Crohn's disease was mapped to chromosome 16 by Hugot and colleagues in 1996 in the first genome-wide linkage analysis in this disease using affected pedigrees: a total of 25 Caucasian families with at least two affected siblings were analysed using 270 highly polymorphic microsatellite markers across the genome (Hugot et al. 1996). A stepwise approach involving additional families and denser sets of markers resolved a pericentromeric region of chromosome 16, dubbed 'inflammatory bowel disease locus 1' (IBD1). The number of alleles 'shared identical by descent' among affected siblings were analysed, in other words those alleles inherited by both siblings that are copies of the same parental alleles with no assumption about the mode of inheritance. Other investigators subsequently replicated and to some extent refined the region of linkage for IBD1, clearly showing that the linkage was with Crohn's disease and not ulcerative colitis (Ohmen et al. 1996; Cavanaugh et al. 1998; Cavanaugh 2001).
In 2001 evidence was published from different groups resolving a specific gene and sequence variants underlying IBD1 in Crohn's disease (Hampe et al. 2001; Hugot et al. 2001; Ogura et al. 2001a). Hugot and coworkers sought to use a positional cloning strategy, with linkage analysis and linkage disequilibrium mapping: they interrogated the 20 Mb region of IBD1 (a large region thought to contain some 250 genes) with 26 microsatellite markers, analysing two sets of affected families (Hugot et al. 2001). Sequencing of cloned DNA associated with disease resolved a number of SNPs and a specific gene, NOD2, which encodes 'nucleotide-binding oligomerization domain containing 2' protein, whose role in the inflammatory response to bacteria would suggest a highly plausible model for disease pathogenesis. The exonic sequence of NOD2 was determined for 50 unrelated patients with further variants identified. Overall, highly significant associations were found for two nonsynonymous SNPs and a single nucleotide insertion polymorphism, with a gene dosage effect such that highest risks were seen in individuals homozygous for a given variant or having two different variants (compound heterozygotes) (Fig. 9.19A).
In the same issue of Nature that Hugot and colleagues published their results, Ogura and coworkers published a paper in which they identified the same gene and insertion polymorphism as being important in determining Crohn's disease, except that here the approach had been different (Ogura et al. 2001a). Ogura and colleagues had sought to identify novel Nod1 -like genes and found a sequence with high homology on chromosome 16q12, leading to the identification of NOD2, a gene found to be expressed mainly in monocytes, which activated NFkB, and was important to endotoxin responsiveness (Inohara et al. 2001; Ogura et al. 2001b). Given that NOD2 was lying under the linkage peak, Ogura and colleagues then proceeded to investigate it as a candidate gene for Crohn's disease by resequencing the exons and flanking intronic sequence. Among 12 patients with Crohn's disease three were found to have a specific insertion polymorphism in exon 11, the same variant identified by Hugot and colleagues. This C nucleotide insertion (rs2066847; also called 3020insC as the insertion is at nucleotide position 3020) alters the encoded amino acid at that codon from leucine to proline (CUU to CCU) and results in a frameshift such that the next codon is a stop codon (UGA) (Fig. 9.20). This truncates the NOD2 protein by 33 amino acids and, as in the other study, the variant was significantly associated with Crohn's disease but not ulcerative colitis, being preferentially transmitted from heterozygous parents to affected children and, in three Caucasian cohorts, being found at a significantly higher frequency among cases of Crohn's disease (8.2%) versus controls (4%) (Ogura et al. 2001a).
A further study by Hampe and colleagues, also published in the same year, identified the same insertion polymorphism among British and German affected sib pairs after adopting a candidate gene approach (Hampe et al. 2001). In a meta analysis published in 2004, Economou and colleagues found that from 39 studies of Caucasians of non-Jewish descent, the three NOD2 variants identified by Hugot and others were all significantly associated with Crohn's disease. Possession of one of the high risk alleles was associated with a 2.4 (95% CI 2-2.9) increased odds of disease compared to individuals without the alleles, rising to 17.1 (10.7-27.2) for carriers of at least two of the risk alleles (Fig. 9.19B) (Economou et al. 2004). Subsequent genome-wide association studies repeatedly demonstrated association at this locus, showing the NOD2 locus was one of the most strongly associated genomic regions in Crohn's disease susceptibility (Box 9.5).
Are these NOD2 alleles functionally important? Intensive research has established that NOD2 is critical to the innate and acquired immune response through its role as an intracellular protein responsible for detecting bacteria and inducing a proinflammatory response (Inohara et al. 2005; Kobayashi et al. 2005; Watanabe et al. 2005; Strober et al. 2006). NOD2 is a member of a family of pattern recognition receptors and comprises a central nucleotide binding domain (NBD), N-terminal caspase recruitment domains, and a leucine-rich repeat (LRR) region (see Fig. 9.20). The three genetic variants associated with Crohn's disease discussed to date are found in sequence encoding the LRR, a region responsible for bacterial recognition, specifically recognizing a short motif called muramyl dipeptide (MDP) found in peptoglycans in bacterial cell walls. On recognizing MDP,
Number of chromosomes
Was this article helpful?