|
|
||||||||



* Department of Medicine and
Department of Pediatrics, Harvard Medical School and Division of Rheumatology, Immunology and Allergy, Brigham and Womens Hospital; and
Department of Neurology, MassGeneral Institute for Neurodegenerative Disease, Massachusetts General Hospital, Boston, MA 02115
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Among the >400 GPCRs encoded by mammalian genomes, the closest homologs of the cys-LT receptors are the purinergic (P2Y) receptors for extracellular nucleotides. Eight such GPCRs (P2Y1, P2Y2, P2Y4, P2Y6, P2Y11, P2Y12, P2Y13, and P2Y14) have confirmed nucleotide specificity (reviewed in Ref. 11). Some P2Y receptors are expressed in the CNS and others are expressed by vascular structures, epithelial cells, and hematopoietic cells. Cellular damage due to vascular injury and shear stress causes the release of large amounts of nucleotides, which can also be released in a regulated fashion by activation of secretory cell types such as MCs (12) or platelets. Thus, during tissue injury or inflammatory responses, extracellular concentrations of nucleotides can reach micromolar concentrations, with even higher local concentrations at cell-cell junctions (11). In these contexts, P2Y receptors can function as "danger sensors," activating pain fibers (13), and regulating platelet aggregation (14, 15). P2Y receptors can also induce migration and phagocytic function of microglia (16, 17) and cause the migration of monocytes (18) and activation of macrophages (19), MCs (20), and myeloid dendritic cells (21). Although most P2Y receptors preferentially recognize adenine nucleotides (ATP or ADP), some recognize uracil nucleotides. The human P2Y4 receptor binds UTP and the P2Y14 receptor recognizes the uridine diphosphate (UDP) metabolite UDP-glucose (21). A recently deorphanized GPCR, GPR17, is expressed in the CNS and confers calcium signals in response to uracil-containing nucleotides and to LTD4 when expressed by transfection into an astrocyte cell line (22). The P2Y6 receptor is the only known UDP-specific GPCR and is strongly expressed by vascular smooth muscle (23) and the spleen (24). Many cells of myelomonocytic origin (osteoclasts (25), microglia (17), macrophages (19)) express P2Y6 receptors. Lung epithelial cells constitutively express P2Y6 receptors (26), while both mouse and human intestinal epithelium inducibly express P2Y6 receptors with inflammation (27). The expression of P2Y6 receptors is required for IL-8 production by a monocytic cell line stimulated with bacterial LPS (19) and by human intestinal epithelial cells stimulated with neutrophil-derived antimicrobial peptides (27). Thus, among the P2Y receptor class, P2Y6 receptors may be specialized to amplify the responses of the innate immune system to inflammatory stimuli that induce release of endogenous uracil nucleotides. The distributions of P2Y6 receptors and CysLT1R overlap, and many cell types (e.g., smooth muscle, leukocytes, microglia (17, 28) express both receptors.
MCs are tissue-resident effector cells that potently induce innate immune responses and are central to the pathophysiology of both allergic and nonallergic inflammatory diseases. Primary human MCs derived in vitro from cord blood (hMCs) express both CysLT1R and CysLT2R (8, 29) and respond to exogenous cys-LTs by proliferating (30) and generating cytokines (31), including large quantities of the chemokine MIP-1β by a signaling mechanism augmented by cell priming by IL-4. LTD4-mediated signaling in MCs requires CysLT1R-mediated phosphorylation of ERK, reflecting transactivation of the Kit receptor tyrosine kinase (30). Previously, we had demonstrated that UDP, like LTD4, caused calcium flux, ERK phosphorylation, and cytokine generation by hMCs, each of which was attenuated by pretreatment of the cells with MK571, an antagonist that blocks the CysLT1R (5) and several P2Y receptors (32), but not CysLT2R (6). We hypothesized that the parallels between cys-LT- and UDP-induced signaling events in MCs reflect cross-talk between the CysLT1R and one or more UDP-responsive receptors. We now report that CysLT1R and the P2Y6 receptor require one anothers presence on MCs for maximal ligand-initiated cytokine generation, ERK activation, and cytoprotection. Pharmacologic blockade of P2Y6 receptors in LAD2 cells, a well-differentiated human MC sarcoma line (33), diminishes their ERK activation in response to LTD4. Likewise, UDP-dependent ERK activation of mouse bone marrow-derived MCs (mBMMCs) requires the presence of CysLT1R and endogenous cys-LT production. Under conditions of starvation from cytokines, both UDP and LTD4 sustain the survival of MCs, a function requiring CysLT1R, extracellular nucleotides, LTC4 synthase (LTC4S), and the intact functions of P2Y6 receptors. LTD4-mediated MIP-1β production by LAD2 cells is strikingly attenuated by knockdown or blockade of P2Y6 receptors, by depletion of extracellular nucleotides, and by inhibition of LTC4 synthesis. Collectively, these observations suggest that P2Y6 receptors are components of CysLT1R-dependent inflammatory signaling in MCs and that the two receptors cross-talk through their respective endogenous ligands to enhance cell activation and survival.
