Metagenomic description of the whole gut microbiome was recently provided (22) and attempts are made to establish correlations between variations in microbiome composition and the occurrence of diseases. Beside this correlative approach, it is essential to develop analytical metagenomics, particularly in the context of niches of the gut that are of major functional importance, such as the crypts that account for epithelial regeneration. In a crypt-specific core microbiota may reside important elements of crypt homeostasis, and in its dysbiotic variations may reside pathologies, particularly IBDs and cancer. We are applying a cellular microbiology approach to decipher the cross-talks established between the microbiota and the gut mucosa. This approach combines bacterial genetics and (meta)genomics, cell and tissue biology with a strong focus on stem cells and epithelial homeostasis. Some of our major ongoing projects are summarized below:
1 – Identification of the core of genes necessary to Lactobacillus casei to survive on and colonize gut surfaces.
The bacterial factors that enable commensal bacteria to maintain in the gut and thus to benefit the host are not known. To identify the bacterial genes that contribute to this persistence, in collaboration with the group of Jean-François Cavin (ENSBANA, Dijon), we set up reverse genetics of Lactobacillus casei and in vivo screening for bacterial persistence in the ileum. 70 mutants were shown to be deficient for ileal establishment. Among the major affected pathways are: cell wall maturation and turn over, sugar transport, carbohydrates metabolism, biosynthesis of selected aminoacids such as methionine, cysteine and asparagine, and regulatory functions (23). We are exploring the pathways highlighted by this screening and analyzing the respective role of the host itself and its resident microbiota in the modulation of L. casei establishment in the gut. Following the complex set up of a dedicated procedure (24), we have sequenced the entirety of our initial mutants library (i.e. ca 10,000 transposon mutants), thereby allowing the ordering of a fully annotated library of 1,100 mutants in individual orfs. This library can now be used for other purposes (i.e. analysis of probiotic functions, impact on host metabolism, etc…).
2 – Identification of a « crypt-specific core microbiota, CSCM » in the murine intestine. In an attempt to explore the microbial content of colonic crypts of the mouse gut that contain the stem cells, thereby accounting for epithelial regeneration, we proceeded to targeted molecular microbial diagnostic. We showed that only the caecal and colonic crypts harbor a living resident flora in the mouse. CSCM is unexpectedly dominated by strictly aerobic genus such as Acinetobacter, Delftia and Stenotrophomonas (25). We suggest that the CSCM plays both a protective and a homeostatic role in the crypt, thus severe consequences may result from its displacement by a more aggressive genus (i.e. pathobionts). We are currently characterizing the CSCM in human colonic crypts and its variations in pathological situations such as colon cancer.
3 – A microbiota-dependant pathway of cytoprotection of intestinal adult stem cells. The beneficial effect of the microbiota on gut epithelial homeostasis occurs, in part, through direct sensing of bacterial products by host receptors, including Toll-like receptors (TLRs) and Nod receptors. The intestinal crypt harbors adult stem, Paneth, and dividing epithelial cells, which altogether support epithelial regeneration. The crypt could indeed be a major site of interaction between diffusing bacterial products and the host, thus offering a unique potential for the microbiota to influence the pathways of regeneration. Making use of the ex vivo method of crypt culture developed by the group of Hans Clevers (26), we have demonstrated that murine crypt stem cells expressed high levels of Nod2, and that upon elicitation by muramyl-dipeptide (MDP) they showed a pattern of increased survival, particularly in the presence of a toxic stress such as the administration of the DNA intercalating agent doxorubidine (27). We are decrypting the cross talks between bacterial MAMPs (microbial associated molecular patterns), particularly the cytoprotective role of MDP, its molecular bases that seem to encompass triggering of autophagy, and possible enhancement by chemical changes on the MDP scaffold. We will also study the possible role of endotoxin, its hexacylated form showing cytotoxic activity to stem cells (Nigro et al., unpublished). We will experimentally address this novel angle to understand the pathogenesis of Crohn’s disease in relation to Nod2 mutations, hypothesizing that the pathogenic scheme may now integrate delayed epithelial repair following an event of crypt “aggression”.
4 – Can crypt dysbiosis account for oncogenesis of the colon ?
Our data are opening the way to important developments, such as substitution of pathobionts to the CSCM, thereby facilitating the occurrence of colorectal cancer. Our underlying hypothesis is that colon cancer being a “stem cell cancer”, only stem cells, due to their extended life span, can recapitulate sufficient microbial signals to accumulate mutations and epigenetic modifications participating to oncogenesis. We attempt to identify the existence of a CSCM associated with colonic crypts from healthy tissues, polyps and tumors in human biopsy and colectomy samples. We expect in polyps and tumors a crypt microbiota that strongly differs from the CSCM observed in healthy subjects, possibly marked by the presence of pathobionts such as Streptococcus gallolyticus, a species epidemiologically associated with the occurrence of colon cancer (28), or others. Taking S. gallolyticus as a model, the culture of primary human colon epithelium from crypts allows us to study the direct effect of putatively oncogenic bacteria on intestinal stem cells.
We are also developing an experimental approach in the mouse encompassing a model of oncogenic dysbiosis based upon Streptococcus gallolyticus gut colonization, and on the direct interaction of either S. gallolyticus or intestinal secretions of S. gallolyticus-colonized gut on organoids and purified stem cells. We analyze the direct genotoxic stress induced by S. gallolyticus on stem cells and its effect on DNA repair pathways, and attempt to define the epigenetic modifications that are likely to appear in stem cells undergoing these interactions.
5 – Deciphering the regulation of expression of epithelial antimicrobial peptides (AMP).
Our demonstration that pathogens like Shigella can suppress AMPs expression was a strong indication of the role played by AMPs in the control of bacterial density and invasive capacities (29). We consequently initiated an in-depth study of the transcriptional regulation of the genes encoding antimicrobial peptides (i.e. hBD1, 2, 3 and cathelicidin LL37). We have deciphered the genetic and epigenetic bases of their regulation and identified cues allowing to disconnect the induction of antimicrobial peptides and inflammatory genes (i.e.: IL-1b, IL-8, CCL20). We have constructed stable reporter epithelial cell lines in which lentivirus bearing transcriptional fusions combining a gene encoding eGFP and the promoter regions of hBD1,2,3 and LL37 encoding genes. We use these reporters to study transcriptional regulation, including genome-wide gene silencing by siRNA, and dissection of epigenetic regulatory mechanisms.
Conclusions and perspectives.
Our project bears on the demonstration that crypt homeostasis depends on signals “emitted” by the microbiota, thereby stressing the depth of our symbiosis with the microbial world. An extension of this paradigm is that loss or subversion of the microbiota-crypt homeostasis may account not only for inflammatory bowel diseases (IBD), but also for colon cancer. This fundamental knowledge is also the basis for translational research, particularly the search for molecules that boost antimicrobial defenses and comfort homeostasis.