The C3H/HeJ (C3H) and CBA/J (CBA) mouse strains are classical mouse models of cancer susceptibility, exhibiting high risks for both spontaneous and chemically induced liver cancer. By analysis of backcrosses and intercrosses between C3H or CBA and resistant B6 mice, we have mapped a potent modifier of hepatocellular carcinoma development to distal chromosome 1, linked to the marker D1Mit33 with combined LOD_W scores of 5.9 (C3H) and 6.5 (CBA). We previously identified this region as one of two that modify susceptibility in the more distantly related C57BR/cdJ (BR) strain. Congenic B6.C3H(D1Mit5-D1Mit17) and B6.BR(D1Mit5-D1Mit17) mice developed significantly more liver tumors than B6 mice did (6- to 13-fold, P < 10^–11, in males; 3- to 4-fold, P < 10^–3, in females). Thus, distal chromosome 1 carries one or more genes that are sufficient to confer susceptibility to liver cancer.

---

THE C3H/HeJ (C3H) and CBA/J (CBA) mouse strains are classical models of liver carcinogenesis, highly susceptible to both spontaneous and carcinogen-induced liver cancer (ANDERVONT 1950; FLAKS 1968; SMITH et al. 1973). The genetic basis for their susceptibility has not been established. Identifying the number and types of genes underlying their susceptibility is likely to have public health implications, as liver cancer in C3H-derived B6C3F₁ mice is the single most common carcinogenic response to the >500 compounds that have been tested in chronic bioassays by the National Toxicology Program (ASHBY and TENNANT 1991; http://ntp-server.niehs.nih.gov/; January 2004).

By 2 years of age, 30–50% of C3H mice spontaneously develop hepatocellular carcinoma (HCC), the most common form of liver cancer in mice and humans (STORER 1966; SMITH and WALFORD 1978). In contrast, <5% of 2-year-old B6 mice develop HCC (FRITH and WILEY 1982). On the basis of modeling studies, we have suggested that the majority (85%) of the difference in susceptibility between B6 and C3H is controlled by one locus (DRINKWATER and GINSLER 1986). This quantitative trait locus (QTL) has been named Hcs7 (Hepatocarcinogen sensitivity 7) on the basis of a preliminary report of its location (BENNETT et al. 1993).

Hcs7 appears to control early stages of tumor development. Preneoplastic lesions, induced by treatment of male mice with N,N-diethylnitrosamine (DEN) or N-ethyl-N-nitrosourea (ENU) at 12–15 days of age, grow about twofold faster in 12- to 32-week-old C3H mice than in B6 mice (HANIGAN et al. 1988; PUGH and GOLDFARB 1992). The implication that the net growth of preneoplastic lesions in intact B6 livers is limited is supported by the results of partial hepatectomy of DEN-treated livers. The growth stimulus provided by partial removal of the liver causes a dramatic increase in lesion number and size in B6, but not C3H, mice (HANIGAN et al. 1990), suggesting that partial hepatectomy activates in B6 a growth pathway that is already active in C3H lesions. This strain-specific difference in growth control is also evident in untreated livers that have undergone partial hepatectomy. The level of DNA synthesis in these normal hepatocytes is over twofold higher in C3H mice than in B6 mice (BENNETT et al. 1995), indicating that Hcs7 may play a role in normal hepatocyte growth control.

Hcs7 does not significantly affect carcinogen metabolism and tumor initiation. C3H mice are more susceptible than B6 to liver tumor induction by a variety of carcinogens that differ in their metabolic activation, and they are also more susceptible to spontaneous tumors (NAGASAKI et al. 1975; DRINKWATER and GINSLER 1986; HOOVER et al. 1986; WISEMAN et al. 1986). In addition, similar numbers of DNA adducts and preneoplastic lesions form in the livers of carcinogen-treated B6 and C3H mice (DRINKWATER and GINSLER 1986). Although 12-week-old C3H mice develop 25-fold more detectable preneoplastic foci than do age-matched B6 mice, this difference diminishes dramatically with age. This result suggests that similar numbers of lesions are initiated in B6 and C3H livers, but that B6 hepatic foci take longer on average to grow to detectable size, supporting the hypothesis that Hcs7 controls preneoplastic growth (HANIGAN et al. 1988).

Studies of chimeric mice indicate that Hcs7 affects tumorigenesis from within the liver, probably at the level of the hepatocyte (CONDAMINE et al. 1971; LEE et al. 1991). In chimeric mice derived from aggregated C3H and B6 embryos, spontaneous and DEN-induced tumors develop mainly from C3H hepatocytes—even in livers derived predominantly from B6 hepatocytes (CONDAMINE et al. 1971; LEE et al. 1991). In addition, the effect of Hcs7 appears tissue specific. C3H mice are more susceptible than B6 to tumorigenesis in the liver, but they are more resistant than B6 to colon cancer and comparable to B6 in susceptibility to lung cancer (TURUSOV et al. 1982; MALKINSON and BEER 1983; DRINKWATER and GINSLER 1986).

C57BR/cdJ (BR) mice, originally derived from the same breeding pair that generated B6 mice (BECK et al. 2000), are up to 50-fold more susceptible than B6 to liver tumorigenesis (KEMP and DRINKWATER 1989). The loci responsible for this difference have been mapped to chromosome 17 (Hcf1) and distal chromosome 1 (Hcf2; POOLE and DRINKWATER 1996). Analysis of chimeras showed that the net effect of these factors is intrinsic to the liver and may be cell autonomous (CHIAVEROTTI and DRINKWATER 2003). A number of other loci, on chromosomes 2 (Hcs4), 4 (Hcr1), 5 (Hcs5), 7 (Hcs1), 8 (Hcs2), 10 (Hcr2), 12 (Hcs3), and 19 (Hcs6) have been implicated in the control of liver tumorigenesis on the basis of mapping crosses involving a variety of strains and carcinogens (GARIBOLDI et al. 1993; MANENTI et al. 1994; LEE et al. 1995). Some of these loci may also control the growth rate of preneoplastic cells, which differs among the strains used in these crosses (A/J, DBA/2, BALB/c, and C3H; DRAGANI et al. 1987, 1991; LEE and DRINKWATER 1995b).

