|
|
||||||||


* Instituto de Biología y Genética Molecular, Consejo Superior de Investigaciones Científicas;
Facultad de Medicina, Universidad de Valladolid, Valladolid, Spain; and
Institute of Infectious Disease and Molecular Medicine, Division of Immunology, Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa
| Abstract |
|---|
|
|
|---|
R induced AA release, short-term, low-grade PG biosynthesis, and IL-10 production, whereas zymosan, which contains ligands of both the mannose receptor and the human β-glucan receptor dectin-1, induced a wider set of responses including cyclooxygenase 2 induction and biosynthesis of leukotriene C4 and IL-12p70. The cytosolic phospholipase A2 inhibitor pyrrolidine 1 completely inhibited AA release stimulated via all receptors, whereas the spleen tyrosine kinase (Syk) inhibitors piceatannol and R406 fully blocked AA release in response to immune complexes, but only partially blocked the effect of zymosan. Furthermore, anti-dectin-1 mAb partially inhibited the response to zymosan, and this inhibition was enhanced by mAb against DC-specific ICAM-3-grabbing nonintegrin (SIGN). Immunoprecipitation of DC lysates showed coimmunoprecipitation of DC-SIGN and dectin-1, which was confirmed using Myc-dectin-1 and DC-SIGN constructs in HEK293 cells. These data reveal a robust metabolism of AA in human DC stimulated through both opsonic and nonopsonic receptors. The Fc
R route depends on the ITAM/Syk/cytosolic phospholipase A2 axis, whereas the response to zymosan involves the interaction with the C-type lectin receptors dectin-1 and DC-SIGN. These findings help explain the distinct functional properties of DC matured by immune complexes vs those matured by β-glucans. | Introduction |
|---|
|
|
|---|
PG biosynthesis involves several steps catalyzed by different enzymes, but the limiting step for the biosynthesis of eicosanoids is hydrolysis of AA from phospholipids by cytosolic phospholipase A2 (cPLA2). Cyclooxygenases (COX) 1 and 2 (COX-1 and -2) convert the AA released by cPLA2 to PG endoperoxide H2, which is the precursor of series 2 prostanoids such as PGD2 and PGE2. Unlike COX-1, COX-2 is an inducible enzyme involved in the sustained production of prostanoids by many cell types. Notably, COX-2 activity is necessary for strong Ab response following vaccination, especially when vaccines are poorly immunogenic or the target population is poorly responsive to immunization (6). In addition to the COX-2 route for AA metabolism, there are pathways dependent on constitutively expressed 5-lipoxygenase and COX-1, which are triggered shortly after cell activation. With regard to 5-lipoxygenase products, deficient extracellular export of leukotriene (LT) C4 is associated to a decreased migratory response of DC (7) and cysteinyl-LT increase IL-10 production by myeloid DC (8). Recent studies have disclosed lipoxins as a unique class of lipoxygenase interaction metabolites with a strong ability to suppress the production of IL-12 and the function of DC, a phenomenon termed DC paralysis (9).
In previous studies, we have observed a robust release of AA in human monocytes (10) and polymorphonuclear leukocytes (PMN) (11) by two pathogen-associated molecular signatures, namely, peptidoglycan (PGN) and mannose-based polymers, thus defining pathogen-associated molecular patterns as chief stimuli for AA metabolism. In the present study, we addressed the effect of a set of stimuli acting through receptors involved in recognition of microbial components by DC. We found that ligands of opsonic and nonopsonic receptors such as the human β-glucan receptor dectin-1 (12), the mannose receptor (MR), and Fc
Rs elicited a robust release of AA. Binding of zymosan and mannose-based stimuli (13, 14) was followed by the induction of COX-2 expression and the production of IL-12p70, whereas Fc
R cross-linking was associated with IL-10 generation. The Syk kinase inhibitors piceatannol and R406 completely inhibited AA release in response to Fc
R cross-linking, but they only partially blocked the response to zymosan. Anti-dectin-1 mAb inhibited the effect of zymosan and this inhibition was further enhanced by mAb against DC-specific ICAM-3-grabbing nonintegrin (SIGN) (15). Immunoprecipitation of DC lysates with anti-DC-SIGN mAb showed coprecipitation of dectin-1. These data show for the first time the occurrence of a robust metabolism of AA in human DC in response to immune complexes (IC), β-glucan particles, and mannose-based molecular patterns.
