Diet, Gut Microbiota, and Butyrate: Investigating the Epigenetic Mechanism that Prevents Colorectal Inflammation and Tumorigenesis

<p>Long-Term Goals, Central Hypothesis, and Supporting ObjectivesWhether dietary fiber protects against colorectal cancer (CRC) has been controversial because of conflicting results from human epidemiologic studies. However, these studies and mouse models of CRC have not controlled the composition of gut microbiota, which ferment fiber into short-chain fatty acids such as butyrate. Butyrate is noteworthy because it has energetic and epigenetic functions in colonocytes and tumor-suppressive properties in CRC cell lines. We recently developed a gnotobiotic mouse model of CRC colonized with wild-type or mutant strains of a butyrate-producing bacterium to demonstrate that fiber does have a potent tumor-suppressive effect but in a microbiota- and butyrate-dependent manner. The next challenge is to characterize the mechanism of tumor suppression. Our overarching hypothesis is that the fiber-microbiota-butyrate axis epigenetically regulates the expression of genes involved in cell proliferation, apoptosis, and inflammation in both neoplastic colonocytes and immune cells within the tumor microenvironment. Although butyrate is a well-known histone deacetylase (HDAC) inhibitor, the target genes important for tumor suppression in vivo have yet to be identified. It also is not known whether the source of dietary fiber is a significant factor or if other bioactive food components can interact with fiber and butyrate to increase the efficacy of tumor suppression.</p><p> To address these gaps in our knowledge and achieve the objectives of this application, we propose an experimental plan that includes the following Aims:</p><p> Aim 1: Characterize fiber-microbiota-butyrate induced changes in histone acetylation and gene expression during tumor suppression. Germfree mice will be colonized with specific gut microbiota, including either a wild-type or mutant strain of butyrate-producing bacterium, while maintained in gnotobiotic isolators. These mice will receive control (low-fiber) or high-fiber diets and will be treated with AOM/DSS (azoxymethane/dextran sodium sulfate) to induce colorectal tumors. ChIP-seq will be used to identify histone acetylation profiles for every gene in the genome. H3K9ac and H3K27ac will be analyzed because they are highly enriched at gene promoters and enhancers of transcribed genes, respectively. H3K27ac is also a marker of variant enhancer loci that differ between normal colonic tissue and CRC in human clinical samples. These epigenomic data will be integrated with transcriptome profiles obtained from the same samples using RNA-seq. In addition to normal colonic tissues and colorectal tumors, flow-sorted immune cells within the tumor microenvironment will be analyzed because tumor suppression could be mediated, in part, through decreased inflammation. </p><p>Aim 2: Determine the relative importance of the fiber-microbiota-butyrate axis in anti-inflammatory-mediated tumor suppression. Butyrate is known to have anti-inflammatory effects, and we hypothesize that this contributes to tumor suppression. Gnotobiotic mice colonized with bacteria that ferment fiber into either low or high levels of butyrate will be treated with AOM alone or AOM/DSS to induce colorectal tumors. The former is associated with less inflammation than the latter, and this comparison will provide insight into whether the protective effect is linked to inflammation. Additionally, for the AOM/DSS model, the tumor burden of wild-type mice will be compared to RAG-2- and IL-10-deficient mice because their immune responses are modulated in opposite directions. RAG-2 mutants lack functional T cells and B cells and have a diminished immune response, whereas IL-10 mutants lack a crucial immunosuppressive cytokine and have an aberrantly heighted immune response. The RAG-2-deficient mice will interrogate the function of T cells and B cells in tumor suppression, and this is expected to complement the histone acetylation and transcriptome studies on these cells in Aim 1. IL-10-deficient mice are particularly relevant to our experimental design because IL-10 regulates the inflammatory response of immune cells to commensal bacteria in the colon. Tumors will be characterized histopathologically and using markers of cell proliferation and apoptosis. The immune response will be analyzed by flow cytometry and multiplexed biomarker assays for 50 cytokines/chemokines.</p><p>Aim 3: Evaluate different sources of fiber and omega-3 fatty acids in tumor suppression. First, we will compare two sources of dietary fiber in our gnotobiotic mouse model of CRC. We hypothesize that fructo-oligosaccharides (FOS)/inulin will have a stronger tumor-suppressive effect than wheat bran because the former is fermented by gut microbiota to a greater extent than the latter. Second, we will investigate whether omega-3 fatty acids increase the efficacy of the fiber-microbiota-butyrate axis in tumor suppression. Our gnotobiotic mouse model of CRC will be provided two omega-3 fatty acids, alone or in combination with a high-fiber diet. Based on our preliminary studies in CRC cell lines, our hypothesis is that a diet where fiber/butyrate and docosahexaenoic acid (DHA) are combined will have a stronger protective effect than either bioactive food component alone. Our cell line data suggest that another omega-3 fatty acid, eicosapentaenoic acid (EPA), will not interact with fiber and butyrate to the same extent. To address mechanism, we will assess whether DHA influences butyrate metabolism or if it has a significant effect on inflammation. The former will be accomplished by measuring intracellular butyrate concentrations by LC-MS, while the latter will be investigated by flow cytometry and multiplexed biomarker assays for 50 cytokines/chemokines. </p>