Published online before print February 24, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Unité Mixte de Recherche 7151 Centre National de la Recherche Scientifique–Université Paris 7, and laboratoire d’Immunologie Cellulaire et Immunopathologie de l’Ecole Pratique des Hautes Etudes, Institut Universitaire d’Hématologie, hôpital Saint-Louis, France
1Correspondence: CNRS UMR 7151, Institut Universitaire d’Hématologie, Centre Hayem, hôpital Saint-Louis, 1 avenue Claude Vellefaux, 75475 Paris Cedex 10, France. E-mail: sarah.boudaly{at}paris7.jussieu.fr
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: granulocytes • phagocytosis • cell activation • human
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, IL-12, as well as CXC and CC chemokines such as IL-8/CXC chemokine ligand 8 (CXCL8), growth-related gene product /CXCL1, the 10-kDa interferon-
(IFN-)-inducible protein 10/CXCL10, macrophage-inflammatory protein-1 (MIP-1)/CC chemokine ligand 3 (CCL3), MIP-1ß/CCL4, MIP-3/CCL20, and MIP-3ß/CCL19 [2
3
4
5
]. Migration of other leukocytes, including lymphocytes, occurs with a time lag following accumulation of PMNs, the effector and regulatory functions of which could contribute to the downstream development of T lymphocyte-dependent immune responses [6
]. Dendritic cells (DCs) are instrumental in the uptake of antigenic material, which they process and present to CD4+ and CD8+ T lymphocytes [7 ]. Immature (i)DCs are distributed in peripheral tissues, where they capture apoptotic or necrotic cells and/or microorganisms. In case of injury or infection, a DC precursor influx occurs at the injured site within minutes to hours, depending on the nature of the tissue [8 9 10 ]. Uptake of antigenic material induces DCs to be activated and to migrate to secondary lymphoid organs, where they initiate activation of naive T lymphocytes. The nature of the signals delivered to DCs upon antigen uptake—nature of the microorganisms [11 ], necrotic or apoptotic cells [12 13 14 ], the microenvironment [15 , 16 ], and endogenous inflammatory danger signals [17 ]—is known to influence the phenotypic and functional changes of DCs, which are crucial for triggering activation of antigen-specific effector T lymphocytes or to induce tolerance [18 ].
The early and massive recruitment of PMNs at infected or injured tissues, the tissue distribution of iDCs, as well as the influx of DC precursors allow colocalization of both populations at the early stages of these inducible processes. Using freshly isolated PMNs and iDCs derived from monocytes isolated from the blood of healthy human donors, we analyzed the potential of PMNs to influence the phenotype and functions of DCs in a coculture system. In addition, heat-killed Candida albicans was chosen as a model microorganism and source of antigen to examine whether PMNs could transfer antigens to DCs for subsequent presentation to T lymphocytes.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
95%. Monocyte-derived DCs are referred to as DCs hereafter.
Isolation and characterization of PMNs
Less than 4 h after blood sampling, PMNs were separated from PBMCs by Ficoll-Paque centrifugation and lysis of red blood cells with buffered NH4Cl. After extensive washing, enriched PMNs were suspended in RPMI 1640 and 10% FCS and identified on morphology after May-Grünwald-Giemsa staining and by fluorescein-activated cell sorter (FACS) immunophenotyping (see below). To assess apoptosis, PMNs were incubated with the TACSTM Annexin-V-fluorescein isothiocyanate (FITC) detection kit (R&D System) or the Annexin-V-phycoerythrin (PE) detection kit I (BD Biosciences PharMingen, San Diego, CA), according to the manufacturers’ recommendations. Cell viability was evaluated by propidium iodide (PI) or 7-amino-actinomycin D (7-AAD) staining. Cells were examined with a FACSCalibur (Becton Dickinson, San Jose, CA) immediately after labeling.
To get live or apoptotic/dead PMN populations, freshly isolated PMNs were cultured for 24 h in complete medium, 10% FCS, and 10 ng/ml GM-CSF before they were fractionated into Annexin-V– (live) and Annexin-V+ cells by magnetic separation with Annexin-V Microbeads and MS separation columns (Miltenyi Biotec, Auburn, CA). PMN viability was then evaluated by Annexin-V-FITC/PI labeling and May-Grünwald-Giemsa staining. Only fractions with 
10% Annexin-V+ PMNs were referred to as "live," whereas Annexin-V+ apoptotic/dead cell fractions were referred to as "apoptotic" PMNs.