| Materials and Methods |
|---|
|
|
|---|
LTD4 and MK571 were purchased from Cayman Chemical. Polyclonal anti-peptide Abs were raised against conserved sequences of human and mouse CysLT1R (RB34, against extracellular domain 3), CysLT2R (RB19, against extracellular domain 2), and the P2Y6 receptor (RB165, against the second intracellular loop of the human P2Y6 receptor CQRYLGICHPLAPWHKRGG by Orbigen). An additional Ab against the C terminus of the human CysLT1R (Cayman Chemical) was used for the fluorescence lifetime imaging microscopy (FLIM) assays. Monoclonal anti-β1 and polyclonal anti-
4 integrin Abs were purchased from BD Pharmingen. Alexa Fluor 488-conjugated goat anti-rabbit IgG and Cy3-conjugated donkey anti-mouse IgG were purchased from Invitrogen. Alexa Fluor 488 and Cy3 (Amersham Biosciences) were used for direct labeling according to the manufacturers protocols. Recombinant human and mouse cytokines were all purchased from R&D Systems. The fura 2-AM was purchased from Molecular Probes. The P2Y receptor antagonist suramin and the P2Y6 receptor-selective antagonist MRS2578 (34) were purchased from Sigma-Aldrich.
Derivation and culture of MCs
The use of cord blood for this study was covered under a protocol for the use of discarded human materials approved by the Partners Healthcare Human Research Committee. hMCs were derived from cord blood mononuclear cells as previously described (35). mBMMCs were derived from bone marrow (29) and studied at 4–6 wk when virtually all stained with toluidine blue. LAD2 cells (provided by Dr. A. Kirshenbaum, National Institutes of Health, National Institute of Allergy and Infectious Diseases, Bethesda, MD) were maintained in SpemPro34 medium containing stem cell factor (SCF; 100 ng/ml) (33). LAD2 cells were used for the experiments involving calcium flux and cytokine generation due to their robust responses to cys-LTs in these assays (36), and hMCs were used as a comparison group for studies of P2Y6 receptor expression. mBMMCs from gene-targeted mice and wild-type controls were used for the studies of ERK activation and cytoprotection to determine the requirements for CysLT1R, CysLT2R, and LTC4S in this process.
Assessment of apoptosis
Apoptosis assays were performed by the cytofluorographic detection of annexin V binding according to the suppliers instructions (BD Biosciences). The cells were maintained for 48–72 h in the presence or absence of SCF (10 ng/ml) or various concentrations of LTD4 or UDP.
Cell activation
LAD2 cells were maintained for 72 h in the presence of recombinant human IL-4 to amplify their production of MIP-1β (30). For cross-linkage of the high-affinity Fc receptor for IgE (Fc
RI), cells were incubated overnight in the presence of a purified human myeloma IgE (Calbiochem). The cells were washed and resuspended in fresh StemPro medium containing SCF (100 ng/ml) at a density of 5 x 105/ml. Triplicate samples of cells were stimulated with exogenous LTD4 (100 or 500 nM), UDP (1 µM), or with a cross-linking rabbit anti-human IgE polyclonal IgG Ab in the wells of a 96-well flat-bottom plate. The supernatants were harvested at 6 h, a time point at which MIP-1β production was maximal. Calcium flux was measured as described previously (8).
