nificant difference was found between UC patients and control group (P > 0.05). In CD and UC patients, no significant association with G908R variant was found. L1007finsC SNP showed an association with CD (9.8%) compared with controls (2.9%, P = 0.002) and UC patients (2.3%, P = 0.01). Moreover, in CD patients, G908R and L1007finsC mutations were significantly associated with different phenotypes compared to CD wild-type patients. No association of IBD with the TLR4 SNPs was found in either cohort (allele frequencies: D299G-controls 3.9%, CD 3.7%, UC 3.4%, P > 0.05; T399I-controls 2.9%, CD 3.0%, UC 3.4%, P > 0.05).
CONCLUSION:These findings confirm that, in our IBD patients selected from Southern Italy, the NOD2/ CARD15, but not TLR4 SNPs, are associated with increased risk of CD. Abstract AIM: To evaluate the role of genetic factors in the pathogenesis of Crohn's disease (CD) and ulcerative colitis (UC), we investigated the single nucleotide polymorphisms (SNPs) of NOD2/CARD15 (R702W, G908R and L1007finsC), and Toll-like receptor 4 (TLR4) genes (D299G and T399I) in a selected inflammatory bowel disease (IBD) population coming from Southern Italy. METHODS: Allele and genotype frequencies of NOD2/ CARD15 (R702W, G908R and L1007finsC) and TLR4 (D299G and T399I) SNPs were examined in 133 CD patients, in 45 UC patients, and in 103 healthy controls. A genotype-phenotype correlation was performed. RESULTS: NOD2/CARD15 R702W mutation was significantly more frequent in CD (9.8%) than in controls (2.4%, P = 0.001) and in UC (2.3%, P = 0.03). No sig-