We report here the first linkage analyses of a QTL that explains most of the difference in liver cancer susceptibility between B6 and C3H or CBA mice. We have mapped Hcs7 to distal chromosome 1 by analysis of backcrosses and intercrosses between B6 mice and both susceptible strains. We have also generated congenic animals that carry, on a B6 background, a 70-cM region of distal chromosome 1 derived from either C3H or BR. These congenic mice confirm the location, potency, and independent action of the C3H chromosome 1 modifier.

Mice:

B6, BR, C3H, and CBA mice were purchased from the Jackson Laboratory (Bar Harbor, ME) and bred in our facilities. All mice were housed in plastic cages on corncob bedding (Bed O'Cobs, Anderson Cob Division, Maumee, OH), fed Wayne Breeder Blox (Figure 1 ; Table 1 ; 11% fat; Continental Grain, Chicago) or Purina 5020 (Table 2 ; 9% fat; St. Louis) diet, and given acidified tap water ad libitum. Mice were inspected daily and weighed monthly.

View larger version (36K): In this window In a new window Download PPT slide   FIGURE 1.— A modifier of liver tumor multiplicity on distal chromosome 1. The log relative significance [log(0.05/genome-wide P value)] is plotted for markers on chromosome 1 (left) and for markers on all other chromosomes (right). Values corresponding to genome-wide P values of 0.05 and 0.01 are indicated by the dotted horizontal lines. (A) C3H crosses: solid circle, B6C3F1 x B6; open circle, B6C3F2; solid square, B6.C3H-Ch1 N5 backcross. (B) CBA crosses: solid diamond, B6CBF1 x B6; open diamond, B6CBF2.

 

View this table: In this window In a new window   TABLE 1 Linkage of DNA markers to liver tumor susceptibility in congenic backcross mice

 

View this table: In this window In a new window   TABLE 2 Liver tumor susceptibility in inbred parental and chromosome 1 congenic mice

 
Congenic B6.C3H(D1Mit5-D1Mit17) (abbreviated as B6.C3H-Ch1) and B6.BR(D1Mit5-D1Mit17) (B6.BR-Ch1) were generated as follows. B6 and C3H or B6 and BR mice were mated to yield B6C3F₁ or B6BRF₁ animals. F₁ males were then backcrossed to B6 females. Mice carrying an 70-cM region of distal chromosome 1 derived from the C3H or BR strains were selected for additional backcrossing. The marker loci D1Mit5 and D1Mit17 were used to select the endpoints of the congenic region. Four to five independent lines were maintained during the generation of B6.C3H-Ch1 and B6.BR-Ch1 congenics. After two more rounds of mating, N₄ congenic male mice from each line were crossed to B6 females to generate experimental progeny (Table 1). N₄ congenic mice were also used in continued backcrossing to yield animals (N₁₀) that were then intercrossed to yield fully homozygous animals carrying the selected C3H or BR region on a B6 background (B6.C3H-Ch1 and B6.BR-Ch1; Table 2). One of each set of congenic lines was chosen for further analysis on the basis of phenotypic validation in the N₄ backcross, progeny testing, and thorough genotypic validation at N₁₀.

Tumor induction and assessment:

Tumors were induced by a single intraperitoneal injection of DEN (Eastman Kodak, Rochester, NY; 0.05 µmol/g body weight for B6CBF₁ x B6 backcross and B6CBF₂ intercross mice; 0.1 µmol/g body weight for all other mice except B6C3F₂; DRINKWATER and GINSLER 1986) or ENU (0.25 µmol/g body weight for B6C3F₂ mice; DRINKWATER and GINSLER 1986) dissolved in trioctanoin (Pfalz and Bauer, Stamford, CT or Sigma, St. Louis) 12 ± 1 days after birth. Male mice were killed by CO₂ asphyxiation at 31–32 weeks of age; females were killed at 49–50 weeks. Livers were removed and weighed; all tumors with diameters >2 mm (B6C3F₁ x B6 backcross mice; DRINKWATER and GINSLER 1986) or 1 mm (all other mice; HANIGAN et al. 1990) on the surface of the liver were counted. Liver tumors were sampled at random and fixed in buffered formalin, and embedded sections were stained with hematoxylin and eosin. Spleens were collected as a source of DNA and frozen on dry ice. All tumors in a given cross were scored by a single observer blind to genotype.

Genotyping:

Spleen DNA was prepared as follows: 5 mm³ of spleen was placed in 500 µl of a lysis solution (1% SDS, 150 mM NaCl, 100 mM EDTA, 20 mM Tris-Cl, pH 8.0) plus 25 µl proteinase K (10 mg/ml; 0.5 mg/ml final) and incubated at 55° for 3–20 hr. Cellular debris was precipitated with 250 µl 6.25 M ammonium acetate and pelleted. DNA was precipitated from the supernatant with 700 µl 100% isopropyl alcohol, and the DNA pellet was washed with 70% ethanol. The resulting genomic DNA pellet was resuspended in 250 µl of TE (10 mM Tris, 1 mM EDTA, pH 8.0).

Microsatellite markers (MCALEER et al. 1992; DIETRICH et al. 1996) were amplified using 1 or 2 µl of spleen DNA (100 ng), 125–190 nM each primer, 50 µM dNTPs (Amersham, Piscataway, NJ), PCR buffer (Roche, Indianapolis; 10 mM Tris/HCl, 1.5 mM MgCl₂, 50 mM KCl, pH 8.3), and 0.024 units/µl Taq polymerase (Roche) in a total reaction volume of 20.5 (1 µl spleen DNA) or 21.5 (2 µl DNA). The reactions were incubated in thermocyclers at 94° for 3 min; followed by 40 cycles of 94° for 30 sec, 55° for 40 sec, and 72° for 60 sec; and followed by 7 min at 72°. The products were separated by electrophoresis through a 7% acrylamide gel.

