If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
To whom correspondence should be addressed: Laboratory of Allergic Diseases, NIAID, National Institutes of Health, Bldg. 10, Rm. 11C206, 10 Center Dr. MSC 1881, Bethesda, MD 20892-1881. Tel.: 301-496-8757; Fax: 301-480-8384;
* This work was supported in part by the NIAID, National Institutes of Health Intramural Program. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ** Supported by a fellowship from Boehringer Ingelheim Fonds.
Mast cell degranulation following FcϵRI aggregation is generally believed to be dependent on phosphatidylinositide 3-kinase (PI 3-kinase)-mediated phospholipase C (PLC)γ activation. Here we report evidence that the PLCγ1-dependent pathway of FcϵRI-mediated activation of mast cells is independent of PI 3-kinase activation. In primary cultures of human mast cells, FcϵRI aggregation induced a rapid translocation and phosphorylation of PLCγ1, and subsequent inositol trisphosphate (IP3) production, which preceded PI 3-kinase-related signals. In addition, although PI 3-kinase-mediated responses were completely inhibited by wortmannin, even at high concentrations, this PI 3-kinase inhibitor had no effect on parameters of FcϵRI-mediated PLCγ activation, and had little effect on the initial increase in intracellular calcium levels that correlated with PLCγ activation. Wortmannin, however, did produce a partial (∼50%) concentration-dependent inhibition of FcϵRI-mediated degranulation in human mast cells and a partial inhibition of the later calcium response at higher concentrations. Further studies, conducted in mast cells derived from the bone marrow of mice deficient in the p85α and p85β subunits of PI 3-kinase, also revealed no defects in FcϵRI-mediated PLCγ1 activation. These data are consistent with the conclusion that the PLCγ-dependent component of FcϵRI-mediated calcium flux leading to degranulation of mast cells is independent of PI 3-kinase. However, PI 3-kinase may contribute to the later phase of FcϵRI-mediated degranulation in human mast cells.
Aggregation of the high affinity receptor for IgE (FcϵRI) on the surface of mast cells initiates a cascade of intracellular signaling events leading to degranulation and the release of proinflammatory mediators. Activation of PLC
). PLC catalyzes the hydrolysis of membrane-associated PIP2 thereby generating diacylglycerol and inositol trisphosphate; essential factors, respectively, for the activation of protein kinase C and intracellular calcium mobilization (
), or recruitment of cytosolic tyrosine kinases as is the case with members of the immunoglobulin superfamily. Generally, PLCγ1 and PLCγ2 isoforms are independently expressed in specific cells types. For example, T cells express only PLCγ1 (
), however, our data with human mast cells indicated that this is not the case. This finding prompted us to investigate the pathways of activation of PLCγ1 and PLCγ2 and the role of PI 3-kinase in the intracellular events leading to degranulation in CD34+ peripheral blood progenitor-derived human mast cells (HuMCs) after FcϵRI aggregation. To further investigate the role of PI 3-kinase in FcϵRI-dependent PLCγ1-mediated mast activation, we examined these responses in BMMCs deficient in the p85α and p85β subunits of PI 3-kinase. As will be shown, the immediate PLCγ1-mediated component of mast cell degranulation following FcϵRI aggregation is independent of PI 3-kinase, however, PI 3-kinase may contribute to a later phase of the degranulation response.
Cell Isolation and Culture—HuMCs were developed from CD34+ cells in StemPro-34 culture media (Invitrogen) containing l-glutamine (2 mm) (Biofluids, Rockville, MD), penicillin (100 units/ml) (Biofluids), streptomycin (100 μg/ml) (Biofluids), IL-6 (100 ng/ml) (PeproTech), and stem cell factor (100 ng/ml) (PeproTech), as described (
). IL-3 (30 ng/ml) was included for the first week of culture. Experiments were conducted on these cells 8–10 weeks after the initiation of culture, at which point, the population was greater than 99% mast cells.
For studies using mouse BMMCs the following KO mice were obtained: p85α-/-:p85β+/+ (BALB/c) (
). The absence or presence of p85α and p85β in these mice was confirmed by PCR and, in the case of p85α, by immunoblot analysis (data not shown). BMMCs were obtained by flushing bone marrow cells from the femurs of either these KO mice or wild type mice, then culturing for 4–6 weeks in RPMI 1640 supplemented with 10% fetal calf serum, glutamine (4 mm), sodium pyruvate (1 mm), penicillin (100 units/ml), streptomycin (100 μg/ml), nonessential amino acids (1 mm), HEPES (25 mm), β-mercaptoethanol (50 mm), and mouse recombinant IL-3 (30 ng/ml) (Peprotech). At this point, the murine mast cell population was greater than 99% pure. RBL 2H3 cells and the U937 human monocytic cell line were cultured in Iscove's media, supplemented with fetal calf serum (10%) (Biofluids), l-glutamine (2 mm), penicillin (100 units/ml), and streptomycin (100 μg/ml). Cultures were maintained at 37 °C in a humidified incubator of 95% air, 5% CO2.