| Materials and Methods |
|---|
|
|
|---|
Zymosan, the soluble β-glucan laminarin, mannan from Saccharomyces cerevisiae, OptiPrep, porcin mucin-3, and piceatannol were purchased from Sigma-Aldrich. Anti-human CD206/MR, anti-human CD209/DC-SIGN, and anti-CD45 mAb were obtained from BD Pharmingen. Anti-human dectin-1 (GE2 mAb) was generated as previously described (16). Goat anti-dectin-1 Ab was purchased from R&D Systems. The cPLA2 inhibitor pyrrolidine 1 was purchased from Calbiochem. The Syk kinase inhibitor R406 (17) was a generous gift from Dr. E. Masuda (RIGEL Inc., San Diego, CA). Preparation of IgG/OVA equivalence IC and opsonization of zymosan particles were conducted as described elsewhere (18). IL-12p70, IL-10, and PGE2 were assayed with Biotrack ELISA systems from Amersham Biosciences according to the manufacturers instructions. The detection limit of these assays is 3 pg/ml for IL-12p70 and IL-10 and 2.5 pg/ml for PGE2. ELISA for LTC4 and PGD2 were obtained from Cayman Chemical. A pEF-hemagglutinin (HA) expression vector encoding porcine spleen tyrosine kinase (Syk) was provided by Dr. T. Mustelin (Burnhan Institute, La Jolla, CA). The cDNA of human β-glucan receptor was inserted into the EcoRI/XbaI sites of a pEF4-Myc vector. A pcDNA3 vector encoding human DC-SIGN cDNA and anti-DC-SIGN (MR1 mAb) for immunoprecipitation assays were provided by Dr. A. Corbí (Centro de Investigaciones Biológicas, Madrid, Spain). Endotoxin levels in the reagents were below 1 ng/ml as determined by the Limulus amebocyte lysate assay (Cambrex). Moreover, addition of 200 µg/ml polymyxin B to these reagents did not modify the effect of the different stimuli, excluding a possible involvement of LPS in the responses studied.
Harvesting of monocyte-derived DC
Mononuclear cells were collected from buffy coats of healthy donors by centrifugation on Ficoll-Hypaque cushions. The mononuclear cell ring was recovered in 3 ml of OptiPrep and then layered below a solution of 7 ml of Ficoll (density 1.072 g/ml). A discontinuous gradient was formed by adding a solution of OptiPrep (density, 1.068 g/ml) containing 0.5% BSA and 1 mM EDTA, followed by a HEPES-buffered saline solution. The mixture was centrifuged for 25 min at 725 x g at 19°C and the lymphocyte-depleted cell solution was collected and treated again to enhance the purity of the cell preparation. Cells were maintained for 2 h at 37°C to allow the adherence of monocytes. Nonadhered cells were removed and the remaining cells were cultured in the presence of GM-CSF (800 U/ml) and IL-4 (500 U/ml) for 5 days. Maturation was achieved by incubation in the presence of recombinant human TNF-
(100 U/ml) and assessed by flow cytometry of CD40, CD80, CD83, and CD86. Labeling of monocyte-derived DC with [3H]AA was conducted as previously described (18).
Phagocytosis of Alexa Fluor 488-labeled zymosan particles by DC
DC (3 x 105 in 0.5 ml) were incubated with zymosan particles conjugated with Alexa Fluor 488 at the concentration of five particles per cell at 4°C to address particle binding or at 37°C to address both binding and phagocytosis as previously reported (18, 19). For inhibition studies with laminarin, mannan, and mAb against dectin-1 and DC-SIGN, DC were incubated for 30 min before the addition of zymosan particles. One hour after addition of zymosan particles, cells were taken from the plates and transferred into FACS tubes (BD Pharmingen) for analysis by flow cytometry in a Beckman Coulter Epics XL cytofluorometer.
