Ulcerative colitis (UC) is a multifactorial disorder with both genetic and environmental factors. HLA-B*52 and DRB1*1502 are reported to be strongly associated with UC in Japan. However, the actual susceptible gene has not been identified yet. In this study, to map precisely the susceptible locus for UC, we performed association mapping in the chromosome 6p using 24 microsatellite markers distributed over 16 Mb. A total of 183 patients with UC and 186 healthy controls (HC) were included in this study. In all, 15 markers around the human leukocyte antigen (HLA) region showed statistical significance in the genotypic differentiation test concerned with the allelic distribution between the UC and HC. Especially, the markers between the centromeric region of HLA class I and the telomeric region of class III showed remarkably low P-values and the allele239 of C2-4-4 in class I marker showed the strongest association (Pc=2.9 × 10−9: OR=3.74, 95% CI=2.50–5.60). Furthermore, we found strong linkage disequilibrium (LD) between the allele239 of C2-4-4 and HLA-B*52 in haplotype analysis. These results provide evidence that, in Japanese, important determinants of disease susceptibility to UC may exist in HLA, especially between the centromeric region of class I and the telomeric region of class III, under the strong LD with HLA-B*52.
Little is known about the pathogenesis of ulcerative colitis (UC) and Crohn’s disease (CD), which are collectively referred to as inflammatory bowel diseases (IBD). Epidemiological data including familial aggregation, increased concordance rates in monozygotic twins compared with dizygotic twins, ethnic differences in disease prevalence, and an association with recognized genetic syndromes have suggested that a genetic background of susceptibility is important in the pathogenesis of IBD.1, 2, 3, 4, 5, 6, 7, 8, 9, 10 Previous genome-wide linkage studies in Western Europe, North America and Australia have demonstrated several loci relevant to IBD.11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22 These studies have used panels of multiply affected relative pairs and hypothesis-free, nonparametric methods of linkage analysis to detect genomic regions showing significant linkage. Putative susceptibility loci have been found on chromosomes 1, 5, 6, 12, 14, 16, and 19. The chromosome 6p is named IBD3 and overlaps with the human leukocyte antigen (HLA) region. It has been reported that HLA class I genes are associated with UC in many populations, such as B52 in Japanese,23 B35 in Ashkenazi Jews,24 Cw7 in Finns,25 and A33 in Asian Indians.26 It has also been reported that HLA class II genes are associated with UC, such as DRB1*1502, DQB1*0601 and DPB1*0901 in Japanese,27, 28, 29 DR2 in ethnically matched patients in California30 and DRB1*0103 in European Caucasian patients with UC.11 The discrepancies in the results between the different ethnic groups may be attributed to genetic heterogeneity, the effect of recombination on an ancestral haplotype structure, and differences of allele frequencies in each population.
It is known that a conserved haplotype of HLA A*24-B*52-DRB1*1502-DQB1*0601-DPB1*0901 is strongly associated with UC in Japan,29 which includes almost the entire HLA region. Therefore, we considered that the susceptible gene for UC in Japanese is located within or near the HLA region and that it would be important to map the short arm of chromosome 6 adequately to elucidate the susceptible locus of UC.
Recently, a large number of new polymorphic microsatellite markers have been identified throughout the HLA region.31, 32, 33, 34 They provide valuable information for focusing on the critical regions of diseases, such as Behçet disease,35 psoriasis vulgaris,36 type I diabetes,37 rheumatoid arthritis 38 and nonmelanoma skin cancer.39
In the present study, we employed a two-stage approach for association mapping using microsatellite markers spanning 16 Mb of chromosome 6p including the HLA region to elucidate precisely the susceptible locus for Japanese UC.
Hardy–Weinberg proportion test
This study was performed using 24 microsatellite markers (20 markers and an additional four markers for fine mapping) in patients with UC and the healthy controls (HC) in stages 1 and 2. No markers deviated from the Hardy–Weinberg equilibrium (HWE) (P>0.05) in HC in either stage. However, three markers (C1-2-A, DQ.CAR, TNFa) in the UC of stage 1 and four markers (DQ.CAR, C1-2-A, D6S1568, D6S1629) in the UC of stage 2 significantly deviated from the HWE (Figure 1).