Cell Activation—HuMCs were sensitized overnight in culture media containing chimeric human Fc anti-4-hydroxy-3-nitrophenylacetyl (NP)-specific IgE (NP-IgE) (Serotec, Raleigh, NC) (1 μg/ml) and then triggered by the addition of 10 μl of 10× NP-BSA (30:1) conjugate (Biosearch Technologies Inc. Novoto, CA) (final concentration 100 ng/ml), as described (
). When the effects of inhibitors were examined, these, or controls, were added 10 min prior to the addition of NP-BSA. BMMCs were similarly treated, however, these cells were sensitized using mouse monoclonal anti-dinitrophenyl IgE and then triggered by the addition on dinitrophenyl-human serum albumin.
Subcellular Fractionation—To separate the membrane and cytosolic fractions, sensitized HuMCs or BMMCs were re-suspended in 200 μl of HEPES/BSA (0.04%) and activated as above. At predetermined times, the tubes were transferred into ice and equal volumes of ice-cold lysis buffer (Tris-HCl, 20 mm; dithiothreitol, 2 mm; EGTA, 1 mm; EDTA, 2 mm; pH 7.5), containing protease and phosphatase inhibitors were added (
). The cells were disrupted by sonication twice on wet ice for 10 s and then the tubes were spun at 20,800 × g for 15 min at 4 °C and the supernatants (cytosolic fractions) and the pellets (membrane fractions) were re-suspended in sample buffer. The samples were then boiled for 3 min and centrifuged at 20,800 × g for 5 min prior to loading the samples onto gels. For inhibitor studies, the cells were pretreated with PP2, piceatannol, or wortmannin (Calbiochem, Carlsbad, CA) for 10 min prior to the addition of antigen.
Gel Electrophoresis—All gel electrophoresis supplies were obtained from Invitrogen. Proteins were separated on 4–12% NuPage BisTris gels and probed with the following primary antibodies: anti-phosphotyrosine (4G10) conjugated to biotin, anti-human p72syk (UBI, Lake Placid, NY), anti-rodent p72syk, anti-PI 3-kinase p85α and p110α, -β, and -δ, anti-PLCγ1, anti-PLCγ2, (Santa Cruz Biotechnology, Santa Cruz, CA), anti-phospho-PLCγ1 (pY783), (BIOSOURCE, Camrillo, CA), anti-phospho-Src (pY416), anti-phospho-AKT (polyclonal pS473 and monoclonal pS473), anti-phospho-p70 S6 kinase (pT389) (Cell Signaling Technology, Beverly, MA), and anti-PI 3-kinase p85β (Acris, Bad Nauheim, Germany). Following rinsing and incubation with the appropriate horseradish peroxidase-conjugated secondary antibody or streptavidin, the immunoreactive bands were visualized utilizing a Renaissance Western blot chemiluminescence kit (PerkinElmer Life Sciences, Boston, MA). To confirm equal protein loading onto gels, membranes were stripped and re-probed as described (
) with anti-p72syk. Alternatively, identically loaded membranes were probed for normalization. For translocation experiments, protein equivalents were established by probing the membrane extracts with anti-Kit (CD117) (Santa Cruz Biotechnology) and/or heat shock protein (HSP)90 (BD Transduction Laboratories, San Diego, CA). Kit was used as a marker for membrane proteins and HSP90 as a marker for the presence of cytosolic proteins in the membrane fraction. Although, in some experiments, HSP90 was detected in the membrane fraction, this was generally a minor component of total cellular protein and was unaffected by antigen challenge. To quantitate changes in protein phosphorylation, the ECL films were scanned using an ImageQuant 5.0 scanner (Amersham Biosciences).
