Abstract
Ag-mediated activation of mast cells initiates signaling events leading to Ca2+ response, release of allergic mediators from cytoplasmic granules, and synthesis of cytokines and chemokines. Although microtubule rearrangement during activation has been described, the molecular mechanisms that control their remodeling are largely unknown. Microtubule nucleation is mediated by complexes that are formed by γ-tubulin and γ-tubulin complex proteins. In this study, we report that, in bone marrow–derived mast cells (BMMCs), γ-tubulin interacts with p21-activated kinase interacting exchange factor β (βPIX) and G protein–coupled receptor kinase-interacting protein (GIT)1. Microtubule regrowth experiments showed that the depletion of βPIX in BMMCs stimulated microtubule nucleation, whereas depletion of GIT1 led to the inhibition of nucleation compared with control cells. Phenotypic rescue experiments confirmed that βPIX and GIT1 represent negative and positive regulators of microtubule nucleation in BMMCs, respectively. Live-cell imaging disclosed that both proteins are associated with centrosomes. Immunoprecipitation and pull-down experiments revealed that an enhanced level of free cytosolic Ca2+ affects γ-tubulin properties and stimulates the association of GIT1 and γ-tubulin complex proteins with γ-tubulin. Microtubule nucleation also was affected by Ca2+ level. Moreover, in activated BMMCs, γ-tubulin formed complexes with tyrosine-phosphorylated GIT1. Further experiments showed that GIT1 and βPIX are involved in the regulation of such important physiological processes as Ag-induced chemotaxis and degranulation. Our study provides for the first time, to our knowledge, a possible mechanism for the concerted action of tyrosine kinases, GIT1/βPIX proteins, and Ca2+ in the propagation of signals leading to the regulation of microtubule nucleation in activated mast cells.
Introduction
Mast cells play a crucial role in allergy, as well as in innate and adaptive immune responses. They express plasma membrane–associated high-affinity IgERs (FcεRIs), the aggregation of which triggers mast cell activation, resulting in degranulation, the release of proinflammatory mediators, and the production of various cytokines. The first step in FcεRI signaling is tyrosine phosphorylation of the FcεRI β- and γ-subunits by the Src family nonreceptor kinase Lyn. This is followed by enhanced activity of tyrosine kinase Syk and the phosphorylation of transmembrane adaptors, which organize and coordinate further signals, resulting in a Ca2+ efflux from the endoplasmic reticulum (ER). Depletion of Ca2+ from the ER lumen induces Ca2+ influx across the plasma membrane, leading to enhancement of the free cytoplasmic Ca2+ concentration, a step that is important in further signaling events (1).
Microtubules, built up from αβ-tubulin heterodimers, are important for mast cell degranulation, because the movement of secretory granules depends on intact microtubules (2, 3), and compounds inhibiting tubulin polymerization suppress degranulation (4). It was reported that FcεRI aggregation leads to reorganization of microtubules (3, 5) and that the influx of Ca2+ plays a decisive role in microtubule remodeling (6, 7). An important role in Ca2+ influx was reported for stromal-interacting protein 1 (STIM1), the Ca2+ sensor in ER that associates with the end binding protein 1 located on the tips of growing microtubules (8). Depletion of STIM1 resulted in the inhibition of microtubule reorganization in stimulated cells (6). Although these data point to the necessity of the microtubule network for mast cell degranulation, the molecular mechanisms responsible for microtubule reorganization in activated mast cells are still largely unknown. The role of Ca2+ in this process is also unclear.
In cells, microtubules are dominantly nucleated from centrosomes or the other microtubule-organizing centers. One of the key components for microtubule nucleation is γ-tubulin (9), a highly conserved member of the tubulin superfamily. In cytosol, γ-tubulin exists as a γ-tubulin ring complex (γTuRC; with size ∼2.2 MDa) comprising γ-tubulin small complexes (∼280 kDa), composed of two molecules of γ-tubulin, one molecule of γ-tubulin complex protein (GCP)2, one molecule of GCP3, and some other proteins (10). Cumulative data indicate that protein tyrosine kinases phosphorylate γ-tubulin or associated proteins and, in this way, modulate γ-tubulin functions (5, 11, 12). The significance of Src kinases for microtubule nucleation from centrosomes was ascertained by microtubule regrowth experiments (13). Identification of tyrosine kinase substrates that regulate γ-tubulin functions should help to elucidate the mechanisms involved in the regulation of microtubule nucleation.
In this study, we examined the hypothesis that phosphotyrosine (P-Tyr) proteins associated with γ-tubulin modulate microtubule nucleation in activated mast cells. We identified p21-activated kinase (PAK) interacting exchange factor β (βPIX) and G protein–coupled receptor kinase-interacting protein (GIT)1 as signaling proteins that interact with γ-tubulin and associate with centrosomes. GIT1 is phosphorylated on tyrosine in activated cells and interacts with γ-tubulin in a Ca2+-dependent manner. Our data suggest a novel signaling pathway for microtubule rearrangement in mast cells, where tyrosine kinase–activated GIT1 and βPIX work in concert with Ca2+ signaling to regulate microtubule nucleation. Through this pathway, Ag-induced signaling pathways leading to chemotaxis and degranulation could be regulated.
Materials and Methods
Reagents
BAPTA-AM, fibronectin, nocodazole, puromycin, ionomycin, and Hygromycin B were from Sigma. IL-3 and stem cell factor were from PeproTech EC. Protein A immobilized on Trisacryl GF-2000 and SuperSignal West Pico Chemiluminescent reagents were bought from Pierce. Protease-inhibitor mixture tablets (Complete EDTA-free) were from Roche Molecular Biochemicals, GFP-trap_A was purchased from ChromoTek, and Ni-NTA-agarose was from QIAGEN. Restriction enzymes were from New England Biolabs, and Glutathione Sepharose 4 Fast Flow was from GE Healthcare Life Sciences. Oligonucleotides were synthesized by East Port.
Abs
The following anti-peptide Abs prepared to human γ-tubulin were used: mouse mAb TU-31 (IgG2b) and mAb TU-30 (IgG1) to the sequence 434–449 (14) and mAb GTU 88 (IgG1; Sigma, cat. no. T6657) to the sequence 38–53. α-Tubulin was detected with rabbit Ab from GeneTex (GTX15246). Rabbit Abs to βPIX (HPA004744), GFP (G1544), GAPDH (G9545), and actin (A2066) were from Sigma. Rabbit Ab to GIT1 (sc-13961) and mAb to GCP4 (IgG1; sc-271876) were from Santa Cruz. Rabbit mAb to GIT1 (ab156001) was from Abcam. Rabbit mAb D11B8 to GIT2 (8072) was from Cell Signaling, and rabbit Ab to tRFP (AB234) was from Evrogen. mAb 4G10 (IgG2b) to P-Tyr conjugated with HRP (16-184) was from Upstate Laboratories, and mAb to 6xHis was from BD Biosciences (San Jose, CA; IgG2b; 8916-1). mAb GCP2-02 (IgG1) to GCP2 protein was described previously (15). Rabbit Ab to nonmuscle myosin H chain (BT-561; Biomedical Technologies) and mAb NF-09 (IgG2a) to neurofilament NF-M protein (16) served as negative controls in the immunoprecipitation experiments. Rabbit Ab to GST was from Dr. Petr Dráber (Institute of Molecular Genetics). The DY488-conjugated anti-rabbit and the DY549-conjugated anti-mouse Abs were from Jackson ImmunoResearch. Secondary HRP-conjugated Abs were from Promega Biotech.
Cell cultures and transfection
Bone marrow–derived mast cells (BMMCs) were isolated from the femurs and tibias of 6–8-wk-old BALB/c mice. All mice were maintained and used in accordance with the Institute of Molecular Genetics guidelines (permit number 12135/2010-17210) and national guidelines (2048/2004-1020). The cells were differentiated in suspension cultures for 6–8 wk, as previously described (6).
Stable cell lines derived from mouse BMMCs and Lyn−/− BMMCs were donated by Dr. M. Hibbs (Ludwig Institute for Cancer Research, Melbourne, Australia) (17). In this article, the cells are denoted as BMMC lines (BMMCLs) or Lyn−/− BMMCLs. Both the Ca2+ response (18) and Ag-induced degranulation (L. Dráberová and P. Dráber, unpublished observations) were decreased in Lyn−/− BMMCLs compared with BMMCLs. Cells were cultured in freshly prepared culture medium (RPMI 1640 supplemented with 20 mM HEPES [pH 7.5], 100 U/ml penicillin, 100 mg/ml streptomycin, 100 mM MEM nonessential amino acids, 1 mM sodium pyruvate, 10% FCS, and 10% WEHI-3 cell supernatant as a source of IL-3). In some cases, cells were cultivated for 30 min in serum-free and Ca2+-free medium to which Ca2+ was freshly added to a final concentration of 1.8 mM. Alternatively, cells were incubated with 1 μM ionomycin in the presence or absence of Ca2+. Cells were grown at 37°C in 5% CO2 in air and passaged every 2 d.
HEK 293-FT (HEK) cells (Promega Biotech) were grown at 37°C in 5% CO2 in DMEM supplemented with 10% FCS and antibiotics. The cells used for lentivirus production were at passage 4–15. HEK cells were transfected with 17 μg DNA/9-cm tissue culture dish using 51 μg polyethylenimine (Polysciences) and serum-free DMEM. After 24 h, the transfection mixture was replaced with fresh medium supplemented with serum, and cells were incubated for an additional 24 h.
