Comparative Prevalence, Multiplicity, and Progression of Spontaneous and Vinyl Carbamate-Induced Liver Lesions in Five Strains of Male Mice

Transcription

1 TOXICOLOGIC PATHOLOGY, vol 30, no 5, pp , 2002 Copyright C 2002 by the Society of Toxicologic Pathology DOI: / Comparative Prevalence, Multiplicity, and Progression of Spontaneous and Vinyl Carbamate-Induced Liver Lesions in Five Strains of Male Mice KIMIMASA TAKAHASHI, 1 GREGG E. DINSE, 2 JULIE F. FOLEY, 3 JERRY F. HARDISTY, 4 AND ROBERT R. MARONPOT 3 1 Nippon Veterinary and Animal Science University, Tokyo, Japan 2 Biostatistics Branch, National Institute of Environmental Health Sciences, NIH, Research Triangle Park, North Carolina 3 Laboratory of Experimental Pathology, National Institute of Environmental Health Sciences, NIH, Research Triangle Park, North Carolina, and 4 Experimental Pathology Laboratories, Inc, Research Triangle Park, North Carolina ABSTRACT The overall and age-speci c prevalences and multiplicities of spontaneous and chemically induced hepatocellular neoplasia were compared among male B6D2F1, B6C3F1, C3H (C3H/HeNCrl MTV-), B6CF1, and C57BL/6 (C57BL/6NCrl) mice following a single intraperitoneal injection of 0.03 l M vinyl carbamate (VC)/g body weight or vehicle alone at 15 days of age. Additional groups of B6C3F1, C3H, and C57BL/6 males received 0.15 l M VC/g body weight at 15 days of age. For male B6D2F1, B6C3F1, C3H, B6CF1, and C57BL/6 mice, the estimated overall prevalences (and multiplicities) of hepatocellular adenomas or carcinomas in vehicle controls were 14.1% (0.19), 12.3% (0.15), 8.2% (0.10), 7.2% (0.09), and 2.4% (0.02), respectively. The analogous estimates in the low-dose group were 59.2% (1.19), 72.9% (4.07), 48.6% (1.99), 22.8% (0.29), and 43.9% (0.82). Analogous estimates for B6C3F1, C3H, and C57BL/6 mice in the high-dose group were 45.3% (4.29), 59.7% (6.63), and 46.8% (1.74), respectively. Age-speci c multiplicity estimates suggested a progression from altered hepatocellular foci (AHF) to hepatocellular neoplasms. Further evidence of progression was provided by the temporal occurrence of hepatocellular adenomas before carcinomas, and the apparent origination of carcinomas within adenomas. Pulmonary metastases were observed in many of the mice with hepatocellular carcinomas. These ndings con rm previous observations of strain differences in liver neoplasm response, suggest a progressive development from AHF to adenomas, and ultimately to carcinomas, and show sensitivity to VC-induced hepatocarcinogenesi s in all 5 strains. Keywords. Hepatocarcinogenesis ; tumor progression; neonatal mouse model. INTRODUCTION Mouse liver tumors occur spontaneously with purported preferential strain sensitivity (12) but the strain responsiveness to chemical induction has not been extensively investigated and may not always re ect the spontaneous liver tumor susceptibility (18, 34). Although interpretation of strain differences and the usage of mice for assessing potential health hazards to humans continue to be actively debated (1, 28), models of mouse hepatocarcinogenesis remain valuable in the study of factors in uencing carcinogenesis. Mechanistic and molecular events associated with mouse liver tumor induction have been previously investigated, and strain differences in clonal expansion of initiated cells (12) and chemical and strain differences in frequencies and patterns of ras oncogene activation have been documented (30). The differential strain susceptibility to liver tumor induction may be a consequence of postinitiation events, because sensitive and resistant strains are equally capable of forming DNA adducts following treatment with agents such as diethylnitrosamine (DEN) and ethylnitrosourea (ENU) (13, 20). Activation of ras protooncogenes has been identi ed as an early event and potentially important factor associated with a signi cant number of spontaneous and chemically induced mouse hepatocellular neoplasms (32, 33, 41). Strain differences exist, however, in ras activation, and mouse hepatocel- Address correspondenc e to: R. R. Maronpot, National Institute of Environmental Health Sciences, PO Box 12233, 111 Alexander Drive, Research Triangle Park, North Carolina 27709; maronpot@niehs.nih.go v lular neoplasms may occur without it (4, 