Genotypic differentiation test
A genotypic differentiation test for the allelic distribution between the UC and HC groups was performed. Eight markers (C1-2-A, D6S2444, C3-2-11, D6S1629, D6S2445, D6S273, TNFa, DQ.CAR) showed significantly different distributions of alleles between UC and HC in stage 1. In stage 2, nine markers (C1-2-A, D6S273, C3-2-11, TNFa, DQ.CAR, D6S1629, D6S2444, D6S1683, D6S306) showed significantly different allelic distributions between UC and HC (Figure 2). The allelic distributions in UC at markers between the centromeric region of class I and the telomeric region of class III were clearly different from those in HC, whereas the allelic distributions in UC at markers outside of the HLA region were not strongly different from those in HC in either stage. C1-2-A in class I showed the lowest P-value in both stages (Pstage 1=0.00104, Pstage 2=0.00039).
We replicated similar results in the two-stage study, suggesting that the susceptible gene to UC was most likely localized in the HLA region, especially around the C1-2-A.
Allelic association test
Each allele of the microsatellite markers was named according to the size of the amplified fragment and the allele frequencies were compared between UC and HC in each marker. Among alleles in the 20 markers, the strongest association was found in allele234 of C1-2-A in both stages 1 and 2 (allele234: Pcstage 1=0.000014, Pcstage 2=0.000003, Pctotal=3.9 × 10−9: ORtotal=2.99, 95% CItotal=2.13–4.19) (Table 1, Figure 3).
Fine mapping with additional microsatellite markers
The results of the genotypic differentiation test and allelic association test at C1-2-A showed the lowest P-value among markers tested in stage 1 and were replicated in stage 2, suggesting that a susceptible gene for UC is likely to exist near C1-2-A. Therefore, we performed further fine mapping with the additional four markers (C1-4-1, C1-2-5, C1-3-1, C2-4-4) near C1-2-A. Final evaluations using the 20 and the additional four markers were made in a combined cohort of both stages 1 and 2. The four markers did not deviate from the HWE in HC. In contrast, C1-2-5 and C2-4-4 deviated from the HWE in UC (Figure 1). The allelic distributions at the additional four markers in UC significantly deviated from those in HC on the basis of the genotypic differentiation test (Figure 2). The strongest association was found in allele239 of C2-4-4 with the allelic association test (allele239: Pc=2.9 × 10−9: OR=3.74, 95% CI=2.50–5.60) (Table 1, Figure 3).
Linkage disequilibrium analysis
Because a remarkable association between HLA-B*52 and Japanese patients with UC was reported in a previous study,29 we performed a linkage disequilibrium (LD) analysis between HLA-B*52 and each allele of all the microsatellite markers used in this study. All alleles that were positively associated with UC showed LD with HLA-B*52 (data not shown). A tight LD was observed between HLA-B*52 and allele239 of C2-4-4 (R=0.858, P=1.1 × 10−35, Pc=6.8 × 10−34).
Analysis in the subgroups of UC
When patients with UC were classified according to disease phenotypes (proctitis, left-sided colitis, total colitis or segmental type) and the allelic association test was performed in a combined cohort of both stages 1 and 2, the frequency of allele234 of C1-2-A was higher in the left-sided colitis than in proctitis (P=0.0149, Pc=0.0447). No significant differences were found in the other alleles (data not shown).