FACS analysis and cytology
Cells were harvested, washed twice in phosphate-buffered saline, 2% FCS, and labeled with PE-, FITC-, or allophycocyanin (APC)-conjugated monoclonal antibodies (mAb; listed in Table 1
) for FACS analysis. Intracellular labeling of IL-12p40/p70 was performed using the Cytofix/Cytoperm GolgiPlug kit assay (Becton Dickinson). Briefly, cells were incubated for 6 h with the GolgiPlug reagent (containing Brefeldin A) to inhibit protein transport. Cells were stained with mAb specific for surface markers and then fixed and permeabilized according to the manufacturer’s recommendation before labeling with an anti-IL-12p40/p70 antibody.
|
View this table: [in a new window] |
Table 1. List of the mAb Used
|
Coculture of DCs and PMNs
DCs were cultured separately or with freshly isolated, allogeneic PMNs, prepared, and characterized as described above. Cocultures were conducted for 18 h at a 10:1 PMN/DC ratio unless otherwise stated. Cultures were performed in 24-well tissue-culture plates (Dutscher S.A.) in 1.5 ml complete medium containing 10 ng/ml GM-CSF and 5 ng/ml IL-4. Allogeneic PMNs were used here, as autologous ones could not be kept cryopreserved for the 6 days needed to derive DCs from monocytes. In all cultures, total cell concentration was kept at 7.5 x 105 cells/ml. Under all the conditions used, the DC and PMN populations recovered were devoid of T lymphocytes, which could react to allogeneic cells. In some cases, 5 µg/ml anti-CD18 or irrelevant IgG1 mAb were added to cocultures.
To avoid cell contact, PMNs were cultured in 0.4 µm pore Transwell® chambers (Becton Dickinson), inserted into wells containing DC cultures. FACS analysis at the end of the culture showed the lack of PMNs mixed with DCs, indicating that the membrane totally blocked migration of both. GM-CSF and IL-4 were present in the medium on both sides of the membrane to ensure equal availability of the cytokines in both compartments.
DCs cultured with 100 ng/ml lipoplysaccharide (LPS; Sigma Chemical Co., St. Louis, MO), with or without 100 ng/ml IFN- (R&D Systems), were used as positive controls of DC activation in most experiments. Polyriboinosinic polyribocytidylic acid (Poly I:C; 2 µg/ml, Sigma Chemical Co.) was used as control in one set of experiments.
Mixed leukocyte reaction
Control DCs or DCs cocultured for 18 h with fresh PMNs at a 10:1 PMN/DC ratio and depleted from PMNs with CD15 Microbeads (Miltenyi Biotec) were mixed with cryopreserved, allogeneic T lymphocytes (105 DCs/106 T lymphocytes) in 1.5 ml complete medium, in which FCS was substituted by 10% heat-inactivated human AB serum (Etablissement Français du Sang). Supernatants were harvested after 48 h and 72 h and assayed for the presence of IFN- using an enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems).
T lymphocyte response to C. albicans antigens
PMNs were cultured for 18 h in complete medium plus 10 ng/ml GM-CSF in the absence (control PMNs) or the presence of heat-killed C. albicans (Ca-PMNs) CAF yeasts in a 5:1 yeast/PMN ratio (the CAF strain was kindly provided by Dr. Thierry Jouault, Institut Pasteur, Lille, France). Yeasts were killed by heating 20 x 106 yeasts/ml at 95°C for 15 min. After overnight culture, PMNs were sorted with CD15 Microbeads to remove free yeast, the remaining number of which was less than one yeast/10 PMNs, as assessed by microscopy. Live or apoptotic PMNs were then separated as described above and mixed with DCs for 3 h at 10:1 or 1:1 ratios, after which DCs were sorted with a FACSVantage (Becton Dickinson) or PMN-depleted with CD15 Microbeads. To avoid cell contacts, PMNs or Ca-PMNs were cultured in 0.4 µm pore Transwell® chambers (Becton Dickinson), inserted into wells containing DCs as described above. It was verified that the 0.4-µm pores did not allow passive transfer of free yeast in the DC compartment. The DCs were added to previously cryopreserved, autologous T lymphocytes (105 DCs/106 T lymphocytes) in 1.5 ml complete medium with 10% heat-inactivated human AB serum. Control DCs were incubated for 3 h with C. albicans (five yeasts:one DC) before being added to the T lymphocytes. Supernatants were harvested after 24 h, 48 h, and 72 h and assayed for the presence of IL-2 and IFN- using an ELISA kit (R&D Systems).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
95% were PMNs, as indicated by cytology, high and homogenous expression of surface CD15, CD11b, and CD18, and lack of detectable CD2+, CD14+, CD19+, and CD56+ cells (Fig. 1A
). After 6 days culture, CD14–CD1a+ DCs were >95% pure, with no detectable natural killer cells or T or B lymphocyte (Fig. 1B)
.