14, 30). In the course of generating liver tumors for molecular studies in susceptible and resistant strains, we noted that liver tumors in more resistant strains exhibited a longer latency than in susceptible strains. This study of the pathogenesis of liver tumor induction in susceptible and resistant strains was performed to help clarify differential strain susceptibility. Vinyl carbamate (VC), a potent rodent carcinogen, has induced skin, lung, thymic, Harderian gland, and hepatic tumors in mice as well as hepatic tumors, ear duct carcinomas, and neuro brosarcomas of the ear in rats (4, 7, 41, 42). VC is mutagenic in Salmonella typhimurium in the presence of metabolic activating enzymes and a potent in vivo and in vitro inducer of sister chromatid exchanges in mouse cells (7, 8). It is believed to be the proximate carcinogenic metabolite of ethyl carbamate (2, 19). Its known mutagenic and metabolic pro le and the organ-speci c liver tumor response in neonatally exposed mice make vinyl carbamate a potentially useful agent with which to study strain sensitivities and pathogenesis of hepatic neoplasia. The neonatal mouse liver tumor model (36) is particularly appealing for studying hepatocarcinogenesis, because there is a single, brief exposure to the carcinogen, and subsequent tumor development is not complicated by continuous exposure. The purpose of this study was to compare the agespeci c prevalence, multiplicity, and progression of vinyl carbamate-induced liver neoplasms among 5 strains of mice. Mouse strains were selected based on commercial availability and anticipated differences in sensitivity to carcinogeninduced hepatocarcinogenesis. The working hypothesis was /02$3.00 $0.00

2 600 TAKAHASHI ET AL TOXICOLOGIC PATHOLOGY that all mouse strains are susceptible to genotoxic hepatocarcinogens, but apparent differences in susceptibility are a consequence of tumor latency. An experimental approach was selected whereby carcinogen exposure was limited to a single, brief neonatal exposure to VC without any subsequent treatment. For 3 (B6C3F1, C3H, and C57BL/6) of the 5 strains, 2 different neonatal doses of VC were compared, whereas only 1 dose was used for the other 2 strains (B6D2F1 and B6CF1). MATERIALS AND METHODS Background Studies were conducted over a 4-year period to compare strain differences in liver tumor responses and ras oncogene pro les following a singleintraperitoneal injection of thehepatocarcinogen, VC, or saline vehicle. Pure VC was synthesized by Midwest Research Institute (Kansas City, MO). All studies followed the same experimental protocol. The time intervals between sacri ces were narrow to maximize the opportunity to collect proliferative hepatic lesions throughout tumor progression. Three to 5 mice per group were randomly selected for sacri ce at 3- to 5-week intervals based on the ndings at the previous sacri ce intervals taking into account the differential lesion progression among strains. The rst sacri ce for B6C3F1, C57BL/6, and C3H mice was performed at 36 days of age. The rst sacri ce for B6D2F1 and B6CF1 mice was performed at 190 days of age. Moribund sacri ces were performed ad hoc as necessary. The nal sacri ce for each dose group and strain occurred when 6 or fewer mice remained and ranged from 448 to 869 days of age. The molecular study results have been reported, and summaries can be found in a comprehensive review (30). Details of the histopathologic ndings in the liver are the subject of the present paper. Regarding strain selection, we examined the commonly used B6C3F1 hybrid and the 2 parental inbred strains (C57BL/6 and C3H). In addition, the B6D2F1 hybrid (a cross of C57BL/6 and DBA/2) was included because it has been proposed as an alternative to the B6C3F1 hybrid. Finally, the B6CF1 hybrid was included because it is derived from 2 inbred strains (C57BL/6 and Balb/C) that are considered to be of relatively low susceptibility to spontaneous or chemically induced hepatocarcinogenesis. Because of cost considerations, the study was restricted to males. Animals and Animal Husbandry Eighty virgin female C57BL/6 (C57BL/6NCrl) mice were mated to male BALB/c (BALB/cAnNCrl) mice to produce B6CF1 hybrids; 80 virgin female C57BL/6 mice were mated to male DBA/2 (DBA/2NCrl) mice to produce