PKCbeta modulates antigen receptor signaling via regulation of Btk membrane localization
2001; Springer Nature; Volume: 20; Issue: 20 Linguagem: Inglês
10.1093/emboj/20.20.5692
ISSN1460-2075
AutoresShin‐Wook Kang, Matthew I. Wahl, Julia Chu, Jiro Kitaura, Yuko Kawakami, R. Kato, Ruby S. Tabuchi, Alexander Tarakhovsky, Toshiaki Kawakami, Christoph W. Turck, Owen N. Witte, David J. Rawlings,
Tópico(s)Blood disorders and treatments
ResumoArticle15 October 2001free access PKCβ modulates antigen receptor signaling via regulation of Btk membrane localization Shin W. Kang Shin W. Kang Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Department of Biological Chemistry, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Matthew I. Wahl Matthew I. Wahl Howard Hughes Medical Institute, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Julia Chu Julia Chu Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Jiro Kitaura Jiro Kitaura Division of Allergy, La Jolla Institute for Allergy and Immunology, San Diego, CA, 92121 USA Search for more papers by this author Yuko Kawakami Yuko Kawakami Division of Allergy, La Jolla Institute for Allergy and Immunology, San Diego, CA, 92121 USA Search for more papers by this author Roberta M. Kato Roberta M. Kato Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Ruby Tabuchi Ruby Tabuchi Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Alexander Tarakhovsky Alexander Tarakhovsky Rockefeller University, New York, NY, USA Search for more papers by this author Toshiaki Kawakami Toshiaki Kawakami Division of Allergy, La Jolla Institute for Allergy and Immunology, San Diego, CA, 92121 USA Search for more papers by this author Christoph W. Turck Christoph W. Turck Howard Hughes Medical Institute, University of California, San Francisco, CA, 94143-0724 USA Search for more papers by this author Owen N. Witte Owen N. Witte Howard Hughes Medical Institute, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author David J. Rawlings Corresponding Author David J. Rawlings Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Molecular Biology Institute, Los Angeles, CA, 90095-1752 USA Jonsson Cancer Center, University of California, Los Angeles, CA, 90095-1752 USA Present address: Division of Immunology/Rheumatology, Department of Pediatrics, Box 356320, School of Medicine, University of Washington, Seattle, WA, 98195-6320 USA Search for more papers by this author Shin W. Kang Shin W. Kang Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Department of Biological Chemistry, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Matthew I. Wahl Matthew I. Wahl Howard Hughes Medical Institute, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Julia Chu Julia Chu Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Jiro Kitaura Jiro Kitaura Division of Allergy, La Jolla Institute for Allergy and Immunology, San Diego, CA, 92121 USA Search for more papers by this author Yuko Kawakami Yuko Kawakami Division of Allergy, La Jolla Institute for Allergy and Immunology, San Diego, CA, 92121 USA Search for more papers by this author Roberta M. Kato Roberta M. Kato Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Ruby Tabuchi Ruby Tabuchi Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author Alexander Tarakhovsky Alexander Tarakhovsky Rockefeller University, New York, NY, USA Search for more papers by this author Toshiaki Kawakami Toshiaki Kawakami Division of Allergy, La Jolla Institute for Allergy and Immunology, San Diego, CA, 92121 USA Search for more papers by this author Christoph W. Turck Christoph W. Turck Howard Hughes Medical Institute, University of California, San Francisco, CA, 94143-0724 USA Search for more papers by this author Owen N. Witte Owen N. Witte Howard Hughes Medical Institute, University of California, Los Angeles, CA, 90095-1752 USA Search for more papers by this author David J. Rawlings Corresponding Author David J. Rawlings Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA Molecular Biology Institute, Los Angeles, CA, 90095-1752 USA Jonsson Cancer Center, University of California, Los Angeles, CA, 90095-1752 USA Present address: Division of Immunology/Rheumatology, Department