|Home | About | Journals | Submit | Contact Us | Français|
An abnormal function of the intestinal barrier has been found not only in patients with inflammatory bowel disease (IBD) but even in their healthy relatives, suggesting that this condition may precede disease onset by years.1 A genetic alteration in the intestinal permeability has also been proposed to exist in patients with coeliac disease. In support of this proposal, polymorphisms in the MYO9B gene (the gene for myosin IXb involved in cytoskeleton remodelling) were found to be associated with increased susceptibility to coeliac disease.2
The MYO9B gene has recently been investigated in relation to IBD and produced discordant results. No association was observed in a Norwegian population,3 but shortly afterwards an international collaboration group performed a statistically powerful study on samples collected from the UK, Netherlands, Canada and Italy in which MYO9B was found to be associated with IBD, and with ulcerative colitis and Crohn's disease considered separately in some populations.4 In that study, a stronger effect was seen in ulcerative colitis than in Crohn's disease. Our aim in the present study was to evaluate the described MYO9B associations in a large sample of Spanish patients with IBD.
We performed a case‐control study of 627 patients with ulcerative colitis and 677 with Crohn's disease recruited from three Spanish hospitals; 990 blood donors of the same ethnicity were used as controls. Written informed consent was obtained from all subjects and ethical approval for the study was obtained from the ethics committees of the hospitals. The diagnosis of ulcerative colitis and Crohn's disease was based on standard clinical, radiological, endoscopic and histological criteria. Clinical data from essentially the same cohort can be found in a previous report.5
Table 11 shows the distribution of MYO9B polymorphisms in patients with ulcerative colitis and Crohn's disease and in healthy controls. All single nucleotide polymorphisms (SNPs) conformed to Hardy‐Weinberg predictions in our control sample. In all cases a strong association was seen when patients with ulcerative colitis were compared with healthy controls. However, the association with Crohn's disease was almost negligible. Moreover, when patients with ulcerative colitis were compared with those with Crohn's disease, significant differences were observed. Stratification of the patients by clinical characteristics did not show differences between the groups.
Table 22 shows the MYO9B haplotype distribution estimated by the expectation‐maximisation algorithm. Only the most strongly associated SNPs, in tight linkage disequilibrium as shown in the previous exhaustive research in coeliac disease susceptibility, were included in our study. It is therefore not surprising that the haplotypes show results which closely mirror those found with individual polymorphisms. In particular, the rs2305767G allele is an almost perfect marker of the first haplotype (GCG) and rs1457092A allele of the second haplotype (AAA). The AAA haplotype confers a strong predisposition to ulcerative colitis compared with the GCG haplotype (p=0.0001).
This study shows a clear association of MYO9B polymorphisms with ulcerative colitis in the Spanish population but not with Crohn's disease. This lack of association does not seem to be derived from a low statistical power (68% for rs1457092 with OR 1.20), because the size of our Crohn's disease sample was high enough to show a strong difference with ulcerative colitis. A lower association has previously been observed with Crohn's disease than with ulcerative colitis, and no association with Crohn's disease was found in the Canadian/Italian sample.4
The diverse results found in the previous studies are difficult to explain. It may be that the low intrinsic OR renders most of the studies underpowered and an association is found only with luck. Another explanation might be the heterogeneous nature of the disease. There is also no easy explanation for the strikingly different susceptibility to Crohn's disease and ulcerative colitis. To our knowledge, no report has claimed that there is a difference in barrier function between the two diseases. However, this is not the first time that genetic differences have been reported between these two forms of IBD.6 Since twin concordance rate data for ulcerative colitis suggest that the heritable component is less important than for Crohn's disease, it has been proposed that environmental factors have a stronger impact in ulcerative colitis.7 It is therefore intriguing that the MYO9B gene, which seems specifically to affect susceptibility to ulcerative colitis, is probably involved in loss of tolerance to environmental agents. However, functional analyses are needed to explore further the role of this gene in ulcerative colitis and other diseases.
The authors thank Carmen Martínez Cuervo for her expert technical assistance.
This work was supported by grants SAF2003‐08522 and SAF2006‐00398. AM has a FIS contract (CP04/00175) and EU works for the “Fundación para la Investigación Biomédica‐Hospital Clínico San Carlos”.
Competing interests: None.