GST Fusion Protein Capture Studies—HuMCs were stimulated and processed as for immunoprecipitation studies. After 2 min of challenge with NP-BSA (100 ng/ml), boiling lysis buffer was added to the cells and processed as above. The supernatants were diluted 10× in Tris-buffered saline containing Triton X-100 (0.1%) and protease and phosphatase inhibitors, and precleared with GST-agarose for 1 h at room temperature under rotation. After centrifugation, the supernatants were removed and incubated with 10 μg of agarose beads-GST-SH2-SH2 domains of PLCγ1 or the GST-SH3 domain of PLCγ1 (Santa Cruz) for 2 h at room temperature under rotation. The beads were washed 3× with Triton X-100 (0.1% in Tris-buffered saline) and the proteins were solubilized and separated by electrophoresis as above. The samples were then probed with anti-Vav, anti-SLP76 (UBI), and anti-LAT (gift from Larry Samelson, NCI, National Institutes of Health) antibodies. Fusion proteins alone were also run on the gels as controls for nonspecific binding of the primary and secondary antibodies.
IP3Assay—Cellular IP3 concentrations were determined utilizing a commercially available kit (Amersham Biosciences) according to the manufacturers instructions. Briefly, following cell activation, cells were extracted with trichloroacetic acid on ice for 15 min. IP3 was then extracted from the trichloroacetic acid precipitate using 1,1,2-trichlorofluorethane-trioctylamine. The IP3 containing fractions were then incubated for 1 h at 4 °C with membrane fractions of calf cerebellum containing the IP3 receptor and tracer d-myo-[H3]inositol 1,4,5-trisphosphate. The mixtures were then centrifuged at 3700 × g for 10 min and the tracer remaining bound to the membranes was measured with a β-scintillation counter. The results were expressed as picomoles of IP3 per 106 cells.
Calcium Measurements—HuMCs and BMMCs were sensitized as above and then re-suspended in HEPES buffer containing 0.04% BSA, 0.3 mm sulfinapyrazole, and 0.5 μm Fura-2 AM (Molecular Probes, Eugene, OR). Cells were then incubated for 30 min, washed twice in the same buffer (without Fura-2 AME), and plated in black 96-well plates (Packard Biosciences, Meriden CT; CulturPlate™ -96F) at a density of 10,000 cells/100 μl/well. An additional aliquot of cells was treated similarly, except that Fura-2 AM was omitted from the protocol (non-loaded cells) and plated alongside Fura-2-loaded cells. Fluorescence was determined at 510 nm with alternating excitation at 340 and 380 nm in the Wallac Victor2 1420 Multilabel Counter (PerkinElmer Life Sciences). The instrument was set at 37 °C and a top reading mode. After subtraction of background fluorescence of non-loaded cells, data were calculated as the ratio of fluorescence at 340 and 380 nm excitation wavelengths.
Degranulation Experiments—Degranulation was monitored by determining the release of β-hexosaminidase as described (
) containing BSA (0.04%), HuMCs were suspended in the same buffer at 50,000 cells/ml, then triggered by the addition of NP-BSA in a 100-μl total volume. BMMCs were suspended at a concentration of 1 million cells/ml and then triggered in a volume of 200 μl. When the effects of inhibitors were examined, these compounds (or carrier buffer for controls) were added 10 min prior to the addition of NP-BSA. The experiments were terminated by centrifugation at 4 °C and then aliquots of the supernatants were removed for β-hexosaminidase assay. The remaining cells were lysed by the addition of distilled water and freeze-thawing. The β-hexosaminidase content of the supernatants and cell lysates was then determined as described and the release calculated as the % of total β-hexosaminidase (cells + supernatant) found in the supernatant.
Phosphatidylinositol 3,4,5-Trisphosphate (PIP3) Production—HuMCs were sensitized as above and then transferred to media without growth factors for 2 h. Following rinsing in HEPES buffer containing 0.04% BSA, the cells were loaded with [32P]orthophosphate (200 μCi/ml) (Amersham Biosciences) for 1 h in the same buffer, rinsed, then 2.5–3 × 106 cells placed in 1.5-ml polyethylene tubes prior to stimulation for the indicated times with NP-BSA (100 ng/ml) in the absence or presence of wortmannin (100 nm). PIP3 production in HuMCs and BMMCs was subsequently measured as described in Ref.
Expression and Activation of PLCγ —To first examine which isoforms of PLCγ were expressed in HuMCs, cell lysates were probed with anti-PLCγ1 and PLCγ2 antibodies. Extracts of other cell types known to express these isoforms were used as positive controls. As expected, the U937 cell line, mouse BMMCs, and RBL 2H3 cells expressed both PLCγ isoforms (Fig. 1a). Similarly both PLCγ1 and PLCγ2 were found in HuMCs.