Mice
Cysltr1–/– mice (C57BL/6-Cysltr1tm1Ykn), Cysltr2–/– mice (C57BL/6-Cysltr2tm1Ykn), and Ltc4s–/– mice (C.129S7-Ltc4stm1Blam), each backcrossed for at least 10 generations onto a BALB/c background), were derived as previously described (3, 9, 37). All animal studies were approved by the Animal Care and Use Committee of the Dana-Farber Cancer Institute.
SDS-PAGE immunoblotting
Cytokine-starved mBMMCs were stimulated for 15 min with various doses of LTD4, SCF (100 ng/ml), or medium, lysed, and processed for Western blotting as described previously (29). Blots were probed with anti-Active ERK, and total ERK (Cell Signaling Technologies). The signals were detected by chemiluminescence and quantitated by densitometry. For each genotype, the density of the corrected phosphorylated ERK band was set as the control and used as a reference for the stimulated samples in each experiment.
Flow cytometry
Surface CysLT1R and CysLT2R were detected in unpermeabilized cells using polyclonal Abs RB34 and RB19, respectively, directed against extracellular loops of the respective GPCRs (32). P2Y6 receptor protein was detected on permeabilized cells with polyclonal Ab RB165 as detailed elsewhere (28). Specificity of the staining was confirmed by demonstrating diminished levels of staining in cells treated with sequence-selective short hairpin RNAs (shRNAs). Nonspecific rabbit and mouse IgG (BD Biosciences) were used as controls. Flow cytometry analyses were performed on a FACSort Calibur flow cytometer and data were analyzed with CellQuest Pro software (BD Immunocytometry Systems). Data were calculated as net mean fluorescence intensity (MFI) (MFI of the primary Ab staining – MFI staining of the IgG control).
Preparation of lentiviral particles and transfection
shRNA constructs were purchased from Open Biosystems. The constructs were designed to include a hairpin of 21 bp of a sense strand and an antisense strand and a 6-bp loop. The sequences were (reading from 5'–3'): CysLT1R, CCGGGCGTGACTTATGTACCCAG AACTCGAGTTCTGGGTACATAAGTCACGCTTTTT; CysLT2R, CCGGCCTGCAGGATTA TGTCTTAT TCTCGAGAATAAGACATAATCCTGCAGGTTTTT; and P2Y6 receptor, CCGGGCAGCCTTCATATTT GCCATTCTCGAGAATGGCAAATATGAAGGCTGCTTTTT. Each hairpin sequence was cloned in frame with a lentiviral vector (pLKo1; Open Biosystems), and infectious virus was derived as described elsewhere (10). Viral stocks were added directly to the medium at a quantity sufficient to achieve a multiplicity of infection of 10, based on preliminary experiments in which FACS analysis was used to assess the effect of the transfection on the expression of the target receptors.
Immunostains and confocal imaging
LAD2 cells were fixed in a suspension with 2% paraformaldehyde in PBS on ice for 10 min. The cells were washed once with wash buffer (PBS containing 0.1% sodium azide) and resuspended in blocking buffer (PBS containing 5% horse serum), immobilized on round 12-mm coverslips by cytocentrifugation, and postfixed with methanol at –20°C for 10 min. The slides were washed twice with wash buffer and then blocked with blocking buffer, with shaking at room temperature for 30 min. The cells were stained with primary Abs that were labeled directly with Alexa Fluor 488 or Cy3 according to the manufacturers specifications. The cells were counterstained with DRAQ5 nuclear stain (1/1000; Biostatus). Anti-CysLT1R (RB34) and polyclonal anti-P2Y6 receptor Abs were used at 5 µg/ml and species-matched control IgG was used at the same concentration. Subcellular localization of the receptors was assessed using a Nikon TE2000-U inverted microscope with a Nikon C1 plus laser-scanning confocal system as detailed elsewhere (10).
