Main questions of the project: How can the interactions with infectious or commensal microbes at a given time point of life induce the modulation of existing chronical situations (infectious or inflammatory) in a given individual? And how, from our results, can we propose therapeutic solutions through innovative nucleic acids based vaccination and induction of specific immune cells by synthetic lipid derivatives of natural compounds?

In the light of the discovery of the new regulatory double positive T cell population (DP8α), which development seems to be strictly linked to the composition of the gut microbiota with all the consequences that the presence or not of such a population has on the course of local inflammation, it is tempting to speculate that such regulatory cells may exist elsewhere in the organism with comparative important role in other kind of immune or inflammatory processes.
To evaluate this hypothesis, preliminary data showed that DP8α cells can be found in another well-known granulomatous inflammatory site: the tuberculous granuloma. We thus analyzed the presence of DP8α within mycobacterial granulomas induced in vitro, and checked whether a mycobacteria can induce the proliferation of these cells. The answer is positive for both questions, mycobacteria-specific DP8α cells exist within tuberculous granuloma and they are specific for BCG antigenic compounds. Interestingly these BCG-induced DP8α share the same surface markers as the previously described gut DP8α (Foxp3 negative). These recent results reinforce our hypothesis that Foxp3-negative regulatory DP8α cells may constitute a novel family of regulatory cells, composed of different sub-populations sharing common functions and cell surface markers, but with punctual differences as a sign of an adaptation to the place where they have to work, or to the kind of immune or inflammatory response they have to modulate.

1. Characterization of the DP8α family

Identification of the components of Faecalibacterium prausnitzii involved in human regulatory T cells modulation
Many data support a major role of F. prau in health and disease: (i) it is a dominant member of the intestinal microbiota in healthy adults and a major representative of Firmicutes phylum; (ii) its abundance is decreased in many inflammatory diseases especially IBD; (iii) it exhibits in vitro and in vivo anti-inflammatory properties, (Miquel Curr Opin Microbio, 16: 1, 2013); (iv) it contributes to the induction of a specific subset of IL10-producing regulatory T cells (Treg) of the human colon characterized by a double positive CD4CD8αα phenotype (DP8α) Sarrabayrousse et al PLoS Biology 12:e1001833, 2014). Moreover, we showed that presentation of the dead bacteria by monocytes stimulated the growth of these Treg. These results suggest that cell-wall-associated or intracellular components could contribute to the immunomodulatory properties of F. prau, as demonstrated previously for pathogenic bacteria.
This project will be done in collaboration with the INRA group of P Langella using the following strategies:
Components of the cell wall of F. prau and an expression library of the bacteria into E. coli, both produced by the INRA group, will be tested for their capacity to stimulate colonic DP8α Treg cell lines upon presentation by antigen presenting cells. The in vivo relevance of the components identified with this approach will then be tested in a mouse DSS colitis model that has been shown to respond to F. prau and its culture supernatant.

Functional characterization of DP8α cells
– Identification of a Microarray signature for F prau-specific DP8α Treg cells: This analysis will be done by comparing the transcriptome of lamina propria lymphocytes (LPL) lines: DP8α and CD4, activated or not. A preliminary study done by Q-PCR with DP8α and CD4 LPL lines from 2 donors has revealed significant differences between these two subsets. We also hope to find out specific biomarkers of these cells that could allow their quantification by Q-PCR or Immunohistochemistry within patient’s biopsies.
– Analysis of the differentiation of colonic DP8α Treg by F prau: naïve CD4 cells will be activated by different types of presenting cells (monocytes, macrophages, dendritic cells) differentiated and/or loaded with F. prau or other bacteria as controls in the presence or not of SCFA, TGF-b and IDO. We already observed an expression of CD8α by a small fraction of naïve CD4 stimulated by monocytes in the presence of F prau and an increase of this expression by butyrate. We will study the TCR specificity of DP8α cells induced and their regulatory phenotype. If DP8α Treg is confirmed the mechanism of their induction by F prau will be studied: by looking to the role of cytokines, PRRs activation, regulation of histone deacetylation (role of butyrate)…
Then molecular biology tools will be used to try to functionally manipulate these cells to either boost their regulatory function in situations where it is useful (inflammatory diseases) or block their regulatory function to limit their modulatory effect on anti-infectious responses for instance.
– Pathophysiological role of DP8α Treg in inflammatory diseases. In collaboration with clinicians we are looking for correlations between the frequency of these cells in blood and the presence and or evolution of inflammatory diseases associated with gut microbiota dysbiosis, especially decreased levels of F prau: colorectal cancers, spondyloarthritis, diabetes, obesity and allergies. Total DP8α cells and F prau-specific ones are being quantified in patients blood respectively by flow cytometry and in vitro analysis of DP8α cell responses to the bacteria.

2. Role of DP8α cells in the modulation of existing infectious and inflammatory responses

The third step to acquire a full description of the role of microbiota induced regulatory T cells in the modulation of distant immune responses, is to analyze their behaviour in natural or in vitro models of inflammation or infection.
The in vitro model of mycobacterial granuloma, developed by our team 10 year ago is used to evaluate how mycobacteria-induced DP8α cells interfere with the host immune response to mycobacteria. Indeed, a special feature of tuberculous granulomas is that despite a “supposed” fully competent immune response, the granulomatous response is able at best to constrain the bacilli, kill some, but not fully eradicate all of them since some acquire a live, dormant stage for sometimes decades within the granulomatous structures. Granulomas thus appear as an equilibrium between the immune response and a bacteria which does not succeed to escape but is not completely eradicated.
The first question that will be asked is whether mycobacteria-induced DP8α are responsible for the fact that the granulomatous response is not 100% efficient to eradicate internalized bacilli. DP8α will thus be removed from total T cell used to form a granuloma, or functionally inactivated when a functional target will have been identified (see above).
Once the role of intragranuloma DP8α will be identified, the following question that we will ask is whether microbiota-induced DP8α, that were shown to have the ability to circulate in blood vessels, may interfere with a mycobaterial granuloma in its classical equilibrium state, and whether such a modulation could explain why the bacteria are able sometimes after several decades of dormancy, to reactivate from these granulomas. In other words, could gut–induced regulatory cells be responsible for a down modulation of the granulomatous response and is this responsible for the switch from asymptomatic stage to clinical tuberculosis in infected individuals?
To test this hypothesis, DP8α will be induced by F. prau from PBMCs of control individuals. PBMCs will then be infected with M tuberculosis until granuloma will form. Then F prau DP8α will be added to the reaction and the progression of M. tuberculosis counts will be analyzed. In the same type of experiments, the very new glycolipid family arising from B. Pitard’s group containing various aminosugars as polar headgroups and linkers linking the polar headgroup and the hydrophobic moieties including various acyl chain lengths or composition, will be assessed in a structure-function relationships for their impact on the granuloma.
On the side of pure inflammatory models, we have shown several years ago that E.coli pathovar purified from Crohn’s disease patients are able to induce a granulomatous response strongly mimicking those found in Crohn’s disease patients, which is not the case of commensal E. coli strains (Meconi et al Cell Microbiology 2007). We more recently showed, in a collaborative study with the group of P. Desreumaux in Lille, that environmental factors such as Alun could alone induce a strong inflammatory granulomatous-like response (Pineton de Chambrun, Mucosal Immunol, 3:589, 2014). To evaluate the role of gut-DP8α in such microbe-induced or chemical component-induced inflammatory granulomatous reactions, and how they can be functionally manipulated to better prevent strong inflammation, we will reproduce them in vitro, and investigate whether DP8α cells are already induced in such granulomas. We will then try to add gut-DP8α, i.e. F prau-induced DP8α, to evaluate their ability to modulate such “pure” inflammatory granulomas.