Interacting QTLs for cholesterol gallstones and gallbladder mucin in AKR and SWR strains of mice
Henning Wittenburg1,2,
Frank Lammert2,
David Q.-H. Wang2,
Gary A. Churchill1,
Renhua Li1,
Guylaine Bouchard2,
Martin C. Carey2 and
Beverly Paigen1
1 Jackson Laboratory, Bar Harbor, Maine 04609
2 Department of Medicine, Harvard Medical School, Gastroenterology Division, Brigham and Womens Hospital, and Harvard Digestive Diseases Center, Boston, Massachusetts 02115
 |
ABSTRACT
|
---|
We employed quantitative trait locus (QTL) mapping in a backcross between gallstone-susceptible SWR/J and gallstone-resistant AKR/J inbred mice to identify additional susceptibility loci for cholesterol gallstone formation. After 12 wk of feeding the mice a lithogenic diet, we phenotyped 330 backcross progeny for gallstones, gallbladder mucin accumulation, liver weight, and body weight. Marker-based regression analysis revealed significant single QTLs associated with gallstone formation on chromosome 9 and the liver weight/body weight ratio on chromosomes 5 and X. A search for gene pairs detected significant gene-gene interactions for mucin accumulation between loci on chromosomes 5 and 11 and suggestive gene-gene interactions linked to gallstone formation between the QTL on chromosome 9 and loci on chromosomes 6 and 15. These findings uncover new QTLs for cholesterol gallstones, reveal independent loci for mucin accumulation, and demonstrate the importance of considering gene-gene interactions in cholesterol cholelithiasis. According to standard nomenclature, the gallstone QTL on chromosome 9 is named Lith5.
genetics; inbred mouse strains; gene-gene interaction; obesity
 |
INTRODUCTION
|
---|
CHOLESTEROL GALLSTONES are extraordinarily common among individuals in Western countries (35) and cause frequent pain and complications that often require surgical removal of the gallbladder. Therefore, cholesterol gallstones are not only a clinical, but also a major public health problem (10). To find new means of prevention and nonsurgical management of this prevalent disorder, a better understanding of the true etiology and pathophysiology of cholesterol gallstone formation as well as detection of the underlying genetic defects are essential.
Epidemiological studies together with ethnic differences reveal a hereditary background for gallstones in many humans (summarized in Ref. 35). Apart from very rare mutations in single genes (31, 39), lithogenicity of bile is caused most likely by a complex genetic predisposition plus a "cholelithogenic" environment such as diet, pregnancies, certain medications, obesity, and weight loss (35). The genetic era in cholesterol gallstone research began with the detection of the first cholesterol gallstone (Lith) genes by quantitative trait loci (QTL) mapping in crosses between gallstone-susceptible and gallstone-resistant inbred strains of mice (16, 28, 36). QTL mapping led to the identification of a number of genetic loci for this trait in mice, and some of them colocalize with putative candidate genes. Inferences on the latter are based on our current understanding of the pathophysiology of cholesterol gallstone formation (18, 35). The major murine cholesterol gallstone QTLs determined so far are Lith1 (on chromosome 2), Lith2 (on chromosome 19), Lith3 (on chromosome 17), and Lith4 (on chromosome X) (18). Since the detection of the first Lith genes, knowledge of the mechanisms of cholesterol gallstone formation has been challenged significantly by the rigorous phenotypic characterization of the gallstone-susceptible C57L/J and the gallstone-resistant AKR/J inbred mouse strains together with their F1 progeny (11, 22, 48, 49).
Based on the concept that only a critical subset of genes is rate limiting in controlling a complex trait, we have predicted that murine Lith genes are concordant with putative orthologous human LITH genes (18), in a fashion corresponding to what has been shown recently (44) for genetic loci influencing salt (NaCl)-induced hypertension. Because of the high degree of conservation between mouse and human genomes (29), the detection of murine cholesterol gallstone genes may enable the identification of orthologous human LITH genes and permit directed studies of these genes. However, the total number of Lith loci in the mouse is still unknown, and the genes responsible for the QTLs that have been detected await definitive identification. In addition, no prior QTL analysis has examined gene-gene interactions that might affect cholesterol gallstone formation, and it remains unclear whether independent genetic loci act upon intermediate steps in cholesterol gallstone pathogenesis such as gallbladder mucin accumulation, which is apparently required as a nucleation matrix and annealing agent for cholesterol monohydrate crystals in cholesterol supersaturated bile (25, 26). Therefore, additional QTL analyses are warranted to find the entire ensemble of single and interacting genetic loci that are involved in cholesterol gallstone formation in inbred mice and to confirm Lith loci from previous QTL crosses (36).
Here, we report the mapping of QTLs for cholesterol gallstones, gallbladder mucin accumulation, liver weight, and body weight in a backcross between gallstone-susceptible SWR/J and gallstone-resistant AKR/J inbred mice, a model that was used previously to identify genetic loci associated with obesity (51). Our results provide new insights into the genetic mechanisms underlying cholesterol gallstone formation in inbred mice by 1) identifying new Lith loci that colocalize with obesity QTLs from previous crosses of the same strains; 2) uncovering gene-gene interactions; and 3) detecting genetic loci associated with mucin accumulation that are independent of gallstone QTLs.
 |
MATERIALS AND METHODS
|
---|
Mice and diets.
Mice of inbred strains AKR/J and SWR/J were obtained from The Jackson Laboratory. We bred (SWR x AKR) F1 progeny and both F1 x AKR and AKR x F1 backcross progeny (female F1 progeny backcrossed to male AKR and vice versa, respectively) in our colony. All animals were maintained in a temperature-controlled room (2223°C) with alternating 12:12-h light/dark cycles. The animals were fed low-cholesterol (<0.02%) laboratory chow (Purina 5001; PMI Feeds, Richmond, IN) until 68 wk of age followed by a lithogenic diet containing 15% butterfat, 1% cholesterol, 0.5% cholic acid, 2% corn oil, 50% sucrose, 20% casein, and essential vitamins and minerals (16) for 12 wk and had free access to food and water. After fasting overnight with free access to water, the animals were anesthetized by intraperitoneal injection of 35 mg/kg body wt of pentobarbital (Abbott Laboratories, North Chicago, IL) and then killed by cervical dislocation. All animal protocols were approved by the Institutional Animal Care and Use Committees of The Jackson Laboratory and Harvard University.
Phenotyping.