FLIM analysis
The cells were fixed and stained for FLIM analysis as described above for confocal imaging, except for the omission of the nuclear stain. Cells were double stained with anti-C-terminal CysLT1R Ab directly labeled with Cy3 (as the acceptor fluorophore) and anti-P2Y6 receptor Ab labeled with Alexa Fluor 488 (as the donor fluorophore). Positive controls were cells stained with primary rabbit anti-human
4 integrin Ab and mouse anti-human β1 integrin Ab, followed by counterstaining with secondary anti-rabbit-Alexa Fluor 488 (donor) and anti-mouse-Cy3 (acceptor) conjugates at a concentration of 1 µg/ml each. Additional experiments were performed with directly labeled primary Abs (data not shown). Alexa Fluor 488 and Cy3 were chosen because the excitation spectrum of Cy3 overlaps with the emission spectrum of Alexa Fluor 488, making them a good pair for demonstrating fluorescence resonance energy transfer (FRET). Primary Ab was omitted from some samples as a control. Images were taken using 488- and 543-nm single-photon excitation to excite Alexa Fluor 488 and Cy3, respectively (Zeiss LSM510/NLO). Emissions were separated in multitrack mode. FLIM was performed using a femtosecond-pulsed Ti:Sapphire laser (Chameleon; Coherent), with a multichannel plate photon counting detector (R3809; Hamamatsu). A time-correlated single-photon counting board and software (SPC830; Becker and Hickl) was used for acquisition, and images were analyzed using SPC Image software (version 2.6.1.2711; Becker & Hickl) with mono- and bi-exponential lifetime curve fits (38, 39). At least three fields were analyzed for each condition in each experiment. Lifetimes were recorded as whole-cell measurements. FRET is indicated by shortening of the donor fluorophores lifetime and will occur if the two fluorophores are within 10 nm of each other. The lifetime of the donor fluorophore (Alexa Fluor 488) was measured in the absence of the acceptor fluorophores (negative control) in each experiment.
Statistical methods
Data are expressed as mean ± SEM from at least three experiments, except where otherwise indicated. Differences between treatment groups were determined with the Student t test. For the FRET studies, differences in lifetimes were assessed by ANOVA, with a Bonferroni analysis to correct for multiple comparisons.
| Results |
|---|
|
|
|---|
To determine whether MCs expressed P2Y6 receptors, we developed an anti-peptide polyclonal Ab (RB165) against a sequence corresponding to the second intracellular loop of the mouse and human P2Y6 receptor. After purification, the Ab was used to stain Chinese hamster ovary (CHO) cells stably expressing human P2Y6 receptors (8) and the parent cell line as positive and negative controls, respectively. The same Ab was used in SDS-PAGE immunoblotting assays to detect P2Y6 receptor protein in lysates. The Ab detected a strong 34-kDa band in lysates of CHO cells transfected with human P2Y6 cDNA, corresponding to the published size of the P2Y6 receptor protein in intestinal epithelial cells (26) (Fig. 1A). The band was weakly detectable in some samples of the parent CHO cell line, suggesting low-level endogenous expression. Permeabilized transfectants displayed a robust signal for P2Y6 receptor protein on FACS compared with the untransfected parent line (Fig. 1A). Primary hMCs, LAD2 cells, and mBMMCs all expressed P2Y6 receptor protein as determined by FACS analysis with RB165 (Fig. 1B). Confocal imaging of LAD2 cells (Fig. 1C) and mBMMCs (data not shown) revealed that P2Y6 receptor protein localized mainly to plasma membrane (Fig. 1C, arrows), particularly within cytoplasmic protrusions (microplicae).
|
LTD4 induces calcium signaling in MCs exclusively through CysLT1R (8). In primary hMCs, this response exhibits partial desensitization with UDP. To determine whether LTD4-induced calcium signaling involved the P2Y6 receptor, LAD2 cells were loaded with fura 2-AM dye and stimulated with LTD4 or UDP. CHO cells expressing the human CysLT1R (CHO-CysLT1R cells) were used as a control. Some cell samples were treated with various concentrations of MRS2578 (a selective antagonist of P2Y6 receptor signaling) (34) or MK571 (a competitive antagonist of CysLT1R that also interferes with some P2Y receptors (32)). LTD4 (10–500 nM) induced calcium flux in both LAD2 cells (Fig. 2A) and CysLT1R-expressing CHO cells (Fig. 2B), which was blocked virtually completely by MK571 (1 µM; Fig. 2A). Calcium flux induced in LAD2 cells by 500 nM LTD4 was unaltered by treatment of the cells with MRS2578 at lower doses (1 and 5 µM), but attenuated when the cells were treated with a higher dose (10 µM, as shown for one experiment, Fig. 2A). The same dose of MRS2578 failed to attenuate LTD4-mediated calcium flux in the CHO-CysLT1R cells (Fig. 2B). UDP at a high dose (100 µM) weakly induced calcium flux in LAD2 cells that was blocked by MRS2578 (1 µM; data not shown).
