Diagnosis, prognosis and treatment of a disease related to a decrease of <i>F. prausnitzii</i>

Abstract

The invention relates to a method comprising a step of determining the number, concentration and/or proportion of T lymphocytes with a CD4.sup.+ CD8.sup.low phenotype and further expressing CCR6 and/or CXCR6, for (i) diagnosing, (ii) prognosing outcome of, or (iii) predicting the risk of developing a disease related to a decrease of F. prau. The invention also concerns the treatment of said disease by administering a population of these specific T lymphocytes. The Inventors have indeed identified two markers, CCR6 and CXCR6, enabling to select a population of F. prau-specific cells among CD4.sup.+ CD8.sup.low T lymphocytes, from a blood sample and without needing to assess their F. prau specificity. T lymphocytes with a CD4.sup.+ CD8.sup.low CCR6+ CXCR6+ phenotype are for example significantly decreased in IBD patients. The disease related to a decrease of F. prau is particularly an inflammatory bowel disease (IBD), such as Crohn's disease.

Claims

1. A method of treating a subject afflicted with a disease characterized by a decrease of F. prausnitzii, wherein said method comprises: a) determining the number and/or concentration and/or proportion of T lymphocytes with a CD4.sup.+ CD8.sup.low CCR6.sup.+ phenotype, a CD4.sup.+ CD8.sup.low CXCR6.sup.+ phenotype and/or a CD4.sup.+ CD8.sup.low CCR6.sup.+ CXCR6.sup.+ phenotype, in a biological sample from the subject, optionally, comparing the result of step a) with i) a control standard value corresponding to the number and/or concentration and/or proportion of these T lymphocytes typically found in a biological sample of the same nature from a healthy subject, and/or ii) a control standard value corresponding to the number and/or concentration and/or proportion of these T lymphocytes typically found in a biological sample of the same nature from a patient suffering from said disease characterized by a decrease of F. prausnitzii; b) deducting from the result(s) of step a) and/or step b) where appropriate, if the subject is afflicted with a disease characterized by a decrease of F. prausnitzii, and c) administering a suitable treatment to the subject deduced in step c) to be afflicted with a disease characterized by a decrease of F. prausnitzii, wherein said suitable treatment is selected from the group consisting of an immunosuppressant, a probiotic, an antibiotic, F. prausnitzii, a fragment of F. prausnitzii, a pharmaceutical composition comprising isolated T lymphocytes with a CD4.sup.+ CD8.sup.low CCR6.sup.+ phenotype, a CD4.sup.+ CD8.sup.low CXCR6.sup.+ phenotype and/or a CD4.sup.+ CD8.sup.low CCR6.sup.+ CXCR6.sup.+ phenotype and their combinations.

2. The method according to claim 1, wherein the disease characterized by a decrease of F. prausnitzii is an inflammatory bowel disease (IBD).

3. The method according to claim 1, wherein the biological sample is a blood sample.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1: Colonic lamina propria DP8a T cells express high levels of CCR6, CXCR3 and CXCR6. A. Gating strategy for the detection of CCR6-, CXCR3- and CXCR6-expressing DP8a cells. Colonic lamina propria were dissociated as described in Methods. A representative example is shown. B. The frequency for each subset is represented as a mean of 5 donors.

(2) FIG. 2: DP8a T cells circulate in blood and express CCR6, CXCR3 and CXCR6, but at lower levels than in colonic lamina propria. A. The gating strategy to detect circulating DP8a cells is shown. B. The frequency within DP8a cells of indicated markers is shown. C. The frequency of the various CCR6, CXCR3 and CXCR6 subsets in blood DP8a is shown for a representative donor and as a mean of 8 donors.

