DRH strain rats were established by inbreeding a closed colony of Donryu rats continuously fed the chemical hepatocarcinogen 3′-methyl-4-dimethylaminoazobenzene for over 10 years. They are highly resistant to chemical induction of liver cancer and preneoplastic lesions. We studied the genetic basis of DRH resistance to preneoplastic lesions by analyzing 108 (F344 × DRH)F2 male rats fed 3′-methyl-4-dimethylaminoazobenzene for 7 weeks. Five parameters of preneoplastic liver lesions were selected for quantitative analysis: (a) number of glutathione S-transferase placental form-positive foci per unit area of liver section; (b) percentage area occupied by the foci; (c) average size of foci;(d) glutathione S-transferase placental form mRNA level; and (e) γ-glutamyltranspeptidase mRNA level. Furthermore, O6-methylguanine DNA methyltransferase and mannose 6-phosphatase/insulin-like growth factor 2 receptor mRNA levels were quantified. Composite interval mapping analysis showed that there were two remarkably significant clusters of quantitative trait loci affecting preneoplastic liver lesions on chromosomes 1 and 4. These clusters were designated collectively as Drh1 and Drh2,respectively. The functions of the recessive DRH allele of Drh1 and the semidominant DRH allele of Drh2 were to suppress the phenotypes of precancerous lesions. Each cluster showed two to three subpeaks in linkage likelihood plots, suggesting the presence of several closely linked quantitative trait loci affecting preneoplastic lesions. Possible candidate genes at each locus will be discussed. Expression of O6-methylguanine DNA methyltransferase and mannose 6-phosphatase/insulin-like growth factor 2 receptor did not affect DRH resistance to hepatocarcinogenesis, although they were polymorphic between DRH and F344 rats.

Chemical-induced hepatocarcinogenesis is a multistep process subdivided into at least three stages: (a) initiation;(b) promotion; and (c) progression(1). Genotoxic carcinogens induce irreversible DNA damages, generating initiated cells that can proliferate clonally in the presence of promoter substances until they acquire autonomic growth capability. Cancer cells thus established show further increases in their malignant characteristics and heterogeneity by subsequent genetic changes during growth. All of these steps involve interactions with host biochemical, endocrinological, immunological, and microenvironmental regulatory systems. Polymorphism in host genes involved in such regulation may be the basis of the strain differences observed frequently among inbred strains. Initiated hepatocytes are shown to express a large amount of GST-P3(2, 3) or GGT (4) and to form immunohistochemically discrete foci of GST-P- and/or GGT-positive cells, i.e., EAF. The mechanisms responsible for GST-P induction are not well understood, but the number of EAF and the size and percentage area occupied by EAF are regarded as quantitative parameters of the promotion stage of hepatocarcinogenesis. As a rough estimate, of several million liver cells initiated by a carcinogen,104 EAF and a few final cancer masses are formed. To dissect such a complicated process, genetic analysis of available susceptible or resistant inbred strains will be of great advantage.

Genetic predisposition to chemically-induced liver cancers has been studied mainly in mice (as reviewed in Refs. 5, 6, 7). To date, about 10 loci have been mapped, but none of these have yet been cloned. Two rat strains, Copenhagen (8) and BN(9), have been shown to be resistant to chemical induction of HCC. Recently, De Miglio et al.(10) mapped a set of BN QTL affecting the volume and/or volume fraction of HCC. These QTL consist of one susceptibility locus and at least three resistance loci. The consistent conclusion of these studies is that genetic predisposition to chemically induced HCC in rodents is a polygenic trait.

The inbred rat strain DRH was established from a closed colony of Donryu rats continuously fed 3′-Me-DAB and selected for reduced HCC induction during inbreeding for more than 10 years(11, 12, 13). In contrast to carcinogen-susceptible parental Donryu rats, DRH rats showed a remarkably lower incidence of hepatic tumors when given liver carcinogens such as 3′-Me-DAB, DAB, 3′-ethyl-DAB, 2-acetylaminofluorene,7,12-dimethylbenz(a)anthracene, or N-nitrosodimethylamine (14). Such resistance is also evident in the induction of preneoplastic lesions. After 6–8 weeks of 3′-Me-DAB administration, Donryu rat liver showed more than 50-fold induction of the GST-P mRNA level above that of the untreated control, whereas DRH liver showed only 2–3-fold induction(15). DRH resistance was suggested to be dominant because reciprocal DRH × F344 F1 rats are induced much less GST-P mRNA than F344 by short-term feeding of 3′-Me-DAB (15). These characteristics indicated that the DRH rat may be a promising model in which to investigate host genetic control in hepatocarcinogenesis.

In this study, we explored the QTL negatively affecting 3′-Me-DAB-induced preneoplastic liver lesions in (F344 × DRH)F2 rats using number of EAF per unit area of liver section, the percentage area occupied by EAF, and the average size of EAF as quantitative parameters. Furthermore, mRNA levels of GST-P, GGT, MGMT, and M6P/IGF2R were quantified. Two clusters of DRH-derived QTL yielding resistance were mapped on RNO1 and RNO4 and designated as Drh1 and Drh2 (DRH resistance to hepatocarcinogenesis 1 and 2), respectively.

Chemicals.

3′-Me-DAB was obtained from Tokyo Kasei Kogyo Co., Ltd. (Tokyo, Japan). Rabbit-antirat GST-P and antirat GGT antibodies were obtained from Medical and Biological Laboratories, Co. (Nagoya, Japan).

Animals and Treatments.

Inbred DRH and F344 rats were purchased from SEAC Yoshitomi,Ltd. (Fukuoka, Japan) at 4 weeks of age and allowed to acclimatize for 1 week before use. F2 rats were generated by intercrossing male and female (F344 × DRH)F1 rats. Starting at 5 weeks of age, male F2 rats were fed a diet containing 0.06%3′-Me-DAB for 8 weeks. All rats were housed individually in suspended wire-bottomed cages in a room with a constant temperature of 24°C,55% humidity, and a 12-h light (6 a.m. to 6 p.m.)/dark cycle. All rats were sacrificed under ether anesthesia after 8 weeks of treatment,and a full postmortem examination was carried out.

Histopathological Analysis.