|
UDP and cys-LTs support survival of MCs through CysLT1R-P2Y6 receptor cooperation and endogenous cys-LTs
CysLT1R-mediated ERK phosphorylation promotes proliferation of MCs (30). ERK signaling is also essential for the survival of MCs in vivo. To determine whether LTD4 and UDP could enhance the survival of MCs and whether these effects involved cooperation between P2Y6 receptors, CysLT1R, and endogenously produced cys-LTs, mBMMCs were withdrawn from IL-3 and maintained in medium supplemented with SCF (10 ng/ml) as a positive control, UDP (0.01, 0.1, or 1 µM), or LTD4 (100 or 500 nM). Entry into apoptosis was monitored by annexin V staining 48 h later. At the highest doses tested, both LTD4 and UDP substantially attenuated apoptosis of wild-type mBMMCs (Fig. 3, A and B). LTD4- and UDP-mediated cytoprotection were abolished in mBMMCs lacking CysLT1R or LTC4S (Fig. 3, A and B). CysLT2R-null mBMMCs were not different from wild-type cells in their cytoprotective responses to LTD4 or UDP. In wild-type mBMMCs, both UDP- and LTD4-mediated cytoprotection were sensitive to blockade of P2Y6 receptors by MRS2578 (1 µM; Fig. 3, C and D).
|
We had previously reported that UDP-mediated signaling and cytokine generation by hMCs was sensitive to blockade by MK571 (8, 30). Because chemokine generation by MCs in response to LTD4 requires both calcium- and ERK-dependent transcriptional mechanisms (30) and because interference with P2Y6 receptors blunted ERK signaling in response to LTD4 (Fig. 2), we sought to clarify whether P2Y6 receptors regulated CysLT1R-dependent MIP-1β production by MCs and whether this response involved endogenous nucleotides. LAD2 cells were primed with IL-4 for 72 h to augment their generation of cytokines (30) and were treated with recombinant lentiviruses containing shRNAs directed toward the CysLT1R, CysLT2R, and P2Y6 receptor sequences, or with a control lentivirus without shRNA for the last 48 h of priming. FACS analyses confirmed the knockdown of the target proteins, without altering the nontarget receptors (as shown for CysLT1R and P2Y6 receptors, Fig. 4A). Stimulation of the empty vector-treated control cells with LTD4-induced dose-dependent secretion of MIP-1β, exceeding the quantities generated in response to the positive control cells subjected to Fc
RI cross-linkage. MIP-1β production in response to LTD4 at 100 or 500 nM was abrogated by the knockdown of CysLT1R, but not knockdown of CysLT2R (Fig. 4B). Knockdown of P2Y6 receptors decreased production of MIP-1β in response to LTD4 by >50% but did not interfere with Fc
RI-mediated MIP-1β production (Fig. 4B). UDP tended to weakly induce MIP-1β production by itself at 1000 nM; this response was abrogated by knockdown of either CysLT1R or the P2Y6 receptor, but not by the CysLT2R knockdown.
