Genomics of Coeliac Disease - Molecular Signatures of the Pathogenesis

Abstract

Coeliac disease, the intolerance for dietary gluten common in Western populations, is a multifactorial disorder, meaning that it is caused by the interaction of environmental factors and multiple susceptibility genes. The aim of this thesis was to gain insight into the pathogenesis by means of a genomics approach. We carried out integrated genetic analysis and gene expression studies to identify, and gain better understanding of genes and mechanisms behind the pathology and etiology of coeliac disease. In the introducing chapter we review genetic studies on coeliac disease and introduce new methods for gene expression profiling. First, we examined whether tissue heterogeneity (‘patchiness’) may have an effect on gene expression measurements performed on duodenal biopsies. We observed that mucosal inflammation and differentiation (represented by IFNG and TM4SF4 gene expression, respectively) were inversely correlated, depending on the extent of mucosal restructuring. However, the extent of variance in gene expression measurements was similar in cases and controls, and thus independent of lesion severity. Consequently, we proposed a model where gene expression mosaicism in the duodenum might result into tissue heterogeneity (‘patchiness’). Second, we applied microarrays to study genome-wide gene expression in the duodenums of coeliac disease patients. Alterations induced by the pathology pointed to enhanced cell proliferation in the mucosa and arrested terminal differentiation of enterocytes. This impaired differentiation affected the uptake and processing of lipids, sterols, sugars, peptides, and iron. These deficiencies contribute to the wide variety of clinical features in coeliac disease. We also observed an impaired detoxification system, previously also described for inflammatory bowel disease (IBD). This suggests that, like coeliac disease, IBD is also characterized by a differentiation defect. Third, we analyzed the tight junction gene network using genetics and gene expression profiling. The expression pattern of genes of the claudin family appeared to have been evolutionary conserved for over 75 million years. Three pairs of claudins showed tight co-regulation. Patients showed variable patterns of claudin gene expression marked by extreme outliers. Particularly genes involved in signal transduction and regulation of the cytoskeleton showed trends of changed gene expression. Genetic association analysis of this gene network was based on the use of single nucleotide polymorphisms (SNPs) and information on linkage disequilibrium from the International HapMap Consortium. We identified two tight junction genes that were associated with coeliac disease, and to a lesser extent with IBD. This suggested that both gastrointestinal disorders, coeliac disease and IBD share a genetic barrier defect. Fourth, we examined candidate genes and gene families in the inflammatory and immune pathways. Both the IFNG gene (pivotal in the Th1 adaptive response), and SPINK4 (expressed by goblet cells) showed differential expression that correlated well with the extent of mucosal remodeling. Weak genetic association was only observed for INFG in the Dutch population, but not for the positional candidate genes SPINK1, -2, -4, and –5. Finally, in the General Discussion, we evaluate our approach and our interpretations of the data, which are than incorporated in possible models of the pathogenesis of coeliac disease.