 |
|
|
 |
LettersThe Crohn's Associated NOD2 3020InsC Frameshift Mutation Does Not Confer Susceptibility to Ankylosing Spondylitis To the Editor: Spondyloarthropathies (SpA) are a group of relatively common diseases of unknown etiology, whose manifestation is thought to result from the interaction of strong environmental factors with the genetic background in predisposed individuals. These disorders are characterized by inflammation of the spine and sacroiliac joints and by peripheral arthritis. In addition, SpA often show extraskeletal manifestations like iritis/uveitis, psoriasis, and, remarkably, gut inflammation1. The association between bowel and joint disease has been extensively investigated, and is strikingly exemplified by one of the most common SpA, ankylosing spondylitis (AS): 20% of patients with AS show intestinal inflammation, and inflammatory gut lesions are found on histology in about 60% of cases, even in the absence of any abdominal symptoms. On the other hand, SpA also represent the most frequent extraintestinal manifestation of both ulcerative colitis (UC) and Crohn's disease (CD), the 2 main types of inflammatory bowel disease (IBD)2: in the case of CD, complications due to SpA are generally present in 20% of the patients, and radiological findings can be observed in more than 40% of asymptomatic subjects. CD and AS, or at least their common features, are believed to result from an aberrant response to bacteria, and this finds strong confirmation in animal models, where germ-free conditions prevent the manifestation of both diseases3. Together with the overlap in their clinical features, it is therefore conceivable that CD and AS might share a common genetic predisposing background. In the case of AS, a role for HLA-B27 histocompatibility gene has been well established both in humans and in transgenic animal models4. However, despite extensive research, a clear pathogenetic role for HLA-B27 molecule has not yet been established, and from twin and family studies it is estimated that more than 50% of the AS-predisposing genetic background remains to be identified5. As for CD, a major breakthrough has come from the recent discovery of mutations in the NOD2 gene in a subset of patients with CD6-8. The product of NOD2 gene appears to be involved in the inflammatory response to bacteria via the activation of the nuclear factor-kB (NF-kB) pathway, following its interaction with bacterial components such as lipopolysaccharide. NOD2 is therefore regarded as an intracellular sensor of bacteria, and alterations in its ability to activate NF-kB might lead to unwanted reactions underlying inflammatory diseases9. Thus, based on its function, and the overlapping features of CD and AS, we were prompted to assess whether NOD2 could also directly influence genetic susceptibility to AS. Three NOD2 polymorphisms have been shown to associate with increased susceptibility to CD6-8, the strongest association being observed with an insertion mutation, 3020insC, which determines a premature truncation of the corresponding protein. As a preliminary test of our hypothesis, we focused our analysis on identification of this variant, and a polymerase chain reaction-restriction fragment length polymorphism investigation was set up to characterize all subjects enrolled in this study (details available upon request). Fifty-three patients with AS with no CD manifestation (38 HLA-B27+ and 15 HLA-B27-) and 58 matched healthy controls were recruited and genotyped for NOD2 3020InsC. The results are shown in Table 1. The 3020InsC mutation was present with a frequency of 2.5% in the control sample (3/116 chromosomes), consistent with reports for other populations6-8. In the AS group, only a single instance of 3020insC was found, corresponding to a frequency lower than 1% (1/106 chromosomes). No differences were observed between HLA-B27+ and HLA-B27- patients (data not shown). Thus, the frequency of 3020InsC did not differ substantially in our AS and control groups, and is far from reaching that previously reported in patients with CD (8-16% in different studies6-8).
As a result, our analysis does not support the hypothesis of a pathogenetic role for NOD2 in the development of AS. It is conceivable that the size of our sample only allowed detection of major genetic effects (with an estimated 80% power to exclude an association with relative risk > 5), and extended analysis of larger samples and different ethnic groups will be necessary to conclusively rule out a NOD2 contribution to AS. However, the fact that we found only a single 3020insC mutation out of 106 AS chromosomes (53 patients) strongly suggests that opposite trends are unlikely to be observed in future studies. The virtual absence of 3020insC mutation in AS is not surprising, as lack of association with NOD2 has also been reported for ulcerative colitis6-8, the other common form of IBD and, more recently, for psoriasis10. It is tempting to speculate that this observation might reflect the presence of different pathogenetic mechanisms, or the involvement of non-NOD2 response-eliciting bacteria, as the basis of ankylosing spondylitis, ulcerative colitis, and psoriasis, in contrast to Crohn's disease. MAURO D'AMATO, PhD, Division of Molecular Pathology, Microbiology and Tumor Biology Center-MTC, Karolinska Institute, Nobels väg 16, SE-17177, Stockholm; ROSA SORRENTINO, PhD, Institute of Cell Biology and Development, University of Rome "La Sapienza," Rome, Italy; SVEN PETTERSSON, MD, PhD, Microbiology and Tumor Biology Center-MTC, Karolinska Institute, Stockholm, Sweden. |