B6D2F1 hybrids; and 85 virgin female C57BL/6 mice were mated to male C3H (C3H/HeNCrl MTV-) mice to produce B6C3F1 hybrids. Pregnant C57BL/6 and C3H mice were purchased to obtain offspring for these inbred strains. All mice were purchased from Charles River Laboratories (Raleigh, NC). All mice were housed in an environment maintained at 22 2 C, with 50% 20% relative humidity and 12-hour light/dark cycles, and in compliance with NIH guidelines for the humane care and use of laboratory animals. NIH-07 pellet feed and deionized water were supplied ad libitum. All pups were given a single 0.1-ml intraperitoneal injection of either 0.03 or 0.15 l M VC/g body weight, or saline on day 15 of age. Mice were examined daily for moribundity and mortality and were weighed monthly for the rst 14 months and then bimonthly. The numbers of mice used for our strain comparison are detailed in Table 1. Mice sacri ced when moribund or those found dead were not included in the analysis of altered hepatocellular foci (AHF) because they were evaluated by a pathologist not involved in the evaluation of the scheduled sacri ces. All mice were used in the analysis of hepatocellular tumors. Pathology At necropsy livers were weighed, and livers and lungs were examined for the presence of neoplasms and other lesions. All grossly visible liver masses were recorded. Representative samples of each liver mass plus standardized sections from the left, median, and right lobes were xed in 10% neutral buffered formalin for hours and then transferred to 70% ethanol. Fixed tissues were trimmed, embedded in paraf- n within 72 hours, sectioned, and stained with hematoxylin and eosin. Published criteria for diagnosis of hepatoproliferative lesions in the mouse were followed (29, 31). For each mouse, AHF present in the stardardized sections were phenotypically classi ed as basophilic, eosinophilic, clear cell, vacuolated, or mixed, and then they were counted. Consequently, equivalent amounts of tissue from each mouse were evaluated for AHF. Representative and equivalent-sized sections of the left and right lung lobes were examined for the presence of metastases and other lesions. Statistical Analyses Our analysis focused on 2 measures of response: prevalence and multiplicity. We estimated the prevalence and multiplicity of hepatocellular adenomas (HA), hepatocellular carcinomas (HC), and total hepatocellular neoplasms (HA/HC). Prevalence refers to the proportion of tumorbearing mice. Our analysis de ned multiplicity as the average number of neoplasms per mouse, as opposed to an alternative convention that de nes multiplicity as the average number of neoplasms per tumor-bearing mouse. Estimates based on the former are more stable because the reference sets are larger. In addition to neoplasm multiplicity, we also estimated the multiplicity of AHF, which was de ned as the average number of AHF per set of standardized liver sections. Both overall and age-speci c estimates of prevalence and multiplicity were calculated. We used logistic regression to model neoplasm prevalence as a continuous function of age (9). Our model automatically adjusts for differential survival by including the time of death as an explanatory variable. This analysis assumes that all hepatocellular neoplasms are incidental (ie, they do not affect longevity), in which case the mice found dead, sacri ced at random intervals, and sacri ced when moribund provide unbiased cross-sectional information on neoplasm status at multiple times. The data suggest that this incidental neoplasms assumption was reasonable. In a similar manner, we used Poisson regression (16) to model multiplicity as a continuous function of age. Whereas only a linear time term was included in the logistic prevalence