Cell activation
Cells were sensitized with DNP-specific IgE and activated with Ag (DNP-BSA conjugate, 30–40 mol DNP/mol BSA), as described (6). Alternatively, cells were activated by pervanadate. Pervanadate solution was freshly prepared by mixing sodium orthovanadate solution with H2O2 to get a 10 mM final concentration of both components. The pervanadate solution was incubated for 15 min at room temperature and then diluted 1:100 in buffered saline solution (6).
DNA constructs
To prepare N-terminally EGFP-tagged human γ-tubulin (TUBG1; RefSeq ID: NM_001070.4), pH3-16 plasmid containing the full-length cDNA of human γ-tubulin (19) was digested with EcoRI/BglII restriction enzymes, and the fragment was inserted into pEGFP-C1 (Clontech), resulting in the plasmid pEGFP-hTUBG1_1-451. To prepare truncated forms of EGFP-tagged γ-tubulin, fragments encoding aa regions 1–440, 1–422, and 1–382 were amplified by PCR from the pH3-16 using forward primer 5′-AGTCAAGCTTATGCCGAGGGAAATCATC-3′ and the following reverse primers: 5′-CTAAGATCTCTACCGTGTGGCCGCATG-3′ (aa 1–440); 5′-GTCAGATCTCTAGTCCATCTCATCAAAGTTGTCC-3′ (aa 1–422), and 5′-CAAAGATCTCTAGGTGTGGTTGGCCATCAT-3′ (aa 1–382), respectively. Sites recognized by restriction endonucleases are underlined. PCR products were digested with HindIII/BglII and inserted into pEGFP-C3 (Clontech), resulting in the plasmids pEGFP-hTUBG1_1-440, pEGFP-hTUBG1_1-422, and pEGFP-hTUBG1_1-382, respectively. The C-terminal part of γ-tubulin (aa 362–451) was amplified from pH3-16 using forward 5′-TCGGATCCAGGAAGTCTCCCTACCT-3′ and reverse 5′-TTCTCGAGTCACTGCTCCTGGGTG-3′ primers. The PCR product was digested with BamHI/XhoI and ligated into pGEX-6P-1 (Amersham Biosciences), resulting in the plasmid pGST-hTUBG1_362-451. The shorter C-terminal part of γ-tubulin (aa 378–451) was amplified from pGST-hTUBG1_362-451 using forward 5′-GAATTCGGCCAACCACACCAGC-3′ and reverse 5′-GGATCCTCACTGCTCCTGGGTGC-3′ primers. The C-terminal part of α-tubulin (aa 378–451) was amplified from pGST-hTUBA1B_364-451 (V. Sulimenko, manuscript in preparation) using forward 5′-GAATTCGAGCAACACCACAGCC-3′ and reverse 5′-GGATCCTTAGTATTCCTCTCCTTCTTCC-3′ primers. PCR products were ligated into pCR2.1 (Invitrogen) by TA cloning, and fragments were excised with EcoRI/BamHI and ligated into pEGFP-C1, resulting in the plasmids pEGFP-hTUBG1_378-451 and pEGFP-hTUBA1B_378-451. To prepare N-terminally 6xHis-tagged γ-tubulin, the coding sequence was amplified from pH3-16 using forward 5′-AAGCATGCCGAGGGAAATCATCAC-3′ and reverse 5′-CTAAGCTTTCACTGCTCCTGGGTGC-3′ primers. The PCR product was digested with SphI/HindIII and ligated to pQE-82L (QIAGEN), resulting in the plasmid 6xHis-hTUBG1_1-451. γ-Tubulin fragment (aa 1–225) was amplified from pH3-16 using forward 5′-AAGCATGCCGAGGGAAATCATCAC-3′ and reverse 5′-AATAAGCTTTCAGGAGAAGGATGGGTTC-3′ primers. The PCR product was digested with SphI/HindIII and ligated to pQE-82L (QIAGEN), resulting in the plasmid 6xHis-hTUBG1_1-225. γ-Tubulin fragment (aa 223–451) was amplified from pH3-16 using forward 5′-AATGGTACCTCCTTCTCCCAGATCAAC-3′ and reverse 5′-GATTAAGCTTTCACTGCTCCTGGGTGC-3′ primers. The PCR product was digested with KpnI/HindIII and ligated to pQE-80 (QIAGEN), resulting in the plasmid 6xHis-hTUBG1_223-451. A cassette encoding mouse γ-tubulin fused to TagRFP was digested out from the pmTubg1-TagRFP construct (20) by EcoRI/NotI and ligated to pCDH-CMV-MCS-EF1-puro vector (System Biosciences), resulting in the lentiviral construct pmTubg1-TagRFP-puro.
A cassette encoding mouse GIT1 (mGIT1; Git1; RefSeq ID: NM_001004144.1) fused to GFP was amplified from the TurboGFP-tagged Git1 cDNA ORF clone (MG210556; OriGene Technologies) using forward 5′-GTCTAGAGAGGAGATCTGCCGCCG-3′ and reverse 5′-GCCGGGAATTCGTTTAAACTCTTTC-3′ primers. The PCR product was ligated into pCR 2.1 by TA cloning, and the fragment was excised with XbaI/EcoRI and ligated into pCDH-CMV-MCS-EF1-puro, resulting in the lentiviral construct pmGIT1-GFP-puro. The coding sequence of human GIT1 (hGIT1; GIT1, transcript variant [tv]1; RefSeq ID: NM_001085454.1) was excised from phGIT1(tv1)-neo (M. Černohorská, manuscript in preparation) with NheI/NotI and ligated into the pCDH-CMV-MCS-EF1-hygro vector (System Biosciences), resulting in the lentiviral construct phGIT1(tv1)-hygro.
A cassette encoding mouse βPIX (mβPIX; Arhgef7, tv3; RefSeq ID: NM_017402.4) fused to GFP was amplified from the TurboGFP-tagged Arhgef7 cDNA ORF clone (MG223397; OriGene Technologies) using forward 5′-TATGCTAGCGTCGACTGGATCCGG-3′ and reverse 5′-GCCGGGAATTCGTTTAAACTCTTTC-3′ primers. The PCR product was ligated into pCR 2.1 by TA cloning, and the fragment was excised with NheI/EcoRI and ligated into pCDH-CMV-MCS-EF1-puro, resulting in the lentiviral construct pmβPIX(tv3)-GFP-puro. The coding sequence of human βPIX (hβPIX; ARHGEF7, tv1; RefSeq ID: NM_003899.3) was excised from GST-hβPIX(tv1) (21) with BamHI/NotI and ligated into the pCDH-CMV-MCS-EF1-hygro vector. For a phenotypic rescue experiment, two silent point mutations (a1620g, t1623g) were generated in this construct by site-directed mutagenesis using the QuikChange II XL Site-Directed Mutagenesis Kit (Stratagene), according to the manufacturer’s protocol. The resulting lentiviral construct was named phβPIX(tv1)mut-hygro.
All constructs were verified by sequencing. Plasmid pFYSH2 encoding the GST-tagged SH2 domain of mouse Fyn kinase was described previously (5). The lentiviral vector pCT-Mito-GFP containing cytochrome oxidase c subunit VIII (COX8) tag for visualization of mitochondria was obtained from System Biosciences (CYTO102-PB-1).
Lentiviral infection
Lentiviral infections were done as described previously (6), using HEK packaging cells for virus preparation. The transfection mixture was replaced after 3 d with fresh complete medium containing 5 μg/ml puromycin. Stable selection was achieved by culturing cells for 1–2 wk in the presence of puromycin. In phenotypic rescue experiments, medium containing puromycin was supplemented with 1 mg/ml Hygromycin B, and stable selection was achieved by culturing cells for 1–2 wk. BMMCL stably expressing TagRFP-tagged mouse γ-tubulin (BMMCL-γTb) was prepared by lentiviral transduction with pmTubg1-TagRFP-puro. To follow the distribution of GIT1 or βPIX in living cells, BMMCL-γTb was transduced with pmGIT1-GFP-puro or pmβPIX(tv3)-GFP-puro. After 3 d, cells expressing fluorescently tagged proteins were flow sorted using the BD Influx cell sorter (BD Biosciences). GFP and TagRFP emission was triggered by 488- and 561-nm lasers; fluorescence was detected with 525/50 and 585/20 band-pass filters. Only double-positive cells were used for subsequent live-cell imaging.
RNA interference
A set of five mouse Arhgef7 (National Center for Biotechnology Information RefSeq: NM_001113517.1, NM_001113518.1, NM_017402.4) short hairpin RNA (shRNA) constructs cloned into the lentiviral pLKO.1 vector (TRCN0000110025, TRCN0000110026, TRCN0000110027, TRCN0000110028, and TRCN0000110029) were purchased from Open Biosystems. With the exception of TRCN0000110025, the vectors target all transcript variants of Arhgef7. A set of five murine GIT1 (National Center for Biotechnology Information RefSeq: NM_001004144.1) shRNA constructs cloned into the pLKO.1 vector (TRCN0000106120, TRCN0000106121, TRCN0000106122, TRCN0000106123, and TRCN0000106124) were also purchased from Open Biosystems. Immunoblotting experiments revealed that cells with the greatest reduction in βPIX protein were obtained with TRCN0000110026 (βPIX-KD1) and TRCN0000110027 (βPIX-KD2). Similarly, cells with the greatest reduction in GIT1 protein were obtained with TRCN0000106122 (GIT1-KD1) and TRCN0000106121 (GIT1-KD2). The stable selected cells with the greatest reduction in βPIX or GIT1 were used for additional experiments. Cells transduced with empty lentiviral pLKO.1 vector (Addgene) or pLKO.1 vector containing nontarget shRNA (Sigma) were used as negative controls.
