Development, validation and implementation of an in vitro model for the study of metabolic and immune function in normal and inflamed human colonic epithelium.

Ulcerative colitis (UC) and Crohn's disease (CD), collectively referred to as inflammatory bowel disease (IBD), are chronic immune disorders affecting the gastrointestinal tract. The aetiology of IBD remains an enigma, but increasing evidence suggests that the development of IBD may be triggered by a disturbance in the balance between gut commensal bacteria and host response in the intestinal mucosa. It is now known that epithelial cells have the capacity to secrete and respond to a range of immunological mediators and this suggests that these cells play a prominent role in the pathogenesis of IBD. Current knowledge about the intestinal epithelium has mainly been obtained using models based on animal cells, transformed human intestinal cell lines and isolated cells from resected colonic bowel segments. Species difference, malignant origin and confounders related to surgery, obviously make these cell models however less applicable for patophysiological studies. Consequently, there was a clear need for models of representative intestinal epithelial cells that would allow functional and dynamic studies of the differentiated human colonic epithelium in vitro. The primary purpose of this thesis was to explore and validate the optimal conditions for establishing a model based on short-term cultures of human colonic epithelial cells obtained from endoscopical biopsies. The cell cultures were accordingly used to describe the interplay between proinflammatory cytokines and colonic epithelium, with focus on alterations in viability, butyrate metabolism and secretion of a chemokine and metalloproteinases (MMP). Finally, the model was used to characterize expression and activation of receptors like toll like receptor (TLR)9 and peroxisome activated proliferators (PPAR)- known to be important players in regulation of innate and adaptive immune responses in human colonic epithelium. The results showed that it is possible to establish short-term cultures of representative, viable human colonic epithelial cells from endoscopic mucosal biopsies of patients with IBD. Short-time isolation by EGTA/EDTA from colonic biopsies allowed establishment of small scale cultures of epithelial cells which were viable and metabolic active for up to 48 hours in vitro. The cell model preserved important cellular metabolic and immunological functions of the human colonic epithelium, including the ability to oxidate butyrate, detoxificate phenolic compounds and secrete the chemokine interleukin (IL)-8 in vitro. Tumour necrosis factor (TNF)-α and interferon (IFN)-γ are pro-inflammatory cytokines, which are present in increased amounts in inflamed colonic mucosa. The precise mechanisms of cytokine-mediated mucosal injury are unknown, but one might be that TNF-α and IFN-γ directly impair epithelial cell function similar to effects seen on distinct target cells in other autoimmune diseases. Using the model, both cytokines were found directly to impair the viability of colonic epithelial cells and to induce secretion of IL-8 in vitro. Interestingly, the cells from inflamed IBD mucosa were less sensitive to cytokine-induced damage, which suggests that an intrinsic defense mechanism is triggered in these cells, perhaps as a result of exposure to toxic luminal factors or high local cytokine levels in vivo. TNF-α and IFN-γ may also be involved in regulation of intestinal inflammation through stimulation of MMP expression and proteolytic activity. We found that colonic epithelial cells express a range of MMPs and moreover that expression of distinct MMPs is increased in cells from inflamed IBD mucosa. Using a functional peptide cleavage assay it was shown that epithelial cells secreted proteolytic active enzymes and that the functional MMP activity was increased in inflamed IBD mucosa. This suggests that colonic epithelial cells, like myofibroblasts and immune cells, may contribute to local intestinal mucosal damage, through secretion of active MMPs. Disturbance of recognition and discrimination of potentially harmful pathogens from commensals in the intestinal mucosa have increasingly been implicated in the pathogenesis of IBD. Our results revealed that colonic epithelial cells express TLR9, a key pattern recognition receptor. Interestingly, the differentiated epithelial cells, which have been exposed to the luminal bacterial flora in vivo, were unresponsive to TLR9 ligand stimulation, contrasting findings in the epithelial cell line HT-29 that is cultured continuously in bacteria free environment. These findings suggest, theoretically, that colonic epithelium may regulate immune responses to microbial antigens including commensal bacterial DNA through modulation of the TLR9 pathway. Currently, the results are in line with the emerging view, that the epithelium represents an important frontline cellular component of the innate immune system in the gut. PPARγ is a nuclear receptor involved in the regulation of lipid and carbonhydrate metabolism. Recent studies in rodent colitis models suggest that PPARγ also is involved in modulation of inflammatory processes in the colon. Using the model, we characterise expression and activity of PPARs in human colonic epithelium and, additionally, evaluated the functional significance of a possible imbalanced PPARγ regulation in relation to inflammation. Our experiments showed that colonic epithelial cells express PPARγ and furthermore that PPARγ signalling was impaired in inflamed UC epithelium. It was possible to restore PPARγ signalling in the cell cultures by stimulation with rosiglitazone (a synthetic PPARγ ligand) in vitro. Hence, these experiments prompted us to design a small controlled, clinical study exploring the possible stimulatory effects of rosiglitazone (a PPAR ligand) in vivo. Interestingly, it was found that topical application of rosiglitazone in patients with active distal UC reduced clinical activity and mucosal inflammation similar to the effects measured in patients treated with mesalazine enemas. Moreover, rectal application of rosiglitazone induced PPARγ signalling in the epithelium in vivo, supporting the view that activation of PPARγ may be a new potential therapeutic target in the treatment of UC. Overall, the in vitro model of representative human colonic epithelial cells has shown to be a useful technique for detailed studies of metabolic and immunological functions that are important for homeostasis of the colonic epithelium. Currently, the findings support the view that intestinal epithelial cells actively participate in immunological processes in the colonic mucosa. Additionally, the model seems to be applicable for generating and evaluating new therapeutic approaches from laboratory bench to bed line as illustrated by the PPARγ study. It is therefore probable, that studies in models of representative colonic epithelial cells, as the one described here, could contribute with important knowledge about the pathogenesis of human inflammatory colonic diseases also in the future.