3 Vol. 30, No. 5, 2002 MOUSE LIVER LESIONS 601 model, which forces monotonicity, a quadratic time term was considered in the Poisson multiplicity model, which allows the slope to change direction. For example, if liver foci can regress over time, then a plot of multiplicity against age might initially increase but later turn downward. A quadratic term was included only if it signi cantly improved the t of the model. RESULTS Liver Lesions in Control Mice Among the subtypes of AHF, basophilic and mixed foci were most commonly observed in these strains. The overall prevalences (and the earliest ages of observation) of hepatocellular adenomas or carcinomas in B6D2F1, B6C3F1, C3H, B6CF1, and C57BL/6 mice were 14.1% (15 months), 12.3% (10 months), 8.2% (13 months), 7.2% (21 months), and 2.4% (14 months), respectively. Morphological features of hepatocellular neoplasms were similar among all 5 strains of mice. Pulmonary metastases of hepatocellular carcinomas were seen in B6C3F1 and C3H mice (Table 1). As for other tumors present in the liver, hemangiosarcomas were observed in 1 (0.7%) of the B6C3F1 mice, 5 (5.2%) of the B6CF1 mice, 2 (3.1%) of the B6D2F1 mice, and 2 (1.2%) of the C57BL/6 mice; lymphomas in 1 (0.7%) of the B6C3F1 mice, 5 (5.2%) of the B6CF1 mice, 1 (1.6%) of the B6D2F1 mice, and 9 (5.4%) of the C57BL/6 mice; and histiocytic sarcomas in 4 (2.4%) of the C57BL/6 mice. Liver Lesions in VC-Treated Mice Among the subtypes of AHF, basophilic foci occurred most commonly followed by mixed foci. Treated mice had earlier times of rst AHF observation (3 13 months) and higher overall multiplicities ( ) than the respective controls (8 17 months and AHF). In B6C3F1, C3H, and C57BL/6 mice, this difference was more prominent in the high-dose than low-dose groups. The overall prevalences (and earliest ages of observation) of hepatocellular adenomas or carcinomas in B6D2F1, B6C3F1, C3H, B6CF1, and C57BL/6 mice treated with 0.03 l M VC/g body weight were 59.2% (14 months), 72.9% (10 months), 48.6% (9 months), 22.8% (19 months), and 43.9% (12 months), respectively. Similarly, in B6C3F1, C3H, and C57BL/6 mice treated with 0.15 l M VC/g body weight, the overall prevalences (and earliest age of observation) were 45.3% (7 months), 59.7% (7 months), and 46.8% (7 months), respectively. Histomorphological features of hepatocellular neoplasms were similar among all 5 strains and similar to those observed in the vehicle controls. Among VC-treated mice, a few pulmonary metastases of hepatocellular carcinomas were seen (Table 1), with the highest metastatic frequency in the B6D2F1 and B6C3F1 strains. In addition to hepatocellular neoplasms, myeloid leukemia was found in 1 (0.7%) of the high-dose C3H mice and 1 (0.4%) of the high-dose C57BL/6 mice; hemangiosarcoma in 1 (1.4%) of the low-dose B6C3F1 mice, 1 (0.8%) of the highdose B6C3F1 mice, 2 (1.5%) of the low-dose B6D2F1 mice, 2 (1.9%) of the low-dose C57BL/6 mice, and 3 (1.3%) of the high-dose C57BL/6 mice; lymphoma in 1 (0.8%) of the highdose B6C3F1 mice, 1 (0.9%) of the low-dose B6CF1 mice, 2 (1.5%) of the low-dose B6D2F1 mice, 2 (1.9%) of the lowdose C57BL/6 mice, and 5 (2.2%) of the high-dose C57BL/6 mice; hemangioma in 3 (2.3%) of the low-dose B6D2F1 mice and 3 (1.3%) of the high-dose C57BL/6 mice; and histiocytic sarcomas in 1 (0.9%) of the low-dose C57BL/6 mice. Age-Speci c Prevalence of Hepatocellular Neoplasms The prevalence of hepatocellular neoplasms varied with age, treatment, and strain (see Figure 1). All prevalence estimates increased with age and VC dose. The age effects varied across strains, even among controls. Relative to the baseline rates among controls, the increase with age was more rapid and occurred earlier among mice given the low dose of VC, and the effect was even more marked in the high-dose group. The smallest prevalence estimates and the smallest increases with age occurred in B6CF1 mice. The largest age effects occurred in C3H and B6C3F1 mice, with intermediate effects observed in B6D2F1 and C57BL/6 mice. In the most extreme case, based on the estimated prevalences, virtually all high-dose C3H mice had at TABLE 1. Hepatocellular proliferative lesions and pulmonary metastases in ve strains of male mice. Strain Number examined for AHF Overall multiplicity of AHF Number Overall prevalence (%) of: Overall multiplicity of: examined for tumors HA HC HA/HC HA HC HA/HC Total mice with HC Mice with HC in HA Control B6D2F B6C3F C3H B6CF C57BL/ Low-dose B6D2F B6C3F C3H B6CF C57BL/ High-dose B6C3F C3H C57BL/ Mice with PM Note: AHF altered hepatocellular foci; HA hepatocellular adenoma; HC hepatocellular carcinoma; HA/HC hepatocellular adenoma or carcinoma; PM pulmonary metastases; Overall multiplicity the number of tumors (or AHF) divided by the number of mice (or standardized liver sections).