(3) FIG. 3: F. prau-specific DP8a T cells are enriched in CCR6 and CXCR6 cells. CD4.sup.+ T cells were stained with VPD and stimulated by autologous monocytes (ratio monocytes:CD4.sup.+ T cells of about 1:1) loaded 6 h with F. prau (ratio monocytes:bacteria of about 1:10) in the presence of 0.5 microg/ml anti-CD28 (Miltenyi Biotec). Unloaded monocytes or monocytes in the presence of CD3/CD28 beads (ratio T cells:beads of 3:1) were used as negative and positive controls, respectively. Proliferation was measured at day 5 by dilution of VPD. CCR6 (B) and CXCR6 (C) expression was assessed in proliferated (VPD.sup.low) versus non-reactive (VPD.sup.high) DP8a cells (A). Two-sided paired t-test; *: p<0.05 was considered significant.

(4) FIG. 4: Circulating F. prau-reactive DP8a cells are present in the CCR6+ population. Sorted CCR6-negative or CCR6-positive, as well as CXCR6-negative and positive DP8a T cells were stained with VPD and stimulated by autologous monocytes (ratio monocytes:DP8a T cells of about 1:1) loaded ON with F. prau (ratio monocytes:bacteria of about 1:10; middle panel) in the presence of 0.5 microg/ml anti-CD28 (Miltenyi Biotec). Unloaded monocytes (monocytes alone) or monocytes in the presence of CD3/CD28 beads (ratio T cells:beads of 3:1) were used as negative and positive controls, respectively. Proliferation was measured at day 5 by dilution of VPD.

(5) FIG. 5: Clonal responses and phenotype of circulating DP8a cells. A. Gating strategy used for cloning. B. Production of IL-10 by a panel of DP8a clones in the presence (gray bars) or in the absence (black bars) of 1 microg/ml coated anti-CD3 Ab. C. Percentages of clones among all 36 F. prau-specific DP8a clones tested expressing the indicated subsets. Non-indicated subsets were not detected in the clone panel.

(6) FIG. 6: F-prau-specific DP8a clones inhibit CD4 T cell proliferation. A. CD4.sup.+ T cells were stained with VPD and stimulated by CD3/CD28 beads (ratio CD4 T cells:beads of 3:1) in the presence or in the absence of each of all 19 clones tested (ratio 1:1). Proliferation of CD4 T cells was measured at day 5. sem bars are from 3 different donors of CD4 T cells. B. VPD.sup.+ CD4 T cells were stimulated as above at indicated CD4 T cells:DP8a clones ratios and proliferation was measured at day 5 in the presence of 4 different DP8a clones. Two-sided unpaired t-test; error bars: sem; *: p<0.05 was considered significant.

(7) FIGS. 7 and 8: Circulating CCR6.sup.+/CXCR6.sup.+ DP8a cells are specifically decreased in IBD patients. PBMCs derived from IBD patients, healthy donors or infectious colitis patients were stained for CD3, CD4, CD8a, CCR6 and CXCR6.

(8) FIG. 7: The frequency of CCR6.sup.+/CXCR6.sup.+ DP8 cells within CD3.sup.+ T cells was determined and plotted for 10,000 CD3.sup.+ cells. A diagnostic threshold was positioned at 7.875 CCR6.sup.+/CXCR6.sup.+ DP8 cells per 10,000 CD3.sup.+ cells using GraphPad Prism version 6.0.

(9) FIG. 8: The frequency of CD4.sup.+ cells within CD3.sup.+ T cells was also determined. A One-way ANOVA was used. error bars: sem; ****: p0.0001, ***: p=0.001, **: p=0.01.

(10) FIG. 9: ROC curve of infectious colitis patients versus IBD patients. This curve was performed using GraphPad Prism version 6.0. and allows for the determination of a diagnostic threshold at <7.875 CCR6.sup.+/CXCR6.sup.+ DP8a cells per 10,000 CD3.sup.+ cells with a specificity of 100%. The sensitivity is 72.22% (Area: 0.931, p<0.0001).

(11) FIG. 10: Circulating CCR6.sup.+/CXCR6.sup.+ DP8a cells are significantly decreased in obese patients with or without type 2 diabetes, but not in T1D (Type 1 diabetic) patients. PBMCs derived from healthy donors, non-diabetic obese patients, pre-diabetic obese patients, obese and type 2 diabetic patients or type 1 diabetic patients were stained for CD3, CD4, CD8a, CCR6 and CXCR6.