Previous association studies carried out using modern molecular genotyping methods provided strong evidence that genes within or near the HLA region are important determinants of susceptibility and disease behavior in IBD. In the present study, we performed an association mapping using microsatellite markers spanning 16 Mb in the IBD3 to narrow the susceptible locus in Japanese patients with UC. These markers were densely distributed in the HLA region at appropriate distances. The average number of alleles at these microsatellite markers was 10.2 and these had been selected as informative polymorphic genetic markers.35, 36, 37, 38, 39 Microsatellite markers are generally considered to be useful for association mapping, because of the higher levels of heterozygosity and longer extent of useful levels of LD than single-nucleotide polymorphisms (SNPs).41
In this study, several markers in the HLA region showed significant deviations from the HWE in UC. These markers also showed significant results in the genotypic differentiation and allelic association tests. Among these markers, C1-2-A showed the strongest association with UC on the basis of odds ratio, and all significant results of C1-2-A in stage 1 were replicated in stage 2, suggesting that these results had high reliability and that the susceptible gene was near C1-2-A. Then further mapping was performed in a combined cohort of stages 1 and 2. We set up an additional four markers near the C1-2-A, and found that allele239 of C2-4-4 (282 kb telomeric from the HLA-B) showed the strongest association (Pc=2.9 × 10−9: OR=3.74, 95% CI=2.50–5.60). These results strongly suggest that candidate susceptibility gene for UC is likely to exist between the centromeric region of class I and the telomeric region of class III. Interestingly, this genomic region corresponds with psoriaris vulgaris36 and nonmelanoma skin cancer.39 However, it is likely that the ancestral haplotypes for UC, psoriaris vulgaris and nonmelanoma skin cancer population groups are different, because the associated alleles for the same markers are not concordant among these diseases. Moreover, we found strong LD (R=0.858) between the allele239 of C2-4-4 and HLA-B*52 which has been reported to be associated with UC. Because 90% of HLA-B*52 form the DRB1*1502-HLA-B52* haplotype in the Japanese population,42 it is thought that the allele239 of C2-4-4 also shows strong LD with DRB1*1502 which has been also reported to be associated with UC. Thus, the associations of the allele239 of C2-4-4, HLA-B*52, and DRB1*1502 with UC appear to reflect the LD with an identical risk allele of a true susceptibility gene for UC.
It has been hypothesized that UC is not a single, homogenous disorder, but consists of heterogeneous groups that may be produced by extensive genetic heterogeneity. Therefore, it is necessary to investigate the allele frequencies in subgroups of UC stratified according to the clinical phenotypes. When UC was classified by the location of the lesions, only the allele234 of C1-2-A was higher in the left-sided colitis than in proctitis. No significant differences were found in other markers in the allele or carrier frequencies between the respective subgroups. Considering these data, it seems that the gene near C1-2-A may affect the disease location to some extent.
The genetic heterogeneity is just one of the factors that will confound attempts to isolate complex disease genes. UC possesses a complex genetic trait wherein several genes interact with different environmental factors and determine the clinical phenotype of the disease. Moreover, there is a possibility that combinations of alleles may interact on the same or the opposing haplotype to cause disease. Therefore, the mapping of complex trait loci is a difficult task and requires a large sample size and a dense genetic map. Further analysis is needed to narrow the susceptible locus using the SNPs for high-resolution mapping in the centromeric region of class I to the telomeric region of class III.
In conclusion, the susceptible gene of Japanese UC is localized in HLA, especially between the centromeric region of class I and the telomeric region of class III, under the strong LD with HLA-B*52.
Materials and methods
Blood samples in stage 1 were obtained from 92 HC and 91 patients with UC who visited Tohoku University Hospital from May 1998 to December 1999, and samples in stage 2 were obtained from 94 HC and 92 patients who visited the hospital from January 2000 to January 2002. There was no overlap of subjects between stages 1 and 2. All patients and HC were Japanese. The characteristics of the subjects are summarized in Table 2. No differences were found in the age of onset, mean age, or clinical and analytical characteristics between the patients of stages 1 and 2. The diagnosis of UC was made on the basis of clinical symptoms and endoscopic, radiographic, and histological findings according to conventional criteria. The study protocols were approved by the Ethics Committee of Tohoku University School of Medicine. All patients and HC gave written and informed consent to participate in this study.
Genotyping for microsatellite alleles
Genomic DNAs were obtained from peripheral blood mononuclear cells by standard phenol–chloroform extraction and ethanol precipitation or by utilizing the NA-1000 Automated Nucleic Acid Extraction Machine (Kurabo, Osaka, Japan). Microsatellite markers and primer sequences used in this study have been previously reported.31, 32, 33, 34, 35, 36, 37, 38, 39 In all, 24 microsatellite markers were distributed over a 16 Mb including the HLA region. To determine the number of repeat units of the microsatellite loci exhibiting polymorphisms, we synthesized forward primers by labeling the 5′-ends with the fluorescent reagents 6-FAM, HEX, and TET (PE Applied Biosystems, Foster City, CA, USA). The PCR reaction mixture contained 12.5 ng of genomic DNA, 1 μl of dNTP (2.5 mM each), 1 μl of 10 × PCR buffer (100 mM Tris-HCl, pH 8.3, 500 mM KCl, 15 mM MgCl2), 5 pmol of forward and reverse primers and 0.25 U of AmpliTaq Gold (PE Applied Biosystems) in a total volume of 10 μl. Amplification was carried out using a PTC-200 DNA Engine (MJ Research, Inc. Waltham, MA, USA) under the following conditions: 40 cycles of denaturing at 94°C for 30 s, annealing at 53–67°C for 30 s, extension at 72°C for 30 s with a final incubation at 72°C for 10 min. The amplified products were denatured for 2 min at 96°C, mixed with formamide containing stop buffer, applied with a size standard marker of GS500 TAMRA (PE Applied Biosystems) to each lane and run on a 4% polyacrylamide denaturing sequencing gel containing 8 M urea in an automated DNA sequencer. The fragment sizes of PCR products were determined automatically using GeneScan software (PE Applied Biosystems).