B6C3F₁ x B6 backcross progeny were genotyped at the following 107 markers: D1Mit1, -5, -10, -13, -14, -15, -17, -26, -36; D2Mit1, -7, -13, -48, -49, -53, -74; D3Mit3, -7, -11, -17, -19, -42, -45, -55; D4Mit12, -13, -39, -42; D5Mit11, -24, -32, -63; D5Nds2; D6Mit1, -10, -14, -15, -16, -25, -29; D7Mit7, -21, -56; D7Nds1, -4, -5; D8Mit4, -8, -13, -16, -33, -42, -46; D9Mit2, -4, -12, -17, -18, -19; D9Nds2; D10Mit3, -10, -12, -14, -31; D10Nds1; D11Mit2, -5, -12, -14, -20, -41; D12Mit5, -34, -46; D12Nds2, -11; D13Mit3, -8, -13, -35; D14Mit5, -7, -14, -28; D15Mit6, -42; D15Nds2; D16Mit4, -9, -30; D17Mit1, -3, -6, -10, -18, -23, -35; D18Mit4, -8, -9, -17, -22, -33; and D19Mit1, -11, -16. A subset of 34 animals, randomly chosen, was tested at every marker locus. The remaining 24 progeny were tested only at marker loci on chromosomes showing significant linkage to the tumor multiplicity phenotype. B6C3F₂ mice were typed at the following 50 markers: D1Mit3, -5, -7, -13, -14, -15, -17, -21, -23, -26, -33, -34, -36, -46, -54, -61; D2Mit7, -13, -48; D3Mit7, -11, -17; D4Mit12; D5Nds2; D5Mit24; D6Mit33, -29; D7Nds1; D8Mit4, -33; D9Mit4; D10Mit68; D11Mit14, -20, -41; D12Mit5, -46; D13Mit3, -13, -35, -51; D14Mit14, -28; D15Mit2, -6; D16Mit4; D17Mit3, -23, -68; and D18Mit17, -33.

Fifty-three B6CBF₁ x B6 backcross progeny were genotyped at the following 74 marker loci: D1Mit3, -5, -13, -17, -30, -33, -60, -113, -150; D2Mit1, -35, -48, -49, -57, -62, -148; D3Mit62, -6, -9, -11, -14, -17, -19; D4Mit9, -16, -33; D5Mit61, -95; D5Nds2; D6Mit1, -9, -10, -15, -17, -25; D7Mit34, -56; D7Nds2, -4; D8Mit3, -41, -88; D9Mit2, -6, -10; D10Mit3, -10, -31, -72; D11Mit19, -23; D11Nds1; D12Mit5, -12, -20, -34; D13Mit3, -13, -30; D14Mit7, -14, -28, -62; D15Mit3, -43; D16Mit9, -30; D17Mit16, -38, -70; D18Mit4; and D19Mit10, -13, -31. Ninety-five B6CBF₂ intercross progeny were genotyped at the following 12 marker loci: D1Mit3, -5, -13, -17, -33, -60, 113, -150; D12Mit5, -12, -34; and D13Mit13.

Spleen DNA from 138 B6 x B6.C3H-Ch1 and 149 B6 x B6.BR-Ch1 N₅ progeny were genotyped at the following markers: D1Mit5; D1Mit285 or -89 (both at 63 cM); D1Mit33; and D1Mit17 or -117 (both at 106 cM).

Two or three animals from the B6.C3H-Ch1 and B6.BR-Ch1 homozygous congenic lines at generation N₁₀ were tested at marker loci spaced approximately every 5 cM throughout the congenic region. The markers used for B6.C3H-Ch1 were D1Mit64, -66, -231, -211, -233, -5, -19, -23, -215, -83, -10, -135, -285, -91, -218, -100, -105, -33, -399, -15, -13, -206, -166, -461, -17; markers used for B6. BR-Ch1 were D1Mit5, -19, -23, -215, -83, -10, -135, -285, -105, -33, -399, -143, -206, -17. Underlining indicates markers that were used to identify breeders during backcrossing to generate the congenic lines.

Linkage analysis:

We used a nonparametric approach to assess linkage between the marker loci and the quantitative trait loci that determine liver tumor multiplicity (KRUGLYAK and LANDER 1995; POOLE and DRINKWATER 1996). For backcross and intercross mice, the data for each marker were analyzed using the Wilcoxon rank sum or Jonckheere-Terpstra tests, respectively, to obtain the test statistic Z_W (LEHMAN 1998). The genome-wide, null distribution of |Z_W| was determined empirically for each experiment by permutation of the phenotypic data (100,000 permutations for each cross) as described by CHURCHILL and DOERGE (1994). For each permutation, max_cross(|Z_W|) was recorded and this distribution was used to determine the genome-wide significance (two-sided) for linkage to each marker (LYSTIG 2003). Linkage of markers to minor quantitative trait loci was assessed by a conditional permutation test in which the data were stratified by the genotype at the marker nearest the major quantitative trait locus (DOERGE and CHURCHILL 1996). This analysis should also reveal interactions between major and minor loci. Equivalent LOD (logarithm of odds) scores (KRUGLYAK and LANDER 1995), LOD_W, were estimated from LOD_W = 0.5 (log₁₀ e) (Z_W)². These analyses were performed using Qlink 3.2 software, which is available from the authors (http://mcardle.oncology.wisc.edu/qlink).

B6C3F₁ x B6 backcross and B6C3F₂ intercross:

The high sensitivity of C3H mice is governed largely by a single locus (DRINKWATER and GINSLER 1986). To map this QTL, we generated 58 B6C3F₁ x B6 backcross male progeny, injected them with DEN at 12 days of age, and counted their liver tumors at 32 weeks of age. The tumor multiplicity phenotypes were correlated with genotypes at 107 microsatellite markers spread at 15-cM intervals throughout the 19 autosomes. (Previous analysis of F₁ mice had shown that the locus was not carried by the X or Y chromosomes; DRINKWATER and GINSLER 1986.) Segregation analysis reveals a significantly linked region centered at D1Mit15 at 88 cM on chromosome 1, with a LOD_W of 3.06 (genome-wide P value = 0.0067; Figure 1A). B6C3F₁ x B6 mice heterozygous for the C3H allele at D1Mit15 developed twofold more tumors than their siblings that were homozygous B6 at this locus (40 ± 20 vs. 20 ± 21). Inbred C3H and B6 mice, treated in parallel, developed 78 ± 30 and 1.4 ± 1.6 tumors, respectively.

We independently tested this region's ability to modify liver tumor multiplicity and determined the effect of locus dosage by analyzing 57 B6C3F₂ intercross mice treated with ENU. Unlike DEN, which requires metabolic activation, ENU is a direct-acting alkylating agent. However, the two carcinogens yield identical patterns of ethylated bases in DNA (BERANEK et al. 1980) and highly similar strain distribution patterns for liver tumor induction among BXH recombinant inbred strains (LEE and DRINKWATER 1995a). Intercross mice were injected intraperitoneally at 12 days and killed at 32 weeks of age. Again, C3H alleles on distal chromosome 1 were most tightly linked with liver tumor susceptibility (Figure 1A). Specifically, D1Mit13 at 63 cM yielded a significant LOD_W score of 2.85 (P_genome = 0.007; Figure 1A). This susceptibility locus appears semidominant. B6C3F₂ mice heterozygous for the C3H allele at D1Mit13 developed threefold more tumors than B6 homozygotes did (10 ± 10 vs. 3.1 ± 5), and homozygosity for the C3H allele increased the number of tumors an additional twofold (19 ± 14 vs. 10 ± 10). Inbred C3H and B6 mice, treated in parallel with ENU, developed 20 ± 12 and 1.5 ± 1.6 tumors, respectively.

