부산대학교 도서관

Inflammatory Bowel Disease-associated colorectal cancer is a known and serious complication of Inflammatory Bowel Disease (IBD) affecting the colon. IBD itself is increasing in prevalence and now affects up to 1 in 125 people in the UK. The patients most likely to develop colorectal cancer as a result of IBD are those with a greater severity and extent of colonic disease, with a longer duration of disease. However, relatively little is known about the pathogenesis of IBD-associated colorectal cancer in comparison with its sporadic cancer counterpart. We do know, however, that the predominant gene altered in these cancers is p53, rather than Adenomatous polyposis coli (Apc), and that mutations in Dock2 occur in around 10% of IBDassociated colorectal cancers. Dock2 is a gene that activates Rac, a pathway involved in proliferation, migration and apoptosis of cells. The aims of my project were twofold. First, I wanted to examine the role of Dock2 in a mouse model of IBD-associated colorectal cancer, hypothesising that loss of Dock2 increases tumourigenesis. Secondly, I wanted to develop a more pathophysiologically-relevant Dextran Sodium Sulphate (DSS) mouse model of colorectal cancer, involving loss of p53 in the colonic epithelium rather than loss of Apc (which is the more commonly represented model in the literature), as well as examining the effect of loss of Dock2 in this p53-mediated model. As hypothesised, we found that loss of Dock2 increases tumourigenesis in an APCmediated model of tumourigenesis in vivo. Strikingly, this increased tumourigenesis is associated with an immune cell infiltrate, an interferon gamma signature, and a significant upregulation of Ido1 (a gene that catalyses the conversion of amino acid tryptophan to kynurenine). In vitro, we showed that epithelial IDO1 is induced by administration of interferon gamma. We have also demonstrated the possible source of this interferon gamma as gamma delta T cells, which are increased in tumours of mice lacking Dock2. Separately, we have shown that epithelial loss of p53 leads to a reliable model of IBDassociated colorectal cancer that accurately reflects human disease, and significantly also models tumour invasion, something that we have not observed in the APCmediated model. Disappointingly however, additional loss of Dock2 does not lead to increased tumourigenesis in this model. My findings give insights into potential mechanisms of increased tumourigenesis after Dock2 loss, as well as providing a novel p53-mediated mouse model of tumourigenesis that can be used in future experiments aimed at understanding mechanisms behind this important complication of IBD. Potentially, this model could be used to identify biomarkers or further explore tumour initiation following p53 loss.