Inhibition of Anti-IgM-induced Translocation of Protein Kinase C βI Inhibits ERK2 Activation and Increases Apoptosis*

Expression of the COOH-terminal residues 179–330 of the LSP1 protein in the LSP1+ B-cell line W10 increases anti-IgM- or ionomycin-induced apoptosis, suggesting that expression of this LSP1 truncate (B-LSP1) interferes with a Ca2+-dependent step in anti-IgM signaling. Here we show that inhibition of Ca2+-dependent conventional protein kinase C (cPKC) isoforms with Gö6976 increases anti-IgM-induced apoptosis of W10 cells and that expression of B-LSP1 inhibits translocation of PKCβI but not of PKCβII or PKCα to the plasma membrane. The increased anti-IgM-induced apoptosis is partially reversed by overexpression of PKCβI. This shows that the B-LSP1-mediated inhibition of PKCβI leads to increased anti-IgM-induced apoptosis. Expression of constitutively active PKCβI protein in W10 cells activates the mitogen-activated protein kinase ERK2, whereas expression of B-LSP1 inhibits anti-IgM-induced activation of ERK2, suggesting that anti-IgM-activated PKCβI is involved in the activation of ERK2 and that inhibition of ERK2 activation contributes to the increased anti-IgM-induced apoptosis. Pull-down assays show that LSP1 interacts with PKCβI but not with PKCβII or PKCα in W10 cell lysates, while in vitro LSP1 and B-LSP1 bind directly to PKCβI. Thus, B-LSP1 is a unique reagent that binds PKCβI and inhibits anti-IgM-induced PKCβI translocation, leading to inhibition of ERK2 activation and increased apoptosis.

Many mouse and human B-lymphoma cell lines are susceptible to anti-IgM-induced apoptosis (1)(2)(3)(4)(5)(6). Multiple mIgM 1coupled signal transduction pathways such as increased [Ca 2ϩ ] i , and production of ceramides and reactive oxygen species mediate the apoptotic effect of mIgM stimulation (7)(8)(9). However, anti-IgM treatment also activates potentially antiapoptotic signaling pathways such as activation of phosphatidylinositol 3-kinase and its downstream target Akt/PKB or activation of PKC (10 -12). Thus, the outcome of anti-IgM signaling depends on a balance of pro-apoptotic and anti-apoptotic signals.
Activation and translocation of PKC by phorbol ester protects normal immature and mature mouse B-lymphocytes, the mouse B-lymphoma cell line WEHI-231, human B-lymphoma cell lines and human B-CLL cells from anti-IgM-induced apoptosis (1,4,13,14). The precise PKC isoform involved in protection from anti-IgM-induced apoptosis is not yet known. Evidence for a role of one or more of the Ca 2ϩ -dependent cPKC isoforms ␣, ␤I, ␤II, or ␥ in this protection comes from experiments showing that the protective action of phorbol esters on immature B-lymphocytes (13) is abrogated by a cPKC-specific inhibitor. This inhibitor also renders mature Blymphocytes susceptible to anti-IgM-induced apoptosis (13), suggesting that anti-IgM-induced activation of cPKC isoforms plays an important role in regulating susceptibility of different B-lymphocyte lineage cells to anti-IgM-induced apoptosis.
The mouse leukocyte-specific protein 1 (LSP1) is a 330-amino acid residue intracellular protein expressed in B-and T-lymphocytes and in macrophages and neutrophils (15)(16)(17)(18)(19). Transfection experiments using the LSP1 ϩ B-lymphoma cell line WEHI-231/89 or a single cell subclone, W10, showed that expression of an LSP1 truncate containing residues 179 -330 (designated B-LSP1) significantly increased the extent of apoptosis induced by anti-IgM or by ionomycin but not by sorbitol, nocodazole, C 2 -ceramide, or H 2 O 2 (20). Expression of B-LSP1 had no effect on anti-IgM-induced growth arrest. Consistent with a role of LSP1 in the early induction phase of apoptosis, we found that, after anti-IgM treatment, the number of cells showing loss of mitochondrial membrane potential (⌬⌿ m ) increased faster in the B-LSP1 transfectant than in the parental cells (20). From these experiments we concluded that LSP1 regulates an early step in the induction of anti-IgM-mediated apoptosis, downstream of the anti-IgM-induced increase in [Ca 2ϩ ] i , but upstream of the loss of ⌬⌿ m . Given the protective role of the Ca 2ϩ -dependent cPKC isoforms in anti-IgM-induced apoptosis, we asked whether the expression of B-LSP1 inhibits anti-IgM-induced translocation of cPKC isoforms and, if so, whether this increases anti-IgM-induced apoptosis by inhibiting the cPKC-regulated activation of the MAP kinase ERK2.