Mice were phenotyped over a period of 11 mo essentially as described previously (16, 49). In brief, following cholecystectomy, gallbladder volume was measured gravimetrically. The gallbladder was cut at the fundus to collect and count stones. The bile was characterized by one of two authors who examined for mucin gel, cholesterol monohydrate crystals, and sandy and true stones under a polarizing light microscope using a modified scoring system (49) based on microscopic appearances of bile as proposed previously (1). After drying overnight at room temperature, stones were weighed. Additionally, we determined liver weight and body weight of the animals. The amount of gallbladder mucin was semiquantitatively scored microscopically as follows: 0, indicating no visible mucin; 1, a small amount of mucin gel in part of the gallbladder; 2, less than 50% of the gallbladder filled with mucin gel; 3, more than 50% of the gallbladder filled with mucin gel; 4, gallbladder completely filled with mucin gel ("mucin score"). Mucin was verified by periodic acid/Schiff (PAS) staining (49). To characterize the backcross progeny for cholesterol crystallization sequences (49) and formation of gallstones, we used a score for absence (score = 0) or presence (score = 1) of cholesterol monohydrate crystals and presence of stones (score = 2) ("gallstone score").
Genotyping.
Genomic DNA was prepared from tails or spleens of mice by standard phenol:chloroform extraction (46). For genotyping individual animals we employed the polymerase chain reaction (PCR) (45) of microsatellite markers (simple sequence length polymorphisms) that discriminate between AKR and SWR alleles. Primer pairs were obtained from Research Genetics, Huntsville, AL. A list of the markers is available upon request from the corresponding author. All reported map positions were retrieved from the 2001 Mouse Genome Database (http://www.informatics.jax.org).
Mapping panels.
Because we were unable to identify polymorphisms between strains AKR and SWR for the hepatic lipase (Lipc) and the sterol regulatory element binding protein (SREBP) cleavage-activating protein (SCAP) genes (Scap), these candidate genes were mapped by linkage analysis using the (C57BL/6JEi x SPRET/Ei) F1 x SPRET/Ei (BSS) panel form the Jackson Laboratory (38). After standard PCR (denaturation time 5 min at 95°C, 35 cycles with 30 s at 94°C, 30 s at 55°C, and 1 min at 72°C, followed by a final extension at 72°C for 7 min), allele detection was performed by 4% agarose gel electrophoresis (Lipc) or single-strand conformation polymorphism (SSCP) analysis (Scap) as described previously (19). Primer pairs with the sequences 5' CAGCTTTATCCTTGACTGGTTTTAG 3' (forward primer) and 5' CCCTTCAAAGGTGAGTGTCTC 3' (reverse primer) for Lipc and 5' TTCTGCCTCTTTGCTGTTGTG 3' (forward primer) and 5' CTCCCATGTCTGAAGAGAGC 3' (reverse primer) for Scap identified polymorphisms in intronic sequences of the genes between C57BL/6JEi and SPRET/Ei. Primer sequences for Scap were described previously to amplify 96 bp of Scap cDNA from codon 417 to 448 and 66 bp of intronic sequence (14). Because this fragment contains part of the sterol-sensing domain of SCAP (7) and codon 443 appears to have an important function for the feedback inhibition of SCAP activity (14), we employed the same primer pairs for direct sequencing of the amplified region between mouse strains AKR and SWR. PCR products were extracted using the QIAquick Gel Extraction Kit (Qiagen, Santa Clarita, CA). Intronic sequence of Lipc for primer design was retrieved from the Celera Mouse Genome Database (fragment accession number GA_72972673). Identity of the PCR product was confirmed by direct sequencing (data not shown).
Sequencing of the HMG-CoA reductase (Hmgcr) gene promoter.
Since SWR and AKR mice differ in HMG-CoA reductase (HMGCR) activity (22), and our initial analysis revealed an apparent QTL colocalizing with the Hmgcr gene, we sequenced the promoter region of this rate-limiting enzyme of cholesterol synthesis in the parental strains of our QTL cross. The 5' flanking region of Hmgcr was amplified from DNA of mice of strains SWR and AKR by standard PCR (45), using primers reported previously to cover the minimal promoter region that contains major regulatory elements (15). We extracted the PCR product as described above for direct sequencing with primers 5' TGGGCTTTTTCTTTTCAACC 3' (forward) and 5' CTCACCTCCGGATCTCAATG 3' (reverse).
Statistical analysis.
Results are expressed as means ± SD. We analyzed the phenotypic data by using one-way ANOVA to determine the effect of continuous values on the categorical data for gallstone score. Correlation analysis between continuous values was performed by calculating Pearsons correlation coefficient. We assessed differences among strains by chi-square test. P < 0.05 was considered statistically significant.
A multiple imputation algorithm (42) was applied to account for missing marker genotypes. Results reported here represent an average of 256 imputations. Significant loci displayed F statistics exceeding the 95th percentile of the permutation distribution. Suggestive loci exceeded the 63rd percentile and may be expected to occur once in a genome-wide scan of randomized data (23).
To identify single and interacting QTLs and to determine the combined contributions of all QTLs to the phenotypic variation, we used a three-stage statistical analysis developed recently (42, 44). In the first stage, we tested each marker for association with the traits by marker-based regression analysis and performed interval mapping using the pseudo-marker algorithm (42) employing an one-way ANOVA F statistic. Thresholds for significance were determined by permutation analysis (8). The second stage was a simultaneous search for gene pairs. First, we screened all pairs of loci using a two-way ANOVA to compare the full two-locus model with interactions to the null hypothesis of no genetic effect at either locus ("Fall" statistic). Because of the large number of locus pairs tested, significance thresholds for the Fall statistic were determined by permutation analysis (8). For those marker pairs that passed this screen, we computed a second F statistic ("Fint") to compare two models, with one assuming gene-gene interactions, the other assuming two loci with additive effects. In case of significance for both the Fall and the Fint statistics, we concluded that an interacting QTL pair affects the phenotype. The third stage of analysis integrated the single and interacting QTLs from the first two stages by multiple regression analysis. The corrected sums of squares were used to determine the contribution of each QTL in combination with all other QTLs (for details see Ref. 44; software available at http://www.jax.org/research/churchill).
 |
RESULTS
|
---|
Phenotypes of parental strains and F1 progeny.