 View larger version (41K): [in a new window] |
Figure 1. Characterization of freshly prepared PMNs and DCs. (A) PMN phenotype and morphology. Dot-plots represent Annexin-V-FITC (AV) and PI staining and double-labeling with PE-conjugated anti-CD11b and FITC-conjugated anti-CD15 mAb. Quadrants are established according to negative control labeling. Histograms represent labeling for the indicated markers (shaded areas, specific labeling; dotted lines, isotype control labeling). May-Grünwald-Giemsa-stained PMNs (lower left panel) are shown at x100 original magnification. (B) DC phenotype. Dot-plots represent double-labeling with PE-conjugated anti-CD14 and FITC-conjugated anti-CD1a antibodies. Histograms represent labeling for the indicated markers as in A. (C) Evaluation of apoptotic and dead PMNs. Dot-plots represent AV and PI staining of freshly isolated PMNs (t0) or PMNs cultured for 18 h without (t18) or with GM-CSF and IL-4 (t18 GM-CSF+IL-4). Numbers in quadrants indicate cell percentages. Data are from one of 20 independent experiments.
|
Clustering and phenotypic changes of cocultured PMNs and DCs
We next cocultured fresh PMNs with DCs for 18 h at a 10:1 ratio. Characteristically, the cells then formed heterologous clusters, whereas they did not aggregate when cultured separately (Fig. 2A
). FACS analysis of the cocultures was performed after gating the FSClow/CD1a–CD15+ and FSChigh/CD1a+CD15– populations, corresponding to PMNs and DCs, respectively (Fig. 2B)
.
 View larger version (57K): [in a new window] |
Figure 2. Phenotypic and morphological characterization of the cells in cocultures of PMNs with DCs, which were cultured separately (control DCs and control PMNs) or cocultured for 18 h at a 10:1 ratio. (A) Photographs of cultures (x40 original magnification). (B) PMNs and DCs appear as distinct populations in FACS. Dot-plots represent forward-scatter/side-scatter (FSC/SSC) and CD1a/CD15 double-labeling of total nongated cells. Gates are set according to isotype control labeling. Numbers in quadrants indicate cell percentages. Data are from one of 10 experiments.
|
 View larger version (42K): [in a new window] |
Figure 3. Phenotype of cocultured PMNs and DCs, which were cocultured for 18 h at a 10:1 ratio. As negative controls, DCs or PMNs were cultured separately in the presence of GM-CSF and IL-4. DCs were cultured with LPS as DC activation control. (A) PMN labeling. Overlay histograms represent irrelevant (dashed lines) and specific labeling of CD1a–CD15+-gated cells from coculture (shaded areas) or control PMNs (bold lines). (B) Assessment of apoptotic and dead PMNs. Dot-plots represent Annexin-V-FITC and PI staining of the CD1a–CD15+ cell-gated population from PMNs cultured separately (Control PMNs) or with DCs (PMN/DC). Numbers in quadrants indicate cell percentages. (C) Markers of DC activation. Overlay histograms represent irrelevant (dashed lines) or specific labeling of CD1a+CD15–-gated cells from PMN/DC coculture (shaded areas), LPS-activated DC (thin lines), or control DCs (bold lines). As isotype control histograms strictly overlap (data not shown), only one of those is shown for the sake of clarity. Data are from one of 10 experiments.
