overall purpose of this investigation was to develop additional data regarding the mode of action of PCB-induced hepatocarcinogenesis in rodents by the use of immunohistochemical and other staining methods on tissues obtained from the study reported by Mayes et al. (1998). Here we have measured increases in cell proliferation rates and tested these for correlations with tumorigenesis, according to Aroclor type and dose. Also, the time courses of increased cell proliferation rates and development of altered foci were examined. In addition, because decreases in hemoglobin and blood cell parameters, and the presence of gray or brown pigment filling the cytoplasm of cells lining the hepatic sinusoids were reported (Brunner et al., 1996; Mayes et al., 1998
), we used iron staining to determine whether iron accumulation was associated with PCB exposures. We found that hepatocyte-proliferation increases correlated with subsequent tumor formation in both sexes and that iron accumulations in hepatocytes occurred in females, especially in groups that subsequently developed the greatest numbers of tumors.
MATERIALS AND METHODS
The in-life study design, test materials, and diet have been previously described (Mayes et al., 1998). The dosage groups for males and females were dietary levels of PCBs as follows: control; 50, 100, and 200 ppm Aroclor 1016; 50 and 100 ppm Aroclor 1242; 25, 50, and 100 ppm Aroclor 1254; and 25, 50, and 100 ppm Aroclor 1260. Liver sections were fixed in 10% neutral buffered formalin, sectioned, stained by immunohistochemistry, and counterstained with hematoxylin.
Cell proliferation was evaluated by immunohistochemical staining according to the method previously described by Iatropoulos et al. (1997). Sections from the formalin-fixed left and median lobes of the livers were stained using the immunohistochemical technique, anti-proliferating cell nuclear antigen (PCNA) clone PC10 (Zymed). Staining for PCNA was done for livers from 3 animals/group for the 26- and 52-week interim sacrifices. The methods used for counting numbers of PCNA-staining cells excluded cells in identifiable hepatocellular altered foci.
The quantitation of PCNA-positive nuclei in the immunostained sections was performed in the following manner: a square graticule with 25 equal subdivisions was used for counting labeled cells at 200 x magnification, providing a square counting area that usually contained between 25 to 50 hepatocytes. All PCNA-labeled and unlabeled nuclei in this graticule area were counted. Counts were obtained from 12 to 14 graticule areas for each 2 liver sections per animal, resulting in the counting of about 1000 nuclei per animal, and the labeling index was expressed as the percent of hepatocytes labeled. Mean labeling indices were compared among the groups using one-way analysis of variance (ANOVA) followed by Dunnett's method for multiple comparisons (Dunnett, 1955), where each exposure group is compared to the control. In order to assess any dose-response relationships relative to controls, trend analysis was conducted on selected groups using linear regression, assuming equal spacing between groups. Correlation coefficients were computed to assess any linear relationship between mean PCNA labeling indices and tumor incidence. Trend and correlation p-values were 2-tailed and considered significant at the 0.05 level. For testing of difference between regression coefficients, the t-test method described by Zar (1984) was used. Further testing of this difference used an analysis of covariance model (ANCOVA), in which tumor incidence was included as the dependent variable, sex the independent variable, and the PCNA labeling index as the covariate.
The Prussian-blue method for iron employed a reaction of potassium ferrocyanide with pretreatment by dilute acid to solubilize loosely bound ferric iron such as that in hemosiderin. The slides were then counterstained with nuclear fast red. Sections from 3 animals/group from the formalin-fixed left and median lobes of the livers for all of the 26-week groups and the high-dose animals for each of the Aroclors at 52 weeks of exposure were stained along with the controls.
GSTP staining using anti-GSTP rabbit polyclonal antibody (Novacastra) was performed on slides from the formalin-fixed left and median lobes of the livers from 6 animals/group exposed for 26 weeks to high-dose Aroclors 1254 and 1260, as well as the male and female controls. For the 52-week sacrifice, slides from 3 animals from each of these groups were stained. GSTP staining was performed on slides from all 6 animals from female groups exposed for 78 weeks to the highest dose of each of the Aroclors. For terminal sacrifice (104 weeks), staining was performed on slides from the high-dose Aroclor 1254 and 1260 groups.