Table 1 summarizes the phenotypes for SWR and AKR parental strain mice as well as their reciprocal F1 progeny, i.e., SWR females crossed to AKR males [(SWR x AKR) F1] and AKR females crossed to SWR males [(AKR x SWR) F1]. The parental strains for this QTL cross were chosen based on a previous small strain survey of female mice fed the lithogenic diet for 18 wk; in that experiment, strain SWR was gallstone susceptible, and strain AKR was gallstone resistant (16). Here, we confirm that SWR females display a greater gallstone prevalence rate than AKR males (Table 1). However, AKR females show the same gallstone prevalence rate as SWR males (Table 1). The prevalence of gallstones was higher in AKR x SWR F1 than in reciprocal F1 (Table 1). The reciprocal F1 mice, AKR x SWR or SWR x AKR, have exactly the same genotypes except for mitochondrial DNA, the X and Y chromosomes, and any imprinted chromosomal regions. These non-Mendelian factors affecting gallstone formation might be interesting to pursue further, but we chose to avoid them by backcrossing only (SWR x AKR) F1 progeny to AKR mice.
Phenotypes of backcross progeny.
Both male and female (SWR x AKR) F1 mice were backcrossed to AKR, and the backcross progeny did not differ in any phenotype based on parental gender (data not shown). When we initiated backcrossing F1 offspring to AKR, both male and female backcross progeny were fed the lithogenic diet and phenotyped. Once we had recognized that male backcross offspring had a higher prevalence rate of cholesterol gallstones than females, only males were collected for the remainder of the experiment. However, even though females developed fewer gallstones than males, they affected the results in this QTL cross, and the 57 female offspring that had been phenotyped were included in the final analysis. Figure 1A shows the distribution of the scores for formation of cholesterol crystals and stones ("gallstone score") among the 330 backcross progeny, and Fig. 1B shows the distribution of mucin accumulation in the gallbladder among the backcross progeny ("mucin score"). Both traits varied widely among the backcross progeny, which differed genetically but shared the same environment, supporting genetic determinants of cholesterol gallstone formation and gallbladder mucin accumulation in this backcross.

View larger version (9K):
[in this window]
[in a new window]
|
Fig. 1. Distributions of the gallstone score (A) and the mucin score (B) among the 330 reciprocal backcross progeny. The scoring systems employed are described in MATERIALS AND METHODS.
|
|
At 12 wk on the lithogenic diet, mice weighed an average of 32.3 ± 5.3 g. Although we detected an association for the occurrence of stones (score = 2) with higher body weight (F = 4.3, P = 0.015), this parameter did not correlate significantly with either stone weight or stone number. Mice with gallstones were characterized by an average of 10.5 ± 21.5 stones per gallbladder. The total stone weight per mouse with gallstones averaged 1.4 ± 1.2 mg and gallbladder volumes of all animals averaged 12 ± 6.9 µl. Higher gallbladder volumes correlated positively (r = 0.24, P = 0.009) with stone number, but not with stone weight or gallstone score. Higher mucin scores correlated positively with gallstone scores (F = 21.4, P < 0.001), but not with either gallstone weight or number. Taking all mice, the liver weights averaged 2.0 ± 0.4 g, which after being normalized for body weights, did not correlate significantly with either gallstone weight, gallstone number, or gallstone score. The lack of correlation between liver weight and gallstone formation is reassuring that diet-induced liver toxicity did not confound the genetic analysis of gallstone development in this cross, a possible concern because mice of the gallstone-susceptible strain SWR develop higher weight livers when subjected to lithogenic diet feeding (Table 1). Despite the enrichment of the diet with 0.5% cholic acid, we did not observe diarrhea in any of the animals.
QTL analysis.
In a QTL analysis, most linkage information is contained in those animals that fall into the extremes of the phenotype distribution of the trait of interest. This leads to the concept of "selective genotyping," a method to economize on the number of progeny to be genotyped (24). Therefore, to identify putative QTLs, we genotyped those backcross progeny that represented the 40% of animals with the largest deviation from the mean to the upper end and to the lower end of the distribution of each trait (gallstone weight, gallstone score, mucin accumulation, gallbladder volume, and body weight). Since different animals fall into the two categories for the different traits, a total of 224 mice (68% of the backcross progeny) were genotyped. However, a preliminary analysis was performed after the genotyping of the upper and lower 5% of the phenotype distribution for gallstone weight (see DISCUSSION). The initial genome-wide screen was carried out with 92 microsatellite markers that were spaced at about 20 cM intervals throughout the genome. After the discovery of the QTLs, eight markers were added on chromosomes 5 and 9 for higher resolution of the regions.
Figure 2 shows results for the genome-wide scans by interval mapping. We detected a significant single QTL on chromosome 9 for the gallstone score. Significant F statistics after permutation testing were F > 11.0 (P < 0.05), and the maximum F value was obtained for marker D9Mit307 (as defined by marker regression analysis) at cM 43 (F = 12.2). Figure 3 shows the average gallstone score phenotype for animals with homozygous (AK/AK) and heterozygous (AK/SW) alleles at D9Mit307. Heterozygosity at D9Mit307 was associated with higher gallstone scores, indicating that the SWR allele increased gallstone formation. Our analysis revealed four suggestive single QTLs for body weight on chromosomes 1, 2, 6, and 8 with F statistics ranging from 9.3 to 9.7 ("suggestive" threshold as determined by permutation testing was F > 6.4). For relative liver weight (corrected for body weight), two significant single QTLs (significance threshold F > 10.8) were detected on chromosome 5 [F = 11.1, D5Mit100 (cM 82)] and on chromosome X (F = 11.4, DXMit55, cM 1). Another suggestive single QTL associated with mucin accumulation in the gallbladder was detected on chromosome 11 [F = 9.7, D11Mit42 (cM 72)].

View larger version (35K):
[in this window]
[in a new window]
|
Fig. 2. Interval mapping genome scans to detect single quantitative trait loci (QTLs) for gallstone score, mucin accumulation in the gallbladder, liver weight, and body weight in genotyped backcross progeny. The genome-wide scan was calculated by one-way ANOVA F statistics. The abscissa indicates chromosomal locations, and the horizontal line indicates genome-wide significant (P < 0.05) or suggestive (P < 0.37) thresholds.
|
|

View larger version (13K):
[in this window]
[in a new window]
|
Fig. 3. Effects of homozygous and heterozygous alleles at D9Mit307 on gallstone score for genotyped backcross progeny. The abscissa shows homozygous (AK/AK) or heterozygous (AK/SW) state at the indicated locus, and the ordinate shows gallstone score. Diamonds indicate the mean phenotypic effects for all animals that are homozygous or heterozygous at the locus. Standard error intervals are indicated by "+".
