Introduction

Inflammatory bowel diseases (IBDs), which are usually classified into two clinical entities—Crohn disease (CD; MIM 266600) and ulcerative colitis (UC)—are chronic conditions characterized by remitting and relapsing inflammation of the small and/or large intestines. Familial aggregation and twin studies indicate a presence of genetic factors susceptible to this condition. Genome-wide linkage analyses have localized genes conferring susceptibility to IBD to several possible candidate loci on chromosomes 1, 3, 5, 6, 7, 10, 12, 14, 16, 19, and 22 (Hugot et al. 1996; Satsangi et al. 1996; Cho et al. 1998; Duerr et al. 2000; Hampe et al. 1999; Rioux et al. 2000, 2001).

Among the several candidate loci, susceptible genes at two distinct loci were recently identified through the evidences of strong association with the CD phenotype. In the IBD5 locus on chromosome 5, single-nucleotide polymorphisms (SNPs) in two candidate genes, SLC22A4 and SLC22A5, both of which encode organic cation transporters, revealed significant associations with CD (Peltekova et al. 2004). A C1672T substitution in exon 9 of the SLC22A4 gene and a G-207C in the SLC22A5 promoter region were indicated as functional and causative mutations to increase susceptibility to CD. The other gene identified from chromosome 10 was the DLG5 gene, encoding a scaffolding protein involved in the maintenance of epithelial integrity. Risk-associated variants, including a G113A substitution in exon 3 of the DLG5 gene, constructed two distinct haplotypes with a replicable distortion in transmission (Stoll et al. 2004).

To investigate a possible role of these candidate gene loci, one corresponding to SLC22A4 and SLC22A5 and another corresponding to DLG5, in the pathogenesis of CD in Japanese, we examined SNPs of these three genes in a large number of clinical samples. We here report an absence of DNA substitutions or lack of association for the candidate-causative SNPs, which were indicated in the previous reports, in the Japanese CD patients. However, we observed a weak association of other genetic substitutions in these genes of Japanese patients with CD. Our results indicate that the reported substitutions in the three genes are unlikely to be causative to Japanese CD patients, but the candidacy of these two loci for Japanese CD cannot be totally excluded.

Materials and methods

Subjects and DNAs

Japanese blood samples were obtained with written informed consent from 484 CD patients at the Social Insurance Central General Hospital and from 345 unaffected control individuals belonging to the Osaka-Midosuji Rotary Club. All CD cases were diagnosed at the Inflammatory Bowel Unit of the Social Insurance Hospital by clinical, radiological, endoscopic, and histological findings according to the Lennard-Jones’ criteria (Lennard-Jones 1989). Patients with indeterminate colitis were excluded. DNAs were prepared from these samples according to standard protocols.

DNA sequencing

To search genetic variations in these candidate loci including the five reported variants, C1672T in exon 9 of SLC22A4 and G-207C in the SLC22A5 promoter, as well as G113A in exon 3, C4136A in exon 23 and 35delA in intron 26 of DLG5, we carried out direct sequencing of those regions in 48 individuals with confirmed diagnosis of CD by means of the BigDye Terminator RR Mix (Applied Biosystems, USA) with ABI 3700 sequencers using the primers listed in Table 1.

Table 1 List of primers in this study

Markers

SNPs in the SLC22A4 and SLC22A5 genes were screened according to methods described previously (Saito et al. 2002). We selected 17 SNPs, including six in SLC22A4 (SLC22A4 1–6), three in SLC22A5 (SLC22A5 1–3), and eight in DLG5 (DLG5 1–8) (Table 2). Information for each SNP in the SLC22A4, SLC22A5, and DLG5 chosen for this study was obtained from the Japanese SNP (JSNP) database (http://snp.ims.u-tokyo.ac.jp) (Hirakawa et al. 2002; Haga et al. 2002).

Table 2 List of genotyped single-nucleotide polymorphisms (SNPs) in SLC22A4, SLC22A5 and DLG5

SNP analysis and genotyping

We amplified multiple genomic fragments using 20 ng of genomic DNA for each polymerase chain reaction (PCR), as described previously (Ohnishi et al. 2001). We genotyped all participants for a total of 17 SNPs indicated in Table 2 by means of the Invader assay (Mein et al. 2000). The C4136A in exon 23 and 35delA in intron 26 (deletion of an adenine at the 35th nucleotide in intron 26) of DLG5 were examined by direct sequencing using the same primers.

Statistical analysis

Genotype distributions and allele frequencies of each SNP were compared, respectively, between the cases and the controls, as described elsewhere (Yamada et al. 2001). Haplotype frequencies were estimated using the expectation-maximization algorithm (Ott 1977).