of Pediatrics, Box 356320, School of Medicine, University of Washington, Seattle, WA, 98195-6320 USA Search for more papers by this author Author Information Shin W. Kang1,2, Matthew I. Wahl3, Julia Chu1, Jiro Kitaura4, Yuko Kawakami4, Roberta M. Kato1, Ruby Tabuchi1, Alexander Tarakhovsky5, Toshiaki Kawakami4, Christoph W. Turck6, Owen N. Witte3 and David J. Rawlings 1,7,8,9 1Department of Pediatrics, University of California, Los Angeles, CA, 90095-1752 USA 2Department of Biological Chemistry, University of California, Los Angeles, CA, 90095-1752 USA 3Howard Hughes Medical Institute, University of California, Los Angeles, CA, 90095-1752 USA 4Division of Allergy, La Jolla Institute for Allergy and Immunology, San Diego, CA, 92121 USA 5Rockefeller University, New York, NY, USA 6Howard Hughes Medical Institute, University of California, San Francisco, CA, 94143-0724 USA 7Molecular Biology Institute, Los Angeles, CA, 90095-1752 USA 8Jonsson Cancer Center, University of California, Los Angeles, CA, 90095-1752 USA 9Present address: Division of Immunology/Rheumatology, Department of Pediatrics, Box 356320, School of Medicine, University of Washington, Seattle, WA, 98195-6320 USA *Corresponding author. E-mail: [email protected] The EMBO Journal (2001)20:5692-5702https://doi.org/10.1093/emboj/20.20.5692 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Mutations in Bruton's tyrosine kinase (Btk) result in X-linked agammaglobulinemia (XLA) in humans and X-linked immunodeficiency (xid) in mice. While targeted disruption of the protein kinase C-β (PKCβ) gene in mice results in an immunodeficiency similar to xid, the overall tyrosine phosphorylation of Btk is significantly enhanced in PKCβ-deficient B cells. We provide direct evidence that PKCβ acts as a feedback loop inhibitor of Btk activation. Inhibition of PKCβ results in a dramatic increase in B-cell receptor (BCR)-mediated Ca2+ signaling. We identified a highly conserved PKCβ serine phosphorylation site in a short linker within the Tec homology domain of Btk. Mutation of this phosphorylation site led to enhanced tyrosine phosphorylation and membrane association of Btk, and augmented BCR and FcϵRI-mediated signaling in B and mast cells, respectively. These findings provide a novel mechanism whereby reversible translocation of Btk/Tec kinases regulates the threshold for immunoreceptor signaling and thereby modulates lymphocyte activation. Introduction Aggregation of antigen receptors on B cells leads to the concerted activation of non-receptor tyrosine kinases, including Src, Syk and Tec family kinases, and lipid kinases including phosphoinositide 3-kinase (PI3K) (Kurosaki, 1999; Rawlings, 1999; Yang et al., 2000). These events initiate the formation of a multimeric signaling complex termed the signalosome. The signalosome promotes enhanced catalytic activity of phospholipase C-γ2 (PLCγ2) and the generation of inositol-1,4,5-trisphosphate (IP3) and diacylglycerol (DAG). These second messengers lead to intracellular calcium mobilization and protein kinase C (PKC) activation, respectively (Fruman et al., 2000). Synergistic signaling by PKC isoforms and calcium is important in the activation of transcription factors leading to diverse biological responses in activated B cells (Dolmetsch et al., 1997; Healy et al., 1997). Notably, the signalosome complex integrates both positive and negative signal transducers to produce a graded response that modulates and, ultimately, terminates B-cell receptor (BCR) signaling (Cyster and Goodnow, 1997). While signalosome activation events have been relatively well characterized, the events important for negative regulation of BCR signaling remain elusive. Bruton's tyrosine kinase (Btk) plays a pivotal role in the regulation of pre-B and mature BCR signaling, and is a major component of the BCR signalosome (Fruman et al., 2000; Guo et al., 2000). Mutation of Btk results in X-linked agammaglobulinemia (XLA) in humans and X-linked immunodeficiency (xid) in mice. These diseases are associated with impaired maturation of B cells, diminished immunoglobulin production, compromised T-independent immune response and marked attenuation of