The B-cell antigen receptor signals through a preformed transducer module of SLP65 and CIN85
2011; Springer Nature; Volume: 30; Issue: 17 Linguagem: Inglês
10.1038/emboj.2011.251
ISSN1460-2075
AutoresThomas Oellerich, Vanessa Bremes, Konstantin Neumann, Hanibal Bohnenberger, Kai Dittmann, He‐Hsuan Hsiao, Michael Engelke, Tim Schnyder, Facundo D. Batista, Henning Urlaub, Jürgen Wienands,
Tópico(s)Immune Cell Function and Interaction
ResumoArticle5 August 2011Open Access The B-cell antigen receptor signals through a preformed transducer module of SLP65 and CIN85 Thomas Oellerich Thomas Oellerich Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Vanessa Bremes Vanessa Bremes Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Konstantin Neumann Konstantin Neumann Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Hanibal Bohnenberger Hanibal Bohnenberger Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Kai Dittmann Kai Dittmann Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author He-Hsuan Hsiao He-Hsuan Hsiao Bioanalytical Mass Spectrometry Group, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany Search for more papers by this author Michael Engelke Michael Engelke Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Tim Schnyder Tim Schnyder Lymphocyte Interaction Laboratory, Cancer Research UK, London Research Institute, London, UK Search for more papers by this author Facundo D Batista Facundo D Batista Lymphocyte Interaction Laboratory, Cancer Research UK, London Research Institute, London, UK Search for more papers by this author Henning Urlaub Corresponding Author Henning Urlaub Bioanalytical Mass Spectrometry Group, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany Bioanalytics, Department of Clinical Chemistry, University Medical Center, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Jürgen Wienands Corresponding Author Jürgen Wienands Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Thomas Oellerich Thomas Oellerich Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Vanessa Bremes Vanessa Bremes Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Konstantin Neumann Konstantin Neumann Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Hanibal Bohnenberger Hanibal Bohnenberger Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Kai Dittmann Kai Dittmann Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author He-Hsuan Hsiao He-Hsuan Hsiao Bioanalytical Mass Spectrometry Group, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany Search for more papers by this author Michael Engelke Michael Engelke Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Tim Schnyder Tim Schnyder Lymphocyte Interaction Laboratory, Cancer Research UK, London Research Institute, London, UK Search for more papers by this author Facundo D Batista Facundo D Batista Lymphocyte Interaction Laboratory, Cancer Research UK, London Research Institute, London, UK Search for more papers by this author Henning Urlaub Corresponding Author Henning Urlaub Bioanalytical Mass Spectrometry Group, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany Bioanalytics, Department of Clinical Chemistry, University Medical Center, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Jürgen Wienands Corresponding Author Jürgen Wienands Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany Search for more papers by this author Author Information Thomas Oellerich1, Vanessa Bremes1, Konstantin Neumann1, Hanibal Bohnenberger1, Kai Dittmann1, He-Hsuan Hsiao2, Michael Engelke1, Tim Schnyder3, Facundo D Batista3, Henning Urlaub 2,4 and Jürgen Wienands 1 1Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Göttingen, Germany 2Bioanalytical Mass Spectrometry Group, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany 3Lymphocyte Interaction Laboratory, Cancer Research UK, London Research Institute, London, UK 4Bioanalytics, Department of Clinical Chemistry, University Medical Center, Georg-August-University Göttingen, Göttingen, Germany *Corresponding authors: Institute of Cellular and Molecular