EXAMPLE

Example 1: CCR6.SUP.+./CXCR6.SUP.+ DP8a T Cells and IBD

(12) Material and Methods

(13) (i) Reagents

(14) Human Peripheral blood mononuclear cells (PBMCs), purified monocytes or lymphocytes were cultured in RPMI-1640 supplemented with 5% human Serum, 2 mM L-glutamine and 10 g/ml penicillin-streptomycin (Gibco). rhIL-2 was used for the culture and expansion of T cells. Violet Proliferation Dye 450 (VPD) (1 M, BD Bioscience), anti-CD3/anti-CD28 activation beads (Gibco), Brefeldin A (10 mg/ml, Sigma-Aldrich), and 4% paraformaldehyde (Sigma-Aldrich) were used.

(15) (ii) Bacterial Cultures

(16) F. prausnitzii A2-165 were obtained from Commensal and Probiotic-Host Interactions Laboratory, UMR1319 Micalis, INRA, Jouy-en-josas, France. F. prausnitzii was grown for 20 h at 37 C. in LYBHI medium (brain-heart infusion medium supplemented with 0.5% yeast extract; Difco), cellobiose (1 mg/ml; Sigma-Aldrich), maltose (1 mg/ml; Sigma-Aldrich), and cysteine (0.5 mg/ml; Sigma-Aldrich) in an anaerobic chamber. F. prausnitzii was used after sonication.

(17) (iii) Cell Separation

(18) PBMCs were isolated by Ficoll gradient centrifugation from healthy donor blood (EFS, Nantes, France), IBD patients or infectious colitis patients. This latter study was approved by the ethics committee of the Comit de Protection des Personnes Ile-de-France IV (Suivithque). Several patients also came from the CHU of Nantes hospital. All patients signed informed consent forms. Patients' and donors' characteristics are shown in supplementary Table I.

(19) Monocytes and CD4 T cells were purified using CD14 and CD4 microbeads, respectively, according to the supplier's instructions (Miltenyi).

(20) Normal colonic mucosa was obtained from colorectal cancer patients from surgically resected tissue, taken approximately 10 cm downstream of the tumor. The lamina propria was separated from the epithelium after incubation in 1 mM EDTA PBS buffer (20 min) and then minced into approximatively 1 mm.sup.2 fragments and washed with RPMI containing penicillin (10%) and gentamycin (0.1 mg/ml; Sigma-Aldrich). Tissue fragments were digested with collagenase IV (1 mg/ml; Sigma-Aldrich), with shaking at 37 C. Mucus and large debris were removed by filtration through a 40 mm-cell strainer (BD). Viable cells were obtained by Ficoll gradient centrifugation. Cells were then cultured for 7 days in RPMI with 10% FBS, 2 antibiotics, 1 fungizone and 150 UI/ml IL-2, before stainings were performed. During this time, some samples got infected and were discarded. Uninfected samples recovered from collagenase-treatment and re-expressed markers which were then studied. This study was approved by the ethics committee of the CHU de Nantes. All patients signed informed consent forms.

(21) (iv) Antibodies

(22) For surface staining, cells were harvested, washed and stained for 30 min at 4 C. in PBS 0.1% BSA with the following Abs: anti CD3-PECy7 (clone UCHT1, Becton Dickinson), anti CD4-FITC (clone 13B8.2, Beckman Coulter), anti CD8a-APC (clone B9.11, Beckman Coulter), BV605 (clone SK1, Becton Dickinson) or -BV421 (clone RPA-T8, Becton Dickinson) CCR6-BV421 (clone 11A9, Becton Dickinson) or -PE (clone G034E3, Biolegend), CXCR3-BV785 (clone G025H7, Biolegend), CXCR6-APC (clone K041E5, Biolegend), anti 7-PE (clone FIB504, Becton Dickinson),

(23) For intracellular staining, cells were harvested, fixed in 4% paraformaldehyde, washed and stained for 30 min at RT in PBS 0.1% BSA 0.1% saponin with anti-IFNg-APC (clone B27, Becton Dickinson).