Patients with UC were classified according to the location and the extent of the inflammatory lesions (proctitis, left-sided colitis, total colitis, or segmental type), need for intensive intravenous steroid therapy, and need for surgical treatments. Steroid therapy was administered according to an intensive intravenous regimen described previously.40 Surgical operation was carried out in patients with severe UC who did not respond to the steroid therapy.
A two-stage approach for association analysis was performed. The results of stage 1 were reconfirmed in stage 2 in order to exclude false-positive results. Allele and carrier frequencies (proportion possessing one or two copies of an allele) were estimated by direct counting. The exact test of Hardy–Weinberg proportion (HWP) for multiple alleles and the genotypic differentiation test for the allelic distribution between the UC and HC groups were performed by the Markov chain method using the Genepop software package.43, 44, 45 The Markov chain method has the advantage of providing a complete enumeration for testing HWP in cases where the number of alleles and the sample size is small. A P-value of less than 0.05 was regarded as statistically significant. Subsequently, an allelic association test was performed by comparing the allele frequencies of each marker between the UC and the HC groups with the χ2-test using a 2 × 2 contingency table. The P-value was corrected by multiplying the number of alleles observed in each locus tested (corrected P-value: Pc-value). A Pc-value of less than 0.05 was regarded as statistically significant. The strength of association was estimated by the odds ratio (OR). Moreover, an additional four microsatellite markers (C1-4-1, C1-2-5, C1-3-1, C2-4-4) were estimated in the same way for further investigation. Comparisons within each stratum (classified according to the clinicopathological features) were made by the χ2-test for independence.
The LD analysis with HLA-B was performed using haplotype frequency estimation software (EH program). This program uses an expectation–maximization algorithm to estimate the haplotype frequencies based on a maximum likelihood approach.46, 47 HLA-B genotyping was performed by the PCR-SSP (sequence–specific primer) method.48 Each haplotype was analyzed using a 2 × 2 contingency table with the χ2-test, and P was corrected by the number of haplotypes observed between each marker and HLA-B (Pc). The degree of LD (R) was calculated in HC by the EH program and assessed by the χ2-test. This measure of R has a range of −1 to 1, and is equal to 0 when there is no LD.
Maratka Z, Sera J . Familiar occurrence of ulcerative colitis. Gastroenterologia 1965; 103: 321–325.
Mayberry JF . Recent epidemiology of ulcerative colitis and Crohn's disease. Int J Colorectal Dis 1989; 4: 59–66.
Orholm M, Munkholm P, Langholz E, Nielsen OH, Sorensen IA, Binder V . Familial occurrence of inflammatory bowel disease. N Engl J Med 1991; 324: 84–88.
Satsangi J, Grootscholten C, Holt H, Jewell DP . Clinical patterns of familial inflammatory bowel disease. Gut 1996; 38: 738–741.
Cipolla C, Magliocco A, Oliva L, Cottone M . Familial aggregation of inflammatory bowel disease in a Mediterranean area. Eur J Epidemiol 1996; 12: 205–210.
Russel MG, Pastoor CJ, Janssen KM et al. Familial aggregation of inflammatory bowel disease: a population-based study in South Limburg, The Netherlands. The South Limburg IBD Study Group. Scand J Gastroenterol Suppl 1997; 223: 88–91.
Tysk C, Lindberg E, Jarnerot G, Floderus-Myrhed B . Ulcerative colitis and Crohn's disease in an unselected population of monozygotic and dizygotic twins. A study of heritability and the influence of smoking. Gut 1988; 29: 990–996.
Thompson NP, Driscoll R, Pounder RE, Wakefield AJ . Genetics versus environment in inflammatory bowel disease: results of a British twin study. BMJ 1996; 312: 95–96.
Orholm M, Binder V, Sorensen TI, Rasmussen LP, Kyvik KO . Concordance of inflammatory bowel disease among Danish twins. Results of a nationwide study. Scand J Gastroenterol 2000; 35: 1075–1081.
Roth MP, Petersen GM, McElree C, Feldman E, Rotter JI . Geographic origins of Jewish patients with inflammatory bowel disease. Gastroenterology 1989; 97: 900–904.
Ahmad T, Satsangi J, McGovern D, Bunce M, Jewell DP . Review article: the genetics of inflammatory bowel disease. Aliment Pharmacol Ther 2001; 15: 731–748.
Hampe J, Schreiber S, Shaw SH et al. A genomewide analysis provides evidence for novel linkages in inflammatory bowel disease in a large European cohort. Am J Hum Genet 1999; 64: 808–816.
Hampe J, Shaw SH, Saiz R et al. Linkage of inflammatory bowel disease to human chromosome 6p. Am J Hum Genet 1999; 65: 1647–1655.
Rioux JD, Silverberg MS, Daly MJ et al. Genomewide search in Canadian families with inflammatory bowel disease reveals two novel susceptibility loci. Am J Hum Genet 2000; 66: 1863–1870.
Dechairo B, Dimon C, van Heel D et al. Replication and extension studies of inflammatory bowel disease susceptibility regions confirm linkage to chromosome 6p (IBD3). Eur J Hum Genet 2001; 9: 627–633.
Satsangi J, Parkes M, Louis E et al. Two stage genome-wide search in inflammatory bowel disease provides evidence for susceptibility loci on chromosomes 3, 7 and 12. Nat Genet 1996; 14: 199–202.
Curran ME, Lau KF, Hampe J et al. Genetic analysis of inflammatory bowel disease in a large European cohort supports linkage to chromosomes 12 and 16. Gastroenterology 1998; 115: 1066–1071.
Cho JH, Nicolae DL, Gold LH et al. Identification of novel susceptibility loci for inflammatory bowel disease on chromosomes 1p, 3q, and 4q: evidence for epistasis between 1p and IBD1. Proc Natl Acad Sci USA 1998; 95: 7502–7507.
Duerr RH, Barmada MM, Zhang L et al. Linkage and association between inflammatory bowel disease and a locus on chromosome 12. Am J Hum Genet 1998; 63: 95–100.
Mirza MM, Lee J, Teare D et al. Evidence of linkage of the inflammatory bowel disease susceptibility locus on chromosome 16 (IBD1) to ulcerative colitis. J Med Genet 1998; 35: 218–221.
Cho JH, Nicolae DL, Ramos R et al. Linkage and linkage disequilibrium in chromosome band 1p36 in American Chaldeans with inflammatory bowel disease. Hum Mol Genet 2000; 9: 1425–1432.
Cho J . Linkage of inflammatory bowel disease to human chromosome 6p. Inflamm Bowel Dis 2000; 6: 259–261.
Asakura H, Tsuchiya M, Aiso S et al. Association of the human lymphocyte-DR2 antigen with Japanese ulcerative colitis. Gastroenterology 1982; 82: 413–418.
Delpre G, Kadish U, Gazit E, Joshua H, Zamir R . HLA antigens in ulcerative colitis and Crohn's disease in Israel. Gastroenterology 1980; 78: 1452–1457.
Leidenius MH, Koskimies SA, Kellokumpu IH, Hockerstedt KA . HLA antigens in ulcerative colitis and primary sclerosing cholangitis. Apmis 1995; 103: 519–524.
Habeeb MA, Rajalingam R, Dhar A, Kumar A, Sharma MP, Mehra NK . HLA association and occurrence of autoantibodies in Asian-Indian patients with ulcerative colitis. Am J Gastroenterol 1997; 92: 772–776.
Futami S, Aoyama N, Honsako Y et al. HLA-DRB1*1502 allele, subtype of DR15, is associated with susceptibility to ulcerative colitis and its progression. Dig Dis Sci 1995; 40: 814–818.
Yoshitake S, Kimura A, Okada M, Yao T, Sasazuki T . HLA class II alleles in Japanese patients with inflammatory bowel disease. Tissue Antigens 1999; 53: 350–358.
