Thapsigargin-Induced Degranulation of Mast Cells Is Dependent on Transient Activation of Phosphatidylinositol-3 Kinase¹

Mast cells play a crucial role in the initiation of allergic responses. Upon exposure to multivalent allergens, they are activated and rapidly secrete preformed mediators, such as histamine, from cytoplasmic granules. This process is initiated by allergens that cross-link IgE molecules that bind with high affinity to the FcεR1 on the surface of mast cells (1, 2). Recently, it has been shown that mast cell degranulation is responsible for the coating of nonimmunogenic surgically implanted biomaterials (3). Specifically, release of histamine at the site of the implant appears to act as a chemoattractant for phagocytic cells that encapsulate the implant and can lead to its failure (3). It is therefore important to elucidate the intracellular pathways that regulate mast cell degranulation. Studies to date indicate that this degranulation process is strictly dependent on the influx of extracellular calcium. The depletion of extracellular calcium by the chelator EGTA results in the complete inhibition of Ag-induced degranulation (4). In general, receptors that are capable of stimulating the release of intracellular calcium induce the tyrosine phosphorylation of phospholipase C-γ1 (PLC-γ1)⁴ and PLC-γ2 (5), resulting in the generation of the two second messengers inositol-1,4,5-trisphosphate (IP₃) and diacylglycerol (DAG) (6). IP₃ binds to its receptor in the membrane of the endoplasmic reticulum and induces the release of intracellular calcium, whereas DAG associates with certain isoforms of the serine/threonine protein kinase C (PKC), thereby promoting their activation (6). The IP₃-induced emptying of intracellular calcium stores then triggers the entry of extracellular calcium through store-operated calcium channels in the plasma membrane (7).

We recently generated a mouse containing a targeted disruption of the gene encoding the hemopoietic specific src homology 2-containing inositol phosphatase, SHIP. Bone marrow-derived mast cells (BMMCs) from these mice were found to readily degranulate with either IgE (4) or steel factor (SF) (8), two proteins that do not by themselves stimulate degranulation in normal murine BMMCs. This hyperresponsiveness was due in large part to markedly elevated levels of PI-3 kinase-generated phosphatidylinositol-3,4,5-trisphosphate (PIP₃), because SHIP was not present to hydrolyze PIP₃ to PI-3,4-P₂. Specifically, in SF-induced degranulation of SHIP^−/− BMMCs, we found that PI-3 kinase-generated PIP₃ was critical both for a step upstream of intracellular calcium release and between intracellular calcium release and extracellular calcium entry (8). Our working hypothesis based on these results was that the markedly increased, membrane-anchored PIP₃ in SHIP^−/− BMMCs attracted substantially more pleckstrin homology (PH) domain-containing proteins, such as PLC-γ (9) and Btk (10), to the plasma membrane to mediate these calcium fluxes. Thus, SHIP appears to function in normal BMMCs to restrict the entry of extracellular calcium by reducing the level of PI-3 kinase-generated PIP₃ (4, 8).

One approach to studying the regulation of mast cell degranulation downstream of extracellular calcium entry is by bypassing the activation of plasma membrane receptors by using calcium ionophores that transport calcium ions across the plasma membrane or via chemicals that induce the release of calcium from intracellular stores. In this study, the effect of the tumor promoter thapsigargin on mast cell degranulation was evaluated in SHIP^+/+ and SHIP^−/− BMMCs. Thapsigargin is a specific inhibitor of the sarcoplasmic/endoplasmic reticulum calcium-dependent ATPase, which pumps calcium that leaks from the endoplasmic reticulum back into this organelle (11). Adding thapsigargin to mast cells thus results in the draining of calcium ions from the endoplasmic reticulum, capacitative entry of extracellular calcium through store-operated calcium surface channels, and subsequent degranulation (12).

The results presented in this work demonstrate that although thapsigargin induces degranulation in a cell surface receptor-independent fashion, this process is still dependent on the activation of PI-3 kinase in normal BMMCs and is strongly enhanced in SHIP^−/− BMMCs. In fact, the independence of thapsigargin-mediated degranulation from surface receptors as well as PLC-γ enabled us to identify an additional PI-3 kinase-dependent step within the mast cell degranulation process downstream from intracellular calcium release and influx of extracellular calcium. We also show that thapsigargin stimulation of BMMCs leads to the activation of protein kinase B (PKB) and survival of cells in the absence of cytokines, thereby providing a possible mechanism for the tumor-promoting ability of this molecule.