For quantification of foci, identification of a GSTP-positive focus was based upon the presence of 3 or more contiguous stained cells. The livers were scanned with a Umax Powerlook III using a transparency adaptor and MagicScan software provided by the manufacturer, and were further processed to determine the areas with Image-Pro software from Media Cybernetics, Silver Spring, MD. Positively stained cells that were in a regional staining pattern surrounding a central vein were not considered foci according to the description of Thomas et al., (1998). Mean numbers of GSTP-positive foci (foci/cm2) were compared between the exposure groups and control using Student's t-test and 1-way analysis of variance (ANOVA), followed by Dunnett's method for multiple comparisons (Dunnett, 1955). Ito et al. (1992) has described this method of quantitation, using foci/cm2. Since the purpose of this measurement was to determine the time course for focus development rather than a quantitative comparison among exposure groups, more extensive measurements involving 3-dimensional aspects of focus stereology were not used.
The TdT-mediated dUTP-biotin nick-end labeling (TUNEL) immunohistochemical technique employed polyclonal rabbit antibody (Trevigen), according to the method of Gravrieli et al. (1992). The terminal deoxynucleotidyl transferase (TdT) binds specifically to the 3'-OH end of DNA, which is formed during apoptosis, and catalyzes the addition of biotin-labeled deoxynucleotides. Staining was performed on slides from the formalin-fixed left and median lobes of the livers of 6 animals/group, who were exposed for 26 weeks to high-dose Aroclors 1254 and 1260 (with the exception of 5 animals/group in the male high-dose Aroclor 1254 group), as well as the male and female controls. For the 52-week sacrifice, slides from 3 animals from each of these groups were stained. Due to the small numbers of apoptotic cells, quantitative analysis was not possible. However, a semiquantitative method was used, which involved counting the numbers of apoptotic bodies seen when scanning a defined area of each liver section. This area was the width of the viewing field at 100 x magnification following the entire periphery of the liver.
RESULTS
The animal livers from the 26-week sacrifices showed no increased numbers of PCNA-labeled nuclei in any of the Aroclor groups compared to the controls for either sex, but several groups at 52-weeks were elevated (see Table 1). The only detectable change at 26 weeks was a borderline statistically significantly decreased number of PCNA-labeled cells associated with dose in the female Aroclor 1016 (p = 0.06) and 1260 (p = 0.04) groups, as determined by trend analysis. At 52 weeks, the average numbers of PCNA-labeled cells in female livers were about 2-fold greater than the control for the highest doses of Aroclors 1242, 1254, and 1260, and these changes were statistically significant. For Aroclor 1016, the highest dose resulted in an approximate 1.6-fold increase in PCNA-labeled cells relative to the control group. For the males, there was a dose-related increase in numbers of PCNA-labeled cells for groups exposed to Aroclors 1254 and 1260. Increases were statistically significant compared to controls for the mid- and high-dose groups exposed to Aroclor 1260. Analysis of the labeling indices at 52 weeks for trends revealed statistically significant PCB dose-response for Aroclors 1254 and 1260 in both sexes and for Aroclor 1242 in females. For the female animals from the 52-week sacrifice, the labeling indices were increased in a dose-related manner in all Aroclor groups, although for Aroclor 1016, this did not reach statistical significance.
|
|
Immunohistochemical staining for GSTP was done to identify the time course of development of altered hepatic foci and neoplasms (Moore et al., 1987; Sato, 1988
) in relation to other events. Most animals sacrificed at 26 weeks from the high-dose Aroclor 1254 and 1260 male and female groups did not reveal increased foci compared to controls. However, 2 out of 6 females in the Aroclor 1254 high-dose group showed some small altered foci containing 310 cells (Fig. 2A
) and centrilobular GSTP staining (Fig. 2B
), and in one of these animals, there were several foci, and the centrilobular GSTP staining was extensive.
|
|
The sections stained by the TUNEL method showed very little evidence of apoptosis and no apparent differences between groups; however, due to the very few numbers of apoptotic cells, only a semiquantitative approach was used. By this method, there appeared to be a decreased number of apoptotic cells only in the male high-dose Aroclor-1254 group compared to control at 26 weeks (data not shown). No differences relative to controls were apparent in the male high-dose Aroclor-1260 group or the female animals exposed to the high doses of Aroclor 1254 or 1260. Similar analysis at 52 weeks suggested that apoptosis was increased in the females exposed to Aroclor 1254 or 1260, but due to a high degree of variability, these findings were uncertain (data not shown). However, there were no apparent large effects of PCBs on apoptosis.