Immunoblots and immunoprecipitations
Proteins were separated by electrophoresis in SDS-PAGE and transferred to nitrocellulose membranes. The membranes were used for immunodetection of COX-2 with a goat Ab (SC-1745) from Santa Cruz Biotechnology. The phosphorylation of cPLA2 was assayed using anti-cPLA2 phospho-specific Ab (Ser505; catalog no. 2831 from Cell Signaling Technology). For the assay of phosphotyrosine, antiphosphotyrosine mAb (4G10) from Upstate Biotechnology was used. Syk phosphorylation was addressed using phospho-specific anti-human Syk Ab reactive to the activation loop site pY525/526 (ref. 2711; Cell Signaling Technology). Quantitation of the blots was conducted using Bio-Rad Quantity One gel imaging software. Immunoprecipitations were conducted as previously described (20). Briefly, cells were lysed in a medium containing 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, 1 mM Na3VO4, 10 µg/ml aprotinin and leupeptin, 100 µg/ml soybean trypsin inhibitor, and 1 mM PMSF and clarified by centrifugation at 15,000 rpm for 20 min. The clarified lysates were preabsorbed on protein G-Sepharose and then incubated with precipitating mAb for 4 h, followed by overnight incubation with protein G-Sepharose beads. IC were extensively washed, suspended in Laemmli sample buffer, and subjected to SDS-PAGE. Blots were stained to assess the input protein and the coimmunoprecipitation of C-type lectins. Experiments to address the effect of the ectopic expression of the C-type lectin receptors and the potential interactions with Syk were conducted in HEK293 cells. Transfections were conducted by the calcium phosphate method with 5 µg of plasmid DNA completed up to 15 µg with pEF4 empty vector in plates containing 3 x 106 cells.
Real-time RT-PCR of COX-2
Purified RNA was depleted of genomic DNA by treatment with DNase (Turbo-DNA free; Ambion) and used for reverse transcriptase reactions. The resulting cDNA was amplified in a PTC-200 apparatus equipped with a Chromo4 detector (Bio-Rad) using SYBR Green I mix containing HotStart polymerase (ABgene). The sets of primers for PCR were selected in exons 1 and 2 and were: forward 5'-CAATTGTCATACGACTTGCA-3' and reverse 5'- GTGGGAACAGCAAGGATTTG-3'. Cycling conditions were as follows: 94°C for 5 min, followed by 40 cycles of 94°C for 30 s, 60°C for 30 s, and extension at 72°C for 30 s. GAPDH was used as a housekeeping gene to assess the relative abundance of the different mRNA, using the comparative cycle threshold method.
Confocal microscopy
DC were seeded in glass coverslips for 1 h and then stimulated with Alexa Fluor 488-labeled zymosan particles at the concentration of five particles per cell. Cells were fixed with 10% formaldehyde in PBS and stained with different mAb and goat anti-mouse IgG Ab labeled with Alexa Fluor 594. The coverslips were observed by confocal microscopy using a Bio-Rad Laser Scanning System Radiance 2100 with LaserSharp2000 software coupled to a Nikon inverted microscope, with a x60 oil immersion objective. Green fluorescence (zymosan) was monitored at 488 nm of argon excitation using HQ500LP and HQ540SP blocking filters (Chroma Technology). Red fluorescence was monitored at 590 nm using a HQ570LP blocking filter.
Statistics
For statistical analysis of data, paired and unpaired Students t tests were performed (PRISM version 4.0; GraphPad) as appropriate. Values of p < 0.05 were considered significant.
| Results |
|---|
|
|
|---|
Since the functional properties of DC change along the process of maturation, experiments were conducted in both immature and mature DC. TNF-
-induced maturation of DC was characterized by an increased expression of CD83 and a parallel decrease of the surface display of CD14. The expression of DC-SIGN, MR, and dectin-1 increased above the levels detected in monocytes upon the addition of GM-CSF and IL-4, but decreased after treatment with TNF-
. Of the different types of Fc
R, Fc
RII showed the highest level of expression, whereas the expression of Fc
RI and Fc
RIII was low (Fig. 1A).
