Published online before print May 29, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Immunoregulation Laboratory, Centre Hospitalier de l’Université de Montréal Research Center, University of Montréal, Québec, Canada; and
Danone Research, Palaiseau, France
1 Correspondence: Centre de Recherche du CHUM, Laboratoire Immunorégulation (M4211K), 1560, rue Sherbrooke est, Montréal, Québec, Canada H2L 4M1. E-mail: m.sarfati{at}umontreal.ca
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, IL-10, and IL-17 and instead, displayed suppressor function. These data demonstrate that phenotypic DC maturation combined to an appropriate cytokine profile is insufficient to warrant Th1, IL-10-secreting T regulatory Type 1 (Tr1), or Th17 polarization. We propose that commensal flora and as such, probiotics manipulate DCs by a yet-unidentified pathway to enforce gut tolerance.
Key Words: probiotics • IL-12 • IL-10 • suppressive T cells • tolerance • Foxp3
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Dendritic cells (DCs) are specialized in antigen capture and presentation to naive T cells. As such, DCs orchestrate the fate of T cell response [6 ]. A direct communication occurs in the gut between intestinal microbiota and DCs, and DCs extend processes across the epithelium to directly capture bacteria in gut lumen; the number of DC extensions is regulated by the epithelial response to TLR-mediated signals [7 , 8 ]. In the intestine, TGF-β-rich milieu, a particular DC subset (i.e., myeloid CD11b+CD103+ cells) in the presence of retinoic acid is instrumental to the conversion of naive CD4+CD25– T cells into CD4+CD25+forkhead box p3 (Foxp3)high regulatory T cells (Tregs) [9 ]. Together with IL-10-secreting T regulatory Type 1 (Tr1) and thymic-derived, naturally occurring Tregs, they likely contribute to the establishment of gut peripheral tolerance [10 , 11 ]. However, several mechanisms apart from Tregs are contributing to the maintenance of gut homeostasis. These include the production of antimicrobial peptide defensins, mucus secretion in the gut lumen that limits access to epithelial cell surfaces, IgA production, and the restriction to the local lymph nodes of the mucosal immune responses to noninvasive pathogens [12 13 14 15 ].
In mice, monocytes give rise to mucosal but not conventional DCs [16
]. In the course of persistent inflammation, monocytes are massively recruited to the inflamed sites and develop into DCs [17
]. In the present study, we thought to investigate how some of the natural components of the colonic microbiota, also termed probiotics, directly regulate maturation and function of human monocyte-derived DCs and affect the outcome of the T cell response in the absence of inflammatory signals. Indeed, different strains and species of probiotics have been reported to induce a large variation in the cytokine and chemokine response by DCs stimulated in the presence of inflammatory conditioned medium or maturation factors (for instance, IL-1, TNF-
, and PGE2) [18
19
20
21
]. Under inflammatory conditions, gram– but gram+ noninvasive bacteria induced IL-12 release by DCs that drove Th1 polarization. In contrast, selected bacteria enhanced IL-10 and promoted Tr1 development.
We here evaluated the functional consequences of direct DC exposure to eight different commensal-related bacteria (CB) that represent the dominant (i.e., Bifidobacteria and Bacteroides) and subdominant human microflora (i.e., Lactobacilli and Streptococci). We found that all CB strains induced a complete functional DC maturation program in the absence of exogenous inflammatory mediators. Although the different CB strains elicited a common DC phenotypic maturation and chemokine receptor-switch profile, they may be categorized into three major groups according to the amount of cytokine (IL-12 and IL-10) and chemokine they triggered in DCs. Yet, the eight CB strains instruct mature DCs (mDCs) to promote the generation of non-Tr1 and Foxp3low/– T cells that inhibited the proliferation of target T cells. We propose that the immune response evoked by a direct contact between symbiotic bacteria and DCs enforces gut tolerance under steady-state conditions.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-irradiated CB (L. casei DN-114 001, L. casei #2 DN-114 086, Lactobacillus rhamnosus DN-116 047, Lactobacillus plantarum DN-121 022, B. animalis DN-173 010, Bifidobacterium adolescentis DN-150 017, Streptococcus thermophilus DN-001 621, and Bacteroides thetaiotaomicron, a gram– commensal-related species) were used in the cultures at a CB:DC ratio of 200:1 unless indicated otherwise. Staphylococcus aureus cowan I (SAC) was purchased from Calbiochem-Behring (San Diego, CA, USA) and used at a dilution of 1/5000 (0.02% wt/vol). Cytokines and chemokines were measured in culture supernatants using commercially available ELISA kits for CCL1, CCL13, CXCL5, IL-1β, IL-10, IL-12p70, and IL-17 (R&D Systems, Minneapolis, MN, USA), IFN-, IL-6, IL-13, TGF-β, and TNF- (BioSources, San Diego, CA, USA), and CXCL9 and CXCL10 (BD Biosciences, San Jose, CA, USA).
Cell preparation, purification, and culture conditions
Human PBMC were isolated by density gradient centrifugation of heparinized blood from healthy volunteers using Lymphoprep (Axis-Shield PoS AC, Norway). Permission to use human primary cells for the described experiments was obtained from the Ethic Research Committee of the Centre de Recherche du Centre Hospitalier de l’Université de Montréal (CRCHUM; Canada). Monocyte-derived DCs were prepared from enriched monocytes as described in ref. [22
] following 5–6 days culture in the presence of 25 ng/ml IL-4 and 25 ng/ml GM-CSF (R&D Systems). CD4+ T cells or naïve CD4+ T cells were isolated by human CD4 selection cocktail or by human CD4+ naïve T cell enrichment kit, respectively, according to the protocol provided by StemCell Technologies (Canada). Purified, naïve T cells were >98% CD4+ and CD45RA+ (that contained less than 2–5% CD45ROlowCD45RA+) and less than 0.1% single CD45RO+. The frequency of CD25lowFoxp3low among these highly purified, naïve CD4+ T cells was less than 0.5%.
Immature DCs (iDCs) were cultured with CB in HB101 complete medium (Irvine Scientific, Santa Ana, CA, USA) for 48 h. Naïve CD4+ T cells were cocultured for 5–7 days with mDCs at a DC:T ratio of 1:10 in RPMI-1640 medium supplemented with 10% FCS (Gibco-Invitrogen, Carlsbad, CA, USA). Primed T cells were expanded for 3–7 days in the presence of IL-2 (50 U/ml, R&D Systems). Primed T cells were assessed for functional activities. T cells (2x105/ml) were restimulated for 2–4 days with anti-CD3 (OKT-3; 100 ng/ml, Janssen-Ortho Inc., Canada) plus anti-CD28 (500 ng/ml, BD Biosciences) in the presence of mitomycin C (Calbiochem-Behring)-treated, allogeneic monocytes (2x104/ml). To evaluate regulatory function, freshly isolated, nonfractionated CD4+ T cells (2x105/ml) were cultured alone (target x1) or together with target cells (target x2) or primed T cells at a 1:1 or 1:3 cell ratio and stimulated for 3–5 days with OKT3 (100 ng/ml) plus anti-CD28 (500 ng/ml) in the presence of mitomycin C-treated monocytes (2x104/ml) or anti-CD3 (UCHT-1; 25 ng/ml; provided by Dr. Claire Hivroz, Institut Curie, Paris, France), immobilized on mouse L fibroblast expressing CD32/CD80. Exogenous IL-2 (50 U/ml) was added during the effector:target (E:T) cell coculture. Cytokine production was measured in the culture supernatant by ELISA (R&D Systems), and cell proliferation was assessed by 3H-thymidine (GE Healthcare, UK) incorporation. Proliferation and IFN- production of CD3-activated CD4+ T cells alone (target x1) and CD4+ T cells mixed at a cell ratio of 1:1 or 1:3 (target x2) were used as a reference value to evaluate suppressive function of primed T cells.
Antibodies and flow cytometry analysis
CB-treated DCs were stained with directly conjugated antibodies for CD80, CD83, CD86, and MHC class II (ID Labs, Canada) and CD40, CCR5, CCR7, CXCR4, and DC-SIGN (BD Biosciences). CD4, CD45RA, and CD45RO (Biolegend, San Diego, CA, USA) and Foxp3 (eBioscience, San Diego, CA, USA) antibodies were used to verify the purity of naïve T cells. Primed and expanded T cells were stained with antibodies for CD25, CD103, CD152, CCR4, and CCR5 (BD Biosciences) and glucocorticoid-induced TNFR (GITR; R&D Systems).
Quantitative RT-PCR (qRT-PCR) analysis
Total RNA was extracted from primed and expanded T cells, and qRT-PCR for Foxp3 expression was performed according to the manufacturer’s protocol (Qiagen, Valencia, CA, USA; InvitroGen, Carlsbad, CA, USA; and Applied Biosystems, Foster City, CA, USA) [23
].
Statistical analysis
Statistical analysis was done using two-tailed Student’s t-test assuming equal variance between the compared groups. Statistical significance is indicated in the figures (*, P<0.05; **, P<0.01; ***, P<0.001).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Monocyte-derived DCs were cultured individually with eight different CB strains or SAC particles as a surrogate for gram+ pathogenic bacteria. Bifidobacteria and Bacteroides belong to the dominant species of the human microflora, whereas Lactobacilli and Streptococci are representative of subdominant human flora. Dose-response studies and kinetics established a 200:1 CB:DC ratio and 48 h stimulation to be optimal for the induction of costimulatory molecule expression (i.e., CD80, CD86, CD40), MHC class II, and chemokine receptor switch (CCR7 up-regulation) in response to all CB examined (Fig. 1
). In fact, no significant difference was noted between various CB strains and SAC particles (1/5000 dilution, 0.02% wt/vol), with the exception of B. animalis, which was the least potent inducer of CD86. Of interest, we noticed that L. casei and B. aminalis, but not SAC, down-regulated CCR5, CXCR4, and DC-SIGN expression, three surface molecules involved in HIV recognition and transport [24
]. Previous studies have used a maturation cocktail (IL-1 and TNF-) concomitant to CB exposure or a 1000:1 irradiated CB:DC ratio combined with a prolonged exposure time (i.e., 72 h) to activate DC cytokine production [19
, 20
]. Here, we show that a direct contact with CB during 48 h triggered cytokine and chemokine release in the absence of inflammatory signals (Fig. 2
). CB were distinguished according to their ability to preferentially induce IL-12p70 or IL-10 secretion (Fig. 2)
. Notably, L. casei and S. thermophlius activated DCs to produce significant amounts of IL-12p70 and CXCL10 but low quantities of IL-10. Conversely, B. animalis and B. adolescentis triggered high IL-10 and IL-6 but low IL-12 release. The proinflammatory cytokine TNF- and the neutrophil-attracting chemokine CXCL5 were preferentially induced by the latter group. In contrast to the gram+ CB, the B. thetaiotaomicron, an abundant gram– CB, induced a low amount of IL-12p70 and IL-10 and significant quantities of IL-1β and IL-6. Finally, TGF-β was barely detectable in the culture supernatant of CB-activated DCs (Fig. 2D)
. The differential IL-12/IL-10 cytokine ratio elicited by gram+ CB was not altered by extending the culture period to 72 h (Fig. 2E
and 2F)
. It is important to note that our preliminary studies aiming to compare DC activation by live versus -irradiated bacteria (i.e., L. casei and B. animalis) yielded a comparable cytokine profile (Fig. 2G)
. However, -irradiated bacteria revealed a lower DC-activating potential than live bacteria. Considering that the irradiation procedure maintained the bacteria structure integrity and that DC viability was preserved at a 200:1 CB:DC ratio, these experimental conditions have been selected and used throughout the study.
 View larger version (64K): [in a new window] |
Figure 1. Gram+ and gram– CB induce phenotypic DC maturation. DCs were cultured for 48 h in the absence (iDC) or the presence of CB strains (L. casei DN-114 001, L. casei #2 DN-114 086, L. rhamnosus DN-116 047, L. plantarum DN-121 022, B. animalis DN-173 010, B. adolescentis DN-150 017, S. thermophilus DN-001 621, and B. thetaiotaomicron, a gram– commensal-related strain) or SAC (0.02% wt/vol). (A) CD80 and CD86 expression was determined on DCs directly activated by L. casei or B. animalis at various bacteria:DC ratios. (B) Expression of CD40, CD80, CD83, CD86, MHC class II, CCR7, CCR5, CXCR4, and DC-SIGN in iDC, SAC (0.02% wt/vol), and L. casei- and B. animalis-activated DCs (200:1). (C) CD86 and MHC class II expression in DCs treated with six different gram+ and one gram– CB (200:1). These data are one representative experiment of three to six.
|
 View larger version (38K): [in a new window] |
Figure 2. Gram+ and gram– CB induce differential DC cytokine and chemokine profiles. DCs were cultured for 48 h in the absence (iDC) or the presence of gram+ and gram– CB strains at a CB:DC ratio of 200:1 or SAC (0.02% wt/vol). (A and C) Production of IL-12p70, IL-10, IL-6, and IL-1β and TNF- (A only). (B) CXCL10, CXCL5, CCL1, CCL13, and CXCL9. (D) TGF-β was measured using ELISA in the 48-h culture supernatants. Data are shown as mean ± SEM of five independent experiments. (E) IL-12p70/IL-10 ratio of SAC and L. casei-, S. thermophilus-, B. animalis-, and B. adolescentis-activated DCs. (F) Time kinetics of IL-10 and IL-12p70 secretion in L. casei- and B. animalis-treated DCs. Data are one representative experiment of three. (G) Cytokine profile of DCs activated for 48 h with live or -irradiated L. casei and B. animalis at various CB:DC ratios. Data showed one representative of three experiments. (H) Classification model of commensal-related bacteria according to DC cytokine- and chemokine-release profiles. Statistical significance between two groups was shown as *, P < 0.05; **, P < 0.01; and ***, P < 0.001.
|
Commensal-related, bacteria-treated mDCs favor CD4+ T cell hyporesponsiveness
Based on their differential cytokine profile, we predicted that CB-treated, phenotypical mDCs were equipped to elicit Th1, Tr1, or Th17 responses. More precisely, we postulated that L. casei promoted IL-12-dependent Th1, Bifidobacterium IL-10-dependent Tr1, and Bacteroides IL-1- and IL-6-driven Th17 development. To verify our hypothesis, we stimulated highly purified adult allogeneic, naive CD45RA+CD4+Foxp3low T cells (Fig. 3A
) for 5 days with CB-treated DCs at a DC:T cell ratio of 1:10. At the end of primary cultures, viable, activated T cells were recovered and cultured in IL-2 for 7–9 days. T cells expanded 
12-fold (Fig. 3B)
. Primed CD4+ T cells were then restimulated with anti-CD3 and anti-CD28 (Fig. 3C)
. Naïve CD4+ T cells primed with SAC-activated DCs proliferated and secreted substantial amounts of IFN-. Unexpectedly, DCs exposed to five different CB strains primed naïve T cells to differentiate into T cells that acquired similar functional properties—i.e., hypoproliferation and low cytokine secretion. Notably, T cells primed with CB-treated DCs produced small amounts of IFN- and almost undetectable IL-10 or IL-17 in secondary cultures. However, we noticed that IL-10 and IL-13 were increased at an early-time point in T cells primed with iDC, delineating the known potential of iDC to drive T cell polarization toward IL-10-producing T cells (Tr1; Fig. 3D
). Thus, in spite of their mature phenotype and ability to secrete large quantities of IL-12, IL-10, or IL-6, DCs exposed to five gram+ and gram– CB did not instruct T cells to acquire the Th1, Tr1, or Th17 cytokine profile. Rather, they promoted T cell hyporesponsiveness.
 View larger version (33K): [in a new window] |
Figure 3. CB-activated mDCs promote the development of hyporesponsive T cells. DCs were cultured for 48 h in the absence (iDC) or the presence of CB strains at a CB:DC ratio of 200:1 or SAC (0.02% wt/vol) and then cultured with allogeneic, naïve CD4+ T cells (DC:T ratio of 1:10). (A) Phenotype of naïve CD4+ T cells. Expression of CD45RA/CD45RO and CD45RA/Foxp3 was determined by flow cytometry before and after naïve T cell purification. (B) Recovery of viable T cells was determined after 5–7 days of T cell primary culture, followed by 7–9 days expansion in IL-2. (C) Cell proliferation, IFN-, IL-10, and IL-17 secretion was assessed in T cell secondary cultures after 3–5 days restimulation with anti-CD3, anti-CD28, and mitomycin C-treated monocytes as APCs. Data are presented as mean ± SEM of five independent experiments. Statistical significance was calculated by Student’s t-test (*, significance in SAC-treated DCs when compared with iDC or CB-treated DCs). (D) Time kinetics of IFN-, IL-10, and IL-13 secretion was determined in T cell secondary cultures after restimulation with anti-CD3 immobilized on a mouse L fibroblast expressing CD32/CD80. The results show one representative experiment of two.
|
production of target T cells (Fig. 4A)
. The suppressive function was comparable with that of T cells primed with iDC. Cocultures of target cells with autologous, freshly isolated CD4+ T cells (target x2) or T cells primed with SAC-treated DCs showed no significant decrease and sometimes increase in cell proliferation and IFN- production as compared with target cells alone (target x1), largely excluding a competition for APCs or nutrient deprivation in the coculture system. The inhibitory effect of suppressor T cells was overcome by the presence of IL-2 during cocultures, suggesting the possible IL-2 consumption by suppressor T cells as a potential mechanism for the inhibition of target cell proliferation (Fig. 4B)
. These generated suppressor T cells did not produce TGF-β nor did they display cytotoxic activities against the K562 cell line (data not shown).
 View larger version (36K): [in a new window] |
Figure 4. Gram+ and gram– CB instruct DCs to promote the development of suppressor T cells. Allogeneic, naïve CD4+ T cells were primed with SAC or CB-treated DCs and expanded in IL-2 (as described in the legend to Fig. 3
). Primed T cells were assessed for regulatory function. (A, B) Freshly isolated CD4+ T cells were cultured alone (target x1), together with target cells (target x2) or primed T cells at an E:T cell ratio of 1:1 or 1:3, and stimulated for 4 days with anti-CD3 immobilized on a mouse L fibroblast expressing CD32/CD80 (left panel) or stimulated for 3–5 days with anti-CD3, anti-CD28, and mitomycin C-treated monocytes (right panel). Exogenous IL-2 was added during the E:T cell coculture (B). Cell proliferation was assessed by 3H-thymidine uptake for the last 6 h of cocultures. IFN- production was measured at the end of 3–5 days of culture. Data are presented as one representative experiment of five. (C) Foxp3 expression was determined by Q-PCR in T cells primed with CB-treated DCs after expansion in IL-2. CD4+ CD25+ T cells were used as a positive control. Data are presented as mean ± SD of three independent experiments. (D) Phenotypic analysis of suppressor CD4+ T cells. Expression of CD25, CTLA-4, CD103, CCR4, and CCR5 (left panel) and GITR (right panel) was analyzed on T cells at the end of expansion in IL-2. Data are one representative experiments of five. Numbers on the right panel indicate the frequency (%) of GITR-positive cells among CD4+ T cells.
|
We thus propose that direct exposure to commensal flora in the absence of inflammation rendered DCs phenotypically and functionally mature. Regardless of their cytokine and chemokine profile, all CB-treated mDCs induced hyporesponsive T cells that acquired suppressive activity, a mechanism whereby DCs may participate to the maintenance of gut homeostasis.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
DCs play a key role in eliciting adaptive immune response and tolerance. A common view proposes that DCs exist in two functional states (mature vs. immature) to exert opposite functions [25 ]. iDCs that reside in peripheral tissues express low cell-surface levels of costimulatory molecules and are tolerogenic. The DC maturation status, which is correlated to immunogenicity, is generally defined according to several criteria: mDCs up-regulate costimulatory molecules and MHC class II expression; adapt their chemokine receptors and migrate to the regional lymph node; lose their antigen uptake capacity and acquire antigen-presenting function and ability to induce naïve T cell proliferation; and drive T cell polarization toward Th1, Th2, or Th17 [6 , 25 ]. Although DCs should receive at least one and preferentially two TLR signals to sustain Th1 responses [26 27 28 ], our present data indicated that CB-treated DCs fulfill the three first criteria and thus, may be considered as functionally mature cells. According to the cytokine profile evoked by CB strains, these mDCs were equipped to drive Th1 or Th17 responses [28 , 29 ]. As depicted in Figure 2H , L. casei and S. thermophilius triggered high amounts of IL-12 and CXCL10 and low IL-10. Gram+ and gram– CB strains induced substantial amounts of IL-1β and IL-6 and low TGF-β release. Yet, none of the CB-primed DCs generated Th1 or Th17 effector T cells. TGF-β is required together with IL-6 for murine Th17 development, whereas it inhibits human Th17 polarization [29 , 30 ]. Moreover, the use of monocyte-derived DCs instead of conventional blood DCs or monocytes may have precluded Th17 polarization [29 ]. Thus, CB-treated, functional mDCs were not considered as immunogenic. Our present data are in apparent contradiction with some reports indicating that CB instruct DCs to promote naïve T cell differentiation into Th1 [19 ]. The latter studies were all carried out with DCs activated in the presence of inflammatory cytokines and FCS-containing medium. Hence, serum promotes DC maturation [31 ].
Monocyte-derived mDCs that do not prime T cell responses and instead induce CD8+ T cell tolerance have been described previously [32 ]. In other instances, phenotypical mDCs that do not release the IL-12p70 proinflammatory molecule are designated "semi-mature" or "partially mature" tolerogenic DCs [25 , 33 ]. In mice, the CD103+ CX3CR1+ tolerogenic DC subset, residing in the lamina propria, expresses large amounts of MHC class II, CD80, CD86, as well as CCR7, a prerequisite for their constant trafficking to the lymph node at steady-state, whereby they promote the conversion of naïve T cells into Foxp3highCD103+ Tregs [9 , 34 ]. In fact, the tolerogenic DC task in the lamina propria is shared by a particular macrophage subset (CD11b+CD11c–programmed death ligand 1+) that secretes a large amount of TGF-β and IL-10 [35 ]. Interaction of CD103 on DCs and E-cadherin on epithelial cells allows them to be in close contact to and be conditioned by gut epithelium [36 , 37 ]. Under steady-state conditions, thymic stromal lymphopoietin released by gut epithelial cell lines down-regulates IL-12 production by DCs and skews human T cell responses toward Th2 or Tregs, delineating an additional mechanism to ensure gut homeostasis [38 ].
Here, we showed that CB-treated DCs did not require prior exposure to epithelium-conditioned medium, TGF-β, or IL-10 to convert naïve CD4+ T cells into hyporesponsive T cells that exerted suppressive function. Our newly described Tregs are Foxp3lowCD25low; they differed from Tr1 (IL-10-producing cells) or thymic-derived Tregs (CD4+CD25highFoxp3high) and were characterized by a moderate but significant up-regulation in GITR expression and lack of Th1 and Th17 cytokine profile. Global gene array analysis indicated T cells primed with L. casei or B. animalis-stimulated DCs displayed a homogenous gene profile that instead can be distinguished from T cells primed with SAC- or polyinosinic:polycytidylic acid-activated DCs (data not shown). Yet, the nature of and the precise inhibitory signals used by these newly generated suppressor T cells to exert their regulatory function remain to be identified. Of note, addition of IL-2 during T suppressor/T effector cocultures overcame the inhibition of cell proliferation and IFN- production. These data corroborate with recent studies demonstrating that CD25+CD4+Foxp3+ T cells exert their suppressive function by inducing target cell apoptosis through IL-2 deprivation [39
]. IL-2 was shown to inhibit commensal-driven intestinal inflammation. This occurs via TLR-independent pathways, as triple IL-2–/–, MyD88–/–, and IL-10–/– mice still develop severe IBD, strongly in favor of a critical role of Treg in controlling the disease [4
]. However, MyD88–/– and IL-10–/– mice are resistant to chemically induced colitis, underscoring the importance of a direct commensal flora, TLR-dependent mechanism in the protection of gut mucosa [4
]. Probiotics administration ameliorates established murine colitis by inducing Tregs expressing TGF-β [40
]. Taken all, it would be difficult to envision an anti-inflammatory mechanism that is unique to TLR agonists in CB, as all TLR ligands are shared by commensal bacteria and pathogens.
Direct immunoregulatory properties of probiotic-treated DCs have also been documented in vivo. Murine bone marrow-derived DCs pulsed with two different Lactobacilli strains protect mice against trinitrobenzene sulfonic acid-induced colitis [41 ]. Such a protective effect is abolished when DCs are generated from MyD88–/–, NOD2–/–, and TLR2–/– mice, providing direct evidence that DCs require a pattern recognition receptor-mediated microbiota signal to display anti-inflammatory function. TLR2 signal appears to be implicated in the ability of Bifidobacterium and not Lactobacilli to prime human DCs for the generation of IL-10-producing Tr1 cells in the presence of maturation factors and superantigens [18 , 20 ]. However, probiotic administration to IL-10-deficient mice that develop spontaneous colitis could resolve the disease, suggesting the IL-10-independent mode of action of probiotics [42 ]. Yet, Tr1 may develop independently of IL-10 in vivo [43 ]. Our data indicate that direct contact between CB and DCs in the absence of exogenous inflammatory mediators during DC activation failed to induce Tr1 but still promoted the generation of hyporesponsive suppressor T cells. In support to our findings, regular intake of L. rhamnosus in humans leads to general T cell unresponsiveness [44 , 45 ].
Whether probiotics signal through TLR and/or NOD on epithelial cells and DCs, we propose that direct CB uptake by DCs is sufficient to trigger phenotypic maturation and cytokine release but insufficient to drive effector Th1/Th17 cell responses. Rather, CB instruct DCs to generate immunosuppressive T cells at steady-state, and as such, probiotics may be helpful to enforce gut tolerance.
Received January 9, 2008; revised April 24, 2008; accepted April 25, 2008.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
receptor enables human dendritic cell maturation with IL-12p70 production and immunity to antibody-coated tumor cells Proc. Natl. Acad. Sci. USA 102,2910-2915This article has been cited by other articles:
 |
|
 |
|
  F. Hacini-Rachinel, S. Nancey, G. Boschetti, F. Sardi, R. Doucet-Ladeveze, P.-Y. Durand, B. Flourie, and D. Kaiserlian CD4+ T Cells and Lactobacillus casei Control Relapsing Colitis Mediated by CD8+ T Cells J. Immunol., November 1, 2009; 183(9): 5477 - 5486. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
