Investigation of the phenotypic and genotypic determinants of disease susceptibility and progression in Crohn’s Disease
Phillips, Anne Mairead
The inflammatory bowel diseases (IBD), encompassing Crohn’s disease (CD) and ulcerative colitis (UC), are chronic inflammatory disorders of the gastrointestinal tract. Their aetiology is not fully understood but is thought to be a combination of the effect of environmental factors in a genetically susceptible person. The work presented is an examination of the phenotypic characteristics of CD in the Scottish population, and an investigation into genetic factors that may influence susceptibility and progression. An IBD cohort from Dundee was recruited (CD=367, UC=265), and extensive phenotypic information collected from these patients together with genomic DNA. Together with the Edinburgh CD cohort already established, the total CD population (n=1155) was examined for time to disease progression (stricturing and/or penetrating disease, according to the Montreal classification) and first resection; a multivariate analysis was performed for factors influencing these outcomes. In this Scottish CD population, the median time to disease progression and first resection was 14.2 years and 8.9 years respectively. The major factor influencing risk of resection and disease progression was disease location, with patients having pure ileal (L1) disease or ileocolonic (L3) disease being more susceptible than those with pure colonic (L2) disease. Compared with L2 disease, the hazards ratios (HR) for disease progression were 4.7 and 2.8, and risk of resection 5.2 and 2.6 for L1 and L3 disease respectively. Disease progression and risk of resection are surrogate markers of disease severity. To try to better understand the determinants of severe disease, a novel score for disease severity was developed and applied to the Dundee CD cohort. This composite score encompassed the variables of medical and surgical management, disease behaviour and location, nutritional status as well as hospitalisations, with a total score that could range from 1 to 16. A score of 7 or more was found to define the 50% of patients with the most severe disease. This cut-off was used to divide patients into less severe and more severe categories; phenotypic and genetic factors were examined for correlation with more severe disease. Genetic factors examined were the 32 most significant CD susceptibility single nucleotide polymorphisms (SNPs) uncovered by recent genome-wide association scans (GWAS). Factors correlated with more severe disease included disease location (L1, odds ratio (OR) 2.20, p=0.0025), age group at diagnosis (p=0.0004) and two CD susceptibility SNPs (rs9286879 and rs17582416; p=0.0085 and p=0.045 respectively). NOD2 was the first IBD susceptibility gene identified. In order to further define pathways involving NOD2, a yeast two-hybrid screen in our laboratory using NOD2 cDNA as the bait had already identified an interaction between NOD2 and UDP-Nacetyl- alpha-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase (GALNT2). This enzyme is involved in O-glycosylation, important in the post translational modification of mucins. A GALNT2 genotype/phenotype analysis on the Edinburgh IBD population was completed, with the Dundee IBD population used as a replication cohort. In the Edinburgh IBD population, the GALNT2 tagging SNP rs7536663 was associated with CD susceptibility (OR 1.38, p=0.0008 vs controls), but replication was not achieved in the Dundee cohort (p=0.469). There was no association of any of the GALNT2 SNPs with UC. The GALNT2/NOD2 interaction was further investigated by completing coimmunoprecipitation between the two genes to characterise the level and type of interaction. An interaction between GALNT2 and NOD2 was confirmed in mammalian cells, with the interaction being at the N-terminal end of the NOD2 protein. GALNT2 expression in a cell line and biopsies was investigated by quantitative polymerase chain reaction and immunohistochemistry respectively. There were no statistically significant changes in GALNT2 or NOD2 mRNA expression in the LS174T cell line after stimulation with specific ligands for NOD2 and GALNT2. GALNT2 protein expression was characterised in intestinal biopsy samples to be predominantly in the lamina propria, with some expression in the enterocytes. To further define the contribution of mucin genes to IBD susceptibility, tagging SNPs across the MUC2, MUC3A and MUC19 genes were genotyped in the Edinburgh IBD cohort and examined for a link with IBD, CD and UC susceptibility, but associations were not found. In view of the strong association with CD susceptibility of a SNP near the MUC19 locus in a recent GWAS, tagging SNPs across the leucine rich repeat kinase-2 (LRRK2) gene, near the MUC19 gene, were also genotyped and examined in the Dundee cohort for an association with IBD, CD and UC susceptibility, but was also negative when corrected for multiple testing. The studies presented allow an improved understanding of the influence of phenotypic characteristics on disease progression, need for surgery and severity in CD. The role of disease location has been determined to be particularly critical, in keeping with other published studies. A detailed examination of the influence of specific genes on disease susceptibility has failed to definitely demonstrate an association between germline variation in GALNT2, MUC2, MUC3A, MUC19 or LRRK2 and IBD, CD or UC susceptibility. An interaction in mammalian cells between NOD2 and GALNT2 has been shown, but further work is required to demonstrate that this is a biologically relevant interaction.