|
mature DC. On this basis and unless otherwise indicated, experiments were conducted with 5-day immature DC. [3H]AA release induced by zymosan and IC steadily increased up to
90 min (Fig. 1B) and was observed with concentrations as low as 0.1 mg/ml zymosan and 10 µg/ml IC (Fig. 1C). Mucin-3, a glycoprotein from the gastrointestinal tract, which is an endogenous ligand of the MR, also showed a robust [3H]AA-releasing activity (Fig. 1C), whereas mannan induced [3H]AA release to a lower extent (Fig. 1D). Given that complement coating of both zymosan and IC modifies the ability of these stimuli to release AA in human monocytes (18, 20), zymosan and IC were treated with normal human serum to allow the covalent coupling of C3bi moieties. Unlike the results observed in monocytes, neither C3bi coating nor opsonization with rabbit IgG modified the ability of these stimuli to release [3H]AA (Fig. 1D). These data indicate that the ability of zymosan to induce AA release in DC is independent of its ability to interact with complement factors and IgG.
Initial attempts to delineate the pathway involved in AA release were conducted with the cPLA2 inhibitor pyrrolidine 1. As shown in Fig. 2A, pyrrolidine 1 fully inhibited [3H]AA release, which suggests a complete dependence of [3H]AA release from cPLA2-catalyzed reactions (Fig. 2A). Zymosan-induced [3H]AA release was inhibited by laminarin, mannan, and anti-dectin-1 and anti-DC-SIGN mAb, which was most evident when the inhibitors were used in combination (Fig. 2, B and C). However, when zymosan was used at nonsaturating levels, inhibition with laminarin was almost complete (Fig. 2B). These data indicate receptor cooperation in zymosan-induced AA release. To obtain further insight into the type of receptors involved in the recognition of zymosan by DC, the binding of Alexa Fluor 488 zymosan was studied in the presence of different inhibitors. These experiments were conducted at both 4 and 37°C to differentiate binding from the combination of binding and phagocytosis, which would occur at 37°C. Different agonist:antagonist ratios were used and the experiments were performed under identical conditions to those used for AA release. As shown in Fig. 2D, mannan, laminarin, anti-DC-SIGN, and anti-dectin-1 mAb could block zymosan binding. Combination of these inhibitors enhanced the blockade of zymosan binding at both 4 and 37°C, although the amount of particles associated with the cells was higher at 37°C than at 4°C (325 vs 21 mean fluorescence intensity units, respectively), consistent with enhanced uptake at 37°C. Taken together, this data show the existence of a cPLA2-dependent route for AA release in DC, which can be triggered by engaging Fc
R and by binding to dectin-1 and DC-SIGN in the case of zymosan particles.
|
, and IC did not stimulate IL-12p70 production, whereas zymosan and LPS were robust stimuli (Fig. 3A). In contrast, IL-10 was induced by all of the stimuli (Fig. 3B). Treatment with pyrrolidine 1, piceatannol, and the COX-2 inhibitor SC-236 inhibited the production of IL-12p70 in response to zymosan, whereas only SC-236 partially inhibited the effect of LPS. Pyrrolidine 1 also inhibited the production of IL-10 elicited by zymosan, but was ineffective in the case of IC (Fig. 3B). These data would suggest a partial dependence of cytokine production on the biosynthesis of eicosanoids by DC.
|
To delineate the routes of AA metabolism downstream of cPLA2 that can be operative in DC, the induction of COX-2 and the production of PGE2, PGD2, and LTC4 were addressed. Zymosan and mucin-3 induced COX-2 protein expression in immature DC (Fig. 4A), whereas both IC and TNF-
at the concentration used to induce DC maturation failed to do so (Fig. 4B, right panels). COX-2 protein induction was less prominent in mature DC than in immature DC (Fig. 4A) and showed a dose dependence similar to that observed for [3H]AA release (Fig. 4, B and C). Although COX-2 mRNA was detected as soon as 30 min after the addition of zymosan (Fig. 4D), COX-2 protein was only detectable after 4 h and persisted up to 48 h (Fig. 4E).