|
|
View this table: [in a new window] |
Table 2. Effect of a Blocking Anti-CD18 Antibody on DC Marker Expression
|
 View larger version (55K): [in a new window] |
Figure 4. DC activation is elicited by live but not apoptotic PMNs, which were cultured for 24 h before sorting of Annexin-V– (live PMNs) and Annexin-V+ (apoptotic PMNs) and cultured with DCs at a 10:1 PMN/DC ratio. (A) Dot-plots represent Annexin-V-PE (AV) and 7-AAD staining of sorted PMNs. Numbers in quadrants indicate cell percentages. Photographs represent May-Grünwald-Giemsa staining (x100 original magnification) of the indicated populations. Overlay histograms represent irrelevant (dashed lines) or specific labeling of control DCs (bold lines) or DCs cocultured for 3 h (shaded areas) with live (upper) or apoptotic (lower) PMNs. Data are from one of four experiments. (B) Overlay histograms represent irrelevant (dashed lines) or specific labeling of control DCs (bold lines) or DCs cultured for 18 h (shaded areas) with LPS, live, or apoptotic-sorted PMNs. Data are from one of two experiments.
|
 View larger version (36K): [in a new window] |
Figure 5. Assessment of DC function.(A and B) Evaluation of IL-12-producing DCs, which were cultured for 18 h separately (control), with LPS + IFN-, or with PMNs (10:1 PMN/DC ratio), before labeling with anti-CD1a-APC-, anti-CD15-FITC-, and anti-IL-12-PE-conjugated mAb. PMNs were (A) freshly isolated or (B) sorted into Annexin-V– live PMNs and Annexin-V+ apoptotic PMNs as in Figure 4
. Dot-plots (gates are set according to isotype-control labeling) and histograms (dashed lines, control labeling; shaded areas, specific labeling) represent labeling of CD1a+-gated cells with the indicated mAb. Numbers in quadrants indicate cell percentages. Data are from one of (A) four or (B) two experiments. (C) Evaluation of the T lymphocyte stimulatory capacity. Allogeneic T lymphocytes were cultured alone (T) or with allogeneic PMNs (T+PMN), control DCs (T+DC), or DCs cocultured for 18 h with PMNs [T+(PMN DC)]. Controls also comprised DCs or PMNs cultured alone or together (PMNDC). IFN- in 48- and 72-h culture supernatant production was measured by ELISA. Data are from one of two experiments.
|
 View larger version (38K): [in a new window] |
Figure 6. HLA-DR and CD86 up-regulation on DCs depends on contacts with PMNs and requires CD18 ligation. DCs were cultured for 18 h separately or with 10 PMNs per DC under conditions that allowed or did not allow cell contact. (A) Representative labeling. Histograms represent irrelevant (dashed lines) or specific staining of CD1a+CD15–-gated cells from the coculture (shaded areas) and of nonstimulated (bold lines) or Poly I:C-activated (thin lines) DCs. Data are from one of five experiments. (B) Effect of an anti-CD18 mAb. DCs were cultured for 18 h with LPS (DCs+LPS) or with 10 PMNs per DC in the presence of the anti-CD18 mAb (DCs+PMNs+anti-CD18) or of an irrelevant IgG1 (DCs+PMNs+IgG1) at 5 µg/ml. Overlay histograms represent specific labeling of CD1a+CD15–-gated, nonstimulated, control DCs (bold lines), LPS-activated, or PMN-exposed DCs (shaded area), respectively. Dotted lines represent irrelevant labeling of control DCs (...) or LPS-activated or PMN-exposed DCs (.....). Data are from one of four experiments.
|
90% fresh PMNs were initially live (see Fig. 1A
and 1C
), percentages of early-to-late apoptotic PMNs increased in culture with or without DCs (Fig. 3B)
in a proportion that varied, depending on the donor. To analyze whether the presence of a low proportion of apoptotic PMNs was responsible for the DC phenotypic changes, DCs were cultured for 3 h or 18 h with sorted Annexin-V– (live) and Annexin-V+ (apoptotic) PMNs. As shown in Figure 4A
, PMNs referred to as live were nonapoptotic and exhibited the typical morphology of live PMNs, whereas PMNs referred to as apoptotic comprised >50% of Annexin-V+ cells, which displayed morphological features of apoptosis such as heterochromatin condensation. In contrast to live PMNs, enriched, apoptotic PMNs did not elicit HLA-DR and CD86 up-regulation nor any change of CD80 and CD83 expression on DCs (Fig. 4A
and 4B
; and data not shown) whatever the culture duration. That HLA-DR and CD86 up-regulation was less pronounced on DCs after 18 h than after 3 h coculture with live PMNs probably relates to the increased susceptibility to apoptosis of PMNs after magnetic sorting, which represents a drastic procedure for these cells. From these data, one may infer that the minority of apoptotic PMNs that appears during coculture cannot account for the DC phenotypic changes occurring in cocultures with fresh PMNs.
