| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Institut National de la Santé et de la Recherche Médicale (INSERM) U774, Institut Pasteur de Lille, Lille, France;
INSERM U564, University Hospital of Angers, Angers, France;
INSERM U459, Centre Hospitalier Régional Universitaire (CHRU), Lille, France;
Plateau d’Imagerie Cellulaire, Institut de Biologie de Lille, Lille, France;
¶ Centre d’Immunologie Pierre Fabre, St Julien en Genevois, France; and
|| Clinique des Maladies Respiratoires, CHRU, Lille, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Migration is an integral part of DC functions (5). Precursor cells first migrate from the bone marrow to resident sites in tissues such as lung. At steady state, due to the environment, immature DC and/or their precursors are continuously recruited to the airway mucosa where inhaled Ags are sampled (6). Upon activation by inflammatory mediators or microbial components, they undergo a complex process of maturation allowing their migration toward the regional lymph nodes (7) through a modification of adhesion molecule and chemokine receptor expression (8). Moreover, DC maturation is associated with high surface expression of MHC and costimulatory molecules, as well as secretion of immunomodulatory cytokines refers mainly to IL-10 and IL-12. These factors drive the differentiation and the polarization by DC of naive T cells into Th1, Th2, or alternatively into regulatory T cells. Recent data suggest that DC behavior is locally controlled by lung environment and particularly by bronchial epithelial cells (BEC) (9, 10, 11). In addition to GM-CSF, these BEC produce several cytokines and chemokines and express adhesion molecules potentially involved in DC traffic, as shown after exposure to diesel exhausted particles or allergens (6, 11). In both allergic patients and nonatopic subjects, allergen-exposed BEC trigger the recruitment of monocyte-derived DC (MDDC) precursors through the secretion of CCL2, CCL5, and CXCL10. In contrast, the migration of CD34+-derived LC precursors, depending on CCL20 secretion, is only observed with BEC from allergic patients. The recruitment of DC precursors by BEC suggests that local DC differentiation/maturation could be regulated by a confined crosstalk with BEC (12).
In this context, we demonstrated that BEC bind and are directly activated by outer membrane protein A (OmpA) from Klebsiella pneumoniae (KpOmpA), a mechanism implicated in neutrophil migration and the initiation of the innate immune response (13). OmpA has been shown to bind to and activate macrophages and immature MDDC in a TLR-2-dependent manner (14, 15). Moreover, KpOmpA is a carrier molecule suitable to generate a CD4+ and CD8+ T cell responses in different models of immunization (16, 17, 18, 19, 20). This model of exposure to a pathogen-associated molecular pattern (PAMP) derived from a pathogen with a lung tropism appears to us particularly relevant to evaluate the role of BEC in linking innate and adaptive immunity within airway mucosa.
To define this role, we studied in vitro the crosstalk between BEC and DC after exposure to KpOmpA, and particularly the regulation of DC traffic and functions by BEC. For this purpose, we analyzed the capacity of human BEC upon exposure to KpOmpA, to produce chemokines, to control the recruitment and the differentiation/maturation process of DC in an in vitro model of reconstituted bronchial epithelium. The in vivo relevance of this mechanism was also defined. The results showed that bronchial epithelium regulates the functions of DC and that this process might affect the development of the T cell response.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
All procedures were reviewed and approved by the Hospital Institutional Review Board and written informed consent was obtained from all subjects included in the study.
Human bronchial biopsies were obtained by fiberoptic bronchoscopy from 25 nonatopic patients who were being investigated for idiopathic chronic cough (n = 5) or for bronchopulmonary carcinoma (n = 20). For 9 patients, the diagnosis of cancer was not confirmed. For the patients without tumor, there is no obvious abnormality in the airways. There are 12 smokers among these patients with a smoking history of 11 ± 3 packs/yr (mean ± SEM). These patients did not receive any anti-inflammatory or immunomodulatory medication. In patients with cancer, biopsies were collected far from the carcinoma.