(24) Fluorescence was measured on FACS LSR II flow cytometer and analyzed using Diva software (Becton Dickinson).

(25) (v) T Cell Culture and Stimulation

(26) Purified CD4.sup.+ T cells or PBLs were stained with violet proliferation dye (VPD) and stimulated by autologous monocytes (ratio 10 lymphocytes:1 monocyte) loaded ON or not with bacteria (10 bacteria:1 monocyte). As a positive control, T cells were stimulated by CD3/CD28 beads (3 T cells:1 bead). VPD dilution was assessed 5 days later.

(27) To obtain F. prau-specific T cell clones, VPD-stained CD4.sup.+ T cells were stimulated as above. At day 5, VPD.sup.low DP8 T cells were sorted and cloned using a FACS Aria and amplified using irradiated allogeneic PBMCs and LAZ cells (B-EBV cell line), in the presence of 1 mg/ml PHA and 150 IU/ml rhIL-2.

(28) T cell clones were stimulated by autologous monocytes (ratio 2-3 lymphocytes:1 monocyte) loaded ON or not with bacteria (10 bacteria:1 monocyte). For IFNg detection, T cell clones were stimulated for 6h in the presence of 10 mg/ml brefeldin A before intracellular staining of cytokines. For IL-10 detection, clones were stimulated 48 h by 1 mg/ml coated anti-CD3 (OKT3), before IL-10 measurement by ELISA.

(29) (vi) ELISA

(30) DP8a T cell clones were stimulated or not using coated anti-CD3 (clone OKT3, 1 mg/ml, eBioscience) for 48 h at 37 C. Supernatants were harvested and tested for their IL-10 content using the Ready-Set-Go ELISA according to the manufacturer's guidelines (eBioscience).

(31) (vii) Statistical Analysis

(32) Statistical analysis was performed using GraphPad Prism version 6.0. Most comparisons were performed using 2-sided t-test or one-way ANOVA, as indicated in figure legends. p<0.05 was considered statistically significant.

(33) Results

(34) Colonic Lamina Propria DP8a T Cells Express High Levels of CCR6, CXCR3, CXCR6 and b7 Integrin.

(35) To study DP8a T cells within colonic lamina propria lymphocytes (LPLs) for the expression of intestine homing markers, LPL were dissociated using collagenase IV, from 5 human colon tissue resections. Filtered cells were ficolled and cultured as described in the Methods section. It was then stained the CD3.sup.+ DP8a cells for CCR6, CXCR3 and CXCR6 at day 7 because the collagenase enzyme temporarily stripped most surface markers from the cell surface (see FIG. 1A et B). As a mean, 78% of LPL-derived DP8a cells expressed CCR6 (composed of CCR6.sup.+/CXCR3.sup./CXCR6.sup., CCR6.sup.+/CXCR3.sup.+/CXCR6.sup., CCR6.sup.+/CXCR3.sup./CXCR6.sup.+ and CCR6.sup.+/CXCR3.sup.+/CXCR6.sup.+ DP8 cells), 66% expressed CXCR6 (composed of CCR6.sup./CXCR3.sup./CXCR6.sup.+, CCR6.sup./CXCR3.sup.+/CXCR6.sup.+, CCR6.sup.+/CXCR3.sup./CXCR6.sup.+ and CCR6.sup.+/CXCR3.sup.+/CXCR6.sup.+ DP8 cells) and 76% expressed CXCR3 (composed of CCR6.sup./CXCR3.sup.+/CXCR6.sup., CCR6.sup.+/CXCR3.sup.+/CXCR6.sup., CCR6.sup./CXCR3.sup.+/CXCR6.sup.+ and CCR6.sup.+/CXCR3.sup.+/CXCR6.sup.+ DP8a cells), among which 50% expressed the 3 markers. Interestingly, most LPL-derived DP8a cells (>57%) co-expressed CCR6 and CXCR6. The majority of LPL-derived DP8a cells also expressed the 7 integrin (data not shown).

(36) CCR6, CXCR3 and CXCR6 are Potential Markers for Recirculating Colonic DP8 T Cells

(37) To characterize circulating DP8a cells for the expression of intestine homing markers and thereby detect gut-associated blood DP8a (see FIG. 2A), PBMCs were stained for not only CCR6, CXCR3 and CXCR6, but also CCR9, CCR10 and b7 (see FIG. 2 B). CCR9 and CCR10 were barely expressed by DP8a cells (or by their CD4 counterparts, not shown). On the other hand, CCR6, CXCR3, CXCR6 or b7 positive cells represented a significant proportion of DP8a cells (20-40%; see FIG. 2 B). Moreover, the expression of these 4 markers in DP8a cells was heterogeneous and overall more frequent than in CD4 T cells (see FIG. 2 B and not shown), which could be compatible with the amount of F. prau-specific T cells among circulating DP8a. Indeed, because 1/ the frequency of circulating DP cells can highly vary between individuals and be amplified, especially by viruses, and 2/ the fecal and mucosal F. prau levels varies considerably from one person to another, one can speculate that the expected number of F. prau-specific T cells within DP8a T cells fluctuates significantly between donors. It was therefore studied more closely the frequency of CCR6, CXCR3, CXCR6 and b7 in DP8a cells. Almost 22% of DP8a cells expressed b7 (FIG. 2 B), but sorted DP8a cells displayed F. prau-specific responses in both b7-positive and b7-negative subsets (not shown), eliminating b7 as a marker for F. prau-specific DP8a cells.

(38) Based on these results and on the frequent expression of CCR6, CXCR3 and CXCR6 by colon-derived DP8a cells, it was then further focused on CCR6, CXCR3 and CXCR6 as potential markers for F. prau-specific DP8a cells: in comparison to LPLs, CCR6.sup.+, CXCR6.sup.+ and CXCR3.sup.+ cells represented only 55%, 32% and 37% respectively of DP8a cells derived from PBMCs, and almost 18% of the latter cells expressed the 3 molecules (see FIG. 2 D). Compared with the co-expression of these three markers by a majority of colon DP8 cells, these data suggested that only a fraction of circulating DP8 T cells could be colon-derived T.sub.REGS.

(39) F. prau-Specific DP8 T Cells Express CCR6 and CXCR6

(40) To better determine whether these receptors are preferentially expressed by F. prau-specific DP8a, stimulation of VPD-stained CD4.sup.+ T cells by autologous monocytes loaded with F. prau was performed. Five days later, proliferation to F. prau was assessed and CCR expression was studied in proliferated (F. prau-specific) versus non-proliferated DP8a cells (see FIG. 3A). CCR6-positive DP8a cells represented a (p<0.003; n=4) higher frequency in VPD.sup.low (F. prau-specific) than in VPD.sup.high DP8 cells (see FIG. 3 B). A similar trend was observed for CXCR6 (see FIG. 3 C). In contrast, CXCR3-positive DP8a cells were equally found in VPD.sup.high versus VPD.sup.low fractions (not shown). Altogether these data suggest that F. prau-specific DP8a cells preferentially express CCR6 and CXCR6, but not CXCR3.

(41) It was further studied the F. prau-specificity of both CCR6-negative and CCR6-positive DP8a cells. No F. prau-specific cells were detected in the CCR6-negative fraction of DP8a cells (see FIG. 4). Nevertheless, in the same donor, a significant fraction of CCR6-positive DP8a cells proliferated in response to F. prau. In other words, F. prau-specific cells can be found in the CCR6.sup.+ fraction, but not in the CCR6.sup. fraction, of DP8a cells. Similarly, CXCR6-negative DP8 contained only very few F. prau-reactive, while the F. prau-specific cells were strikingly found in the CXCR6-positive fraction of DP8 cells (See FIG. 4).

(42) To better assess whether F. prau-specific cells were located in the CCR6.sup.+ and CXCR6.sup.+ DP8 subset, clones were generated from an enriched population of specific cells (see FIG. 5A). Thirty-six clones proliferated and produced IFNg in response to autologous monocytes loaded with F. prau (data not shown). Additionally, most of these clones produced IL-10 upon CD3 ligation (see FIG. 5 B). Strikingly, the majority (about 86%) of the tested F. prau-specific clones expressed CCR6, CXCR3 as well as CXCR6 and 89% and 92% expressed CCR6 and CXCR6, respectively (see FIG. 5 C), highly suggestive of a significant correlation between F. prau-specificity and CCR6/CXCR6 expression.

(43) Circulating F-Prau-Specific DP8 Clones Inhibits CD4 T Cell Proliferation

(44) To determine their inhibitory potential, VPD-stained CD4.sup.+ T cells were stimulated by CD3/CD28 beads in the presence or in the absence of one F. prau-specific DP8 clone. Most of the 19 tested clones substantially inhibited CD4 T cell proliferation (see FIG. 6A) and still displayed inhibitory functions at 1:20 DP8a:CD4.sup.+ T cell ratio (see FIG. 6B), demonstrating their potent regulatory function.

(45) Circulating F-Prau-Specific DP8a Clones Display T.sub.R1-Like TREG Properties

(46) To establish the TRI-like features of these clones, it was first assessed their ability to produce IL-10. Most of these clones produced IL-10 upon CD3 ligation and stimulation by F. prau-loaded monocytes (data not shown). Moreover, CD4.sup.+ T cells proliferation induced by CD3/CD28 beads was substantially inhibited in the presence of all the tested clones (data not shown), an inhibition still exhibited at 1:10 DP8:CD4.sup.+ T cell ratio, demonstrating their potent regulatory function. It was also checked for the expression of CD39 and CD73 by the clones, two ectoenzymes, which in cooperation are involved in the differentiation and function of regulatory T.sub.R1 cells. F. prau-specific DP8 clones expressed CD39 at heightened levels, as compared to T.sub.H1 (CD4.sup.+/CXCR3.sup.+) or Foxp3.sup.+ T.sub.REG (CD4.sup.+/CD25.sup.high/CD127.sup.low) clones generated in parallel of DP8 clones. DP8 clones also expressed CD73 on a fraction of their cells, ranging from 5 to 42% of the clone cells. T.sub.H1 and FoxP3.sup.+ T.sub.REG cell clones used also expressed CD73 within this range. ROR is known to be expressed by gut-derived T.sub.REGS. It was expressed by the DP8 clones as well as the FoxP3.sup.+ T.sub.REG clone, as compared to the T.sub.H1 clone whose expression was minimal. Because DP8 clones expressed ROR, it was also assessed whether they produced IL-17. Upon potent stimulation, no IL-17 was detected, confirming their regulatory properties. Finally, while F. prau-specific clones lacked CD62L expression, they expressed CD38 therefore exhibiting the CD62L.sup./CD38.sup.+ phenotype, reported to identify mucosally-differentiated cells.

(47) It was further investigated the underlying mechanism for DP8 regulatory phenotype. It had already been shown that it was partly dependent on IL-10 production, since blocking IL-10 and IL-10R incompletely restored CD4 proliferation. It was also determined that these DP8 clones produced no TGF, eliminating a role for this cytokine. It was then assessed whether this mechanism was contact-dependent or not using transwell assays. CD4 proliferation was not inhibited when separated from the DP8 cells by a 1 m-pore membrane, while clearly inhibited when cells were in the same well. Finally, it was investigated whether CD39 was involved in the regulatory mechanism. Indeed, DP8 clones expressed elevated levels of this membrane-bound molecule. A molecule, POM-1, known to inhibit CD39 function through inhibition of ATP hydrolysis, was used in an assay measuring the inhibition by DP8x clones of CD4 proliferation. Strikingly, in the presence of POM-1, proliferation was significantly restored, demonstrating the implication of CD39 in the regulatory mechanism of DP8 cells. The fact that proliferation was both mostly restored by POM-1 and virtually entirely contact-dependent suggests that IL-10 expression happens downstream of CD39 and may depend on CD39 function, as previously described in dendritic cells.

(48) Circulating CCR6.sup.+/CXCR6.sup.+ DP8a Cells are Decreased in IBD Patients

(49) Altogether, these data support that CCR6 and CXCR6 are preferentially expressed by circulating F. prau-specific DP8a T.sub.REGS, which should allow for their quantification/tracking in PBMC samples. The function of these circulating cells seems to mirror those of the colonic lamina propria and it was assessed whether they could help predict colon homeostasis versus inflammation in IBD.

(50) To start assessing a potential role for F. prau-specific DP8a T.sub.REGS in IBD, the frequency of CCR6.sup.+/CXCR6.sup.+ circulating DP8a T cells was determined in 106 IBD patients, as compared to 35 age-matched healthy donors, as well as 12 infectious colitis patients (used as a control for IBD-related inflammation specificity). Strikingly, the frequency of this subset within total CD3.sup.+ T cells was significantly (p<0,0001) decreased in IBD patients (mean=5.9.sup.0/.sub.0001.0), as compared to both that in healthy donors (mean=24.4.sup.0/.sub.0003.5) or in infectious colitis patients (mean=49.5.sup.0/.sub.00023.1) (see FIG. 7). The frequency of this subset was also reduced significantly within CD4.sup.+ T cells or DP8a cells (data not shown), while the fraction of CD4.sup.+ T cells within CD3.sup.+ cells was not altered (FIG. 8).

(51) Additionally, it was determined a diagnostic threshold plotting a ROC curve to compare infectious colitis patients versus IBD patients (FIG. 9), which are indiscernible at the time of the first IBD flare. The threshold was positioned at <7.875 CCR6.sup.+/CXCR6.sup.+ DP8a cells per 10,000 CD3.sup.+ cells, with a sensitivity of 72.22% and a specificity of 100% (Area: 0.931, p<0.0001).

(52) These results show that the frequency of circulating DP8a T cells expressing CCR6 and CXCR6 is decreased in IBD patients, suggesting a role for these cells, or the lack thereof, in the susceptibility to IBD. Moreover, frequencies of these cells in blood are thus of prognosis and diagnosis value, especially at the time of the first IBD flare, which is indiscernible from infectious colitis.

CONCLUSION

(53) The generation of F. prau-specific DP8a clones derived from PBMCs definitely established the presence of such cells in the periphery. Importantly, these circulating cells display the same properties than DP8a cells derived from colonic LPLs: they specifically recognize a commensal colonic bacterium, Faecalibacterium prausnitzii, produce IL-10 (see FIG. 5) and exert potent regulatory functions (See FIG. 6). Nevertheless, in the colon, most DP8a cells are highly specific for F. prau, while in the blood, the frequency of F. prau-specific cells varies much more and can be as low as a few percent. This shows that circulating DP8a cells are heterogeneous, in accordance with previous reports that some of these cells represent clonal expansions of various virus-specific T cells, which could takeover F. prau-specific DP8a cells. Moreover, the variable frequency of circulating F. prau-specific DP8a cells may reflect the highly diverse levels of this bacterium between healthy donors.

(54) In contrast with colonic LPLs, quantifying overall circulating DP8a cells is therefore not sufficient to estimate the frequency of F. prau-specific cells, which appears to be central and could represent a marker for IBD. Unfortunately, assessing the specificity of these cells for F. prau remains cumbersome because 1/ the bacterium is difficult to grow due to its anaerobic properties and 2/ the need for autologous or HLA-matched antigen presenting cells. Therefore, identification of a marker(s) for F. prau-specific DP8a cells is critical to detect these cells in the blood, rather than in colonic biopsies where very little material ends up being available anyway. Therefore, identifying markers for this circulating subset appears to be key to use these cells as a prognosis marker in IBD. Zooming on CCR6- and CXCR6-positive cells within DP8a cells seems to allow for such a closer detection of F. prau-specific cells, than overall DP8a cells. Accordingly, F. prau-specific cells were found only in the CCR6.sup.+ and mainly in the CXCR6.sup.+ fractions of DP8a cells (see FIG. 4), as well as 89% and 92% of the clones specific for F. prau expressed CCR6 and CXCR6, respectively (see FIG. 5C).

(55) Importantly, the frequency of circulating CCR6.sup.+/CXCR6.sup.+ DP8a T cells is low in the majority of IBD patients, as compared to age-matched healthy donors (FIGS. 7 to 9). Therefore, a low abundance of these cells in blood is associated with inflammatory gut disorders suggesting that this T.sub.REG subset plays a role in the control of IBD-related inflammation. The frequency of these cells in blood is therefore of prognosis/diagnostic value (with a threshold that was positioned at 7.875 CCR6.sup.+/CXCR6.sup.+ DP8 cells per 10,000 CD3.sup.+ T cells), especially at the time of the first IBD flare, which is indiscernible from infectious colitis. Moreover, these results showed no significant differences between remission and relapse groups, suggesting that low DP8a numbers is a general, but specific (since infectious disease patients display a higher DP8a frequency, similar to the one of healthy donors), IBD feature.

(56) T regulatory type 1 (T.sub.R1) cells have been described as Foxp3-negative T.sub.REG that suppress T cell responses via the secretion of IL-10 and TGF-b. F. prau-specific DP8a cells had a strong propensity to produce IL-10 upon TCR ligation (see FIG. 5 C). F. prau-specific DP8a cells could therefore be a subset of T.sub.R1-like cells. Nevertheless, whether DP8a cells exert their regulatory function through IL-10 still needs to be established. Moreover, F. prau-specific DP8 cells differ from described T.sub.R1 by their potent proliferative capacity.

(57) Altogether, these data establish 1/ the presence in the blood, of a T.sub.REG subset of colon origin induced by F. prau, 2/ its variable abundance in healthy subjects and 3/ its striking decrease in IBD patients. Hence, a precise and easy follow-up of circulating DP8a cells is now possible thanks to the identification of their specific expression of both CCR6 and CXCR6. These markers may thus be used as a diagnosis tool.

Example 2: CCR6.SUP.+./CXCR6.SUP.+ DP8a T Cells and Obesity

(58) Material and Methods

(59) (I) Cell Separation

(60) PBMCs were isolated by Ficoll gradient centrifugation from healthy donor blood (EFS, Nantes, France), or from obese patients suffering or not of type diabetes treated at the CHU of Nantes hospital. This study was approved by the ethics committee of the CHU de Nantes. All patients signed informed consent forms.

(61) (II) Antibodies

(62) For surface staining, cells were harvested, washed and stained for 30 min at 4 C. in PBS 0.1% BSA with the following Abs: anti CD3-PECy7 (clone UCHT1, Becton Dickinson), anti CD4-FITC (clone 13B8.2, Beckman Coulter), anti CD8a-APC (clone B9.11, Beckman Coulter), -BV605 (clone SK1, Becton Dickinson) or -BV421 (clone RPA-T8, Becton Dickinson) CCR6-BV421 (clone 11A9, Becton Dickinson) or -PE (clone G034E3, Biolegend), CXCR3-BV785 (clone G025H7, Biolegend), CXCR6-APC (clone K041E5, Biolegend), anti 7-PE (clone FIB504, Becton Dickinson),

(63) Fluorescence was measured on a LSR II flow cytometer and analyzed using Diva software (Becton Dickinson).

(64) (III) Statistical Analysis

(65) Statistical analysis was performed using GraphPad Prism version 6.0. Most comparisons were performed using 2-sided t-test or one-way ANOVA, as indicated in figure legends. p<0.05 was considered statistically significant.

(66) Results

(67) As shown in FIG. 10, circulating CCR6.sup.+/CXCR6.sup.+ DP8a cells are significantly decreased in obese patients with or without type 2 diabetes, but not in T1D patients (Type 1 diabetic patients). This population of cells is therefore useful in the methods according to the invention, wherein the disease associated with a decrease of F. prausnitzii is obesity.