Regarding DC function, we examined whether fresh PMNs induced DCs to produce IL-12p40/p70. As PMNs can also produce IL-12 under some experimental conditions [6
], this was evaluated by intracellular labeling. Figure 5A
indeed shows the presence of IL-12+ cells among cocultured DCs but not DCs cultured separately. As indicated by the lack of IL-12+/CD15+ cells, IL-12 labeling could not be accounted for by the presence of residual PMNs in the CD1a+ cell-gated population. Again, in experiments with sorted Annexin-V– and Annexin-V+ PMNs, IL-12+ DCs were found only in cocultures with the live PMNs (Fig. 5B)
. Consistent with the phenotypical and functional changes, PMN-exposed DCs elicited greater IFN- production from allogeneic T lymphocytes than control DCs (Fig. 5C)
. This cannot be accounted for by the presence of residual PMNs, which do not stimulate T lymphocytes by themselves.
Altogether, these data indicate that cross-talk can take place between iDCs and PMNs. Indeed, DCs exposed to live PMNs undergo phenotypic and functional changes known to be associated with enhanced T lymphocyte-activating capacity. Interactions with DCs also affect PMNs in that they at least prevent phenotypic changes related with apoptosis.
Role of CD18 in PMN-induced DC activation
To investigate the mechanisms of these interactions, a Transwell® system was used, in which DCs and PMNs were separated by a 0.4-µm membrane (Fig. 6A
). Only CD40 up-regulation on DCs was then noted in cocultures relative to DCs cultured separately, whereas HLA-DR and CD86 expression was unchanged. In contrast, when allowed to contact PMNs, DCs also displayed increased HLA-DR and CD86 levels, as in the first series of experiments. These data indicate that CD40 up-regulation on DCs is induced by PMN-released, soluble factors, whereas that of HLA-DR and CD86 depends on direct contacts with PMNs.
CD11a/CD18 (lymphocyte function-associated antigen-1) and CD11b/CD18 (complement receptor 3) are the most important integrins involved in PMN adhesion and in transmitting activation signals to PMNs [27 , 28 ]. We thus examined whether these integrins played a role in the contact interactions of PMNs with DCs. Indeed, coculture in the presence of an anti-CD18-blocking mAb, but not of irrelevant, control IgG1, resulted in inhibition of HLA-DR and CD86 increased expression (Fig. 6B and Table 2 ), whereas CD40 up-regulation was unaffected (Table 2) .
This finding indicates that CD18 ligation is necessary to induce HLA-DR and CD86 increased expression on DCs.
We then examined the possibility that CD18 ligation delivers signals to PMNs, which promote their production of cytokines, some of which could eventually activate HLA-DR and CD86 up-regulation on DCs. However, supernatants collected after 18 h coculture of PMNs with DCs did not affect HLA-DR and CD86 expression by DCs cultured independently (data not shown), indicating that DC activation in the presence of PMNs is not a result of cytokines produced by the PMNs in response to CD18 ligation.
DCs cocultured with C. albicans-loaded PMNs induce T lymphocyte reactivity
We first evaluated whether DCs loaded with heat-killed C. albicans elicited responses of autologous T lymphocytes, the donors of which were known to be sensitized to the fungus. The readout was IL-2 and/or IFN- production in culture supernatants. As expected for a recall antigen, the DCs indeed efficiently stimulated the T lymphocytes in a dose-dependent manner (Fig. 7A
), and five yeasts per DC were selected as the concentration to be used in further experiments.
 View larger version (25K): [in a new window] |
Figure 7. DCs cocultured with C. albicans-loaded PMNs elicit T lymphocyte IL-2 and IFN- production. (A) Dose-response effect of heat-killed C. albicans antigens presented by DCs to T lymphocytes. DCs preincubated with different yeast concentrations were cultured with T lymphocytes, the IL-2 and IFN- production (data not shown) of which was evaluated after 48 h. Data represent means ± SEM of three experiments. (B and C) T lymphocyte responses. T lymphocytes were cultured with DCs separately (DC), DCs and heat-killed C. albicans (DC+Ca), DCs cocultured for 3 h with control PMNs (PMNDC) or Ca-PMNs (Ca-PMNDC) at a 10:1 PMN/DC ratio, or with control PMNs (PMN) or Ca-PMN in the absence of DCs. (B) IL-2 production was evaluated after 24 h, 48 h, and 72 h, and (C) IFN- production was evaluated after 48 h. (D) Comparisons of the effects of cultured and fractionated live and apoptotic PMNs (1:1 PMN/DC ratio). Data are means ± SEMfrom (B and C) five or (D) three experiments. (E) Effect of cell contact on the transfer of antigen from PMNs to DCs. In addition to the conditions described in B and C, control (TW-PMNDC) or Ca-PMNs (TW-Ca-PMNDC) were separated from DCs by a Transwell® membrane. Data represent means from two experiments, the observed values of which differed only by 20% from the mean. *, P < 0.05, using the paired Student’s t-test.