4 602 TAKAHASHI ET AL TOXICOLOGIC PATHOLOGY FIGURE 1. Age-speci c prevalence of hepatocellular neoplasms in male mice by dose and strain. (.....) control, ( ) 0.03 l M VC/g BW, (-.-.-.) 0.15 l M VC/g BW. least 1 hepatocellular adenoma or carcinoma by 300 days of age. Age-Speci c Multiplicity of Hepatocellular Lesions The effects of age and VC dose on the multiplicity of hepatocellular lesions varied with strain (see Figure 2). The average number of hepatic lesions was extremely small at all ages and for all strains of control mice, as evidenced by the bare visibility of the dashed curves in Figure 2 in any of the subplots. These negligible age-speci c multiplicities did not appear to increase with dose in B6CF1 mice, and the low dose of VC exerted minimal effect in B6D2F1 mice. In contrast, the multiplicity estimates varied greatly with age and dose in B6C3F1 and C3H mice and moderately in C57BL/6 mice. FIGURE 2. Age-speci c multiplicity of hepatocellular lesions in male mice by dose and strain. (.....) control, ( ) 0.03 lm VC/g BW, (-.-.-.) 0.15 l M VC/g BW.

5 Vol. 30, No. 5, 2002 MOUSE LIVER LESIONS 603 In these last 3 strains of mice, the estimated multiplicities of AHF and hepatocellular adenomas clearly increased with age initially and later decreased. The multiplicity of hepatocellular carcinomas appeared to peak at some intermediate age. There was also a de nite dose effect in strains (B6C3F1, C3H, and C57BL/6) receiving 2 different doses of VC. The multiplicity by age curves shift with dose the peaks are higher and occur earlier as VC dose increases. For example, the maximum estimated number of AHF per standardized liver section in B6C3F1 mice is over 3 times greater in the high-dose group, and this peak occurs approximately 200 days earlier in the high-dose group, compared to the low-dose group. Progression of Proliferative Hepatocellular Lesions The progressive development of hepatocellular proliferative lesions over time re ects the likely pathogenesis of hepatocellular carcinomas and is most apparent in the 0.15 l M/g VC dose in the C3H, B6C3F1, and C57BL/6 strains. AHF developed rst, followed subsequently by hepatocellular adenomas, and then carcinomas (see Figure 2). In some cases, the multiplicity of adenomas rose as the multiplicity of AHF declined, suggesting a progression from AHF to adenoma. This trend is best seen in high-dose B6C3F1 mice. Similarly, the multiplicity of adenomas appeared to decrease as multiplicity of hepatocellular carcinomas increased, especially in B6C3F1 and C3H mice. Further evidence of progressive lesion development is provided by the large number of hepatocellular carcinomas arising within adenomas (Table 1). This diagnosis was made when a small area of carcinoma was localized within a hepatocellular adenoma. DISCUSSION The present study demonstrates that all 5 strains of male mice develop spontaneous hepatocellular neoplasia as well as an increased frequency of hepatocellular neoplasia following a single neonatal dose of VC. Our study results are consistent with progressive lesion development from AHF to hepatocellular adenomas to hepatocellular carcinomas. Furthermore, reduced latency and increased lesion multiplicity occurred as a function of VC dose in the 3 strains that had 2 dose groups (B6C3F1, C3H, and C57BL/6). The single exposure neonatal mouse model has previously been used in the study of hepatocarcinogenesis utilizing DEN (17, 24 26, 36) and ENU (13, 21, 23). Treatment of 12- or 15-day-old mice with hepatocarcinogens corresponds to exposure at a time of normal rapid proliferation of hepatocytes presumably allowing the xation of initial critical genetic damage before reversal by DNA repair mechanisms. Subsequent stochastic or environmentally induced genetic damage is presumably necessary for emergence and progression of proliferative lesions. Two critical genetic events for development of foci and 4 for development of hepatocellular carcinomas may occur in neonatal mice following treatment with DEN (36). Strain differences