Livers were removed, and 5-mm-thick slices were cut from the right lateral median and right lateral lobes of individual rats. These slices were fixed in ice-cold acetone and embedded in paraffin for subsequent immunohistochemical examination of GST-P. Small pieces from the remaining left lateral lobe of livers were quickly frozen and kept at−80°C until RNA extraction. Paraffin-embedded blocks were sectioned at a thickness of 3–4 μm and stained immunohistochemically for GST-P by using the avidin-biotin peroxidase complex method (Vecstatin ABC elite kit; Vector Laboratories, Burlingame, CA) as described previously(16). For quantitative assessment of lesions, the areas of liver sections and the numbers and areas of GST-P-positive foci, i.e., EAF, were measured, and the numbers of EAF/cm2 liver area and the percentage of liver area occupied by EAF were calculated using an image analyzer (IPAP-WIN;Sumika Technos Co., Ltd., Hyogo, Japan). Liver lesions were diagnosed according to the criteria described by Squire and Levitt(17) and the descriptions given by the Institute of Laboratory Animal Resources (18).

Northern Analysis.

Aliquots of 20 μg of total RNA extracted from individual liver pieces by the guanidine isothiocyanate-phenol-chloroform procedure(19) were electrophoresed in 1.0% agarose gels and transferred onto nylon membranes (Hybond N; Amersham Pharmacia Biotech,Buckinghamshire, United Kingdom) prehybridized in a solution containing 4× SSC, 50% formamide, 0.5% SDS, 5× Denhardt’s solution, and 20μg/ml salmon sperm DNA (20). The membranes were then hybridized with a mixture of 32P-labeled nick-translated probes of GST-P, GGT, MGMT, or M6P/IGF2R(14) at 42°C for 16–24 h. The membranes were washed in 1× SSC containing 0.1% SDS at 42°C for 30 min, and this was repeated two or three times. The radioactivity on the hybridized membranes was quantified using a BAS2000 bio-imaging analyzer (Fuji Photo Film Co., Tokyo, Japan). To minimize data fluctuation between the blots, 108 RNA samples were separated on 9 membrane filters and hybridized with a single lot of radiolabeled probes on the same day.

Genetic Analysis.

For linkage analysis, we used the simple sequence repeat(microsatellite) length polymorphism assay, using genomic DNA extracted from the kidney as a template. All primers for microsatellite analysis were purchased from Research Genetics, Inc. (Huntsville, AL.). Methods for PCR and agarose electrophoresis of PCR products have been described previously (21). The relative map positions of microsatellite loci were based on Brown et al.(22) or obtained from the World Wide Web.4Of 536 microsatellite loci examined, 119 (22.2%) were polymorphic between DRH and F344 and were readily recognizable on agarose gel electrophoresis. To find the loci associated with either susceptibility or resistance to preneoplastic liver lesions, 76 marker loci (2–14 markers for each chromosome) were analyzed. Approximate coverage was 94.6% of the entire rat genome, assuming that each marker would detect linkage within a circle of 20 cM. Of 108 F2 rats,we selected 20 each with extremely low or high GST-P or GGT mRNA levels and percentage liver area occupied by EAF. A preliminary genome-wide scan was performed by genotyping them for 66 markers. Linkage was evaluated by χ2 test for goodness of fit against an expected 1:2:1 ratio of F/F, F/D, and D/D genotypes (F represents the F344 allele, and D represents the DRH allele). When a marker locus showed a possible linkage (χ2test, P < 0.05) in the primary screen, all 108 rats were genotyped for all available polymorphic loci on the same chromosome. Any double recombinants were retyped to minimize typing errors. Genotype data were analyzed using the Mapmaker/QTL program,using five quantitative parameters for preneoplastic liver lesions. When more than one possible peak of linkage was found on the chromosome, composite interval mapping was carried out with Cartographer QTL software (23) to determine whether there were multiple independent loci. In these analyses, two quantitative parameters, GST-P and GGT mRNA levels, were transformed to logarithms,and size of EAF was transformed to the square root to obtain a better fit to the normal distribution.

Statistical Analysis.

In the preliminary genome scan, linkage was evaluated by the nonparametric Kruskal-Wallis test. According to the criteria of Lander and Kruglyak (24), linkage in the F2generation was taken as significant when P was<5.2 × 10−5 (lod score, 4.3)and was suggestive when P was <1.6 × 10−3 (lod score, 2.8). Phenotypic differences between the three genotypes (F/F, F/D, and D/D) were analyzed by the unpaired Student’s t test. Correlations between different phenotypic parameters were evaluated by correlation analysis with StatView-J 4.11 software (Abacus Concepts, Inc., Berkeley, CA) on a Macintosh personal computer.

Genetic Crosses.

Our preliminary study showed that reciprocal DRH × F344 F1 rats exhibit equally low induction of liver GST-P mRNA by a short-term feeding of 3′-Me-DAB (15),suggesting that the resistance is generally dominant without sex-dependent regulation or modification. In this study, we further scrutinized the mode of genetic resistance and the map location of relevant genes using 108 male (F344 × DRH)F2 rats fed 3′-Me-DAB for 8 weeks.

Parameters of Preneoplastic Liver Lesions.

In this study, we focused on genetic control of preneoplastic liver lesions after feeding rats 3′-Me-DAB for 8 weeks. First, five phenotypic parameters that have been shown to be remarkably different between DRH and other HCC-susceptible rat strains were selected(14, 15), i.e.: (a) the number of EAF per unit area of liver section; (b) the percentage of liver area occupied by EAF; (c) the average size of the foci; (d) the GST-P mRNA level; and (e) the GGT mRNA level. These parameters were closely correlated with each other,as shown in Table 1 (r = 0.460–0.892). Therefore, none of them seemed to be independent traits. The correlation between the number and size of the EAF was just above the level of significance(r = 0.460), whereas that between the number of EAF and the percentage of liver area occupied by EAF was very high(r = 0.892). GST-P and GGT mRNA levels both correlated best with the percentage of liver area occupied by EAF,suggesting that they were dependent on the total mass of EAF.

Our previous study showed that MGMT and M6P/Igf2Rexpression is polymorphic between DRH and F344 or Donryu rats,suggesting that MGMT and M6P/Igf2R may play a role in DRH resistance (14). However, the levels of MGMT and M6P/Igf2R mRNA were not correlated with any of the five parameters examined in the present study (Table 1). These observations clearly excluded the possibility that MGMT and M6P/Igf2R may be responsible for DRH resistance.

QTL Analysis.

A preliminary genome-wide screen with 20 animals showing extremely low and high phenotypic values revealed significant linkage on RNO1 and RNO4 for all five parameters of preneoplastic liver lesions (data not shown). Assuming a level of suggestive significance of P < 1.6 × 10−3, no linkage for any of these phenotypic parameters was suggested on other chromosomes. RNO2 and RNO18 contained marker loci with 0.001 < P < 0.01. Subsequently, we genotyped all 108 F2 rats for the 27 microsatellite loci on RNO1 and RNO4, which were polymorphic between DRH and F344.

The results of mapping were first analyzed by interval mapping with MapMaker/QTL software (Fig. 1). On RNO1 and RNO4, we observed three and two peaks, respectively. Interval mapping by MapMaker/QTL assumes one locus/chromosome. To further investigate whether these peaks represented independent QTL,composite interval mapping with QTL Cartographer software (Version 1.13; Ref. 23) was performed. Fig. 2 shows a linkage plot of the results of composite interval mapping.

On the distal segment of RNO1, there were apparently three peaks of linkage. The first peak was closely linked to D1 Mgh10, at which the number of EAF/cm2 liver section showed a peak lod score of 4.6 (Fig. 2,A); we named this locus Drh1a. At this position, the percentage area and mRNA levels of GST-P and GGT formed a minor peak on interval mapping (Fig. 1,A), but the peaks were below the level of significance on composite interval mapping (Fig. 2,A). The second peak was rather broad on the segment between D1Wox10and D1Rat302. At the position of D1Rat302, lod scores for the number of EAF, percentage area, and mRNA level of GST-Pwere 8.0, 7.2, and 5.9, respectively. The lod score for the level of GGT mRNA was below the level of significance (Fig, 2A);it was designated as Drh1b. The third peak was linked with D1 Mgh12. Although the presence of a quantitative trait locus at this position was merely suggestive of the size of the EAF on QTL Cartographer, we tentatively named it Drh1c because lod scores for percentage area, GST-P mRNA, and size of EAF were above 4.3 in Mapmaker/QTL analysis. These three QTL/peaks were located close together, and their phenotypic effects were pleiotropic. As shown in Table 2, at all three Drh1 loci, all genetic traits were recessive because phenotypic values of F2 rats homozygous for the DRH-derived allele were consistently smaller than those of rats heterozygous and homozygous for the F344-derived allele.

On the other hand, we detected two peaks on the distal segment of RNO4(Figs. 1,B and 2,B). The first peak was closely linked to D4Rat37, at which significant linkage was observed for percentage area (lod score, 7.8), number of foci (lod score, 7.8),mRNA level of GST-P (lod score, 4.6), and size of foci (lod score,3.8). This locus was named Drh2a. The second peak(Drh2b) was 4 cM distal from D4Rat48, at which significant linkage was observed for percentage area (lod score, 5.8)and number of EAF (lod score, 5.8). The interval between Drh2a and Drh2b was as little as 10 cM,but combination of the parameters showed that linkage to each loci was different. The DRH allele of both Drh2a and Drh2bwas semidominant, as shown in Table 2.

Cartographer QTL analysis showed that there were five putative QTL affecting the parameters of precancerous lesions in (DRH × F344)F2 intercrosses. The map positions of each locus were exactly the same as those suggested by Mapmaker analysis. Each putative loci had pleiotropic effects on phenotypic traits as seen in Table 2. Fig. 3 shows the percentage of phenotypic variation explained by these loci. The loci explaining >20% phenotypic variance for the number of EAF were Drh1a, Drh1b, Drh2a, and Drh2b; those for percentage area of EAF were Drh1a, Drh1b, Drh2a, and Drh2b; and those for GST-P mRNA were Drh1b, Drh2a, and Drh2b. For GGT mRNA, low peaks of ∼15% were observed at Drh1a and Drh1b. For the size of EAF, similar low peaks of ∼15% were seen at Drh1c and Drh2a.

MGMT and M6P/Igf2R mRNA were not correlated with any of five traits of precancerous lesions; therefore, they did not seem to be responsible for DRH resistance (Table 1). They were,however, located on RNO1, so that we performed QTL analysis of these loci. The semidominant MGMT locus was mapped 8 cM proximal to Drh1a with a lod score of 18.6 (data not shown). None of five preneoplastic lesion traits were comapped at MGMT. Another semidominant M6P/IgF2R locus was mapped on the short arm of RNO1, very close to D1 Mit9 (data not shown). Again, none of the five preneoplastic lesion traits were comapped with this locus.

DRH strain rats are a unique experimental model in which to study chemically induced liver carcinogenesis. They have been selected from a closed colony of Donryu rats fed 3′-Me-DAB by sister-brother mating for over 40 generations. Their resistance to hepatocarcinogens has been suggested to be genetically dominant (15). In the present study, we focused on genetic control in the induction of preneoplastic lesions. Five phenotypic parameters were selected to quantitatively describe the preneoplastic lesions; i.e., the number of GST-P-positive foci, the percentage of liver area occupied by EAF, the size of the foci, and the amounts of GST-P as well as GGT mRNA. Their phenotypic values were mutually closely interrelated [for instance,percentage of liver area occupied by EAF and GST-P mRNA(r = 0.763)], but still they seemed under differential genetic control by combinations of host loci. No single phenotypic parameter was controlled by a single locus. The contributions of the loci to phenotype variance are summarized in Fig. 3. For preneoplastic liver lesions, five QTL were mapped by composite interval mapping of 108 F2 rats. Whether these loci are directly correlated with susceptibility to HCC remains unclear, but answers to this question will soon be provided by a concurrent study in our laboratory.

DRH rats are resistant to a broad spectrum of chemical carcinogens:(a) DAB; (b) 3′-Me-DAB; (c)3′-ethyl-DAB; (d) 2-acetylaminofluorene; and (e) N-nitrosodimethylamine (14). To explain the mechanism of resistance, the biochemical pathway of 3′-Me-DAB has been studied extensively. The level of metabolic activation of DAB is lower and the level of metabolic inactivation of DAB is higher in the DRH rat liver than in the parental Donryu rat liver (14). However,there is no difference in DNA-adduct formation of 3′-Me-DAB between the carcinogen-sensitive parental strain Donryu and carcinogen-resistant DRH rat livers at several time points during 3′-Me-DAB administration(14). Therefore, the critical event of resistance in the DRH strain seems to occur after formation of DNA-adducts. In support of this hypothesis, we found that the expansion of EAF in the DRH rat liver after 3′-Me-DAB administration was significantly less than that seen in Donryu rats under the same experimental conditions(15). The target of resistance was not specific to the liver because DRH females given 7,12-dimethylbenz(a)anthracene were highly resistant to mammary cancers (14).

In this study, we used F344 rats as 3′-Me-DAB-sensitive parents rather than Donryu rats. This was because the parental Donryu strain was not inbred but was a closed colony. Multiple QTL affecting preneoplastic liver lesions were mapped in the (F344 × DRH)F2 intercross rats. None of the phenotypic traits was controlled by a single quantitative trait locus. All QTL were pleiotropic and formed clusters on RNO1 and RNO4. It was not possible to separate them clearly by likelihood plots, even by composite interval mapping, and therefore they were tentatively designated as Drh1a, Drh1b, and Drh1c and Drh2a and Drh2b. Drh1a, Drh1b, and Drh1c were localized on the distal segment of RNO1. Drh1a was at the major peak of the linkage curve, especially for the trait of number of EAF. The DRH allele of Drh1reduced all phenotypic values in a recessive manner. At the position of Drh1a, there are several putative candidate genes affecting carcinogenesis: (a) c-Ha-ras; (b) Hrev107; (c) Ins2; (d) Igf2; (e) H19; (f) Tapa1; and (g) Gst-p. The cellular proto-oncogene c-Ha-ras is activated in mouse HCC induced by chemical carcinogens (25, 26, 27) but is activated relatively infrequently in human HCC and rat HCC (28). Hrev107(29) is a putative tumor suppressor gene that inhibits the malignant phenotype of Ha-ras-transformed cells. Other possible tumor suppressor genes are H19(30) and TAPA1(31). The segment containing Igf2 and H19 shows parental imprinting. The observation that reciprocal DRH × F344 F1 rats are equally resistant to 3′Me-DAB(15) excluded Igf2 and H19from being candidate genes. MGMT is one of the key enzymes involved in DNA repair, and mice with targeted disruption of MGMT are highly susceptible to carcinogen-induced HCC as well as thymic lymphoma (32, 33). DRH rats showed far lower levels of MGMT induction than Donryu rats after 3′-Me-DAB feeding (14). MGMT was mapped only 8 cM proximal to Drh1a. However, we excluded MGMTfrom the candidate genes because the mRNA level in F2 was not correlated with other preneoplastic lesion parameters (Table 1), and the peak of Drh1a was evidently discrete from that of MGMT.

The linkage peak for Drh1b was about 15 cM distal from that of Drh1a, where percentage area, GST-P mRNA, and the number of EAF showed significant linkage. Aldh1 and Anx1were shown to be localized at this position in a recent RH mapping study (34). Drh1c showed a small peak at the telomeric end of RNO1, which was suggested to be correlated with the size of foci. Cyp2C and Cyp17 are possible candidate genes. These cytochrome P450 enzymes may affect the susceptibility to cancer through metabolic activation of carcinogens.

Two QTL, Drh2a and Drh2b, were mapped to the distal segment of RNO4. One of the candidate genes for Drh2ais growth hormone-releasing hormone receptor (Ghrhr). The little mice (35), which are defective in Ghrhr, develop fewer liver cancers than their wild-type counterparts. The body weight of DRH strain rats is consistently larger than that of F344 rats. Interestingly, interval mapping using body weight before 3′-Me-DAB administration showed a low peak of QTL with a lod score of 2.4 at Dhr2a (data not shown). Polymorphism at Ghrhr may affect the number and progression of liver preneoplastic lesions. At Drh2b, Tgfα may be a possible candidate gene. Tgfα is a ligand of the epidermal growth factor receptor and stimulates the growth of liver cells in an autocrine loop (1, 36).

De Miglio et al.(10) recently reported genetic resistance of BN rats to chemically induced HCC in (BN × F344) × F344 rats. The QTL on RNO7 and RNO10 are significantly associated with tumor volume fraction, and the presence of other QTL is also suggested on RNO4 and RNO8. The BN quantitative trait locus on RNO7 is a susceptibility locus, and the QTL on RNO10, RNO4, and RNO8 are resistance loci. The observation that these QTL affect tumor size but not the number of tumors may suggest that they modify the growth of transformed hepatocytes, but not the frequency of malignant transformation. Effects are also seen on volume but not on the number of EAF (10). No LOH at these loci is observed in HCCs developed in resistant (BN × F344)F1 hybrid rats (37). Judging from the chromosomal locations of these loci, DRH resistance seemed distinct from that of BN rats. However, in the study by De Miglio et al.(10), the recessive resistance locus on RNO1 might have been overlooked because they examined backcrosses to susceptible F344 rats rather than F2 rats.

Genetic predisposition to HCC in mice has been studied extensively by two groups. Drinkwater et al.(38) identified a susceptibility gene in C3H mice in a cross between C3H and C57BL and later mapped the gene on MMU1 (39). Gariboldi et al.(40) and Manenti et al.(41) identified six susceptibility loci for urethane-induced liver cancers: (a) Hcs1 (MMU7);(b) Hcs2 (MMU8); (c) Hcs3(MMU12); (d) Hcs4 (MMU2); (e) Hcs5 (MMU5); and (f) Hcs6 (MMU19). Using N,N-diethylnitrosamine, Lee et al.(42) and Poole and Drinkwater (43) mapped resistance genes to HCCs to Hcr1 (MMU4) and Hcr2(MMU10). Of these, Hcs1 was on RNO1 but was more centromeric than the Drh1 cluster. Hcs5 was on RNO4 but was also more centromeric than the Drh2 cluster. Hcs6was on MMU 19 and was on the telomeric portion of RNO1. Its map position suggested that it might be homologous to either Drh1b or Drh1c. In humans, there have been reports of familial clusters of HCC (44, 45, 46). Polymorphisms in drug-metabolizing enzyme genes have been shown to be correlated to genetic risk of HCC in humans (47). To explore putative tumor suppressor genes of HCC, LOH has been studied extensively in human HCC (48, 49, 50, 51, 52, 53) and also in rodent models (37, 54). The telomeric segment of RNO1 is homologous to 11p15, 11q13, 9p21, and 10q23–26,and the telomeric segment of RNO4 is homologous to 3p, 12p12,10q11.2, 3p25, and 12p13.2-p11.2. Frequent LOH in these segment has not been reported in human HCC, except for 11p15 in hepatoblastomas (55). The difference in species and agent of induction may be responsible for these discrepancies.

Genetic resistance of DRH strain rats was more complicated than we had first assumed. Combination of recessive and semidominant QTL could explain a large proportion of the phenotypic variance between DRH and F344 rats. However, there were some ambiguities in the number of independent QTL even with the composite interval mapping analysis. To clarify these problems, a far larger set of genetic analysis is required. Furthermore, it remains unclear whether such resistance has been acquired by mutational events in the germ line or by long-term selection. There are several promising candidate genes for these QTL. This study provided useful information for the analysis of genetic predisposition to chemically induced HCC.

Fig. 1.

Interval mapping by Mapmaker QTL. A,chromosome 1. B, chromosome 4. Each plot line represents a preneoplastic lesion trait. Number of foci/cm2 liver section, ; % liver area occupied by foci in log,; size of foci transformed to square root,. . . . . .; mRNA of GST-P in log, - - - - -; and mRNA of GGT in log, -.-.-.-.-. Horizontal bold line, the significance level of linkage; horizontaldotted line, the suggestive level of linkage in F2intercross. Bar, 10 cM.

Fig. 1.

Interval mapping by Mapmaker QTL. A,chromosome 1. B, chromosome 4. Each plot line represents a preneoplastic lesion trait. Number of foci/cm2 liver section, ; % liver area occupied by foci in log,; size of foci transformed to square root,. . . . . .; mRNA of GST-P in log, - - - - -; and mRNA of GGT in log, -.-.-.-.-. Horizontal bold line, the significance level of linkage; horizontaldotted line, the suggestive level of linkage in F2intercross. Bar, 10 cM.

Close modal
Fig. 2.

Composite interval mapping by Cartographer QTL. A, chromosome 1. B, chromosome 4. See the Fig. 1 legend for definition of the lines. This analysis was an approximation because genotype data for all 108 F2 rats were calculated only on RNO1 and RNO4. For more precise data, genotype data for all chromosomes, even those with a lower likelihood of linkage, had to be considered.

Fig. 2.

Composite interval mapping by Cartographer QTL. A, chromosome 1. B, chromosome 4. See the Fig. 1 legend for definition of the lines. This analysis was an approximation because genotype data for all 108 F2 rats were calculated only on RNO1 and RNO4. For more precise data, genotype data for all chromosomes, even those with a lower likelihood of linkage, had to be considered.

Close modal
Fig. 3.

Percentage of variance of phenotype explained by each locus (calculated by Mapmaker/QTL for each trait).

Fig. 3.

Percentage of variance of phenotype explained by each locus (calculated by Mapmaker/QTL for each trait).

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Supported by a grant-in-aid from the Ministry of Education, Culture, Sports and Science, Japan and a grant for cancer research from the Ministry of Health and Welfare, Japan.

The abbreviations used are: GST-P, glutathione S-transferase placental form; 3′-Me-DAB,3′-methyl-4-dimethylaminoazobenzene; GGT, γ-glutamyltransferase; QTL,quantitative trait loci; EAF, enzyme-altered foci; HCC, hepatocellular carcinoma; MGMT, O6-methylguanine DNA methyltransferase; M6P/IGF2R, mannose 6-phosphatase/insulin-like growth factor 2 receptor; lod, logarithm of odds; RNO, rat chromosome; MMU,mouse chromosome; BN, Brown Norway; DAB, 3,3′-diaminobenzidine; LOH,loss of heterozygosity.

http://waldo.wi.mit.edu/rat/public/images/csomes/(November 22, 1998).

Table 1

Correlation between parameters of preneoplastic liver lesions in(F344 × DRH) F2a

% liver area occupied by fociAverage sizeGST-P mRNAGGT mRNAMGMT mRNAM6P/IGF2R mRNA
No. of foci/cm liver area r = 0.892 r = 0.460 r = 0.654 r = 0.613 r = 0.199 r = 0.107 
% liver area occupied by foci  r = 0.767 r = 0.763 r = 0.725 r = 0.100 r = 0.117 
Average size   r = 0.655 r = 0.600 r =−0.23 r = 0.125 
GST-P mRNA    r = 0.761 r = 0.082 r = 0.051 
GGT mRNA     r = 0.070 r = 0.056 
MGMT mRNA      r =−0.030 
% liver area occupied by fociAverage sizeGST-P mRNAGGT mRNAMGMT mRNAM6P/IGF2R mRNA
No. of foci/cm liver area r = 0.892 r = 0.460 r = 0.654 r = 0.613 r = 0.199 r = 0.107 
% liver area occupied by foci  r = 0.767 r = 0.763 r = 0.725 r = 0.100 r = 0.117 
Average size   r = 0.655 r = 0.600 r =−0.23 r = 0.125 
GST-P mRNA    r = 0.761 r = 0.082 r = 0.051 
GGT mRNA     r = 0.070 r = 0.056 
MGMT mRNA      r =−0.030 

Correlation coefficient: r > 0.7, highly significant; 0.4 < r < 0.7, significant; r < 0.4, not significant.

Table 2

Phenotype-genotype correlation at each quantitative trait locus

LocusAlleleNNo. of foci/cm liver area% liver area occupied by fociAverage size (mm)aGST-P mRNAGGT mRNA
Drh1a D/D 22 41 ± 20 3.4 ± 2.5 7.5 ± 2.5 2.3 ± 2.9 2.2 ± 1.8      
 D/F 60 86 ± 36]b 7.8 ± 4.7]b 8.6 ± 2.9 4.4 ± 3.5 4.6 ± 3.2]c       
 F/F 26 99 ± 36 8.5 ± 4.6 8.0 ± 2.4 5.3 ± 3.7 5.4 ± 3.9      
Drh1b D/D 26 46 ± 27 3.8 ± 3.0 6.7 ± 2.0 2.1 ± 2.7 2.46 ± 1.74      
 D/F 56 87 ± 35]b 7.9 ± 4.7]c 8.7 ± 2.8 4.6 ± 3.5]d 4.55 ± 3.27]d       
 F/F 26 99 ± 37 8.6 ± 4.6 8.7 ± 2.8 5.3 ± 3.7 5.6 ± 4.0      
Drh1c D/D 32 60 ± 38 4.4 ± 3.2 6.8 ± 1.9 2.7 ± 3.0 2.6 ± 1.9      
 D/F 58 86 ± 34]d 7.9 ± 4.5]b 8.8 ± 2.9]c 4.7 ± 3.5]d 4.7 ± 3.5]d       
 F/F 18 97 ± 43 9.2 ± 5.4 8.8 ± 2.5 5.4 ± 4.1 5.7 ± 3.7      
Drh2a D/D 36 55 ± 31 4.2 ± 3.3 7.0 ± 2.5 2.3 ± 2.1 3.1 ± 3.1      
 D/F 47 87 ± 38]c 7.4 ± 4.4]c 8.2 ± 2.4 4.5 ± 3.5]c 4.4 ± 3.1      
 F/F 25 103 ± 33 10.5 ± 4.4]c 10.1 ± 2.7]d 6.4 ± 3.9 5.6 ± 3.6      
Drh2b D/D 33 58 ± 32 4.6 ± 3.5 7.1 ± 2.4 2.5 ± 2.3 3.4 ± 3.2      
 D/F 53 83 ± 37]d 7.1 ± 4.4]d 8.1 ± 2.5 4.3 ± 3.6 4.1 ± 3.1      
 F/F 22 106 ± 35 10.8 ± 4.6]d 10.1 ± 2.8]d 6.6 ± 3.6 5.9 ± 3.7      
LocusAlleleNNo. of foci/cm liver area% liver area occupied by fociAverage size (mm)aGST-P mRNAGGT mRNA
Drh1a D/D 22 41 ± 20 3.4 ± 2.5 7.5 ± 2.5 2.3 ± 2.9 2.2 ± 1.8      
 D/F 60 86 ± 36]b 7.8 ± 4.7]b 8.6 ± 2.9 4.4 ± 3.5 4.6 ± 3.2]c       
 F/F 26 99 ± 36 8.5 ± 4.6 8.0 ± 2.4 5.3 ± 3.7 5.4 ± 3.9      
Drh1b D/D 26 46 ± 27 3.8 ± 3.0 6.7 ± 2.0 2.1 ± 2.7 2.46 ± 1.74      
 D/F 56 87 ± 35]b 7.9 ± 4.7]c 8.7 ± 2.8 4.6 ± 3.5]d 4.55 ± 3.27]d       
 F/F 26 99 ± 37 8.6 ± 4.6 8.7 ± 2.8 5.3 ± 3.7 5.6 ± 4.0      
Drh1c D/D 32 60 ± 38 4.4 ± 3.2 6.8 ± 1.9 2.7 ± 3.0 2.6 ± 1.9      
 D/F 58 86 ± 34]d 7.9 ± 4.5]b 8.8 ± 2.9]c 4.7 ± 3.5]d 4.7 ± 3.5]d       
 F/F 18 97 ± 43 9.2 ± 5.4 8.8 ± 2.5 5.4 ± 4.1 5.7 ± 3.7      
Drh2a D/D 36 55 ± 31 4.2 ± 3.3 7.0 ± 2.5 2.3 ± 2.1 3.1 ± 3.1      
 D/F 47 87 ± 38]c 7.4 ± 4.4]c 8.2 ± 2.4 4.5 ± 3.5]c 4.4 ± 3.1      
 F/F 25 103 ± 33 10.5 ± 4.4]c 10.1 ± 2.7]d 6.4 ± 3.9 5.6 ± 3.6      
Drh2b D/D 33 58 ± 32 4.6 ± 3.5 7.1 ± 2.4 2.5 ± 2.3 3.4 ± 3.2      
 D/F 53 83 ± 37]d 7.1 ± 4.4]d 8.1 ± 2.5 4.3 ± 3.6 4.1 ± 3.1      
 F/F 22 106 ± 35 10.8 ± 4.6]d 10.1 ± 2.8]d 6.6 ± 3.6 5.9 ± 3.7      

Data shown are ×100 original figures.

P < 0.0001.

0.001 > P > 0.0001.

0.01 > P > 0.001.

We are grateful to Dr. Zhao-Bang Zeng (Department of Statistics,North Carolina State University, Raleigh, NC) for helpful discussion.

1
Mills J. J., Jirtle R. L., Boyer I. J. Mechanisms of liver tumor promotion Jirtle R. L. eds. .
Liver Regeneration and Carcinogenesis
,
:
199
-226, Academic Press San Diego, CA  
1995
.
2
Moore M. A., Nakagawa K., Satoh K., Ishikawa T., Sato K. Single GST-P positive liver cells: putative initiated hepatocytes.
Carcinogenesis (Lond.)
,
10
:
2107
-2111,  
1987
.
3
Cameron R. G. Identification of the putative first cellular step of chemical hepatocarcinogenesis.
Cancer Lett.
,
47
:
163
-167,  
1989
.
4
Tsuji S., Ogawa K., Takasaka H., Sonoda T., Mori M. Clonal origin of γ-glutamyl transpeptidase positive lesions induced by initiation-promotion in ornithine carbamyltransferase mosaic mice.
Jpn. J. Cancer Res.
,
79
:
148
-151,  
1988
.
5
Chiaverotti T. A., Carabeo R. A., Drinkwater N. R. Genetic and hormonal regulation of murine hepatocarcinogenesis Hiai H. Hino O. eds. .
Animal Models of Cancer Predisposition Syndromes
,
:
131
-142, Karger Basel  
1999
.
6
Dragani T. A., Canzian F., Pierotti M. A. A polygenic model of inherited predisposition to cancer.
FASEB J.
,
10
:
865
-870,  
1996
.
7
Feo F., Pascale R. M. Polygenic control of susceptibility and resistance to hepatocarcinogenesis Dragani T. A. eds. .
Human Polygenic Diseases Animal Models
,
:
189
-206, Harwood Academic Publishers Amsterdam  
1998
.
8
Wood G. A., Korkola J. E., Lee V. M., Sarma D. S. R., Archer M. C. Resistance of Copenhagen rats to chemical induction of glutathione S-transferase 7-7-positive liver foci.
Carcinogenesis (Lond.)
,
18
:
1745
-1750,  
1997
.
9
Pascale R. M., Simile M. M., De Miglio M. R., Muroni M. R., Gaspa L., Dragani T. A., Feo F. The BN rat strain carries dominant hepatocarcinogen resistance loci.
Carcinogenesis (Lond.)
,
17
:
1765
-1768,  
1996
.
10
De Miglio M. R., Canzian F., Pascale R. M., Silime M. M., Muroni M. R., Calvisi D., Romeo G., Feo F. Identification of genetic loci controlling hepatocarcinogenesis on rat chromosomes 7 and 10.
Cancer Res.
,
59
:
4651
-4657,  
1999
.
11
Yoshimoto F., Masuda S., Higashi T., Nishii T., Takamisawa I., Tateishi N., Sakamoto Y. Comparison of drug-metabolizing activities in the livers of carcinogen-sensitive parent rats and carcinogen-resistant descendants.
Cancer Res.
,
45
:
6155
-6159,  
1985
.
12
Higashi T., Fukui R., Sekiyama M., Yoshimoto F., Tateishi N., Sakamoto Y. Decreased induction of γ-glutamyltransferase activity by 3′-methyl-4-dimethylaminoazobenzene in the liver of rats given carcinogen-containing diet for several generations.
Jpn. J. Cancer Res.
,
77
:
139
-144,  
1986
.
13
Yano M., Higashi T., Kano S., Tateishi N., Kido Y., Mori T., Higashi K., Sakamoto Y. Decreased activation of carcinogens in the liver of carcinogen-resistant rats.
Cancer Res.
,
49
:
5352
-5357,  
1989
.
14
Yan Y., Higashi K., Yamamura K., Fukamachi Y., Abe T., Gotoh S., Sugiura T., Hirano T., Higashi T., Ichiba M. Different responses other than the formation of DNA-adducts between the livers of carcinogen-resistant rats (DRH) and carcinogen-sensitive rats (Donryu) to 3′-methyl-4-dimethylaminoazobenzene administration.
Jpn. J. Cancer Res.
,
89
:
806
-813,  
1998
.
15
Denda A., Kitayama W., Konishi Y., Yan Y., Fukamachi Y., Miura M., Gotoh S., Ikemura K., Abe T., Higashi T., Higashi K. Genetic properties for the suppression of development of putative preneoplastic glutathione S-transferase placental form-positive foci in the liver of carcinogen-resistant DRH strain rats.
Cancer Lett.
,
140
:
59
-67,  
1999
.
16
Yamamoto K., Yokose Y., Nakajima A., Eimoto H., Shiraiwa K., Tamura K., Tsutsumi M., Konishi Y. Comparative histochemical investigation of the glutathione S-transferase placental form and γ-glutamyltranspeptidase during N-nitrosobis(2-hydroxypropyl)amine-induced lung carcinogenesis in rats.
Carcinogenesis (Lond.)
,
9
:
339
-404,  
1988
.
17
Squire R. A., Levitt M. H. Report of a workshop on classification of specific hepatocellular lesions in rats.
Cancer Res.
,
35
:
3214
-3223,  
1975
.
18
Institute of Laboratory Animal Resources. Histologic typing of liver tumors of the rat.
J. Natl. Cancer Inst.
,
64
:
177
-206,  
1980
.
19
Chirgwin J. M., Przybyla A. E., MacDonald R. J., Rutter W. J. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease.
Biochemistry
,
18
:
5294
-5299,  
1979
.
20
Konishi T., Karasaki Y., Nomoto M., Ohmori H., Shibata K., Abe T., Shimizu K., Itoh H., Higashi K. Induction of heat shock protein 70 and nucleolin and their intracellular distribution during early stage of liver regeneration.
J. Biochem. (Tokyo)
,
117
:
1170
-1177,  
1995
.
21
Yamada Y., Matsushiro H., Ogawa M. S., Okamoto K., Nakakuki Y., Toyokuni S., Fukumoto M., Hiai H. Genetic predisposition to Pre-B lymphomas in SL/Kh strain mice.
Cancer Res.
,
54
:
403
-407,  
1994
.
22
Brown D. M., Matise T. C., Koike G., Simon J. S., Winer E. S., Zangen S., McLaughli M. G., Shiozawa M., Atkinson O. S., Hudson J. R., Jr., Chakravarti A., Lander E. S., Jacob H. J. An integrated genetic linkage map of the laboratory rat.
Mamm. Genome
,
9
:
521
-530,  
1998
.
23
Basten C. J., Weir B., Zeng Z-B. QTL Cartographer, Version 1
13
North Carolina State University Program in Statistical Genetics. Raleigh, NC  
1999
.
24
Lander E. S., Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results.
Nat. Genet.
,
11
:
241
-247,  
1995
.
25
Ferreira-Gonzalez A., Deangelo A. B., Nasim S., Garrett C. T. Ras oncogene activation during hepatocarcinogenesis in B6C3F1 male mice by dichloroacetic and trichloroacetic acids.
Carcinogenesis (Lond.)
,
16
:
495
-500,  
1995
.
26
Dragani T. A., Manenti G., Colombo B. M., Falvella F. S., Gariboldi M., Pierotti M. A. , and Della Porta, G.
Incidence of mutations at codon 61 of the Ha-ras gene in liver tumors of mice genetically susceptible and resistant to hepatocarcinogenesis. Oncogene
,
6
:
333
-338,  
1991
.
27
Sohn Y-W., Lee G-H., Liem A., Miller J. A. Activation of H-ras oncogenes in male B6C3F1 mouse liver tumors induced by vinthionine or 2-chloroethyl methyl sulfide.
Carcinogenesis (Lond.)
,
17
:
1361
-1364,  
1996
.
28
Buchmann A., Bauer-Hofmann R., Mahr J., Drinkwater N. R., Luz A., Schwarz M. 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
:
911
-915,  
1991
.
29
Sers C., Emmenegger U., Husmann K., Bucher K., Andres A. C., Schafer R. Growth-inhibitory activity and downregulation of the class II tumor-suppressor gene H-rev 107 in tumor cell lines and experimental tumors.
J. Cell Biol.
,
136
:
935
-944,  
1997
.
30
Sohda T., Iwata K., Soejima H., Kamimura S., Shijo H., Yun K. In situ detection of insulin-like growth factor II (IGF2) and H19 gene expression in hepatocellular carcinoma.
J. Hum. Genet.
,
43
:
49
-53,  
1998
.
31
Reid L. H., Davies C., Cooper P. R., Crider-Miller S. J., Sait S. N., Nowak N. J., Evans G., Stanbridge E. J., deJong P., Shows T. B., Weissman B. E., Higgins M. J. A 1-MB physical map PAC contig of the imprinted domain in 11p15.5 that contains TAPA1 and the BWSCR1/WT2 region.
Genomics
,
43
:
366
-375,  
1997
.
32
Sakumi K., Shiraishi A., Shimizu S., Tsuzuki T., Ishikawa T., Sekiguchi M. Methylnitrosourea-induced tumorigenesis in MGMT gene knockout mice.
Cancer Res.
,
57
:
2415
-2418,  
1997
.
33
Iwakuma T., Sakumi K., Nakatsuru Y., Kawate H., Igarashi H., Shiraishi A., Tsuzuki T., Ishikawa T., Sekiguchi M. High incidence of nitrosamine-induced tumorigenesis in mice lacking DNA repair methyltransferase.
Carcinogenesis (Lond.)
,
18
:
1631
-1635,  
1997
.
34
Watanabe T. K., Bihoreau M-T., McCarthy L. C., Kiguwa S. L., Hishigaki H., Tsuji A., Browne J., Yamasaki Y., Mizoguchi-Miyakita A., Oga K., Ono T., Okuno S., Kanemoto N., Takahashi E-I., Tomita K., Hayashi H., Adachi M., Webber C., Davis M., Kiel S., Knights C., Smith A., Critcher R., Miller J., Thangarajah T., Day P. J. R., Hudson J. R., Jr., Irie Y., Takagi T., Nakamura Y., Goodfellow P. N., Lathrop G. M., Tanigami A., James M. R. A radiation hybrid map of the rat genome containing 5,255 markers.
Nat. Genet.
,
22
:
27
-36,  
1999
.
35
Godfrey P., Rahal J. O., Beamer W. G., Copeland N. G., Jenkins N. A., Mayo K. E. GHRH receptor of little mice contains a missense mutation in the extracellular domain that disrupts receptor function.
Nat. Genet.
,
4
:
227
-232,  
1993
.
36
Aihara T., Noguchi S., Miyoshi Y., Nakano H., Sasaki Y., Nakamura Y., Monden M., Imaoka S. Allelic imbalance of insulin-like growth factor II gene expression in cancerous and precancerous lesions of the liver.
Hepatology
,
28
:
86
-89,  
1998
.
37
Gariboldi M., Pascale R., Manenti G., De Miglio M. R., Calvisi D., Carru A., Dragalio T. A., Feo F. Analysis of loss heterozygosity in neoplastic nodules induced by diethylnitrosamine in the resistant BFF1 rat strain.
Carcinogenesis (Lond.)
,
20
:
1363
-1368,  
1999
.
38
Drinkwater N. R., Ginsler J. J. Genetic control of hepatocarcinogenesis in C57BL/6J and C3H/HeJ inbred mice.
Carcinogenesis (Lond.)
,
7
:
1701
-1707,  
1986
.
39
Bennett L. M., Winkler M. L., Drinkwater N. R. A gene that determines the high susceptibility of the C3H/HeJ strain of mouse to liver tumor induction is located on the chromosome one.
Proc. Am. Assoc. Cancer Res.
,
34
:
144
1993
.
40
Gariboldi M., Manenti G., Canzian F., Falvella F. S., Pierotti M. A., Della Porta G., Binelli G., Dragani T. A. Chromosome mapping of murine susceptibility loci to liver carcinogenesis.
Cancer Res.
,
53
:
209
-211,  
1993
.
41
Manenti G., Binelli G., Gariboldi M., Canzian F., De Gregorio L., Falvella F. S., Dragani T. A., Pierotti M. A. Multiple loci affect genetic predisposition to hepatocarcinogenesis in mice.
Genomics
,
23
:
118
-124,  
1994
.
42
Lee G. H., Bennett L. M., Carabeo R. A., Drinkwater N. R. Identification of hepatocarcinogen-resistance genes in DBA/2 mice.
Genetics
,
139
:
387
-395,  
1995
.
43
Poole T. M., Drinkwater N. R. Two genes abrogate the inhibition of murine hepatocarcinogenesis by ovarian hormones.
Proc. Natl. Acad. Sci. USA
,
93
:
5848
-5853,  
1996
.
44
Fernandez E., La Vecchia C., D’Avanzo B., Negri E., Francheschi S. Family history and the risk of liver, gallbladder, and pancreatic cancer.
Cancer Epidemiol. Biomark. Prev.
,
3
:
209
-214,  
1994
.
45
Alberts S. R., Lainer A. P., MeMahon B. J., Harpster A., Bulkow L. R., Heywars W. L., Murray C. Clustering of hepatocellular carcinoma in Alaska native families.
Genet. Epidemiol.
,
8
:
127
-139,  
1991
.
46
Shen F-M., Lee M-K., Gong H-M., Cai X-Q., King M-C. Complex segregation analysis of primary hepatocellular carcinoma in Chinese families: interaction of inherited susceptibility and hepatitis B viral infection.
Am. J. Hum. Genet.
,
19
:
88
-93,  
1991
.
47
Zenklusen J. C., Rodriguez L. V., LaCava M., Wang Z., Goldstein L. S., Conti C. J. Novel susceptibility locus for mouse hepatomas: evidence for a conserved tumor suppressor gene.
Genome Res.
,
6
:
1070
-1076,  
1996
.
48
Wang H. P., Roger C. E. Deletion in human chromosome arms 11p and 13q in primary hepatocellular carcinomas.
Cytogenet. Cell Genet.
,
48
:
72
-78,  
1988
.
49
Zang W., Hirohashi S., Tsuda H., Shimosato Y., Yokota J., Terada M., Sugimura T. Frequent loss of heterozygosity on chromosome 16 and 4 in human hepatocellular carcinoma.
Jpn. J. Cancer Res.
,
81
:
108
-111,  
1990
.
50
De Souza A. T., Hankins G. R., Washington M. K., Orton T. C., Jirtle R. L. M6P/IGF2R gene is mutated in human hepatocellular carcinomas with loss of heterozygosity.
Nat. Genet.
,
11
:
447
-449,  
1995
.
51
Pineau P., Nagai H., Prigent S., Wei Y., Gyapay G., Weissenbach J., Tiollais P., Buendia M. A., Dejean A. Identification of three distinct regions of allelic deletions on the short arm of chromosome 8 in hepatocellular carcinoma.
Oncogene
,
18
:
3127
-3134,  
1999
.
52
Mills J. J., Falls J. G., De Souza A. T., Jirtle R. L. Imprinted M6p/Igf2 receptor is mutated in rat liver tumors.
Oncogene
,
16
:
2797
-2802,  
1998
.
53
Yuan B. Z., Miller M. J., Keck C. L., Zimonjic D. B., Thorgeirsson S. S., Popescu N. C. Cloning, characterization, and chromosomal localization of a gene frequently deleted in human liver cancer (DLC) homologous to rat RhoGAP.
Cancer Res.
,
58
:
2196
-2199,  
1998
.
54
Sargent L., Dragan Y., Xu Y-H., Sattler G., Wiley J., Pitot H. C. Karyotypic changes in a multistage model of chemical hepatocarcinogenesis in the rat.
Cancer Res.
,
56
:
2985
-2991,  
1996
.
55
Albrecht S., von Schweinitz D., Waha A., Kraus J. A., von Deimling A. V., Pietsch T. Loss of maternal alleles on chromosome arm 11p in hepatoblastoma.
Cancer Res.
,
54
:
5041
-5044,  
1994
.