DISCUSSION
PCB-induced increased cell proliferation has been previously reported. Deml and Oesterle (1982) found an enhanced mitotic index in males, but not female, Sprague-Dawley rats 24 h after a single dose of a German commercial mixture containing 50% chlorine. Studies of Wölfle et al. (1988) reported that one week of 3,4,3',4'-tetrachlorobiphenyl administration increased DNA synthesis in the livers of male Wistar rats, and Chen et al. (1995) reported similar findings in male F344 rats. In mice, 100 ppm Aroclor 1254 in the diet for 2 months was found to increase mitotic rates in C57BL/10ScSn but not DBA/2 mice (Smith et al., 1990). In the present study, the percentages of PCNA-positive cells in livers from PCB-exposed Sprague-Dawley rats at 52 weeks were increased in a dose-related manner for all Aroclors, although some of these increases were not statistically significant. At 52 weeks, there was a statistically significant positive correlation between numbers of PCNA-positive cells and dose for all groups except for the male and female Aroclor 1016 and male 1242 groups (Table. 1
). In contrast, PCB exposures did not increase cell proliferation at 26 weeks (the earliest time measured), and, if anything, there was suggestive evidence of decreased cell proliferation in 2 of the Aroclor groups.
In Figure 3, the percentages of PCNA-positive cells at 52 weeks are plotted with incidences of tumorigenesis reported by Mayes et al. (1998) for each of the Aroclor dose groups. Statistical analysis revealed very strong positive correlations between PCNA labeling and liver tumor incidence for both males and females (p < 0.001). The labeling index was maximally increased about 2-fold above control levels for both males and females; however, the increased tumor rates were much greater in females. The tumors in males increased from 7% in controls to 12 and 20% in high-dose Aroclor 1254 and 1260, respectively, whereas in females the increases were from 1 to 56% and 48%. The regression coefficients for the 2 sexes were found to be statistically significantly different (p < 0.0001). Also, in an analysis of covariance model, there was a significant interaction between PCNA and sex for tumor incidence. Consequently, the increased cell proliferation rates do not explain the sex difference in PCB-induced tumorigenesis, although the labeling indices do correlate with tumorigenesis for each of the sexes individually. This finding indicates that in females there may be another tumor-enhancing effect produced by PCBs that is independent of enhanced cell proliferation, and our finding of PCB-related hepatocyte iron accumulation provides one possible explanation for the gender difference.
|
The finding of increased iron in Kupffer cells appears to be the source of the pigment localized in the sinusoidal lining cells that had been reported in the bioassay (Mayes et al., 1998). As documented in the bioassay (Brunner et al., 1996
), this pigment occurred in the livers from the females from the 26- and 52-week sacrifices exposed to Aroclors 1242, 1254, and 1260. Increases in pigment were not detectable in the males in any dosage group, even at the terminal sacrifice. Similarly in our study, staining for iron in male liver sections did not reveal accumulations of iron in the Kupffer cells at either 26 or 52 weeks of PCB exposures, whereas increases were found in females from the 26- and 52-week sacrifices exposed to Aroclors 1242, 1254, and 1260. This Kupffer cell siderosis in the females usually indicates iron derived from the phagocytosis of erythrocytes. However, in this case, hepatocellular siderosis appeared to be even more prominent at 26 weeks, whereas accumulation of iron in the Kupffer cells was much more prominent at 52 weeks.
The accumulation of iron in the liver was accompanied by a mild decrease in mean corpuscular hemoglobulin and hemoglobulin concentration, especially in females, where it occurred earlier and to a greater extent than in males (Brunner et al., 1996; Mayes et al., 1998
). In females, this finding was present at 26 weeks at all doses of Arolcor and in the high-dose Aroclor 1242 and 1260 groups. In some of these groups, there were also small decreases in total hemoglobulin. These changes may reflect reduced available iron for red cells due to sequestration of iron in the liver, and they argue against an elevation in intestinal iron absorption as the cause of hepatocellular siderosis (Lee, 1994
).
The induction of porphyria by PCBs in rats has been previously described (Goldstein et al., 1974; Strik and Debets, 1980
), and in this bioassay (Mayes et al., 1998
), large dose-related increases in total liver porphyrins were also found (Brown et al., in press). Porphyrins were elevated beginning at 6 months of exposure (the earliest time measured), especially in females exposed to Aroclor 1254, with smaller increases in the Aroclor 1260 and 1242 groups, and the least for the Aroclor 1016 groups (Brown, personal communication). It has been shown that iron loading also greatly enhances porphyria induced by exposure to 100 ppm Aroclor 1254 for 5 weeks in Ah-responsive male mice (Smith et al., 1995
). Also, iron-treated mice with PCB exposure exhibited an inhibition of uroporphyrinogen decarboxylase, whereas no inhibition occurred with exposure to PCBs or iron alone.
The localization of iron in hepatocyte lysosomes associated with porphyria has been reported to be involved in the mode of action for hexachlorobenzene (HCB)-induced liver cancer (Tangeras, 1986). A theory regarding the production of hepatic tumors by HCB-induced oxidative damage, involving cytochrome P450 induction, porphyria, inhibition of uroporphyrinogen decarboxylase, and iron effects, was proposed by Smith and De Matteis (1990). It has been hypothesized that the induction of P450 and iron act synergistically to produce oxidative damage leading to tumor development. Iron loading has been found to significantly enhance the effects of HCB, and a study of Cyp1a2(-/-) mice found that the presence of CYP1A2 is essential for the production of uroporphyria caused by HCB and Fe (Sinclair et al., 2000
). The uroporphyria in hereditary porphyria cutanea tarda (PCT) is believed to be precipitated by an iron-dependent process (Siersema et al., 1995
) and PCT can be treated by phlebotomy to remove excessive iron (Nordmann, 1999
). Therefore, increased available iron contributed to the porphyrinogenic process in vivo.
The finding of iron staining with a stippled pattern in hepatocytes in the present study would be consistent with an accumulation in subcellular organelles such as lysosomes. Similar iron staining has also been reported for HCB-exposed animals (Kuiper-Goodman, 1977). These findings were not secondary to an increase in total body iron since HCB (and PCBs in our study) produces some evidence of a decrease in red blood cells, hematocrit, and blood hemoglobin (Kuiper-Goodman, 1977). Studies of HCB have shown that female F344 rats are more sensitive than males to the accumulation of hepatic non-heme iron, porphyria and hepatocarcinogenesis produced by HCB (Smith et al., 1985), and Sprague-Dawley rats show a similar sex difference for liver tumor formation (Erturk et al., 1986
). The female rat liver contains greater hemosiderin (Smith et al., 1985
; Vidnes and Hegeland, 1973), and the sex difference of the HCB effect has been ascribed to the greater liver iron content and turnover in females (Smith and De Mattei, 1990).
In the case of PCBs, the porphyrinogenic effect was found to be greater in female F344 rats compared to males, even with iron loading (Franklin et al., 1997), which implies that another female specific factor besides iron content may be involved. In deproteinated liver cytosols from the same bioassay used in the studies reported here, a number of redox-active peaks were found, and the largest were ascribed to the glutathionylated estrogen catechols. Consequently, Brown et al. (in press) has proposed a mode of action for the hepatocarcinogenesis of PCBs due to these estrogen metabolites, with the involvement of reactive oxygen generation.
The findings of PCB-induced hepatic siderosis and porphyria in our study strongly suggests that PCB-related hepatocarcinogenesis may also involve a mechanism including iron-mediated oxidative damage, similar to that proposed for HCB. The presence of hepatocellular siderosis, as opposed to Kupffer cell siderosis, is associated with iron-induced hepatic injury in human diseases (Lee, 1994). Iron pretreatment of Ah-responsive C57BL/10ScSn mice sensitizes them to the induction of oxidative DNA damage and tumorigenicity by Aroclor 1254 (Smith et al., 1990
; Faux et al., 1992
). These investigators also showed a synergistic effect of iron loading on PCB-induced mitotic rate increases following 2 months of Aroclor 1254 exposure. It is possible that iron accumulation works synergistically with generation of hydrogen peroxide via superoxide formed by redox cyclers to ultimately produce the hydroxyl radical (Halliwell and Gutteridge, 1999
).
The mechanism of PCB effects on iron metabolism that lead to hepatic iron accumulation remains to be elucidated. The regulation of iron and heme biosynthesis has been summarized (May et al., 1995; Young and Aisen, 1994
). Normally, iron in hepatocytes is complexed with protoporhyrin as heme and stored as Fe3+ within ferritin shells, if not utilized. This maintains excess iron sequestered in a relatively non-toxic form; however, the large iron accumulations found in this study suggest a significant increase in amount of free iron.
It is well established that iron accumulation in hepatocytes occurs in hereditary PCT (Lee, 1994), which has been associated with liver cancer (Peters et al., 1987
). However, HCB and PCB exposures have not been associated with increased incidences of liver cancer in humans (Kimbrough et al., 1999
; Peters et al., 1987
). Based upon the proposed mode of action, the lack of liver cancer induction in humans may be due to the dose-response relationship whereby environmental exposures, including occupational exposures, do not produce sufficient hepatic iron accumulation or redox cyclers. In addition, these elements of the mode of action may be particularly active in the female Sprague-Dawley rat. Norback and Weltman (1985) tested Aroclor 1260 exposure for 104 weeks in SpragueDawley rats and also found that the incidence of hepatocellular neoplasms in females was much higher than in males. In contrast to these higher incidences of hepatic tumors in female Sprague-Dawley rats, a study of F344 rats with Aroclor 1254 exposure for 1045 weeks (NCI, 1978
) found only small increases in liver tumors for both males and females. These increases were not statistically significantly different from controls, although there was a dose-related increase in hyperplastic nodules that were dose related in both sexes. Schaeffer et al. (1984) did report a relatively high incidence of hepatocellular carcinomas in male Wistar rats exposed for 119 weeks to a German commercial mixture of PCBs containing a 60% chlorine content; however, the first tumors in these males were not observed until after 100 weeks. This delay is consistent with a relatively low incidence of tumors in males after 104 weeks found in the other bioassays (Mayes et al., 1998
; Norback and Weltman, 1985
). Consequently, these reported studies suggest that the greater effect in female rats may be related to the rat strain.
At 26 and 52 weeks of PCB exposures, centrilobular cells staining for GSTP were found, and especially noted in 2/6 female rats, in the 26-week Aroclor 1254 high-dose group. Although areas of this staining were quite extensive in one animal, most were not identified as altered foci. The occurrence of increased centrilobular GSTP staining has been noted to occur following exposures to other Ah-receptor agonists or porphyria-inducing chemicals, including TCDD, chlorinated benzenes, and PCBs (Thomas et al., 1998). Enhanced centrilobular expression of glutathione S-transferase P1-1 was found to be co-localized with c-jun, c-fos, and CYP 1A2 induction in HCB-induced porphyria (Thomas et al., 1998
). Iron loading has also been found to enhance the Aroclor-induced GSTs in the livers of c57BL/10ScSn mice, which was ascribed to the oxidative processes produced by HCB (Madra et al., 1996
). The increases in GSTP centrilobular staining found in our study may, therefore, also represent a response to oxidative stress produced by PCBs in these female rats. It has also been found that PCBs can induce GSTP in primary cultured hepatocytes from male Wistar rats (Aoki et al., 1992
).
It was verified in our study that all identifiable PCB-induced foci and tumors, as identified by conventional staining, contained GSTP-positive cells; however, in some foci there was great variability in degree of staining. At 26 weeks, there were small foci containing 310 cells discernible in 2/6 Aroclor 1254 high-dose female liver, but no differences were observed, compared to controls, in the numbers of GSTP-positive foci in other groups of males or females examined. These were the same animals that exhibited the large numbers of GSTP-positive staining hepatocytes in the centrilobular region. In contrast, the numbers of foci were increased in the females exposed to high-dose Aroclors 1254 and 1260 for 52 weeks, and many of these foci were larger (>10 hepatocytes). By 78 weeks, there were many large foci in all high-dose Aroclor females with the exception of the Aroclor 1016 group, where there was an average of only one focus per section. These findings are in contrast to the relatively rapid development of foci and tumors in the liver by DNA-reactive compounds (Williams, 1999), and provide further support for an epigenetic effect of PCBs.
Increases in cell proliferation and development of GSTP-positive foci showed a similar time course. No increases in cell proliferation were found at 26 weeks; however, for females of these groups, increases in both cell proliferation and GSTP-positive foci were found at 52 weeks. In contrast, although there was clear evidence of increased cell proliferation rates in males at 52 weeks, no increases in GSTP-positive foci were evident.
PCB-induced hepatocellular tumor development has been correlated in this study with enhanced cell proliferation in both sexes. In addition, the hepatic accumulation of iron, primarily in females, has been observed. The mode of action for liver cancer produced by PCBs is therefore hypothesized to involve enhanced cell proliferation and oxidative damage, which may be mediated by iron. In addition, formation of glutathionylated estrogen quinones, along with generation of reactive oxygen species, have been described in these PCB-exposed rats (Brown et al., in press). Porphyria has also been found in this study, and inhibition of uroporphyrinogen decarboxylase, leading to porphyrin accumulation, has been described by other investigators of PCBs and other chlorinated aromatic hydrocarbons that produce hepatic tumors.
The time-course of PCB-related effects found in this study is shown in Figure 4. The accumulation of iron in the hepatocytes appears to be an early event, and at this point small altered foci appear. Also, redox cyclers generate reactive oxygen species such as glutathionylated estrogen quinones, and porphyria develops. Later, foci become more common and larger, which coincides with increased cell proliferation. Finally, very large foci and tumors occur. These events provide additional evidence for an epigenetic mode of action for PCB-induced hepatic tumorigenesis and are consistent with our previous findings of a lack of demonstrable DNA reactivity of PCBs.
|
ACKNOWLEDGMENTS
The authors would like to acknowledge the technical expertise of Beverly Gambrell and Victoria Baker, who provided the excellent staining results for this study. Also, we wish to thank Drs. John F. Brown, Jr., Brian A. Mayes, and Stephen B. Hamilton for the tissues from their study and for their suggestions and encouragement. The funding for this study was provided by the General Electric Company.
NOTES
1 To whom correspondence should be addressed. Fax: (914) 592-6317. E-mail: whysner{at}aol.com.
The major findings of this research were presented at the 39th annual meeting of the Society of Toxicology, March 2000, Philadelphia, PA.
REFERENCES
Aoki, Y., Satoh, K., Sato, K., and Suzuki, K. T. (1992). Induction of glutathione S-transferase P-form in primary cultured rat liver parenchymal cells by co-planar polychlorinated biphenyl congeners. Biochem. J. 281, 539543.[ISI][Medline]
Brown, J. F., Jr., Fish, K. M., Mayes, B. A., Silkworth, J. B., Hamilton, S. B., and Whysner, J. (in press). PCB effects on epigenetic carcinogenic processes. In Recent Advances in the Environmental and Health Effects of PCBs (L. W. Robertson and L. G. Hanson, Eds.).
Brunner, J. M., Sullivan, T. M., Singer, A. W., Ryan, M. J., Toft, J. D., Menton, R. S., Graves, S. W., and Peters, A. C. (1996). An Assessment of the Chronic Toxicity and Oncogenicity of Aroclor-1016, Aroclor-1242, Aroclor-1254, and Aroclor-1260 Administered in Diet to Rats: Chronic Toxicity and Oncogenicity Report. Batelle, Columbus, OH.
Chen, Z.-Y., White, C. C., He, C.-Y., Liu, Y.-F., and Eaton, D. L. (1995). Zonal differences in DNA synthesis activity and cytochrome P450 gene expression in livers of male F344 rats treated with five nongenotoxic carcinogens. J. Environ. Pathol. Toxicol. Oncol. 14, 8399.[Medline]
Deml, E., and Oesterle, D. (1982). Sex-dependent promoting effect of polychlorinated biphenyls on enzyme-altered islands induced by diethylnitrosamine in rat liver. Carcinogenesis 3, 14491453.[ISI][Medline]
Deml, E., and Oesterle, D. (1987). Dose-response of promotion by polychlorinated biphenyls and chloroform in rat liver foci bioassay. Arch. Toxicol. 60, 209211.[ISI][Medline]
Dunnett, C. W. (1955). A multiple comparisons procedure for comparing several treatments, with a control. J. Am. Stat. Assoc. 50, 10961121.[ISI]
Erturk, E., Lambrecht, R. W., Peters, H. A., Cripps, D. J., Gocmen, A., Morris, C. R., and Bryan, G. T. (1986). Oncogenicity of hexachlorobenzene. In Hexachlorobenzene: Proceedings of an International Symposium (C. R. Morris, and J. R. P. Cabral, Eds.), Vol. 77, pp. 417423. IARC Scientific Publication, Lyon.
Faux, S. P., Francis, J. E., Smith, A. G., and Chipman, J. K. (1992). Induction of 8-hydroxydeoxyguanosine in Ah-responsive mouse liver by iron and Aroclor 1254. Carcinogenesis 13, 247250.[Abstract]
Flodström, S., and Ahlborg, U. G. (1992). Relative liver tumor promoting activity of some polychlrorinated dibenzo-p-dioxin, dibenzofuran and biphenyl congeners in female rats. Chemosphere 25, 169172.[ISI]
Franklin, M. R., Phillips, J. D., and Kushner, J. P. (1997). Cytochrome P450 induction, uroporphyrinogen decarboxylase depression, porphyrin accumulation and excretion, and gender influence in a 3-week rat model of Porphyria Cutanea Tarda. Toxicol. Appl. Pharmacol. 147, 289299.[ISI][Medline]
Goldstein, J. A., Hickman, P., and Jue, D. L. (1974). Experimental hepatic porphyria induced by polychlorinated biphenyls. Toxicol. Appl. Pharmacol. 27, 437448.[ISI][Medline]
Gravrieli, Y., Sherm, Y., and Ben-Sasson, S. A. (1992). Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biol. 119, 493501.[Abstract]
Halliwell, B., and Gutteridge, J. M. C. (1999). Free Radicals in Biology and Medicine, 3rd ed., p. 172. Oxford University Press, Oxford, U.K.
Iatropoulos, M. J., Williams, G. M., Wang, C.-X., Brunnemann, K. D., and Leber, A. P. (1997). Assessment of chronic toxicity and carcinogenicity in rats of Wingstay® 100, a rubber antioxidant/antiozonant. Exp. Toxicol. Pathol. 49, 153165.[ISI][Medline]
Ito, N., Hasegawa, R., Imaida, K., Masui, T., Takahashi, S., and Shirai, T. (1992). Pathological markers for non-genotoxic agent-associated carcinogenesis. Toxicol. Lett. 64/65, 613620.
Kimbrough, R. D., Doemland, M. L., and LeVois, M. E. (1999). Mortality in male and female capacitor workers exposed to polychlorinated biphenyls. J. Occup. Environ. Med. 41, 161171.[ISI][Medline]
Kimbrough, R. D., Squire, R. A., Linder, R. E., Strandberg, J. D., Montali, R. J., and Burse, V. W. (1975). Induction of liver tumors in Sherman-strain female rats by polychlorinated biphenyl Aroclor 1260. J. Natl. Cancer Inst. 55, 14531459.[ISI][Medline]
Kuiper-Goodman, T., Grant, D. L., Moodie, C. A., Korsrud, G. O., and Munro, I. C. (1977). Subacute toxicity of hexachlorobenzene in the rat. Toxicol. Appl. Pharmacol. 40, 529549.[ISI][Medline]
Lee, R. G. (1994). Diagnostic Liver Pathology, pp. 237249. Mosby, St. Louis.
Longueville, A., and Crichton, R.R. (1986). An animal model of iron overload and its application to study hepatic ferritin iron mobilization by chelators. Biochem. Pharmacol. 35, 36693678.[ISI][Medline]
Madra, S., Mann, F., Francis, J. E., Manson, M. M., and Smith, A. G. (1996). Modulation of iron of hepatic microsomal and nuclear cytochrome P450, and cytosolic glutathione S-transferase and peroxidase in C57BL/10ScSn mice induced with polychlorinated biphenyls (Aroclor 1254). Toxicol. Appl. Pharmacol. 136, 7986.[ISI][Medline]
May, B. K., Dogra, S. C., Sadlon, T. J., Bhasker, C. R., Cox, T. C., and Bottomley, S. S. (1995). Molecular regulation of heme biosynthesis in higher vertebrates. Prog. Nucleic Acid Res. Mol. Biol. 51, 151.[ISI][Medline]
Mayes, B. A., McConnell, E. E., Neal, B. H., Brunner, M. J., Hamilton, S. B., Sullivan, T. M., Peters, A. C., Ryan, M. J., Toft, J. D., Singer, A. W., Brown, J. F., Jr., Menton, R. G., and Moore, J. A. (1998). Comparative carcinogenicity in Sprague-Dawley rats of the polychlorinated biphenyl mixtures Aroclors 1016, 1242, 1254, and 1260. Toxicol. Sci. 41, 6276.[Abstract]
Moore, M. A., Nakagawa, K., Satoh, K., Ishikawa, T., and Sato, K. (1987). Single GST-P-positive liver cellsputative initiated hepatocytes. Carcinogenesis 8, 483486.[Abstract]
NCI (1978). Bioassay of Aroclor® 1254 for possible carcinogenicity. Report 346, 162.
Norback, D. H., and Weltman, R. H. (1985). Polychlorinated biphenyl induction of hepatocellular carcinoma in the Sprague-Dawley rat. Environ. Health Perspect. 60, 97105.[ISI][Medline]
Nordmann, Y. (1999). Human hereditary porphyrias. In: Oxford Textbook of Clinical Hepatology. Volume II (J. Bircher, J.-P. Benhamou, N. McIntyre, M. Rizzetto, and J. Rodés, Eds.), pp.14241429. Oxford University Press, New York.
Peters, H., Cripps, D. Gocmen, A., Bryan, G. Erturk, E., and Morris, C. (1987) Turkish epidemic hexachlorobenzene porphyria: A 30-year study. Ann. NY Acad. Sci. 514, 183190.[Medline]
Ruch, R. J., and Klaunig, J. E. (1986). Effects of tumor promoters, genotoxic carcinogens, and hepatocytotoxins on mouse hepatocyte intercellular communications. Cell Biol. Toxicol. 2, 469483.[ISI][Medline]
Safe, S. (1989). Polychlorinated biphenyls (PCBs): Mutagenicity and carcinogenicity. Mutat. Res. 220, 3147.[ISI][Medline]
Sato, K. (1988). Glutathione S-transferases and hepatocarcinogenesis. Jpn. J. Cancer Res. (Gann) 79, 556572.[ISI][Medline]
Schaeffer, E., Greim, H., and Goessner, W. (1984). Pathology of chronic polychlorinated biphenyl (PCB) feeding in rats. Toxicol. Appl. Pharmacol. 75, 278288.[ISI][Medline]
Siersema, P. D., Rademakers, L. H., Cleton, M. I., ten Kate, F. J., de Bruijn, W. C., Marx, J. J., and Wilson, J. H. (1995). The difference in liver pathology between sporadic and familial forms of porphyria cutanea tarda: The role of iron. J. Hepatol. 23, 259267.[ISI][Medline]
Sinclair, P. R., Gorman, N., Bement, W. J., and Sinclair, J. F. (2000). Chlorinated biphenyls induce cytochrome P4501A2 and uroporphyrin accumulation in cultures of mouse hepatocytes. Arch. Biocem. Biophys. 281, 225232.
Smith, A. G., Carthew, P., Clothier, B., Constantin, D., Francis, J. E., and Madra, S. (1995). Synergy of iron in the toxicity and carcinogenicity of polychlorinated biphenyls (PCBs) and related chemicals. Toxicol. Lett. 82/83, 945950.
Smith, A. G., and De Matteis, F. (1990). Oxidative injury mediated by the hepatic cytochrome P-450 system in conjunction with cellular iron. Effects on the pathway of heame biosynthesis. Xenobiotica 20, 865877.[ISI][Medline]
Smith, A. G., Francis, J. E., and Carthew, P. (1990). Iron as a synergist for hepatocellular carcinoma induced by polychlorinated biphenyls in Ah-responsive C57BL/10ScSn mice. Carcinogenesis 11, 437444.[Abstract]
Smith, A. G., Francis, J. E., Dinsdale, D., Manson, M. M., and Cabral, J. R. P. (1985). Hepatocarcinogenicity of hexachlorobenzene in rats and the sex difference in hepatic iron status and development of porphyria. Carcinogenesis 6, 631636.[Abstract]
Strik, J. J., and Debets, F. M. (1980). Chemical porphyria. In Topics in Environmental Health. Vol. 4. Halogeneted Biphenyls, Terphenyls, Naphtalenes, Dibenzodioxins andRelatedProducts (Kimbrough, Ed.), pp. 191239. Elsevier, North Holland.
Swierenga, S. H. H., Yamasaki, H., Piccoli, C., Robertson, L., Bourgon, L., Marceau, N., and Fitzgerald, D. J. (1990). Effects of intercellular communication in human keratinocytes and liver-derived cells of polychlorinated biphenyl congeners with differing in vivo promotion activities. Carcinogenesis 11, 921926.[Abstract]
Tangeras, A. (1986). Lysosomes, but not mitochondria, accumulate iron and porphyrins in porphyria induced by hexachlorobenzene. Biochem. J. 235, 671675.[ISI][Medline]
Thomas, R. S., Gustafson, D. L., Ramsdell, H. S., El-Masri, H. A., Benjamim, S. A., and Yang, R. S. H. (1998). Enhanced regional expression of glutathione S-transferase P11 with colocalized AP-1 and CYP 1A2 induction in chlorobenzene-induced porphyria. Toxicol. Appl. Pharmacol. 150, 2231.[ISI][Medline]
Vidnes, A., and Helgeland, L. (1973). Sex and age differences in the hemosiderin content of rat liver. Biochim. Biophys. Acta 328, 365372.[ISI][Medline]
Whysner, J., Montandon, F., McClain, M., Downing, J., Verna, L. K., Steward, R. E., III, and Williams, G. M. (1998). Absence of DNA adduct formation by phenobarbital, polychlorinated biphenyls, and chlordane in mouse liver using the 32P-postlabeling assay. Toxicol. Appl. Pharmacol. 148, 1423.[ISI][Medline]
Williams, G. M. (1999). Chemically induced preneoplastic lesions in rodents as indicators of carcinogenic activity. In The Use of Short- and Medium-term Tests for Carcinogens and Data on Genetic Effects in Carcinogenic Hazards Evaluation (D. B. McGregor, J. M. Rice, and S. Venitt, Eds.), Vol. 146, pp. 185202. IARC Scientific Publications. IARC, Lyon.
Wölfle, D., Münzel, P., Fischer, G., and Bock, K. W. (1988). Altered growth control of rat hepatocytes after treatment with 3,4,3',4'-tetrachlorobiphenyl in vivo and in vitro. Carcinogenesis 9, 919924.[Abstract]
Young, S. P., and Aisen, P. (1994). The liver and iron. In The Liver: Biology and Pathology (I. M. Arias, J. L. Boyer, N. Fausto, W. B. Jakoby, D. A. Schachter, and D. A. Shafritz, Eds.), pp. 597617. Raven Press, New York.
Zar, J. H. (1984). Biostatistical Analysis, 2nd ed., pp. 292295. Prentice-Hall, New Jersey.