EXPERIMENTAL PROCEDURES
Cell Culture and Apoptosis Measurements-Cells were cultured in RPMI 1640 medium as described (20). The W10 cell line is a single cell subclone of the B-lymphoma cell line WEHI-231/89, and the TW10.1 cell line is a stable G418 resistant transfectant derived from W10 cells expressing the LSP1 truncate B-LSP1 containing LSP1 residues 179 -330 (20). To overexpress PKC␤I, TW10.1 cells were co-transfected with the pcDNA3 vector containing a rat PKC␤I cDNA and with the pBABEpuro vector, mixed in a 2:1 molar ratio. Cells were selected in 1 mg/ml * This work was supported by grants from the National Cancer Institute of Canada with funds from the Canadian Cancer Society and from the Cancer Research Society, Inc. 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.
¶ To whom correspondence should be addressed: Toronto Western Hospital, Rm. 13 1 The abbreviations used are: mIgM, membrane immunoglobulin M; LSP1, leukocyte-specific protein 1; PKC, protein kinase C; cPKC, conventional protein kinase C; GST, glutathione S-transferase; MAP, mitogen-activated protein; MBP, myelin basic protein; HBSS, Hanks' buffered saline solution; FSC, forward scatter; SSC, side scatter. G418 and 0.25 g/ml puromycin. The constitutively active PKC␤I⌬NPS construct encoding rat PKC␤I without the pseudosubstrate containing residues 1-30 was expressed following electroporation of W10 cells and selection in 1 mg/ml G418. Parental and transfected cells were cultured in 24-well plates with or without the addition of 5 g/ml goat antimouse IgM (Sigma-Aldrich, Oakville, Ontario, Canada). Apoptosis was assessed using fluorescence-activated cell sorting analysis to determine the fraction of viable cells identified by their normal forward scatter/ side scatter (FSC/SSC) profile after 72 h in culture or by determination of the fraction of cells containing subdiploid DNA 48 h after the initiation of culture (20,21). Inhibitors were added either alone or 15 min before the addition of anti-IgM.
Cell Fractionation and Western Blotting-For PKC translocation experiments, W10 or TW10.1 cells were washed twice in Hank's buffered saline solution (HBSS) and resuspended at 25 ϫ 10 6 cells/ml in HBSS. Three aliquots of 1 ml in microcentrifuge tubes were then warmed at 37°C for 10 min in a water bath. Goat anti-mouse IgM was added to a final concentration of 50 g/ml to two tubes, and incubation at 37°C was then continued. Cells from the third tube served as the unstimulated controls and were recovered by centrifugation for 30 s in a microcentrifuge, washed once in cold HBSS, and stored on ice. Anti-IgM-stimulated cells were recovered after 5 or 20 min of incubation. To prepare plasma membrane fractions, cells were resuspended for 30 min in cold hypotonic fractionation buffer (Buffer A: 5 mM Tris-HCl, pH 7.4, 5 mM KCl, 1.5 mM MgCl 2 , 0.1 mM EGTA) containing a mixture of protease inhibitors (22,23) and then disrupted by 50 strokes in a small Potter homogenizer. The lysates were cleared by centrifugation for 5 min at 500 ϫ g. Supernatants were layered on top of 1.2 ml of 1.2 M sucrose in buffer A and spun at 10,000 ϫ g for 15 min to remove the majority of mitochondria. The lysates were then centrifuged at 100,000 ϫ g for 60 min. The high speed pellet was designated as the plasma membrane fraction and was solubilized in buffer A, while the supernatant was designated as the cytoplasmic fraction. Protein concentrations were determined according to the Bradford method using reagents from Bio-Rad (Oakville, Ontario, Canada).