Bone marrow cells from 4- to 8-wk-old SHIP^+/+ and SHIP^−/− littermates were plated in methylcellulose (Methocult M3234; StemCell Technologies, Vancouver, Canada) supplemented with 30 ng/ml murine IL-3, 50 ng/ml murine SF, and 10 ng/ml human IL-6 for 10–14 days. They were then harvested and grown in suspension in IMDM containing 15% FCS (StemCell Technologies), 150 μM monothioglycerol (Sigma, Oakville, Canada), and 30 ng/ml IL-3. By 8 wk in culture, greater than 99% of the cells were c-kit and FcεR1 positive, as assessed by FITC-labeled anti-c-kit Abs (PharMingen, Mississauga, Canada) and FITC-labeled IgE (anti-erythropoietin) (26), respectively (4, 8).

Polyclonal Abs against S473-phosphorylated PKB (P-PKB), PKB, and T202/Y204-phosphorylated MAPK (P-MAPK) were obtained from New England Biolabs (Mississauga, Canada). The polyclonal anti-ERK-1 Ab was purchased from Upstate Biotechnology (Lake Placid, NY). Compound 3 (bisindolylmaleimide) (catalog no. 203290), thapsigargin (catalog 586005), and LY294002 (catalog no. 440202) were obtained from Calbiochem (San Diego, CA). Wortmannin (catalog no. W1628) and PMA (catalog no. P8139) were purchased from Sigma.

Calcium fluxes were measured according to Huber et al. (8). In brief, SHIP^+/+ and SHIP^−/− BMMCs were incubated with 2 μM fura 2-acetoxymethyl ester (Molecular Probes, Eugene, OR) in Tyrode’s buffer (130 mM NaCl, 5 mM KCl, 1.4 mM CaCl₂, 1 mM MgCl₂, 5.6 mM glucose, and 0.1% BSA in 10 mM HEPES, pH 7.4) at 23°C for 45 min. The cells were then washed, resuspended in 1 ml of the same buffer at 5 × 10⁵ cells/ml in a stirring cuvette. Following stimulation with thapsigargin or SF, cytosolic calcium was measured by monitoring fluorescence intensity at 510 nm, after excitation of the sample with two different wavelengths (340 and 380 nm) using an MC200 monochromator from SLM AMINCO with a 8100 V3.0 software program.

For degranulation studies, 5 × 10⁵ cells/sample were washed with IMDM and starved overnight in IMDM, containing 10% FCS and 150 μM monothioglycerol. The cells were then resuspended in Tyrode’s buffer and treated for 15 min at 37°C with or without 1 μg/ml thapsigargin. The degree of degranulation was determined by measuring release of β-hexosaminidase (13).

SHIP^−/− and SHIP^+/+ BMMCs were starved as above and incubated for 1 min at 37°C with control buffer,1 μg/ml thapsigargin, or 300 ng/ml SF. Plasma membrane-enriched membrane fractions were prepared as described by Miura et al. (14). Briefly, the cells were then pelleted, resuspended at 1.5 × 10⁷ cells/ml in 4°C hypotonic lysis buffer (20 mM Tris-Cl, pH 7.4, 5 mM EDTA, 5 mM EGTA, 5 mM DTT, 5 mM Na₃VO₄, 0.5 mM PMSF, 2 μg/ml leupeptin, and 10 μg/ml aprotinin), allowed to swell for 5 min on ice, and sonicated for 15 × 1-s bursts on ice using an ultrasonic cell disruptor (Heat Systems Ultrasonics, Faimingdale, NY). After centrifugation at 2000 × g for 5 min at 4°C, the supernatant was centrifuged at 100,000 × g for 10 min at 4°C in an airfuge (Beckman Instruments, Fullerton, CA). The pellet was resuspended in 400 μl of hypotonic lysis buffer containing 1% Nonidet P-40 by repeated vortex mixing. After a 60-min incubation at 4°C, the suspension was centrifuged at 100,000 × g for 10 min, and the supernatant was collected as the membrane fraction. This solubilized membrane fraction was then subjected either to Western blot analysis with anti-p85 (Upstate Biotechnology), and the blot reprobed with anti-c-kit as a loading control or to immunoprecipitation by first incubating at 4°C for 1 h with anti-p85 Ab (Upstate Biotechnology) and then with protein A-Sepharose beads at 4°C for 1 h. Beads were then washed and PI-3 kinase assays were performed, as described previously (15).