the sustained calcium signal upon BCR stimulation (Rawlings, 1999). Btk contains an N-terminal Pleckstrin homology (PH) domain, followed by Tec homology (TH), SH3, SH2 and kinase domains (Yang et al., 2000). While mutations within each of the domains can lead to immunodeficiency, the best characterized is a missense mutation, R28C, within the PH domain responsible for xid (Rawlings et al., 1993). Biochemical and crystallographic studies demonstrate that this mutation interferes with the capacity of the Btk PH domain to bind membrane phospholipids including phosphatidylinositol-3,4,5-trisphosphate (PIP3), thereby impairing its ability to translocate to the plasma membrane. Conversely, another PH domain mutation, E41K, enhances PIP3 binding avidity, augmenting Btk membrane translocation (Li et al., 1995; Baraldi et al., 1999). Interestingly, mice expressing this activated allele exhibit an immunodeficiency that is similar to but more severe than xid (Dingjan et al., 1998). This observation highlights the critical role of the Btk PH–phospholipid interaction in the regulation of optimal threshold and amplitude for BCR signaling. Membrane recruitment leads to Btk activation and is associated with rapid, but transient, phosphorylation of two major regulatory tyrosine residues. Upon BCR stimulation, Btk is transphosphorylated by Src family kinases at Y551 in the activation loop of the kinase domain. This promotes Btk catalytic activity and subsequent autophosphorylation at Y223 in the SH3 domain (Park et al., 1996; Rawlings et al., 1996). The sequential tyrosine phosphorylation of Btk following BCR engagement is rapidly attenuated, demonstrating the presence of negative regulatory signals that oppose these activating events (Wahl et al., 1997; Nisitani et al., 1999). Accumulating evidence suggests that PKC serine/threonine isoforms may function to negatively regulate protein tyrosine kinases (Sidorenko et al., 1996; Bagowski et al., 1999). The PKC family is subdivided into classical, novel and atypical isoforms on the basis of the relative requirements for calcium and/or DAG for catalytic activity. The PKC-related isoform PKCμ (also referred to as PKD) is recruited to the BCR and can negatively modulate Syk kinase activity in vitro (Sidorenko et al., 1996). Mice deficient for the classical PKC isoform, PKCβ, display a developmental phenotype similar to xid (Leitges et al., 1996). This observation has suggested a functional link between Btk and PKCβ. Paradoxically, BCR-mediated Btk tyrosine phosphorylation is increased and prolonged in PKCβ-deficient B cells. This complex phenotype suggests that PKCβ exerts a dual function as both a positive and negative regulator of the strength and duration of Btk activation (Tarakhovsky, 1997). The precise biochemical events responsible for the inhibitory function of PKCβ and other PKC isoforms on protein tyrosine kinases remain unknown. In this report we demonstrate that PKCβ is a potent inhibitor of Btk-mediated calcium signaling. To elucidate the underlying mechanism, we mapped the PKCβ phosphorylation site on Btk. A non-phosphorylatable mimetic of Btk displayed a marked increase in phosphotyrosine content, augmented capacity to support BCR-induced calcium mobilization and enhanced high affinity IgE receptor (FcϵRI)-dependent c-Jun N-terminal kinase (JNK) activation. In addition, we provide direct evidence that PKCβ negatively regulates Btk by altering its membrane localization. Taken together, these data demonstrate that PKCβ utilizes a unique regulatory mechanism to modulate the strength and duration of Btk activation. Conservation of the major PKCβ phosphorylation site in nearly all members of the Tec kinase family suggests that this mechanism operates to down-regulate the activity of multiple cell surface receptors over a broad range of immune and hematopoietic cell lineages. Results Pharmacological inhibition of PKC results in enhanced BCR-induced Ca2+ signaling, increased Btk membrane translocation and PLCγ2 tyrosine phosphorylation The overlapping phenotype of Btk and PKCβ-deficient mice suggests that PKCβ is required for peripheral B-cell development and function (Tarakhovsky, 