|
RI-mediated MIP-1β generation. Lack of direct interaction between CysLT1R and P2Y6 receptors
To determine whether P2Y6 receptors colocalized with CysLT1R, we performed confocal imaging of LAD2 cells that were double stained with RB34 and RB165 Abs directly conjugated to Alexa Fluor 488 and Cy3, respectively. CysLT1R protein localized to the plasma membrane (Fig. 5A, left panel), as well as to the nuclear envelope and punctuate intranuclear structures as described previously (32). Double staining revealed colocalization of CysLT1R and P2Y6 receptors primarily at the plasma membrane and within microplicae (Fig. 5A, middle panel, arrows). Double staining of the LAD2 cells with a monoclonal anti-β1 integrin Ab and a polyclonal anti-
4 integrin Ab showed the anticipated prominent colocalization at the plasma membrane (Fig. 5A, right panel, arrows). Multiphoton-based FLIM analysis (which reveals a shortening of the lifetime of the fluorescence emitted by Alexa Fluor 488 if the Cy3-conjugated Ab is within 10 nm of the donor fluorophore (39, 40)) showed no energy transfer between CysLT1R and P2Y6 receptors (Fig. 5B, left and middle panels). In contrast, double staining for both
1 and β1 integrins showed strong energy transfer (Fig. 5B, right panel). These results suggest that despite the cellular (or membrane) colocalization of CysLT1R and P2Y6, there is likely no direct interaction between these receptors.
|
| Discussion |
|---|
|
|
|---|
Although human GPR17 and P2Y4 receptors recognize UDP at high concentrations when expressed in heterologous cell systems (21, 43), P2Y6 receptors are the only high-affinity, UDP-selective GPCR described thus far (22, 23). To determine whether human and mouse MCs express this receptor, we raised an Ab against a conserved sequence of the second intracellular loop of P2Y6 receptors to detect both the mouse and human proteins. We verified that MCs from both species expressed P2Y6 receptors using flow cytometry (Fig. 1B). The specificity of the staining with the Ab was confirmed by its ability to recognize recombinant P2Y6 receptors expressed by transfected CHO cells (Fig. 1A) and by the abrogation of the FACS signal on MCs subjected to the knockdown procedure (Fig. 4A). Confocal imaging revealed prominent P2Y6 receptor staining of LAD2 cells (Fig. 1C). Thus like other cells of the innate immune system, MCs express P2Y6 receptors. Given that this receptor functions in autocrine signal amplification in other effector cell types (25, 26), we sought such functions in MCs stimulated with LTD4, the highest affinity ligand for CysLT1R.
Nucleotides released by MCs in a regulated fashion after activation can propagate Fc
RI-mediated calcium signaling from cell to cell (12). Since there is no accurate method to measure UDP in extracellular fluids, we performed calcium flux assays in the presence of the selective P2Y6 receptor antagonist MRS2578 and the CysLT1R antagonist MK571. The complete blockade of the LTD4-mediated calcium flux by MK571 is consistent with a CysLT1R-depedent event (Fig. 2A), but does not preclude the involvement of extracellular nucleotides since this reagent can also block P2Y receptors. MRS2578 at 1 µM (a dose sufficient to completely block signaling through the recombinant P2Y6 receptors (34) did not alter calcium flux induced by LTD4, but did decrease the response to LTD4 when the concentration was increased to 10 µM (Fig. 2A). MRS2578 does not exert nonselective actions at other P2Y receptors at this high dose (34), and the same dose of MRS2578 failed to alter calcium fluxes at recombinant CysLT1R in CHO cells (Fig. 2B), suggesting that it does not behave as a partial CysLT1R antagonist. We cannot completely exclude another potential off-target effect of MRS2578 in these experiments, but the results support a potential role for endogenous uracil nucleotides in regulating calcium signaling through CysLT1R.
Although mBMMCs do not exhibit the robust cys-LT-induced calcium fluxes of hMCs or LAD2 cells, they do respond to LTD4 with potent phosphorylation of ERK that depends on CysLT1R-dependent transactivation of the c-Kit receptor tyrosine kinase (29). We previously demonstrated that UDP, like LTD4, also induces ERK activation in primary hMCs (30). We found that UDP caused MRS2578-sensitive ERK phosphorylation, indicating a functional response mediated by P2Y6 receptors (Fig. 2C). At a low dose (1 µM), MRS2578 decreased the LTD4-dependent ERK signal (Fig. 2, C and E). It is thus likely that endogenous UDP released by stimulation through CysLT1R amplifies the subsequent phosphorylation of ERK through P2Y6 receptors. As anticipated, mBMMCs from CysLT1R-null mice (but not from CysLT2R-null mice) showed no LTD4-mediated ERK activation. Unexpectedly, however, UDP also failed to induce ERK activation in the absence of CysLT1R and in the absence of LTC4S (Fig. 2, D and E). Although UDP and LTD4 do not induce the release of cys-LTs from MCs in concentrations sufficient to be detected in the medium (30), these findings suggest that both stimuli induce the formation of intracellular LTC4. Because LTC4S is integral to the nuclear envelope (44), while CysLT1R localizes to both the cell surface and nucleus (31), endogenous LTC4 could either serve a synaptic function at the cell surface or could activate intracellular receptors.
ERK signaling is important for MC survival and UDP can act through P2Y6 receptors to promote survival in some cell types (23). We investigated whether LTD4 or UDP would protect cytokine-starved MCs from apoptosis, whether this response also involved reciprocal cross-regulation among CysLT1R, P2Y6 receptors, and endogenous cys-LTs. As was the case for the experiments involving ERK activation, we used mBMMCs to take advantage of cells with deletions of the relevant genes. As anticipated, SCF protected cells from apoptosis irrespective of genotype. Cytoprotection conferred by either LTD4 or UDP required the presence of CysLT1R and LTC4S, but not of CysLT2R (Fig. 3, A and B) and was blocked by interference with UDP signaling through P2Y6 receptors using MRS2578 (Fig. 3, C and D). Thus, the cross-talk among CysLT1R, LTC4S, and P2Y6 at the level of ERK activation (Fig. 2) is paralleled by effects on protection from apoptosis. Our previous study demonstrated that both LTC4S and CysLT1R expression were also necessary for IL-4, an accessory mitogen for MCs, to induce proliferation (29). The profound MC deficiency in the airway of allergen-challenged mice lacking LTC4S (45) could thus reflect a requirement for endogenous cys-LTs to facilitate cell survival and/or proliferation in circumstances where local concentrations of SCF, the obligate MC growth factor, are limited.
MC-derived chemokines play a key role in recruiting blood-borne effector cells during innate immune responses (46). We had previously reported that cys-LTs and UDP could induce chemokine generation by primary hMCs through an IL-4-dependent pathway (30), the latter being essential to amplify ERK activation (47). Because mBMMCs do not generate abundant chemokines when stimulated with cys-LTs (data not shown), we used LAD2 cells for these experiments and manipulated the key receptors using shRNA knockdowns. IL-4-primed LAD2 cells were stimulated with LTD4 or UDP after transfection with the shRNAs. The abrogation of MIP-1β production in response to LTD4 by the knockdown of CysLT1R (Fig. 4B), but not by the CysLT2R knockdown, confirms the essential requirement for CysLT1R in chemokine production that had been inferred from earlier pharmacologic studies (30). Strikingly, P2Y6 receptor knockdown substantially decreased the generation of MIP-1β in response to LTD4, but not to Fc
RI cross-linkage (Fig. 4B). The effect was not due to an off-target effect of the P2Y6 knockdown on CysLT1R expression levels or vice-versa (Fig. 4A). Combined with the marked inhibition of LTD4-dependent MIP-1β production by MRS2578, suramin (a nonselective P2Y receptor antagonist), and depletion of extracellular nucleotides by apyrase (Fig. 4C), these studies indicate that P2Y6 receptors respond to UDP at the plasma membrane after its CysLT1R-dependent release from MCs. Moreover, the fact that MK886 also abrogated the response to LTD4 indicates that endogenous cys-LTs are required for chemokine generation induced by exogenous LTD4, as was the case for ERK activation (Fig. 2) and cytoprotection (Fig. 3).
Structurally homologous GPCRs form multimers that function as signaling units (48, 49, 50, 51, 52), as we had demonstrated previously for CysLT1R and CysLT2R (31). We confirmed that CysLT1R and P2Y6 colocalized at the plasma membranes of LAD2 cells (Fig. 5A), but Ab-based FLIM, a sensitive method for detecting protein-protein interactions in primary cells, showed no evidence for a direct interaction between CysLT1R and P2Y6 receptors (Fig. 5B). The colocalization of CysLT1R and P2Y6 receptors, perhaps within lipid microdomains of the plasma membrane, may facilitate some of the synergy of their respective signaling events, particularly if LTD4 elicits UDP release in close proximity to CysLT1R at the plasma membrane. Since the FLIM assay requires that the Abs be directed against epitopes on the respective GPCRs that are in close proximity to detect oligomerization, we cannot completely exclude the existence of CysLT1R-P2Y6 receptor oligomers. However, the lack of a FRET signal in the same setting where CysLT1R and CysLT2R heterodimerizes (31) is consistent with GPCR oligomerization being tightly regulated and specific.
Our study indicates that CysLT1R and P2Y6 receptors orchestrate a coordinated, complementary series of signaling events in MCs involving their respective endogenous ligands. Given its subcellular distribution, it is likely that the P2Y6 receptor responds primarily to extracellular UDP at the cell surface released in response to CysLT1R-dependent MC activation. The requirement for endogenous cys-LTs and CysLT1R for cytoprotection and cytokine generation could reflect synaptic signaling functions at the cell surface or possibly at intracellular sites (Fig. 6). The interplay between these receptors could extend to smooth muscle, intestinal epithelial cells, macrophages, and other hematopoietic effectors that express both (53, 54, 55). The relevance of this interplay in vivo for the function of mast cells and other cell types remains to be determined. It is important to note that uracil nucleotides and cys-LTs are both implicated in proliferation and/or cytoprotection of epithelial cells and smooth muscle (56, 57). Given the established role of cys-LTs in asthma, it is also noteworthy that bronchoalveolar lavage fluids from allergen-challenged asthmatic individuals contain high levels of extracellular nucleotides and that experimental allergen-induced bronchial inflammation in mice is profoundly inhibited by suramin treatment (58). It will be essential to determine the role for P2Y6 receptors in fibrosis, remodeling, cardiovascular disease, and other pathologic processes in which MCs and cys-LTs are implicated as effectors to reveal their potential as therapeutic targets.
|
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported by National Institutes of Health Grants AI-48802, AI-52353, AI-31599, HL-36110, and EB-00768 and by grants from the Charles Dana Foundation and the Vinik Family Fund for Research in Allergic Diseases. ![]()
2 Address correspondence and reprint requests to Dr. Joshua A. Boyce, Brigham and Womens Hospital, Harvard Medical School, One Jimmy Fund Way, Smith Building Room 626, Boston, MA 02115. E-mail: jboyce{at}rics.bwh.harvard.edu ![]()
3 Abbreviations used in this paper: LT, leukotriene; cys-LT, cysteinyl LT; CysLT1R, type 1 receptor for cys-LT; CysLT2R, type 2 receptor for cys-LT; Fc
RI, high-affinity Fc receptor for IgE; FLAP, 5-lipoxygenase activating protein; GPCR, G protein-coupled receptor; hMC, cord blood-derived human MC; LTC4S, LTC4 synthase; MC, mast cell; P2Y, purinergic; SCF, stem cell factor; shRNA, short hairpin RNA; MC, mast cell; UDP, uridine diphosphate; mBMMC, mouse bone marrow-derived MC; FLIM, fluorescence lifetime imaging microscopy; MFI, mean fluorescence intensity; FRET, fluorescence resonance energy transfer; CHO, Chinese hamster ovary. ![]()
Received for publication September 26, 2008. Accepted for publication November 5, 2008.
| References |
|---|
|
|
|---|
B and increase survival of osteoclasts. J. Biol. Chem. 280: 16909-16915.
RI or Fc
RI. Leuk. Res. 27: 677-682. [Medline]
and induces prostaglandin D2 generation by human mast cells. J. Biol. Chem. 283: 16477-16487. Related articles in The JI:
This article has been cited by other articles:
![]() |
G. Woszczek, L.-Y. Chen, S. Alsaaty, S. Nagineni, and J. H. Shelhamer Concentration-Dependent Noncysteinyl Leukotriene Type 1 Receptor-Mediated Inhibitory Activity of Leukotriene Receptor Antagonists J. Immunol., February 15, 2010; 184(4): 2219 - 2225. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Paruchuri, H. Tashimo, C. Feng, A. Maekawa, W. Xing, Y. Jiang, Y. Kanaoka, P. Conley, and J. A. Boyce Leukotriene E4-induced pulmonary inflammation is mediated by the P2Y12 receptor J. Exp. Med., October 26, 2009; 206(11): 2543 - 2555. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |