Further Evidence of IBD5/CARD15 (NOD2) Epistasis in the Susceptibility to Ulcerative Colitis (original) (raw)

To the Editor:

We read with interest the two recent articles describing analyses of the IBD5 (MIM 606348) risk haplotype and inflammatory bowel disease (IBD) in European cohorts (Giallourakis et al. 2003; Mirza et al. 2003). In both European cohorts, the association with the IBD5 risk haplotype and Crohn disease (CD [MIM 266600]) was replicated. Mirza et al. (2003) additionally provided evidence for interaction between IBD5 and CARD15 (NOD2) (MIM 605956) in CD, a finding not seen in the subsequent paper by Giallourakis et al. (2003) or in our own transmission/disequilibrium testing (TDT) analysis (Negoro et al. 2003). Subsequent genotype-phenotype analysis by Giallourakis et al. (2003) found no association between IBD5 and clinical subgroups of patients with CD. In contrast, our group has recently published a case-control study of U.K. whites demonstrating that the association between IBD5 and CD was confined to those individuals with perianal disease (Armuzzi et al. 2003). It would be interesting to know whether association with this particular phenotype was evaluated in the Giallourakis cohort.

In 187 trios with ulcerative colitis (UC [MIM 191390]), Giallourakis et al. (2003) also reported a novel association between the IBD5 risk haplotype and UC. This association was most pronounced in those individuals possessing one of the three common CD-associated CARD15 variant alleles, suggesting an epistatic relationship between these replicated CD loci in patients with UC. In our British patients, we observed a similar linkage disequilibrium (LD) pattern across this locus. However, using both TDT (105 transmissions to 124 nontransmissions, _P_=.24) and case-control studies, we were unable to demonstrate association between the IBD5 risk haplotype and UC (Armuzzi et al. 2003; Negoro et al. 2003). However, stratification of the IBD5 results in UC by CARD15 status was not performed.

Following the report by Giallourakis et al. (2003), we therefore stratified our trios to assess the transmissions of IGR2060a_1 (an IBD5 risk haplotype–tagging SNP) from heterozygous parents to affected offspring who also carried at least one CARD15 risk allele, revealing 14 transmissions to 12 nontransmissions (_P_=.78). The lack of association may reflect a true relationship, or it may be a type I error due to the relatively weak power of this analysis. We therefore genotyped for the three common CARD15 variants and IBD5 haplotype–tagging SNP to assess any epistatic association in a more powerful case-control study of 278 patients with UC (largely independent of the UC trios) and 232 healthy controls (HC). We found a novel association between the CARD15 702Trp variant and UC (table 1). This association was not significant in the IBD5 wild-type homozygotes but became significant in the IBD5 heterozygotes and even more significant in the IBD5 “risk” homozygotes, supporting the theory of an epistatic relation between the IBD5 locus and CARD15 in the susceptibility to UC (table 1). However, there was no such relationship between UC and the 908Arg or Leu1007fsinsC CARD15 alleles (908Arg: HC 0.65%, UC 0.73%, _P_=.73; Leu1007fsinsC: HC 2.14%, UC 0.74%, _P_=.057), despite IBD5 stratification (data not shown). We found no particular UC phenotype (need for surgery, age at onset, disease distribution) associated with the CARD15 702Trp allele, with or without IBD5 stratification.

Table 1.

Case-Control Analysis of CARD15 702Trp Allele Association with UC, Stratified by IBD5 Status

We believe that these data confirm the presence of an IBD5/CARD15 epistatic relationship in the susceptibility to UC (although the overall population effect is relatively small [table 1]), in contrast to the non-epistatic relationship between IBD5/CARD15 and CD seen in both our populations and that of Giallourakis et al. (2003). Our data, however, suggest that this epistatic effect is seen exclusively with the 702Trp variant, supporting other data that imply that the 702Trp polymorphism may possess unique properties not shared with the other CARD15 CD-associated variants (Bonen et al. 2003; Rahman et al. 2003; Sugimura et al. 2003). Indeed, our data suggest a trend toward undertransmission/reduced allele frequency of 908Arg/Leu1007fsinsC in UC (table 2 and the paragraph above). Further work is needed to determine whether the epistatic IBD5/CARD15 interaction in UC is a “global” CARD15 phenomenon or, as we have suggested, is restricted to the 702Trp allele. These data support the theory that UC and CD are related polygenic conditions that share some, but not all, susceptibility genes.

Table 2.

Transmissions versus Nontransmissions in Common CD-Associated CARD15 Variants from the TDT of 244 UC Trios

Acknowledgments

This work was supported by the British Medical Research Council (LINK Grant, D.Mc.G. and D.v.H.), the Digestive Disorders Foundation/Belmont Trust (D.Mc.G.), the International Organization for the Study of Inflammatory Bowel Disease (D.Mc.G.), the National Association for Colitis and Crohn’s Disease (T.A.), the Broad Foundation (T.A.), the Hamanako Symposium (K.N.), and Oxagen Ltd.

Electronic-Database Information

Accession numbers and URLs for data presented herein are as follows:

1. Online Mendelian Inheritance in Man (OMIM), http://www.ncbi.nlm.nih.gov/Omim/ (for Crohn disease, ulcerative colitis, IBD5, and CARD15 [NOD2])

References

1. Armuzzi A, Ahmad T, Ling KL, De Silva A, Cullen S, Van Heel D, Orchard TR, Welsh KI, Marshall SE, Jewell DP (2003) Genotype-phenotype analysis of the Crohn’s disease susceptibility haplotype on chromosome 5q31. Gut 52:1133–1139 10.1136/gut.52.8.1133 [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Bonen DK, Ogura Y, Nicolae DL, Inohara N, Saab L, Tanabe T, Chen FF, Foster SJ, Duerr RH, Brant SR, Cho JH, Nunez G (2003) Crohn’s disease-associated NOD2 variants share a signaling defect in response to lipopolysaccharide and peptidoglycan. Gastroenterology 124:140–146 10.1053/gast.2003.50019 [DOI] [PubMed] [Google Scholar]
3. Giallourakis C, Stoll M, Miller K, Hampe J, Lander ES, Daly MJ, Schreiber S, Rioux JD (2003) IBD5 is a general risk factor for inflammatory bowel disease: replication of association with Crohn disease and identification of a novel association with ulcerative colitis. Am J Hum Genet 73:205–211 [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Mirza MM, Fisher SA, King K, Cuthbert AP, Hampe J, Sanderson J, Mansfield J, Donaldson P, Macpherson AJ, Forbes A, Schreiber S, Lewis CM, Mathew CG (2003) Genetic evidence for interaction of the 5q31 cytokine locus and the CARD15 gene in Crohn disease. Am J Hum Genet 72:1018–1022 [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Negoro K, McGovern DP, Kinouchi Y, Takahashi S, Lench NJ, Shimosegawa T, Carey A, Cardon LR, Jewell DP, van Heel DA (2003) Analysis of the IBD5 locus and potential gene-gene interactions in Crohn’s disease. Gut 52:541–546 10.1136/gut.52.4.541 [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Rahman P, Bartlett S, Siannis F, Pellett FJ, Farewell VT, Peddle L, Schentag CT, Alderdice CA, Hamilton S, Khraishi M, Tobin Y, Hefferton D, Gladman DD (2003) CARD15: a pleiotropic autoimmune gene that confers susceptibility to psoriatic arthritis. Am J Hum Genet 73:677–681 [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Sugimura K, Taylor KD, Lin YC, Hang T, Wang D, Tang YM, Fischel-Ghodsian N, Targan SR, Rotter JI, Yang H (2003) A novel NOD2/CARD15 haplotype conferring risk for Crohn disease in Ashkenazi Jews. Am J Hum Genet 72:509–518 [DOI] [PMC free article] [PubMed] [Google Scholar]