Immunology, Georg-August-University Göttingen, Humboldtallee 34, Göttingen 37073, Germany. Tel.: +49 551 39 5821; Fax: +49 551 39 5843; E-mail: [email protected] Mass Spectrometry Group, Max Planck Institute for Biophysical Chemistry, Am Fassberg 11, 37077 Göttingen, Germany. Tel.: +49 551 201 1060; Fax: +49 551 201 1197; E-mail: [email protected] The EMBO Journal (2011)30:3620-3634https://doi.org/10.1038/emboj.2011.251 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Spleen tyrosine kinase Syk and its substrate SLP65 (also called BLNK) are proximal signal transducer elements of the B-cell antigen receptor (BCR). Yet, our understanding of signal initiation and processing is limited owing to the incomplete list of SLP65 interaction partners and our ignorance of their association kinetics. We have now determined and quantified the in vivo interactomes of SLP65 in resting and stimulated B cells by mass spectrometry. SLP65 orchestrated a complex signal network of about 30 proteins that was predominantly based on dynamic interactions. However, a stimulation-independent and constant association of SLP65 with the Cbl-interacting protein of 85 kDa (CIN85) was requisite for SLP65 phosphorylation and its inducible plasma membrane translocation. In the absence of a steady SLP65/CIN85 complex, BCR-induced Ca2+ and NF-κB responses were abrogated. Finally, live cell imaging and co-immunoprecipitation experiments further confirmed that both SLP65 and CIN85 are key components of the BCR-associated primary transducer module required for the onset and progression phases of BCR signal transduction. Introduction Two biochemical principles characterize initiation and processing of antigen-induced signalling in lymphocytes. First, antigen receptor-proximal effector proteins assemble into multimeric complexes in a rapid and transient manner. Second, the formation of these early signalosomes is accompanied by subcellular relocation of one or more of the molecules involved, most prominently from the cytosol into signalling-competent membrane microdomains. The activated B-cell antigen receptor (BCR) recruits the tandemly arranged Src homology (SH) 2 domains of spleen tyrosine kinase Syk (Zioncheck et al, 1988; Hutchcroft et al, 1991, 1992) to a phosphorylated immunoreceptor tyrosine-based activation motif (ITAM) in Igα (CD79a) and Igβ (CD79b) (for review, see Engels et al, 2008 and Geahlen, 2009). Syk and its phospho-ITAM-mediated activation provide a common trigger for many B-cell signalling cascades including Ca2+ mobilization (Engelke et al, 2007; Scharenberg et al, 2007), induction of MAP kinase pathways (Hashimoto et al, 1998; Jiang et al, 1998) and nuclear translocation of transcriptionally active NF-κB dimers (Schulze-Luehrmann and Ghosh, 2006). The proximal and dominant Syk substrate in BCR-stimulated cells is the SH2 domain-containing leukocyte adaptor protein of 65 kDa (SLP65; Wienands et al, 1998) alternatively named BLNK (Fu et al, 1998) or BASH (Goitsuka et al, 1998). Nine tyrosine residues become rapidly phosphorylated upon BCR activation (Oellerich et al, 2009). Together with the complex involvement of 29 serine and 6 threonine residues intricate phosphorylation patterns are generated which enable SLP65 to act upstream for triggering the signal as well as downstream during selective processing of the BCR signal (Oellerich et al, 2009). Hence, SLP65 is a genuine signal integrator which is as mandatory as Syk for normal B lymphopoiesis and humoral immunity (Jumaa et al, 1999; Minegishi et al, 1999; Pappu et al, 1999; Hayashi et al, 2000; Xu et al, 2000). Fulfilling the signal integration function requires differential networking of SLP65 with numerous interaction partners. This has been worked out to a great detail for the SLP65-mediated nucleation of the Ca2+ initiation complex. Three distinct phosphotyrosine residues in SLP65 serve as docking sites for the SH2 domains of phospholipase C-γ2 (PLC-γ2) and Bruton's tyrosine kinase (Btk) (Hashimoto et al, 1999; Ishiai et al, 1999; Su et al, 1999; Chiu et al, 2002). Within this trimolecular arrangement Btk phosphorylates and thereby activates PLC-γ2. It remained unclear how the Ca2+ initiation complex anchors inducibly to the plasma membrane to provide PLC-γ2 with access to its substrate phosphatidylinositol-4,5 bisphosphate, which upon hydrolysis yields the second messengers inositol-trisphosphate and diacylglycerol. Mutational analysis revealed that membrane recruitment of SLP65 requires both the N-terminal basic region and the C-terminal SH2 domain (Kohler et al, 2005; Abudula et al, 2007). A ligand for the basic effector region has not yet been identified. Binding of the SH2 domain to a non-ITAM phosphotyrosine in Igα places SLP65 in close vicinity to ITAM-bound Syk and facilitates SLP65 phosphorylation (Engels et al, 2001; Kabak et al, 2002), which in turn augments B-cell responses to T cell-independent antigens (Patterson et al, 2006). Nonetheless, Syk can phosphorylate SLP65 even in the absence of the non-ITAM tyrosine. It is unknown at what subcellular location this crucial step of signal initiation occurs and how the enzyme–substrate interaction is accomplished. However, genetic reconstitution experiments from our group provided evidence that the initiation of signalling requires protein complexes that are already formed before BCR stimulation (Wienands et al, 1996). Collectively, these examples suggest the existence of so far unidentified effector proteins of BCR signal initiation and processing. To fill the gaps in our understanding of proximal BCR signalling, we have now determined the interactome of SLP65 in resting and stimulated B cells by quantitative mass spectrometry. Additional monitoring of the association kinetics revealed a class of steady ligands whose preformed association to SLP65 in resting cells remained unaltered upon BCR activation. One of those permanent SLP65 companions was the Cbl-interacting protein of 85 kDa (CIN85) also named Ruk or SH3KBP1. Steady complex formation between CIN85 and SLP65 orchestrated Syk-mediated phosphorylation and membrane translocation of SLP65 and globally affected the composition of the SLP65 interactome. Mutational analysis showed that CIN85 is a novel key component of the BCR-associated transducer complex. Results SILAC-based interactome analysis provides a library for the SLP65 signalling network To search for missing links of the primary BCR transducer complex, we determined the interactome of SLP65 using stable isotope labelling with amino acids in cell culture (SILAC) in combination with mass spectrometric analysis of affinity-purified protein complexes (Ong et al, 2002; Neumann et al, 2009; Selbach et al, 2009). Therefore, slp65−/− DT40 B cells were reconstituted with an SLP65 variant harbouring an N-terminal One Strep tag that was expressed in almost identical amounts compared with endogenous SLP65 in wild-type cells (see Figure 1A). Cells expressing tagged SLP65 were cultured in SILAC medium containing lysine and arginine amino acids that have incorporated ‘heavy’ isotopes of carbon and nitrogen (13C and 15N). As negative control, DT40 cells expressing non-tagged SLP65 were cultured in the presence of lysines and arginines encompassing carbon 12C and nitrogen 14N, so-called ‘light’ isotopes. Proteins from the two culture conditions contained either ‘heavy’ or ‘light’ lysines and arginines (Supplementary Figure S1). Accordingly, the two culture conditions confer distinct molecular masses on the cellular proteins synthesized; and thus, proteins derived from ‘heavily’ and ‘lightly’ labelled cells can be distinguished by mass spectrometry. For elucidation of the SLP65 interactome in the absence of BCR stimulation, the differentially labelled cells were lysed without further treatment. Proteins were affinity purified with a Strep-tactin column, pooled at a 1:1 ratio and hydrolysed with endoproteinase trypsin. Peptides were identified by liquid chromatography (LC)-coupled tandem mass spectrometry (MS/MS) and allocated to the corresponding protein by database search. Relative quantification of all