The human bronchial epithelial cell line BEAS-2B and primary BEC were cultured as previously described (13). BEC were exposed to different concentrations of KpOmpA. This protein was produced according to pharmaceutical quality standards, intended for clinical trials in human (14). Supernatants of BEC were collected after 24 h of incubation.
Preparation of monocyte-derived dendritic cells and naive T cells
Blood monocytes were purified by positive selection over a MACS column using anti-CD14-conjugated microbeads (Miltenyi Biotec) and were differentiated into DC by standard procedures (21). Briefly, immature DC (high expression of CD1a and CD11c) were obtained after culture for 6 days in RPMI 1640 medium supplemented with 10% heat-inactivated FCS (Invitrogen Life Technologies) and containing 10 ng/ml IL-4 and 25 ng/ml GM-CSF (R&D Systems). Mature MDDC were obtained after 24 h of incubation with 1 µg/ml LPS (from 055B5 Escherichia coli strain; Sigma-Aldrich).
Naive Th cells were isolated from the CD14– fraction by negative selection using the CD4+ T cell isolation kit associated with CD45RO microbeads (Miltenyi Biotec). After magnetic selection of memory T cells, the negative fraction contained >95% of CD4+ CD45RA+ cells, as determined by flow cytometry (data not shown).
Chemokine and cytokine measurements
The concentrations of CCL2/MCP-1, CCL5/RANTES, CCL20/MIP-3
, and CXCL10/IP-10 chemokines were determined in BEC supernatants by ELISA (all obtained from R&D Systems). Murine CCL20 was measured in lung extracts and bronchoalveolar lavage (BAL) fluids (R&D Systems). IL-10, IL-12p70 (Diaclone), IL-6, CCL17/TARC, and CCL22/MDC (R&D Systems) were measured in BEC/DC coculture supernatants, and IL-4, IFN-
, and IL-10 (Diaclone) in DC/T cell coculture supernatants, by sandwich enzyme immunoassay, as described by the manufacturer.
Boyden-type microchamber chemotaxis assays
Each BEC supernatants (1/20 in RPMI 1640 medium containing 0.1% FCS) and CCL5, used as positive control (200 ng/ml; R&D Systems), were added to the lower wells of the chemotaxis chamber (48-well Boyden microchamber; Neuroprobe). MDDC migration (5 x 104 cells/well in 50 µl of RPMI 1640 medium, 0.1% FCS) was performed through a standard 5-µm pore filter (Neuroprobe), at 37°C for 1 h and 30 min. Migrated cells were counted after staining on the lower side of the filter in three randomly selected high power fields (magnification x40). Each assay was performed in quadruplicate. Results are expressed as the difference between mean numbers of cells per high power fields minus the negative control (medium alone).
Identification of BEC-derived mediators implicated in DC precursor recruitment
Chemokine depletion of BEC supernatants was performed with protein G-Sepharose affinity columns (Amersham Biosciences) preincubated with anti-CCL5 (BD Pharmingen) and/or anti-CXCL10 Abs (BD Pharmingen) or rabbit IgG as control (1 µg/µl protein G-Sepharose). Solutions of rhCCL5 and rhCXCL10 (200 ng/ml; R&D Systems) were similarly depleted to control the efficacy of the depletion (78 and 75%, respectively; data not shown).
For neutralization experiments, BEC supernatants were incubated with neutralizing anti-CCL2 (R&D Systems) and anti-CCL7 Abs (BD Pharmingen) (10 µg/ml) or with rabbit IgG, for 1 h at 37°C. Efficacy of neutralization was checked using rhCCL2 or rhCCL7 solutions (200 ng/ml; R&D Systems): 85 and 80% of neutralization were obtained respectively (data not shown).