1997). Paradoxically, engagement of receptors in PKCβ-deficient B cells (Leitges et al., 1996) or PKC-depleted mast cells (Yao et al., 1994) results in enhanced and prolonged Btk tyrosine phosphorylation, suggesting that PKCβ may negatively modulate Btk function. To test this possibility directly, we determined whether Btk-mediated downstream signals were modulated when PKCβ activity was inhibited. Murine (A20) and human (Ramos) B-cell lines were pretreated with PKC inhibitors and analyzed for their Ca2+ responses upon BCR cross-linking. Ligation of the BCR resulted in a typical biphasic calcium response: a transient peak increase in the intracellular Ca2+ concentration followed by low, but sustained levels of Ca2+ (Scharenberg and Kinet, 1998). Strikingly, when cells were pre-treated with Ro318425 (a general PKC inhibitor), the BCR-mediated Ca2+ signal was significantly enhanced in a concentration-dependent manner (Figure 1A). The effect of PKC inhibition is most pronounced during the sustained phase Ca2+ signal versus the initial peak response. Moreover, addition of the PKCβ-specific inhibitor LY379196 also led to a significantly enhanced response, suggesting a direct role for PKCβ in negative feedback regulation of the BCR-induced Ca2+ signal (Figure 1A). Figure 1.Pharmacological inhibition of PKC isoforms results in enhanced BCR-induced Ca2+ signaling. A20 and Ramos B cells were pretreated with either general (Ro318425) or PKCβ-specific inhibitor (LY379196) at doses indicated for 5 min, then activated with anti-Ig cross-linking. (A) In A20 and Ramos B cells, Ca2+ mobilization was monitored by spectrofluorimetry after a 30 s baseline measurement. (B) Ramos cells are stimulated with anti-IgM for 2 min or unstimulated after pretreatment with 5 μM Ro318425 or DMSO. Cells were fractionated by hypotonic lysis, and membrane and cytosolic fractions were analyzed by western blot analysis. Membrane enrichment of Btk was detected by immunoblotting with anti-Btk antibody (left). Fractionation efficiency was evaluated by immunoblotting with anti-transferrin receptor (TrfR). Total cell lysates (TCL) were immunoblotted with anti-phosphotyrosine antibody (middle). Enhancement of pp130 signal was observed in both membrane and cytosolic fractions by anti-phosphotyrosine (PY) blotting (right) and the same membrane was stripped and reblotted with anti-PLCγ2 antibody. (C) Ramos cells were treated as in (B), and lysed in BBS following BCR stimulation. Syk (left) or Lyn (right) was immunoprecipitated and immunoblotted with anti-PY antibody and the same membrane was stripped and reblotted with anti-Syk or anti-Lyn antibodies. Download figure Download PowerPoint The sustained phase of Ca2+ signal is associated with Btk activation, as measured by increased phosphotyrosine content, Btk membrane targeting and increased PLCγ2 activation (Fluckiger et al., 1998; Scharenberg et al., 1998). Accordingly, we considered the possible connection between Btk and PKC-mediated inhibition of Ca2+ signaling. To address the role of PKC in inhibiting Btk membrane targeting, Ramos cells were pretreated with Ro318425 and activated with anti-IgM. Following stimulation, cellular components were fractionated and the distribution of BCR-activated signaling molecules was examined. Constitutively membrane-localized transferrin receptor (TrfR) was used to demonstrate effective partitioning of membrane and cytosolic fractions. While no change in the distribution of TrfR was observed, levels of Btk in the membrane fraction were significantly increased (∼2-fold) in Ro318425-treated cells. No significant changes in Btk distribution were observed in the cytosolic fraction. This is consistent with a previous observation that the membrane-associated fraction constitutes a relatively small component of the overall pool of Btk (Wahl et al., 1997). These data demonstrate that enhancement of BCR-induced Ca2+ signal by PKC inhibition correlates with Btk membrane recruitment. We sought to determine whether PKC inhibitors affect the activity of other key signal transducers. First, we evaluated the global pattern of BCR-induced tyrosine phosphorylation in the presence and absence of PKC inhibitors using anti-phosphotyrosine antibody 4G10 (Figure 1B, second panel). Although the pattern was for the most part unchanged by PKC inhibition, one protein with approximate molecular weight of 130 kDa (pp130) exhibited enhanced tyrosine phosphorylation upon pretreatment with Ro318425 (Figure 1B, second panel). Based upon its apparent molecular weight, we hypothesized that this protein might be PLCγ2, a likely candidate molecule for a functional link between Btk activation and Ca2+ signal (Fluckiger et al., 1998; Scharenberg et al., 1998). Indeed, when the membrane from Figure 1B (first two panels) was stripped and reblotted with anti-PLCγ2 antibody, pp130 co-migrated with PLCγ2 (Figure 1B, panel 3). Identical results were obtained when PLCγ2 was immunoprecipitated and examined for phosphotyrosine content (data not shown). Notably, we observed a significant level of tyrosine phosphorylated PLCγ2 in the cytosolic fraction upon inhibition of PKC. The exact kinetics and sites for tyrosine phosphorylation on PLCγ2 are currently unknown. However, the presence of tyrosine-phosphorylated PLCγ2 in the cytosolic fraction suggests that PLCγ2 may translocate between the membrane and cytosol during its activation/inactivation process. In addition, we evaluated the phosphotyrosine content of Syk and Lyn, two critical non-receptor tyrosine kinases that regulate Btk and/or Ca2+ signaling (Kurosaki, 1999). Neither tyrosine phosphorylation nor in vitro kinase activity (data not shown) was altered by PKC inhibitors (Figure 1C). Together, these results suggested that inhibition of PKC leads specifically to increased membrane targeting of Btk, enhanced phosphorylation of PLCγ2 and augmented BCR-mediated Ca2+ signaling. PKCβ co-expression down-modulates both Btk transphosphorylation and autophosphorylation We utilized a fibroblast expression system to define directly the functional interaction between Btk and PKC isoforms. To study the effect of PKC co-expression on Lyn-mediated Btk activation, Btk, Lyn and PKCβ proteins were coordinately expressed in NIH 3T3 cells using recombinant vaccinia virus. Btk was immunoprecipitated and its tyrosine phosphorylation content was measured by immunoblotting (Figure 2). Btk tyrosine phosphorylation significantly increased with Lyn co-expression (as described previously by Rawlings et al., 1996). An increasing dosage of PKCβ led to a progressive diminution of Btk phosphorylation, returning to basal levels at the highest dosage (Figure 2A). Identical results were obtained through expression of either PKCβI or PKCβII, isoforms differing only by a short alternatively spliced C-terminus. Therefore, PKCβI was used for all subsequent experiments. Figure 2.PKCβ specifically down-modulates both Btk transphosphorylation and autophosphorylation. (A) Btk, Lyn and PKCβ proteins were coordinately expressed in NIH 3T3 cells using recombinant vaccinia virus. Btk was immunoprecipitated and the total tyrosine phosphorylation content was measured by immunoblotting. Btk loading was measure by anti-Btk antibody. Relative tyrosine phosphorylation levels were quantified by densitometric analysis. (B) Left panel: Btk, Lyn and Akt proteins were co-expressed and Btk tyrosine phosphorylation was analyzed as in (A). Right panel: Lyn was immunoprecipitated with anti-Lyn antibody and in vitro kinase assay (IVK) was performed. Bottom: Btk and Lyn were co-expressed with increasing dosage of PKCμ. Btk phosphorylation was analyzed as in (A). (C) Btk-Wt and Btk-E41K were co-expressed with Lyn and increasing dosage of PKCβ as in (A). Btk protein was immunoprecipitated and sequentially immunoblotted with anti-PY, anti-PY551, anti-PY223 and anti-Btk specific antibodies. (D) Btk and Lyn were co-expressed with high dosage PKCβ and cells were treated with increasing doses of Ro318425 for 30 min. Btk was immunoprecipitated and analyzed as in (A). Download figure Download PowerPoint In contrast to PKCβ, co-expression of an alternative serine/threonine kinase, Akt, had no significant effect on Btk phosphorylation (Figure 2B, left panel). In addition, we tested the possibility that PKC co-expression might indirectly affect Btk activation by altering Lyn activity (Figure 2B, right panel). PKCβ expression, however, did not significantly affect the in vitro kinase activity of Lyn under these conditions. Finally, we also tested whether PKCμ, previously implicated as a negative regulator of BCR signaling, could functionally substitute for PKCβ (Sidorenko et al., 1996). In our co-expression system, PKCμ had no significant effect on Btk (Figure 2B, bottom panel). Taken together, these data indicate that modulation of Btk by PKCβ is most likely a direct and specific effect. Btk activation requires sequential phosphorylation of two regulatory tyrosines (Y551 and Y223). The phosphorylation level of Btk Y551 is a relatively direct measure of Btk transphosphorylation by Src family kinases (Rawlings et al., 1996; Wahl et al., 1997). Phosphorylation of Btk Y223 represents subsequent autocatalytic activity following Btk activation (Park et al., 1996). Accordingly, we sought to determine the precise steps at which PKC intervenes to regulate Btk activity. Monoclonal antibodies specific for either phosphorylated Y551 or Y223 were used to measure the effect of PKCβ expression on Lyn-mediated Btk activation. In a similar experimental setup to Figure 2A, Btk was immunoprecipitated and sequentially immunoblotted with antibodies to phosphotyrosine, phospho-Y551, phospho-Y223 and Btk. As before, PKCβ dramatically decreased total Btk tyrosine phosphorylation. Interestingly, PKCβ down-modulated phosphorylation of Y551 and Y223, suggesting that both transphosphorylation and autophosphorylation events were attenuated by PKC-mediated inhibition (Figure 2C). We next repeated the experiment using the activated Btk mutant, Btk-E41K. PKCβ also effectively down-modulated both transphosphorylation and autophosphorylation of Btk-E41K (Figure 2C). This result further demonstrates that PKCβ is a potent negative regulator of Btk. The capacity of PKCβ to alter the enhanced membrane targeting phenotype of Btk-E41K is consistent with the pharmacological inhibitor data, suggesting that PKC may regulate Btk membrane targeting. Finally, we tested whether PKC kinase activity is required for its inhibitory effect. Btk, Lyn and a high dosage of PKCβ were co-expressed in the presence of increasing concentrations of the PKC inhibitor Ro318425. While Lyn-induced Btk tyrosine phosphorylation was significantly down-modulated with PKCβ co-expression, prior treatment with the PKC inhibitor abrogated the inhibitory effect of PKCβ (Figure 2D). Thus, PKCβ potently down-modulates Btk in a kinase activity-dependent manner. Moreover, PKCβ inhibits both transphosphorylation and autophosphorylation of Btk, most likely by affecting the events controlling Btk localization and Lyn-mediated activation of Btk. PKCβ phosphorylates Btk on a single serine residue in the Tec homology domain To elucidate the molecular basis of PKCβ-dependent inhibition of Btk activity, we utilized phosphopeptide mapping to identify potential PKCβ phosphorylation sites. NIH 3T3 cells were infected with vaccinia expressing either Btk alone or Btk and PKCβI. Following in vivo [32P]orthophosphate labeling, Btk was immunoprecipitated and analyzed by two-dimensional tryptic phosphopeptide mapping. Since the pattern of PKCβ-induced Btk phosphopeptides was identical with either wild-type Btk or kinase inactive Btk (Btk-K430R), this mutant was used to minimize the complexity of the phosphopeptide maps (Figure 4A and data not shown). Expression of Btk-K430R alone led to generation of two major tryptic spots (P1, P2), each with low but equal intensity. Co-expression of PKCβ resulted in a marked enhancement of P1 (Figure 3A). Phospho-amino acid analysis indicated that the peptide extracted from P1 contained only phosphoserine (data not shown). The presence of P1 in the absence of PKCβ suggests that endogenous PKC isoforms may basally phosphorylate this peptide. The presence of the second peptide, P2, which also contained only phosphoserine (data not shown), suggests that additional serine/threonine kinases may phosphorylate Btk on alternative peptides. Figure 3.Phosphopeptide mapping of the PKCβ-induced phosphorylation site on Btk. (A) Kinase inactive Btk (KI; Btk-K430R) was expressed in NIH 3T3 cells using recombinant vaccinia virus in the presence or absence of co-expressed PKCβ. Cells (1 × 107) were labeled with [32P]orthophosphate and Btk was immunoprecipitated, digested with trypsin, and tryptic peptides were separated by thin-layer electrophoresis at pH 1.9 followed by chromatography (see Materials and methods). (B) Identification of the domain phosphorylated by PKCβ. Upper panel: schematic representation of IgA cleavage sites within Btk. Bottom panel: partially purified Btk from (A) was incubated with IgA protease. The digested fragments were resolved by SDS–PAGE, and visualized by autoradiography (first panel) and western blot analysis using antibodies against N-terminal (middle) and C-terminal (third panel) regions of Btk. Download figure Download PowerPoint Figure 4.PKCβ phosphorylates S180 in the Tec-linker of Btk. (A) Phosphopeptide mapping analysis was performed on Btk-Wt and Btk-S180A, with or without the co-expression of PKCβ. As shown in Figure 3A, Btk-Wt displays two predominant phospho-tryptic fragments (P1, P2), and P1 is increased with PKCβ co-expression. The putative PKCβ phosphorylation site mutant Btk-S180A fails to induce P1, while P2 is still intact (panel 4). (B) Sequence alignment of murine Tec family kinases and Raja eglanteria Btk ortholog, skate PTK. Alignment of the conserved PKC consensus sequence is shown. (C) PKCβ-mediated negative regulation of Tec was examined in NIH 3T3 cells coordinately expressing Tec, Lyn and PKCβ. Tec was immunoprecipitated and blotted with anti-PY and reblotted with anti-Tec antibodies. Download figure Download PowerPoint To map the domain that is phosphorylated by PKCβ, IgA protease analysis was utilized. IgA protease cleaves Btk within the proline-rich region (Park et al., 1996), generating two major fragments consisting of the N-terminal PH/TH domain and a larger C-terminal fragment (Figure 3B). Autoradiography of the resulting nitrocellulose membrane showed predominant incorporation of [32P]orthophosphate in the 30 kDa (pp30) fragment corresponding to the N-terminal fragment of Btk (Figure 3B). In the presence of PKCβI, 32P incorporation in pp30 increased ∼5-fold, consistent with enhanced Btk phosphorylation (Figure 3B, bottom panel). The identity of pp30 was confirmed by immunoblotting with antibodies specific to either the N-terminal or the C-terminal regions of Btk. Taken together, these results reveal a predominant, PKCβ phosphorylation site within the N-terminal region of Btk (Figure 3B). The specific site of PKCβ-dependent Btk phosphorylation was determined by automated Edman degradation analysis of the phospho-tryptic peptide P1. Analysis revealed a major radioactive peak on the ninth cycle, suggesting that the ninth residue of the tryptic fragment was the site of phosphorylation (data not shown). Our analysis of the Btk/TH tryptic map indicated only one candidate site that clearly met these criteria. This deductive analysis indicated that PKCβ phosphorylates S180 in the region bisecting the Btk motif (BM) and the PRR of the TH domain. To demonstrate definitively that the correct site had been identified, we mutated the S180 to alanine (Btk-S180A), and repeated the phosphopeptide mapping. While Btk-Wt maintained the expected phosphopeptide profile, the S180A mutation resulted in a complete loss of basal and induced phosphorylation at P1 (Figure 4A, bottom panels). Because PKC isoforms frequently phosphorylate their substrates on multiple sites, we made additional mutations at the serine residues in the vicinity of S180, including S174 and S179. The phosphopeptide map of the triple mutant (Btk-S174, 179, 180A) was indistinguishable from that of the Btk-S180A mutant (data not shown). In addition, we have mutated several other potential sites to identify or eliminate additional phosphorylation events on Btk. Phosphopeptide mapping of S14A, S21A, S51A, S158A and S174A revealed an intact PKCβ phosphorylation profile (data not shown). These data clearly demonstrate that S180 represents the major phosphorylation site likely responsible for the negative regulation of Btk by PKCβ.
Referência(s)