|
|
Syk activity is involved in AA release
The protein tyrosine kinase Syk plays a central role in cell signaling through both Fc
R and dectin-1 in murine DC (21), and is a key element in Fc receptor-mediated Ag presentation and DC maturation (22). The involvement of Syk in [3H]AA release and COX-2 induction was then assessed by examining tyrosine phosphorylation of this kinase (a measure of Syk activation) and the effect of Syk inhibitors. As shown in Fig. 6A, both IC and zymosan induced activation of Syk in DC. Treatment of DC with piceatannol significantly inhibited IC-induced, but only marginally affected zymosan-induced cPLA2 phosphorylation (Fig. 6B). In agreement with this result, piceatannol inhibited the release of [3H]AA by 96 and 54% in response to IC and zymosan, respectively. R406, a very specific Syk inhibitor, also inhibited completely the response to IC and reduced zymosan-induced [3H]AA release by 30% (Fig. 2A). Zymosan-induced Syk phosphorylation was also inhibited with the addition of laminarin, but not by anti-DC-SIGN mAb (Fig. 6C). We also examined the effect of Syk inhibition on the induction of COX-2 and observed that piceatannol could inhibit the induction of this protein partially (Fig. 6D). Piceatannol also blocked the induction of COX-2 elicited by PGN, which is in agreement with the recent observation that piceatannol could inhibit LTC4 production in response to PGN (23); however, this inhibitor did not affect COX-2 induction by LPS. HEK293 cells transfected with expression vectors encoding dectin-1 and Syk showed a clear association of dectin-1 and Syk following zymosan addition (Fig. 6E), further supporting the involvement of Syk in dectin-1 signaling. Taken collectively, these results are consistent with the notion that Syk activity is completely necessary for IC-induced AA release, but it is only partially involved in the signaling mechanism whereby zymosan elicits AA release and COX-2 induction in DC.
|
B and we examined the involvement of this pathway in the induction of COX-2 in DC. By analyzing I
B
degradation (Fig. 7A), we observed a robust activation of the NF-
B route by zymosan, which occurred between 30 and 60 min and was followed by resynthesis of I
B
by 2 h (Fig. 7A). Moreover, blockade of this pathway with N-acetyl-leucinyl-leucinyl-norleucinal (ALLN) inhibited the induction of COX-2 protein in response to all stimuli, suggesting an involvement of NF-
B-driven transcriptional regulation (Fig. 7B).
|
The ability of combinations of laminarin/anti-dectin-1 and anti-DC-SIGN mAb to inhibit [3H]AA suggested a cooperation between DC-SIGN and dectin-1 for the generation of this response. Indeed, we could demonstrate that dectin-1 coimmunoprecipitated with DC-SIGN, particularly after the stimulation of DC with zymosan (Fig. 8A). A control anti-GST mAb did not immunoprecipitate DC-SIGN or dectin-1, demonstrating the specificity of this assay. Additional experiments in HEK293 cells transfected with vectors encoding DC-SIGN and Myc-dectin-1 showed a robust coimmunoprecipitation of both C-lectin receptors when immunoprecipitation was conducted with either anti-DC-SIGN mAb or anti-Myc mAb (Fig. 8B). These results are consistent with a system for zymosan recognition in DC involving the interaction of dectin-1 and DC-SIGN.