Sugimura K, Asakura H, Mizuki N et al. Analysis of genes within the HLA region affecting susceptibility to ulcerative colitis. Hum Immunol 1993; 36: 112–118.
Toyoda H, Wang SJ, Yang HY et al. Distinct associations of HLA class II genes with inflammatory bowel disease. Gastroenterology 1993; 104: 741–748.
Tamiya G, Shiina T, Oka A et al. New polymorphic microsatellite markers in the human MHC class I region. Tissue Antigens 1999; 54: 221–228.
Tamiya G, Ota M, Katsuyama Y et al. Twenty-six new polymorphic microsatellite markers around the HLA-B, -C and -E loci in the human MHC class I region. Tissue Antigens 1998; 51: 337–346.
Matsuzaka Y, Makino S, Nakajima K et al. New polymorphic microsatellite markers in the human MHC class II region. Tissue Antigens 2000; 56: 492–500.
Matsuzaka Y, Makino S, Nakajima K et al. New polymorphic microsatellite markers in the human MHC class III region. Tissue Antigens 2001; 57: 397–404.
Ota M, Mizuki N, Katsuyama Y et al. The critical region for Behçet disease in the human major histocompatibility complex is reduced to a 46-kb segment centromeric of HLA-B, by association analysis using refined microsatellite mapping. Am J Hum Genet 1999; 64: 1406–1410.
Oka A, Tamiya G, Tomizawa M et al. Association analysis using refined microsatellite markers localizes a susceptibility locus for psoriasis vulgaris within a 111 kb segment telomeric to the HLA-C gene. Hum Mol Genet 1999; 8: 2165–2170.
Herr M, Dudbridge F, Zavattari P et al. Evaluation of fine mapping strategies for a multifactorial disease locus: systematic linkage and association analysis of IDDM1 in the HLA region on chromosome 6p21. Hum Mol Genet 2000; 9: 1291–1301.
Ota M, Katsuyama Y, Kimura A et al. A second susceptibility gene for developing rheumatoid arthritis in the human MHC is localized within a 70-kb interval telomeric of the TNF genes in the HLA class III region. Genomics 2001; 71: 263–270.
Oka A, Hayashi H, Tomizawa M et al. Localization of a non-melanoma skin cancer susceptibility region within the major histocompatibility complex by association analysis using microsatellite markers. Tissue Antigens 2003; 61: 203–210.
Truelove SC, Jewell DP . Intensive intravenous regimen for severe attacks of ulcerative colitis. Lancet 1974; 1: 1067–1070.
Kruglyak L . Prospects for whole-genome linkage disequilibrium mapping of common disease genes. Nat Genet 1999; 22: 139–144.
Tokunaga K, Ishikawa Y, Ogawa A et al. Sequence based association analysis of HLA class I and II alleles in Japanese supports conservation of common haplotypes. Immunogenetics 1997; 46: 199–205.
Guo SW, Thompson EA . Performing the exact test of Hardy–Weinberg proportion for multiple alleles. Biometrics 1992; 48: 361–372.
Raymond M, Rousset F . An exact test for population differentiation. Evolution 1995; 49: 1280–1283.
Raymond M, Rousset F . Genepop (version 1*2): population genetics software for exact tests and ecumenicism. J Hered 1997; 86: 248–249.
Terwilliger J, Oh J . Handbook of Human Genetic Linkage. Johns Hopkins University Press: Baltimore, 1994.
Xie X . Testing linkage disequilibrium between a disease gene and marker loci. Am J Hum Genet 1993; 53: 1107.
Kinouchi Y, Matsumoto K, Negoro K et al. Hla-B genotype in Japanese patients with Crohn's disease. Dis Colon Rectum 2003; 46: S10–S14.
About this article
Cite this article
Nomura, E., Kinouchi, Y., Negoro, K. et al. Mapping of a disease susceptibility locus in chromosome 6p in Japanese patients with ulcerative colitis. Genes Immun 5, 477–483 (2004) doi:10.1038/sj.gene.6364114
- ulcerative colitis
- human leukocyte antigen
- microsatellite marker
- susceptible gene
Tissue Antigens (2009)
The American Journal of Gastroenterology (2008)
Hepatology Research (2007)
Expert Opinion on Pharmacotherapy (2006)
Digestive Endoscopy (2006)