Pivotal role of the nuclear receptor PPARgamma in colon epithelial cells

Abstract

When symbiosis between the residential intestinal flora and the epithelial cells of the gastrointestinal tract is broken one endpoint may be chronic inflammation, e.g. Inflammatory Bowel Disease (IBD). The disruption of the epithelial lining elicits activation of NF-kappaB and secretion of potent inflammatory mediators. In the case of IBD, elevated levels of NF-kappaB have been observed and interestingly several nuclear receptors have been reported to attenuate NF-kappaB activity. One such receptor is PPARgamma. My thesis presents evidence of a possible molecular mechanism whereby commensal bacteria not only regulate the expression of PPARgamma, but also its ability to export NF-kappaB from the nucleus, independently Of IkappaB, suggesting a novel pathway for nuclear export of NF-kappaB.

In paper I we show that TLR4 signalling can increase expression levels of PPARgamma. Moreover we also show that germ free mice do not express PPARgamma indicating that bacteria can regulate its expression. Interestingly, patients with ulcerative colitis, one of the two main forms of IBD, display a defect in protein levels of PPARgamma in colonic epithelial cells compared to Crohn's disease patients and healthy controls.

In the second paper, we assess the possible anti-inflammatory effects of Bacteroides thetaiotaomicron, a common anaerobic commensal bacteria of the gastrointestinal tract, upon co-infection in CaCo-2 cells with the pathogenic Salmonella enterica. Immunohistochemistry in combination with biochemical and functional experiments revealed that PPARgamma upon exposure to B. thetaiotaomicron, possess the ability to shuttle NFkappaB from the nucleus to the cytosol.

In the third manuscript we extend the observations presented in paper one and assess whether there is a relationship between high levels of NF-kappaB and low PPARgamma levels in the colonic epithelium of IBD patients. No strict correlation was observed between low intestinal levels of PPARgamma and elevated NF-kappaB levels. Individual variations of the absolute NFkappaB levels among IBD patients were however observed In addition, we show that the apoptosis inducing ligand, TRAIL, is expressed at lower levels in inflamed colon mucosa, thereby linking it to chronic inflammation.

A DNA-binding protein may switch between an activator and a repressor, depending on cofactor assembly. This prompted us to assess whether such a factor could be identified for PPARgamma since opposing effects of its function in colonic tumor cells had been reported. One possible factor is the transcriptional co-factor beta-catenin. In the final manuscript we examine the possible interplay between PPARgamma and beta-catenin. Biochemical data show that beta-catenin interacts with PPARgamma which results in elevated protein levels of PPARgamma. Functional data show that beta-catenin and the known chemical agonist of Writ signalling Lithium chloride, can activate a PPARgamma dependent reporter gene in a dose dependent manner. Hence, under conditions where abnormal levels of beta-catenin is observed, e.g. in colon cancer, it is tempting to speculate that the altered beta-catenin levels may change the net function of PPARgamma in favour of cell growth.

In summary, the data presented in this thesis, provide some aspects as to how homeostasis may be tightly regulated by a very limited set of factors and that these factors may be intimately connected to execute multiple functions under stringent control. The fine tuning of these functions can be guided by epigenetic factors, such as microbes and nutrients. The strategic tissue distribution of PPARgamma, and its pleiotropic functions serve as one example of an inflammatory gatekeeper that integrates gut homeostasis and metabolic control.