|
|
The search for interacting QTLs revealed two gene pairs affecting gallstone formation, one member of both pairs being a locus in the central portion of chromosome 9 and the second member loci on chromosomes 6 and 15. The interacting locus on chromosome 9 was defined by marker D9Mit11 (cM 48), mapping only 5 cM distal to marker D9Mit307, which defines the peak of the QTL in the search for single loci. The proximity of the peaks of the loci from the searches for single and interacting QTLs indicates the likelihood of just one QTL on chromosome 9 conferring both the single gene and gene-gene interaction effects. Table 2 shows the effects of allelic substitutions at the interacting loci: the combination of heterozygosity (AK/SW) at D9Mit11 with homozygosity (AK/AK) at D6Mit14 (cM 71; Table 2, top) and heterozygosity at D15Mit26 (cM 29; Table 2, bottom), respectively, is associated with a substantial increase of the gallstone score. As determined by permutation testing, F statistics for marker pairs D6Mit14*D9Mit11 (Fall = 7.4, Fint = 8.8), and D9Mit11*D15Mit26 (Fall = 7.9, Fint = 10.3) fell just below the significance threshold [Fall > 8.5 (P < 0.05)] and therefore had "suggestive" significance. Figure 4 displays that our analysis revealed an interacting gene pair for gallbladder mucin gel accumulation between chromosome 5 and chromosome 11. The significance threshold after permutation testing was Fall > 8.9 (P < 0.05). The results of the F statistics for the marker pair D5Mit100*D11Mit74 for the association with mucin accumulation were Fall = 10.6, and Fint = 31.9, respectively. The positions of these interacting QTLs are shown in Fig. 5. The location of the interacting QTL on chromosome 11 at the proximal end of the chromosome differed from the single QTL at the distal end, indicating the likelihood of two separate QTLs. For this chromosome pair, the combination of homozygosity for AKR alleles at both loci had the largest effect on mucin accumulation (Table 3). The analysis revealed additional complexity with indications for interacting gene pairs between chromosomes 4 and 9 and chromosomes 7 and 12; however, these results were less clear (Fig. 4).

View larger version (123K):
[in this window]
[in a new window]
|
Fig. 4. Pairwise genome scans for the detection of gene-gene interactions affecting mucin accumulation among the 222 genotyped backcross progeny. The boxes ascending the diagonal indicate chromosomes 1 through X and the markers genotyped on the chromosomes, with the sizes of the boxes representing the number of markers for each chromosome. The vertical spectrum band to the right indicates the negative log (P value) for F tests. The negative log (P values) for the Fall statistics are visible below the main diagonal; the orange area shows a significant QTL pair on chromosomes 5 and 11 affecting mucin accumulation (circle). The negative log (P values) of the Fint statistics are shown above the main diagonal, and the red area (circle) indicates that this QTL pair is interacting.
|
|

View larger version (15K):
[in this window]
[in a new window]
|
Fig. 5. Confidence intervals (CI) for gene interactions between chromosomes 5 (ordinate) and 11 (abscissa) affecting the mucin score [calculated by the method of Sen and Churchill (42)]. Black areas indicate 95% CI, and the gray areas indicate 99% CI for interacting regions. At the right of the plot, the gray area represents the single QTL at the distal end of chromosome 11.
|
|
We used a joint statistical model to assess the contribution of the single and interacting QTLs detected in the first two models in combination with all other QTLs for the gallstone score (Table 4) and for gallbladder mucin accumulation (Table 5). The interacting loci on chromosome 6 and 15 display no significance on their own; however, their interactions with the QTL on chromosome 9 made a notable effect on the gallstone score (Table 4). The single QTL on chromosome 9 alone explained only 5.5% of the total variance for the gallstone score, but this effect, together with the interactions with the loci on chromosome 6 and chromosome 15, explained 11.2% of the total variance. The additional variance explained in the model including the gene-gene interactions, confirms the importance of gene-gene interactions for gallstone formation in this cross. The analysis for mucin accumulation showed a much larger effect for the interactions between the QTLs than for each QTL alone (Table 5). The interacting loci together with the single locus on chromosome 11 explained 16.3% of the total variance for gallbladder mucin accumulation.
View this table:
[in this window]
[in a new window]
|
Table 4. Multiple regression model for loci affecting the gallstone score among all 222 animals genotyped from the reciprocal backcross
|
|
View this table:
[in this window]
[in a new window]
|
Table 5. Multiple regression model for loci affecting gallbladder mucin accumulation among all 222 animals genotyped from the reciprocal SWR x AKR backcross
|
|
Mapping of Scap and Lipc.
We identified possible candidate genes for the QTLs by exploring the Mouse Genome Database and the human/mouse homology database (http://www.ncbi.nlm.nih.gov/HomoloGene/). To determine the map positions of the gene for the SREBP cleavage-activating protein (Scap) and of the gene for hepatic lipase (Lipc) relative to the gallstone QTL on chromosome 9, we mapped both employing The Jackson Laboratorys BSS panel. Scap, which had not previously been assigned to a mouse chromosome, was mapped to mouse chromosome 9; centromere, Uqcrc1, 1.06 ± 1.06 cM [95% confidence interval (CI) 03.8 cM], [Scap, Mtap4], 1.06 ± 1.06 cM (03.8 cM), D9Ertd241e. No recombinants were detected between Mtap4 and Scap, suggesting that these two loci are within 3.8 cM of each other (upper 95% confidence limit). The mapping of Scap to mouse chromosome 9 is consistent with the mapping of human SCAP to the homologous region 3p21.3 by in situ hybridization (33). Sequencing of the region that had been shown to be polymorphic between C57BL/6 and SPRETUS and had been used for mapping of Scap revealed no polymorphisms in cDNA or intronic sequence between the parental strains of our cross, AKR and SWR.
Lipc, previously mapped to chromosome 9, was localized to the central portion of the chromosome; centromere, Tpm1, 2.13 ± 1.49 cM (0.37.5 cM), [Lipc, Adam10], 1.06 ± 1.06 cM (03.8 cM), Adam4. We found no recombinants between Adam10 and Lipc, again suggesting that these two loci are within 3.8 cM of each other (upper 95% confidence limit). The mapping data for Scap and Lipc have been deposited in the Mouse Genome Database (accession number J:72264).
Sequencing of the promoter region of Hmgcr.
