Analysis of the IBD5 locus and potential gene-gene interactions in Crohn's disease

Abstract

Background and aims: Genetic variation in the chromosome 5q31 cytokine cluster (IBD5 risk haplotype) has been associated with Crohn's disease (CD) in a Canadian population. We studied the IBD5 risk haplotype in both British and Japanese cohorts. Disease associations have also been reported for CARD15/NOD2 and TNF variants. Complex interactions between susceptibility loci have been shown in animal models, and we tested for potential gene-gene interactions between the three CD associated loci.

Methods: Family based association analyses were performed in 457 British families (252 ulcerative colitis, 282 CD trios) genotyped for the IBD5 haplotype, common CARD15, and TNF-857 variants. To test for possible epistatic interactions between variants, transmission disequilibrium test analyses were further stratified by genotype at other loci, and novel log linear analyses were performed using the haplotype relative risk model. Case control association analyses were performed in 178 Japanese CD patients and 156 healthy controls genotyped for the IBD5 haplotype.

Results: The IBD5 haplotype was associated with CD (p=0.007), but not with UC, in the British Caucasian population. The CARD15 variants and IBD5 haplotype showed additive main effects, and in particular no evidence for epistatic interactions was found. Variants from the IBD5 haplotype were extremely rare in the Japanese.

Conclusions: The IBD5 risk haplotype is associated with British CD. Genetic variants predisposing to CD show heterogeneity and population specific differences.

Figure 1

Agarose gel image of IGR2060a_1 genotyping. First lane is 100 bp DNA size marker, followed by products of 24 individual polymerase chain reactions (PCR). PCR product sizes are control (212 bp), C allele (149 bp), and G allele (105 bp). Arrows mark, in order, CC homozygote, GC heterozygote, GG homo- zygote. Lanes marked “h” are negative controls (containing no DNA).