Results

To examine a possible association of genetic substitutions in the two candidate loci—one including SLC22A4 and SLC22A5 and the other at DLG5—with susceptibility to CD in the Japanese population, we first examined DNA sequences from 48 CD patients in 438–976 bp genomic regions, including the five major genetic variants—C1672T in exon 9 of SLC22A4 and G-207C in the SLC22A5 promoter, G113A in exon 3, C4136A in exon 23, and 35delA in intron 26 of DLG5—that were reported to have significant associations with CD in the Caucasian population (Table 1). Among these five genetic variations reported previously, we found that the three SNPs, C1672T, G-207C, and G113A, were completely absent in the Japanese CD cases. Since the C4136A and 35delA variations were observed in the Japanese population, we carried out genotyping of 484 Japanese CD patients for these variations and found no association of these two reported substitutions to CD in the Japanese population (Table 3).

Table 3 Association of major genetic variants in DLG5 with Crohn disease (CD) in the case-control study

To further verify whether these three genes can be excluded as candidates for Japanese CD, we performed case-control association studies by means of genotyping of 17 JSNPs located within the three genes at the two loci as shown in Table 2. The analyses using allelic, recessive, and dominant models for CD patients versus controls disclosed an association of two SNPs, one at SLC22A4_2 (P=0.028) by dominant model and the other at DLG5_2 (P=0.023) by recessive model, although the associations observed here were much weaker than those for the five genetic variations observed in Caucasian CD cases (Table 4). In addition, we constructed the haplotype structure using the 19 genotyped variations and examined its association with CD but found no significant association with CD (data not shown). Our studies have indicated that the five reported variants are unlikely to be disease causative, but we have not excluded a possibility that these genes may play some role in susceptibility to CD in the Japanese population.

Table 4 Association of SLC22A4 and DLG5 with Crohn disease (CD) in the case-control study

Discussion

Genetic factors that affect susceptibility to CD have been disclosed through genetic linkage and population-based association studies although it is very far from complete understanding of the subject. CARD15 was found to be associated with IBD by means of genome-wide sib-pair analysis (Hampe et al. 2001; Hugot et al. 2001; Ogura et al. 2001). Through the candidate gene approach, various genes, such as mucin 3 (MUC3), tumor necrosis factor (TNF), and HLA class II, were identified as candidate genes susceptible to IBD in some populations (Nakajima et al. 1995; Kyo et al. 1999, 2001; Negoro et al. 1999). In addition, recent studies identified three candidate susceptibility genes at two loci, one was SLC22A4 and SLC22A5 on chromosome 5 corresponding to IBD5 (Peltekova et al. 2004), and the other was DLG5 on chromosome 10 (Stoll et al. 2004).

Our case-control study for SLC22A4, SLCA22A5, and DLG5 showed no evidence of association between SNPs in the SLC22A5 gene and CD and that there might be some associations with SNPs in the two gene loci, SLC22A4 and DLG5, to the disease, if any. In addition, it is notable that the SNPs showing weak and possible associations in our study were different from ones reported previously; three variations, C1672T in exon 9 of SLC22A4, G-207C in the SLC22A5 promoter region, and G113A in exon 3 of DLG5, that showed the strong associations in Caucasian CD were completely absent in Japanese. The two remaining candidate variants, C4136A of exon 23 and 35 delA in intron 26 of DLG5, were found to be polymorphic in Japanese, but no association between these SNPs and Japanese CD was observed.

Interestingly, the genetic variants that showed the strong association in Caucasian but were completely absent in Japanese CD were indicated to interact with other genetic variants of CARD15 that was also indicated to be a candidate susceptible gene to CD. Three major polymorphisms in the CARD15 gene—R702W, G908R, and 1007fs—were confirmed to be independently associated with susceptibility to Caucasian patients with CD (Ahmad et al. 2002; Cuthbert et al. 2002; Lesage et al. 2002). However, our extensive DNA sequence analysis of this gene in more than 400 Japanese CD patients failed to identify such genetic variations except for a single case, indicating no involvement of CARD15 in pathogenesis of Japanese CD (Yamazaki et al. 2002). Ethnic differences in the genetic variations among Caucasian, Asian, and African populations were also shown by others (Bonen et al. 2002; Inoue et al. 2002; Croucher et al. 2003).

We failed to confirm the association of the five candidate genetic variations in the SLC22A4, SLCA22A5, and DLG5 genes in the previous reports to be susceptible to Japanese CD. However, we found a weak association of SNPs in the two genes, SLC22A4 and DLG5, with Japanese CD. The results indicate a possibility that the five SNPs in the previous reports may not be causative, but the SNPs that we found to have possible association with or specific genetic substitutions having linkage disequilibrium with these SNPs in the region may play some role in Japanese CD. Nonetheless, combining the data that there is no association of CARD15 with Japanese CD, it is apparent that there should be a presence of ethnic differences in susceptibility to CD. Further studies including both large-scale genomic and environmental analysis involving a large number of cases and controls are warranted to identify genes susceptible to CD on a worldwide scale, and such studies would eventually shed more light on the etiology of IBD.