The colonic microbiome has been implicated in colorectal cancer (CRC) pathogenesis but the exact mechanisms underlying these observations remain incompletely understood. We have observed striking associations between specific host microbes and aberrant DNA methylation in CRC. For example, Fusobacterium species are substantially enriched in cancers affected by the CpG Island Methylator Phenotype (CIMP). Preliminary data based on sequencing and qPCR validation also show enrichment of bacteria that have been linked to disease in humans or mouse models (E.coli sp., Klebsiella sp. etc.). High levels of these pathogenic bacteria are associated with recurrences in CIMP+ colon cancers. This unexpected link between colonic microbiota and epigenetic control was also seen in an analysis of DNA methylation genome wide in the colonic mucosa of germ free (GF) mice compared to conventionalized mice, where we found that reintroduction of bacteria led to hypermethylation of normally unmethylated CpG island sites (the main anomaly seen in CIMP+ cases). Thus, our preliminary data support a new hypothesis, that the GI microbiota affects colonic neoplasia through inducing or modulating aberrant DNA methylation and epigenetic control. Mechanistically, we propose that multiple parallel mechanisms may be contributing to this link including DNA damage associated recruitment of silencing complexes, and metabolic disturbances whereby bacteria secrete metabolites and/or toxins that diffuse into colonic epithelial cells and affect DNA methylation directly (e.g. 2-hydroxyglutarate [2-HG], which inhibits the TET DNA demethylase enzymes) or indirectly (e.g. butyrate, which is known to modulate epigenetics through inhibition of histone deacetylases). To test these hypotheses, we propose three specific aims: (1) Define the microbiome across the spectrum of CIMP+ tumors. We will use 16S RNA genomic sequencing in an extensive tumor (cancer, precursors and adjacent normal) set simultaneously characterized for CIMP, mutations and gene expression. (2) Impact of CIMP+ associated bacteria on tumorigenesis and DNA methylation in mice. GF Il10-/-;Apcmin/+ mice will be colonized with bacterial candidates (e.g. E.coli, F.nucleatum, K.pneumonia) and tumor incidence, severity, survival as well as DNA methylation and gene expression in normal and tumor tissues will be evaluated. We will also test whether drugs targeting DNA methylation are effective in prevention of bacteria-associated tumorigenesis. (3) Study metabolites by which bacteria influence DNA methylation profiles. We will use metabolomics on cultures of bacteria associated with CIMP and on lysates from CIMP+ and CIMP- cancers to identify metabolites that potentially modulate DNA methylation. These (e.g. 2HG, butyrate) will be tested for effects on DNA methylation (in cell culture and in GF mice) and tumorigenesis (in mice). The proposed research tests a new mechanism for microbiome-associated tumorigenesis and has important implications for detection, prevention and treatment of CRCs.

Public Health Relevance

The bacteria and other microorganisms that normally reside in the human gastrointestinal tract (the microbiome) are known to contribute to colon cancer development through mechanisms that remain incompletely understood. Based on preliminary experiments, we have developed the new hypothesis that specific components of the microbiome contribute to colon cancer development by triggering abnormal DNA methylation, an epigenetic process that controls gene expression long term. This grant will test this hypothesis by discovering those microorganisms specifically enriched in cancers that display profound DNA methylation changes (CpG Island Methylator Phenotype) and measuring their effects on tumor formation, DNA methylation and gene expression in mouse models; the proposed research has important potential implications for early detection, prevention and treatment of colorectal cancers.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Research Project (R01)
Project #
Application #
Study Section
Cancer Genetics Study Section (CG)
Program Officer
Daschner, Phillip J
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Coriell Institute for Medical Research
United States
Zip Code
Good, Charly Ryan; Panjarian, Shoghag; Kelly, Andrew D et al. (2018) TET1-Mediated Hypomethylation Activates Oncogenic Signaling in Triple-Negative Breast Cancer. Cancer Res 78:4126-4137
Zhang, Hanghang; Pandey, Somnath; Travers, Meghan et al. (2018) Targeting CDK9 Reactivates Epigenetically Silenced Genes in Cancer. Cell 175:1244-1258.e26