Research Themes Tumors of the digestive tract Microbiome from association studies towards clinically relevant tools

The presence of various bacterial species is associated with the development of CRC: within the theme we initially focused on S. Gallolyticus (Clin Inf Dis 2011; Cancer Prev Res 2012). Using different techniques and model systems, we have determined that specific virulence features of this bacterium, such as immune evasion by its surface structure and proliferation on tumor metabolites, allow this strain to be strongly associated with CRC (Nat Rev Microbiol 2012; Lancet Inf Dis 2013). We subsequently investigated how lifestyle and dietary factors influenced the microbiome (PLoS ONE 2012; Nutrition and Cancer 2013). Direct mechanisms by which bacteria can cause CRC were investigated in close collaboration with the Sears groups at Johns Hopkins (Clin Inf Dis 2015: the carcinogenic toxin of Bacteroides fragilis is obtained by horizontal gene transfer and specifically binds to the GPR35 receptor on tumor cells (manuscript in preparation). As GPR35 is a risk locus for inflammatory bowel disease, targeting this receptor could be a potential therapeutic strategy. Further therapeutic and diagnostic implications of the microbiome in IBD are currently being investigated in a prospective cohort study, by Boleij on a Veni grant.
Several approaches are used to determine which bacteria are relevant in carcinogenesis and cancer detection building on our previous achievements. We have established a system to test hundreds of naturally occurring gut isolates for their oncogenic effect by using complementary high-throughput gnotobiotic CRC models that are capable of quantifying the full range of possible bacterial influences on CRC, from tumor promotion to inhibition. Next, we will mine the genomes of bacteria with a similar phenotype (CRC promotion/inhibition) for the genes most strongly associated with that phenotype, and use heterologous expression to test if these bacterial genes indeed cause the phenotype. We will then mine publicly available metagenomic sequencing data from the human gut, including CRC patients and healthy controls, for the identified bacterial pathways to assess the involvement of these processes in humans in vivo. Innovation in bioinformatics (metabolome prediction, metagenomic characterization methods) allows rapid progress in this field (Best Prac Res Clin Gastroenterol 2013; Nat Microbiol 2018).
Progressing from single bacterial species towards a more comprehensive view of the microbiome, we investigated bacterial dysbiosis and biofilm formation patients with hereditary CRC. In collaboration with the Sears group, we found that, in patients with FAP, the combination of E Coli and B fragilis was present in the biofilms on the adenomas, and this combination lead to fast tumor onset in animal models (Science 2018). The presence and impact of biofilms in patients with Lynch syndrome is investigated in a large prognostic cohort (current KWF grant). In this cohort we can also explore the potential synergy of microbiome and immune response. Biofilms and their contents are currently being investigated as biomarkers for the improved detection of right-sided CRC precursor’s lesions. Within the population screening program we still miss a significant proportion of these precursors that will subsequently develop into interval cancers. By utilizing the biofilm and potential other microbiome markers we might improve endoscopic detection and prevent CRC. The exploration of the effects of the microbiome on treatment is the next challenge, which will tackled by radiochemotherapy utilizing a combination of patient databases and prospective cohorts.