No other loci were found to interact significantly with the B6 or C3H alleles at D1Mit13. In this intercross analysis, mice lacking C3H alleles on distal chromosome 1 were generally not typed elsewhere. Therefore, the interaction of recessive B6 alleles on chromosome 1 with recessive C3H alleles elsewhere might not have been detected. Interactions with dominant C3H alleles elsewhere should have been detected in the backcross.

B6CBF₁ x B6 backcross and B6CBF₂ intercross:

The CBA inbred strain was derived from the same C line as C3H and is almost identically susceptible to liver cancer, whether spontaneous or induced by a variety of carcinogens (GRASSO and HARDY 1975; DRINKWATER 1989). To map the CBA susceptibility loci, 53 B6CBF₁ x B6 backcross and 95 B6CBF₂ intercross male progeny were injected with DEN at 12 days of age and tumors were counted at 32 weeks of age. The tumor multiplicity phenotypes of 53 backcross animals were correlated with their genotypes at 74 marker loci at 20-cM intervals. Marker D1Mit113 at 93 cM on chromosome 1 yielded a significant LOD_W score of 3.29 (P_genome = 0.0036; Figure 1B). The results of the F₂ intercross confirm the presence of a single strong modifier on chromosome 1. Significant linkage spanned the region between D1Mit13 and D1Mit17, completely overlapping the susceptibility region in C3H, with a peak LOD_W score of 3.21 at D1Mit33 at 82 cM (P_genome = 0.0011; Figure 1B). Animals carrying the dominant CBA allele developed between 2.5- and 3-fold more tumors than B6 homozygotes did (106 ± 43 vs. 42 ± 40, intercross; 69 ± 38 vs. 24 ± 29, backcross). Inbred CBA and B6 mice, treated in parallel, developed 166 ± 157 and 4 ± 6 tumors, respectively. No other loci were found to interact significantly with the chromosome 1 modifier.

Congenic backcross:

The above mapping crosses between B6 and C3H or CBA and previous mapping crosses between B6 and BR (POOLE and DRINKWATER 1996) identified modifiers of liver cancer risk on distal chromosome 1. To verify the existence of these modifiers and test their ability to act alone, we generated congenic animals carrying C3H or BR chromosome 1 regions on a B6 background. We selected for C3H or BR alleles at four marker loci spanning a 70-cM region of chromosome 1, from D1Mit5 to D1Mit17. After three generations of backcrossing, mice from four to five independent congenic lines (then at N₄) were crossed to B6 to generate N₅ backcross progeny. (On average, unlinked C3H or BR alleles outside the congenic region should compose only 3% of the genome in these N₅ congenics, with each subline carrying a different complement of residual heterozygosity.) The N₅ congenics were treated with DEN and their tumors were counted, yielding additional mapping data and an initial assessment of the allele's ability to act independently. Among 138 mice, progeny heterozygous for C3H or BR chromosome 1 alleles developed 4- to 5-fold more tumors than their homozygous B6 siblings did (Table 1), suggesting that the chromosome 1 locus acts independently of other alleles in the donor strain.

Many of the backcross progeny carried newly recombinant chromosomes in the large congenic region. These novel recombinants were used to map the modifier. Once again, peak linkage was near D1Mit33, with a LOD_W of 5.08 (P < 10^–4) for B6.C3H-Ch1 and a LOD_W of 4.52 (P < 10^–4) for B6.BR-Ch1. The data for all of the C3H and CBA mapping crosses are shown in Figure 1. Each cross yielded a highly significant LOD_W for distal chromosome 1. Combined, the three C3H crosses yield a LOD_W of 11.0, and the two CBA crosses yield a LOD_W of 6.5. Combining our previous linkage results for crosses between B6 and BR mice (POOLE and DRINKWATER 1996) with those for the B6.BR-Ch1 backcross yields a peak LOD_W score of 11.2 at D1Mit33.

Susceptibility of congenic mice:

We continued backcrossing the congenic animals carrying C3H or BR chromosome 1 regions to B6 animals and assessed the heterozygous and homozygous effects of each congenic region in a single 10th-generation line (N₁₀; 0.1% unlinked donor genome). At this generation, the congenic region consisted of the selected 70-cM interval and up to 40 additional megabase pairs proximal to D1Mit5.

We found that the C3H and BR chromosome 1 regions impart dramatic susceptibility to both males and females (Table 2). Homozygosity for 70 cM of C3H chromosome 1 caused congenic males to develop 13-fold more tumors than B6 males (P < 10^–11) and congenic females to develop 4-fold more tumors than B6 females (P < 10^–7). [Similar results were obtained with B6, C3H, and B6.C3H-Ch1 animals fed a diet containing 6% rather than 9% fat (data not shown).] These increases account for most of the difference in susceptibility between B6 and C3H, for both genders. Specifically, the 13-fold effect in B6.C3H-Ch1 males accounts for 78% of the 27-fold effect between the B6 and C3H strains (in terms of relative risk), and the 4-fold effect in B6.C3H-Ch1 females accounts for 86% of the 5-fold effect between strains. Homozygosity for BR chromosome 1 resulted in a 6-fold increase in B6.BR-Ch1 males (P < 10^–8) and a 3-fold increase in B6.BR-Ch1 females relative to B6 (P < 10^–3). The effect in males accounts for 100% of the 6-fold difference in susceptibility between B6 and BR, but the 3-fold effect in females accounts for only 42% of the 14-fold difference between B6 and BR females. The discrepancy in females is due to susceptibility alleles on chromosome 17 (POOLE and DRINKWATER 1996). Chromosome 1 alleles from both C3H and BR appear semidominant in females and dominant in males.

Tumors induced in parental and congenic mice were selected randomly and assessed histopathologically. The tumors were all hepatocellular in origin, with the exception of one cholangioma and three sections that exhibited nodules consistent with lymphoma. Among the 211 hepatocellular tumors examined, approximately equal numbers were diagnosed as adenomas and carcinomas. The distribution between tumor types was independent of gender or strain. Hematoxylin- and eosin-stained liver sections from B6, B6.C3H-Ch1 congenic, and C3H mice were also scored for eosinophilic inclusions. Although commonly found in B6 hepatic lesions (KAKIZOE et al. 1989), previous results suggested that these inclusions do not segregate with resistance to liver tumorigenesis (LEE and DRINKWATER 1995a). We observed many inclusions in susceptible B6.C3H-Ch1 livers, confirming the independent segregation of the inclusion and tumor resistance phenotypes (data not shown).

Distal chromosome 1 carries one or more potent modifiers of liver cancer risk that account for most of the difference in tumor multiplicity between the C3H and B6 strains and all of the difference between BR and B6 males. Linkage analysis of crosses between the B6 and C3H or CBA strains indicate that a QTL, Hcs7, lies near D1Mit33 at 82 cM. Our congenic analyses show that the C3H allele of Hcs7 (Hcs7^C3H) is sufficient to confer susceptibility to the resistant B6 strain. The identification of Hcs7 is based on both F₁ and congenic backcrosses, as well as F₂ intercrosses. Its location and independence were confirmed using N₁₀ congenic lines. These methods exceed the most rigorous guidelines for QTL analysis promoted in a recently published white paper by the COMPLEX TRAIT CONSORTIUM (2003). No other loci that are polymorphic in these crosses interact significantly with the Hcs7 modifier. We have been unable to map Hcs7 in several B6 x C3H recombinant inbred strains (LEE and DRINKWATER 1995a), an observation that bears further study and might reveal interactions between recessive B6 and C3H alleles.

Hcs7^C3H has a 13- to 14-fold effect on liver tumor multiplicity in congenic males. The only known modifiers more potent than Hcs7 in the liver are gender and growth hormone deficiency (VESSELINOVITCH and MIHAILOVICH 1967; VESSELINOVITCH 1990; BUGNI et al. 2001). Male mice are much more susceptible to liver tumorigenesis than females, and gonadectomy of either sex reduces this difference. Mutations in Tfm and Ghrhr, genes in the sex hormone and growth hormone pathways, confer 25- to 100-fold reductions in tumor multiplicity in carcinogen-treated mice (KEMP et al. 1989; BUGNI et al. 2001). The Hcs7 region contains no known component of these pathways. Accordingly, Hcs7 appears to have an effect independent of sex: on a B6 background, Hcs7^C3H confers increased tumor multiplicity to a similar degree in both genders (Table 2).

The congenic (N₁₀) Hcs7^C3H modifier appears dominant in males and semidominant in females (Table 2). This difference is unlikely to reflect any real difference in the genders, because previous experiments suggest that Hcs7 acts in a semidominant manner in males (Figure 1A; DRINKWATER and GINSLER 1986). Rather, the apparent dominance in congenic males might reflect our inability to detect some tumors because of their high density in homozygotes under these conditions.

Among loci that have previously been mapped as liver cancer modifiers, only the BR Hcf2 locus maps to the same chromosome as Hcs7 (GARIBOLDI et al. 1993; MANENTI et al. 1994; LEE et al. 1995; POOLE and DRINKWATER 1996). The possibility that Hcs7 and Hcf2 represent the same gene is supported by chimera analysis and the congenic data presented in Tables 1 and 2. In both C3H B6 chimeras and BR B6 chimeras, tumors develop mainly from the cells of the susceptible parent, suggesting that the predominant modifiers in C3H and BR act within hepatocytes (CONDAMINE et al. 1971; LEE et al. 1991; CHIAVEROTTI and DRINKWATER 2003). In addition, both B6.C3H-Ch1 mice and B6.BR-Ch1 mice develop severalfold more tumors than do B6. However, the effect of the BR congenic region is less than that of the C3H region (6-fold vs. 13-fold). This 2-fold difference might be explained by the presence of two (or more) polymorphic modifiers, only one of which is common to BR and C3H. Complexity in polymorphic modifier regions is frequent and might reflect the inheritance of linked gene families among inbred strains (CORMIER et al. 2000; reviewed in BALMAIN 2002). Linked modifiers might also help explain the greater effect of the Hcs7 region in the congenic lines than in the backcross and F₂ mice. In the congenics, the Hcs7 locus might act additively with other minor loci in the region, while in the segregating crosses the linked genes would be separated by recombination at some frequency (RESULTS; POOLE and DRINKWATER 1996). The presence of a linked modifier might also explain the broad peak of the B6C3F₂ cross. Alternatively, the more proximal distribution of this intercross peak might reflect loci that depend on the carcinogen used to induce the tumors. We are resolving this issue by fine-structure mapping.

Much of the Hcs7 region of chromosome 1 is orthologous to human chromosome 1q, which is amplified in about half of all tested hepatocellular carcinomas, independent of hepatitis status (LIN et al. 1999; GUAN et al. 2000; MARCHIO et al. 2000; TORNILLO et al. 2000; WONG et al. 2000; ZONDERVAN et al. 2000). Chromosome 1q is also amplified in >50% of breast cancers and in 20–40% of tumors from a variety of tissues (CLIMENT et al. 2002; HISLOP et al. 2002; SHAM et al. 2002). Our mapping suggests that Hcs7 lies near D1Mit33, at 160 Mb on mouse chromosome 1. The region of mouse chromosome 1 orthologous to human chromosome 1q extends almost uninterrupted from 130 Mb to the end of the chromosome at 197 Mb. A minimal region frequently amplified in human HCC, 1q21–23 (WONG et al. 1999, 2000, 2001; GUAN et al. 2000; MARCHIO et al. 2000), is orthologous to a subset of the congenic region close to D1Mit33, from 168 Mb to 197 Mb (http://www.ensembl.org; October 2003). This region contains a number of intriguing genes common to mouse and man, including pre-B-cell leukemia transcription factor 1 (Pbx1), regulators of G-protein signaling 4 and 5 (Rgs4, Rgs5), TNF ligand superfamily member Dedd, the receptor tyrosine kinase Ddr2, activating transcription factor 6- (ATF6-), and Fas antigen ligand (Tnfsf6). Identification of the gene product(s) of the Hcs7 locus is likely to lead to the identification of molecular targets for the prevention, early diagnosis, detection, or treatment of liver cancer.

The identity of Hcs7 should also help explain the mechanism(s) of action of carcinogenic compounds identified in 2-year bioassays performed by the National Cancer Institute and the National Toxicology Program. About half of the compounds tested are carcinogenic to mice, rats, or both (ASHBY and TENNANT 1991). Of these, 20% induce only liver tumors in B6C3F₁ mice. This response is likely to be mediated by Hcs7, which accounts for much of the susceptibility of B6C3F₁ mice to carcinogenesis by both DEN and ENU (Figure 1A; Tables 1 and 2). Key questions include whether a similar pathway is active in humans and whether humans carry a homolog of the susceptibility gene and a high frequency of functional variants.

The authors are grateful to Cynthia Leary, Mary Winkler, Donal Kaehler, Kim Leutkehoelter, Alissa Barthel, Andrew Kastenmeier, Andrew Schneider, ManChing Leung, Rachel Potempa, Annette Pownell, Matt Gigot, and Khelan Patel for expert technical assistance in various aspects of this work. We also thank our colleagues Bill Dove, Chris Bradfield, Eric Sandgren, Michael Gould, Amy Moser, Jennifer Drew, and Mara Feld for critical advice on this manuscript. This work was supported by grants CA22484, CA07175, CA09135, and CA14520 from the National Cancer Institute of the National Institutes of Health.

This article is dedicated to the memory of our late colleague, Kristin M. Liss.

¹ Present address: Center for Cancer Research, National Cancer Institute, Bethesda, MD 20892.

² Present address: Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, Hamilton, MT 59840.

³ Present address: UCSF Cancer Center, San Francisco, CA 94143.

⁴ Present address: Fairfax Family Practice Center, Fairfax, VA 22033.

ANDERVONT, H., 1950 Studies on the occurrence of spontaneous hepatomas in mice of strains C3H and CBA. J. Natl. Cancer Inst. 11: 581–592.

ASHBY, J., and R. W. TENNANT, 1991 Definitive relationships among chemical structure, carcinogenicity and mutagenicity for 301 chemicals tested by the U.S. NTP. Mutat. Res. 257: 229–306.[Medline]

BALMAIN, A., 2002 Cancer as a complex genetic trait: tumor susceptibility in humans and mouse models. Cell 108: 145–152.[CrossRef][Medline]

BECK, J. A., S. LLOYD, M. HAFEZPARAST, M. LENNON-PIERCE, J. T. EPPIG et al., 2000 Genealogies of mouse inbred strains. Nat. Genet. 24: 23–25.[CrossRef][Medline]

BENNETT, L. M., M. L. WINKLER and N. R. DRINKWATER, 1993 A gene that determines the high susceptibility of the C3H/HeJ strain of mouse to liver tumors is located on Chromosome 1. Proc. Am. Assoc. Cancer Res. 34: 144.

BENNETT, L. M., P. J. FARNHAM and N. R. DRINKWATER, 1995 Strain-dependent differences in DNA synthesis and gene expression in the regenerating livers of C57BL/6J and C3H/HeJ mice. Mol. Carcinogen. 14: 46–52.[Medline]

BERANEK, D. T., C. C. WEISS and D. H. SWENSON, 1980 A comprehensive quantitative analysis of methylated and ethylated DNA using high pressure liquid chromatography. Carcinogenesis 1: 595–606.[Abstract/Free Full Text]

BUGNI, J. M., T. M. POOLE and N. R. DRINKWATER, 2001 The little mutation suppresses DEN-induced hepatocarcinogenesis in mice and abrogates genetic and hormonal modulation of susceptibility. Carcinogenesis 22: 1853–1862.[Abstract/Free Full Text]

CHIAVEROTTI, T. A., and N. R. DRINKWATER, 2003 C57BR/cdJ hepatocarcinogen susceptibility genes act cell-autonomously in C57BR/cdJC57BL/6J chimeras. Cancer Res. 63: 4914–4919.[Abstract/Free Full Text]

CHURCHILL, G. A., and R. W. DOERGE, 1994 Empirical threshold values for quantitative trait mapping. Genetics 138: 963–971.[Abstract]

CLIMENT, J., J. A. MARTINEZ-CLIMENT, D. BLESA, M. J. GARCIA-BARCHINO, R. SAEZ et al., 2002 Genomic loss of 18p predicts an adverse clinical outcome in patients with high-risk breast cancer. Clin. Cancer Res. 8: 3863–3869.[Abstract/Free Full Text]

COMPLEX TRAIT CONSORTIUM, 2003 The nature and identification of quantitative trait loci: a community's view. Nat. Rev. Genet. 4: 911–916.[Medline]

CONDAMINE, H., R. P. CUSTER and B. MINTZ, 1971 Pure-strain and genetically mosaic liver tumors histochemically identified with the beta-glucuronidase marker in allophenic mice. Proc. Natl. Acad. Sci. USA 68: 2032–2036.[Abstract/Free Full Text]

CORMIER, R. T., A. BILGER, A. J. LILLICH, R. B. HALBERG, K. H. HONG et al., 2000 The Mom1^AKR intestinal tumor resistance region consists of Pla2g2a and a locus distal to D4Mit64. Oncogene 19: 3182–3192.[CrossRef][Medline]

DIETRICH, W. F., J. MILLER, R. STEEN, M. A. MERCHANT, D. DAMRON-BOLES et al., 1996 A comprehensive genetic map of the mouse genome. Nature 380: 149–152.[CrossRef][Medline]

DOERGE, R. W., and G. A. CHURCHILL, 1996 Permutation tests for multiple loci affecting a quantitative character. Genetics 142: 285–294.[Abstract]

DRAGANI, T. A., G. MANENTI and G. DELLA PORTA, 1987 Genetic susceptibility to murine hepatocarcinogenesis is associated with high growth rate of NDEA-initiated hepatocytes. J. Cancer Res. Clin. Oncol. 113: 223–229.[CrossRef][Medline]

DRAGANI, T. A., G. MANENTI and G. DELLA PORTA, 1991 Quantitative analysis of genetic susceptibility to liver and lung carcinogenesis in mice. Cancer Res. 51: 6299–6303.[Abstract/Free Full Text]

DRINKWATER, N. R., 1989 Genetic control of hepatocarcinogenesis in inbred mice, pp. 3–17 in Genes and Signal Transduction in Multistage Carcinogenesis, edited by N. H. COLBURN. Dekker, New York.

DRINKWATER, N. R., and J. J. GINSLER, 1986 Genetic control of hepatocarcinogenesis in C57BL/6J and C3H/HeJ inbred mice. Carcinogenesis 7: 1701–1707.[Abstract/Free Full Text]

FLAKS, A., 1968 The susceptibility of various strains of neonatal mice to the carcinogenic effects of 9,10-dimethyl-1,2-benzanthracene. Eur. J. Cancer 4: 579–585.

FRITH, C. H., and L. WILEY, 1982 Spontaneous hepatocellular neoplasms and hepatic hemangiosarcomas in several strains of mice. Lab. Anim. Sci. 32: 157–162.[Medline]

GARIBOLDI, M., G. MANENTI, F. CANZIAN, F. S. FALVELLA, M. A. PIEROTTI et al., 1993 Chromosome mapping of murine susceptibility loci to liver carcinogenesis. Cancer Res. 53: 209–211.[Abstract/Free Full Text]

GRASSO, P., and J. HARDY, 1975 Strain difference in natural incidence and response to carcinogens, pp. 111–131 in Mouse Hepatic Neoplasia, edited by W. H. BUTLER and P. M. NEWBERNE. Elsevier, Amsterdam.

GUAN, X. Y., Y. FANG, J. S. SHAM, D. L. KWONG, Y. ZHANG et al., 2000 Recurrent chromosome alterations in hepatocellular carcinoma detected by comparative genomic hybridization. Genes Chromosomes Cancer 29: 110–116.[CrossRef][Medline]

HANIGAN, M. H., C. J. KEMP, J. J. GINSLER and N. R. DRINKWATER, 1988 Rapid growth of preneoplastic lesions in hepatocarcinogen-sensitive C3H/HeJ male mice relative to C57BL/6J male mice. Carcinogenesis 9: 885–890.[Abstract/Free Full Text]

HANIGAN, M. H., M. L. WINKLER and N. R. DRINKWATER, 1990 Partial hepatectomy is a promoter of hepatocarcinogenesis in C57BL/6J male mice but not in C3H/HeJ male mice. Carcinogenesis 11: 589–594.[Abstract/Free Full Text]

HISLOP, R. G., N. PRATT, S. C. STOCKS, C. M. STEEL, M. SALES et al., 2002 Karyotypic aberrations of chromosomes 16 and 17 are related to survival in patients with breast cancer. Br. J. Surg. 89: 1581–1586.[CrossRef][Medline]

HOOVER, K. L., C. L. HYDE, M. L. WENK and L. A. POIRIER, 1986 Ethionine carcinogenesis in CD-1, BALB/c and C3H mice. Carcinogenesis 7: 1143–1148.[Abstract/Free Full Text]

KAKIZOE, S., S. GOLDFARB and T. D. PUGH, 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: 3985–3989.[Abstract/Free Full Text]

KEMP, C. J., and N. R. DRINKWATER, 1989 Genetic variation in liver tumor susceptibility, plasma testosterone levels, and androgen receptor binding in six inbred strains of mice. Cancer Res. 49: 5044–5047.[Abstract/Free Full Text]

KEMP, C. J., C. N. LEARY and N. R. DRINKWATER, 1989 Promotion of murine hepatocarcinogenesis by testosterone is androgen receptor-dependent but not cell autonomous. Proc. Natl. Acad. Sci. USA 86: 7505–7509.[Abstract/Free Full Text]

KRUGLYAK, L., and E. S. LANDER, 1995 A nonparametric approach for mapping quantitative trait loci. Genetics 139: 1421–1428.[Abstract]

LEE, G. H., and N. R. DRINKWATER, 1995a Hepatocarcinogenesis in BXH recombinant inbred strains of mice: analysis of diverse phenotypic effects of the hepatocarcinogen sensitivity loci. Mol. Carcinogen. 14: 190–197.[Medline]

LEE, G. H., and N. R. DRINKWATER, 1995b The Hcr (Hepatocarcinogen Resistance) loci of DBA/2J mice partially suppress phenotypic expression of the Hcs (Hepatocarcinogen Sensitivity) loci of C3H/HeJ mice. Carcinogenesis 16: 1993–1996.[Abstract/Free Full Text]

LEE, G. H., K. NOMURA, H. KANDA, M. KUSAKABE, A. YOSHIKI et al., 1991 Strain specific sensitivity to diethylnitrosamine-induced carcinogenesis is maintained in hepatocytes of C3H/HeN in equilibrium with C57BL/6N chimeric mice. Cancer Res. 51: 3257–3260.[Abstract/Free Full Text]

LEE, G. H., L. M. BENNETT, R. A. CARABEO and N. R. DRINKWATER, 1995 Identification of hepatocarcinogen-resistance genes in DBA/2 mice. Genetics 139: 387–395.[Abstract]

LEHMAN, E. L., 1998 Nonparametrics: Statistical Methods Based on Ranks. Prentice-Hall, Upper Saddle River, NJ.

LIN, Y. W., J. C. SHEU, G. T. HUANG, H. S. LEE, C. H. CHEN et al., 1999 Chromosomal abnormality in hepatocellular carcinoma by comparative genomic hybridisation in Taiwan. Eur. J. Cancer 35: 652–658.

LYSTIG, T. C, 2003 Adjusted P values for genome-wide scans. Genetics 164: 1683–1687.[Abstract/Free Full Text]

MALKINSON, A. M., and D. S. BEER, 1983 Major effect on susceptibility to urethane-induced pulmonary adenoma by a single gene in BALB/cBy mice. J. Natl. Cancer Inst. 70: 931–936.

MANENTI, G., G. BINELLI, M. GARIBOLDI, F. CANZIAN, L. DE GREGORIO et al., 1994 Multiple loci affect genetic predisposition to hepatocarcinogenesis in mice. Genomics 23: 118–124.[CrossRef][Medline]

MARCHIO, A., P. PINEAU, M. MEDDEB, B. TERRIS, P. TIOLLAIS et al., 2000 Distinct chromosomal abnormality pattern in primary liver cancer of non-B, non-C patients. Oncogene 19: 3733–3738.[CrossRef][Medline]

MCALEER, M. A., T. J. AITMAN, R. J. CORNALL, S. GHOSH, J. R. HALL et al., 1992 Linkage analysis of 84 microsatellite markers in intra- and interspecific backcrosses. Mamm. Genome 3: 457–460.[CrossRef][Medline]

NAGASAKI, H., H. KAWABATA, Y. MIYATA, K. INOUE, K. HIRAO et al., 1975 Effect of various factors on induction of liver tumors in animals by the alpha-isomer of benzene hexachloride. Gann 66: 185–191.[Medline]

POOLE, T. M., and N. R. DRINKWATER, 1996 Two genes abrogate the inhibition of murine hepatocarcinogenesis by ovarian hormones. Proc. Natl. Acad. Sci. USA 93: 5848–5853.[Abstract/Free Full Text]

PUGH, T. D., and S. GOLDFARB, 1992 Growth kinetics of microscopic hepatocellular neoplasms in carcinogen-resistant and carcinogen-responsive strains of mice. Cancer Res. 52: 280–284.[Abstract/Free Full Text]

SHAM, J. S., T. C. TANG, Y. FANG, L. SUN, L. X. QIN et al., 2002 Recurrent chromosome alterations in primary ovarian carcinoma in Chinese women. Cancer Genet. Cytogenet. 133: 39–44.[CrossRef][Medline]

SMITH, G. S., and R. L. WALFORD, 1978 Influence of the H-2 and H-1 histocompatibility systems upon life span and spontaneous cancer incidences in congenic mice. Birth Defects 14: 281–312.

SMITH, G. S., R. L. WALFORD and M. R. MICKEY, 1973 Lifespan and incidence of cancer and other diseases in selected long-lived inbred mice and their F₁ hybrids. J. Natl. Cancer Inst. 50: 1195–1213.

STORER, J. B., 1966 Longevity and gross pathology at death in 22 inbred mouse strains. J. Gerontol. 21: 404–409.

TORNILLO, L., V. CARAFA, J. RICHTER, G. SAUTER, H. MOCH et al., 2000 Marked genetic similarities between hepatitis B virus-positive and hepatitis C virus-positive hepatocellular carcinomas. J. Pathol. 192: 307–312.[CrossRef][Medline]

TURUSOV, V. S., N. S. LANKO, V. A. KRUTOVSKIKH and Y. D. PARFENOV, 1982 Strain differences in susceptibility of female mice to 1,2-dimethylhydrazine. Carcinogenesis 3: 603–608.[Abstract/Free Full Text]

VESSELINOVITCH, S. D., 1990 Perinatal mouse liver carcinogenesis as a sensitive carcinogenesis model and the role of the sex hormonal environment in tumor development. Prog. Clin. Biol. Res. 331: 53–68.[Medline]

VESSELINOVITCH, S. D., and N. MIHAILOVICH, 1967 The effect of gonadectomy on the development of hepatomas induced by urethan. Cancer Res. 27: 1788–1791.[Abstract/Free Full Text]

WISEMAN, R. W., S. J. STOWERS, E. C. MILLER, M. W. ANDERSON and J. A. MILLER, 1986 Activating mutations of the c-Ha-ras proto-oncogene in chemically induced hepatomas of the male B6C3F₁ mouse. Proc. Natl. Acad. Sci. USA 83: 5825–5829.[Abstract/Free Full Text]

WONG, N., P. LAI, S. W. LEE, S. FAN, E. PANG et al., 1999 Assessment of genetic changes in hepatocellular carcinoma by comparative genomic hybridization analysis: relationship to disease stage, tumor size, and cirrhosis. Am. J. Pathol. 154: 37–43.[Abstract/Free Full Text]

WONG, N., P. LAI, E. PANG, L. F. FUNG, Z. SHENG et al., 2000 Genomic aberrations in human hepatocellular carcinomas of differing etiologies. Clin. Cancer Res. 6: 4000–4009.[Abstract/Free Full Text]

WONG, N., W. C. LAM, P. B. LAI, E. PANG, W. Y. LAU et al., 2001 Hypomethylation of chromosome 1 heterochromatin DNA correlates with q-arm copy gain in human hepatocellular carcinoma. Am. J. Pathol. 159: 465–471.[Abstract/Free Full Text]

ZONDERVAN, P. E., J. WINK, J. C. ALERS, J. N. IJZERMANS, S. W. SCHALM et al., 2000 Molecular cytogenetic evaluation of virus-associated and non-viral hepatocellular carcinoma: analysis of 26 carcinomas and 12 concurrent dysplasias. J. Pathol. 192: 207–215.[CrossRef][Medline]

This article has been cited by other articles:

				F. Feo, M. Frau, M. L. Tomasi, S. Brozzetti, and R. M. Pascale Genetic and Epigenetic Control of Molecular Alterations in Hepatocellular Carcinoma Experimental Biology and Medicine, July 1, 2009; 234(7): 726 - 736. [Abstract] [Full Text] [PDF]

				S. E.-M. Peychal, A. Bilger, H. C. Pitot, and N. R. Drinkwater Predominant modifier of extreme liver cancer susceptibility in C57BR/cdJ female mice localized to 6 Mb on chromosome 17 Carcinogenesis, May 1, 2009; 30(5): 879 - 885. [Abstract] [Full Text] [PDF]

				H. Wang, D. Teske, A. Tess, R. Kohlhepp, Y. Choi, C. Kendziorski, and A. R. Moser Identification of Novel Modifier Loci of ApcMin Affecting Mammary Tumor Development Cancer Res., December 1, 2007; 67(23): 11226 - 11233. [Abstract] [Full Text] [PDF]

				M. R. De Miglio, P. Virdis, D. F. Calvisi, M. Frau, M. R. Muroni, M. M. Simile, L. Daino, G. M. Careddu, E. Sanna-Passino, R. M. Pascale, et al. Mapping a Sex Hormone-Sensitive Gene Determining Female Resistance to Liver Carcinogenesis in a Congenic F344.BN-Hcs4 Rat Cancer Res., November 1, 2006; 66(21): 10384 - 10390. [Abstract] [Full Text] [PDF]

				T. L. Thomae, E. A. Stevens, A. L. Liss, N. R. Drinkwater, and C. A. Bradfield The Teratogenic Sensitivity to 2,3,7,8-Tetrachlorodibenzo-p-dioxin Is Modified by a Locus on Mouse Chromosome 3 Mol. Pharmacol., March 1, 2006; 69(3): 770 - 775. [Abstract] [Full Text] [PDF]

International Access Link

Online ISSN: 1943-2361  Print ISSN: 0016-6731
Copyright 2009 by the Genetics Society of America
phone: 412-268-1812   fax: 412-268-1813   email: genetics-gsa{at}andrew.cmu.edu