Preparation of cell extracts
Whole-cell extracts for SDS-PAGE were prepared by washing the cells in cold HEPES buffer (50 mM HEPES [pH 7.6], 75 mM NaCl, 1 mM MgCl2, and 1 mM EGTA), solubilizing them in hot SDS sample buffer without bromophenol blue, and boiling for 5 min. When preparing extracts for immunoprecipitation and GST pull-down assays, cells were rinsed twice in cold HEPES buffer and extracted at a concentration of 15 × 106 cells/ml for 10 min at 4°C with HEPES buffer supplemented with 1% Nonidet P-40 (extraction buffer), protease inhibitor mixture, and phosphatase inhibitors (1 mM Na3VO4, 1 mM NaF). The suspension was spun down (20,000 × g, 15 min, 4°C), and supernatant was collected. In some cases, cell lysates were supplemented with CaCl2 at concentrations ranging from 50 μM to 1.0 mM or with 10 mM EGTA. Alternatively, lysates were supplemented with MgCl2 or ZnCl2 at a concentration of 0.8 mM. When preparing extracts for gel-filtration chromatography, cells were extracted at a concentration of 14 × 107 cells/ml. Protein quantification in lysates and SDS-PAGE samples was assessed with a bicinchoninic acid assay and a silver dot assay, respectively (22).
Gel-filtration chromatography
Gel filtration was performed using fast protein liquid chromatography (AKTA-FPLC system; Amersham) on a Superose 6 10/300 GL column (Amersham). Column equilibration and chromatography were performed in an extraction buffer. The column was eluted at 30 ml/h, and 0.5-ml aliquots were collected. Samples for SDS-PAGE were prepared by mixing with 5× concentrated SDS sample buffer. The following molecular mass standards were used: IgM (900 kDa), thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), and BSA (66 kDa).
Immunoprecipitation, GST pull-down assay, gel electrophoresis, and immunoblotting
Immunoprecipitation was performed as previously described (23). Cell extracts were incubated with Protein A beads (Pierce, Rockford, IL) saturated with mAb TU-31 (IgG2b) to γ-tubulin, rabbit Ab to GIT1, rabbit Ab to βPIX, rabbit Ab to RFP, rabbit Ab to nonmuscle myosin (negative control), mAb NF-09 (IgG2a; negative control), or immobilized protein A alone. Abs to βPIX, GIT1, and tRFP were used at Ig concentrations of 2, 2, and 5 μg/ml, respectively. Ab to myosin was used at a dilution of 1:100. mAb TU-31 and mAb NF-09, in the form of hybridoma supernatants, were diluted 1:2. GFP-tagged proteins were immunoprecipitated using GFP-trap_A, according to the manufacturer’s directions.
For large-scale immunoprecipitation experiments, Lyn−/− BMMCLs (total 6 × 108 cells) were activated by pervanadate, as described (6), and resuspended in 15 ml cold extraction buffer supplemented with protease and phosphatase inhibitors. After 10 min of incubation at 4°C, the suspension was centrifuged at 28,000 × g for 15 min at 4°C. Supernatant was incubated with anti-peptide mAb TU-31 to γ-tubulin immobilized on 5 ml Protein A beads overnight at 4°C. After extensive washing in TBST (10 mM Tris-HCl [pH 7.4] 150 mM NaCl, 0.05% Tween 20), immobilized protein A with bound proteins was loaded into a 10-ml (7 × 1.5 cm) column, washed with 70 ml TBST, and eluted with peptide used for immunization (14) at a concentration of 200 μg/ml in TBST. From the 0.5-ml fractions collected, 1-μl aliquots were spotted onto nitrocellulose and probed with anti–P-Tyr mAb conjugated with HRP (dilution 1:30,000), followed by chemiluminescent detection of bound Ab. Fractions containing tyrosine-phosphorylated proteins were collected (total 2 ml) and loaded onto 50 μl pelleted glutathione-Sepharose with bound GST-tagged mouse Fyn-SH2 domain to concentrate the tyrosine-phosphorylated proteins capable of binding this domain. After 3 h of incubation at 4°C and extensive washing, the proteins associated with the domain were dissolved in 50 μl 2× SDS sample buffer and boiled for 5 min.
Preparation and purification of GST-tagged fusion proteins were described previously, as were pull-down assays with whole-cell extracts (23). For comparison of immunoprecipitation and pull-down assays in the presence or absence of Ca2+, protein extracts were used at the same protein concentration. Gel electrophoresis and immunoblotting were performed using standard protocols. Preparation, purification, and immobilization of 6xHis-tagged proteins onto Ni-NTA agarose were performed according to the manufacturer’s recommendations.
SDS-PAGE and immunoblotting were performed using standard protocols (24). For immunoblotting, mAbs to 6xHis, γ-tubulin (GTU-88), P-Tyr, and GCP4 were diluted 1:50,000, 1:10,000, 1:10,000, and 1:1,000, respectively. mAb to GCP2, in the form of spent culture supernatant, was diluted 1:10. Rabbit Abs to GAPDH, βPIX, GIT1, actin, and GIT2 were diluted 1:50,000, 1:4,000, 1:2,000, 1:2,000, and 1:500, respectively. Rabbit Abs to GST, GFP, and RFP were diluted 1:20,000, 1:5,000, and 1:5,000, respectively. Secondary anti-mouse and anti-rabbit Abs conjugated with HRP were diluted 1:10,000. Bound Abs were detected by SuperSignal West Pico Chemiluminescent reagents (Pierce).
Mass spectrometry
Following large-scale immunoprecipitation and concentration of peptide-eluted samples on the GST-Fyn-SH2 domain, proteins dissolved in SDS sample buffer were separated on preparative 7.5% SDS-PAGE using the Multigel-Long electrophoretic system (Biometra). Gels were stained by Coomassie Brilliant Blue G-250. The bands of interest were excised from the gel, destained, and digested by trypsin. Extracted peptides were analyzed by a MALDI-FTICR mass spectrometer (APEX-Qe) equipped with a 9.4 tesla superconducting magnet (both from Bruker Daltonics, Billerica, MA) at the Core facility of the Institute of Microbiology, Academy of Sciences of the Czech Republic. The obtained data were processed by Data Analysis 4.0 software (Bruker Daltonics) and searched by the ProFound (PROWL) search engine against the nonredundant database of all known Mus musculus proteins.
Degranulation assay
The degree of degranulation was quantified as the release of β-glucuronidase from anti-TNP IgE-sensitized cells activated with Ag (TNP-BSA conjugate, 15–25 mol TNP/mol BSA), using 4-methylumbelliferyl β-d-glucuronide as substrate (25). The total content of the enzyme was evaluated in supernatants from cells lysed by 0.1% Triton X-100.
Chemotaxis assay
The chemotactic response of cells was examined using a 24-well Transwell system with 8-μm-diameter pore size polycarbonate filters (Corning), as described previously (26). TNP-BSA, at a concentration of 250 ng/ml in RPMI 1640, supplemented with 20 mM HEPES and 1% BSA (assay buffer), served as a chemoattractant.
Evaluation of cell proliferation and apoptosis
Cell proliferation was assessed by manual counting of stable cell lines with the greatest reduction in βPIX or GIT1. Cells transduced with empty lentiviral pLKO.1 vector or pLKO.1 vector containing nontarget shRNA (pLKO.1-NT) were used as negative controls. A total of 2 × 105 transfected cells was plated on a 6-cm-diameter petri dish. Cells were counted from 1 to 5 d. Samples were counted in doublets in a total of three independent experiments. Cell viability was evaluated by a trypan blue exclusion test.
One of the classic features of apoptosis is the cleavage of genomic DNA into oligonucleosomal fragments represented by multiples of 180–200 bp. Genomic DNA was isolated from BMMCLs by the QIAamp DNA Mini kit (QIAGEN), according to the manufacturer’s directions, and prepared samples were incubated with RNase A (Fermentas) at a final concentration of 10 μg/ml for 1 h at 37°C. DNA integrity was evaluated by electrophoresis in 1.5% agarose gel in 40 mM Tris-acetate/1 mM EDTA buffer. For staining, we used GelRed Nucleic Acid Gel Stain (Biotium). GeneRuler 1 kb Plus DNA Ladder 75 to 20,000 bp (Thermo Scientific) was used for DNA sizing. Apoptosis was induced in control BMMCLs by treating cells with 10 μM 17-(allylamino)-17-demethoxygeldanamycin (Sigma) for 24 h.
Microtubule regrowth
Microtubule regrowth from centrosomes was followed in a nocodazole-washout experiment. BMMCLs in complete medium were overlaid on fibronectin-coated coverslips (6) and allowed to attach for 1 h at 37°C. Cells were then treated with nocodazole at a final concentration of 10 μM for 50 min at 37°C to depolymerize microtubules. Cells were washed with medium precooled to 4°C (3 × 5 min each) to remove the drug, and microtubule regrowth was allowed for 1–6 min at 26°C. To test the effect of pervanadate, cells treated with 10 μM nocodazole for 40 min at 37°C were transferred into medium containing 10 μM nocodazole and freshly prepared pervanadate. After 10 min of incubation at 37°C, cells were washed (2 × 2 min each) with precooled medium, and microtubule regrowth was allowed in the presence of pervanadate. Samples were fixed in cold methanol, air-dried, and washed in PBS before immunostaining (27). In some cases, nocodazole-washout experiments were performed in the absence of both extracellular and intracellular Ca2+. Cells in complete medium were treated with nocodazole (10 μM, 40 min, at 37°C), transferred into Ca2+- and serum-free medium supplemented with 5 μM BAPTA-AM and 10 μM nocodazole, and incubated for 10 min at 37°C to deplete intracellular Ca2+. After washing in cold Ca2+- and serum-free medium to remove the drug, microtubule regrowth was allowed for 1–6 min at 26°C in Ca2+- and serum-free medium containing 5 μM BAPTA-AM. The nocodazole-washout experiment also was performed in the absence of extracellular Ca2+ only. In control experiments, Ca2+- and serum-free medium was substituted for serum-free medium with 1.8 mM Ca2+.
Immunofluorescence microscopy
Immunofluorescence staining was performed as described (27). To visualize centrosomes and microtubules in fixed cells, coverslips were incubated with mAb TU-30 (spent culture supernatant diluted 1:5) and rabbit Ab to α-tubulin (diluted 1:150). Anti-mouse DY549-conjugated Ab was diluted 1:800, and anti-rabbit DY488-conjugated Ab was diluted 1:300. The preparations were mounted in MOWIOL 4-88 (Calbiochem), supplemented with 4,6-diamidino-2-phenylindole (Sigma), and examined on an Olympus Scan^R system (Olympus) equipped with Acquisition Scan^R program and an oil-immersion objective 60×/1.4 NA. Huygens Deconvolution Software (Scientific Volume Imaging) was used for preparation of immunofluorescence figures.
Quantification of the microtubule regrowth assay was performed on a large number of cells that were selected automatically and compared with the quantification of cells selected manually to confirm the reliability of the automatic method. In the course of automatic quantification, 36 areas/sample were taken in both fluorescence channels, and optical z-sections were acquired at 0.3-μm steps. Maximum intensity projection of γ-tubulin staining was used to identify the position of centrosomes. The sum of α-tubulin immunofluorescence intensities was generated by Sum Slices projection in ImageJ software (National Institutes of Health). The α-tubulin fluorescent signal near the centrosome was measured in separate concentric circles centered at the centrosome with radii of 1, 1.5, and 2 μm (regions of interest; ROIs). Background fluorescence using circles of corresponding sizes was subtracted from each measurement. Measurements were made using ImageJ. In manually performed analyses, only images of cells with a homogenous background around the centrosomes were selected. The sum of α-tubulin immunofluorescence intensities was obtained from 11 consecutive frames (0.2-μm steps), with the middle frame chosen with respect to the highest γ-tubulin intensity. For statistical analysis, the two-tailed, unpaired Student t test was used to compare samples and to obtain p values.
Live-cell imaging
BMMCL-γTb expressing mGIT1-GFP, mβPIX-GFP, or GFP-COX8 tags were plated on 35-mm glass-bottom culture dishes (MatTek; P35G-1.5-14-C) precoated with fibronectin (6), and cells were allowed to attach for 1 h at 37°C. Cells were washed and subsequently incubated in medium for live-cell imaging (RPMI 1640 without phenol red, riboflavin, folic acid, pyridoxal, Fe[NO3]3) supplemented with 20 mM HEPES. Cells were imaged on the Delta Vision Core system (Applied Precision) equipped with a 60×/1.42 NA oil-immersion objective. Optical z-sections were acquired in 0.3-μm steps.
Results
Proteins βPIX and GIT1 interact with γ-tubulin and associate with centrosomes in BMMCLs
We showed previously that, in FcεRI- or pervanadate-activated BMMCLs or Lyn−/− BMMCLs, γ-tubulin interacts with a similar set of tyrosine-phosphorylated proteins (5). In an attempt to identify these proteins, we applied large-scale immunoprecipitation coupled with mass spectrometry (MS). Lysates from pervanadate-activated cells were immunoprecipitated with anti-peptide mAb to γ-tubulin, and the bound proteins were eluted with peptide used for immunization. Eluted proteins were concentrated on the immobilized GST-SH2 domain of Fyn kinase, which effectively binds tyrosine-phosphorylated proteins (23). Bound proteins were separated on SDS-PAGE and subjected to MALDI/MS fingerprint analysis. Because slightly higher recovery of tyrosine-phosphorylated proteins associated with γ-tubulin was observed from activated Lyn−/− BMMCLs (5), we used these cells for MALDI/MS fingerprint analysis. Of the three independent experiments, βPIX (also known as Rho guanine nucleotide exchange factor 7, Arhgef7, β-Pix, COOL-1; gene name Arhgef7; Swiss-Prot identifier Q9ES28) was identified three times. A typical example of MS identification is shown in Supplemental Table I.
In cells, βPIX forms complexes with GITs, which are GTPase-activating proteins for the ADP-ribosylation factor family of small GTP-binding proteins (28). Such complexes might serve as scaffolds to bring together signaling molecules affecting various cellular processes, including cytoskeletal organization (29). To ascertain whether βPIX or GIT1 associates with γ-tubulin in Lyn−/− BMMCL cells, immunoprecipitation experiments were performed with Abs to βPIX and GIT1. Immunoblot analysis revealed the coprecipitation of γ-tubulin with both βPIX (Fig. 1A, lane 3) and GIT1 (Fig. 1B, lane 3). In addition, reciprocal precipitation with Ab to γ-tubulin confirmed an interaction between both βPIX (Fig. 1F, lane 3) and GIT1 (Fig. 1I, lane 3) with γ-tubulin. As expected, Ab to βPIX coprecipitated GIT1 (Fig. 1G, lane 3), and Ab to GIT1 coprecipitated βPIX (Fig. 1E, lane 3). Immunoblot analysis revealed that GIT2 also interacted with γ-tubulin (Supplemental Fig. 1A, lane 3). Negative-control mAb of IgG2b class or rabbit polyclonal Ab did not precipitate GIT1, βPIX, or γ-tubulin (Supplemental Fig. 1D). No association of γ-tubulin with βPIX and GIT1 was detected when precipitation with anti-peptide Ab to γ-tubulin was performed in the presence of immunizing peptide (data not shown). Immunoprecipitation experiments also confirmed interactions between γ-tubulin and βPIX, GIT1, or GIT2 in BMMCLs, proving that the association of γ-tubulin with these proteins is not restricted to cells with depleted Lyn kinase (data not shown). To independently confirm the interaction of γ-tubulin with βPIX and GIT1, immunoprecipitation experiments were performed with BMMCL-γTb. Precipitation with anti-RFP Ab revealed the coprecipitation of endogenous βPIX (Fig. 1J, lane 3) and GIT1 (Fig. 1K, lane 3). Precipitation of γ-tubulin–TagRFP is shown in Fig. 1L (lane 3). Immunoblot analysis also revealed that GIT2 was coprecipitated by anti-RFP Ab (Supplemental Fig. 1B, lane 3). Multiple phosphorylation sites were identified on hGIT1 (30). To determine whether GIT1 can be phosphorylated on tyrosine in activated BMMCLs or Lyn−/− BMMCLs, as well as whether tyrosine-phosphorylated GIT1 interacts with γ-tubulin, pervanadate-activated cells were precipitated with Abs to γ-tubulin and GIT1. Blots were probed with Abs to P-Tyr and GIT1. Both Abs precipitated P-Tyr proteins from both cell lines (Fig. 1M, lanes 1, 2), which were stained with Ab to GIT1 (Fig. 1N, lanes 1, 2). Negative-control Abs did not precipitate tyrosine-phosphorylated GIT1 from activated cells (data not shown). Thus, tyrosine-phosphorylated GIT1 from pervanadate-activated cells is capable of interacting with γ-tubulin. Similar, but less profound, results were obtained after activation of the cells by aggregation of FcεRI (data not shown). To decide whether γ-tubulin and βPIX or GIT1 protein appear in the form of complexes, BMMCL extracts were subjected to gel-filtration chromatography on the Superose 6 column. γ-Tubulin was distributed through a large zone in complexes of various sizes. Large complexes of ∼2 MDa could represent γTuRCs. Importantly, βPIX and GIT1 also were present in the complexes, and their distribution partially overlapped with γ-tubulin. In contrast, control actin was only found in low molecular mass fractions (Supplemental Fig. 1C). Immunoprecipitation experiments with anti–γ-tubulin Ab from fractions containing GIT1 and PIX (fractions 24–26 in Supplemental Fig. 1C) revealed the formation of complexes between γ-tubulin and these proteins (data not shown). Isotype controls for immunoprecipitation experiments are shown in Supplemental Fig. 1D.
γ-Tubulin interacts with βPIX and GIT1 in mast cells. (A–I) γ-Tubulin coprecipitates with βPIX and GIT1 in Lyn−/− BMMCLs. Extracts were precipitated with Protein A–immobilized Abs specific to βPIX (A, D, and G), GIT1 (B, E, and H), or γ-tubulin (C, F, and I). Blots were probed with Abs to γ-tubulin (γ-Tb), βPIX, or GIT1. Load (lane 1), immobilized Abs not incubated with cell extract (lane 2), immunoprecipitated proteins (lane 3), and protein A without Abs, incubated with cell extracts (lane 4). (J–L) TagRFP-tagged γ-tubulin associates with βPIX and GIT1 in BMMCLs. Extracts from BMMCL-γTb were precipitated with Protein A–immobilized Abs specific to RFP (J–L). Blots were probed with Abs to βPIX (J), GIT1 (K), and RFP (L). Load (lane 1), immobilized Abs not incubated with cell extract (lane 2), immunoprecipitated proteins (lane 3), and protein A without Abs, incubated with cell extracts (lane 4). In (J), one quarter of the load was applied compared with (K) and (L). (M and N) GIT1 is a substrate for protein tyrosine kinases in activated Lyn−/− BMMCLs and BMMCLs. Extracts from pervanadate-activated Lyn−/− BMMCLs (lane 1) or BMMCLs (lane 2) were precipitated with Protein A–immobilized Abs specific to γ-tubulin or GIT1. Immobilized Abs not incubated with cell extracts (lane 3). Blots were probed with Abs to P-Tyr (M) and GIT1 (N). Lines on the right indicate the positions of molecular mass markers (in kDa).
For the localization of centrosomes in live cells, BMMCL-γTb were used. The TagRFP-tagged γ-tubulin marked centrosomes and properly nucleated microtubules as documented by microtubule staining of mitotic cells (Fig. 2A–C). Because βPIX (tv3) is the most abundant variant in BMMCLs (V. Sládková, unpublished observations), GFP-tagged mβPIX (tv3) was expressed in BMMCL-γTb. Live-cell imaging revealed the association of βPIX with centrosomes of interphase BMMCLs. Moreover, diffuse staining of the cytoplasm was observed (Fig. 2D–F). Similarly, GFP-tagged GIT1 localized to centrosomes, and granular staining in cytoplasm was detected (Fig. 2G–I). Nocodazole treatment did not affect GIT1 localization on the centrosome, indicating no requirement for intact microtubules (data not shown). In control cells expressing GFP-tagged mitochondrial marker, no association of the expressed protein with the centrosome was detected (Fig. 2J–L). Collectively, the data suggest that βPIX and GITs form complexes with γ-tubulin and that βPIX and GIT1 associate with centrosomes in BMMCLs.
Subcellular localization of GFP-tagged βPIX and GIT1 in BMMCLs. Mitotic BMMCL-γTb stained for α-tubulin (A) to visualize localization of TagRFP-tagged γ-tubulin (B). (C) Superposition of images (α-tubulin, green; γ-tubulin, red), methanol-fixed cells. Localization of βPIX-GFP (D) and γ-tubulin–TagRFP (E) in live BMMCL-γTb cells. (F) Superposition of images (βPIX, green; γ-tubulin, red). Localization of GIT1-GFP (G) and γ-tubulin–TagRFP (H) in live BMMCL-γTb. (I) Superposition of images (GIT1, green; γ-tubulin, red). Localization of control mitochondrial Mito-GFP (J) and γ-tubulin-TagRFP (K) in live BMMCL-γTb. (L) Superposition of images (Mito-GFP, green; γ-tubulin, red). In (D)–(L), the best centrosomal plane is shown. Scale bars, 5 μm.
Ca2+ level affects γ-tubulin properties and its interaction with GITs and GCPs in BMMCLs
During the immunoprecipitation experiments, we noticed that Ca2+ affects the association of γ-tubulin with GITs and GCPs. When BMMCLs were cultivated for 30 min in serum- and Ca2+-free medium to which Ca2+ was freshly added to a final concentration of 1.8 mM, more GIT1 (Fig. 3A) or GIT2 (Fig. 3A) coprecipitated with γ-tubulin. Similarly, more GCP2 (Fig. 3A) or GCP4 (Fig. 3A) was associated with γ-tubulin in the presence of Ca2+. No substantial changes in the amount of βPIX associated with γ-tubulin were observed in the presence or absence of Ca2+ (Fig. 3A, βPIX), and actin (negative control) did not coprecipitate with γ-tubulin (Fig. 3A, Actin). Interestingly, two protein bands with different electrophoretic mobilities were detected in γ-tubulin immunoprecipitates from extracts containing Ca2+: an upper band with slow mobility and a lower band with fast mobility. Sometimes, three closely spaced γ-tubulin bands were detected in the presence of Ca2+. This was an unexpected finding because no such split in γ-tubulin electrophoretic mobilities was observed in loads (Fig. 3A, γ-Tb). Comparable results from immunoprecipitation experiments were obtained when Ca2+ was added to cell extracts. Moreover, Ca2+ similarly affected the electrophoretic mobility of γ-tubulin immunoprecipitated from BMMCs (data not shown). Negative-control Abs did not precipitate βPIX, GIT1, GIT2, γ-Tb, GCP2, or GCP4 from lysates containing Ca2+ (data not shown).
Effect of Ca2+ on coimmunoprecipitation of proteins with γ-tubulin. (A) Ca2+ stimulates coprecipitation of GITs and GCPs with γ-tubulin. BMMCLs were preincubated in the presence or absence of 1.8 mM Ca2+, and extracts were precipitated with Ab to γ-tubulin immobilized on Protein A. Blots were probed with Abs to βPIX, GIT1, GIT2, γ-tubulin (γ-Tb), GCP2, GCP4, or actin. Note the change in electrophoretic mobility of precipitated γ-tubulin and the higher amount of coprecipitated GIT1, GIT2, GCP2, and GCP4 in the presence of Ca2+. (B) Ca2+ affects the electrophoretic mobility of TagRFP-tagged γ-tubulin. Extracts from BMMCL-γTb were supplemented or not with 0.8 mM Ca2+ and precipitated with Ab to RFP immobilized on protein A. Blot was probed with Ab to RFP. Immobilized Ab not incubated with cell extract (lane 1), immunoprecipitated proteins (lanes 2 and 3), and protein A without Ab, incubated with cell extract (lanes 4 and 5). Lines on the left indicate the positions of the molecular mass markers (in kDa).
Immunoprecipitation experiments with Ab to γ-tubulin also were performed with extracts obtained from cells incubated with ionomycin, an effective Ca2+ ionophore, in the presence or absence of Ca2+. In the absence of Ca2+, ionomycin did not affect the mobility of γ-tubulin and coprecipitation of GCP2 and GCP4 with γ-tubulin. However, in the presence of ionophore and Ca2+, large amounts of GCP4 coprecipitated with γ-tubulin. Under such conditions the amount of coprecipitated GCP2 increased only slightly, and the amount of modified γ-tubulin was basically unchanged (Supplemental Fig. 1E).
When TagRFP–γ-tubulin was precipitated in the presence of Ca2+, slow-migrating electrophoretic variants of tagged γ-tubulin were detected (Fig. 3B, lane 3). Together, these data suggest that Ca2+ modulates the interaction of γ-tubulin with GITs and GCPs in BMMCLs and that both endogenous and exogenous γ-tubulin concentrated by immunoprecipitation is affected by an increased concentration of Ca2+ in BMMCLs.
C-terminal region of γ-tubulin is essential for Ca2+-dependent changes in γ-tubulin
To obtain deeper insight into the γ-tubulin region that is sensitive to Ca2+, we performed experiments with truncated γ-tubulins. For this, we used human γ-tubulin 1, which shows 99% identity with mouse γ-tubulin 1 (20). First, we immobilized 6xHis-tagged whole-length γ-tubulin (aa 1–451) or its truncated constructs covering aa regions 1–225 and 223–451 on Ni-NTA agarose and incubated them with BMMCL extracts in the presence or absence of Ca2+. The shift in electrophoretic mobility of γ-tubulin was observed in the C-terminal part of the molecule (Fig. 4A). When γ-tubulin was incubated with Ca2+ in the absence of extract, no mobility shift was observed (Fig. 4A, Control). Next, we prepared EGFP-tagged versions of γ-tubulin truncated from the C-terminal end of molecules containing α-helixes H11 (aa 385–400) and H12 (aa 419–437), which are exposed on the surface of the molecule (Protein Data Bank ID: 3CB2). A schematic diagram of the used constructs is depicted in Fig. 4B. The EGFP-tagged proteins covering γ-tubulin aa regions 1–382, 1–422, 1–440, and 1–451 were precipitated from HEK-293 lysates using GFP-trap; after thorough washing, immobilized constructs were incubated with BMMCL extracts in the presence or absence of Ca2+. Although electrophoretic mobilities were basically not affected in truncated γ-tubulins (aa 1–382 or 1–422), a Ca2+-dependent shift in mobility was seen in whole-length γ-tubulin and in truncated γ-tubulin (aa 1–440) (Fig. 4C). When the EGFP-tagged γ-tubulin C-terminal region (aa 378–451) was used, a shift in electrophoretic mobility was detected, whereas the EGFP-tagged C-terminal region of α-tubulin (aa 378–451) was not affected (Fig. 4D). Mg2+ or Zn2+ ions were not capable of forming γ-tubulin variants; when EGTA was added to deplete Ca2+, the generation of slow-migrating γ-tubulin forms was inhibited. Generation of γ-tubulin electrophoretic variants in the presence of Ca2+ also was observed when lysates from BMMCs were used in pull-down experiments (data not shown). Altogether, these findings indicate that the C-terminal region of γ-tubulin (aa 423–451) is essential for Ca2+-dependent changes in γ-tubulin in the presence of BMMCL or BMMC extracts.
Ca2+-dependent changes in electrophoretic mobility of γ-tubulin C-terminal domain. Immobilized tagged human γ-tubulin fragments were incubated with BMMCL extracts in the presence or absence of 0.8 mM Ca2+ for 30 min at 37°C before immunoblotting. (A) 6xHis-tagged human whole-length γ-tubulin or its fragments encoding aa regions 1–225 and 223–451 immobilized on Ni-NTA agarose and immunoblotted with Ab to 6xHis. Extracts were not added in the Control. (B) Schematic diagram of the constructs used. Positions of helixes H11 (gray) and H12 (dark gray) are highlighted. (C) EGFP-tagged truncated forms of human γ-tubulin encoding aa regions 1–382, 1–422, 1–440, and full-length γ-tubulin (1–451) immobilized on GFP-trap and immunoblotted with Ab to γ-tubulin (γ-Tb). (D) EGFP-tagged human γ-tubulin or α-tubulin fragments encoding aa regions 378–451 immobilized on GFP-trap and immunoblotted with Ab to GFP. Lines on the left indicate the positions of molecular mass markers (in kDa).
Ca2+ modulates the interaction of βPIX with the C-terminal region of γ-tubulin
In HEK cells expressing either GFP-tagged whole-length (aa 1–451) or truncated (aa 1–382) γ-tubulin, anti-βPIX and anti-GIT1 Abs coprecipitated GFP-tagged whole-length γ-tubulin. Coprecipitation of the truncated form of γ-tubulin was undetectable or it was substantially lower (Fig. 5A). This indicated that the C-terminal region of γ-tubulin is important for the interaction with βPIX and GIT1. Because Ca2+ affected the interactions of GIT1 with γ-tubulin and also generated changes in the C-terminal domain of γ-tubulin, we evaluated the possibility that Ca2+ might modulate the binding of GIT1 and βPIX to the γ-tubulin C-terminal region. The pull-down experiments were performed with the GST-tagged C-terminal part of γ-tubulin (aa 362–451) and BMMCL extracts in the presence or absence of Ca2+. βPIX associated with the GST-tagged γ-tubulin domain, but its binding was inhibited in the presence of Ca2+. In contrast, GIT1 bound to the GST-fusion protein both in the absence and presence of Ca2+. Actin (negative control) did not bind to the GST-tagged γ-tubulin domain, and no binding of GIT1 or βPIX was observed when GST alone was used in the pull-down assay (Fig. 5B). These findings suggest that Ca2+ adversely affects the interaction of βPIX with the C-terminal region of the γ-tubulin molecule in BMMCLs.
Effect of Ca2+ on βPIX and GIT1 binding to the C-terminal domain of γ-tubulin. (A) βPIX and GIT1 interact with whole-length γ-tubulin but not with its truncated form lacking a C-terminal region. Extracts from HEK cells expressing either GFP-tagged whole-length (aa 1–451; lanes 1 and 3) or truncated (aa 1–382; lanes 2 and 4) γ-tubulin were precipitated with Protein A–immobilized Abs specific to βPIX or GIT1. Blots were probed with Ab to GFP. Loads (lanes 1 and 2), immunoprecipitated proteins (lanes 3 and 4). (B) Binding of βPIX to the C-terminal region of γ-tubulin is affected by Ca2+. Extracts from BMMCLs (Load) were incubated in the presence or absence of 0.8 mM Ca2+ with the GST-tagged C-terminal region of γ-tubulin (aa 362–451) or GST alone immobilized on glutathione-Sepharose beads. Blots of bound proteins were probed with Abs to βPIX, GIT1, actin, or GST. Lines on the left indicate the positions of molecular mass markers (in kDa).
Nucleation of microtubules in BMMCLs is affected by Ca2+ and cell activation
To evaluate the possibility that the Ca2+ level affects microtubule assembly from centrosomes, we performed the nocodazole-washout assay. Microtubules were depolymerized with nocodazole and allowed to grow in the absence of the drug. To perform microtubule regrowth in the absence of both intracellular and extracellular Ca2+, cells were preincubated with BAPTA-AM, which depleted intracellular Ca2+. Cells for analysis were selected either manually or automatically. The α-tubulin fluorescence of microtubule asters was measured in ROIs centered at centrosomes that were marked by staining for γ-tubulin immunofluorescence, as previously described (13). In control cells, a clearly visible microtubule array, originating from the centrosomes, appeared after 3 min (Fig. 6Aa), and a fully developed microtubule array was detected after 4 min. In the absence of both extracellular and intracellular Ca2+, only small microtubule asters were formed after 3 min of microtubule regrowth (Fig. 6Ac), and a fully developed microtubule array was observed after 6 min. Statistical evaluation of α-tubulin fluorescence in ROIs with diameters of 1, 1.5, or 2 μm after manual selection documented clear differences between regrowth in control and Ca2+-depleted cells (Fig. 6B). Similar results were obtained when analysis was performed on a large number of cells selected automatically (data not shown). When Ca2+ was absent only in the medium, a delay in microtubule regrowth was also observed (data not shown). When cells were activated by pervanadate, a potent protein tyrosine phosphatase inhibitor that mimics, in part, the stimulatory effect of Ag (31), microtubule regrowth was increased (Fig. 6C). The extent of microtubule regrowth could be modulated by mechanisms regulating either microtubule nucleation or microtubule dynamics. It was reported previously that microtubule dynamics is regulated at the cell periphery (32) and that a delay in microtubule regrowth is associated with defects in microtubule nucleation (13, 33). Together, these data suggest that both Ca2+ level and cell activation affect microtubule nucleation in BMMCLs.
Effect of Ca2+ depletion and cell activation on microtubule regrowth. (A) Double-labeling of α-tubulin (a and c) and γ-tubulin (b and d) in a microtubule regrowth experiment in the presence or absence of Ca2+. Both extracellular and intracellular Ca2+ were depleted, and cells were fixed at 3 min of microtubule regrowth. Scale bar, 5 μm. (B) Statistical analysis of α-tubulin fluorescence intensity in 1-, 1.5-, and 2-μm ROIs of BMMCLs nucleated in the absence of Ca2+ relative to BMMCLs nucleated in the presence of Ca2+. Four independent experiments were performed, each involving 30 control cells and 30 Ca2+-depleted cells (- Ca2+). Data are mean ± SD (n = 120). (C) Statistical analysis of α-tubulin fluorescence intensity in 1-, 1.5-, and 2-μm ROIs of pervanadate-activated BMMCLs relative to unstimulated BMMCLs. Cells were fixed at 2.5 min of microtubule regrowth. Three independent experiments were performed, each involving 40 control cells and 40 pervanadate-activated cells (+ Pv). Data are mean ± SD (n = 120). ***p < 0.001.
Opposite regulatory roles of βPIX and GIT1 in nucleation of microtubules in BMMCLs
Because βPIX and GIT1 interact with γ-tubulin, we compared microtubule regrowth from centrosomes in BMMCLs with a reduced level of βPIX or GIT1 in nocodazole-washout experiments. The βPIX- and GIT1-deficient cells were produced using lentiviral vectors. A typical result of immunoblotting experiments after depletion of βPIX or GIT1 is shown in Fig. 7A and Fig. 7B, respectively. At the best silencing, the amount of βPIX in βPIX-KD1 cells reached 14.7 ± 2.9% and the amount of GIT1 in GIT1-KD1 cells was 10.7 ± 2.8% (mean ± SD; n = 5) compared with the expression level in control cells with an empty pLKO.1 vector. Three independent experiments were performed with selected stable cells with reduced levels of βPIX, GIT1, and corresponding control cells. α-Tubulin immunofluorescence was measured at 1.5 min after washout in a 1.0-μm ROI. Control experiments revealed similar nucleation of microtubules when empty pLKO.1 vector or pLKO.1 vector with nontarget shRNA (pLKO.1-NT) was used (Supplemental Fig. 2H). Results of regrowth experiments are presented after analysis of a large number of cells selected automatically, but similar results were obtained after the analysis of a limited number of cells selected manually.
βPIX and GIT1 depletion affects microtubule regrowth. (A) Immunoblot analysis of cells with reduced levels of βPIX. Whole-cell lysates from cells infected with empty pLKO.1 vector (Control) or from cells selected after the knockdown of βPIX by shRNA1 (βPIX-KD1) or shRNA2 (βPIX-KD2). (B) Immunoblot analysis of cells with reduced levels of GIT1. Whole-cell lysates from cells infected with empty pLKO.1 vector (Control) or from cells selected after knockdown of GIT1 by shRNA1 (GIT1-KD1) or by shRNA2 (GIT1-KD2). Numbers under the blots indicate the relative amounts of βPIX (A) or GIT1 (B) normalized to control cells and to the amount of GAPDH in individual samples (Fold). (C–F) Distribution of α-tubulin fluorescence intensities (arbitrary units [AU]) in 1-μm ROI at 1.5 min of microtubule regrowth are shown as box plots (three independent experiments, >700 cells counted for each experimental condition). (C) Box plot of βPIX-depleted cells (mβPIX-KD1; n = 4501) relative to control cells (Control, pLKO.1; n = 2788). (D) Box plot of βPIX-depleted cells rescued by hβPIX (mβPIX-KD1 + hβPIX; n = 2269) relative to control cells (Control, pLKO.1 + pCDH; n = 2645). (E) Box plot of GIT1-depleted cells (mGIT1-KD1; n = 3641) relative to control cells (Control, pLKO.1, n = 2938). (F) Box plot of mGIT1-depleted cells rescued by hGIT1 (mGIT1-KD1 + hGIT1; n = 4503) relative to control cells (Control, pLKO.1 + pCDH; n = 2,688). In (C)–(F), bold and thin lines within the box represent the mean and median (the 50th percentile), respectively. The bottom and top of the box represent the 25th and 75th percentiles. Whiskers below and above the box indicate the 10th and 90th percentiles. ***p < 1 × 10−15.
βPIX depletion resulted in an increase in microtubule regrowth (Fig. 7C). Typical staining of α-tubulin in control cells and βPIX-depleted cells is shown in Supplemental Fig. 2Ab and 2Ac, respectively. The increase in microtubule regrowth also was observed in cells with a lower level of βPIX depletion, denoted βPIX-KD2 (Supplemental Fig. 2B). For phenotypic rescue experiments, two silent point mutations were introduced into hβPIX to prevent its depletion by the shRNA. Rescue experiments revealed that the introduction of hβPIX to BMMCLs with depleted mβPIX levels led to restoration of nucleation capacity, as in control cells (Fig. 7D). Microtubule regrowth after control βPIX depletion in rescue experiments is shown in Supplemental Fig. 2C, and an immunoblot from the rescue experiment is shown in Supplemental Fig. 2D. The data obtained suggest that βPIX represents a negative regulator of microtubule nucleation from the centrosomes in BMMCLs.
In contrast, GIT1 depletion resulted in a decrease in microtubule regrowth (Fig. 7E). Typical staining of α-tubulin in GIT1-depleted cells is shown in Supplemental Fig. 2Ad. The decrease in microtubule nucleation also was observed in cells with a lower level of GIT1 depletion, denoted GIT1-KD2 (Supplemental Fig. 2E). Rescue experiments confirmed that the introduction of hGIT1 into BMMCLs with a reduced level of mGIT1 restored nucleation capacity to that observed in control cells (Fig. 7F). Microtubule regrowth after control GIT1 depletion in rescue experiments is shown in Supplemental Fig. 2F, and an immunoblot from the rescue experiment is shown in Supplemental Fig. 2G. These experiments indicate that GIT1 represents a positive regulator of microtubule nucleation from the centrosomes in BMMCLs. Altogether, the results of these experiments suggest that βPIX and GIT1 differentially regulate microtubule nucleation.
Differential effect of βPIX and GIT1 depletion on degranulation and cell motility
We performed immunoprecipitation experiments to evaluate how the depletion of mβPIX affects the interaction of GIT1 with γ-tubulin, as well as how the depletion of GIT1 affects the interaction of βPIX with γ-tubulin. Although decreased levels of βPIX had no effect on the interaction between mGIT1 and γ-tubulin (Fig. 8A, lanes 4, 5, GIT1), depletion of GIT1 led to the increased association of βPIX with γ-tubulin (Fig. 8B, lanes 4, 5, βPIX). These results indicate that GIT1 modulates the formation of γ-tubulin/βPIX complexes.
Depletion of GIT1 promotes interaction of βPIX with γ-tubulin. (A) Extracts from BMMCLs infected with empty pLKO.1 vector (Control) or BMMCLs with reduced levels of βPIX (βPIX-KD1) were precipitated with Protein A–immobilized Ab to γ-tubulin. Blots were probed with Abs to βPIX, GIT1, and γ-tubulin (γ-Tb). Load of control cells (lane 1), load of βPIX-KD1 cells (lane 2), immobilized Ab not incubated with cell extract (lane 3), precipitated proteins from control cells (lane 4), precipitated proteins from βPIX-KD1 cells (lane 5), Protein A without Ab, incubated with extract from control cells (lane 6), and Protein A without Ab, incubated with extract from βPIX-KD1 cells (lane 7). (B) Extracts from BMMCLs infected with empty pLKO.1 vector (Control) or BMMCLs with reduced levels of GIT1 (GIT1-KD1) were precipitated with Protein A–immobilized Ab to γ-tubulin. Blots were probed with Abs to βPIX, GIT1, and γ-tubulin (γ-Tb). Load of control cells (lane 1), load of GIT1-KD1 cells (lane 2), immobilized Ab not incubated with cell extract (lane 3), precipitated proteins from control cells (lane 4), precipitated proteins from GIT1-KD1 cells (lane 5), Protein A without Ab, incubated with extract from control cells (lane 6), and Protein A without Ab, incubated with extract from GIT1-KD1 cells (lane 7).
Because microtubules are important for cell proliferation and survival, we assessed how depletion of βPIX or GIT1 affects these processes. Proliferation of βPIX-depleted cells was comparable to negative-control cells (pLKO.1 and pLKO.1-NT). In contrast, depletion of GIT1 resulted in partial inhibition of proliferation (Supplemental Fig. 3A). The viability of cells in the evaluated time interval was >99, >99, >98, and >95% for pLKO.1, pLKO.1-NT, βPIX-depleted cells, and GIT1-depleted cells, respectively. No differences between control cells (pLKO.1 and pLKO.1-NT) and βPIX- or GIT1-depleted cells were detected when genomic DNAs were analyzed for the presence of oligonucleosomal fragments characteristic of apoptotic cells (Supplemental Fig. 3B).
Finally, we measured the role of GIT1 and βPIX in Ag-induced chemotaxis and degranulation. Data in Fig. 9A show that cells with reduced GIT1 exhibited a significantly stronger Ag-mediated chemotactic response than did the control cells. In contrast, βPIX-depleted cells showed a less efficient chemotactic response. Interestingly, the general migration of GIT1-deficient cells also was enhanced, as indicated by the greater migration of cells in the absence of Ag (Fig. 9A, Control). Rescue experiments confirmed that the introduction of hGIT1 into BMMCLs with a reduced level of mGIT1 restored the chemotactic response to that observed in control cells. Similarly, the chemotactic response was rescued after the introduction of hβPIX into BMMCLs with a reduced level of mβPIX (Fig. 9A, Ag). In contrast to chemotaxis, cells with depleted GIT1 exhibited reduced degranulation, whereas βPIX-deficient cells showed significantly higher degranulation compared with control cells (Fig. 9B). Similar results were obtained when pLKO.1 or pLKO.1-NT were used as negative controls (Supplemental Fig. 3C). Two cell lines with reduced GIT1 (GIT1-KD1, GIT1-KD2) gave comparable results in the degranulation assay. Similarly, no statistically significant differences were detected in degranulation between two cell lines with reduced βPIX (βPIX-KD1, βPIX-KD2) (Supplemental Fig. 3C). Rescue experiments confirmed that introduction of hGIT1 into BMMCLs with reduced levels of mGIT1 restored degranulation to that observed in control cells. Degranulation also was recovered after the introduction of hβPIX into BMMCLs with a reduced level of mβPIX (Fig. 9B). Altogether, the results suggest that βPIX and GIT1 differentially regulate Ag-induced chemotaxis and degranulation in BMMCs.
Differential effect of βPIX and GIT1 depletion on Ag-induced chemotaxis and degranulation. IgE-sensitized control cells or cells deficient in GIT1 (mGIT1-KD1) or βPIX (mβPIX-KD1) were used. hGIT1 or hβPIX was applied in rescue experiments. (A) The cells were analyzed in chemotactic assays with chemotaxis medium alone (Control) or supplemented with Ag (TNP-BSA, 250 ng/ml). The numbers of cells migrating into lower wells were determined after 8 h. Mean ± SE was calculated from three independent experiments performed in duplicates or triplicates. (B) IgE-sensitized cells were stimulated with various concentrations of Ag, and the amount of β-glucuronidase released from the cells was determined after 30 min. Mean ± SE was calculated from three independent experiments performed in duplicates or triplicates. The statistical significance of differences between control cells and βPIX-deficient cells is shown in the upper part, whereas those between control and GIT1-deficient cells are shown in the lower part. *p < 0.05, **p < 0.01, ***p < 0.001.
Discussion
Ag-induced activation of mast cells leads to rapid cytoskeleton rearrangements and degranulation. Accumulating data point to the importance of microtubules in this process (3, 4, 7, 34). We showed previously that the stimulation of mast cells or basophils through FcεRI aggregation or by pervanadate exposure triggers the generation of complexes containing γ-tubulin, tyrosine-phosphorylated proteins, and tyrosine kinases (5, 35) and the reorganization of microtubules (5, 6). In this article, we report on GIT1 and βPIX as signaling proteins interacting with γ-tubulin in a Ca2+-dependent manner, associating with centrosomes, and modulating microtubule nucleation from the centrosomes of BMMCLs. βPIX and GIT1 represent negative and positive regulators of microtubule nucleation, respectively. Our study provides a possible mechanism for the concerted action of tyrosine kinases, GIT1 and βPIX proteins, and Ca2+ in the propagation of signals leading to microtubule nucleation in activated mast cells.
Several lines of evidence indicate that the association of βPIX and GIT1 with γ-tubulin is specific. First, βPIX was repeatedly identified by MALDI/MS fingerprint analysis after immunoprecipitation of activated Lyn−/− BMMCLs with an anti-peptide mAb to γ-tubulin, elution of bound proteins with peptide, and concentration of tyrosine-phosphorylated proteins on an immobilized SH2 domain. Second, reciprocal precipitation experiments confirmed an interaction between βPIX or GIT1 and γ-tubulin. Third, TagRFP- or GFP-tagged γ-tubulins interacted with GIT1 and βPIX. Fourth, βPIX and GIT1 were associated with the GST-tagged C-terminal region of γ-tubulin. Finally, live-cell imaging revealed the localization of βPIX and GIT1 to centrosomes, where γ-tubulin is accumulated. Because the identification of βPIX in γ-tubulin complexes was carried out in lysates from Lyn−/− BMMCLs, we also analyzed GITs/βPIX–γ-tubulin complexes in BMMCLs. Although the differential expression of GIT1 and GIT2 was described in mouse tissues (36), our data document the expression of both proteins in BMMCLs and Lyn−/− BMMCLs. The results also demonstrate that the deletion of Lyn kinase does not affect the formation of complexes containing γ-tubulin, βPIX, and GITs.
GITs are multidomain proteins, and several signaling molecules, including βPIX, PAK, focal adhesion kinase, phospholipase Cγ, MAPK 1, and the synaptic protein Piccolo, associate with GIT1 through its Spa2 homology domain (30). Our data demonstrate that GIT1 is substrate for tyrosine kinases in pervanadate or FcεRI aggregation-activated BMMCLs. Moreover, tyrosine-phosphorylated GIT1 forms complexes with γ-tubulin. GIT1 was shown to be phosphorylated in cells in a Src kinase–dependent manner (37), and different studies pointed to the relevance of tyrosine phosphorylation in the regulation of GIT1 functions. It was shown that tyrosine phosphorylation of GIT1 is required for intramolecular conformational changes in GIT1 and release its autoinhibitory interaction (38). Phosphorylation of GIT1 by Src family kinases is required for its association with focal adhesion kinase (39), as well as for phospholipase Cγ activation (40). In contrast, tyrosine phosphorylation of βPIX, in a Src-dependent manner, weakens its ability to bind GIT1 (41). Association of Src family kinases with γ-tubulin complexes was reported in activated RBL-2H3 (35), activated BMMCLs (5), and in differentiating P19 embryonic carcinoma cells (23, 42). Thus, tyrosine phosphorylation of GIT1 in stimulated mast cells might lead to its activation.
There are reports suggesting an important role for tyrosine kinases in the regulation of microtubule nucleation from centrosomes. Fyn kinase was found on centrosomes in myelocytic leukemia cells HL-60 (43) and in human T lymphocytes (44). Androgen and Src signaling modulated microtubule nucleation during interphase by promoting the centrosomal localization of γ-tubulin (13) via activation of the MAPK/Erk signaling pathway (45). It also was shown that Syk is catalytically active at the centrosome (46). Previous studies showed that early stages of BMMCL activation, when microtubule formation is stimulated, are characterized by the concentration of tyrosine-phosphorylated proteins in the centrosomal region (5).
Although the FcεRI proximal signaling pathways involved in mast cell activation are well defined (1), the signaling events downstream of Ca2+ influx are less well studied. Ca2+-dependent kinases or phosphatases might participate in microtubule stability in activated mast cells (47). It was reported that STIM1-regulated Ca2+ influx is essential for reorganization of microtubules in activated mast cells (6). In this study, we show that the depletion of Ca2+ substantially delays microtubule regrowth. In contrast, elevated levels of Ca2+ are accompanied by increased amounts of GCP2, GCP4, and GITs in complexes with γ-tubulin. Interestingly, Ca2+ inhibits binding of βPIX to the C-terminal region of γ-tubulin. Direct Ca2+-dependent interaction between βPIX and calmodulin was reported recently (48). Live-cell imaging with EGFP-calmodulin chimeras revealed the association of calmodulin with centrosomes after activation of RBL-2H3 (49). A truncated splice variant of the FcεRIβ binding calmodulin, Gab2, and Fyn kinase was found to accumulate in the centrosome–Golgi region and stimulate microtubule formation and degranulation in human LAD-2 mast cells. It was suggested that the truncated splice variant may act to propagate Ca2+ signals for microtubule nucleation (7). It also was reported that the Fyn kinase/Gab2/PI3K pathway is critical for the formation of microtubules (3, 50). Our data support the importance of Ca2+ for microtubule nucleation in BMMCLs.
Interestingly, elevated levels of Ca2+ affected γ-tubulin properties manifested by changes in the electrophoretic mobility of γ-tubulin in BMMCLs. This conclusion is supported by several findings. First, both endogenous and exogenous γ-tubulins were modified. Second, 6xHis-tagged whole-length γ-tubulin or its C-terminal part (aa 223–451) was affected. Third, EGFP-tagged whole-length γ-tubulin or its C-terminal region (aa 378–451) was modified. Fourth, experiments with EGFP-tagged C-terminal regions of γ-tubulin or α-tubulin revealed that the mobility shift is specific for γ-tubulin. Finally, the presence of EGTA inhibited the mobility shift of γ-tubulin, and elevated concentrations of Mg2+ or Zn2+ did not demonstrate this effect. It was shown that Ca2+ has high-affinity binding sites on the C-terminal regions of αβ-tubulin dimers (51). However, it is unlikely that only the direct binding of Ca2+ to the C-terminal region of γ-tubulin causes observed changes, because no mobility shift was detected when Ca2+ alone was added to 6xHis-tagged γ-tubulin (Fig. 4A, Control). Unique changes in γ-tubulin properties in BMMCLs warrant further investigation. Ca2+-dependent modifications in the C-terminal region of γ-tubulin could affect γ-tubulin binding characteristics and, thus, modulate the nucleation of microtubules in activated cells.
Our data from live-cell imaging revealed that GIT1 and βPIX are located on centrosomes in BMMCLs, which could reflect their regulatory roles in microtubule nucleation. Although GIT1/βPIX complexes are usually connected with cell migration, the centrosomal function of the complexes has only been reported in fibroblasts. GIT1 targeting to the centrosome served as a scaffold for βPIX. Associated PAK, activated via a process not requiring Rho GTPases, phosphorylated Aurora A in mitosis (52). Our data demonstrate that GIT1 and βPIX regulate interphase microtubules of mast cells. Although GIT1 and βPIX are both associated with centrosomes in BMMCLs, they play opposite regulatory roles in microtubule nucleation from centrosomes. Depletion of βPIX led to increased nucleation, whereas depletion of GIT1 resulted in decreased nucleation. The obtained data were confirmed by rescue experiments. GIT1/βPIX could regulate the nucleation of microtubules via several mechanisms. They may directly promote the assembly of γTuRCs and/or the recruitment of γTuRCs to the centrosome. Thus, kinases associated with GIT1/βPIX may phosphorylate components of γTuRC to promote the assembly of the complex, or they may regulate the association or activity of NEDD1/GCP-WD, the attachment factor that lies most proximal to γTuRC and is required for the centrosomal recruitment of γTuRC (53). Alternatively, GIT1/βPIX may indirectly affect this process by regulating the assembly of the pericentriolar matrix or the activity of a centrosomal protein(s) required for the anchorage of γTuRCs (54).
It is generally thought that the GIT1/βPIX complex serves as a signaling cassette that elicits changes in cell shape and migration (29). Our data indicate that depletion of βPIX enhances degranulation in Ag-activated cells. This could reflect increased microtubule nucleation. In contrast, depletion of GIT1 resulted in reduced degranulation, which could be related to decreased levels of microtubule nucleation in such cells. Both proteins are also involved in chemotaxis toward Ag, where they play opposite regulatory roles: GIT1 is a negative regulator, whereas βPIX serves as a positive regulator of chemotaxis. The negative regulatory role of GIT2 in cell motility was reported in other cell types (55). In contrast, it was shown that fMLP-induced fibronectin-mediated chemotaxis was unaffected after depletion of GIT2 and was inhibited after depletion of GIT1 (56). The difference in the results presented in this study and those of another study (56) may be attributed to the different experimental cell models, different chemoattractants used, and the presence or absence of fibronectin during the chemotactic assays. Although microtubules have long been implicated in cell motility, their role in this process varies with cell types. The mechanism of their involvement in cell motility is poorly understood. It was suggested that microtubules normally act to restrain cell motility (57). This could explain the observed opposite effects in degranulation and chemotaxis after the depletion of GIT1 and βPIX proteins. Because the GIT/PIX complex affects actin-dependent processes, including cell motility (29), the observed changes in chemotaxis could reflect changes in the microtubules, as well as in the actin cytoskeleton.
In conclusion, our data suggest a novel signaling pathway for microtubule rearrangement in mast cells; tyrosine kinase–activated GIT1 and βPIX, in concert with Ca2+ signaling, regulate microtubule nucleation. Enhanced levels of Ca2+ affect γ-tubulin properties, resulting in greater binding of GCPs to γ-tubulin. Presumably, through this action, GIT1 and βPIX are involved in the regulation of such important processes in mast cells physiology as is Ag-induced degranulation and chemotaxis. Interference with the microtubular network via specific regulators of microtubule nucleation in mast cells could open up rational new approaches to the treatment of inflammatory and allergic diseases.
Disclosures
The authors have no financial conflicts of interest.
Acknowledgments
We thank Dr. M. Hibbs for the BMMCLs, Dr. P.L. Hordijk (Sanquin Research and Landsteiner Laboratory, University of Amsterdam, Amsterdam, the Netherlands) for the GST-tagged βPIX construct, and Dr. Petr Dráber for providing BMMCs and the Ab to GST, as well as for critical reading of the manuscript.
Footnotes
This work was supported in part by Grants P302/12/1673, P302/11/P709, P302/14/09807S, and 15-22194S from the Grant Agency of the Czech Republic, Grants LD13015 and LH12050 from the Ministry of Education, Youth, and Sports of the Czech Republic, Grant NT14467 from Ministry of Health of the Czech Republic, Grant GAUK 796913 from the Grant Agency of Charles University, and by Institutional Research Support (RVO 68378050). We also acknowledge support of the COST Action BM1007 Mast Cells and Basophils–Targets for Innovative Therapies.
The online version of this article contains supplemental material.
Abbreviations used in this article:
- BMMC
- bone marrow–derived mast cell
- BMMCL
- BMMC line
- BMMCL-γTb
- BMMCL stably expressing TagRFP-tagged mouse γ-tubulin
- ER
- endoplasmic reticulum
- GCP
- γ-tubulin complex protein
- GIT
- G protein–coupled receptor kinase-interacting protein
- HEK
- HEK 293-FT
- hGIT1
- human GIT1
- hβPIX
- human βPIX
- mGIT1
- mouse GIT1
- mβPIX
- mouse βPIX
- MS
- mass spectrometry
- PAK
- p21-activated kinase
- βPIX
- p21-activated kinase interacting exchange factor β
- P-Tyr
- phosphotyrosine
- ROI
- region of interest
- shRNA
- short hairpin RNA
- STIM1
- stromal-interacting protein 1
- γTuRC
- γ-tubulin ring complex
- tv
- transcript variant.
- Received September 26, 2014.
- Accepted February 27, 2015.
- Copyright © 2015 by The American Association of Immunologists, Inc.
References
In this issue