|
(data not shown). Direct presentation of C. albicans antigens by DCs cultured with C. albicans-exposed PMNs was ruled out, as after sorting of the DCs from the PMNs (see Materials and Methods), there remained less than one yeast/10 PMNs (and therefore, per DC), which was below the threshold for direct presentation by DCs (see Fig. 7A
). This indicates that T lymphocyte responses did not result from direct or indirect presentation of PMN-derived antigens but rather to C. albicans antigen transfer from the PMNs to the DCs. We then examined the relative contribution of live and apoptotic PMNs in this effect. Experiments in which sorted live or apoptotic Ca-PMNs were used showed that both could indeed transfer antigens to DCs to the same extent (Fig. 7D)
. Finally, when using the two-chamber system described above, we found that separation of DCs and Ca-PMNs completely abrogated the DC capacity to stimulate T lymphocytes (Fig. 7E) , indicating that antigen acquisition by DCs depended then on direct contact with PMNs.
Thus, PMNs are capable to efficiently transfer C. albicans antigens to DCs, enabling the latter cells to elicit T lymphocyte responses.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Although circulating PMNs undergo constitutive apoptosis [29 , 30 ], their survival is prolonged in tissues [31 ] where they are exposed to antiapoptotic signals delivered by microbial molecules and/or to endogenous signals from the microenvironment. Hence, it is reasonable to assume that DCs and PMNs could interact in sites where they at least colocalize transiently. To address this issue, we used an experimental system in which PMNs and monocyte-derived iDCs were cocultured in the presence of GM-CSF and IL-4, which were maintained in cultures for the sake of the DCs [19 ]. Consistent with the known antiapoptotic and activating signals delivered to PMNs by GM-CSF [21 , 32 , 33 ] or IL-4 [23 , 32 , 34 , 35 ], we found that this cytokine combination indeed delayed PMN apoptosis. Thus, our model can reproduce in vitro the context of potential interactions between iDCs and PMNs exposed to endogenous cytokines and/or chemokines.
Upon coculture with live PMNs, DCs exhibited increased HLA-DR, CD86, and CD40 surface expression and produced IL-12, indicating that signals from PMNs can activate iDCs to become competent APCs. The inability of apoptotic PMNs to activate DCs shown here is consistent with the reported capacity of such PMNs to elicit down-regulation of costimulatory molecules on DCs [36 ]. We found that CD40-increased expression appears to be mediated by soluble factors released by the PMNs, whereas HLA-DR and CD86 increase on DCs requires direct contact with PMNs. In this respect, our data differ from findings that supernatants of activated PMNs up-regulate CD86 and CD40 on, as well as IL-12 production by, DCs [37 ], a discrepancy, which may be accounted for by the fact that we used human leukocytes, whereas murine Toxoplasma gondii-infected murine cells were used in the latter report. Our results are in line with a recent report [38 ] about the interactions of human PMNs and DCs, although resting PMNs had no effect there, and only PMNs treated with proinflammatory LPS elicited DC activation. This model differs from ours in which in the absence of inflammatory factors, GM-CSF and IL-4 were present to maintain DC and PMN survival and function [20 21 22 23 ]. As apoptotic PMNs were not accounted for in the latter report [38 ], it cannot be ruled out that without LPS, most resting PMNs underwent apoptosis then, during coculture with DCs. In any case, both studies show that PMNs can induce DC activation. This may result from the cytokines and chemokines produced by activated PMNs [3 4 5 ] or from their strong secretory activity and release of exosomes [39 ] with DC stimulatory capacity, such as that which we reported for mast cell-derived exosomes [40 ].