Equal amounts of protein were separated on 12.5% SDS-polyacrylamide gels, transferred to polyvinylidene difluoride membranes, and analyzed by using antibodies specific for PKC␣ (Sigma-Aldrich) or PKC␤I or PKC␤II (Santa Cruz Biotechnology Inc., Santa Cruz, CA). Blots were developed with ECL reagent (Amersham Pharmacia Biotech, Oakville, Ontario, Canada) followed by exposure to film or with SuperSignal West Femto reagent (Pierce) followed by imaging in a Bio-Rad Fluor-S max . Film exposures were converted to .tif files using a scanner, and differences in band intensities were quantitated using Quantity One software from Bio-Rad. Each gel contained a set of 2-fold dilutions of a total cell lysate from W10 or TW10.1 cells to construct a standard curve used for quantitation of protein signals. Total lysates were prepared by lysing 10 7 cells directly in 1 ml of Laemmli sample buffer.
Activation of ERK2-Cells (25 ϫ 10 6 cells/ml) were warmed to 37°C for 10 min and then stimulated with 50 g/ml anti-IgM with or without pretreatment with 0.25 M Gö6976 or 25 M PD098059 for 15 min. Cells were harvested at different times after addition of anti-IgM, pelleted for 30 s in a microcentrifuge, lysed in 1ϫ sample buffer, and immersed in boiling water for 3 min. Phosphorylation of ERK2 was then analyzed on polyacrylamide gels with an acrylamide:bisacrylamide ratio of 118:1 (24), followed by Western analysis with anti-ERK2 antibodies (Santa Cruz Biotechnology Inc., catalog no. sc-154).
ERK2 kinase activity was measured in an in vitro immune kinase assay. Cells were harvested as above and resuspended in lysis buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, 0.2 mM NaVO 3 , 1 mM dithiothreitol, 10 g/ml aprotinin, 2 g/ml leupeptin, 1 g/ml pepstatin, and 1 mM phenylmethylsulfonyl fluoride). After incubation on ice for 15 min, the insoluble material was pelleted for 15 min at 13,000 rpm in a microcentrifuge and ERK2 was precipitated from the soluble lysates by addition of 5 g of anti-ERK2 antibody followed by incubation at 4°C. After 1 h, 20 l of protein G-agarose slurry (Pierce) was added and incubation was continued for an additional 30 min. Protein G-agarose beads were recovered by centrifugation, washed three times in lysis buffer, and washed once in kinase buffer (20 mM HEPES, pH 7.2, 5 mM MgCl 2 , 1 mM EGTA). To measure kinase activity, beads were resuspended in 30 l of kinase buffer supplemented with 2 mM sodium vanadate, 5 mM ␤-2-mercaptoethanol, 7 g of myelin basic protein (MBP), and 5 Ci of [␥-32 P]ATP (3000 mCi/mmol) and incubated for 15 min at 30°C. The reaction was stopped by adding 30 l of 3ϫ Laemmli sample buffer prewarmed to 55°C, followed by immersion in boiling water for 3 min. Phosphorylation of MBP was then analyzed on 12.5% SDS-acrylamide gels and quantitated using a phosphoimager (Personal FX, Bio-Rad).