Having demonstrated that PLCγ1 and PLCγ2 were both expressed in HuMCs, we next determined whether aggregation of FcϵRI resulted in activation of these molecules as assessed by PLCγ translocation and phosphorylation, IP3 production, and intracellular calcium mobilization. HuMCs were sensitized overnight and activated by the addition of NP-BSA (
). This resulted in a rapid translocation of both PLCγ1 and PLCγ2 to the mast cell plasma membrane, which was observed as early as 5 s after cell activation (Fig. 1b). Translocation of PLCγ2, however, was not as marked as the translocation of PLCγ1. We monitored PLCγ phosphorylation in HuMC lysates by probing with an activation state-specific anti-phospho-PLCγ1 antibody (Fig. 1c). There was little constitutive phosphorylation of PLCγ1 in resting HuMCs. However, FcϵRI aggregation induced a marked increase in phosphorylation of the active regulatory tyrosine residue of PLCγ1. This was followed by a noticeable, but incomplete, dephosphorylation by 120 s after FcϵRI aggregation. In these studies, no phosphorylation of PLCγ2 was observed in HuMCs, utilizing conditions (
) under which PLCγ2 phosphorylation was observed in mouse BMMCs (data not shown). However, we cannot rule out the possibility that PLCγ2 is activated in HuMCs following FcϵRI aggregation but is under the level of detection.
We then confirmed the production of IP3, known to be PLCγ-dependent in HuMCs following FcϵRI aggregation. From Fig. 1d it can be seen that aggregation of FcϵRI resulted in a rapid increase in the production of IP3 that maximized between 10 and 20 s after FcϵRI aggregation. IP3 levels subsequently decreased but still remained elevated for up to 120 s. Finally, as before (
), we observed a rapid increase in intracellular free Ca2+ levels following FcϵRI aggregation (Fig. 1e). The initial increase closely followed the kinetics of PLCγ1 activation described above, however, levels continued to increase until 120 s after FcϵRI aggregation. This was followed by a slow decrease in levels.
Interaction of PLCγ1with Adaptor Molecules—In activated T cells, PLCγ1 is recruited to the cell membrane in a PI 3-kinase-independent manner via the adaptor molecules LAT, SLP76, and Vav (reviewed in Ref.
). To examine whether the activation of PLCγ1 in HuMCs following FcϵRI aggregation may also involve similar interactions, we examined the ability of the PLCγ1 SH3 domain and dual SH2 domains to bind to these molecules in extracts of activated HuMCs. Cells were stimulated for 120 s, then proteins were extracted under denaturing conditions. Following capture by the GST-SH3 domain and GST-SH2-SH2 fusion proteins, the samples were probed with antibodies to SLP76, Vav, and LAT. We observed no direct binding of PLCγ to LAT via the SH2 or SH3 domains under these conditions (data not shown) but we did observe an inducible binding of both SLP76 and Vav to the SH2 domains of PLCγ1 (Fig. 2a). Thus, in HuMCs, as in T cells, following cell activation, PLCγ1 is capable of directly associating with SLP76 and Vav.
Expression and Activation of PI-3 Kinase—To next investigate the role of PI 3-kinase in the regulation of PLCγ activation leading to degranulation of HuMCs, we first examined the expression of PI 3-kinase in HuMCs. Lysates were probed for the α and β isoforms of the PI 3-kinase p85 subunit, and the α, β, and δ isoforms of the PI 3-kinase p110 subunit utilizing lysates of U937 cells, mouse BMMCs, and RBL 2H3 cells as positive controls (Fig. 3a). In HuMCs, the p85α and p85β and the p110α, -β, and -δ subunits (Fig. 3a) were all expressed. The anti-p85β antibody used in these studies did not recognize rodent p85β. However, expression of p85β in BMMCs has been confirmed utilizing an anti-p85 polyclonal antibody that recognizes both the α and β isoforms (data not shown).
To establish that PI 3-kinase was activated in HuMCs following FcϵRI aggregation, HuMCs were sensitized and then loaded with [32P]orthophosphate prior to triggering as above. The formation of PI phosphates was then examined 2 min after FcϵRI aggregation. At this point, we observed a marked increase in PIP, PIP2 (Fig. 3b), and PIP3 (Fig. 3c) formation. As previously observed in mouse BMMCs (
), PIP3 was a relatively minor product compared with PIP and PIP2. These responses were blocked by preincubation of the cells with the PI 3-kinase inhibitor wortmannin, thus confirming PI 3-kinase activity following FcϵRI aggregation in HuMCs. To further support this conclusion, control and activated HuMC lysates were probed with antibodies against the phosphorylated forms of AKT and p70 S6 kinase, which have been utilized as markers for PI 3-kinase activity (
), there was a marked increase in the phosphorylation of both AKT and p70 S6 kinase (Fig. 3, d and e). From Fig. 3d, it can be seen that wortmannin blocked the increase in phosphorylation of both AKT and p70 S6 kinase over a similar concentration range to that previously reported for the effect of wortmannin on PI 3-kinase-dependent responses (
). Taken together, the above data demonstrate that both PLCγ and PI 3-kinase are activated following FcϵRI aggregation in HuMCs.