Cell-surface, molecule-dependent interactions between PMNs and other myeloid cells have already been reported [24
, 41
, 42
]. CD28+ PMNs have been shown to activate Leishmania-infected macrophages to produce IFN- via the CD28-CD80/CD86 pathway [24
]. As a result of the lack of CD28 expression on PMNs before and during coculture with DCs, this pathway cannot be involved in the interactions reported here. In contrast, we show involvement of the CD18-ß2 chain of CD11/CD18 integrins. In our model, ligation of CD18 appears necessary to elicit HLA-DR and CD86 up-regulation on DCs, which indicates that the CD18 pathway plays a role in PMN/DC interactions. Cell contact-dependent interactions mediated by engagement of CD18-ß2 integrin on PMNs have already been shown to facilitate their interaction with Küpffer cells and to be crucial for bacterial clearance in the liver [41
, 42
]. It is interesting that van Gisbergen et al. [38
] have shown that PMNs and DCs interacted via the membrane-activated complex 1 (CD18/CD11b) CD11b subunit on PMNs and DC-specific intercellular cell adhesion molecule-grabbing nonintegrin (SIGN; CD209) on DCs. However, we could not inhibit this interaction with an anti-DC-SIGN mAb (data not shown) as they did, which could be a result of the fact that we used a different clone (1B10 [43
], gift from Ali Amara, Institut Pasteur, Paris, France, instead of AZN-D1) or that the molecules involved in the interaction between PMNs and DCs differ whether the PMNs are activated or not by a proinflammatory factor.
To examine whether interactions with PMNs influenced DC function, we investigated if PMNs could transfer antigens to DCs for eventual presentation to and stimulation of T lymphocytes. Hence, C. albicans was chosen as a model microorganism, because of its well-documented capacity to interact with PMNs and DCs [44 45 46 47 ] and as most adults are already sensitized against it, which makes normal donor selection easier. This fungus expresses different pathogen-associated molecules, including ß-glucans, mannans, mannoproteins, and phospholipomannans, which interact with phagocytic cells in an opsonin-independent manner via counter-receptors such as ß-glucan receptors (CD11b/CD18 [48 ], lactosylceramide [49 ], scavenger receptors [50 ], Dectin-1 [51 ]), mannose receptors [52 ], and Toll-like receptors 2 and 4 [53 ]. PMNs [54 ] and DCs [46 , 55 ] can capture and kill C. albicans. Here, using heat-killed C. albicans, we found that PMNs could deliver antigens to DCs via a cell contact-dependent mechanism, which allowed the DCs to present antigen to and elicit reactivity from T lymphocytes. This finding indicates that antigenic material from microorganisms, which are first captured by PMNs, can be further available for uptake and presentation by professional APCs such as DCs. In vivo, such transfer could occur in tissues from newly recruited PMNs to immature resident DCs, which will next migrate to the draining lymphoid organ. Alternatively, PMNs could transport microorganisms first captured in peripheral tissues to lymphoid organs, where they would encounter DCs, as recently shown in mice infected with Mycobacterium bovis BCG [56 ].
It has been reported that DCs can capture antigens from dead or dying cells for subsequent presentation to T lymphocytes [57 ]. Although this mechanism appears to be involved in our model, we found that DCs can also acquire antigens from live PMNs. The mechanisms of antigen transfer to DCs from live or from apoptotic/dead PMNs should be different. In the case of C. albicans, live PMNs can release soluble antigen following phagocytosis [58 ], which raises the possibility that DCs acquire degraded Candida antigens released by the PMNs. However, our data do not support this hypothesis, as transfer is dependent on cell contacts. We propose that live PMNs serve as donor cells for antigen uptake by DCs after having captured and concentrated the antigens by two mechanisms at least; they could concentrate killed C. albicans on their surface via their large spectrum of pathogen-recognition receptors for fungal products; and/or they could bind and concentrate microorganisms by a recently reported mechanism, which consists of the release of protein and chromatin granules forming extracellular traps [59 ]. As professional phagocytes, live or dead PMNs could be an important source of antigens for DC presentation and possibly cross-presentation to major histocompatibility complex class-I restricted CD8 lymphocytes, which remain to be shown [57 , 60 61 62 ]. Whether the findings reported here apply to ex vivo-isolated DC subsets will be investigated further. Indeed, it has been shown that human blood CD11c+ DC are able to phagocyte apoptotic and necrotic cells [63 ].
Altogether, this study demonstrates that PMNs can well-regulate DC functions and that they deliver activation signals and antigens to DCs, which will then efficiently stimulate appropriate T lymphocyte responses. This opens new insights into the cooperation between the innate and acquired arms of immunity. It is interesting that although apoptotic PMNs are able to transfer antigens, they do not deliver activation signals to DCs. The consequences of the interactions of PMNs with DCs on the immune response may thus vary, depending on the ratio of live and apoptotic PMNs in the environment. It is then possible that at the site of an injury, evolution with time of the relative proportion of live and apoptotic PMNs may regulate the level and type of DC activation, resulting eventually in the induction of different types of adaptive immune responses.
Received September 22, 2005; revised December 9, 2005; accepted January 10, 2006.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(MIG), IFN-inducible T cell chemoattractant (I-TAC), and IFN--inducible protein-10 (IP-10) chemokines by human neutrophils J. Immunol. 162,4928-4937 (MIP-3) (CCL20) and MIP-3ß (CCL19) by human peripheral blood neutrophils in response to microbial pathogens Infect. Immun. 71,524-526 production in LPS-stimulated neutrophils is enhanced by the chemoattractant formyl-methionyl-leucyl-phenylalanine and IFN- through independent mechanisms Eur. J. Immunol. 32,3515-3524[CrossRef][Medline] secretion and restriction of Leishmania growth J. Immunol. 169,920-928 RIII) expression J. Immunol. 153,1254-1263[Abstract]RIIIb (CD16) surface expression following delayed apoptosis in response to GM-CSF and sodium butyrate J. Leukoc. Biol. 65,875-882[Abstract]This article has been cited by other articles:
 |
|
 |
|
  M. R. Murawski, G. N. Bowen, A. M. Cerny, L. J. Anderson, L. M. Haynes, R. A. Tripp, E. A. Kurt-Jones, and R. W. Finberg Respiratory Syncytial Virus Activates Innate Immunity through Toll-Like Receptor 2 J. Virol., February 1, 2009; 83(3): 1492 - 1500. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kesteman, G. Vansanten, B. Pajak, S. M. Goyert, and M. Moser Injection of lipopolysaccharide induces the migration of splenic neutrophils to the T cell area of the white pulp: role of CD14 and CXC chemokines J. Leukoc. Biol., March 1, 2008; 83(3): 640 - 647. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Beauvillain, Y. Delneste, M. Scotet, A. Peres, H. Gascan, P. Guermonprez, V. Barnaba, and P. Jeannin Neutrophils efficiently cross-prime naive T cells in vivo Blood, October 15, 2007; 110(8): 2965 - 2973. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Sukhumavasi, C. E. Egan, and E. Y. Denkers Mouse Neutrophils Require JNK2 MAPK for Toxoplasma gondii-Induced IL-12p40 and CCL2/MCP-1 Release J. Immunol., September 15, 2007; 179(6): 3570 - 3577. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. C. Maffia, S. E. Zittermann, M. L. Scimone, N. Tateosian, N. Amiano, D. Guerrieri, V. Lutzky, D. Rosso, H. E. Romeo, V. E. Garcia, et al. Neutrophil Elastase Converts Human Immature Dendritic Cells into Transforming Growth Factor-{beta}1-Secreting Cells and Reduces Allostimulatory Ability Am. J. Pathol., September 1, 2007; 171(3): 928 - 937. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Mosca, N. Polentarutti, G. Mangano, C. Apicella, A. Doni, F. Mancini, M. De Bortoli, I. Coletta, L. Polenzani, G. Santoni, et al. Regulation of the microsomal prostaglandin E synthase-1 in polarized mononuclear phagocytes and its constitutive expression in neutrophils J. Leukoc. Biol., August 1, 2007; 82(2): 320 - 326. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Yu, M. J. Ruddy, G. C. Wong, C. Sfintescu, P. J. Baker, J. B. Smith, R. T. Evans, and S. L. Gaffen An essential role for IL-17 in preventing pathogen-initiated bone destruction: recruitment of neutrophils to inflamed bone requires IL-17 receptor-dependent signals Blood, May 1, 2007; 109(9): 3794 - 3802. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Mathias, B. J. Perrin, T.-X. Liu, J. Kanki, A. T. Look, and A. Huttenlocher Resolution of inflammation by retrograde chemotaxis of neutrophils in transgenic zebrafish J. Leukoc. Biol., December 1, 2006; 80(6): 1281 - 1288. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