LSP1/PKC Interactions-To probe for LSP1/PKC interactions in cell lysates, 4 ϫ 10 7 W10 cells were lysed in lysis buffer (20 mM Tris, pH 7.5, 0.15 M NaCl, 5 mM EDTA) containing 0.5% Nonidet P-40 and a mixture of protease inhibitors as described (22,23) and incubated with 1-2 g of a GST-LSP1 fusion protein containing the intact LSP1 (residues 1-330). After 1 h of incubation at 4°C, 10 l of a 1:1 slurry of glutathione-Sepharose beads (Sigma-Aldrich) was added and incubated for an additional 30 min at 4°C. The beads were recovered by brief centrifugation in a microcentrifuge and washed five times with 1 ml of lysis buffer ϩ 0.5% Nonidet P-40, and the recovered proteins were analyzed by Western blotting. To determine binding of LSP1 with PKC␤I in vitro, 1 g of GST fusion protein containing LSP1 residues 1-330, 1-178, or 179 -330 was mixed with 50 ng of recombinant human PKC␤I (PanVera Corp., Madison, WI) in 250 l of 1ϫ PKC buffer (20 mM Tris, pH 7.5, 5 mM MgCl 2 ). After incubation for 1 h at 4°C, glutathione-Sepharose beads were added and the incubation was continued for an additional 30 min. Beads were recovered by centrifugation and washed four times in 1 ml of 1ϫ PKC buffer, and the amounts of bound PKC␤I and recovered GST and GST fusion protein were analyzed by Western blotting.

Inhibition of cPKC Isoforms Increases Anti-IgM-induced
Apoptosis-Treatment of mature B-cells with a cPKC-specific inhibitor renders these cells susceptible to anti-IgM-induced apoptosis, suggesting that anti-IgM-activated cPKC plays a role in the protection of B-cells from anti-IgM-induced cell death (13). To determine whether activated cPKC has a similar role in the B-lymphoma cell line W10, we treated these cells with anti-IgM in the presence or absence of Gö6976, a PKC inhibitor that preferentially inhibits the cPKC isoforms ␣, ␤I, and ␤II (25,26). The extent of apoptosis was determined by measuring changes in the FSC/SSC profile of the cells at 72 h (20,21). Treatment of W10 cells with 0.25 M Gö6976 or with 5 g/ml anti-IgM results in only a slight reduction of viable cells; 70.6% of anti-IgM-treated cells and 76.4% of Gö6976-treated cells display a normal FSC/SSC profile after 72 h of culture (Fig. 1). In contrast, the addition of both agents reduces the number of cells with a normal FSC/SSC profile to 10.4%. The addition of Gö6976 could be delayed for at least 20 min after addition of anti-IgM, which shows that the protective action of cPKC depends on a relatively late event and that activation of cPKC immediately following anti-IgM stimulation is not suffi- Since the cPKC␥ isoform is not expressed in these cells (27), these results suggest that the anti-apoptotic effect of activated cPKC is due to activation of one or more of the PKC␣, -␤I, or -␤II isoforms.
Expression of LSP1 Residues 179 -330 Inhibits Translocation of PKC␤I to the Plasma Membrane-Increased anti-IgM-induced apoptosis was evident after treatment of W10 cells with Gö6976 or after transfection with B-LSP1 containing LSP1 residues 179 -330. To determine whether expression of B-LSP1 inhibits the activation of cPKC, we measured the extent of anti-IgM-induced translocation of PKC␣, -␤I, and -␤II to the plasma membrane fractions of W10 cells and of the transfectant TW10.1. Translocation of PKC is often used as a measure of activation (28). Plasma membrane fractions were prepared from unstimulated cells and from cells stimulated at 37°C with anti-IgM for 5 or 20 min. Equal amounts of plasma membrane protein were analyzed for the presence of PKC␣, -␤I, and -␤II by Western blotting. All three PKC isoforms tested translocated to the plasma membrane after treatment of W10 cells with anti-IgM (Fig. 2). Translocation was evident after 5 min of stimulation and did not change significantly for the next 15 min. Translocation of PKC␣ and PKC␤II in TW10.1 cells did not differ significantly from that measured in W10 cells. Interestingly, translocation of PKC␤I was significantly inhibited in TW10.1 cells, when measured 5 or 20 min after addition of anti-IgM. We conclude from these data that expression of B-LSP1 specifically inhibits anti-IgM-induced PKC␤I translocation to the plasma membrane.