Relative Kinetics of the Activation of Src Kinases, PLCγ1, and PI 3-Kinase following FcϵRI Aggregation—To determine the sequence of activation events after aggregation, and the relationship between the activation of PI 3-kinase and PLCγ, we next examined the temporal relationships between the FcϵRI-dependent phosphorylation of PLCγ, and phosphorylation of AKT and p70 S6 kinase, and compared these kinetics to those of the activation of src kinases that represents one of the earliest FcϵRI-induced signaling events in mast cells. Following FcϵRI aggregation, the increase in tyrosine phosphorylation of both PLCγ1 and, as expected, src kinase was very rapid, maximizing within 5 to 10 s (Fig. 4a). In contrast, the kinetics of phosphorylation of AKT and p70 S6 kinase, whereas virtually coincidental, was not apparent until ∼30 s after FcϵRI aggregation (Fig. 4b). Thus, PI 3-kinase-dependent responses, as monitored by the phosphorylation of AKT and p70 S6 kinase, were delayed (Fig. 4c), appearing after the activation of PLCγ. These data are consistent with the conclusion that PLCγ activation is coincidental with the activation of src kinases but precedes the activation of PI 3-kinase.
The Effect of Src Kinase, p72syk, and PI 3-Kinase Inhibitors on FcϵRI-mediated Protein Phosphorylation and PLCγ Activation—The above data suggest that PLCγ activation is independent of PI 3-kinase but may be regulated by src kinase(s). To verify this conclusion, inhibitors of src kinases, p72syk, and PI 3-kinase, respectively, PP2, piceatannol, and wortmannin, were next employed.
We first examined the effect of PP2 on the phosphorylation of PLCγ1 and AKT. These responses were completely blocked in a concentration-dependent manner (Fig. 5a) suggesting that the activation of both PLCγ1 and PI 3-kinase is dependent on src kinases. Src kinase(s) are known to regulate signaling events in activated mast cells in part through p72syk (
). To, therefore, determine whether the regulation of PLCγ activation by src kinases required p72syk activation, we next examined the ability of piceatannol to block FcϵRI-dependent PLCγ activation. The phosphorylation of PLCγ1 and AKT observed following FcϵRI aggregation was again completely inhibited in a concentration-dependent manner by piceatannol (Fig. 5b). These data are consistent with the conclusion that both src kinases and p72syk activation precedes PLCγ phosphorylation.
In addition to tyrosine phosphorylation, PLCγ activation also requires membrane translocation. We, thus, next examined the relative abilities of PP2 and piceatannol to prevent the FcϵRI-dependent membrane translocation of PLCγ and PLCγ-dependent IP3 production. Both PP2 and piceatannol blocked the membrane translocation of PLCγ (Fig. 5c) and IP3 production (Fig. 5d) following FcϵRI aggregation, indicating that these responses are also dependent on both src kinases and p72syk activation.
Last, and following the data presented in Fig. 3 that shows that PI 3-kinase activation follows PLCγ activation, we examined the effect of wortmannin on PLCγ1 activation. As before, in these experiments wortmannin blocked the phosphorylation of AKT (Fig. 6a) confirming that PI 3-kinase was inhibited. Wortmannin, however, failed to inhibit the phosphorylation of PLCγ1 (Fig. 6a), the membrane translocation of PLCγ1 and PLCγ2 (Fig. 6b), and the production of IP3 (Fig. 6c) induced by FcϵRI aggregation. Taken together, these data demonstrate that the activation of PLCγ1 and PLCγ2 following FcϵRI aggregation is independent of PI 3-kinase activity but regulated by src kinases and p72syk.
The Effect of Wortmannin on FcϵRI-mediated Calcium Mobilization and Degranulation—The PLCγ-dependent increase in IP3 is responsible for the initial increase in cytosolic free calcium levels observed following FcϵRI aggregation in mast cells (
). We thus next examined the effect of wortmannin on intracellular calcium levels in FcϵRI-activated HuMCs. From Fig. 7a, it can be seen that the initial increase in intracellular calcium levels, which correlated with activation of PLCγ (0–120 s), was relatively unaffected by wortmannin, however, the calcium flux at later time points, which occurred after PLCγ activation had already declined (>120 s), was partially sensitive to wortmannin, at least at higher concentrations (100 nm). This agrees with our previous observation on the effect of wortmannin on FcϵRI activation of interferon γ-treated HuMCs (
Taken together, the above data are consistent with PLCγ1 being activated in HuMCs in a PI 3-kinase independent manner (Figs. 3 and 6), although PI 3-kinase is activated temporally following PLCγ activation (Fig. 3) upon FcϵRI aggregation in HuMCs and may play a role in the delayed influx of extracellular calcium. This being the case, if mast cell degranulation following FcϵRI aggregation is solely regulated by PLCγ1, then inhibitors of PI 3-kinase should have no effect on degranulation. To examine this hypothesis, we activated HuMCs via FcϵRI in the presence of wortmannin, which, as described earlier, completely blocked PI 3-kinase-dependent responses (Figs. 2 and 6) and, as previously reported (
), degranulation of RBL 2H3 cells (data not shown) but not PLCγ activation (Fig. 6). As can be seen from Fig. 7b, wortmannin blocked degranulation by ∼50% with an IC50 for this component of ∼3 nm, which is similar to that required to inhibit known effects of PI 3-kinase. However, 50% of the degranulation response was refractory to the effects of wortmannin. This again supports our previous observations on the effects of wortmannin on FcϵRI-mediated activation of interferon γ-treated HuMCs (
). The inhibition of FcϵRI-mediated degranulation tended to be more apparent at later time points (>60 s) than at earlier time points (<60 s). These observations are consistent with the conclusion that the PLCγ1-dependent component of HuMC activation is PI 3-kinase independent. However, the data also suggests that these may be a latent PI 3-kinase dependent component to HuMC activation.
We have shown that the PLCγ1-dependent, PI 3-kinase independent pathway, events require initial activation of src kinases and p72syk (see Fig. 5). To determine whether activation of src kinases and p72syk also precedes activation of the PLCγ1 independent, PI 3-kinase-dependent pathway, we again employed PP2 and piceatannol. As can be seen, from Fig. 7, c and d, both reagents completely inhibited HuMC degranulation in a concentration-dependent manner. This indicates that both pathways are regulated by src kinase(s) and p72syk.
Activation of PLCγ1and Degranulation in Mast Cells Deficient in the p85 Subunit of PI 3-Kinase—To confirm that FcϵRI-mediated PLCγ1 activation was independent of PI 3-kinase we adopted knock-out approaches. Because of technical difficulty in utilizing HuMCs for these studies, we conducted these experiments in BMMCs derived from the bone marrow of mice deficient in the p85α (
) subunits of PI 3-kinase. FcϵRI aggregation in BMMCs derived from wild type mice revealed identical kinetics of AKT phosphorylation (Fig. 8, a and b), PLCγ1 phosphorylation and translocation (Fig. 8, c to f), IP3 production (Fig. 9, a and b), calcium flux (Fig. 9, c and d) and degranulation (Fig 9, e and f) to those observed in activated HuMCs (Figs. 1, 2, 3). As was the case with HuMCs (Fig. 3), PI 3-kinase, as monitored by AKT phosphorylation, was delayed compared with the parameters of PLCγ1 activation (Fig. 8, c and d).
In p85α-/- BMMCs, we observed a substantial reduction in the phosphorylation of AKT following FcϵRI aggregation (Fig. 8a), however, in the p85β-/- BMMCs, little reduction in the AKT phosphorylation was observed (Fig. 8b). In contrast, there was no decrease in PLCγ1 activation as monitored by the phosphorylation (Fig. 8, c and d) and translocation (Fig. 8, e and f) of PLCγ1 and by peak IP3 production (Fig. 9, a and b) in either the p85α-/- BMMCs or, the p85β-/- BMMCs. Similarly, there were no defects in FcϵRI-mediated translocation of PLCγ2 observed in these BMMCs. Finally, maximal FcϵRI-dependent degranulation in the p85α-/- and p85β-/- BMMCs was not different from that observed in wild type BMMCs (Fig. 9, e and f).
In this paper we examined the role of PI 3-kinase in the regulation of PLCγ activation leading to degranulation of mast cells in primary culture following FcϵRI aggregation. Generally, only one of the two isoforms of PLCγ is expressed in a particular cell type of hematopoietic lineage. However, data from mouse BMMCs (
), our studies demonstrated the expression of both PLCγ1 and PLCγ2 isoforms in CD34+-peripheral blood-derived human mast cells (Fig. 1). FcϵRI-dependent activation of PLCγ was demonstrated by the rapid membrane translocation of PLCγ1 and PLCγ2, rapid tyrosine phosphorylation of PLCγ1, and a concomitant increase in IP3 and intracellular calcium. These events appear to be minimally regulated by PI 3-kinase as demonstrated by studies utilizing PI 3-kinase inhibitors and studies conducted in BMMCs derived from knock-out mice.
It should be noted that in mast cells, the FcϵRI-mediated calcium signal is regulated by both PLCγ and sphingosine kinase (
). Indeed, it has been argued that the initial phase of FcϵRI-mediated calcium mobilization in human cord blood-derived mast cells is regulated by a sphingosine kinase-PLD pathway rather than a PLCγ-dependent pathway (
). Thus, we believe that the kinetics of PLCγ1 activation in the CD34+ peripheral blood-derived HuMCS were consistent with PLCγ regulating the immediate rather than a latent calcium signal.
The expression and activation of PI 3-kinase, as monitored by the production of PIP, PIP2, and PIP3 and phosphorylation of AKT and p70 S6 kinase, in FcϵRI-activated HuMCs (Fig. 3) is consistent with a role for this enzyme in HuMC activation. However, when the kinetics of activation of PI 3-kinase were compared with the activation of PLCγ1 and PLCγ2, noticeable differences were observed (Fig. 4). Indices of PLCγ1 and PLCγ2 activation including translocation, phosphorylation, and IP3 production were all immediate responses, maximizing between 5 and 10 s after FcϵRI aggregation. In contrast, indices of PI 3-kinase activation including phosphorylation of AKT and p70 S6 kinase, were delayed, appearing only after the activity of PLCγ was already maximal (Fig. 4). Moreover, activation of PLCγ more closely correlated with the activation of src kinases. These observations led us to conclude that PLCγ activation following FcϵRI aggregation is primarily regulated by src kinases rather than PI 3-kinase.
The conclusion that the FcϵRI-mediated PLCγ activation in HuMCs was independent of PI 3-kinase was supported by the studies utilizing the PI 3-kinase inhibitor wortmannin (
). This compound was observed to potently inhibit PI 3-kinase activity in HuMCs as monitored by the phosphorylation of AKT and p70 S6 kinase. However, wortmannin, had little effect on FcϵRI-dependent phosphorylation and translocation of PLCγ, PLCγ-dependent IP3 production, and initial intracellular calcium mobilization in HuMCs, even at concentrations up to 1000 nm (Figs. 6 and 7). In contrast, the src kinase inhibitor, PP2, completely blocked all FcϵRI-induced responses examined including total tyrosine phosphorylation, PLCγ phosphorylation (Fig. 5), translocation, and activation (Fig. 6). The aforementioned responses were also blocked by the p72syk inhibitor, piceatannol (Fig. 5), demonstrating that the regulation of PLCγ activity by src kinases was mediated via the activation of p72syk.
Further supporting evidence for FcϵRI-dependent activation of PLCγ1 being regulated independently of PI 3-kinase came from studies conducted in BMMCs derived from mice deficient in the α (
) isoforms of the p85 regulatory subunit of PI 3-kinase. In the p85α-/-, but not the p85β-/- BMMCs, we observed a marked reduction in FcϵRI-mediated AKT phosphorylation, compared with that observed in wild type BMMCs, demonstrating a defect in PI 3-kinase activation in these cells (Fig. 8). This contrasts with a previous report where phosphorylation of AKT in response to Kit activation, but not FcϵRI aggregation, was reduced in p85α-/- mast cells derived from fetal liver (
). The reason for this disparity is unclear, but it may reflect phenotypic differences in BMMCs versus fetal liver-derived mast cells. In contrast to AKT phosphorylation, there was no defect in FcϵRI-mediated PLCγ activity in the p85α-/- and p85β-/- BMMCs as monitored by PLCγ1 phosphorylation, PLCγ translocation, and IP3 production (Figs. 8 and 9). Taken together, the data from the PI 3-kinase-defective BMMCs strongly supports the conclusion that in mast cells, FcϵRI-mediated PLCγ1 activation is independent of PI 3-kinase.
In B cells, the phosphorylation and membrane translocation of PLCγ2 following activation of the BCR occurs in a PI 3-kinase-dependent manner (
). In T cells, PI 3-kinase does not appear to be required for TCR-mediated activation of PLCγ1 and this has been explained by the fact that Tec kinases appear to be constitutively associated with the T cell plasma membrane (
). Similarly, we also observed constitutive association of the Tec kinases, Btk, Itk, and Tec, with the plasma membrane in HuMCs, with only Tec showing an increase in translocation at later time points following FcϵRI aggregation (data not shown). The fact that PLCγ1 activation in FcϵRI-activated BMMCs derived from Btk knock-out mice is normal (
), even suggests that PLCγ1 activity is not regulated by Btk.
TCR-mediated PLCγ1 activation is a result of activation of lck and ZAP70 followed by translocation and anchoring at the plasma membrane via the adaptor molecules LAT, GAD(s), and SLP76, in a PI 3-kinase independent manner (
). In the activated HuMCs, we observed that the SH2 domain of PLCγ1 could bind directly SLP76 and Vav in activated HuMCs (Fig. 2). These studies were conducted under denaturing conditions to prevent endogenous indirect protein-protein interactions. Therefore, although we did not observe direct binding to LAT in these studies, as both Vav and SLP76 directly associate with LAT, it is likely that PLCγ can also indirectly bind to the membrane via this molecule. BMMCs derived from mice deficient for Vav1 (
) show a reduction in PLCγ1 and PLCγ2 activation and subsequent inhibition of degranulation following FcϵRI aggregation. Thus, based on the data presented in this article, HuMCs display more of the T cell phenotype for PLCγ activation than the B cell, platelet, and macrophage phenotype. These data further support the concept that the expression and mode of activation of PLCγ1 and PLCγ2, in hematopoietic cells, is highly dependent upon the particular cell type (
Despite demonstrating that PLCγ activation following FcϵRI aggregation in HuMCs is independent of PI 3-kinase, we observed that wortmannin inhibited degranulation following FcϵRI aggregation in the HuMCs by ∼50% (Fig. 8) and the remaining 50% was refractory to the effects of even high concentrations of wortmannin. The IC50 for the inhibitory component was very similar to that previously described for wortmannin on PI 3-kinase-mediated responses in RBL 2H3 cells (
). In addition, we observed that, although the initial PLCγ-dependent increase in intracellular calcium in HuMCs was refractory to wortmannin, the later phase, which is largely because of influx of extracellular calcium, was also sensitive to this inhibitor. However, the effect of wortmannin on calcium mobilization was only observed at higher concentrations (100 nm). This observation is similar to that previously described in other systems where, the effects of wortmannin on calcium mobilization was demonstrated to be because of inhibition of PI 4-kinase that regulates the intracellular pools of inositol phospholipids (
). As we observed no defect of FcϵRI-mediated calcium flux in the p85 KO BMMCs, it is possible that the effect of wortmannin on the latent FcϵRI-mediated calcium response is also independent of inhibition of PI 3-kinase.
Finally, recent studies in mouse BMMCs have revealed that two complementary pathways contribute to FcϵRI-mediated mast cell degranulation (
). One of these pathways requires the activation of p72syk and the src kinase p56lyn and leads to PLCγ activation with a resulting increase in intracellular calcium concentrations. The adaptor molecules Vav, SLP76, and LAT are essential for PLCγ-dependent degranulation by this pathway. The second pathway is regulated by the src kinase fyn, which phosphorylates the adaptor molecule Gab2 in turn recruiting other signaling molecules. Our data obtained in HuMCs certainly supports the conclusion from this study that two alternative pathways contribute to FcϵRI-mediated mast cell activation. These observations and the ability of wortmannin to block other PI kinases at higher concentrations may help to explain conflicting data regarding the role of PI 3-kinase in the activation of mast cells. In this context, wortmannin has been reported to inhibit PLCγ activation (
). The degree to which wortmannin inhibits cell activation in these studies may, therefore, be dependent upon the degree to which the alternative signaling pathways, described above, are activated and utilized for subsequent mediator release.
In conclusion, the data presented in this study suggest that the PLCγ-mediated component of degranulation in FcϵRI-activated HuMCs is independent of PI 3-kinase, thus these cells display more of the T cell model for signaling than the model proposed for B cells, macrophages, and platelets (
). Furthermore, our data supports the concept of two complementary pathways contributing to the regulation of degranulation of mast cells following FcϵRI aggregation: a PLCγ-dependent, PI 3-kinase-independent pathway that is insensitive to wortmannin, and a wortmannin-sensitive pathway that may or may not be dependent on PI 3-kinase.