BEC/DC precursors coculture model
BEC were cultured on the lower side of a 5-µm pore filter coated with collagen G matrix (type I and III collagen) (Biochrom KG). After confluence, BEC were starved overnight in 50% RPMI 1640 medium, 0.1% FCS, and transepithelial resistance was controlled (data not shown). Biotinylated KpOmpA (20 µg/ml) was added in the lower chamber containing BEC, and day 3 MDDC precursors (1 x 106 cells) in the upper chamber for 24 h in RPMI 1640 medium containing 0.1% FCS. To evaluate the role of ICAM-1, BEC were preincubated with blocking anti-ICAM-1 Ab or an isotype control (5 µg/ml; R&D Systems) for 2 h, and washed before coculture with MDDC precursors.
Confocal microscopy and flow cytometry
The epithelial layer containing migrated MDDC precursors was assessed by confocal microscopy. MDDC precursors were stained with FITC-labeled anti-CD11c mAb, BEC with PE-labeled anti-CD49c mAb or the corresponding isotype controls (BD Biosciences), and biotinylated KpOmpA was revealed with allophycocyanin-labeled streptavidin. The epithelial layer containing migrated MDDC precursors was assessed by confocal microscopy (Imaging Core Facility of Calmette Campus). Imaging was done using a DM-IRE2 inverted microscope with SP2-AOBS scan-head (Leica). Acquisitions were performed using a 100x, 1.4 NA objective. The laser excitation wavelengths were selected with an AOBS, and their power was modulated by an AOTF (AOBS and AOTF are two specific acousto-optical systems). Excitation power was between 100 and 400 µW. Three-dimensional reconstruction results of the epithelial layer were obtained using Imaris software (Bitplane). KpOmpA capture was also quantified by flow cytometry in comparison with a control, a biotinylated-mouse IgG1 isotype control (n = 3).
Day 3 MDDC precursors, and DC from the DC/BEC cell coculture model were assessed for phenotypic analysis. Epithelial layer containing migrated MDDC precursors was dissociated using trypsin solution (Invitrogen Life Technologies). Cells were stained with allophycocyanin-labeled anti-CD11c, FITC-labeled anti-CD86, anti-HLA-DR and anti-CD1a, PE-labeled anti-CD80, anti-CD54, and anti-CD40 mAbs or with the isotype controls (BD Biosciences) and with PE-labeled anti-CCR7 mAb (R&D Systems), and analyzed by flow cytometry (FACSCalibur; BD Biosciences). To measure DC recruitment in cocultures, the percentage of intraepithelial DC (CD11c+ cells) was determined since the number of epithelial cells collected on each Transwell did not differ. Moreover, the level of DC markers was reported as the difference between median fluorescence intensity (MFI) with specific Ab minus the isotype control MFI (
MFI).
Evaluation of DC ability to induce T cell proliferation and polarization
MDDC were collected after 24 h coculture of DC precursor (day 3) with BEC or after culture in medium alone. For the DC/BEC cell coculture model, we collected both the DC present on the basolateral pole of BEC, by vigorous washing of the upper chamber of the Transwell and the epithelial layer. In this case, we positively selected the DC with CD11c microbeads to remove the BEC. Importantly this treatment does not modify the DC function. The purity of this preparation was up to 92%. In a first set of experiments, basolateral DC were incubated with unlabeled autologous naive T cells (DC/T lymphocyte ratio: 1:20) in 1 ml of RPMI 1640 medium supplemented with 10% FCS during 5 days. Moreover, we evaluated the capacity of the different DC to induce the proliferation and the cytokine production of heterologous CFSE-labeled T cells. For both experiments, T cells were collected after a 5-day incubation and resuspended at 2 x 106 cells/ml in fresh medium. Restimulation was performed by the addition of anti-CD3 plus anti-CD28 mAbs (2 µg/ml) (BD Biosciences). Cytokine production was first analyzed by ELISA in 24-h supernatants for unlabeled T cells. Cytokine synthesis was also measured after a 6-h incubation by intracellular flow cytometry with allophycocyanin-labeled anti-IL-4, anti-IL-10, and anti-IFN- mAb for CFSE-labeled heterologous T cells identified with PE-Cy5-anti-CD4 mAb (all obtained from BD Biosciences).
KpOmpA intranasal injection in mice
Female BALB/c mice, 6–10 wk old (IFFA Credo), were anesthetized by i.p. injection of ketamine/xylazine (Sigma-Aldrich) (2.5% v/v in PBS); 30 µl of BSA (100 µg) or KpOmpA solution (100 µg) was administrated intranasally on days 0, 1, and 2. Analyses were performed 24 h later. Experiments were approved by the animal ethical committee.
Cell infiltrate and inflammation in mouse lung tissue
BAL was first collected; then, right lungs were collected in PBS and mechanically dissociated to extract proteins from lung tissue. With the left lobe, total cells were isolated and then stained with allophycocyanin-labeled anti-CD11c and FITC-labeled anti-I-Ad mAbs, or with the isotype control (BD Biosciences), and analyzed by flow cytometry. Results are expressed as the percentage of CD11c+ I-Ad+ cells in the lung.
Left lung specimens were fixed with Immunohistofix and embedded in Immunohistowax (Aphase). After permeabilization, sections were incubated overnight with purified anti-mouse I-Ad (BD Biosciences) and biotinylated rat anti-mouse CD11c mAb and with rat anti-mouse CCL20 mAb (R & D Systems). For CCL20, Ab binding was detected after a 2-h incubation with biotin-conjugated goat anti-rat IgG (dilution: 1/400; Sigma-Aldrich) at room temperature, followed by incubation (or directly incubated for the biotinylated primary Ab) with Vectastain ABC kit for 30 min (Vector Laboratories). Color development was obtained with a Fast-Red solution (Sigma-Aldrich). Counterstaining was performed using Gill’s hematoxylin (Labonord).
Statistical analysis
Results are expressed as mean ± SEM. Statistical analysis was performed using the Wilcoxon test for paired data or the Mann-Whitney U test for unpaired data (SPSS Software, Windows).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
We previously described that KpOmpA induced the production of CXCL1 and CXCL8 by BEC, two chemokines involved in neutrophil recruitment (13). KpOmpA also induced a dose-dependent production of CCL2, CCL5, CCL20, and CXCL10 but not CCL7 by BEAS-2B cells after 24 h of stimulation (with a maximal activity at 20 µg/ml) associated with specific mRNA expression at 6 h (data not shown). Similar results were obtained with BEC primary cultures. Indeed, 20 µg/ml KpOmpA significantly increased the production of CCL2, CCL5, CCL20, and CXCL10 compared with unstimulated cells after 24 h (Fig. 1A). In our hands, LPS was a poor inducer of these chemokines in BEC compared with unstimulated cells (CCL2: 68 ± 12 vs 57 ± 20; CXCL10: 68 ± 48 vs 66 ± 33; CCL20: 1295 ± 110 vs 1117 ± 240, respectively). Taken together, these data show that KpOmpA triggers the production by BEC of chemokines potentially involved in DC migration.
|
We next tested the effects of supernatants from BEC primary culture on the migration of MDDC precursors (days 1, 3, 5), immature (day 6), and mature MDDC obtained after 24 h of stimulation with LPS. The precursors and the immature DC were responsive to CCL5 with a maximal activity at day 3 and mature DC to CCL19 (data not shown). Whereas supernatants from unstimulated cells had no marked effect, KpOmpA-treated BEC supernatants have a strong chemotactic activity on MDDC precursors taken at day 3, and at a lower level at day 5 (p < 0.05 vs unstimulated BEC) (Fig. 1B). In contrast, supernatants of KpOmpA-activated BEC did not modulate the migration of immature (day 7) (Fig. 1B), day 1 precursors, or mature DC (data not shown). No migration of MDDC precursors was observed in response to CCL20 or KpOmpA alone (data not shown). As previously demonstrated (12), immunodepletion of KpOmpA-stimulated BEC supernatants with anti-CXCL10 and anti-CCL5 Abs reduced the number of recruited day 3 precursors (32 and 46% inhibition, respectively; Fig. 1C). Moreover, there is an additive effect of CXCL10 and CCL5 depletion on cell migration (76% inhibition). The neutralizing anti-CCL2 Ab had a weak effect on day 3 precursor migration (Fig. 1C) and neutralizing anti-CCL7 or -CCL20 Abs had no effect on chemotactic activity of BEC (data not shown).
MDDC precursors migrate into bronchial epithelial layer and capture KpOmpA
To confirm that KpOmpA modulated MDDC precursor migration, we used a polarized model of coculture on Transwell devices, using BEC and day 3 MDDC precursors. At baseline, approximatively 25% of the cells present in the epithelial layer were MDDC. Exposure to KpOmpA significantly increased the number of CD11c+ precursors (41 ± 5%) present within the epithelium (p < 0.05, Fig. 2A). In addition, intraepithelial MDDC (CD11c+ cells) captured KpOmpA as illustrated after three-dimensional image reconstruction, by the association between CD11c labeling and allophycocyanin-conjugated KpOmpA (Fig. 2B). Flow cytometry analysis revealed that 
25% of intraepithelial DC (Fig. 2D) captured KpOmpA. Basolateral DC were also labeled but with a lower intensity (Fig. 2C).
|
We next analyzed the differentiation of day 3 MDDC precursors, recruited by BEC within the epithelial layer (intraepithelial), present on the other side of the Transwell (basolateral) or cultured without BEC (Fig. 3). The expression of the co-stimulatory molecules CD80 and CD86 were significantly up-regulated on MDDC precursors in contact with the basolateral side of unstimulated BEC, compared with MDDC precursors cultured without BEC (Figs. 3 and 4). This effect was more pronounced on intraepithelial DC (p < 0.05) (Fig. 3). The same profile was obtained for HLA-DR, ICAM-1, and the maturation markers, namely CD83 and CCR7 (p < 0.05). In contrast, BEC did not significantly modulate the expression of CD1a as well as the level of CD11c expression by DC (not shown). Addition of KpOmpA to MDDC precursors cultured without BEC increased the expression of CD80, CD86, HLA-DR, ICAM-1, CD83, and CCR7 (p < 0.05), but not CD1a. KpOmpA exposure has an additive effect with BEC on the modulation of DC phenotype, particularly on the expression of CD80 and HLA-DR (p < 0.05) and, in a lower manner, of CD83, CD86, and CCR7. As previously demonstrated, KpOmpA increased ICAM-1 expression on BEC or MDDC precursors alone (13) and the coculture still enhanced ICAM-1 expression on both types of cells (data not shown). Since 
2 integrin-ICAM-1 interactions result in cell activation, we evaluated the potential implication of this adhesion molecule in the crosstalk between DC and BEC. Pretreatment of BEC with neutralizing anti-ICAM-1 Ab significantly reduced the expression of CD80, CD86, and HLA-DR (46, 68, and 43% inhibition, respectively) whereas this treatment did not block the effect of KpOmpA on DC alone (Fig. 4). In our hands, supernatants of unstimulated and KpOmpA-activated BEC had a weak effect on DC phenotype compared with the coculture model (data not shown).
|
|
|
First, the capacity of basolateral MDDC cocultured with BEC to stimulate T cells was evaluated by incubating these cells with autologous CD4+CD45RA+ T lymphocytes for 5 days. The MDDC cocultured with BEC did not modify the production of IL-4 and IFN-, but increased IL-10 production by T cells after restimulation with anti-CD3 plus anti-CD28 mAbs. KpOmpA activation of MDDC alone or in cocultures did not significantly modulate the capacity of DC to regulate IL-4 and IFN- secretion whereas there is a trend to increase the levels of IFN- (Fig. 6A). In contrast, treatment with KpOmpA did not affect IL-10 production in coculture with BEC whereas it increased its production with day 3 MDDC precursors.
|
production was also increased by KpOmpA exposure; however, it does not differ with the different kinds of DC (data not shown). IL-4 and IL-10 production was similar in the presence of unstimulated basolateral DC and DC alone whereas KpOmpA induced a strong increase of both positive IL-4 and IL-10 T cell percentage only with basolateral DC. In contrast, intraepithelial DC in unstimulated conditions have an enhanced activity to induce T cell cytokine production whereas they do not respond to KpOmpA stimulation. The production of IFN- was not markedly affected by the coculture with BEC (data not shown). Intranasal injection of KpOmpA induces the recruitment of CD11c+ cells and I-Ad+ cells into the airways
In a previous study, we demonstrated that one intratracheal injection of KpOmpA triggered neutrophil influx to bronchial epithelium and into the bronchial lumen (13) whereas no significant migration of myeloid DC was observed in this situation (data not shown). However, after three injections of KpOmpA, no more influx of neutrophils was observed in the BAL (3.3% at day 3 compared with 1.9% with BSA and 44% at day 1 with KpOmpA) or in the bronchial wall (data not shown). In contrast, the number of CD11c+ I-Ad+ cells in lung homogenates was significantly increased after KpOmpA exposure (Fig. 7A). Moreover, important infiltrates of CD11c+ cells and I-Ad+ cells were present mainly in bronchial mucosa (Fig. 7, B and C). In contrast, only a few infiltrating leukocytes were observed in lung sections of mice primed with BSA (Fig. 7C).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
We first focused on the production of chemokines involved in DC migration, and showed that KpOmpA increased the secretion of CCL2, CCL5, CCL20, and CXCL10 by BEC. These results extended previous data showing that KpOmpA directly stimulates BEC to produce cytokines and chemokines, a process probably related, at least, to NF-
B activation (13). Moreover, KpOmpA exposure of BEC led to an increased recruitment of MDDC precursors, through the coordinate activity of CXCL10 and CCL5. Surprisingly, CCL2 secretion is not involved in the effect of KpOmpA-stimulated BEC, in contrast to the data obtained with allergen-exposed BEC (12). Compared with Der p1-allergen, KpOmpA is a stronger inducer of CXCL10 production by BEC, this effect compensating the lack of CCL2 activity in KpOmpA-derived MDDC precursor recruitment. In addition, as previously demonstrated in the skin, we suspect that KpOmpA is as a migration-promoting stimulus for LC (19), a cell only responding to CCL20 (23) in bronchial mucosa. CCL20 expression, but not that of inflammation-related chemokines, determines LC homing in the epidermis (23), the tonsils and probably also in the lung (24). CCL20 secretion by BEC was previously reported after exposure to allergen, but only in allergic asthmatic patients and to diesel exhaust particles (11, 12). Therefore, CCL20 secretion is only induced by strong signals in BEC such as pathogen exposure or by proinflammatory mediators present within the lung environment. In summary, KpOmpA represents a strong inducer of myeloid DC migration within the bronchial epithelium.
After pathogen exposure, the lung vascular compartment is enriched in a population of mononuclear cells (25), which are able to differentiate into MHC class II+ DC when exposed to the appropriate growth factors, including GM-CSF, a cytokine produced by BEC (26). This enriched population of DC precursors provides a readily available source to replenish pulmonary DC, both as part of the normal turnover of these cells and during inflammatory reactions (25, 27). In this study, we showed that airway epithelium controls the mobilization of these DC precursors through chemokine secretion. These results were confirmed in an in vitro model of polarized epithelium where DC precursors invaded the epithelial layer after KpOmpA exposure. Randolph et al. (28) reported that monocytes transmigrating through endothelium differentiated into DC after a complex set of signals. Whereas endothelium delivers the first signal for DC differentiation/maturation process, we hypothesized that airway epithelium may generate the second one (29). The importance of this crosstalk has been suggested in intestinal mucosa where transepithelial dendrite formation appears dependent on tight junction protein expression and on the chemokine CX3CL1 (30, 31). In our model of airway epithelium, we showed that BEC control the recruitment of MDDC precursors, the capture of KpOmpA, and the differentiation/maturation process. After 24 h of incubation with BEC, MDDC expressed higher levels of costimulatory molecules and produced higher amounts of chemokines than DC differentiated without BEC. Increased expression of CD83 and CCR7 confirmed DC maturation, and suggested that these DC could migrate to the draining lymph nodes, in response to CCR7 ligands. Moreover, ICAM-1 neutralization on BEC strongly reduced the effect of bronchial epithelium on CCL17 and CCL22 secretion and on the phenotype of recruited MDDC. This suggests that ICAM-1/LFA-1 interactions between BEC and DC precursors are involved in DC activation as well as in the subsequent recruitment of T cells and eosinophils (32). Local administration of KpOmpA in the experimental model also shows that this PAMP induces the recruitment of myeloid DC and the production of CCL20 by BEC in vivo. We have also reported that KpOmpA-exposed animals have an increased expression of ICAM-1 (13) suggesting that both CCL20 and ICAM-1 are involved in the in vivo DC/BEC crosstalk. GM-CSF production by BEC is also important in the interactions between DC and BEC, because addition of neutralizing anti-GM-CSF Ab during the coculture inhibited the secretion of CCL17 and CCL22 whereas this treatment did not affect the expression of membrane DC markers (data not shown). Other adhesion molecules or tight junction proteins may be also involved in the DC/BEC interactions. Because some DC function such as cytokine production (particularly IL-10 and IL-12) are down-regulated by BEC, additional experiments are required to evaluate the potential implication of BEC-derived inhibitors in this process (such as arachidonic acid metabolites).
An intriguing question concerns the consequence of DC recruitment and maturation into airways from healthy donors. In mice, airway exposure to an Ag, such as OVA, induces DC-dependent development of a specific tolerance through ICOS-L expression and IL-10 production (33). In contrast, association of OVA with TLR2 or TLR4 ligands triggers airway sensitization and the development of an allergic reaction, after subsequent OVA instillation (34). Concerning the polarization of Th cell, murine resting airway DC preferentially skews the T cell response toward a Th2 profile suggesting that lung environment favors this type of response (35, 36). Thus, we evaluated the capacity of MDDC precursors cocultured with BEC to activate and polarize T cells in the context or not of an exposure to a PAMP. At steady state, BEC promoted an increased production of IL-10 by T cells, suggesting that BEC may be involved in tolerance induction. In the absence of proinflammatory signals, partially mature DC probably induce an abortive proliferative response of unfit T cells. Airway exposure to a protein Ag in mice leads to the development of Ag-specific tolerance associated with induction of regulatory T cells, as shown by Akbari et al. (33). In this situation, tolerogenic lung DCs have a phenotype of mature DC expressing IL-10 and ICOS-L. We have also analyzed the T cell response in a situation with stronger T cell activation: a MLR. In this case, coculture with BEC did not affect their capacity to induce T cell proliferation whereas their ability to induce IL-4 and IL-10 production by T cells was modified. Basolateral DC responded more vigorously to KpOmpA whereas intraepithelial DC had a high basal activity but did not respond anymore to KpOmpA. IFN- production was not markedly affected by the coculture with BEC. The fact that intraepithelial DC appear more mature than the basolateral cells (Fig. 3) could explain the difference in cytokine production by T cells. However, these differences are not sufficient to modify the T cell proliferation. These data also demonstrated that bronchial DC might favor the development of Th2 cells as previously reported in experimental models (35, 36). The implicated mechanism remains to be determined. In our model, IL-10 production by DC was not detected whereas IL-12 (a pro-Th1 factor) secretion seemed to be down-regulated. In our in vivo model, KpOmpA induced DC recruitment within bronchial mucosa and facilitated the development of IgG1 (an isotype associated with a Th2 response) Ab production when this PAMP was co-administrated intranasally with OVA in naive mice (our unpublished observations). These results suggest that KpOmpA, through its effect on BEC, may break down or block the Ag-specific tolerance, and induce a local immune response to the Ag coupled to KpOmpA according to the presence or not of this TLR ligand. Taken together, these data show that BEC not only regulate inflammatory processes (37) but also control the local immune response by its action on DC traffic and functions.
In addition to its role in the innate immune response, bronchial epithelium may link innate and acquired immunity through its dialog with DC. Moreover, the BEC/DC crosstalk may be implicated in the control of airway tolerance and, in the context of a PAMP exposure, may favor the development of the immune response. Thus, bronchial epithelium targeting represents a novel strategy in vaccination process via airway mucosa.
|
Acknowledgments |
|---|
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 S.T. was a recipient of a fellowship from Agence de l’Environnement et de la Maîtrise de l’Energie. C.S. was supported by Région Nord-Pas-de-Calais PhD fellowship. 
2 Address correspondence and reprint requests to Dr. Philippe Gosset, Institut National de la Santé et de la Recherche Médicale, Unité 774, Institut Pasteur, 1, rue du Professeur Calmette, BP 245, 59019 Lille cedex, France. E-mail address: Philippe.Gosset{at}pasteur-lille.fr
3 Abbreviations used in this paper: DC, dendritic cell; BAL, bronchoalveolar lavage; BEC, bronchial epithelial cell; KpOmpA, outer membrane protein A from Klebsiella pneumoniae; LC, Langerhans cell; DC, dendritic cell; MDDC, monocyte-derived DC; PAMP, pathogen-associated molecular pattern.
Received for publication December 5, 2005. Accepted for publication July 31, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
: II. functional analysis. Blood 90: 1458-1470. /CCL20 in response to cytokines and ambient particulate matter. Am. J. Respir. Cell Mol. Biol. 28: 648-654. is involved in the constitutive trafficking of epidermal langerhans cells. J. Exp. Med. 190: 1755-1768. is expressed at inflamed epithelial surfaces and is the most potent chemokine known in attracting Langerhans cell precursors. J. Exp. Med. 192: 705-718. This article has been cited by other articles:
|
|
 |
|
  S. Podgrabinska, O. Kamalu, L. Mayer, M. Shimaoka, H. Snoeck, G. J. Randolph, and M. Skobe Inflamed Lymphatic Endothelium Suppresses Dendritic Cell Maturation and Function via Mac-1/ICAM-1-Dependent Mechanism J. Immunol., August 1, 2009; 183(3): 1767 - 1779. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Rate, J. W. Upham, A. Bosco, K. L. McKenna, and P. G. Holt Airway Epithelial Cells Regulate the Functional Phenotype of Locally Differentiating Dendritic Cells: Implications for the Pathogenesis of Infectious and Allergic Airway Disease J. Immunol., January 1, 2009; 182(1): 72 - 83. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
6. *, p < 0.05, **, p < 0.01 compared with unstimulated cells. B, Migration of MDDC collected at different time points during the differentiation was evaluated in response to BEC supernatants. Results are expressed as the mean ± SEM of the number of cells by high power fields minus the medium alone (n = 5). C, The effect of CCL2 neutralization and depletion in either CCL5, CXCL10, and of both CCL5 and CXCL10 in KpOmpA-stimulated BEC supernatants was defined on day 3 precursor recruitment. Results are expressed as mean ± SEM (n = 5).