To determine whether the increased anti-IgM-induced apoptosis of TW10.1 cells is due to the inhibition of PKC␤I activation, we transfected TW10.1 cells with an expression vector encoding intact rat PKC␤I and selected four colonies in which the expression levels of PKC␤I were 2-3 times higher than in the untransfected TW10.1 cells. We also isolated four control colonies that express the puromycin resistance gene but have no increased levels of PKC␤I. Stimulation with anti-IgM showed that overexpression of PKC␤I rendered the TW10.1 cells less susceptible to anti-IgM-induced apoptosis (Fig. 1). Although only 16.8% of TW10.1 cells or 17.1% of the puromycin resistant control colonies (data not shown) were viable after 72 h of anti-IgM stimulation, 35.4% of PKC␤I-overexpressing TW10.1 transfectants remained viable. This shows that the B-LSP1-mediated inhibition of PKC␤I contributes significantly to the B-LSP1-mediated increase in anti-IgM-induced apoptosis of TW10.1 cells. The partial reversal may be related to the level of overexpression of PKC␤I. Alternatively, it may indicate that expression of B-LSP1 also affects anti-IgM-induced apoptosis through mechanisms that do not involve PKC␤I activation.
Inhibition of PKC␤I Translocation Inhibits Anti-IgM-induced ERK2 Activation-To determine whether the specific inhibition of anti-IgM-induced translocation of PKC␤I affects a known cPKC-regulated, anti-IgM-stimulated signaling pathway, we measured the anti-IgM-induced activation of the MAP kinase ERK2 in W10 and TW10.1 cells. Activation of ERK2 is associated with survival in many cell types (29,30), and results in Fig. 1 show that this is the case in W10 cells as well. Treatment of W10 cells with 25 M PD098059, an inhibitor specific for MEK1 (31), the direct activator of ERK2, does not significantly affect cell viability as 89.1% of W10 cells displayed a normal FSC/SSC profile after 72 h of culture. However, in cultures treated with PD098059 and 5 g/ml anti-IgM, only 34.3% of cells were viable, showing that anti-IgM-induced activation of ERK2 protects W10 cells from anti-IgM-induced apoptosis.
Anti-IgM stimulation activates ERK2 in a PKC-independent manner through the RAS/Raf-1/MEK1 pathway (12,32,33) and in a PKC-dependent manner through activation of Raf-1 (34 -36). Different PKC isoforms can contribute to Raf-1 activation when overexpressed in COS cells (37), but the specific PKC isoform involved in anti-IgM-induced Raf-1/MEK1/ERK2 activation in B-cells is not yet known. Fig. 3A shows that treatment of W10 cells with 0.25 M Gö6976 or with 25 M PD098059 significantly inhibits anti-IgM-induced ERK2 activation as measured by the appearance of phosphorylated ERK2, which has a slightly lower mobility on SDS-PAGE than the non-phosphorylated form of ERK2. These results confirm that, in W10 cells, activated cPKC isoforms contribute to the activation of ERK2 and that activation of ERK2 depends on active MEK1. It is interesting to note that the effect of Gö6976 is more pronounced after 20 min than after 2 min of anti-IgM stimulation, whereas the MEK1 inhibitor PD098059 acts equally well at both time points. Increasing the pre-incubation time for Gö6976 to 35 min gave a similar result (data not shown). This suggests that the early activation of MEK1/ERK2 by anti-IgM stimulation is less dependent on activation of cPKC, whereas at later times activated cPKC contributes significantly to ERK2 activation. The early activation of ERK2 activation may be more dependent on activated RAS.