SHIP^+/+ and SHIP^−/− BMMCs were starved as above and incubated for various times at 37°C with 1 μg/ml thapsigargin. The cells were then solubilized with 0.5% Nonidet P-40 in 4°C phosphorylation solubilization buffer (16), and subjected to Western blot analysis, as described previously (16).

SHIP^+/+ and SHIP^−/− BMMCs were washed with IMDM and incubated at 5 × 10⁵ cells/ml in IMDM containing 10% FCS and vehicle (DMSO) or various concentrations (0.015–0.06 μg/ml) of thapsigargin at 0.4 ml/well in Falcon 3047 24-well flat-bottom plates. Viability was assessed by trypan blue exclusion.

We recently demonstrated that both IgE and SF stimulate a much more profound calcium influx into SHIP^−/− than SHIP^+/+ BMMCs, and this leads to a substantially greater degranulation of the SHIP^−/− cells (4, 8). Moreover, the enhanced calcium entry and degranulation of the SHIP^−/− BMMCs by these two agonists were shown to be dependent on elevated PIP₃ levels in these cells (8). This is consistent with our finding that SHIP is the primary enzyme responsible for hydrolyzing SF-induced PIP₃ in these BMMCs and that PIP₃ reaches substantially higher levels in response to SF or IgE when SHIP is absent (8, 17). To gain some insight into the regulation of the degranulation process downstream of calcium release from intracellular stores, we investigated the degranulation potential of thapsigargin in SHIP^+/+ and SHIP^−/− BMMCs. We found that while SHIP^+/+ BMMCs only degranulated to ∼10% in response to this agent, stimulation of SHIP^−/− BMMCs resulted in about 55% degranulation (Fig. 1,A). Based on our previous findings with IgE- and SF-stimulated BMMCs (4, 8), we anticipated an increased thapsigargin-induced extracellular calcium influx in SHIP^−/− BMMCs. However, the calcium entry induced by thapsigargin in SHIP^+/+ and SHIP^−/− BMMCs was identical (Fig. 1,B). For comparison, a SF-induced calcium influx was conducted with the same cells and, as reported previously (8), SF stimulated a far greater influx of calcium into SHIP^−/− than into SHIP^+/+ BMMCs (Fig. 1 B). These results suggested that there was at least one additional, calcium-independent pathway present in thapsigargin-stimulated mast cells that contributes to the enhanced degranulation in SHIP^−/− BMMCs.

Since one characteristic of SHIP^−/− BMMCs is their augmented, PI-3 kinase-generated PIP₃ levels in response to various stimuli (8, 17), we asked whether PI-3 kinase activation might be involved in the thapsigargin-induced degranulation process. To address this, SHIP^+/+ and SHIP^−/− BMMCs were preincubated with various concentrations of the PI-3 kinase inhibitors, LY294002 and wortmannin, and thapsigargin-stimulated degranulation was assessed. As shown in Fig. 2,A, thapsigargin-induced degranulation of SHIP^+/+ BMMCs was significantly inhibited by LY294002, with 10 μM LY294002 giving a maximal reduction of ∼75% (Fig. 2,A, left panel). In SHIP^−/− BMMCs, thapsigargin-induced degranulation was even more inhibited by LY294002. However, higher concentrations of this PI-3 kinase inhibitor were required to achieve maximal inhibition, i.e., 97% at 100 μM LY294002 (Fig. 2,A, right panel). Similar results were obtained with wortmannin (Fig. 2,B). Both LY294002 and wortmannin reduced the thapsigargin-induced degranulation of the two cell types to ∼3%, suggesting that this residual degranulation may be PI-3 kinase independent. These results suggested that thapsigargin was capable of stimulating PI-3 kinase activity in both SHIP^+/+ and SHIP^−/− BMMCs, and that this activation plays a critical role within the thapsigargin-mediated degranulation process. Importantly, as was found previously with IgE- or SF-stimulated SHIP^+/+ and SHIP^−/− BMMCs (4, 8), thapsigargin-induced PI-3 kinase activity in SHIP^+/+ and SHIP^−/− BMMCs was similar, as assessed both by the level of p85 associated with plasma membrane-enriched membrane preparations (Fig. 3,A) and by PI-3 kinase assays with p85 immunoprecipitates from these membrane preparations (Fig. 3,B). Densitometric analysis of our Western blots (Fig. 3,A) revealed that unstimulated levels of p85 were similar in the two cell types and that thapsigargin recruited p85 to a similar degree in SHIP^+/+ and SHIP^−/− cells (∼1.4-fold over unstimulated levels). Moreover, this p85 recruitment was substantially less than that recruited by SF (∼2.9-fold over unstimulated levels), consistent with the greater effect of SF on PKB activation in SHIP^+/+ BMMCs (see below, Fig. 4). Because of the modest PI-3 kinase activation induced by thapsigargin and the many washing steps involved in the PI-3 kinase assay (15), we obtained some variation in our assay results (e.g., in the top panel of Fig. 3 B, there appears to be more PI-3 kinase activity in thapsigargin-stimulated SHIP^+/+ cells, while in the bottom panel the activation levels look similar in the two cell types). Averaging the densitometric results of five separate experiments revealed no significant difference in thapsigargin-induced PI-3 kinase activity. We also conducted PI-3 kinase assays with anti-phosphotyrosine (4G10) (four separate experiments) and anti-p85 (Upstate Biotechnology) (four separate experiments) immunoprecipitates from whole cell lysates and obtained a similar degree of variation and no significant difference in the SHIP^+/+ and SHIP^−/− BMMCs (data not shown). Thus, our finding that higher concentrations of LY294002 and wortmannin are required to inhibit thapsigargin-induced degranulation in SHIP^−/− BMMCs is consistent with there being higher levels of PIP₃ in SHIP^−/− BMMCs following thapsigargin exposure due to the loss of SHIP rather than higher PI-3 kinase activity.

Since we had shown previously that PI-3 kinase activation was essential for SF-mediated calcium mobilization (8), we investigated whether PI-3 kinase inhibition reduced thapsigargin-induced calcium influx. As shown in Fig. 3,C (left panel), preincubation of SHIP^−/− BMMCs with 50 μM LY294002 had no effect on the initiation of calcium mobilization by thapsigargin and only a minor effect on the overall intracellular calcium level after stimulation. The same concentration of this inhibitor, however, was capable of almost completely abolishing SF-stimulated calcium entry into the same cells (Fig. 3 C, right panel). The same result was obtained in SHIP^+/+ BMMCs (data not shown). These data suggested that in thapsigargin-treated mast cells, PI-3 kinase plays a role in mediating the degranulation process. However, unlike in SF-stimulated mast cells, this enzyme appears to be acting downstream of calcium entry into thapsigargin-treated cells.

One of the major targets of PI-3 kinase activation is PKB (18, 19). Because optimal thapsigargin-induced degranulation was dependent on PI-3 kinase (Fig. 2), we asked whether thapsigargin stimulated PKB activation. To study this, SHIP^+/+ BMMCs were stimulated with thapsigargin for different times, and the activation of PKB was assessed in total cell lysates using a phospho-specific Ab recognizing PKB phosphorylated at Ser⁴⁷³ (P-PKB). Intriguingly, a very transient phosphorylation of PKB was detected only at 1 and 2 min after stimulation (Fig. 4,A, upper panel), a PKB activation pattern totally distinct from that elicited by SF with these cells (Fig. 4,B, upper panel). To verify equal loading, the blots were reprobed with anti-PKB Abs (Fig. 4, A and B, lower panels). Because thapsigargin has been shown previously to activate MAPK in the rat hippocampal cell line, H19-7 (20), we asked whether MAPK was also activated by thapsigargin in murine BMMCs and whether this activation followed the same transient pattern. Total cell lysates from thapsigargin-treated SHIP^+/+ BMMCs were assessed by Western blot analysis with a phospho-specific Ab recognizing doubly phosphorylated (Thr²⁰²/Tyr²⁰⁴) ERK-1 and ERK-2 (P-ERK-1 and P-ERK-2). Interestingly, a sustained MAPK phosphorylation/activation was observed in thapsigargin-stimulated SHIP^+/+ BMMCs (Fig. 4,C), paralleling the prolonged calcium mobilization induced by this drug (Fig. 1,B, left panel). To verify equal loading, the blot was reprobed with anti-ERK-1 Abs (Fig. 4 C, lower panel).

Recently, Yano et al. identified a calcium-triggered signaling cascade in which calcium/calmodulin-dependent kinase kinase activates PKB in a PI-3 kinase-independent fashion (21). We therefore investigated whether the transient PKB activation observed in BMMCs in response to thapsigargin was dependent on PI-3 kinase activation. Specifically, SHIP^+/+ BMMCs were stimulated with 1 μg/ml thapsigargin for 1 or 5 min in the presence or absence of the two PI-3 kinase inhibitors, LY294002 (50 μM) and wortmannin (50 nM), and cell lysates were examined for P-PKB expression. As shown in Fig. 5 (upper panel), both inhibitors were capable of inhibiting thapsigargin-induced PKB phosphorylation, indicating PI-3 kinase dependence. Equal loading was verified by reprobing with anti-PKB Abs (Fig. 5, lower panel). Interestingly, thapsigargin stimulation in the presence of the PI-3 kinase inhibitors had no effect on the activation of MAPK (data not shown).

In previous reports, we demonstrated that SHIP deficiency in murine BMMCs results in the potentiation of PI-3 kinase-mediated effects, such as IgE- and SF-stimulated calcium mobilization and degranulation (4, 8). Since we have shown in this study that PI-3 kinase-dependent PKB activation takes place in SHIP^+/+ BMMCs in response to thapsigargin (Figs. 4,A and 5), we asked whether thapsigargin treatment of SHIP^−/− BMMCs might result in a more pronounced activation of PKB. To investigate this, SHIP^+/+ and SHIP^−/− BMMCs were stimulated with 1 μg/ml thapsigargin for various times and cell lysates subjected to Western blot analysis using anti-P-PKB Abs. As shown in Fig. 6,A, PKB phosphorylation in SHIP^−/− BMMCs was easily detected after a 5-s exposure (upper panel), whereas 3 min of exposure was required to barely see phosphorylation of PKB in SHIP^+/+ BMMCs (middle panel). This demonstrates that PKB is much more activated in SHIP^−/− BMMCs in response to thapsigargin. Interestingly, even in the absence of thapsigargin, the level of PKB phosphorylation was significantly higher in SHIP^−/− BMMCs (see Fig. 6,A, middle panel), most likely because the cells were starved in the presence of 10% FCS, which contains substantial levels of SF (22). Also of interest, PKB phosphorylation was lower following 30 and 60 min of thapsigargin treatment than in unstimulated SHIP^−/− BMMCs (see Fig. 6,A, middle panel), and this could be due to activation of the serine/threonine phosphatase, PP2A, which has been implicated in the in vivo dephosphorylation/inactivation of PKB (23). Equal loading of the gels shown in the upper and middle panels of Fig. 6,A was confirmed by reprobing with anti-PKB Abs (Fig. 6,A, lower panel). Thapsigargin stimulation in the presence of the PI-3 kinase inhibitor LY294002 resulted in the inhibition of PKB phosphorylation in SHIP^+/+ (Fig. 5) as well as in SHIP^−/− BMMCs (data not shown). To determine whether all signaling pathways were activated more strongly in SHIP^−/− BMMCs, the same cell lysates were analyzed by Western blotting with anti-P-MAPK Abs. As shown in Fig. 6,B (upper panel), phosphorylation of the MAPKs, ERK-1 and ERK-2, was the same in the two cell types, consistent with the notion that the loss of SHIP primarily enhances PIP₃-mediated pathways. The membrane was reprobed with anti-ERK-1 Abs to demonstrate equal loading (Fig. 6 B, lower panel).

Having established, using thapsigargin as a probe, that a PI-3 kinase-regulated step in degranulation is present downstream of extracellular calcium entry, we asked what PIP₃-binding protein(s) could be mediating this process. Recently, certain conventional (PKC α and βII), novel (PKC δ), and atypical (PKC ζ) PKC isotypes have been reported to be regulated by PI-3 kinase via the downstream 3-phosphoinositide-dependent protein kinase, PDK-1, which binds PIP₃ in the plasma membrane (24, 25, 26, 27, 28). Because PDK-1 is involved in the phosphorylation/activation of PKB (26, 29) and because we found that thapsigargin increased PKB phosphorylation more in SHIP^−/− than in SHIP^+/+ BMMCs, we investigated whether a pan-specific PKC inhibitor, compound 3, could inhibit thapsigargin-induced degranulation. As shown in Fig. 7,A, compound 3 markedly inhibited thapsigargin-induced degranulation in both SHIP^+/+ and SHIP^−/− BMMCs. To gain some insight into which PKC might be involved in this process, we assessed degranulation in the presence of the phorbol ester, PMA, which specifically activates DAG-dependent PKC isoforms. Although 50 nM PMA had no statistically significant effect on degranulation by itself in SHIP^+/+ or in SHIP^−/− BMMCs, it slightly enhanced thapsigargin-induced degranulation (Fig. 7 B). It did not, however, increase thapsigargin-induced degranulation in SHIP^+/+ cells to the same level observed in SHIP^−/− BMMCs. This suggests that a PDK-1-dependent DAG-independent PKC, such as PKC ζ (28), might be important in this step.

Because we found that thapsigargin stimulated PKB, a kinase shown to enhance survival of many cell types (29), we asked whether thapsigargin might promote the survival of our SHIP^+/+ and SHIP^−/− BMMCs following withdrawal of IL-3. As can be seen in Fig. 8,A, thapsigargin at concentrations between 0.02 and 0.06 μg/ml dramatically enhanced the survival of both SHIP^+/+ and SHIP^−/− BMMCs in the absence of IL-3. Also, as predicted based on the greater activation of PKB in SHIP^−/− BMMCs, thapsigargin was more effective at promoting survival of SHIP^−/− BMMCs at low concentrations of the tumor promoter. This is also shown in a time course study with SHIP^+/+ and SHIP^−/− BMMCs using 0.02 μg/ml thapsigargin (Fig. 8 B).

The present study was aimed at evaluating the effect of thapsigargin, a non-phorbol ester-type tumor promoter originally isolated from the plant Thapsia garganica L. (Apiaceae) (11), on mast cell degranulation using SHIP^+/+ and SHIP^−/− BMMCs. Interestingly, as was found previously with IgE and SF (4, 8), thapsigargin induced a more profound degranulation in SHIP^−/− BMMCs (Fig. 1,A). However, unlike IgE and SF, this thaspigargin-induced increased degranulation did not correlate with an elevated calcium influx into SHIP^−/− BMMCs (Fig. 1,B), suggesting that calcium entry is a crucial trigger, but not the only factor responsible for degranulation. Consistent with this observation, Ludowyke et al. (30) reported that the calcium ionophore, A23187, caused degranulation of the rat mast cell line RBL-2H3, but far less than that elicited with IgE plus cross-linking Ag. However, they found that A23187 plus the phorbol ester PMA (which did not stimulate degranulation by itself) induced degranulation to the same extent as IgE plus cross-linking Ag (30) and proposed that in addition to calcium, other signals have to be generated for optimal degranulation. Related to this, we found in the present study that LY294002 and wortmannin inhibited thapsigargin-induced degranulation in both SHIP^+/+ and SHIP^−/− BMMCs (Fig. 2), suggesting that activation of PI-3 kinase by thapsigargin provides a signal in addition to calcium mobilization to attain maximal degranulation. This is consistent with a report by Marquardt et al. (31), who showed that A23187-induced degranulation of BMMCs could be attenuated using the PI-3 kinase inhibitor, wortmannin. Intriguingly, we found that LY294002 had no significant effect on the calcium influx induced by thapsigargin (Fig. 3 B), suggesting a PI-3 kinase-dependent step downstream of calcium mobilization. This downstream step could not be detected with PI-3 kinase inhibitors in our earlier studies employing IgE or SF to trigger mast cell degranulation because the degranulation pathway initiated by these two agonists was blocked by PI-3 kinase inhibitors at two much earlier steps (intracellular calcium release and extracellular calcium entry) (4, 8). Thus, thapsigargin has proven extremely useful, by bypassing PLC-γ-induced calcium release and PI-3 kinase-mediated extracellular calcium entry, in revealing this third PI-3 kinase-regulated step in degranulation.

PI-3 kinase phosphorylates PI-4,5-P₂ at the 3′ position of the inositol ring to generate PIP₃, which then serves as a substrate for SHIP, yielding PI-3,4-P₂ (32). We demonstrated recently that the loss of SHIP enhances PI-3 kinase-induced cellular responses by elevating PIP₃ levels (4, 8). Because several signaling proteins containing PH domains, such as PLC-γ (9), the tyrosine kinase Btk (10), and the serine/threonine kinase PKB (33, 34), are capable of binding to and becoming activated by PI-3 kinase-generated phosphoinositides, PI-3 kinase is an important switch for the initiation of various pathways. Related to this, we observed transient PI-3 kinase-mediated activation of PKB in response to thapsigargin. This is especially interesting given that PKB activation requires colocalization of the PH-containing kinase PDK-1 (which phosphorylates PKB at Thr³⁰⁸) at the plasma membrane. PDK-1 has been shown to phosphorylate/activate various PKC isoforms (24, 25, 26, 27, 28). This, coupled with our data showing that the pan-specific PKC inhibitor, compound 3, prevents thapsigargin-induced degranulation and that PMA doesn’t bring thapsigargin-induced degranulation in SHIP^+/+ BMMCs to levels obtained with SHIP^−/− BMMCs suggests that a PIP₃-binding DAG-independent PKC isotype, such as PKC ζ, might be connecting the activation of PI-3 kinase with the degranulation machinery (Fig. 7). Consistent with this model, Cissel et al. (35) have demonstrated in the rat RBL-2H3 mast cell line, using a variety of pharmacological activators and inhibitors of signaling molecules, that thapsigargin-induced degranulation is regulated by both phospholipase D and PKC in a PI-3 kinase-dependent manner. As well, it has been known for some time that activation of PKC is a critical event for effective degranulation to occur (30). Moreover, as mentioned earlier, the combination of calcium-mobilizing probes and pharmacological PKC activators has been shown to lead to a synergistic increase in mast cell degranulation (30).

With respect to thapsigargin being a tumor promoter, the activation of PKB and the enhanced survival in the absence of exogenous cytokines offer an interesting new possibility for the tumorigenicity of thapsigargin. PKB is known to be a key survival kinase required for the inhibition of apoptosis in both hemopoietic cells and other cell types (36, 37). Seven targets of PKB have been identified to date, and they are the Bcl-2 family member, Bad (38), glycogen synthase kinase-3 (39), caspase-9 (40), a forkhead transcription factor (FKHRL1) (41), Iκκα (42), endothelial NO synthase (eNOS) (43, 44), and Raf (45). Phosphorylation of these proteins by PKB inactivates them, thus promoting survival in certain cell types. Because we also found that thapsigargin activates MAPK (Fig. 4), which has been shown to promote survival in some cell types and proliferation in others (46, 47), the initiation of antiapoptotic (and/or cell proliferation) pathways might be the main mechanism(s) by which thapsigargin mediates its tumor-promoting activity. However, it is worth noting that thapsigargin has been reported to activate both a tyrosine as well as a serine/threonine kinase, leading to transcriptional activation of the glucose-regulated protein GRP78 promoter (48). Consistent with this, Chao et al. (20, 49, 50) have reported the activation of Src tyrosine kinase as well as Raf-1 and MAPK serine/threonine kinases in response to thapsigargin in H19-7 cells. Thus, we cannot rule out at this time that activation of a tyrosine kinase such as Src is also involved in the tumor-promoting activity of thapsigargin. In contrast to our studies, Conus et al., who assessed the role of calcium in the regulation of PKB and p70^S6K in BALB/c-3T3 fibroblasts, found that thapsigargin stimulated full activation of p70^S6K, whereas little or no activation of PKB was observed (51). However, because a different source of cells was used and the time point assessed was 5 min (51), a time in which PKB activation is almost back to baseline in BMMCs (Figs. 4 and 5), their results do not necessarily contradict ours.

In summary, we have shown that the tumor promoter thapsigargin activates a PI-3 kinase-dependent survival pathway, thereby providing a new model for its tumor promotion. Moreover, we have identified a PI-3 kinase-dependent pathway important for primary mast cell degranulation that is downstream of intracellular calcium release and extracellular calcium entry. This pathway might involve PDK-1 and a DAG-independent PKC isoform. Further studies are currently underway to identify the PKC isoform(s) involved in this pathway.

We thank Vivian Lam for excellent technical support and Christine Kelly for typing the manuscript.