Phospholipase C-γ2 couples Bruton's tyrosine kinase to the NF-κB signaling pathway in B lymphocytes

Mutations in the gene encoding Bruton's tyrosine kinase (BTK) interfere with B cell proliferation and lead to an X-linked immunodeficiency in mice characterized by reduced B cell numbers. Recent studies have established that BTK transmits signals from the B cell antigen receptor (BCR) to transcription factor NF-κB, which in turn reprograms a set of genes required for normal B cell growth. We now demonstrate that induction of NF-κB via this pathway requires the intermediate action of the -γ2 isoform of phospholipase C (PLC-γ2), a potential phosphorylation substrate of BTK. Specifically, pharmacologic agents that block the action of either PLC-γ2 or its second messengers prevent BCR-induced activation of IκB kinase. Moreover, activation of NF-κB in response to BCR signaling is completely abolished in B cells deficient for PLC-γ2. Taken together, these findings strongly suggest that PLC-γ2 functions as an integral component of the BTK/NF-κB axis following BCR ligation. Interference with this NF-κB cascade may account for some of the B cell defects reported forplc-γ2 − /− mice, which develop an X-linked immunodeficiency-like phenotype.


Mutations in the gene encoding Bruton's tyrosine kinase (BTK) interfere with B cell proliferation and lead to an X-linked immunodeficiency in mice characterized by reduced B cell numbers. Recent studies have established that BTK transmits signals from the B cell antigen receptor (BCR) to transcription factor NF-B, which in turn reprograms a set of genes required for normal B cell growth. We now demonstrate that induction of NF-B via this pathway requires the intermediate action of the -␥2 isoform of phospholipase C (PLC-␥2), a potential phosphorylation substrate of BTK. Specifically, pharmacologic agents that block the action of either PLC-␥2 or its second messengers prevent BCR-induced activation of IB kinase. Moreover, activation of NF-B in response to BCR signaling is completely abolished in B cells deficient for PLC-␥2. Taken together, these findings strongly suggest that PLC-␥2 functions as an integral component of the BTK/NF-B axis following BCR ligation.
Interference with this NF-B cascade may account for some of the B cell defects reported for plc-␥2 ؊/؊ mice, which develop an X-linked immunodeficiency-like phenotype.
The generation and survival of B lymphocyte subpopulations is contingent upon the expression of a functional B cell antigen receptor complex (BCR) 1 (1,2). BCR engagement directs B cell biological responses by initiating biochemical signaling cascades involving the cytoplasmic protein tyrosine kinases Lyn, Syk, and BTK (3)(4)(5). BTK plays an integral role in transducing BCR-directed signals, because mutations in the btk gene result in the B cell deficiencies X-linked agammaglobulinemia (XLA) in man and X-linked immunodeficiency (xid) in mice (6 -10). B cells from xid mice are defective in survival and proliferation, implicating BTK in these biological processes (10 -12). However, the molecular mechanisms by which BTK effects B cell proliferation and survival are not well understood.
Like BTK, transcription factor NF-B has been implicated in the regulation of genes essential for B cell responses including proliferation and survival (13)(14)(15). In resting cells, NF-B is sequestered in the cytoplasmic compartment via its association with a family of inhibitory proteins, termed IBs (16). Recent studies have identified a cytokine-inducible IB kinase complex (IKK) consisting of two catalytic (IKK␣ and IKK␤) and one regulatory subunit (IKK␥) (17). In response to NF-B activating signals, IKK phosphorylates and targets IB for degradation (17). We and others (18,19) have recently shown that BTK couples the BCR to IKK and NF-B. However, the biochemical mechanism by which BTK activates NF-B remains largely undefined. BTK, in concert with the protein tyrosine kinase Syk and the adaptor protein BLNK, has recently been demonstrated to phosphorylate and activate PLC-␥2 (22)(23)(24). In response to BCR signals, PLC-␥2 catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate, generating inositol 1,4,5-trisphosphate and diacylglycerol. Inositol 1,4,5-trisphosphate induces the release of Ca 2ϩ from intracellular stores, and diacylglycerol facilitates the activation of PKC isoenzymes (20,21). Thus, BTK-dependent activation of PLC-␥2 is essential for BCR-initiated calcium fluxes (22). However, the functional consequences of PLC-␥2 signaling in the activation of nuclear factors that direct B cell responses are not known.