We determined the involvement of PKC␤I in anti-IgM-induced ERK2 activation by two approaches. First, W10 cells were transfected with a pcDNA3-based expression vector encoding a constitutively active PKC␤I protein (PKC␤I⌬NPS, lacking the NH 2 -terminal pseudosubstrate region) and two colonies expressing the PKC␤I⌬NPS protein were selected. Two G418-resistant colonies that do not express PKC␤I⌬NPS were selected as control cells. Fig. 3B shows that ERK2 activity is 4 -5 times higher in the PKC␤I⌬NPS-expressing transfectants than in the control cells, showing that activation of PKC␤I leads to activation of the ERK2 pathway. Second, we analyzed ERK2 activation in W10 and TW10.1 cells in which the anti-IgM-induced translocation of PKC␤I is inhibited. ERK2 activation was measured 2, 10, and 20 min after anti-IgM treatment, using the decreased mobility of phosphorylated ERK2 on SDS-PAGE as a read-out (Fig. 3C) or by measuring ERK2 activity in an immune kinase assay (Fig. 3, D and E). In W10 cells, ERK2 is activated efficiently after 2 min of anti-IgM stimulation and does not change significantly over the next 15 min. The extent of activation of ERK2 in TW10.1 cells is similar to that found in W10 cells when measured 2 or 10 min after stimulation but after 20 min is significantly less than in W10 cells. These data show that inhibition of cPKC with Gö6976 or inhibition of PKC␤I by expression of B-LSP1 both inhibit only the late but not the early activation of ERK2 by anti-IgM. This is strong evidence that anti-IgM-induced activation of PKC␤I contributes significantly to the late activation of ERK2. Given that inhibition of ERK2 by PD098059 increases the extent of anti-IgM-induced apoptosis, we suggest that the inhibition of ERK2 activation in TW10.1 cells contributes to the increase in anti-IgM-induced apoptosis. The designation of ERK2 as an antiapoptotic protein does not agree with a report showing that inhibition of anti-IgM-induced activation of ERK2 in WEHI-231 cells protects from anti-IgM-induced apoptosis (38). These discordant findings may be related to the different methods used to inhibit ERK2 activation. Whereas we established a protective role for ERK2 using the MEK1 inhibitor PD098059 to inhibit ERK2, the pro-apoptotic role of ERK2 was established using expression of the phosphatase MKP-1 to inhibit ERK2. However, this phosphatase is not specific for ERK2 and inhibits other members of the MAP kinase family, p38 and stress-activated protein kinase/c-Jun NH 2 -terminal kinase 1 as well (39).
PKC␤I Interacts with LSP1 Residues 179 -330 -We used pull-down experiments to determine whether LSP1 interacts with PKC␤I. A GST fusion protein containing the intact LSP1 or the GST protein was mixed with Nonidet P-40-soluble lysates from W10 cells. After incubation for 1 h, the GST protein and the LSP1/GST fusion proteins were recovered using glutathione-Sepharose beads and analyzed by Western blotting. Using this protocol PKC␤I but not PKC␣ or PKC␤II were recovered from lysates when mixed with the LSP1/GST fusion protein. No detectable amounts of PKC␣, -␤I, or -␤II were recovered using the GST protein, indicating a specific interaction of LSP1 with PKC␤I (Fig. 4A). To determine whether PKC␤I and LSP1 interact directly, we performed in vitro binding assays using human recombinant PKC␤I and GST fusion proteins containing different LSP1 sequences. One g of GST or GST fusion protein was incubated with 50 ng of recombinant human PKC␤I. Fig. 4B shows that PKC␤I interacts preferentially with intact LSP1 or with the COOH-terminal domain residues 179 -330. In replicate experiments, binding to the NH 2 -terminal residues 1-178 is not significantly higher than binding to the GST protein alone, indicating that the preferential binding site or sites for PKC␤I are located between LSP1 residues 179 and 330.