in genetic predisposition and susceptibility to hepatocarcinogenesis have been attributed to differences in carcinogen metabolism (10), hepatocarcinogen-susceptibilit y genes (13), differences in DNA methylation (6), and differential growth rate of proliferative lesions (11, 20, 22, 23, 26). Differences as great as 20-fold in the growth rate of AHF have been reported between mouse strains (20). Initiation/promotion studies have shown that liver tumor promotion is dependent on the mouse strain and the carcinogen used for neonatal initiation (40). In our investigation, overall prevalences of hepatocellular neoplasia in vehicle controls (Table 1) were highest in B6C3F1 and B6D2F1 males. C57BL/6 mice, generally considered relatively resistant to hepatocellular neoplasia (13), also developed a few liver neoplasms. All 5 strains of control mice had both hepatocellular adenomas and carcinomas, though the overall multiplicities were low. Spontaneous AHF, the putative preneoplastic hepatic lesions, were observed in 4 strains but not in B6CF1 mice. A single intraperitoneal injection of VC on day 15 of age resulted in an increased overall prevalence of AHF and hepatocellular neoplasms with a robust response in all strains except B6CF1. Neoplasm latency was clearly reduced relative to vehicle controls (Figure 1), and treatment resulted in increased lesion multiplicity except in B6CF1 mice (Figure 2). The most dramatic effect of treatment on neoplasm prevalence was seen in C3H and B6C3F1 mice, with a modest response in C57BL/6 and B6D2F1 mice, and a marginal one in B6CF1 mice. The effect of treatment on neoplasm multiplicity was robust in C3H and B6C3F1 mice, moderate in C57BL/6 mice, and marginal in B6CF1 and B6D2F1 mice. Two different doses of VC were used in 3 mouse strains (B6C3F1, C3H, and C57BL/6). In general, the higher dose of VC resulted in decreased latency and increased multiplicity of AHF, hepatocellular adenomas, and hepatocellular carcinomas. Dose-related increased lesion frequency has been documented in neonatal B6C3F1 mice treated with DEN (25, 35). The morphological features of altered foci and hepatocellular neoplasms were similar among the 5 strains and consistent with published descriptions (3, 5, 15, 22, 38). Eosinophilic globular cytoplasmic inclusions were noted in the majority of hepatocellular adenomas in all strains except C3H (data not shown). Similar cytoplasmic inclusions have been previously reported in liver tumors from different strains, and their presence was associated with a decreased rate of cellular proliferation within the lesions (22). Based on the temporal patterns of occurrence of hepatic proliferative lesions in this study, evidence is substantial for the progression of AHF to adenomas and adenomas to carcinomas. The multiplicity of adenomas tended to increase at the same ages that the multiplicity of AHF decreased, suggesting conversion of foci into adenomas. This was most clearly seen in B6C3F1 mice, though the multiplicity peaks for adenomas tended to be to the right of those for AHF in C3H and C57BL/6 mice as well. A similar but less dramatic temporal pattern suggested conversion of adenomas to carcinomas. Documentation of hepatocellular carcinomas arising within adenomas provides additional evidence of progression. Although regression of hepatocellular neoplasms in mice has been documented when a continuous treatment is stopped (27), we suggest that regression of neoplasms in the present study is unlikely, because the experimental model involved a single neonatal dose of the carcinogen. Similar sequential morphogenesis of proliferative hepatocellular lesions has been previously reported in the infant mouse model (17, 36) and treated adult mice (5, 15). Speci c lesions are presumed to progress independently of one another, because AHF,

6 604 TAKAHASHI ET AL TOXICOLOGIC PATHOLOGY adenomas, and carcinomas were frequently present simultaneously in the same liver. Furthermore, the ras oncogene pro le in multiple liver neoplasms occurring in individual mice supports the presumption of individual lesion independence (30). Pulmonary metastases occurred in approximately 29% and 25% of vehicle control B6C3F1 and C3H mice with hepatocellular carcinomas, respectively. These metastases of carcinomas in the low-dose VC-treated mice were present in all but the C3H strain with a maximum of 29% metastases in B6C3F1 mice. The frequency did not increase in the higher VC dose. These rates of pulmonary metastases are similar to those previously reported in mice (15, 37, 39). With respect to mouse hepatic neoplasia, the ndings of the present study considered with other published reports imply that all mouse strains can spontaneousl y develop liver neoplasms, all mouse strains respond to various hepatocarcinogens, strain differences in response are primarily differences in latency, and the dose of carcinogen in uences prevalence and multiplicity. REFERENCES 1. Alden CL, Smith PF, Piper CE, Brej L (1996). A critical appraisal of the value of the mouse cancer bioassay in safety assessment. Toxicol Pathol 24: Allen J, Langenbach R, Nesnow S, Sasseville K, Leavitt S, Campbell J, Brock K, Sharief Y (1982). Comparative genotoxicity studies of ethyl carbamate and related chemicals: Further support for vinyl carbamate as a proximate carcinogeni c metabolite. Carcinogenesis 3: Becker FF (1982). Morphological classi cation of mouse liver tumors based on biological characteristics. Cancer Res 42: Buchmann A, Bauer-Hofmann R, Mahr J, Drinkwater NR, Luz A, Schwarz M (1991). Mutational activation of the c-ha-ras gene in liver tumors of different rodent strains: Correlation with susceptibility to hepatocarcinogenesis. Proc Natl Acad Sci USA 88: Butler WH, Newberne PM (1975). Mouse Hepatic Neoplasia. Elsevier Scienti c Publishing Company, Amsterdam. 6. Counts J, Goodman J (1993). Comparative analysis of the methylation status of the 5 anking region of the Ha-ras in B6C3F1, C3H/He and C57BL/6 mouse liver. Cancer Lett 75: Dahl G, Miller E, Miller J (1980). Comparative carcinogenicities and mutagenicities of vinyl carbamate, ethyl carbamate and ethyl N-hydroxyethylcarbamate. Cancer Res 40: Dahl G, Miller J, Miller E (1978). Vinyl carbamate as a promutagen and a more carcinogenic analog of ethyl carbamate. Cancer Res 38: Dinse GE, Haseman JK (1986). Logistic regression analysis of incidentaltumor data from animal carcinogenicity experiments. Fundam Appl Toxicol 6: Diwan BA, Rice JM, Ward JM (1990). Strain-dependent effects of phenobarbital on liver tumor promotion in inbred mice. Prog Clin Biol Res 331: Dragani TA, Manenti G, Della Porta G (1991). Quantitative analysis of genetic susceptibility to liver and lung carcinogenesis in mice. Cancer Res 51: Drinkwater NR, Bennett LM (1991). Genetic control of carcinogenesis in experimental animals. In: Modi cation of Tumor Development in Rodents, Ito N, Sugano H (eds). Karger, Basel, pp Drinkwater NR, Ginsler JJ (1986). Genetic control of hepatocarcinogen - esis in C57BL/6J and C3H/HeJ inbred mice. Carcinogenesis 7: Fox TR, Schumann AM, Watanable PG, Yano BL, Maher VM, McCormick JJ (1990). Mutational analysis of the H-ras oncogene in spontaneous C57BL/6 C3H/He mouse liver tumors and tumors induced with genotoxic and nongenotoxi c hepatocarcinogens. Cancer Res 50: Frith CH, Wiley L (1982). Spontaneous hepatocellular neoplasms and hepatic hemangiosarcoma s in several strains of mice. Lab Anim Sci 32: Frome E (1983). The analysis of rates using Poisson regression models. Biometrics 39: Goldfarb S, Pugh TD, Koen H, He YZ (1983). Preneoplastic and neoplastic progression during hepatocarcinogenesi s in mice injected with diethylnitrosamine in infancy. Environ Health Perspect 50: Grasso P, Hardy J (1975). Strain differences in natural incidence and response to carcinogenesis. In: Mouse Hepatic Neoplasia, Butler WH, Newberne PM (eds). Elsevier, New York, pp Guengerich FP, Kim DH (1991). Enzymatic oxidation of ethyl carbamate to vinyl carbamate and its role as an intermediate in the formation of 1,N6- ethenoadenosine. Chem Res Toxicol 4: Hanigan MH, Kemp CJ, Ginsler JJ, Drinkwater NR (1988). Rapid growth of preneoplastic lesions in hepatocarcinogen-sensitiv e C3H/HeJ male mice relative to C57BL/6J male mice. Carcinogenesis 9: Hanigan MH, Winkler ML, Drinkwater NR (1993). Induction of three histochemically distinct populations of hepatic foci in C57BL/6J mice. Carcinogenesis 14: Kakizoe S, Goldfarb S, Pugh TD (1989). Focal impairment of growth in hepatocellular neoplasms of C57BL/6 mice: A possible explanation for the strain s resistance to hepatocarcinogenesis. Cancer Res 49: Kemp CJ, Drinkwater NR (1989). Genetic variation in liver tumor susceptibility, plasma testosterone levels, and androgen receptor binding in six inbred strains of mice. Cancer Res 49: Koen H, Pugh TD, Goldfarb S (1983). Hepatocarcinogenesi s in the mouse. Combined morphologic-stereologi c studies. Am J Pathol 112: Koen H, Pugh TD, Nychka D, Goldfarb S (1983). Presence of alphafetoprotein-positive cells in hepatocellular foci and microcarcinomas induced by single injections of diethylnitrosamine in infant mice. Cancer Res 43: Lee GH, Bennett LM, Carabeo RA, Drinkwater NR (1995). Identi cation of hepatocarcinogen-resistanc e genes in DBA/2 mice. Genetics 139: Malarkey DE, Devereux TR, Dinse GE, Mann PC, Maronpot RR (1995). Hepatocarcinogenicit y of chlordane in B6C3F1 and B6D2F1 male mice: Evidence for regression in B6C3F1 mice and carcinogenesis independent of ras proto-oncogene activation. Carcinogenesis 16: Maronpot RR, Boorman GA (1996). The contribution of the mouse in hazard identi cation studies. Toxicol Pathol 24: Maronpot RR, Boorman GA, Gaul BW (eds). (1999). Pathology of the Mouse. Cache River Press, Vienna, IL. 30. Maronpot RR, Fox T, Malarkey DE, Goldsworthy TL (1995). Mutations in the ras proto-oncogene : Clues to etiology and molecular pathogenesi s of mouse liver tumors. Toxicology 101: Maronpot RR, Haseman JK, Boorman GA, Eustis SE, Rao GN, Huff JE (1987). Liver lesions in B6C3F1 mice: The National Toxicology Program, experience and position. Arch Toxicol Suppl 10: Reynolds SH, Stowers SJ, Maronpot RR, Anderson MW, Aaronson SA (1986). Detection and identi cation of activated oncogenes in spontaneously occurring benign and malignant hepatocellular tumors of the B6C3F1 mouse. Proc Natl Acad Sci USA 83: Reynolds SH, Stowers SJ, Patterson RM, Maronpot RR, Aaronson SA, Anderson MW (1987). Activated oncogenes in B6C3F1 mouse liver tumors: Implications for risk assessment. Science 237: Ruebner BH, Gershwin ME, French SW, Meierhenry E, Dunn P, Hsieh LS (1984). Mouse hepatic neoplasia: Differences among strains and carcinogens. In: Mouse Liver Neoplasia. Current Perspectives, Popp JA (ed). Hemisphere Publishing Company, New York, pp Vesselinovitch SD (1987). Certain aspects of hepatocarcinogenesi s in the infant mouse model. Toxicol Pathol 15: Vesselinovitch SD, Mihailovich N (1983). Kinetics of diethylnitrosamine hepatocarcinogenesi s in the infant mouse. Cancer Res 43:

7 Vol. 30, No. 5, 2002 MOUSE LIVER LESIONS Vesselinovitch SD, Mihailovich N, Rao KV (1978). Morphology and metastatic nature of induced hepatic nodular lesions in C57BL C3H F1 mice. Cancer Res 38: Ward JM (1980). Morphology of hepatocellular neoplasms in B6C3F1 mice. Cancer Lett 9: Ward JM, Goodman DG, Squire RA, Chu KC, Linhart MS (1979). Neoplastic and nonneoplastic lesions in aging (C57BL/6N C3H/HeN)F1 (B6C3F1) mice. J Natl Cancer Inst 63: Weghorst CM, Pereira MA, Klaunig JE (1989). Strain differences in hepatic tumor promotion by phenobarbital in diethylnitrosamine- and dimethylnitrosamine-initiate d infant male mice. Carcinogenesis 10: Wiseman RW, Stowers SJ, Miller EC, Anderson MW, Miller JA (1986). Activating mutations of the c-ha-ras protooncogen e in chemically induced hepatomas of the male B6C3F1 mouse. Proc Natl Acad Sci USA 83: You M, Candrian U, Maronpot RR, Stoner GD, Anderson MW (1989). Activation of the Ki-ras protooncogen e in spontaneously occurring and chemically induced lung tumors of the strain A mouse. Proc Natl Acad Sci USA 86: