Toxicologic Pathology

Toxicologic Pathology http://tpx.sagepub.com/

The Nongenotoxic Hepatocarcinogens Diethylhexylphthalate and Methylclofenapate Induce DNA Synthesis Preferentially in Octoploid Rat Hepatocytes Susan C. Hasmall and Ruth A. Roberts Toxicol Pathol 2000 28: 503 DOI: 10.1177/019262330002800401 The online version of this article can be found at: http://tpx.sagepub.com/content/28/4/503

Published by: http://www.sagepublications.com

On behalf of:

Society of Toxicologic Pathology

Additional services and information for Toxicologic Pathology can be found at: Email Alerts: http://tpx.sagepub.com/cgi/alerts Subscriptions: http://tpx.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://tpx.sagepub.com/content/28/4/503.refs.html

Downloaded from tpx.sagepub.com by guest on July 14, 2011

Safety Evaluation

—

Pharmaceutical

The

Nongenotoxic Hepatocarcinogens Diethylhexylphthalate and Methylclofenapate Induce DNA Synthesis Preferentially in Octoploid Rat Hepatocytes SUSAN C. HASMALL AstraZeneca Central

AND

RUTH A. ROBERTS

, Macclesfield, Cheshire, Toxicology Laboratory , Alderley Park

United

Kingdom

ABSTRACT

Diethylhexylphthalate (DEHP), a rodent carcinogen, and 1,4-dichlorobenzene (DCB), a noncarcinogen in rat liver, are potent hepatomitogens. We have reported previously that 7-day dosing with DEHP induced a higher bromodeoxyuridine labeling index (LI) in binuclear octoploid (2×4N) rat hepatocytes than did DCB, suggesting that induction of DNA synthesis in 2×4N hepatocytes might represent a more substantial carcinogenic risk. We compared 2 additional rodent hepatocarcinogens, methylclofenapate (MCP) and phenobarbitone, with ethylene thiourea (ETU), a noncarcinogenic hepatomitogen in rat. All 3 chemicals increased hepatic LI; the 8N population had the highest LI, but only the carcinogens increased LI in the 2×4N and 4N populations. To identify the target population for induction of DNA synthesis, we used a 1-hour pulse label at the peak of induction. The results were consistent with the 7-day data, and again the highest LI was in the 8N population. The nongenotoxic rodent carcinogens MCP and DEHP induced a significant increase in the LI in the 2×4N population, whereas ETU and DCB did not. These data support the hypothesis that increased DNA synthesis within the minority 2×4N population may be more significant for subsequent hepatocarcinogenesis. Keywords. Methylclofenapate; 1,4-dichlorobenzene; ethylene thiourea; phenobarbitone; diethylhexylphalate; hepatocyte ploidy; nongenotoxic hepatocarcinogenesis; DNA synthesis

INTRODUCTION The mechanisms through which nongenotoxic chemicals induce liver tumors in rodents are unclear. Many nongenotoxic liver carcinogens stimulate hepatic DNA synthesis (2), and it is generally accepted that increased DNA synthesis is necessary for nongenotoxic hepatocarcinogenesis (1). However, hepatic DNA synthesis can be increased by several chemicals that are not hepatocarcinogenic. For example, ethylene thiourea (ETU) and 1,4dichlorobenzene (DCB) are not hepatocarcinogenic in Fischer 344 (F344) rats (17, 18), but at the dose levels and routes used in the US National Toxicology Program (NTP) bioassays, both ETU and DCB induced increased DNA synthesis in rat liver (21). The induction of DNA synthesis by DCB is similar in timing and magnitude to that induced by diethylhexylphthalate (DEHP), a nongenotoxic rodent peroxisome proliferator and hepatocarcinogen (16). Despite their differing hepatocarcinogenicity, both DCB and DEHP at the doses and routes used in the NTP bioassays induced similar profiles of S-phase labeling index (LI). A large and rapid peak during the first 7 days ( 1115 % and 1151 % of control for DEHP and DCB, respectively) was followed by a return to control levels (21). These data suggest that the size of the S-phase re-

sponse does not necessarily determine hepatocarcinogenic risk and that the subpopulation in which S-phase is induced may be a better correlate with subsequent he-

patocarcinogenicity. The rodent hepatocyte population is heterogeneous, comprising mononuclear and binuclear cells of several ploidy classes (6, 28). We have questioned whether the ploidy and/or nuclearity of the hepatocytes in which a chemical induces DNA synthesis is correlated with the hepatocarcinogenic potential of that chemical. After 7 days, DEHP and DCB had similar effects on the distribution of S-phase between hepatocytes of different ploidy and nuclearity classes (10). However, there were some key differences.

DCB increased the proportion of cells in the mononuclear octoploid population, and the 7-day LI in the binucleated octoploid population (2X4N) of DEHP-treated animals was twice as high as that found in DCB-treated animals. To confirm and extend these studies, we compared the effects on ploidy/nuclearity and on the distribution of Sphase labeled cells of a second noncarcinogenic hepatomitogen, ETU, with those of 2 nongenotoxic liver carcinogens, methylclofenapate (MCP) and phenobarbitone. Phenobarbitone, a barbiturate drug that induces the activity of the cytochrome P-4502B and C subfamilies (19), is a rat liver mitogen and hepatocarcinogen (23, 25). MCP, like DEHP, is a member of the peroxisome-prolif-

Address correspondence to: Dr Susan Hasmall, Zeneca Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire SK10 4TJ, United Kingdom; e-mail: [email protected].

503 Downloaded from tpx.sagepub.com by guest on July 14, 2011

504

erating class of nongenotoxic hepatocarcinogens (24). The pleiotropic effects of these chemicals such as peroxisome proliferation, liver enlargement, and subsequent hepatocarcinogenesis are thought to be mediated, at least in part, by a member of the nuclear steroid receptor superfamily known as the peroxisome proliferator-activated receptor alpha (11, 13). Styles et al (30, 31) found that a single dose of MCP induced DNA synthesis predominantly in the binuclear tetraploid (2X2N) population. In contrast, Millar et al (15) found that after repeated daily doses of MCP the majority of S-phase labeled hepatocytes were of the 4N population. In neither of these studies was the proportion of labeled cells within each ploidy/ nuclearity subpopulation determined. To confirm these findings and extend these studies, we compared the effects of MCP, ETU, and phenobarbitone on ploidy/nuclearity and on the proportion of S-phase labeled cells within each individual ploidy/nuclearity subpopulation.

Again, hepatocytes

were

exposed

to

bromodeoxyuridine

(BrdU) for the full 7 days of compound administration, allowing BrdU labeling of all DNA-replicating cells. However, during the 7-day period, BrdU-labeled cells could be disseminated into different nuclearity and/or ploidy populations. Therefore, in a second study we used a short BrdU labeling window (1 hour) during the peak of DNA synthesis to determine the nuclearity and ploidy of the target population for MCP-, DEHP-, phenobarbitone-, ETU-, and DCB-induced DNA synthesis. The data obtained add to previous findings, extending our understanding of chemically induced hepatic DNA synthesis. MATERIALS

AND

METHODS

Chemicals. MCP was supplied by Lancaster Synthesis Ltd (Morecombe, Lancashire, UK). ETU, phenobarbitone, DEHP, and DCB were purchased from Sigma-Aldrich Chemical Co (Dorset, UK). Unless stated otherwise, all other chemicals were of the highest available purity and were purchased from Sigma-Aldrich. Animals. Male F344 rats (6-8 weeks old on arrival) were purchased from Harlan Olac Ltd (Bicester, United Kingdom) and acclimatized to the appropriate control diet for 1 week prior to dosing. Animals were housed 3 or 4 per cage, in appropriate conditions of temperature and humidity and with a 12-hour light/dark cycle. Food and water were available ad libitum. Animal Procedures. Seven-Day Studies: Dose levels, routes and regimes were chosen to mimic those employed in previously published cancer bioassays (3, 18, 23-25). MCP (0.1 % w/w in diet) was dosed in parallel with the first control group, and ETU (83 ppm diet) and phenobarbitone (500 mg/ml drinking water) were dosed in parallel with the second control group. BrdU (15 mg/ml in saline) was administered to all animals throughout the 7day dosing period by continuous infusion from Alzet osmotic minipumps (Alza, Palo Alto, CA, USA) implanted

subcutaneously. Thirty-Hour Studies: MCP (120 mg/kg in corn oil), ETU (8 mg/kg in corn oil), and phenobarbitone (65 mg/ kg in saline) were each dosed in parallel with 1 of 3 separate vehicle control groups, 4 rats/group. DEHP (950

mg/kg) and DCB (300 mg/kg in corn oil) were dosed in parallel with a fourth vehicle control group, 3 rats/group. As described previously (9), dose levels were chosen to correlate with those used in published cancer bioassays (3, 16, 17, 23, 25). All rats were dosed by gavage at 0 and 24 hours. To permit identification of the target population for DNA synthesis, BrdU (45 mg/kg in saline) was given intraperitoneally to all animals 1 hour prior to termination, and the hepatic labeling index was determined for the 1-hour period 29-30 hours after the initial dose. Determination of Ploidy, Nuclearity, and LI. Rats were anesthetized with ether, and hepatocytes were prepared by a 2-step collagenase perfusion as previously described (5). Cells were prepared for flow cytometry by the method of Styles et al (30) with minor modifications (10). Ploidy was analyzed on the basis of propidium iodide fluorescence using flow cytometry, and cells were sorted into ploidy classes on individual glass slides. A minimum of 10,000 cells were counted to determine ploidy profiles. Cell nuclearity and LI were determined for slide-mounted hepatocytes as described previously (10). Nuclearity profiles were determined from the proportion of mono- and binuclear cells within each ploidy class together with the ploidy profile. Both LI and nuclearity were determined for a minimum of 2,000 cells in each of the diploid, tetraploid, and octoploid subpopulations. Ploidy/Nuclearity of BrdU-Labeled and Unlabeled Cells after 30 Hours of Exposure. To permit accurate comparison between chemicals, where the S-phase response and mean LI may differ, we examined the effect of each chemical on distribution of labeled and unlabeled cells. The contribution of each subpopulation to the LI of the whole population was determined and the number of BrdU-positive cells within each subpopulation was expressed as a percentage of the total number of BrdUpositive cells. Likewise, the contribution of each subpopulation of unlabeled cells to the whole population was expressed as a percentage of the total number of unlabeled cells. Statistical Methods. The proportion of cells in each nuclearity and ploidy subpopulation and the LI for each subpopulation were compared with the appropriate control group values using generalized linear models with a binomial error and logit link (20). In addition, the distributions of labeled and unlabeled cells in each ploidy/ nuclearity class were compared using analyses of variance for each group separately. All proportions were transformed using a double arcsine transformation (8) before analysis. All analyses were carried out using the GENMOD procedure (27). The mean LIs were analyzed using Student’s t-test. All values are expressed as mean ± SD, and difference were considered significant at the level of p ~ 0.05. RESULTS

Ploidy

and

Nuclearity Profile after

7

Days of Treatment

Treatment of rats with MCP, ETU, or phenobarbitone for 7 days had no significant effect on the ploidy profile as compared with corn oil-treated controls (data not


505 TABLE 1.-Mean (± SD) hepatocyte exposure to BrdU and chemical.

labeling

index for 7

days

of

°n = 4. ’~

p ~ 0.05

as

compared

with concurrent vehicle control.

control groups (Figure 3). DEHP did reduced the binuclear octoploid population but had no effect on the proportion of cells in any other subpopulation. The ploidy and nuclearity profile for the saline and corn oil vehicle control groups were not significantly different from each other. LI

after

of Pulse Label hepatic LI after 1 hour of pulse label

1 Hour

The mean sured 29-30 hours after

FIGURE I.-Effect of chemicals on the proportion of hepatocytes within each hepatocyte subpopulation after a 7-day exposure period. DNA content was analyzed and cells separated into diploid (2N), tetraploid, and octoploid populations by flow cytometry. The tetraploid and octoploid populations were subdivided into mononuclear and bin4. uclear cells by light microscopy. Bars represent mean ± SD, n *Significantly different from concurrent control at p = 0.05. =

shown). MCP and phenobarbitone did induce significant changes in nuclearity (Figure 1). MCP reduced the binuclear tetraploid (2 X 2N) population and increased the mononuclear octoploid (8N) population. Phenobarbitone

mea-

with MCP, DEHP, DCB, ETU, or phenobarbitone is shown in Table 2. MCP, DEHP, DCB, and ETU all induced a significant increase in mean hepatic LI. Phenobarbitone induced only a slight increase in LI during the 29-30-hour labeling period, and this effect was not significant; therefore, phenobarbitonetreated animals were excluded from analysis of nuclearity and distribution of LI at the 30-hour time point. The LI within each ploidy and nuclearity subpopulation is shown in Figure 4. Statistical analyses showed that the increase in LI was not distributed evenly among the nuclearity treatment

similarly increased the proportion of cells in the 4N population. ETU had no effect on the nuclearity profile as compared with the control (Figure 1). LI

after

7

Days of Treatment

Seven days of continuous exposure to MCP, ETU, or phenobarbitone induced a significant increase in the mean hepatic LI (Table 1). The LI within each ploidy and nuclearity subpopulation is shown in Figure 2. As predicted from previous studies (10, 15), the increase in LI was not distributed evenly among the subpopulations; in general, LI was higher in cells of higher ploidy. MCP induced a significant increase in LI in all hepatocyte subpopulations. Phenobarbitone increased the proportion of labeled cells in both of the tetraploid (2X2N and 4N) and octoploid (2X4N and 8N) populations but not in the diploid (2N) population. ETU increased LI significantly in the 2N, binuclear tetraploid (2X2N), and mononuclear octoploid populations (8N).

Ploidy and Nuclearity Profile after

30 Hours

of

Treatment Treatment of rats with MCP, DEHP, DCB, or ETU for 30 hours had no significant effect on the ploidy profile compared with the respective vehicle controls (data not shown). In addition, MCP, ETU, and DCB had no effect on the nuclearity profile compared with the respective

FIGURE 2.-LI within cells of each hepatic ploidy and nuclearity subafter 7 days of exposure to chemical and BrdU. LI is exa percentage of each subpopulation. Bars represent mean values ± SD, n 4. *Significantly different from concurrent control at p S 0.05.

population pressed as


=

506 TABLE 2.-Mean (±SD) hepatocyte labeling index for 1-hour label 30 hours after the first of 2 daily doses of chemical.

°n?3. * p ~ 0.05

FIGURE 3.-Effect of chemicals on the proportion of hepatocytes within each hepatocyte subpopulation 30 hours after the first of 2 daily doses. DNA content was analyzed and cells separated into diploid (2N), tetraploid, and octoploid populations by flow cytometry. The tetraploid and octoploid populations were subdivided into mononuclear and binuclear cells by light microscopy. Bars represent mean ± SD, n 2: 3. *Significantly different from concurrent vehicle control at p :5 0.05.

classes and that the profiles of LI induced by MCP, DEHP, DCB, and ETU differed. MCP and DEHP had similar effects on LI; both increased LI in all subpopulations of hepatocytes except the 2 X 2N population. The highest LI was seen in the 2X4N and 8N populations. In addition, MCP and DEHP significantly increased the LI in both the 2N and 4N populations as compared with controls. DCB induced a significant increase in LI in mononuclear cells only. DCB appeared to increase LI in binuclear octoploid cells, but this increase was not significant. Neither DCB nor DEHP increased the LI in the binucleated tetraploid population. The LI profile induced by ETU was not significantly different from control values despite a significant increase in the mean hepatic LI (2.5-fold over control), suggesting that ETU induced Sphase randomly across all subpopulations.

Distribution of BrdU-Labeled and Unlabeled Cells 30 Hours of Exposure

as

compared

pulse

with concurrent vehicle control.

the labeled population was made up of octoploid cells (2X4N and 8N) compared with the distribution of unlabeled hepatocytes. Similarly, with DEHP a significantly smaller proportion of the labeled population were 4N cells and a greater proportion were octoploid (2X4 and 8N). In the concurrent control groups, the distribution of labeled cells among the hepatocyte subpopulations was the same as the distribution of unlabeled cells. Furthermore, following DCB-induced cell labeling there were no significant differences in ploidy/nuclearity profiles between BrdU-labeled and unlabeled cells. For ETU, the distribution of labeled cells among the hepatocyte subpopulations was different from the distribution of unlabeled cells, but the changes were the same as observed in the concurrent corn oil control.

after

The contribution of each subpopulation to the LI in the whole population during the 1-hour pulse label was compared with the contribution of unlabeled cells to the whole population of unlabeled cells (Figure 5). In all treated and control groups, the majority of labeled and unlabeled cells were in the 4N population. Following MCP- and DEHP-induced cell labeling, there were marked differences in the profiles of labeled and unlabeled cells. For MCP, a significantly smaller proportion of the labeled population was made up of tetraploid cells (2X2N and 4N) and a significantly larger proportion of

FIGURE 4.-LI within cells of each hepatic ploidy and nuclearity sub30 hours after the first of 2 daily doses. LI is expressed as a percentage of each subpopulation. Bars represent mean values ± SD, n 2: 3. *Significantly different from concurrent vehicle control at p = 0.05.

population


507

high LI in polyploid cells after 7 days of dosing with either of the peroxisome proliferators MCP or WY14,643 (15). Taken together, these data provide evidence for selective induction of DNA synthesis in specific target a

populations. The labeled cells

seen at

the

7-day

time

point may

not

represent the major target population for induced DNA

FIGURE 5.-Distribution of BrdU-labeled

(M)

and unlabeled

(D)

he-

patocytes among hepatocyte subpopulations. The number of labeled cells within each subpopulation is expressed as a percentage of the total number of labeled cells and

compared with the distribution of unlabeled

cells expressed as a percentage of the total number of unlabeled cells. Bars represent mean values ± SD, n 2: 3, for each chemical. *Distributions of labeled and unlabeled cells are significantly different at p =

0.05.

DISCUSSION

Induction of rodent liver cancer by nongenotoxic chemicals is frequently associated with increased DNA synthesis, but previous data (10, 21), have suggested that noncarcinogenic mitogens and nongenotoxic hepatocarcinogens may differ in their target populations. To investigate this potential correlation, we examined the effect of 3 nongenotoxic hepatomitogens on hepatocyte ploidy/ nuclearity and distribution of LI to correlate these changes with carcinogenic potential. The data presented confirm that the rodent hepatocarcinogens MCP and phenobarbitone induced a significant increase in the proportion of cells in the 4N population. The noncarcinogenic liver

mitogen ETU did not alter the ploidy/nuclearity profile after 7 days of exposure. Similarly, MCP and phenobarbitone resulted in a high LI in the 2X4N and 4N subwhereas the nonhepatocarcinogen ETU did this effect. This general elevation in LI in 8N hepatocytes coupled with a specific LI in 2X4N hepatocytes for hepatocarcinogens is consistent with our previous observation for DEHP and DCB (10). DEHP induced a much greater increase in LI in the 2X4N population than did DCB, a chemical that is not hepatocarcinogenic in the rat. Furthermore, these data are consistent with the absence of labeling in the 2X2N population and

populations, not cause

synthesis because label may have been disseminated into other subpopulations by nuclear and/or cytoplasmic division. To examine this possibility, we used a 1-hour pulse label at the end of a 30-hour dosing period, during which the label should remain within the target population. All chemicals, with the exception of phenobarbitone, induced a significant increase in mean hepatic LI. Consistent with the 7-day data, label was not distributed randomly among the hepatocyte subpopulations, suggesting that some hepatocyte subpopulations may be more susceptible to chemically induced DNA synthesis. The LI was always highest in 8N hepatocytes and, despite the increase in mean hepatic LI, there was no significant increase in LI in the 2X2N population. In 2X4N hepatocytes, LI was higher after exposure to the hepatocarcinogens MCP and DEHP than after exposure to the nonhepatocarcinogens DCB and ETU. These data suggest that changes in hepatic nuclearity/ ploidy take place during longer exposure to chemicals and reflect a trend towards increasing ploidy. These findings are consistent with the MCP-induced increase in octoploid cells and corresponding decrease in tetraploid cells seen after 5 days but not after 2 days of dosing (14) and with findings of increased polyploidy induced by In addition, the shift from the 2X2N to the 8N population seen after 7 days is consistent with the reduced binuclearity seen after treatment with cyproterone acetate, pregnenolone-16a carbonitrile, and a-hexa-

Wy 14,643 (12).

chlorocyclohexane (29). This increasing polyploidization, coupled with a high LI in cells of higher ploidy, might be predicted to result from DNA synthesis in cells of lower ploidy increasing the ploidy status of the cells. However, overall the effects of MCQ ETU, DCB, and DEHP

the distribution of LI during a 1-hour pulse window were similar to the 7-day LI profiles of MCP and ETU and to those reported previously for DCB and DEHP (10). The highest LI was in the higher ploidy/ nuclearity classes, particularly the 8N populations of hepatocytes. Hence, the cells of higher ploidy may be the result of induction of S-phase rather than being the product of DNA synthesis in cells of a lower ploidy status. The similarity of the distribution of LI during the 1-hour and the 7-day labeling window suggests that most BrdUlabeled hepatocytes remain within the same ploidy/nuclearity subpopulation. Thus, when chemicals induce Sphase in the 8N population, the cells may undergo both nuclear and cytoplasmic division to produce progeny of the same ploidy and nuclearity. In addition the shift towards polyploidization suggests some binuclear cells may undergo DNA synthesis followed by mitosis, giving rise to 2 progeny each with the same cellular ploidy as the parent cell but with increased nuclear ploidy. Whether this is a consequence of accelerating the rate of age-related increase in polyploidization (6, 7) or a direct effect on

labeling


508 of the chemical remains unclear. The relationship between increased ploidy and the high incidence of diploid cells reported in experimental hepatocarcinogenesis (26) is also unclear. Many human hepatocarcinomas are heterogenous, with both diploid and aneuploid regions, and it has been suggested that these regions may arise from changes in growth patterns and dedifferentiation (22). Similarly, dedifferentiation in rodent tumors may give rise to diploid cells. Comparison of the distribution of BrdU-labeled and unlabeled cells after exposure to mitogens illustrated some clear differences (Figure 5). The increased proportion of labeled cells in the 2X4N and 8N populations after exposure to MCP and DEHP suggests that these rodent hepatocarcinogens stimulated DNA synthesis in both the 8N and 2X4N octoploid populations. In contrast, the noncarcinogens ETU and DCB did not increase the proportion of labeled 2X4N cells. Furthermore, the distributions of labeled and unlabeled cells after treatment with either DCB or ETU were similar to those of the respective controls. These data highlight differences in the effects on hepatic nuclearity and LI of the nongenotoxic rodent carcinogens MCP, phenobarbitone, and DEHP and the nonhepatocarcinogenic mitogens ETU and DCB. The differences in the effects of chemicals on DNA

populations are consisthat DEHP induced Sphase in almost twice as many 2X4N hepatocytes as did DCB (10). However, it is not clear whether the differences detected are related mechanistically to the differing carcinogenic potential of these mitogens. The regulatory factors involved in cell cycle control may have different effects in hepatocytes of differing nuclearity and ploidy, hence some populations may be more susceptible to chemically induced DNA synthesis. The tumor suppressor gene, p53, may play a role in controling hepatocyte proliferation and apoptosis, but it appears only to be effective once hepatocytes have been recruited into the cell cycle (4). The increasing polyploidization that occurs in hepatocytes as part of the ageing process seems unaffected by p53 (4). However, p53 deficiency may affect the ratio of mononuclear to binuclear hepatocytes within individual ploidy classes. Furthermore, the rate of agerelated increase in ploidy is reduced in p53-deficient mice (32). Differences in the response of hepatocytes of different ploidy and nuclearity to regulatory factors involved in cell cycle control could place some populations, eg, the 2X4N population, at greater risk of unrestricted cell proliferation. In summary, we have shown that diverse chemical mitogens, MCP, ETU phenobarbitone, DEHP, and DCB, induce DNA synthesis more readily in hepatocytes with higher ploidy levels. The similarity in distribution of LI after I hour of pulse labeling and after 7 days of labeling suggests that on stimulation of DNA synthesis hepatocytes may undergo both nuclear and cytoplasmic division, producing progeny within the same subpopulation. In addition, the nongenotoxic rodent liver carcinogens MCP and DEHP induced a significant increase in the LI in the 24N population, whereas ETU and DCB did not. synthesis

within different target

tent with our

previous findings

These data support the hypothesis that increased DNA synthesis within the minority 24N population may be more significant for carcinogenic outcome than is the overall LI. ACKNOWLEDGMENTS We thank Mr Andrew Heuze for performing the preparative flow cytometry and Mr I. Pate of the Zeneca CTL statistics department for assistance with statistical

analyses. REFERENCES 1. Ames BN, Gold LS (1990). Too many rodent cacinogens: Mitogenesis increases mutagenesis. Science 249: 970-971. 2. Ashby J, Tennant R (1991). Definitive relationships among chemical structure, carcinogenicity and mutagenicity for 301 chemicals tested by the US NTP. Mutat Res 257: 229-306. 3. Azarnoff DL, Reddy J, Hignite C, Fitzgerald T (1976). Structure activity relationships of clofibrate-like compounds and lipid metabolism. Proc 6th Int Pharmacol Meet 4: 137-147. 4. Bellamy COC, Clarke AR, Wyllie AH, Harrison DJ (1997). p53 deficiency in liver reduces local control of survival and proliferation, but does not affect apoptosis after DNA damage. FASEB J 11: 591-599. 5. Berry MN, Friend DS (1969). High yield preparation of isolated rat liver parenchymal cells: A biochemical and fine structural study. J Cell Biol 43: 506-520. 6. Brodsky WY, Uryvaeva IV (1977). Cell ploidy: Its relation to tissue growth and function. Int Rev Cytol 50: 275-332. 7. Carriere R (1969). The growth of liver parenchymal nuclei and its endocrine regulation. Int Rev Cytol 25: 201-277. 8. Freeman M, Tukey J (1950). Transformations related to the angular and the square root. Ann Math Stat 21: 607. 9. Hasmall SC, Pyrah ITG, Soames AR, Roberts RA (1997). Expression of the Immediate early genes, c-fos, c-jun and c-myc: A comparison in rats of nongenotoxic hepatocarcinogens with noncarcinogenic liver mitogens. Fundam Appl Toxicol 40: 129-137. 10. Hasmall SC, Roberts RA (1997). Hepatic ploidy, nuclearity and distribution of DNA synthesis: A comparison of nongenotoxic hepatocarcinogens with noncarcinogenic liver mitogens. Toxicol Appl Pharmacol 144: 287-293. 11. Issemann I, Green S (1990). Activation of a member of the steroid hormoe receptor superfamily by peroxisome proliferators. Nature 347:645-650. 12. Lalwani ND, Dethloff LA, Haskins JR, Robertson DG, De La Inglesia FA (1997). Increased nuclear ploidy, not cell proliferation, is sustained in the peroxisome proliferator-treated rat liver. Toxicol Pathol 25: 165-176. 13. Lee S, Pineau T, Drago J, Lee EJ, Owens JO, Kroetz DL, Fernandez-Salguero PM, Westphal H, Gonzalez FJ (1995). Targeted disruption of the alpha isoform of the peroxisome proliferator-activated receptor gene in mice results in abolishment of the pleiotrophic effects of peroxisome proliferators. Mol Cell Biol 15: 30123022. 14. Millar RT, Glover SE, Stewart WS, Corton JC, Popp JA, Cattley RC (1996). Effect of the expression of c-met, c-myc and PPAR-A in liver and liver tumours from rats chronically exposed to the hepatocarcinogenic peroxisome proliferator WY-14,643. Carcinogenesis 17: 1337-1341. 15. Millar RT, Shah RS, Cattley RC, Popp JA (1996). The peroxisome proliferators WY-14,643 and methylclofenapate induce hepatocyte ploidy alterations and ploidy-specific DNA synthesis in F344 rats. Toxicol Appl Pharmacol 138: 317-323. 16. National Toxicology Program (1982). Carcinogenesis Bioassay of Di(ethylhexyl)phthalate (CAS No. 117-81-7) in F344 Rats and 1 Mice (Feed Study). Technical Report No 217, National B6C3F Institutes of Health Publication No 82-1773, National Institutes of Health, Research Triangle Park, North Carolina.


509 17. National Toxicology Program (1987). Toxicology and Carcinogenesis Studies of 1,4-Dichlorobenzene (CAS No. 106-46-7) in F344/ n Rats and B6C3F 1 Mice. Technical Report No 319, National Institutes of Health Publication No 87-2575, National Institutes of Health, Research Triangle Park, North Carolina. 18. National Toxicology Program (1992). Toxicology and Carcinogenesis Studies of Ethylene Thiourea (CAS No. 96-45-7) in F344 Rats and B6C3F 1 Mice. Technical Report No 388, National Institutes of Health, Research Triangle Park, North Carolina. 19. Nebert D, Gonzalez F (1987). P450 genes: Structure, evolution, and regulation. Annu Rev Biochem 56: 945-993. 20. Nelder JA, Wedderburn RWM (1972). Generalised linear models. J R Stat Soc 135: 370-384. 21. Odum J, Foster JR, Hasmall SC, Tittensor S, Soames AR, Elcombe CR (1997). Markers of non-genotoxic carcinogenesis: 2. F344 rat liver. Hum Exp Toxicol 16: 56. 22. Oriyama T, Yamanaka N, Fujimoto J, Ichikawa N, Okamoto E (1998). Progression of hepatocellular carcinoma as reflected by nuclear DNA ploidy and cellular differentiation. J Hepatol 28: 142149. 23. Phillips J, Price R, Cunninghame M, Osimitz T, Cockburn A, Gabriel K, Preiss F, Butler W, Lake B (1997). Effect of piperonyl butoxide on cell replication and xenobiotic metabolism in the livers of CD-1 mice and F344 rats. Fundam Appl Toxicol 38: 64-74. 24. Reddy JK, Lalwani ND, Reddy K, Qureshi SA (1982). Excessive accumulation of autofluorescent lipofuscin in the liver during he-

25.

patocarcinogenesis by methyl clofenapate and other hypolipidemic peroxisome proliferators. Cancer Res 42: 259-266. Rossi L, Ravera M, Repetti G, Santi L (1977). Long-term administration of DDT or phenobarbital-Na in Wistar rats. Int J Cancer

19:179-185. 26. Saeter G, Schwarze PE, Nesland JM, Juul N, Pettersen EO, Seglen PO (1988). The polyploidizing growth pattern of normal rat liver is replaced by divisional, diploid growth in hepatocellular nodules and carcinomas. Carcinogenesis 9: 939-945. 27. SAS Institute (1989). SAS/STAT User’s Guide, version 6, 4th ed. SAS Institute, Cary, North Carolina. 28. Schulte-Hermann R (1974). Induction of liver growth by xenobiotic compounds and other stimuli. CRC Crit Rev Toxicol 3: 97-158. 29. Schulte-Hermann R, Hoffmann V, Landgraf H (1980). Adaptive responses of the rat liver to the gestagen and antiandrogen cyproterone acetate and other inducers. III. Cytological changes. ChemBiol Interact 31: 301-311. 30. Styles JA, Kelly M, Elcombe CR (1987). A cytological comparison between regeneration, hyperplasia and early neoplasia in the rat liver. Carcinogenesis 8: 391-399. 31. Styles JA, Kelly M, Pritchard NR, Elcombe CR (1988). A species comparison of acute hyperplasia induced by the peroxisome proliferator methylcofenapate: Involvement of the binucleated hepatocyte. Carcinogenesis 9: 1647-1655. 32. Yin L, Ghebranious N, Chakraborty S, Sheehan CE, Ilic Z, Sell S (1998). Control of mouse hepatocyte proliferation and ploidy by p53 and p53ser246 mutation in vivo. Hepatology 27: 73-80.