|
|
| Discussion |
|---|
|
|
|---|
-matured DC showed a diminished induction of COX-2 upon zymosan challenge. Because PGE2 is important for DC migration from peripheral tissues toward secondary lymph organs, these results suggest an autocrine mechanism of PG production which helps us to understand the functional responses of DC. These are novel findings, since most studies on the role of AA in DC biology have focused on the responses to eicosanoids and the process of DC migration mediated by E prostanoid (25, 26), cysteinyl-LT (7, 8), LTB4 (27), and lipoxin receptors (9) rather than on eicosanoid production. It has, however, been proposed that the differentiation of monocytes into DC is associated with a decreased expression of cPLA2, due to the action of IL-4, an essential component of the cytokine mixture used to promote DC differentiation (4, 5). A possible reason for this discrepancy could be the use of different experimental approaches. Zelle-Rieser et al. (5) cultured monocytes for 5 days in the presence of either GM-CSF or GM-CSF plus IL-4 and the differences in PGE2 production in response to LPS were attributed to the effect of IL-4 on cPLA2. However, although LPS is a well-established inducer of COX-2 expression and delayed prostanoid production, it has only a priming effect (28, 29) or no effect at all on early AA release and LT biosynthesis in PMN, macrophages (30, 31), and DC. Our data show stimulus-specific patterns for AA metabolism in DC similar to those recently reported in a murine macrophage cell line, where there were depicted a pattern associated with TLR agonist pathways and another one linked to stimuli able to generate their response via a sustained intracellular Ca2+ elevation (32). Our results can be summarized in three different programs of DC activation: 1) a response triggered by Fc
R cross-linking, which leads to Syk-dependent cPLA2 activation, early PG biosynthesis, and IL-10 production; 2) a program involving cPLA2 activation, early biosynthesis of LTC4 and PG, delayed PG production linked to COX-2 induction, and IL-12 production, which is triggered by ligands of C-type lectin receptors; and 3) a route activated by LPS which involves IL-12 and COX-2 induction in the absence of acute AA release. To explain the distinct pattern of AA metabolism induced by LPS, it should be taken into account that cooperation of secreted phospholipases A2 and cPLA2 is involved in the delayed phase of prostanoid production in different cell types (33, 34, 35, 36, 37, 38). This depends on COX-2 induction and occurs without accompanying LT synthesis. The existence of a similar mechanism in DC might explain our findings.
The time course of AA release observed in the present study is similar to that reported in monocytes stimulated with IC (18), PMN treated with mannan and PGN (11), and RAW264.7 macrophages activated with TLR agonists. In contrast, stimulation of RAW264.7 cells via purinergic receptors shows optimal AA release at
10 min (32) and activation of mast cells via Fc
RI releases AA in 2–3 min (39). Whether the different time courses observed upon stimulation of Fc
RI and Fc
R are due to differences in the structure of the ITAMs or to other elements involved in the signaling cascades is not known.
The association between COX-2 induction and IL-12 production may be linked to posttranscriptional mechanisms or to a mechanism of transcriptional regulation common to both proteins. The first hypothesis is supported by the synergistic effect of PGE2 on TNF-
-induced production of IL-12 (40) and by the diminution of IL-12p70 protein we observed upon COX-2 inhibition (Fig. 3A). In contrast, the regulation of COX-2 and IL-12 by NF-
B transcription factors, which may show stimulus-specific activation patterns, would be consistent with the second hypothesis. For example, TNF-
activates RelA but not c-Rel and is a weak inductor of IL-12p35 mRNA (41) and COX-2. In contrast, LPS activates both RelA and c-Rel (42) and is a strong inducer of both IL-12 p35 and COX-2 expression. In the absence of chromatin remodeling, tightly packaged nucleosomes are also an obstacle for accessibility of transcription factors to regulatory sequences in COX-2 (43), IL-12p35 (44), IL-12p40 (45), and IL-10 (46), which may explain transcriptional repression in the presence of active transcription factors. Taken together, these data indicate that a complex combination of cell responses involving activation of transcription factors and covalent modification of histones could explain the distinct transcriptional responses observed in DC.
The inhibition of IL-10 production by pyrrolidine 1 suggests a connection between IL-10 production and AA metabolism. This might be explained by two different mechanisms: 1) an indirect one, where IL-10 is up-regulated by IL-12 as a feedback mechanism to inhibit the proinflammatory response (47) and/or 2) a direct induction of IL-10 by PGE2 (48). These findings are relevant to understand DC function, because DC matured in the presence of PGE2, although phenotypically identical in many aspects to DC matured with other stimuli, are more sensitive to chemokine-mediated migration and are more efficient in eliciting T cell responses (25). In addition, DC matured in medium supplemented with PGE2 generate optimal yields of DC producing IL-12p70 (49). Autocrine production of PGE2 is also a central factor for DC to switch from a strongly adhesive to a highly migratory phenotype through a mechanism involving podosome disassembly and loss of
5β1 integrin-mediated adhesion to the substrate (50). Taken together, these data suggest that part of the specific responses elicited via innate immune system receptors on DC might be mediated by endogenous PGE2.
Our data underscore the ability of zymosan to induce AA release from DC, as compared with monocytes and PMN, where C3bi coating is required for this response (18, 51). This could be explained by the high levels of expression of dectin-1 on DC; however, the identification of the receptor(s) involved in zymosan recognition by these cells has not yet been resolved. In fact, dectin-1-deficient mice mount a normal cytokine response to Saccharomyces cerevisiae zymosan (52), although the response to Sparassis crispa glucan is abolished (53). Another explanation for our findings could be cooperation of dectin-1 with other receptors and/or regulatory molecules. Indeed, dectin-1 can interact with other proteins such as the leukocyte tetraspanin CD37, which has been shown to inhibit dectin-1-mediated IL-6 production (54). In addition, dectin-1, TLR2, and the MR have been shown cooperate in the recognition of Candida albicans mannans and glucans (55, 56). Given that DC show high levels of DC-SIGN expression, cooperation of dectin-1 and DC-SIGN in triggering AA release seems likely. This hypothesis is supported by the inhibition of [3H]AA release by combinations of anti-DC-SIGN and anti-dectin-1 mAb and by the coimmunoprecipitation of dectin-1 and DC-SIGN in DC and in cells overexpressing these receptors. These findings suggest a system for zymosan recognition in DC involving the synergistic interaction of several receptors.
Unlike monocytes (18), coupling of C3bi to IgG does not result in loss of AA-releasing activity in DC. The C3bi moiety incorporated into IC binds to CR3 (Mac-1, CD11b/CD18,
Mβ2 integrin) and drives the immune load from Fc
R to CR3, thus promoting phagocytosis by monocytes in the absence of inflammatory response. Notably, productive binding of C3bi to leukocyte-expressed CR3 needs an interaction with Fc
RIII to form a CR3/Fc
RIII complex, the formation of which is impeded in DC (57). DC show a high expression of CD11c, i.e., the
-chain of CR4, another integrin with C3bi-binding properties. In addition, DC β2 integrins are in an inactive state which hampers binding to cognate ligands (58). In the light of these data, plausible explanations for the distinct responses of monocytes and DC to C3bi-coupled IC could be an impaired binding to CR3, a preferential binding to CR4 or a constitutively inactive state of DC integrins.
Stimuli leading to AA release also induce Syk phosphorylation, although their ability to phosphorylate Syk does not correlate with their ability to induce the release of AA. Inhibition of Syk blocks AA release in response to IC, but only partially in response to zymosan. Since blockade of DC-SIGN shows an additive effect with laminarin and since DC-SIGN cannot activate Syk, signals triggered through DC-SIGN binding might account for the Syk-independent component of this response (59). This concept is supported by a recent report demonstrating that DC-SIGN activates the phospholipase C
route by a mechanism involving Src family kinases, but not Syk (60). Syk activation itself is not sufficient to induce COX-2, as judged from the absence of COX-2 induction by IC. Because monocyte-derived DC lose the expression of Fc
RI and Fc
RIII (42, 61, 62), it seems likely that the isolated cross-linking of Fc
RIIA and triggering of Syk is sufficient for the activation of the cPLA2, whereas COX-2 induction requires additional signals to initiate
B-driven transcriptional activation, as it has been reported in other systems (63).
Recent studies on dectin-1-mediated signaling suggest the occurrence of two different pathways: a dectin-1/Syk pathway, which depends on CARD9, and a dectin-1/TLR2/MyD88 pathway that might be at least partially independent from Syk and CARD9 (21, 64). The first pathway is involved in the production of IL-2, IL-23, and IL-10, whereas the MyD88-dependent module is involved in the production of IL-6, IL-12, and TNF-
. Our data allow the integration of AA metabolism within this dual signaling scheme (Fig. 10): AA release showing a partial dependence on Syk activity and COX-2 induction showing a full dependence on NF-
B activation triggered by TLR2 engagement (65). A route of dectin-1 signaling involving TLR2-independent NFAT activation has also been recently reported. Although this mechanism might explain a portion of COX-2 induction by zymosan in murine macrophages, it is significantly less potent than the canonical TLR2 route (66). Taken collectively, our data reveal an active metabolism of AA in human DC, which is triggered by physiological stimuli of DC. Stimulation of Fc
Rs is associated with the activation of the cPLA2 enzyme and early production of COX-1 metabolites, whereas fungal stimuli are also inducers of COX-2 expression and endow DC with the ability to produce high-level, long-lasting PG biosynthesis.
|
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported by grants from Plan Nacional de Salud y Farmacia (Grant SAF2004-01232), Fundación Ramón Areces, Junta de Castilla y León (Grant CSI05C05), and Red Temática de Investigación Cardiovascular. I.V. was the recipient of a grant from Banco de Santander-Central-Hispano. N.F. is under contract within the Ramón y Cajal Program (Ministerio de Educación y Ciencia of Spain and Fondo Social Europeo). ![]()
2 I.V. and N.F. contributed equally to this work. ![]()
3 Address correspondence and reprint requests to Dr. M. Sánchez Crespo, Instituto de Biología y Genética Molecular, C/ Sanz y Forés s/n, 47003-Valladolid, Spain. E-mail address: mscres{at}ibgm.uva.es ![]()
4 Abbreviations used in this paper: DC, dendritic cell; AA, arachidonic acid; ALLN, N-acetyl-leucinyl-leucinyl-norleucinal; COX, cyclooxygenase; cPLA2, cytosolic phospholipase A2; DC-SIGN, DC-specific ICAM-3-grabbing nonintegrin; IC, immune complex; LT, leukotriene; MR, mannose receptor; PGN, peptidoglycan; PMN, polymorphonuclear leukocytes; HA, hemagglutinin. ![]()
Received for publication May 8, 2007. Accepted for publication February 14, 2008.
| References |
|---|
|
|
|---|
R and the complement receptor 3 pathways. J. Biol. Chem. 278: 52179-52187.
activation. Blood 110: 561-567.
-receptor. J. Immunol. 153: 796-803. [Abstract]
cooperate to activate human dendritic cells: synergistic activation of interleukin 12 production. J. Exp. Med. 186: 1603-1608.
-primed IL-12 production by IFN regulatory factor 1. J. Exp. Med. 198: 1265-1276.
and interleukin-10. J. Exp. Med. 183: 2559-2569.
is necessary for platelet-activating factor biosynthesis, efficient neutrophil-mediated bacterial killing, and the innate immune response to pulmonary infection: cPLA2-
does not regulate neutrophil NADPH oxidase activity. J. Biol. Chem. 280: 7519-7729.
receptor-mediated phagocytosis by human blood dendritic cells. J. Immunol. 157: 541-548. [Abstract]
R receptors activate MAP kinase and up-regulate the cyclooxygenase pathway without increasing arachidonic acid release in monocytic cells. Eur. J. Immunol. 32: 383-392. [Medline]
B. J. Immunol. 171: 3047-3055. This article has been cited by other articles:
![]() |
Y. Alvarez, C. Municio, S. Alonso, M. Sanchez Crespo, and N. Fernandez The Induction of IL-10 by Zymosan in Dendritic Cells Depends on CREB Activation by the Coactivators CREB-Binding Protein and TORC2 and Autocrine PGE2 J. Immunol., July 15, 2009; 183(2): 1471 - 1479. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Alvarez, C. Municio, S. Alonso, J. A. S. Roman, M. S. Crespo, and N. Fernandez Cyclooxygenase-2 Induced by Zymosan in Human Monocyte-Derived Dendritic Cells Shows High Stability, and Its Expression Is Enhanced by Atorvastatin J. Pharmacol. Exp. Ther., June 1, 2009; 329(3): 987 - 994. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. A. Barrett, A. Maekawa, O. M. Rahman, K. F. Austen, and Y. Kanaoka Dectin-2 Recognition of House Dust Mite Triggers Cysteinyl Leukotriene Generation by Dendritic Cells J. Immunol., January 15, 2009; 182(2): 1119 - 1128. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |