| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
R-Induced Actin Polymerization1
,
,¶,||,,¶,||,,¶,||
* Department of Biochemistry, University of Lausanne, BIL Biomedical Research Center, Epalinges, Switzerland;
Department of Clinical Pathology, Centre Medical Universitaire, Geneva, Switzerland;
Institut Cochin, Département de Biologie Cellulaire, Paris, France;
Institut National de la Santé et de la Recherche Médicale Unité 567, Paris, France;
¶ Centre National de la Recherche Scientifique, Unité Mixte de Recherche, Paris, France;
|| Université Paris 5, Faculté de Médecine René Descartes, Paris, France;
# Institute for Research in Biomedicine, Bellinzona, Switzerland,
** Third Medical Department, Technical University of Munich, Klinikum Rechts der Isar, Munich, Germany; and
The Campbell Family Institute for Breast Cancer Research and Ontario Cancer Institute, University Health Network, University of Toronto, Toronto, Ontario, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
B activation and cytokine production. Bcl10 is a phosphoprotein, but the physiological relevance of this posttranslational modification remains poorly defined. In this study, we report that Bcl10 is rapidly phosphorylated upon activation of human T cells by PMA/ionomycin- or anti-CD3 treatment, and identify Ser138 as a key residue necessary for Bcl10 phosphorylation. We also show that a phosphorylation-deficient Ser138/Ala mutant specifically inhibits TCR-induced actin polymerization yet does not affect NF-B activation. Moreover, silencing of Bcl10, but not of caspase recruitment domain-containing MAGUK protein-1 (Carma1) induces a clear defect in TCR-induced F-actin formation, cell spreading, and conjugate formation. Remarkably, Bcl10 silencing also impairs Fc
R-induced actin polymerization and phagocytosis in human monocytes. These results point to a key role of Bcl10 in F-actin-dependent immune responses of T cells and monocytes/macrophages. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
R-induced NF-B activation and cytokine secretion (1, 2, 3). Bcl10 may also be important for the progression of particular forms of B cell lymphoma. Overexpression and nuclear translocation of Bcl10 has been reported to occur frequently in mucosa-associated lymphoid tissue (MALT)6 lymphomas and in NK/T cell lymphomas (4, 5). Moreover, activated B cell-type diffuse large B cell lymphomas were recently shown to critically depend on Bcl10 expression for their survival (6).
Despite these recent insights into the biological relevance of Bcl10, the molecular function of Bcl10 is only partly understood. Upon TCR triggering, Bcl10 is recruited to the TCR/CD3 complex by a caspase recruitment domain (CARD)-dependent interaction with its binding partner CARD-containing MAGUK protein-1 (Carma1), and activates the NF-B-regulating IB kinase (IKK) complex through the paracaspase protein Malt1 (MALT lymphoma translocation protein-1) (7, 8, 9). The C-terminal portion of Bcl10, which shares no obvious sequence homology to any other known protein, is rich in Ser and Thr residues and is subject to phosphorylation events that are observed upon Bcl10 overexpression, but also specifically triggered upon Ag receptor engagement or TNF-
stimulation (10, 11, 12, 13, 14, 15, 16, 17). Recently, Wegener et al. (18) showed that IKK
-mediated phosphorylation of Bcl10 plays a negative regulatory role in T cell activation by interfering with the Bcl10-Malt1 interaction. It remains unclear, however, whether Bcl10 phosphorylation specifically regulates NF-B activation or also as yet undefined biological functions of Bcl10.
Stimulation of surface receptors of immune cells triggers not only changes in gene expression but also morphological changes via the reorganization of the actin cytoskeleton, which contributes to efficient cellular activation. In immune cells, ligand-induced actin polymerization is initiated by the activation of receptor-proximal tyrosine kinases, which in turn control the activity of GDP/GTP exchange factors (GEFs) that regulate the activity of Rho family GTPases (19). Among these, Cdc42 and Rac proteins are proposed to control actin cytoskeletal changes via regulation of Wiskott-Aldrich syndrome protein (WASP) and the Abi/Wave complex, respectively, which in turn regulate the activity of the Arp2/3 actin nucleation complex (20, 21, 22, 23, 24). To date, a possible role of Bcl10 in the regulation of the actin cytoskeleton has not been explored.
In this study, we reveal a novel function of Bcl10 in the control of TCR- and FcR-induced actin polymerization. We show that Bcl10 is phosphorylated rapidly upon TCR engagement, and identify Ser138 as a potential site of T cell activation-induced phosphorylation. Moreover, we show that Ser138-dependent Bcl10 phosphorylation is critical for TCR-induced actin polymerization, but not necessary for NF-B activation by Bcl10. Importantly, Bcl10 silencing also affects FcR-induced actin polymerization and phagocytosis. Together, these data identify a role for Bcl10 in actin polymerization-dependent cellular immune responses.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
293T and Jurkat cells (J77 clone 20; a gift from O. Acuto, Pasteur Institute, Paris, France) were grown as described previously (14, 25). The monocyte cell line THP-1 (American Type Culture Collection) was grown in RPMI 1640 medium supplemented with 10% FCS at 37°C in 5% CO2. Human CD4+ T cells were isolated from Sepacell RS-2000 filters (donated by the Centre de Transfusion Sanguine du Centre Hôpitalier Universitaire Vaudois) by Ficoll-Paque procedure (Pharmacia Biotech) and positive selection with magnetic anti-human CD4 microbeads on miniMACS columns (Miltenyi Biotec). Mouse primary T cells were isolated from lymph nodes of Bcl10-deficient or littermate control mice (1) by negative selection using anti-B220-coated MACS beads (Miltenyi Biotec). Mice were bred in the animal facilities of the Technical University of Munich Medical School, and experiments with animals were conducted in accordance with the German/institutional guidelines for animal care.
Activation of cells
For short-term activation, T cells were resuspended in RPMI 1640 at 107 cells/200 µl and stimulated during the indicated times using either PMA (10 ng/ml) and ionomycin (1 µM), anti-human CD3 (10 µg/ml OKT3; a gift from S. Valitutti, Institut Claude de Préval, Toulouse, France), or anti-human CD3 (10 µg/ml TR66; Apotech) together with anti-CD28 (10 µg/ml CD28.2; Immunotech) followed by cross-linking with goat anti-mouse IgG1 (5 µg/ml; Southern Biotechnology Associates), or with the chemically synthesized chemokine CXCL12/SDF-1 (100 nM; a gift from I. Clark-Lewis, Biomedical Research Center, University of British Columbia, Vancouver, British Columbia, Canada). In some experiments, Jurkat cells were preincubated with 500 nM GFX (bisindoleylmaleimide hydrochloride; Alexis) or solvent control (DMSO) in RPMI 1640 for 60 min before stimulation. Murine T cells were stimulated with anti-mouse CD3 (10 µg/ml 145-2C11; BD Pharmingen).
For FcR stimulation, THP-1 cells were resuspended in RPMI 1640 without serum at 106 cells/50 µl. Upon addition of 50 µl of IgG-SRBC at (5 x 106 cells/50 µl), the cell mixture was pulse-centrifuged to favor cellular interactions and incubated for the indicated times at 37°C before fixation and analysis of F-actin content (see below).
Expression vectors
Expression vectors for Bcl10, Carma1, and DN-IB have been described previously (13, 14, 25). Ser/Ala point mutants of both human and murine Bcl10 were obtained by a standard double PCR approach and subcloning of sequenced PCR products into expression vectors derived from pCR-3 (Invitrogen Life Technologies) to yield expression constructs with an N-terminal FLAG- or vesicular stomatitis virus glycoprotein tag. For lentiviral expression of Bcl10, constructs were subcloned into the lentiviral vector pRDI_292 (a gift from R. Iggo, University of St. Andrews, Scotland) that allows expression of constructs under the EF1 promoter.
Immunoprecipitation and Western blotting
Cell lysis, immunoprecipitation, phosphatase treatment, and Western blotting were performed as described previously (14, 25). Five microliters of mouse anti-FLAG agarose (Sigma-Aldrich), 1 µg of goat anti-Bcl10 Ab (N-20; Santa Cruz Biotechnology), or 1 µg of anti-phosphotyrosine (4G10; Upstate Biotechnology) previously bound to protein G Sepharose (Amersham Biosciences) were used for immunoprecipitation. Primary Abs used for Western blotting were anti-FLAG (M2; Sigma-Aldrich), affinity-purified polyclonal rabbit Ab (AL114) recognizing the N terminus of mouse and human Bcl10 (14), rabbit anti-Bcl10 (H197; Santa Cruz Biotechnology), rabbit anti-Carma1 antiserum AL 220 (25), mAb anti-Malt1 (a gift from V. Dixit, Genentech), mAb anti-tubulin (T-5168; Sigma-Aldrich), mAb anti-lamin A/C (SC-7292; Santa Cruz Biotechnology), mAb anti-phosphotyrosine (4G10; Upstate Biotechnology), rabbit anti-Vav (C-14; Santa Cruz Biotechnology), mAb anti-phospho-ERK (Sigma-Aldrich), mAb anti-P-IB (5A5; Cell Signaling Technology), and rabbit anti-P-p38 (BioSource International).
In vivo 32P labeling
For in vivo labeling experiments, 80% confluent 293T cells were transfected with PolyFect (Qiagen) following the manufacturer’s instructions. Transfected 293T cells (3 x 105) were washed twice with phosphate-free DMEM and incubated in this medium for 1 h to further deplete the internal phosphate store. Then, [32P]orthophosphate (100 µCi/ml; Amersham Biosciences) was added and the cells were incubated for 2 h.
Phospho-amino acid analysis
Radioactive bands corresponding to Bcl10 were excised from the gel, washed extensively with H2O, and dried in a Speed Vac (Heraeus). The proteins trapped in the gel slices were hydrolyzed in 1 ml of 6 M HCl for 2 h at 95°C under reduced pressure. Amino acids released from the gel slices were recovered with the HCl and dried under vacuum. Samples were resuspended in 5 µl of buffer 1 (8% acetic acid, 2% formic acid) supplemented with 10 µg each of phosphoserine, phosphothreonine, and phosphotyrosine and spotted on phosphocellulose TLC sheets (Sigma-Aldrich). Electrophoresis of the TLC sheets was performed in 8% acetic acid, 2% formic acid at 400 V for 105 min, and continued (same dimension) in 1% pyridine, 10% acetic acid at 400 V for 180 min. Phospho-amino acid standards were visualized with ninhydrine. The radioactivity on the TLC sheet was measured with a Phosphoimager (Storm; APB).
Two-dimesional analysis of phosphorylated Bcl10
Proteins were extracted from washed immunoprecipitates by 100 µl of solubilization buffer S containing 9.0 M Urea, 4% (w/v) CHAPS (Sigma-Aldrich), 65 mM 1,4-dithio-DL-threitol, 0.8% (v/v) Resolytes 4–8 (BDH), 4 mM Tris base (pH 10.5), and 0.001% (w/v) bromophenol blue and incubation at room temperature for 60 min, with vortexing every 10 min. The beads were sedimented by centrifugation, and 60 µl of the supernatant were loaded on 7-cm long immobilized pH gradient gel strips (pH 4–7) (Amersham Biosciences) that had been rehydrated with buffer R (8.0 M Urea, 2% (w/v) CHAPS, 18 mM 1,4-dithio-DL-threitol, 0.8% (v/v) Resolytes 4–8, and 0.001% (w/v) bromophenol blue). In some experiments, cells were directly lysed in buffer S (5 x 106 cells/200 µl) and 150 µl of sonicated and centrifuged extract was loaded on immobilized pH gradient gel strips. Isoelectric focusing was performed at a maximum voltage of 3,500 V, until a volt*hour count of 35,000 was reached. Equilibration and transfer to the second dimension were done as described previously (26). SDS-PAGE for the second dimension was performed on 11% polyacrylamide gels.
NF-B reporter assays
For luciferase assays, Jurkat cells were cotransfected with 4 µg of expression vectors and 1 µg of small interfering RNA (siRNA), as indicated, together with 3 µg of an NF-B luciferase reporter plasmid (NF-Bluc; a gift from V. Jongeneel, Institut Suisse de Bioinformatique, Lausanne, Switzerland) and the phRLTK Renilla luciferase reporter plasmid (1 µg; Promega). Twenty-four hours after electroporation, cells were harvested and stimulated or not with PMA (10 ng/ml) and ionomycin (1 µM) for 24 h or with plate-bound anti-CD3 (OKT3, 10 µg/ml) and anti-CD28 (CD28.2, 10 µg/ml) Ab for 10 h. After stimulation, cells were washed twice with PBS and lysed in 50 µl of passive lysis buffer (Promega). Aliquots of cell lysates (10 µl) were mixed with 50 µl of dual luciferase assay reagent (Promega), and the luciferase activity was determined using a TD 20/20 luminometer (Turner Designs).
Transfection and transduction of Jurkat cells
Transient transfection of Jurkat cells (5 x 106 cells/cuvette) with expression plasmids and/or synthetic siRNA was performed with the Nucleofector system (Amaxa; program O17) according to the manufacturer’s instructions, yielding transfection efficiencies that were routinely between 60 and 70%. For transient gene silencing, siRNA specific for human Bcl10 (sc-29793; Santa Cruz Biotechnology) or negative control (nonsilencing) siRNA (1022076; Qiagen) were used (1 µg/cuvette). For stable siRNA expression, oligonucleotides encoding short hairpin RNAs (shRNAs) targeting nt 143–161 (GTAGAGAAGACACTGAAGA) of the human Bcl10 coding sequence, nt 457–475 of the human Carma1 5'UTR (GCTATGATTTCTCTTGCAT), and nt 129–147 (GCAAACTTTCAGGACTTTGA) of the human Malt1 3'UTR were inserted into pSUPER. The Pol III promoter-shRNA cassettes from these vectors and from a lamin A/C-specific pSUPER control construct were inserted into the lentiviral vector pAB286.1, a derivative of pHR (27) that contains a SV40-puromycin acetyl transferase cassette for antibiotic selection (lamin A/C-pSUPER and pAB286.1 were gifts from R. Iggo, University of St. Andrews, Scotland). Second generation packaging plasmids pMD2-VSVG and pCMV-R8.91 (27) were used for lentivirus production and infection as described elsewhere (www.tronolab.unige.ch). The lentiviral vector pRDI_292 (a gift from R. Iggo) was used to achieve stable expression of FLAG-tagged Bcl10 constructs in Jurkat cells.
IL-2 assay
Jurkat and Raji cells (0.75 x 106 cells/0.5 ml each) were mixed at a ratio of 1:1 in the absence or presence of staphylococcal enterotoxin E. (SEE) (0.2 µg/ml; Toxin Technology) in 24-well plates for 14 h at 37°C. The IL-2 concentration in the supernatants was determined by ELISA (R&D Systems) according to the manufacturer’s instructions.
Determination of cellular F-actin content
Mouse primary T cells or Jurkat T cells (1 x 106/100 µl RPMI 1640) were incubated with or without 10 µg/ml anti-CD3 (145-2C11 or OKT3) for the indicated times at 37°C. The reaction was terminated by addition of 400 µl of 5% paraformaldehyde in PBS. Cells were fixed for 10 min at room temperature, blocked with 1% BSA in buffer A (PBS, 10 mM HEPES (pH 7.3), and 0.5 mM EDTA), permeabilized with buffer B (0.1% saponin in buffer A containing 0.2% BSA), and stained with 2 µg/ml FITC- or TRITC-conjugated phalloidin (Sigma-Aldrich) in buffer B. Cellular F-actin content was determined using a FACScan flow cytometer by gating on living cells based on their forward and side scatter. The relative F-actin content of the cells is proportional to the relative mean fluorescence intensity, which was determined as the ratio of the mean fluorescence intensity of each sample relative to the fluorescence intensity of unstimulated cells. In some experiments, actin polymerization was induced by transfection of enhanced GFP (EGFP)-tagged constitutively active mutants of Rac1 (Rac1.L61) and Cdc42 (Cdc42.L61) (28) into Jurkat cells using the Amaxa system, and relative F-actin content was determined 48 h after transfection by gating on EGFP-positive cells.
Spreading assay and quantification of T cell spreading
Spreading assays were performed essentially as described by Bunnell et al. (29). Briefly, 15-mm glass coverslips placed on 12-well plastic culture plates were treated with 0.01% poly-L-lysine solution (Sigma-Aldrich) for 5 min at room temperature and then coated overnight at 4°C with PBS alone or with anti-CD3 Ab (OKT3) at 20 µg/µl. Wells were washed with PBS before use to remove any excess of Ab, then 200 µl of complete medium was added to each well and equilibrated at 37°C with 5% CO2. Two hundred microliters of Jurkat cell suspension (at 2 x 106 cells/ml) were added into medium-containing plates. To increase the number of cells at the coverslip surface at initial time points, plates with cells were pulse-centrifuged up to 800 rpm, in a 37°C preheated centrifuge with plate adaptors. At time zero, just after pulse centrifugation, no spreading is observed, but enough cells attached to the coverslip for microscopy analysis. To induce spreading, plates were incubated at 37°C for 3 min and fixed by removing 300 µl of medium and gently adding 500 µl of cold, freshly prepared 5% paraformaldehyde in PBS. After 30 min, plates with coverslips were blocked with 1% BSA in buffer A (PBS, 10 mM HEPES (pH 7.3), and 0.5 mM EDTA), permeabilized with buffer B (0.1% saponin in buffer A containing 0.2% BSA), and stained with 2 µg/ml FITC- or TRITC-conjugated phalloidin (Sigma-Aldrich) in buffer B to visualize the actin cytoskeleton. The coverslips were mounted in Fluorsafe (Calbiochem). Samples were examined under an Olympus IX70 inverted microscope, and random fields were collected using a U-PlanApo objective (x40/1.00 Oil Iris Ph3) and a cooled Sensi Cam 12 Bits camera (PCO-CCD Imaging). Images were acquired using Metamorph software and analyzed using ImageJ National Institutes of Health freeware to measure the surface area of plate contact (spreading) of individual cells. Cells that had at least doubled the area of plate contact compared with the average of unstimulated cells were by morphological features always unequivocally positive for spreading, so specific spreading (percentage) was defined as 100 x (number of cells with at least twice the average surface size of plate contact of control cells/number of total cells in the field).
Conjugate formation
Jurkat and Raji cells were labeled with CFSE cell tracker (25 nM; Sigma-Aldrich) and CMTMR cell tracker (500 nM; Molecular Probes), respectively, for 15 min at 37°C. Cells were then washed and incubated for 30 min at 37°C at a density of 106 cells/500 µl. During this incubation, Raji cells were incubated in the absence or presence of SEE superantigen (5 µg/ml; Toxin Technology). Jurkat and Raji cells were then mixed at a ratio of 1:1 (total volume 1 ml), centrifuged for 5 min at 1,000 x g, and 850 µl of medium was removed. Cell pellets were resuspended by vortexing for 2 s and incubated for 15 min at 37°C. After gentle resuspension, cell conjugates were fixed by adding 400 µl of paraformaldehyde 5%. Conjugate formation was analyzed with a FACScan flow cytometer (BD Biosciences). The percentage of conjugates was determined as the number of (CFSE+CMTMR+ events/total number of CFSE+ events) x 100.
Phagocytosis assay
Preparation of IgG-coated SRBC (IgG-SRBC) and phagocytosis assays were performed as described previously (30), except that THP-1 cells were plated onto poly-L-lysine-coated coverslips before the incubation with opsonized SRBCs. To quantitate phagocytosis, the number of internalized SRBCs was counted in 50 cells randomly chosen on the coverslips, and the phagocytic index, i.e., the mean number of phagocytosed SRBCs per cell, was calculated. The index obtained was divided by the index obtained for control lamin-depleted cells and expressed as a percentage of control cells. We also counted the number of cell-associated (bound plus internalized) SRBCs, calculated the association index (mean number of associated SRBCs per cell), and expressed it as percentage of control lamin-depleted cells. To quantitate polymerized actin recruitment, we scored the presence or absence of F-actin accumulations under the particles in 50 cells randomly chosen on the coverslips and calculated an accumulation index, i.e., the mean number of accumulations per cell. The index obtained was divided by the index obtained for control (lamin-depleted) cells and expressed as a percentage of the latter. We checked that lamin-depleted cells and parental nontransduced THP-1 cells showed similar phagocytosis efficiencies (data not shown).
Immunofluorescence microscopy of THP-1 cells
Immunofluorescence, image acquisition, and deconvolution was performed as previously described (Braun et al. (30)), except that the samples were examined under an inverted wide-field microscope (Leica DMB) equipped with an oil immersion objective (x100 PL APO HCX, 1.4 NA) and a cooled CCD camera (MicroMAX; Princeton Instruments). Z-series of images were taken at 0.2-µm increments, and deconvolution was performed by the three-dimensional deconvolution module from Metamorph Software (Universal Imaging). Three-dimensional reconstructions were obtained using the Surpass function of Imaris Software (Bitplane).
Statistics
The statistical significance of the data was tested with an unpaired Student’s t test, and the calculated goodness-of-fit value (p value) is indicated in the figures.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
We had previously observed that ectopic expression of Bcl10 in 293T cells yielded a 30-kDa nonphosphorylated Bcl10 species and two major, more slowly migrating isoforms of Bcl10 that disappeared upon phosphatase treatment (13, 14). Phosphorylation of Bcl10 was detectable by in vivo 32P-labeling of FLAG-Bcl10-transfected 293T cells and occurred exclusively on Ser residues (Fig. 1A). Analysis of various C-terminally truncated forms revealed that the observed Ser phosphorylation sites were present within aas 127 to 207 of the C-terminal portion of Bcl10 (data not shown). Clustered Ser to Ala point mutations within the C-terminal portion of Bcl10 revealed that Ser residues within a cluster comprising Ser134, -136, and -138, and a second cluster comprising Ser167, -170, and -171 most strongly affected Bcl10 phosphorylation in the 293T cell system (Fig. 1, B and C). Ser to Ala point mutation of individual residues within these two clusters demonstrated that Ser138 is critical for the formation of the faster migrating phospho-Bcl10 species, whereas Ser170 and -171 contribute to the formation of the slower migrating species (Fig. 1D).
|
B (8). We therefore analyzed whether T cell activation induced phosphorylation of endogenous Bcl10. Stimulation of Jurkat cells with PMA and ionomycin resulted in a minimal shift in the electrophoretic mobility of endogenous Bcl10 within the first minutes of stimulation (Figs. 1E and 2A, see also Fig. 3, D and E). This minimal shift in electrophoretic mobility was different from the migration pattern of Bcl10 obtained under conditions of overexpression in 293T cells, which induced a substantial shift in the apparent m.w. of Bcl10 (Fig. 1E). However, additional Bcl10 modifications that resulted in a substantial shift in electrophoretic mobility also became apparent upon prolonged stimulation of T cells (15 min and more) even though this concerned a very small fraction of total Bcl10 (see Figs. 2A and 3E, strong exposures). The relation between the altered migration of Bcl10 seen upon prolonged T cell stimulation and the phosphorylated forms of Bcl10 observed in 293T cells is not fully understood. It is likely that in T cells, these later modifications correspond to IKK-dependent Bcl10 phosphorylation events that have been proposed recently to be associated with a negative feedback regulation of the NF-B pathway (18). Similar modifications resulting in substantial m.w. shifts have also been observed by others (16, 18, 31, 32).
|
|
Ser138-dependent Bcl10 phosphorylation is not required for NF-B activation
Bcl10 plays a key role in the Ag receptor-induced activation of the transcription factor NF-B (1). Therefore, we tested the effect of the Bcl10 Ser138 phosphorylation mutant on PMA/ionomycin or CD3/CD28-induced NF-B activation in Jurkat cells in a NF-B luciferase reporter gene assay. Transfection of the Ser138/Ala (S138A) mutant did not inhibit NF-B activation induced by PMA/ionomycin-treatment or Carma1 overexpression in Jurkat cells, whereas a dominant-negative (DN), nonphosphorylatable form of IB impaired it under the same conditions (data not shown). To rule out the possibility that the Bcl10 S138A mutant had a recessive inhibitory effect on NF-B activation, we next analyzed its effect in Jurkat cells in which the expression of endogenous Bcl10 was silenced by the transfection of siRNA specific for human Bcl10 (Fig. 3, A and B). Reconstitution of the cells with either wt or S138A-mutated murine Bcl10 constructs (that are not targeted by the siRNA) restored PMA/ionomycin or CD3/CD28-induced NF-B activation to similar extents, suggesting that Ser138-dependent phosphorylation of Bcl10 was not necessary for NF-B activation (Fig. 3, A and B). Similar results were obtained using an IL-2 reporter construct (data not shown). Next, in Jurkat cells stably transduced with the wt or S138A form of Bcl10 (Fig. 3C, left panel), we tested whether mutation of Ser138 either affected Bcl10’s stability or the cells’ capacity to induce IB phosphorylation, an early event of NF-B activation. Using these cells, we found no difference in the PMA/ionomycin- or anti-CD3/CD28-induced degradation of the wt and the S138A constructs (Fig. 3, C, right panel, and D). Moreover, the cells showed no difference in anti-CD3/CD28- or PMA/ionomycin-induced IB phosphorylation (Fig. 3D and data not shown). Other parameters of cellular activation, such as activation of the MAPKs ERK and p38, were also comparable in cells overexpressing the wt or S138A form of Bcl10 (Fig. 3D). Thus, phosphorylation on Ser138 is not required for Bcl10-mediated NF-B activation. These findings are consistent with the observation that a Bcl10 construct comprising aas 1–116 (comprising only the CARD and the subsequent Malt1 binding region) acts like full-length Bcl10 with respect to its ability to activate NF-B and to synergize with PMA in NF-B induction (33). Moreover, this idea is consistent with the observation that Carma1 silencing did not affect early Bcl10 phosphorylation, whereas it clearly affected IB phosphorylation induced by PMA and ionomycin, as well as later Bcl10 modifications resulting in a more substantial shift in the apparent m.w. of Bcl10 (Fig. 3E).
Finally, we also analyzed the effect of the Bcl10 phosphorylation mutant on superantigen-induced IL-2 production. Jurkat cells stably transduced with the wt form of Bcl10 showed a strongly increased capacity to produce IL-2, whereas the Bcl10 S138A mutant was significantly less potent in increasing IL-2 production than the wt form of Bcl10 (Fig. 3F). Together, these findings suggest that the S138A mutant of Bcl10 affects functional T cell activation in a NF-B-independent manner.
Bcl10 is essential for TCR-induced actin polymerization
Bcl10 phosphorylation occurs rapidly after TCR triggering, suggesting that it may affect an early signaling event other than the activation of NF-B. One of the earliest events upon engagement of the TCR is the reorganization of the cytoskeleton, including the rapid induction of actin polymerization (34). To test whether Bcl10, and in particular its phosphorylation, plays a role in actin polymerization, primary purified T cells from Bcl10-deficient and control mice were stimulated with anti-CD3, and the resulting increase in F-actin was measured using fluorescently labeled phalloidin. In the Bcl10+/+ control cells, CD3 stimulation induced a transient increase in the levels of F-actin, which was absent in the Bcl10–/– T cells (Fig. 4A). Similar results were obtained in Jurkat cells in which Bcl10 expression was silenced by a lentiviral shRNA approach, whereas silencing of Carma1 or Malt1 expression had no significant effect on anti-CD3-induced actin polymerization (Fig. 4B). Silencing of Bcl10 expression did not reflect a general defect in the generation of actin filaments, because actin polymerization induced by the chemokine CXCL12/SDF-1 via the G protein-coupled receptor CXCR4 was not significantly affected by impaired Bcl10 expression (Fig. 4C).
|
Ser138 is crucial for Bcl10’s capacity to control F-actin formation and T cell spreading
Next, we tested the relevance of Ser138 phosphorylation on TCR-induced F-actin formation. In contrast to the wt form of Bcl10, the Ser138/Ala mutant was unable to restore anti-CD3-induced F-actin formation in cells with silenced Bcl10 expression (Fig. 5A). Moreover, the mutant had a DN effect on F-actin generation (Fig. 5A). We also tested the effect of the pan-PKC inhibitor GFX on TCR-induced actin polymerization. Consistent with the observation that the inhibitor did not affect Ser138-dependent Bcl10 phosphorylation (Fig. 2F), we did not observe an inhibitory effect on F-actin formation (Fig. 5B).
|
Bcl10 depletion does not affect activation of Vav nor actin polymerization induced by Cdc42 or Rac1
The molecular control of actin-filament assembly and disassembly upon receptor triggering is highly regulated and coordinated. The Arp2/3 complex is a major regulator of actin filament nucleation (38). The activity of the Arp2/3 complex is regulated by members of the SCAR/WASP family that are in turn controlled by small GTPases such as Cdc42 and Rac1 (24, 39). In T cells, the activation of Rac1 is controlled by the TCR-induced tyrosine phosphorylation of Vav, resulting in the stimulation of its GEF activity (34, 40, 41). Bcl10-deficient cells show a normal profile of total protein tyrosine phosphorylation (1), but the phosphorylation of Vav has not been specifically addressed. To address the effect of Bcl10 silencing on the phosphorylation status of Vav, tyrosine-phosphorylated proteins were immunoprecipitated from lysates of control cells and Bcl10-silenced cells, and analyzed by Western blotting for Vav. In control cells, anti-CD3 stimulation induced a rapid and transient tyrosine phosphorylation of Vav, which was unaltered in Bcl10-silenced cells (Fig. 6A, left panel). Transduction of the cells with the wt or Ser138/Ala mutant did not affect Vav phosphorylation either (Fig. 6A, right panel). Moreover, overexpression of constitutively active forms of Cdc42 or Rac1 induced an increase in F-actin content that was not significantly different in control cells and Bcl10-silenced cells (Fig. 6B). Together, these data suggest that Bcl10 controls actin polymerization downstream or independently of Vav and upstream or independently of the Rho family GTPases Cdc42 and Rac1.
|
R-induced actin polymerization and phagocytosis depend on Bcl10
Recently, Bcl10 has been reported to play a key role in FcR-induced NF-B activation (3). FcRs share a number of common signaling features with the TCR, due to the presence of ITAMs that become phosphorylated by tyrosine kinases of the Src family, and allow the subsequent recruitment of Syk family kinases to activate downstream signaling events via the phosphorylation of adaptor proteins and enzymes with various catalytic activities (42, 43, 44, 45). Triggering of FcR by immune complexes induces phagocytosis that depends on actin polymerization (43). To test whether Bcl10 was essential for this process, lamin A/C, Bcl10, or Carma1 were silenced in the human THP-1 monocytes (Fig. 7A), and F-actin formation or phagocytosis was measured upon incubation of the cells with Ab-coated SRBC (IgG-SRBC) (Fig. 7, B–F). Silencing of Bcl10 strongly impaired the IgG-SRBC-induced increase in total F-actin content (Fig. 7B). Moreover, Bcl10- but not Carma1-silenced monocytes showed an important decrease in phagocytosis, whereas their capacity to associate with the IgG-SRBC was comparable to control (lamin-silenced) cells (Fig. 7, C and D). When we analyzed early steps of phagocytosis, we observed that, whereas Bcl10-silenced cells still formed F-actin cups at sites of particle attachment (Fig. 7E), the cells showed an incomplete engulfment of the IgG-SRBC that was correlated with reduced size of F-actin cups as illustrated on the three-dimensional reconstructions (compare control and Bcl10-silenced cells in Fig. 7F). Thus, Bcl10 is crucial for FcR-induced F-actin increase and the resulting F-actin cup formation and phagocytosis.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Rs on phagocytes. In this study, we present several lines of evidence that support a previously unsuspected and physiologically relevant role for Bcl10 in TCR- and FcR-induced actin polymerization. First, T cells with silenced Bcl10 expression showed impaired TCR-induced actin polymerization, which could be rescued by reconstitution with wt Bcl10. Second, Ser138-dependent phosphorylation of Bcl10 was critical for its capacity to control actin polymerization, but did not affect its function in the NF-B pathway. Third, a S138A mutant of Bcl10 was less potent than the wt form in increasing superantigen-induced IL-2 production. Fourth, T cells with silenced expression of Carma1 or Malt1 showed no defect in TCR-induced actin polymerization, and silencing of Carma1 did not affect initial Bcl10 phosphorylation, suggesting that this function of Bcl10 is fully independent of its binding partners in the NF-B pathway. Furthermore, Bcl10-silenced T cells showed an impaired capacity to spread and to form conjugates with APCs. Finally, Bcl10 silencing led to an impaired FcR-induced increase in F-actin content and altered formation of F-actin-dependent protrusions around the opsonized particle, resulting in defective phagocytosis.
The role of Bcl10 in immune receptor-induced actin polymerization has important functional implications. Bcl10-deficient mice show impaired TCR-induced immune responses that have so far been attributed mainly to a lack of functional NF-B activation and IL-2 production (1). Our findings suggest that the impaired T cell-dependent responses may, at least in part, be due to the impaired capacity of the Bcl10-deficient cells to polymerize actin. This idea is consistent with our observation that T cell spreading and conjugate formation, both known to depend on TCR-induced actin polymerization (21, 34, 35, 46, 47), were affected in Bcl10-silenced cells.
FcR-induced actin remodeling is essential for the internalization of opsonized micro-organisms or particles, and is thus crucial for Ag capture and presentation by APCs (42, 43, 44). Our data identify a new role for Bcl10 as a key regulator of FcR-induced phagocytosis in monocytes, and thus suggest that Bcl10 may also be important for the activation of the adaptive immune system via phagocytic cells.
The signaling cascades leading to actin polymerization downstream of the TCR and the FcR show striking similarities. Indeed, both involve Src- and Syk-family tyrosine kinases and small GTPases such as Cdc42 and Rac that regulate Arp2/3-dependent actin polymerization via protein complexes containing members of the WASP/WAVE family (21, 34, 42, 43, 44). In T cells, the activity of Rac GTPases is controlled by tyrosine phosphorylation and activation of Vav family GEFs (34, 40, 41, 42, 43, 44, 48, 49). Whether Vav plays a similar role downstream of the FcR is controversial (50, 51). In our hands, Bcl10 silencing had no effect on TCR-induced Vav phosphorylation, indicating that Bcl10 controls actin polymerization downstream or independently of Vav. Moreover, the lack of an effect of Bcl10 silencing on Cdc42 and Rac1-induced actin polymerization suggests that Bcl10 acts by targeting an unknown component of the signaling pathway that is upstream or independent of these small GTPases and common to both TCR- and FcR-induced actin polymerization. Another possibility is that Bcl10 phosphorylation affects the stability or stabilization of F-actin filaments rather than their initial formation. Indeed, a physical association of Bcl10 with cytochalasin D-sensitive filaments has been observed in HeLa cells when Bcl10 was overexpressed (52), a condition that favors constitutive Bcl10 phosphorylation. Moreover, in yeast two hybrid assays Bcl10 binds to -actinin (52), an actin-binding protein that is thought to cross-link actin filaments and to link the actin fibrils to the cytoplasmic tail of certain transmembrane receptors (53). Although our data collectively suggest that direct phosphorylation of Ser138 is critical for its role in the signaling pathway controlling actin polymerization, we cannot formally exclude the possibility that mutation of Ser138 indirectly affects phosphorylation on another site by interfering with the recruitment of a kinase, or that the mutation has additional, phosphorylation-independent effects on the conformation and molecular function of Bcl10 that may affect its interaction with other components of the pathway. The functional connection of the Ser138-dependent phosphorylation of Bcl10 and the molecular machinery that polymerizes and stabilizes actin is currently under investigation.
A highly interesting question about the nature of the kinase-targeting Bcl10 remains open. The similarity of the phenotypes of PKC
and Bcl10-deficient mice (1, 54) together with the proposed role for PKC in the regulation of TCR-induced actin polymerization (55), suggest a potential role for PKC in Bcl10 phosphorylation. However, the amino acid sequence of Bcl10 surrounding Ser138 does not match the described consensus motif for PKC-dependent phosphorylation (56), and we and others were unable to demonstrate a direct PKC-dependent phosphorylation of Bcl10 in in vitro kinase assays (Ref. 57 and our unpublished data). Moreover, the pan-PKC inhibitor GFX did not affect the Ser138-dependent initial phosphorylation of Bcl10 nor the TCR-induced F-actin increase, but showed an inhibitory effect on a second, as yet unidentified site of Bcl10 phosphorylation that is thus unlikely to be relevant for actin polymerization. The receptor-interacting protein (RIP) family kinase RIP2 has been shown to associate with Bcl10 and to induce its phosphorylation upon TCR stimulation in the context of NF-B activation (15). Additional experiments are required to identify the RIP2-dependent phosphorylation site(s) in Bcl10 and to assess whether RIP2-mediated Bcl10 phosphorylation contributes to the effect on the actin cytoskeleton described here.
Interestingly, Bcl10 appears to undergo two types of sequential phosphorylation events that regulate distinct molecular functions of Bcl10. We propose that initial, rapidly occurring and Ser138-dependent phosphorylation of Bcl10 is crucial for its capacity to control TCR-dependent actin polymerization, whereas a timely delayed phosphorylation on multiple additional residues may be related to Bcl10’s function in the NF-B pathway. This idea is supported by our observation that Carma1 silencing interfered with the second wave of phosphorylation (Fig. 3E), whereas it neither affected initial Bcl10 phosphorylation nor TCR-induced actin polymerization. Recently, Wegener et al. (18) have proposed a negative regulatory role for this type of delayed, Carma1-dependent Bcl10 phosphorylation in TCR-induced NF-B activation. In particular, a cluster of five Ser residues comprising Ser138 was identified as the target of an IKK-mediated phosphorylation and negative feedback regulation of Bcl10 (18). In our hands, mutation of Ser138 alone did not affect NF-B activation (Fig. 3), suggesting that (an)other Ser residue(s) within the cluster may be more relevant for this function.
In conclusion, we have identified a new function for Bcl10 in receptor-induced actin polymerization and provided evidence that suggests that phosphorylation of Bcl10 on Ser138 is critical for this function. Thus, in addition to its role in controlling NF-B-dependent gene transcription, Bcl10 may control cellular activation via the regulation of actin-dependent cytoskeletal rearrangements.
|
Acknowledgments |
|---|
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported by grants from Swiss Cancer League (to M.Tho. and J.S.), Swiss National Science Foundation (to M.Tho. and J.S.), and Cancer and Solidarité (to J.S.). F.N. is supported by grants from the Ville de Paris and Centre National de la Recherche Scientifique (Action Thématique d’Intérêt Prioritaire Jeune Chercheur Microbiologie). J.R. is supported by a Max-Eder-Program Grant from Deutsche Krebshilfe and by grants from Deutsche Forschungsgemeinschaft. O.G. was supported by a M.D.-PhD fellowship from the Swiss Academy of Medical Sciences. D.R. was supported by a postdoctoral fellowship from the Spanish Ministry of Education and Science. S.H. was supported by a PhD fellowship from the Studienstiftung des Deutschen Volkes. E.L. was supported by a Bourse de Docteur Ingénieur du Centre National de la Recherche Scientifique. 
2 O.G. and L.H. contributed equally to this study.
3 Current address: University Medical Center, 1 rue Michel-Servet, 1211 Geneva 4, Switzerland.
4 Current address: Department of Research, University Hospital, Hebelstrasse 20, CH-4031 Basel, Switzerland.
5 Address correspondence and reprint requests to Dr. Margot Thome, Department of Biochemistry, University of Lausanne, Chemin des Boveresses 155, Epalinges, Switzerland. E-mail address: Margot.ThomeMiazza{at}unil.ch
6 Abbreviations used in this paper: MALT, mucosa-associated lymphoid tissue; CARD, caspase recruitment domain; Carma1, CARD-containing MAGUK protein-1; IKK, IB kinase; Malt1, MALT lymphoma translocation protein-1; GEF, GDP/GTP exchange factor; WASP, Wiskott-Aldrich syndrome protein; siRNA, small interfering RNA; shRNA, short hairpin RNA; EGFP, enhanced GFP; IgG-SRBC, IgG-coated SRBC; wt, wild type; PKC, protein kinase C; DN, dominant negative; RIP, receptor-interacting protein; SEE, staphylococcal enterotoxin E.
Received for publication September 18, 2006. Accepted for publication January 15, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
B and neural tube closure. Cell 104: 33-42. [Medline]RI-mediated nuclear factor B activation and cytokine production from mast cell degranulation. J. Exp. Med. 203: 337-347. B signaling in lymphocytes: a new cast of characters. J. Cell Sci. 117: 31-39. B-activating protein containing a caspase recruitment domain with homology to Herpesvirus-2 protein E10. J. Biol. Chem. 274: 9955-9961. B activation and apoptosis. J. Biol. Chem. 274: 17946-17954. B activation. FEBS Lett. 496: 121-127. [Medline]B activation. J. Biol. Chem. 279: 1570-1574. B signaling intermediates in response to T cell receptor stimulation. Proc. Natl. Acad. Sci. USA 101: 1004-1009. -induced Bcl10 nuclear translocation: Bcl10 is up-regulated by NF-B and phosphorylated by Akt1 and then complexes with Bcl3 to enter the nucleus. J. Biol. Chem. 281: 167-175. B kinase in remodeling Carma1-Bcl10-Malt1 complexes upo
-phosphatase, as indicated, and analyzed by two-dimensional gel electrophoresis and anti-Bcl10 Western blot. C, Kinetics of Bcl10 phosphorylation. Jurkat cells were stimulated with anti-CD3 (OKT3) for the indicated time, and postnuclear cell lysates were analyzed by two-dimensional gel electrophoresis and anti-Bcl10 Western blot. D, Bcl10 phosphorylation occurs in primary T cells. Purified primary human CD4+ T cells (10 x 106) were incubated for 15 min in the presence or absence of PMA and ionomycin or cross-linked anti-CD3 and anti-CD28 Abs, respectively, and cells were lysed and analyzed as in C. E, Mutation of Ser138 into Ala abolishes Bcl10 phosphorylation. Jurkat cells transfected with the indicated FLAG-Bcl10 constructs were stimulated with PMA and ionomycin for 20 min, and anti-FLAG immunoprecipitates were analyzed as in B. F, The pan-PKC inhibitor GFX partially inhibits Bcl10 phosphorylation. Jurkat cells were preincubated for 60 min in the presence of 500 nM GFX or solvent control, stimulated for 15 min with PMA and ionomycin, and phosphorylation of endogenous Bcl10 was assessed as in C. In all parts of the figure, arrowheads indicate the position of unphosphorylated Bcl10 (black arrowhead) and phosphorylated forms of Bcl10 (open arrowheads).