sequenced peptides was performed using MaxQuant software (Cox et al, 2009) and is shown in Supplementary Table 1. An at least five-fold enrichment of heavy versus light peptides was considered to mark those proteins that were specifically co-purified with One Strep-tagged SLP65 from resting cells. To identify inducible SLP65 interactors, we compared affinity-purified SLP65 complexes derived from ‘lightly labelled’ resting cells with those from ‘heavily labelled’ cells stimulated through their BCR for 2, 5, 10 or 20 min. Figure 1.Association kinetics of SLP65 ligands. (A, top part) Immunoblot analysis with anti-SLP65 or anti-actin antibodies (upper and lower panels, respectively) of SLP65-negative DT40 B-cell mutants (lane 1), wild-type DT40 cells (wt, lane 2) and mutants cells that were reconstituted with One Strep-tagged SLP65 (lane 3) and provide the basis for all mass spectrometric ligand identification and quantification analysis. (A, bottom part) Schematic representation of a two-fold triple SILAC approach for relative quantification of individual SLP65 ligands under different stimulation conditions. DT40 B cells expressing One Strep-tagged SLP65 were cultured in SILAC medium containing the differentially labelled amino acids Lys+0/Arg+0 or Lys+4/Arg+6 or Lys+8/Arg+10. In a first approach, cells were left untreated or stimulated through their BCR for 2 or 20 min, respectively. In a complementary experiment to obtain continuous kinetics, cells were left untreated or BCR stimulated for 5 or 10 min. From each approach, the affinity-purified SLP65 signalosomes were mixed at a 1:1:1 ratio. Ligands were identified and relatively quantified by LC-MS/MS using MaxQuant software. Note that not all interaction kinetics of SLP65 ligands could be measured owing to the increased sample complexity in our combined triple SILAC approach that reveals certain technical limitations of the MS instrument to detect low-abundant peptide pairs of ligands that only transiently or weakly interact with SLP65, for example, Btk and Syk. (B–D) The amounts of individual SLP65 ligands measured under distinct stimulation conditions were normalized according to the association quantity observed in resting cells (determined in both approaches). Subsequently, the mean values of the three independent doublet experiments were combined to a single time course of five different stimulation conditions as both sets contained quantitative interaction values from resting B cells which allowed for normalization of the data facilitated by the MaxQuant software. Based on this analysis, SLP65 ligands were classified as being early and transient (B), late and sustained (C) or steady (D). For detailed data sets and statistics, see Supplementary Table 2. (E) Human Ramos B cells left untreated (lanes 1 and 2) or BCR stimulated for 2, 5, 10 or 20 min (lanes 3–6) were lysed and subjected to immunoprecipitations with anti-SLP65 antibodies (lanes 2–6) or isotype-matched control antibodies (lane 1, ctrl). Obtained proteins were analysed by immunoblotting with antibodies directed against CD2AP, Dok-3, CapZ, Sek1/MAPKK4, PLC-γ2 and Grb2 (upper to lower panel, respectively). (F) Lysates of BCR-negative J558L B cells were subjected to immunoprecipitations using antibodies to SLP65 (lane 2) or CD2AP (lane 4) or isotype-matched control antibodies (lanes 1 and 3). Obtained proteins were analysed by immunoblotting with antibodies recognizing CIN85 or SLP65 (lanes 1–2, upper and lower left panels, respectively), or SLP65 and CD2AP (lanes 3–4, upper and lower right panels, respectively). Relative molecular masses of marker proteins are indicated on the left in kDa. Download figure Download PowerPoint In addition to previously reported SLP65 binders, a large number of unknown ligands were identified. Altogether, 29 proteins constituted the SLP65 interactome (Table I). Only a few ligand interactions existed in resting B cells, most notably to UNC119b implicated in the regulation of protein tyrosine kinase activity (Cen et al, 2003; Gorska et al, 2004) and to the SH3 domain-containing adaptor proteins Grb2, CIN85 and CD2-associated protein (CD2AP). CIN85 and CD2AP share a similar domain architecture encompassing three N-terminal SH3 domains followed by a proline-rich region and a C-terminal coiled-coil domain (Dustin et al, 1998; Take et al, 2000; Dikic, 2002). Since both adaptor proteins associate firmly and constitutively with actin-capping proteins Z (CapZ; Hutchings et al, 2003), the detection of CapZ isoforms as part of the SLP65 interactome in resting as well as stimulated B cells might be due to indirect co-purification of these cytoskeleton regulators. However, SLP65 appeared to possess several links to the cytoskeleton as also talin, profilin and Nck all implicated in actin filament dynamics (Buday et al, 2002; Le Clainche and Carlier, 2008) were identified as inducible ligands. Consistent with the role of SLP65 as upstream regulator of B-cell signalling cascades (Engelke et al, 2007), the inducible interactome accommodated positive- and negative-regulatory components of BCR-induced Ca2+ mobilization and MAP kinase activation. Finally, the C-type lectin 17A (CLEC17A) may contribute to SLP65 membrane recruitment as its cytoplasmic tail harbours binding sites for SH2 and SH3 domains. A glycan-binding protein with similar overall structure is also found in mammals and called prolectin (Graham et al, 2009). Table 1. The interactome of SLP65 Ligand IPI number Function References In resting B cells Grb2 00576615 Adaptor protein This study (Fu et al, 1998; Wienands et al, 1998) CIN85 00601097 Adaptor protein This study (Watanabe et al, 2000) CD2AP 00583892 Adaptor protein This study CapZ-α1 00583945 Cytoskeleton regulator This study CapZ-α2 00582081 Cytoskeleton regulator This study CapZ-β1/2 00591422 Cytoskeleton regulator This study Similar to Unc119b 00584927 Src kinase regulator This study Hsp70 00582091 Chaperone This study Hsp105 00590633 Chaperone This study Ig heavy chain-binding protein 00590375 Chaperone This study In activated B cells Igα (CD79A) 00591677 BCR component This study (Engels et al, 2001) CLEC17A 00599471 Transmembrane lectin This study Grb2 00576615 Adaptor protein This study (Fu et al, 1998; Wienands et al, 1998) CIN85 00601097 Adaptor protein This study (Watanabe et al, 2000) CD2AP 00583892 Adaptor protein This study Dok-3 00845156 Adaptor protein This study Nck2 00599713 Cytoskeleton adaptor This study (Fu et al, 1998) CapZ-α1 00583945 Cytoskeleton regulator This study CapZ-α2 00582081 Cytoskeleton regulator This study CapZ-β1/2 00591422 Cytoskeleton regulator This study Talin 00586709 Cytoskeleton regulator This study Profilin 00600305 Cytoskeleton regulator This study Similar to Unc119b 00584927 Src kinase regulator This study Lyn 00579414 Protein tyrosine kinase This study Syk 00603504 Protein tyrosine kinase This study (Abudula et al, 2007; Kulathu et al, 2008) Btka 00595247 Protein tyrosine kinase This study (Hashimoto et al, 1999; Su et al, 1999) PLC-γ2 00587921 Phospholipase This study (Fu et al, 1998) VAV3 00578220 GTP exchange factor This study (Fu et al, 1998; Wienands et al, 1998) Sek1/MAPKK4 00594829 MAP kinase This study Protein phosphatase 2 00651162 Ser/Thr phosphatase This study AIF (programmed cell death 8) 00601063 Apoptosis This study Ig heavy chain-binding protein 00590375 Chaperone This study Hsp70 00582091 Chaperone This study Hsp105 00590633 Chaperone This study WD repeat domain 81 00602009 Unknown This study CCH domain-containing 4 00574478 Unknown This study Hypothetical 00602230 Unknown This study Hypothetical 00597453 Unknown This study Hypothetical 00813608 Unknown This study Two independent SILAC-based LC-MS/MS analyses were performed for resting DT40 B cells and cells stimulated through their BCR for 2, 5, 10 and 20 min. Detailed statistics including total number of all identified and quantified peptides as well as their significance and variability scores are shown in Supplementary Table 1. Listed proteins showed an at least five-fold enrichment of heavy versus light peptides. a The known ligand Btk co-migrated with the dominating protein band of Strep-tagged SLP65 and was thus less well detectable in this set-up but identified in a separate approach with GFP-tagged SLP65. The association kinetics of individual SLP65 interactors defines distinct ligand classes Our description of the SLP65 interactome provided information about the ligands’ identities and whether the corresponding complexes were dependent on BCR activation. However, this type of analysis does not take account of the dynamic nature of signal transduction processes as it lacks information about association kinetics. Thus, we quantified individual SLP65 ligands at different time points of BCR stimulation using a combined triple SILAC/MS approach with B cells that all expressed One Strep-tagged SLP65. Note that tagged SLP65 was expressed by the reconstituted cells in similar amounts compared with endogenous SLP65 in wild-type cells (Figure 1A). Cells were metabolically labelled in ‘light’ (Lys+0/Arg+0), ‘medium’ (Lys+4/Arg+6) or ‘heavy’ (Lys+8/Arg+10) SILAC medium (see Materials and methods for details) and either left untreated or stimulated for 2 and 5, or 10 and 20 min, respectively (see Figure 1A). Affinity-purified SLP65 complexes were mixed at a 1:1:1 ratio (samples 0, 2, 20 and 0, 5, 10 min) and identified as well as quantified as described above. The relative association kinetics uncovered the dynamic nature of SLP65 complex formations and revealed three ligand classes. The class of early but transient interactors (Figure 1B) contained regulatory components of the Ca2+ flux pathway as well as Nck, Grb2 and CLEC17A. A second class of ligands characterized by a late and sustained association (Figure 1C) was represented by CapZ isoforms and the serine/threonine protein phosphatase 2 (PP2). Finally, we detected steady ligands whose association to SLP65 was almost unaffected by BCR engagement in that it was already ‘preformed’ before BCR activation and remained unchanged thereafter (Figure 1D). This class of constant SLP65 interactors contained CD2AP and CIN85. No BCR-induced loss of binding was observed. Co-immunoprecipitation experiments using human and murine B cells (Figure 1E and F) corroborated our proteomic results obtained in the avian system. Moreover, the constitutive association between SLP65 and CIN85/CD2AP adaptors in BCR-negative J558L B cells (Figure 1F, left and right panels, respectively) showed that formation of these steady complexes was also independent of so-called ‘tonic’ BCR signalling in the absence of ligand (Monroe, 2006). In summary, quantitative monitoring of the association kinetics distinguished various classes of SLP65 ligands. Early interaction partners became rapidly engaged upon BCR stimulation and may generally reflect master regulators of BCR-proximal signalling pathways. Late interaction partners may be involved in signal processing. The function of the preformed and constant interactors was less obvious as they may act upstream or downstream of SLP65 or both. Steady ligands CIN85 and CD2AP bind atypical SH3-binding motifs of SLP65 A stimulation-independent assembly of BCR effector proteins into preformed signalling complexes has been suggested as requisite for antigen-induced tyrosine phosphorylation of kinase substrates including SLP65 itself (Wienands et al, 1996, 1998). The identity and architecture of the preformed signalosomes remained unclear ever since. The steady SLP65 ligands CIN85 and/or CD2AP were candidate components as their multifunctionality has been implicated in several aspects of cell surface receptor biology (Dikic, 2002). We wanted to test this possibility by mutational analysis and mapped the interaction sites between SLP65 and CIN85/CD2AP. SLP65 accommodates three evolutionary conserved proline/arginine-based docking motifs (Figure 2A) that match the previously reported atypical recognition consensus PXXXPR of the three SH3 domains of CIN85 and CD2AP (Kowanetz et al, 2003; Kurakin et al, 2003). Exchange of the critical arginine residue to alanine (R-to-A) within the three binding motifs abolished co-immunoprecipitation of SLP65 with both CIN85 and CD2AP from resting and BCR-activated B cells (Figure 2B, left panel). Consistently, glutathione-S transferase (GST) fusion proteins encompassing the three SH3 domains of CD2AP or CIN85 affinity-purified wild-type SLP65 but not the triple R-to-A variant (Figure 2B, middle panel and data not shown). Hence, atypical proline/arginine motifs in SLP65 and the SH3 domains of CIN85/CD2AP were necessary and sufficient for steady complex formation in vivo. Binding studies with recombinantly expressed SLP65 and GST fusion proteins encompassing the triple SH3 domains of CIN85 confirmed that the interaction is direct (Figure 2B, right panel). Furthermore, the interaction is highly selective as shown by SILAC-based mass spectrometric determination of all putative PXXXPR ligands that exist in B cells. Figure 2C shows that biotinylated peptides encompassing the central proline/arginine motif of SLP65 almost exclusively purified CIN85 and CD2AP along with their intimate CapZ binding partners from B-cell lysates cultured in heavy SILAC medium. B cells that were cultured in light SILAC medium and subjected to affinity purification with a mutant form of the peptide served as control and for normalization of the binding scores. The only SH3 domain-containing protein that in addition to CIN85 and CD2AP showed a significant yet four-fold weaker binding score was Grb2. Hence, dependent on protein availability these sites may to some extent also bind Grb2. Collectively, these experiments demonstrated that assembly of the preformed SLP65 signalosome was based on the exquisite binding specificity of the atypical PXXXPR motifs which selectively recruited CIN85 and CD2AP. Figure 2.Atypical SH3 binding motifs of SLP65 selectively and directly recruit the steady ligands CIN85 and CD2AP. (A) Domain architecture of SLP65 and amino-acid sequences of the three atypical proline/arginine motifs (indicated by a–c) in human and murine SLP65. (B, top part) Domain architecture of CIN85/CD2AP adaptors. (B, bottom parts) SLP65-deficient DT40 B cells (slp65−/−) were reconstituted with green fluorescence protein (GFP)-tagged forms of either human wild-type SLP65 or a mutant version harbouring R-to-A amino-acid exchanges in all three atypical SH3 domain-binding motifs of the PXXXPR type (R49, 248 and 313A). Cells were left untreated (0′) or BCR stimulated for the indicated times in minutes (left and middle panels), and subjected either to immunoprecipitation (IP) with anti-GFP antibodies followed by immunoblot analysis of obtained proteins with antibodies recognizing CIN85, CD2AP or SLP65 (left panel, upper, middle and lower blots, respectively), or to affinity purifications using glutathione-S transferase (GST) fusion proteins encompassing the three SH3 domains of CD2AP (GST–CD2AP(SH3)3), or GST alone as negative control, followed by immunoblotting with antibodies recognizing SLP65 or GST (middle panel, upper and lower blots, respectively). Direct protein–protein interactions were tested in vitro by mixing recombinantly expressed HIS-tagged SLP65 with immobilized GST–CIN85(SH3)3, or immobilized GST alone as control, followed by immunoblot analysis of bound proteins with anti-SLP65 or anti-GST antibodies (right panel, upper and lower blots, respectively). Relative molecular masses of marker proteins are indicated on the left in kDa. (C) DT40 B cells were metabolically labelled in heavy (Lys+8/Arg+10) SILAC medium and lysates were incubated with biotinylated peptides encompassing the central PXXXPR motif of human SLP65 (KSPPPAAPSPLPRAGKKPT). For control, DT40 B cells were cultured in light (Lys+0/Arg+0) SILAC medium and their lysates were incubated with biotinylated peptides, in which prolines at positions 10 and 13 were exchanged for alanine (KSPPPAAPSALPAAGKKPT). Purified proteins were mixed at a 1:1 ratio, digested with trypsin and identified by quantitative LC-MS/MS analysis using MaxQuant software. All identified proteins are plotted accor
Referência(s)