In this report, we provide two lines of evidence indicating that BCR-initiated activation of NF-B is mediated by PLC-␥2. First, DT40 chicken B cells deficient for PLC-␥2 fail to translocate NF-B to the nucleus upon BCR activation. Second, pharmacologic inhibition of PLC-␥2 or its second messengers prevents BCR-responsive activation of IKK and phosphorylation of IB␣ in primary B cells. These biochemical findings provide a potential molecular explanation for the B cell defects recently reported for plc-␥2 Ϫ/Ϫ mice, which display an xid-like phenotype reminiscent of animals lacking functional BTK (10).
Splenocytes and primary B lymphocytes were isolated from spleens of C57Bl6 mice. For phospho-IB␣ Western analyses, RBC-depleted splenocytes were cultured and stimulated as indicated. For IKK in vitro kinase assays, B cells were purified by a process of negative selection on an affinity chromatography column (Cedarlane, Ontario, Canada). The purity of B cells isolated in this manner was ϳ90 -95% as verified by fluorescence-activated cell sorter analysis using anti-B220 and anti-IgM antibodies (PharMingen). All purifications were performed at 4°C, and primary cells were used immediately upon purification.
Electrophoretic Mobility Shift Assays (EMSAs)-Nuclear extracts were prepared and used in DNA-binding reactions as described previously (18). For EMSAs, an [␣-32 P]CTP-and [␣-32 P]ATP-labeled doublestranded oligonucleotide probe derived from the B enhancer element of the IL-2R␣ receptor promoter (5Ј-CAACGGCAGGGGAATTC-CCCTCTCCTT-3Ј) was used. To verify equal amounts and integrity of proteins in the nuclear extracts, a control oligonucleotide for NF-Y was used. DNA-binding reactions were resolved by PAGE and visualized by autoradiography.
Western Blot Analyses-For Western blot analysis of RelA and c-Rel, nuclear extracts equivalent to 2 ϫ 10 7 cells were denatured in Laemmli reducing buffer by boiling at 95°C for 3 min, and the proteins were resolved by SDS-PAGE. Proteins were electrotransferred onto nitrocellulose membranes and subjected to immunoblotting with rabbit polyclonal antibodies against RelA, c-Rel, or SP1 as described previously (18). For IB␣ degradation assays, 4 ϫ 10 6 cells/sample were preincubated for 30 min in medium containing 50 M cycloheximide and then stimulated as indicated. Cell extracts were resolved by SDS-PAGE, transferred onto nitrocellulose membranes, probed with antibodies against chicken IB␣ (pp40; gift of C. Chen) and p38 MAPK (Santa Cruz Biotechnology), and detected using the ECL system. Western blot analyses of IB␣ phosphorylation were performed as above and probed with antibodies against mouse IB␣ (Santa Cruz Biotechnology) or phosphorylated Ser-32/Ser-36 IB␣ (Santa Cruz Biotechnology).
Plasmid Constructs and Luciferase Assays-The B reporter plasmid encoding firefly luciferase under the control of a promoter containing six consensus NF-B binding sites (6B) and a control vector containing a Renilla luciferase gene fused to a thymidine kinase promoter have been described previously (25).
The indicated DT40 cell lines were each cotransfected by electroporation (250 V, 960 microfarads, Bio-Rad Gene Pulser) with 5 g of the 6B reporter construct and 1 g of the Renilla construct. 18 h posttransfection, cells were stimulated for 6 h with anti-IgM. Cells were harvested, and levels of both firefly and Renilla luciferase were determined using a Dual Luciferase Reporter Assay System (Promega). Levels of firefly luciferase expression were normalized against Renilla as a control for transfection efficiency.
In Vitro Kinase Assays-In vitro kinase assays were performed on the cytosolic fraction of 5 ϫ 10 6 B cells as described previously (18). Briefly, cell extracts from 0.5 ϫ 10 6 cell equivalents were removed for Western blot analysis, and the remaining cell extract was subjected to immunoprecipitation with anti-IKK␣ plus anti-IKK␤ antibodies (Santa Cruz Biotechnology). The immunocomplexes were then resuspended in 20 l of kinase buffer (20 mM HEPES, pH 7.2, MgCl 2 (2 mM), MnCl 2 (2 mM), dithiothreitol (1 mM), ATP (20 M)) containing 1.0 Ci of [␥-32 P]ATP and 50 g/ml wild type GST-IB␣ substrate. The reaction was allowed to continue for 30 min at 30°C under agitation and then was terminated by the addition of 4ϫ SDS sample buffer. The samples were resolved by 8% SDS-PAGE and stained with Coomassie Brilliant Blue to visualize the GST-IB␣ substrate. The gels were dried and exposed to x-ray film to visualize ␥-32 P-phosphorylated GST-IB␣.