We do not as yet know why expression of LSP1 residues 179 -330 in the LSP1 ϩ TW10.1 cells inhibits translocation of PKC␤I, but we propose that the endogenous, intact LSP1 sequesters inactive PKC␤I to a cytosolic localization and that, in response to anti-IgM-generated signals, the LSP1/PKC␤I complex dissociates, allowing for the translocation of PKC␤I. The release of PKC␤I from B-LSP1 in response to anti-IgM stimulation may be less efficient, leading to inhibition of PKC␤I translocation. LSP1 is a Ca 2ϩ -binding protein and contains two FIG. 3. Inhibition of PKC␤I leads to inhibition of ERK2 activation. A, W10 cells were stimulated with anti-IgM alone (top row) or with anti-IgM and the cPKC inhibitor Gö6976 (middle row) or with anti-IgM and the MEK1 inhibitor PD098059 (bottom row). Cells were harvested just before or 2 or 20 min after addition of anti-IgM and lysed in sample buffer. ERK2 phosphorylation was determined by Western blotting. Phosphorylated ERK2 is indicated with an asterisk (*) and has a slightly lower mobility during PAGE than the non-phosphorylated ERK2. B, W10 were transfected with the constitutively active PKC␤I⌬NPS construct. Two control transfectants and two transfectants overexpressing PKC␤I⌬NPS (ϳ3-fold compared with the endogenous PKC␤I, data not shown) were assayed for ERK2 activity using an in vitro immune kinase assay with MBP as substrate. Both PKC␤I⌬NPS expressing transfectants contained 4 -5-fold more ERK2 activity than the control cells.  4. LSP1 interacts with PKC␤I. A, pull-down experiments were performed by incubating GST or a GST fusion protein containing LSP1 residues 1-330 with Nonidet P-40-soluble lysates from W10 cells as described, and proteins associated with GST or the GST-LSP1 fusion protein were analyzed for the presence of PKC␣, -␤I, and -␤II by Western blotting. TL, total lysate prepared by lysis of W10 cells directly into SDS-PAGE sample buffer. The position of a 73-kDa molecular mass marker is indicated at the left. B, recombinant human PKC␤I (50 ng) was incubated with 1 g of GST or GST fusion proteins containing LSP1 residues 1-330, 1-178, or 179 -330. GST and GST fusion proteins were recovered on glutathione-Sepharose beads and analyzed for the presence of PKC␤I (top panels) and GST or GST fusion proteins (bottom panels) by Western blotting. In the bottom panels, only that portion of the Western blots containing the GST or GST fusion protein is shown.
putative Ca 2ϩ -binding EF-hand motifs near the NH 2 terminus (16). Thus, binding of Ca 2ϩ to the NH 2 -terminal domain may result in a structural change in the COOH-terminal domain, leading to dissociation of PKC␤I. Since both EF-hand motifs are absent from B-LSP1, suggesting that B-LSP1 does not bind Ca 2ϩ , the B-LSP1/PKC␤I complex may not dissociate after the anti-IgM-induced increase in [Ca 2ϩ ] i thereby inhibiting the movement of PKC␤I to the plasma membrane. Alternatively, since the NH 2 -terminal domain of LSP1 contains a phosphorylation site for casein kinase II (40) and several putative PKC phosphorylation sites, it is possible that the binding of PKC␤I to LSP1 residues 179 -330 is regulated by phosphorylation of specific sites in the LSP1 NH 2 -terminal domain. Thus, LSP1 may protect from apoptosis only in response to certain signals such as increased [Ca 2ϩ ] i . Apoptosis induced by sorbitol, nocodazole, C 2 -ceramide, or H 2 O 2 is not regulated by LSP1 (20), possibly because these signals do not lead to dissociation of PKC␤I from LSP1.
The study of PKC isoform-specific functions is hampered by the lack of isoform-specific activators or inhibitors. Our data suggest that B-LSP1 is a unique reagent that specifically inhibits the anti-IgM-induced activation of PKC␤I. Experiments to determine whether B-LSP1 also inhibits PKC␤I translocation in response to other Ca 2ϩ -generating signals such as ionomycin or anti-CD20 stimulation are currently under way. We have used expression of B-LSP1 to inhibit anti-IgM-induced translocation of PKC␤I in a transformed B-cell line, which significantly impacted on the susceptibility of these cells to anti-IgM-induced apoptosis. Many B-lymphoma cell lines are susceptible to anti-IgM-induced apoptosis, a characteristic that forms the basis for using anti-Ig antibodies as therapy for B-lymphoma in vivo (41,42). Our results suggest that the efficiency of anti-Ig therapy may be enhanced by the concomitant inhibition of PKC␤I.