Because the QTL for gallstone weight colocalized with the HMG-CoA reductase gene in a preliminary analysis (18), we sequenced the 5' flanking region of the gene in the parental strains. No sequence differences in the minimal promoter (-489 to transcription start site) were found between SWR and AKR.
 |
DISCUSSION
|
---|
The importance of interacting QTLs in complex genetic traits has been recognized for many years (13, 40), but only recently has quantification become more facile (42, 44). Gene-gene interactions may occur if 1) two proteins encoded by these genes interact physically with each other, such as a ligand and its receptor; 2) proteins function in the same biochemical or biophysical sequences; or 3) proteins with similar functions substitute for each other in a biological-chemical pathway. However, despite the well-recognized significance of gene-gene interactions, searches for them in QTL crosses have not been performed widely, because tools for their statistical analysis were unavailable. Therefore, no gene-gene interactions for Lith genes have been looked for or detected so far. Recently, we have developed a feasible simultaneous search based on a two-way ANOVA F statistic for detecting interacting gene pairs that are otherwise undetectable by conventional single locus analysis (42, 44). This method was used first for analysis of interacting QTL pairs in salt (NaCl)-induced hypertension (44). In the study reported here, we detected two genomic regions on chromosome 6 and 15 that affected the gallstone score by interacting with the gallstone QTL on chromosome 9. The interacting allele on chromosome 6 was conferred by the gallstone-susceptible SWR strain and the allele on chromosome 15 by the resistant AKR strain. This is explained by the 20% gallstone prevalence rate in male AKR mice (36), indicating that these mice also carry gallstone susceptibility alleles. Accordingly, the susceptible alleles of one of the QTLs in a previous cross, Lith3, was contributed by the AKR strain (20), whereas AKR carries gallstone resistance alleles at Lith1 and Lith2 (36). The Lith3 region did not show any association with gallstone formation in our cross, most likely because of the different genetic background from the second parental strain. Gallstone prevalence of (SWR x AKR) F1 progeny was identical to the prevalence of male AKR progeny (Table 1). However, the major gallstone QTL on chromosome 9 was conferred by strain SWR and was, hence, dominant or partially dominant. The observed gene-gene interactions and various segregating gallstone susceptibility alleles contributed by both parental strains can explain the puzzling inheritance of gallstones in this cross. All genomic regions associated with gallstone formation in this work with the exception of chromosome 6 are distinct from Lith genes in our previous crosses (18). The detection of interacting QTL pairs that affect cholesterol gallstone formation in mice emphasizes the importance of both considering gene-gene interactions in this trait as well as in the evaluation of candidate genes. The former is especially important for the construction of congenic strains and the latter for understanding of underlying pathophysiology.
The correlation of obesity and cholesterol gallstone prevalence in humans is well known (3). Analysis of the phenotypes of the backcross progeny in our SWR x AKR backcross indicated a weak association between gallstone formation and higher body weight; however, the QTLs for gallstone formation and body weight did not colocalize in the same genomic regions, with the possible exception for chromosome 6. Interestingly, previous linkage analyses in a AKR/J x SWR/J intercross and in a F1 x SWR backcross (51, 52) revealed QTLs for obesity on chromosome 4, chromosome 9, and chromosome 15 named dietary obesity 1 (Dob1), dietary obesity 2 (Dob2), and dietary obesity 3 (Dob3), respectively. Two of these QTLs, Dob2 and Dob3, overlap with QTLs for gallstone formation in our cross. Obesity at Dob2 was associated with a recessive SWR allele in the previous study; however, since our backcross to AKR did not allow for homozygous SWR alleles, it is not surprising that we failed to find Dob2 in our cross. The interacting gallstone QTL on chromosome 15 and Dob3 were both contributed by the AKR strain. We did not detect Dob1 and Dob3 and can speculate on several reasons for this: 1) we employed a different phenotype, body weight, vs. adipose depot weight; 2) we employed 330 mice compared with 931 mice in the previous intercross; and 3) we backcrossed F1 progeny to AKR rather than SWR in the previous backcross. Colocalization of QTLs for the two traits indicates that some of the genes for obesity and gallstone formation are linked closely and might therefore be inherited together. This would explain further the recent observation from our laboratories that cholesterol cholelithiasis is not necessarily a consequence of obesity per se but depends upon the specific genetic background and ensuing pathophysiological pathways (5). Alternatively, some of the same genes could determine the risk for both the development of obesity and cholesterol gallstone formation. Both traits may also be determined by more than one gene in the same chromosomal region, not an unusual finding in complex traits (27, 32). The detection of the underlying genes may provide valuable insights into the genetic basis of cholesterol gallstone formation in obese humans.
The analysis of the multiple phenotypes in our work revealed a correlation for gallbladder volume and gallstone number. However, we detected no genomic regions associated with these two traits, suggesting that gallbladder volume and gallstone number are controlled indirectly and not by independent genes. After correction for body weight, animals on the lithogenic diet from strain SWR displayed higher liver weights than AKR mice. This difference was more pronounced in females than in males (Table 1). Previous experiments indicate that the liver weight/body weight ratio may be correlated with cholesterol and cholesteryl ester accumulation in livers (34), and this could be explained by different regulation of key enzymes in cholesterol and bile salt metabolism in strains SWR and AKR when challenged with the lithogenic diet (22). However, we have not measured cholesterol contents in the livers of parental or backcross progeny and can only speculate on candidate genes that encode lipid regulatory or metabolic proteins that may underlie the QTLs for liver weight/body weight ratios on chromosome 5 and chromosome X (Table 6).
Finally, we identified genetic loci for accumulation of mucin gel in the gallbladder. We found a positive correlation between mucin scores and gallstone formation; however, the QTLs for these two phenotypes did not overlap, suggesting independent genetic loci affecting the gallbladder mucin phenotype. Even though the QTLs for mucin accumulation and liver weight colocalize on chromosome 5, it seems unlikely that the same genes determine both QTLs, because the two traits are associated with different underlying biological processes. Gallbladder mucin hypersecretion is well known to precede cholesterol gallstone formation (26, 49) and is stimulated by several biliary factors including hydrophobic bile salts (17). Furthermore, a mucin gel layer in the gallbladder seems to be necessary for the nucleation of cholesterol monohydrate crystals in bile that is supersaturated with cholesterol (25). The positive correlation for mucin accumulation in the gallbladder and gallstone formation in our SWR x AKR cross is consistent with this pathophysiological principle and confirms the importance of a preformed mucin gel layer for facilitating cholesterol cholelithiasis. The interacting QTL on chromosome 5 colocalizes with the mucin gene Muc3. Muc3, the mouse ortholog of the major human biliary mucin gene (47), is also expressed in mouse gallbladder (43). Therefore, it is an attractive positional candidate gene for the QTL on chromosome 5, and polymorphisms in the Muc3 gene or its promoter may be linked to gallbladder mucin accumulation. This identification of genetic loci that affect gallbladder accumulation of mucin gel suggests that genes independent of Lith genes influence separate steps in gallstone formation such as lithogenic bile-induced mucin hypersecretion.
The method of QTL mapping reveals large chromosomal regions associated with traits, which contain many genes. Therefore, it is likely that one can identify a large number of possible candidate genes in each of these regions (Table 6). Pathophysiological hallmarks for gallstone susceptibility in the present model are higher de novo cholesterol synthesis (HMGCR activity) and esterification (acyl-CoA:cholesterol acyltransferase 2 activity) rates, together with lower bile salt synthesis rates (cholesterol 7
-hydroxylase and sterol 27-hydroxylase activities) in strain SWR compared with strain AKR when on the lithogenic diet (22). However, none of the QTLs in this cross colocalizes with any of these genes that encode regulatory enzymes in cholesterol metabolism and bile salt biosynthesis. As we suggested for the principal cholelithogenic mechanism in the gallstone-susceptible strain C57L (22), alterations in cholesterol metabolism may be secondary to ensure the supply of cholesterol molecules on demand for biliary cholesterol hypersecretion caused by Lith genes. One attractive positional candidate gene for the cholesterol gallstone QTL on chromosome 9 is the gene encoding hepatic lipase (Lipc), because this hydrolytic enzyme is involved in chylomicron remnant and high-density lipoprotein (HDL) cholesterol uptake by the liver (9, 18). Upon feeding a lithogenic diet, chylomicron remnants are apparently an important source for cholesterol hypersecretion into bile (2, 30). Macrophage-derived HDL was believed to be a major source for biliary cholesterol in the basal state (37, 41), but its role has been questioned recently by the finding that ABCA1 knockout mice display normal biliary cholesterol secretion rates (12). Lipc had been mapped with low resolution in a small backcross panel with a limited number of markers (50). We have redefined its map position within the 95% CI of the QTL on chromosome 9. The map position of the gene encoding the SCAP protein in the human genome (33) suggested a possible location on mouse chromosome 9. Using the BSS panel, we mapped Scap in the mouse, to determine its location relative to the gallstone QTLs. Scap serves as a sterol-sensor and regulates SREBP processing by allowing the site-1 protease to perform the first of two cleavage steps required to release SREBP from the endoplasmic reticulum and enable trafficking to the nucleus (7). SREBPs are transcription factors that mediate feedback regulation of cholesterol synthesis and uptake. Since Scap was found to map within the 95% CI of the QTL on chromosome 9 in our SWR x AKR backcross, its apparent physiological function renders it a very attractive positional candidate gene for the gallstone QTL on chromosome 9. The map positions in the mouse genome of the proteases performing the site-1 and site-2 cleavage of SCAP are currently not known, but it is of great interest if these potential gallstone candidate genes colocalize with Lith genes. The region that we employed for mapping Scap includes part of the sterol-sensing domain, and a point mutation within this region led to resistance of SCAP to suppression by sterols (14). However, even though we detected a polymorphism between C57BL/6 and SPRETUS employed for mapping of Scap, we identified no sequence differences in this region between AKR and SWR, the parental strains of our backcross. This is not surprising, since C57BL/6 and SPRETUS are members of different mouse species, whereas AKR and SWR are more closely related mouse strains (4).
In a preliminary linkage analysis of this SWR x AKR backcross, we genotyped the 5% extremes of the phenotypic distribution as initially proposed by Lander and Botstein (24), and detected two QTLs for gallstone weight on chromosome 13 and chromosome X both with suggestive logarithm of the odds (LOD) scores of 2.0 (18, 21). After increasing the number of genotyped backcross progeny, we were unable to confirm these apparent QTLs and found no association of any chromosomal region with gallstone weight. The most likely explanation for the loss of the QTLs on chromosomes 13 and X after genotyping more animals is that selecting only the extremes of the phenotypic distribution for genotyping overestimated the effects of allelic substitution at these loci and led to false-positive results (51). The colocalization of the QTL on chromosome 13 with the Hmgcr gene prompted us to sequence the minimal promoter region of Hmgcr in the parental strains. We detected no sequence differences between strain AKR and SWR, further supporting an indirect control of HMGCR expression (15).
Taking all findings together, this study advances our knowledge of the genetic background of murine cholesterol gallstone formation. We demonstrate the importance of gene-gene interactions in this complex trait, detect genetic loci for gallbladder mucin gel accumulation that are independent of Lith genes, and we uncover additional genetic loci associated with cholesterol cholelithiasis. According to standard nomenclature (18), the major gallstone QTL identified in this SWR x AKR cross on chromosome 9 is named Lith5. Because of their suggestive significance, assignment of names to the interacting QTLs for gallstone formation will depend on confirmation in ongoing and future crosses. These QTL crosses between mice of genetically and phenotypically diverse inbred strains (6) should enable us to determine how many loci contribute to the phenotypic variation and to find the entire ensemble of single and interacting QTLs for murine cholesterol gallstone formation. The identification of all QTLs affecting cholesterol gallstone formation positions us to take advantage of newly available technologies and database resources such as the whole murine genome sequence to identify the underlying genes and to apply this knowledge to directed studies of the genetic basis of cholesterol cholelithiasis in humans.
 |
ACKNOWLEDGMENTS
|
---|
We thank H. Nelson and K. L. Svenson for excellent technical assistance.
This work was supported in part by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-51568 (to B. Paigen), DK-36588, DK-52911, and DK-34854 (to M. C. Carey), DK-54012 (to D. Q.-H. Wang), and by a grant from the Ellison Medical Foundation (to D. Q.-H. Wang). F. Lammert (La 997/3-1) and H. Wittenburg (WI 1905/1-1) were supported by the Deutsche Forschungsgemeinschaft.
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: B. Paigen, The Jackson Laboratory, 600 Main St., Bar Harbor, ME 04609 (E-mail: bjp{at}jax.org).
10.1152/physiolgenomics.00097.2001.
 |
References
|
---|
-
Akiyoshi T, Uchida K, Takase H, Nomura Y, and Takeuchi N. Cholesterol gallstones in alloxan-diabetic mice. J Lipid Res 27: 915924, 1986.[Abstract]
-
Amigo L, Quinones V, Mardones P, Zanlungo S, Miquel JF, Nervi F, and Rigotti A. Impaired biliary cholesterol secretion and decreased gallstone formation in apolipoprotein E-deficient mice fed a high-cholesterol diet. Gastroenterology 118: 772779, 2000.[ISI][Medline]
-
Attili AF, Capocaccia R, Carulli N, Festi D, Roda E, Barbara L, Capocaccia L, Menotti A, Okolicsanyi L, Ricci G, Lalloni L, Mariotti S, Sama C, and Scafato E. Factors associated with gallstone disease in the MICOL experience. Multicenter Italian Study on Epidemiology of Cholelithiasis. Hepatology 26: 809818, 1997.[ISI][Medline]
-
Beck JA, Lloyd S, Hafezparast M, Lennon-Pierce M, Eppig JT, Festing MF, and Fisher EM. Genealogies of mouse inbred strains. Nat Genet 24: 2325, 2000.[ISI][Medline]
-
Bouchard G, Carver T, Tovbina M, Paigen B, and Carey MC. Genetic factors underlying obesity, but not obesity per se, determine cholesterol gallstone (ChGS) formation in inbred mice (Abstract). Hepatology 32: 391, 2000.
-
Bouchard G, Johnson D, Carver T, Carey MC, and Paigen B. Multiple new murine models of cholesterol gallstones (ChGS): the Jackson Laboratory-Brigham and Womens Hospital strain survey (Abstract). Gastroenterology 120: 72, 2001.
-
Brown MS and Goldstein JL. A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc Natl Acad Sci USA 96: 1104111048, 1999.[Abstract/Free Full Text]
-
Churchill GA and Doerge RW. Empirical threshold values for quantitative trait mapping. Genetics 138: 963971, 1994.[Abstract/Free Full Text]
-
Donner C, Choi S, Komaromy M, and Cooper AD. Accelerated lipoprotein uptake by transplantable hepatomas that express hepatic lipase. J Lipid Res 39: 18051815, 1998.[Abstract/Free Full Text]
-
Everhart JE. Gallstones. In: Digestive Diseases in the United States: Epidemiology and Impact, edited by Everhart JE (US Department of Health and Human Services, PHS NIDDK, NIH publication No 94-1447). US Government Printing Office, 1994, p. 647692.
-
Fuchs M, Ivandic B, Muller O, Schalla C, Scheibner J, Bartsch P, and Stange EF. Biliary cholesterol hypersecretion in gallstone-susceptible mice is associated with hepatic up-regulation of the high-density lipoprotein receptor SRBI. Hepatology 33: 14511459, 2001.[ISI][Medline]
-
Groen AK, Bloks VW, Bandsma RH, Ottenhoff R, Chimini G, and Kuipers F. Hepatobiliary cholesterol transport is not impaired in Abca1-null mice lacking HDL. J Clin Invest 108: 843850, 2001.[Abstract/Free Full Text]
-
Haley CS and Knott A. A simple regression method for mapping quantitative trait loci in line crosses using flanking markers. Heredity 69: 315324, 1992.[ISI]
-
Hua X, Nohturfft A, Goldstein JL, and Brown MS. Sterol resistance in CHO cells traced to point mutation in SREBP cleavage-activating protein. Cell 87: 415426, 1996.[ISI][Medline]
-
Hwa JJ, Zollman S, Warden CH, Taylor BA, Edwards PA, Fogelman AM, and Lusis AJ. Genetic and dietary interactions in the regulation of HMG-CoA reductase gene expression. J Lipid Res 33: 711725, 1992.[Abstract]
-
Khanuja B, Cheah YC, Hunt M, Nishina PM, Wang DQH, Chen HW, Billheimer JT, Carey MC, and Paigen B. Lith1, a major gene affecting cholesterol gallstone formation among inbred strains of mice. Proc Natl Acad Sci USA 92: 77297733, 1995.[Abstract]
-
Klinkspoor JH, Kuver R, Savard CE, Oda D, Azzouz H, Tytgat GN, Groen AK, and Lee SP. Model bile and bile salts accelerate mucin secretion by cultured dog gallbladder epithelial cells. Gastroenterology 109: 264274, 1995.[ISI][Medline]
-
Lammert F, Carey MC, and Paigen B. Chromosomal organization of candidate genes involved in cholesterol gallstone formation: a murine gallstone map. Gastroenterology 120: 221238, 2001.[ISI][Medline]
-
Lammert F, Cohen DE, Paigen B, Carey MC, and Beier DR. The gene encoding the multispecific organic anion transporter (Cmoat) of the hepatocyte canalicular membrane maps to mouse chromosome 19. Mamm Genome 9: 8788, 1998.[ISI][Medline]
-
Lammert F, Nelson HM, Wang DQH, Carey MC, and Paigen B. Quantitative trait locus (QTL) mapping identifies a new murine gallstone gene on chromosome 17 (Abstract). Hepatology 26: 401A, 1997.
-
Lammert F, Wang DQH, Nelson HM, Bouchard G, Carey MC, and Paigen B. Quantitative trait locus mapping in mice identifies X-chromosome-linked cholesterol gallstone genes (Abstract). Hepatology 28: 502, 1998.
-
Lammert F, Wang DQH, Paigen B, and Carey MC. Phenotypic characterization of Lith genes that determine susceptibility to cholesterol cholelithiasis in inbred mice: integrated activities of hepatic lipid regulatory enzymes. J Lipid Res 40: 20802090, 1999.[Abstract/Free Full Text]
-
Lander E and Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 11: 241247, 1995.[ISI][Medline]
-
Lander ES and Botstein D. Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121: 185199, 1989.[Abstract/Free Full Text]
-
Lee SP, Carey MC, and LaMont JT. Aspirin prevention of cholesterol gallstone formation in prairie dogs. Science 211: 14291431, 1981.[ISI][Medline]
-
Lee SP, LaMont JT, and Carey MC. Role of gallbladder mucus hypersecretion in the evolution of cholesterol gallstones. J Clin Invest 67: 17121723, 1981.[ISI][Medline]
-
Legare ME, Bartlett FS II, and Frankel WN. A major effect QTL determined by multiple genes in epileptic EL mice. Genome Res 10: 4248, 2000.[Abstract/Free Full Text]
-
Machleder D, Ivandic B, Welch C, Castellani L, Reue K, and Lusis AJ. Complex genetic control of HDL levels in mice in response to an atherogenic diet. Coordinate regulation of HDL levels and bile acid metabolism. J Clin Invest 99: 14061419, 1997.[Abstract/Free Full Text]
-
Makalowski W and Boguski MS. Evolutionary parameters of the transcribed mammalian genome: an analysis of 2,820 orthologous rodent and human sequences. Proc Natl Acad Sci USA 95: 94079412, 1998.[Abstract/Free Full Text]
-
Mardones P, Quiñones V, Amigo L, Moreno M, Miquel JF, Schwarz M, Miettinen HE, Trigatti B, Krieger M, VanPatten S, Cohen DE, and Rigotti A. Hepatic cholesterol and bile acid metabolism and intestinal cholesterol absorption in scavenger receptor class B type I-deficient mice. J Lipid Res 42: 170180, 2001.[Abstract/Free Full Text]
-
Miller LJ, Holicky EL, Ulrich CD, and Wieben ED. Abnormal processing of the human cholecystokinin receptor gene in association with gallstones and obesity. Gastroenterology 109: 13751380, 1995.[ISI][Medline]
-
Morel L, Blenman KR, Croker BP, and Wakeland EK. The major murine systemic lupus erythematosus susceptibility locus, Sle1, is a cluster of functionally related genes. Proc Natl Acad Sci USA 98: 17871792, 2001.[Abstract/Free Full Text]
-
Nakajima T, Hamakubo T, Kodama T, Inazawa J, and Emi M. Genomic structure and chromosomal mapping of the human sterol regulatory element binding protein (SREBP) cleavage-activating protein (SCAP) gene. J Hum Genet 44: 402407, 1999.[ISI][Medline]
-
Nishina PM, Wang J, Toyofuku W, Kuypers FA, Ishida BY, and Paigen B. Atherosclerosis and plasma and liver lipids in nine inbred strains of mice. Lipids 28: 599605, 1993.[ISI][Medline]
-
Paigen B and Carey MC. Gallstones In: The Genetic Basis of Common Diseases, edited by King RA, Rotter JF, and Motulsky AG. New York: Oxford University Press. In press.
-
Paigen B, Schork NJ, Svenson KL, Cheah YC, Mu JL, Lammert F, Wang DQH, Bouchard G, and Carey MC. Quantitative trait loci mapping for cholesterol gallstones in AKR/J and C57L/J strains of mice. Physiol Genomics 4: 5965, 2000.[Abstract/Free Full Text]
-
Robins SJ and Fasulo JM. High density lipoproteins, but not other lipoproteins, provide a vehicle for sterol transport to bile. J Clin Invest 99: 380384, 1997.[Abstract/Free Full Text]
-
Rowe LB, Nadeau JH, Turner R, Frankel WN, Letts VA, Eppig JT, Ko MS, Thurston SJ, and Birkenmeier EH. Maps from two interspecific backcross DNA panels available as a community genetic mapping resource. Mamm Genome 5: 253274, 1994.[ISI][Medline]
-
Schneider H, Sanger P, and Hanisch E. In vitro effects of cholecystokinin fragments on human gallbladders. Evidence for an altered CCK-receptor structure in a subgroup of patients with gallstones. J Hepatol 26: 10631068, 1997.[ISI][Medline]
-
Schork NJ, Boehnke M, Terwilliger JD, and Ott J. Two-trait-locus linkage analysis: a powerful strategy for mapping complex genetic traits. Am J Hum Genet 53: 11271136, 1993.[ISI][Medline]
-
Schwartz CC, Halloran LG, Vlahcevic ZR, Gregory DH, and Swell L. Preferential utilization of free cholesterol from high-density lipoproteins for biliary cholesterol secretion in man. Science 200: 6264, 1978.[ISI][Medline]
-
Sen S and Churchill GA. A statistical framework for quantitative trait mapping. Genetics 159: 371387, 2001.[Abstract/Free Full Text]
-
Shekels LL, Hunninghake DA, Tisdale AS, Gipson IK, Kieliszewski M, Kozak CA, and Ho SB. Cloning and characterization of mouse intestinal MUC3 mucin: 3' sequence contains epidermal-growth-factor-like domains. Biochem J 330: 13011308, 1998.[ISI][Medline]
-
Sugiyama F, Churchill GA, Higgins DC, Johns C, Makaritsis KP, Gavras H, and Paigen B. Concordance of murine quantitative trait loci for salt-induced hypertension with rat and human loci. Genomics 71: 7077, 2001.[ISI][Medline]
-
Svenson KL, Cheah YC, Shultz KL, Mu JL, Paigen B, and Beamer WG. Strain distribution pattern for SSLP markers in the SWXJ recombinant inbred strain set: chromosomes 1 to 6. Mamm Genome 6: 867872, 1995.[ISI][Medline]
-
Taylor BA and Rowe L. Genes for serum amyloid A proteins map to chromosome 7 in the mouse. Mol Gen Genet 195: 491499, 1984.[ISI][Medline]
-
Vandenhaute B, Buisine MP, Debailleul V, Clement B, Moniaux N, Dieu MC, Degand P, Porchet N, and Aubert JP. Mucin gene expression in biliary epithelial cells. J Hepatol 27: 10571066, 1997.[ISI][Medline]
-
Wang DQH, Lammert F, Paigen B, and Carey MC. Phenotypic characterization of Lith genes that determine susceptibility to cholesterol cholelithiasis in inbred mice: pathophysiology of biliary lipid secretion. J Lipid Res 40: 20662079, 1999.[Abstract/Free Full Text]
-
Wang DQH, Paigen B, and Carey MC. Phenotypic characterization of Lith genes that determine susceptibility to cholesterol cholelithiasis in inbred mice: physical-chemistry of gallbladder bile. J Lipid Res 38: 13951411, 1997.[Abstract]
-
Warden CH, Davis RC, Yoon MY, Hui DY, Svenson K, Xia YR, Diep A, He KY, and Lusis AJ. Chromosomal localization of lipolytic enzymes in the mouse: pancreatic lipase, colipase, hormone-sensitive lipase, hepatic lipase, and carboxyl ester lipase. J Lipid Res 34: 14511455, 1993.[Abstract]
-
West DB, Goudey-Lefevre J, York B, and Truett GE. Dietary obesity linked to genetic loci on chromosomes 9 and 15 in a polygenic mouse model. J Clin Invest 94: 14101416, 1994.[ISI][Medline]
-
West DB, Waguespack J, York B, Goudey-Lefevre J, and Price RA. Genetics of dietary obesity in AKR/J x SWR/J mice: segregation of the trait and identification of a linked locus on chromosome 4. Mamm Genome 5: 546552, 1994.[ISI][Medline]