RESULTS AND DISCUSSION
In prior studies, we established that BTK is required for nuclear translocation of NF-B in BCR-stimulated B cells (18). However, the molecular mechanism by which BTK facilitates NF-B activation is poorly defined. Recent findings suggest that BCR-directed nuclear translocation of NF-B requires the activation of the calcium-responsive phosphatase calcineurin (26). To define further the mechanism employed by BTK to effect NF-B activation, we investigated a role for calcium and calcineurin in BCR-responsive nuclear translocation of NF-B in DT40 B cells.
The DT40 B cell system is amenable to genetic manipulation and has thus proven invaluable for biochemical analysis of BCR-signaling events (27). To determine whether calcium and calcineurin play a role in BCR-responsive activation of NF-B in this cellular background, EMSA analyses were performed on nuclear extracts prepared from DT40 cells preincubated with pharmacological inhibitors of calcium, calcineurin, and PKCs. We used BAPTA-AM/EGTA to chelate intra-and extracellular calcium and cyclosporin A (CsA) to inhibit the calcium-responsive phosphatase calcineurin or bisindolylmaleimide (Bis I), a broad spectrum inhibitor of PKC isoenzymes (Fig. 1). We also treated DT40 cells with PMA and ionomycin, pharmacological agents known to activate NF-B via IKK, as a positive control (28). As expected, BCR cross-linking or PMA/ionomycin treatment resulted in the rapid nuclear accumulation of NF-B (compare lane 1 with 2 and 10). However, BCR-directed nuclear translocation of NF-B was inhibited by treatment with BAPTA-AM/EGTA or CsA (lanes 3 and 4). Bis I treatment significantly, although not completely, inhibited this response (lane 5). However, preincubation with Bis in combination with either BAPTA-AM/EGTA or CsA resulted in a complete block in NF-B nuclear translocation upon BCR activation (lanes 6  and 7). This result demonstrates that inhibition of either calcium or calcineurin and PKC completely abolishes BCR-directed activation of NF-B in DT40 B cells. Upon BCR ligation, BTK activates a distinct set of signal transducers to initiate downstream signaling events (3,29). Of these, Akt, MAPK, and PLC-␥2 have the capability to activate NF-B via IKK. Although both Akt and MAPK have been directly linked to IKK activation (30,31), such a role has not been demonstrated for PLC-␥2. However, our finding that calcium and PKC are essential for nuclear translocation of NF-B in BCR-stimulated DT40 cells implicates PLC-␥2 in this response (Fig. 1). Therefore, we next explored an involvement of PLC-␥2 in NF-B nuclear translocation in B cells stimulated via the BCR.
To determine whether PLC-␥2 is critical for BCR-directed nuclear translocation of NF-B, we used mutant chicken DT40 B cells lacking PLC-␥2 (DT40.PLC-␥2) along with BTK-deficient (DT40.BTK) and parental DT40 B cells. Cells were induced via the BCR, and their nuclear NF-B content was assessed by EMSA ( Fig. 2A). Although BCR cross-linking leads to a marked increase in nuclear NF-B in DT40 cells ( Fig. 2A,  5 and 3 and 6). However, PMA and ionomycin mobilized similar levels of nuclear NF-B in all three cell types ( Fig. 2A, lanes 7-9). These results strongly suggest that like BTK, PLC-␥2 plays an essential role in the transmission of BCR signals to activate NF-B.
To ascertain whether the observed defect was due to delayed kinetics of NF-B activation, we compared BCR-responsive nuclear translocation of NF-B in DT40.PLC-␥2 cells with that in DT40 B cells over a period of 4 h (Fig. 2B). Upon BCR cross-linking, DT40 B cells rapidly translocated NF-B to the nucleus and maintained elevated levels up to 4 h after activation. In contrast, nuclear levels of NF-B did not increase in DT40.PLC-␥2 B cells at any time point within that period (Fig.  2B, compare lanes, 1, 3, 5, 7, and 9 with 2, 4, 6, and 8). To verify further that the NF-B activation defect in DT40.PLC-␥2 B cells was due to PLC-␥2 deficiency, reconstitution experiments were performed. In response to BCR engagement, DT40.PLC-␥2 B cells expressing wild type human PLC-␥2 (DT40.PLC-␥2R (23)) were capable of NF-B nuclear translocation as determined by EMSA and a NF-B responsive luciferase reporter assay (Fig. 2, C and D). These data strongly suggest that PLC-␥2 is critical for transmission of BCR-dependent signals that lead to the nuclear translocation of NF-B.
Members of the NF-B/Rel family of proteins include p50/ NF-B1, p52/NF-B2, RelA, c-Rel, and RelB, which have the capacity to form either homo-or heterodimers (16). NF-B  1-3), stimulated with anti- IgM (lanes 4 -6), or with PMA/ionomycin (lanes 7-9). Nuclear extracts were prepared and used in Western analyses as described under "Experimental Procedures." Blots were stripped and reprobed with an antibody against the constitutive nuclear factor SP-1 to verify extract integrity (lower panel). B, Western blot analysis of c-Rel nuclear levels was performed as in A. . We previously demonstrated that RelA and c-Rel fail to undergo nuclear translocation upon BCR stimulation in BTK-deficient B cells. To test whether BTK-mediated RelA and c-Rel nuclear translocation requires PLC-␥2, we compared the ability of DT40.PLC-␥2, DT40.BTK, and DT40 B cells to translocate these subunits to the nucleus upon BCR-cross-linking (Fig. 3, A  and B). Immunoblotting of nuclear extracts from unactivated (lanes 1-3), anti-IgM stimulated (lanes 4 -6), and PMA/ionomycin treated (lanes 7-9) cells with Rel subunit-specific antibodies revealed that nuclear accumulation of both RelA and c-Rel occurs in DT40 B cells following BCR stimulation (Fig. 3, A and  B, lanes 1 and 4). In contrast, BCR-responsive nuclear translocation of RelA and c-Rel is not observed in either DT40.PLC-␥2 or DT40.BTK B cells. Treatment with PMA/ionomycin induced nuclear translocation of both Rel species (Fig. 3,  A and B, lanes 7-9) in all three cell lines. Furthermore, the observed differences in Rel subunit translocation are not attributable to either variation in total protein content of the nuclear extracts or their integrity, because similar amounts of the constitutively expressed transcription factor SP1 are detectable in all samples (Fig. 3, A and B, lower panels). Thus, BCR-directed nuclear translocation of RelA and c-Rel is PLC-␥2-dependent.
NF-B dimers are found in the cytoplasm of quiescent cells, bound to members of a family of inhibitory molecules termed IBs. BCR-induced nuclear translocation of NF-B is contingent upon the phosphorylation and proteolytic degradation of IB␣, a process that requires BTK. We compared the ability of DT40.PLC-␥2 B cells with DT40.BTK and DT40 B cells to degrade IB␣ in response to BCR activation. Cells were incubated with anti-IgM antibodies or PMA and ionomycin for indicated periods, and cytoplasmic extracts were immunoblotted for chicken IB␣ (Fig. 4, upper panel). As expected, DT40 B cells rapidly degraded IB␣ upon BCR activation. Consistent with the results shown in Fig. 2, DT40.PLC-␥2 B cells failed to degrade IB␣ in response to BCR stimulation (Fig. 4, compare lanes 1-4 with 6 -9 and 11-14). All three cell lines efficiently degraded IB␣ in response to treatment with PMA and ionomycin. Therefore, loss of PLC-␥2 does not affect the downstream components necessary for IB␣ degradation. These results demonstrate that BCR-directed degradation of IB␣ specifically requires PLC-␥2.
Prior biochemical studies have identified several NF-B agonists that converge on IKK␣ and IKK␤ including TNF and IL-1 (17). Additionally, we have recently established that BCR-initiated activation of NF-B by BTK proceeds via IKK (18). To extend our finding that PLC-␥2 is required for BCR-directed nuclear translocation of NF-B, we explored a role for PLC-␥2 in IKK activation. We tested whether pharmacological agents that block PLC-␥2 and its second messengers could prevent BCR-induced activation of IKK in primary B cells (Fig. 5A). In response to activation signals via the BCR or CD40, or treatment with PMA, IKK enzymatic activity was significantly increased as determined by in vitro kinase assays using recombinant GST-IB␣ as the substrate (Fig. 5A, compare lane 1  with 2, 7, and 8). In contrast, incubation of B cells with either the PLC-␥-specific inhibitor (U-73122) or inhibitors of its second messengers (BAPTA-AM/EGTA, CsA, or Bis I) prior to BCR stimulation abolished this activity (Fig. 5A, lanes 3-6). These data implicate PLC-␥2, calcium, calcineurin, and PKC in IKK activation upon BCR ligation. Moreover, they verify the role of these signaling molecules in BCR-responsive activation of IKK in a physiologically relevant background.
To confirm this observation, we performed Western blot analyses of cytosolic fractions from BCR-, CD40-, or PMAstimulated splenocytes using an antibody directed against Ser-32/Ser-36-phosphorylated IB␣ (Fig. 5B, upper panel). Stimulation via either the BCR or CD40 induced phosphorylation of IB␣ (Fig. 5B, compare lanes 1, 2, and 6). BCR-responsive IB␣ phosphorylation was blocked by pretreatment with either BAPTA-AM/EGTA, CsA, or Bis I (Fig. 5B, lanes 3-5). Also, PMA stimulation resulted in IB␣ phosphorylation that was abrogated by pretreatment with the PKC inhibitor Bis I (Fig.  5B, lanes 7 and 8). These observations implicate PLC-␥2, calcium, and PKC in BCR-responsive activation of IKK and phosphorylation of IB␣. Moreover, the observation that cells pre-  6 -10), and DT40.BTK (lanes 11-15) B cells is shown. Cells were translationally arrested prior to stimulation by preincubation with cycloheximide. Cells were then stimulated for the indicated times with anti-IgM or PMA/ionomycin. Cytoplasmic extracts were used for Western analysis as described (18) with antisera against IB␣ (pp40). The blot was stripped and reprobed with an antibody against p38 MAPK to verify the protein content and integrity. Cytosolic extracts were immunoprecipitated with antibodies against IKK␣ and IKK␤, and the resulting immunocomplexes were subjected to in vitro kinase assays containing [␥-32 P]ATP and GST-IB␣ as the substrate. Kinase assays were resolved by SDS-PAGE and subjected to autoradiography. The immunoblot with IKK␤ was performed with 10% of the cell extract used in the in vitro kinase assay to monitor steady state levels of IKK␤. B, EGTA/BAPTA-AM, CsA, or Bis I prevent BCRinduced IB␣ Ser-32/Ser-36 phosphorylation. Western blot analysis of cytosolic extracts from splenocytes is shown. Cells were preincubated with the indicated inhibitors and then stimulated as in A. Cell lysates were used in Western analysis as in Fig. 3. Phosphorylated IB␣ was revealed using an antibody directed against phosphorylated serine residues (32 and 36) of IB␣ (Santa Cruz Biotechnology). The blot was stripped and reprobed with an anti-IB␣ antibody. treated with CsA fail to activate IKK upon BCR cross-linking identifies calcineurin as a critical mediator of this response. This observation is consistent with the recent finding that calcineurin and PKCs synergize to induce IKK activation in T cells (28). Collectively, these data suggest that PLC-␥2 is likely to mediate BCR-responsive activation of IKK, phosphorylation of IB␣, and nuclear translocation of NF-B.
We have found that PLC-␥2 and its downstream signals are essential for BCR-directed activation of IKK and NF-B. Prior studies in BCR-stimulated B cells have revealed that PLC-␥2 is activated via the concerted actions of BTK, Syk, and BLNK (3,32). Therefore, it is likely that PLC-␥2 is the principal BTK signal transducer for BCR-directed activation of IKK and NF-B. Further investigation is required to determine whether additional BTK targets, including Akt and MAPK, synergize with PLC-␥2 to effect nuclear translocation of NF-B in BCRstimulated B cells. However, the placement of PLC-␥2 in the BCR/BTK/NF-B signaling pathway provides the first potential molecular explanation for the similar xid-like B cell deficiencies displayed by plc-␥2 Ϫ/Ϫ and btk Ϫ/Ϫ mice (10,33).