Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Signal Regulatory Proteins Negatively Regulate Immunoreceptor-dependent Cell Activation

Signal Regulatory Proteins Negatively Regulate Immunoreceptor-dependent Cell Activation THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 45, Issue of November 5, pp. 32493–32499, 1999 © 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Signal Regulatory Proteins Negatively Regulate Immunoreceptor-dependent Cell Activation* (Received for publication, May 18, 1999, and in revised form, September 3, 1999) He ´le ` ne Lie ´ nard‡, Pierre Bruhns§, Odile Malbec, Wolf H. Fridman, and Marc Dae ¨ ron¶ From the Laboratoire d’Immunologie Cellulaire et Clinique, INSERM U.255, Institut Curie, 75005 Paris, France Signal regulatory proteins of the a subtype (SIRPa) phoproteins that coprecipitated with SH2 domain-bearing pro- tein tyrosine phosphatases (SHPs) (1–3). SIRP molecules were are ubiquitous molecules of the immunoglobulin super- family that negatively regulate protein tyrosine kinase first identified as SHP substrate 1 (SHPS-1) (2) and brain receptor-dependent cell proliferation. Their intracyto- immunoglobulin-like molecules with tyrosine-based activation plasmic domain contains four motifs that resemble im- motifs (BIT) (4). When cloned, these molecules were found to be munoreceptor tyrosine-based inhibition motifs (ITIMs) the members of a multigene family of at least 15 transmem- and that, when tyrosyl-phosphorylated, recruit cyto- brane immunoglobulin superfamily molecules named collec- plasmic SH2 domain-bearing protein tyrosine phos- tively SIRPs, in which two types, a and b, were recognized, phatases (SHPs). ITIMs are borne by molecules that neg- differing by the presence (in SIRPa) or the absence (in SIRPb) atively regulate cell activation induced by receptors of an intracytoplasmic (IC) domain containing four tyrosine- bearing immunoreceptor tyrosine-based activation mo- based regulatory motifs (3). A neuronal adhesion molecule, a are coexpressed with tifs (ITAMs). Because SIRP previously described as P84, was found to belong to the SIRP ITAM-bearing receptors in hematopoietic cells, we in- family, and the widely expressed integrin-associated protein a could negatively regulate vestigated whether SIRP CD47 was recently identified as a ligand of P84 (5). a tran- ITAM-dependent cell activation. We found SIRP Interestingly, SIRPa were shown to regulate negatively cell scripts in human mast cells, and we show that a chi- proliferation induced by growth factors via protein tyrosine meric molecule having the transmembrane and intracy- kinase receptors (RTKs) and oncogene products. Little is a could inhibit IgE-induced toplasmic domains of SIRP known of the mechanism of inhibition by SIRPa, except that mediator secretion and cytokine synthesis by mast cells. negative regulation was correlated with the tyrosyl phospho- a chimera was coag- Inhibition required that the SIRP rylation of SIRPa and the recruitment of SHP-2 (3, 6). That gregated with ITAM-bearing high affinity IgE receptors eRI). It was correlated with the tyrosyl phosphoryla- SIRPa had inhibitory properties indicated that these molecules (Fc a chimera and the recruitment of SHP-1 bear inhibition motifs rather than activation motifs. SHP-2- tion of the SIRP eRI ITAMs was and SHP-2. The phosphorylation of Fc binding motifs found in SIRPa are indeed reminiscent of im- decreased; the mobilization of intracellular Ca and munoreceptor tyrosine-based inhibition motifs (ITIMs). ITIMs the influx of extracellular Ca were reduced, and the are present in a large group of molecules (7) that negatively activation of the mitogen-activated protein kinases regulate cell activation induced by receptors bearing immuno- Erk1 and Erk2 was abolished. SIRPa can therefore neg- receptor tyrosine-based activation motifs (ITAMs) (8). Negative atively regulate not only receptor tyrosine kinase-de- regulation by ITIM-bearing molecules is correlated with the pendent cell proliferation but also ITAM-dependent cell recruitment of SH2 domain-bearing phosphatases by phospho- activation. rylated ITIMs (9, 10). Thus, killer cell inhibitory receptors (KIRs) inhibit cell-mediated cytotoxicity when they bind to major histocompatibility complex class I molecules on target Signal regulatory proteins (SIRPs) were described as phos- cells (11). Their IC domain contains two ITIMs that, when tyrosyl-phosphorylated, recruit SHP-1 and SHP-2 (12, 13). Likewise, FcgRIIB, a family of low affinity receptors for IgG, * This work was supported by the INSERM, the Association pour la Recherche sur le Cancer, and the Institut Curie. The costs of publica- negatively regulate cell activation via ITAM-bearing receptors tion of this article were defrayed in part by the payment of page when coaggregated with the latter by immune complexes (8). charges. This article must therefore be hereby marked “advertisement” FcgRIIB bear a single ITIM that, when tyrosyl-phosphorylated, in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. recruits selectively the SH2 domain-bearing inositol-5-phos- ‡ Recipient of a Rho ˆ ne-Poulenc Rorer CIFRE contract. § Recipient of a fellowship from the Ministe ` re de l’Enseignement phatase SHIP (14). We found recently that FcgRIIB are also Supe ´ rieur et de la Recherche. capable of inhibiting RTK-dependent cell proliferation (15). To whom correspondence should be addressed: Laboratoire This finding opened the possibility that it might be a general d’Immunologie Cellulaire et Clinique, INSERM U.255, Institut Curie, property of ITIM-bearing molecules to regulate negatively not 26 rue d’Ulm, 75005 Paris, France. Tel.: 33-1-4432-4223; Fax: 33-1- 4051-0420; E-mail: [email protected]. only ITAM-dependent cell activation but also RTK-mediated The abbreviations used are: BIT, brain immunoglobulin-like mole- cell proliferation. If so, one could hypothesize that SIRPa might eRI, high affinity IgE cule with tyrosine-based activation motifs; Fc inhibit ITAM-dependent cell activation. receptors; FcgRIIB and FcgRIIIA, low affinity IgG Receptors; GAM, By contrast with KIRs whose expression is restricted to NK goat anti-mouse Ig; GAR, goat anti-rabbit Ig; HRP, horseradish perox- cells and T cells (11), SIRPa were found to be expressed by all idase; IC, intracytoplasmic; ITAM, immunoreceptor tyrosine-based ac- tivation motif; ITIM, immunoreceptor tyrosine-based inhibition motif; human tissues examined, including hematopoietic cells (3), by KIRs, killer cell inhibitory receptors; MAR, mouse anti-rat Ig; RTK, receptor tyrosine kinase; SH2, Src homology domain 2; SHIP, SH2 domain-bearing inositol-phosphate phosphatase; SHP, SH2 domain- BSA, bovine serum albumin; mAb, monoclonal antibody; FITC, fluores- bearing protein tyrosine phosphatase; SHPS-1, SH2 domain-bearing cein isothiocyanate; PAGE, polyacrylamide gel electrophoresis; RT- phosphatase substrate 1; SIRPs, signal regulatory proteins; TNP, tri- PCR, reverse transcriptase-polymerase chain reaction; TNF, tumor ne- nitrophenyl; MAP, mitogen-activated protein; DNP, 2,4-dinitrophenol; crosis factor; TM, transmembrane. This paper is available on line at http://www.jbc.org 32493 This is an Open Access article under the CC BY license. 32494 Inhibition of Cell Activation by Signal Regulatory Proteins (HRP)-conjugated mouse mAb anti-phosphotyrosine (PY20) were pur- neural and myeloid cells in rats (16), and by myeloid cells, chased from Chemicon (Temecula, CA); mouse mAbs anti-SHP-1 and especially macrophages, but not lymphoid cells in mice (17). anti-SHP-2 were from Transduction Laboratories (Lexington, KY); Hematopoietic cells also express ITAM-bearing receptors. mouse mAbs anti-Erk1/2 and anti-phospho-Erk1/2 were from New Eng- Thus, myeloid cells express ITAM-bearing receptors for the Fc land Biolabs (Beverly, MA); HRP-conjugated polyclonal goat anti- portion of immunoglobulins (FcR) (18, 19). These comprise high mouse Ig (GAM) antibodies and polyclonal goat anti-rabbit Ig (GAR) affinity IgE receptors (FceRI) (20), high affinity IgG receptors antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Immunofluorescence—RBL transfectants were incubated for1hat (FcgRI) (21), high affinity IgA receptors (FcaRI) (22), and low 0 °C with 10 mg/ml 2.4G2 mAb in balanced salt solution containing 5% affinity IgG receptors (FcgRIIIA) (23), which all share the fetal calf serum. Cells were washed and stained by being incubated for ITAM-bearing signal transduction subunit FcRg (18). They 30 min at 0 °C with 50 mg/ml FITC-MAR F(ab9)2. Fluorescence was also include the human-restricted, single chain, low affinity analyzed by flow cytometry using a FACScalibur (Becton-Dickinson, IgG receptors FcgRIIA/C that bear one ITAM in their own IC Mountain View, CA). domain (18). If they could interfere with ITAM-dependent cell RT-PCR Analysis of SIRPa Transcripts—RNA was extracted from 1 3 10 KU812, HMC-1, and cord blood-derived human mast cells. activation, SIRPa would negatively regulate not only the cDNAs were prepared and used as templates to amplify sequences growth but also the many biological functions of hematopoietic corresponding to the TM and IC domains of SIRPa using the two cells triggered by Fc receptors. oligonucleotides SIRPa sense and antisense. Amplified fragments were To investigate the hypothesis that SIRPa might negatively sequenced using the same two oligonucleotides. regulate cell activation induced by ITAM-bearing receptors, we Serotonin Release—RBL transfectants, loaded with [ H]serotonin constructed an experimental model in mast cells. Mast cells (Amersham Pharmacia Biotech), were incubated for1hat37 °C with IgE anti-DNP and with 0 or 3 mg/ml 2.4G2 F(ab9)2, washed, and chal- were chosen as an assay system because we found SIRPa lenged for 30 min at 37 °C with TNP-MAR F(ab9)2 or with MAR F(ab92) transcripts in human mast cells. We report here that, when and DNP-BSA. The percentage of [ H]serotonin released was measured coaggregated with FceRI, a chimeric molecule whose IC domain as described (30). was that of human SIRPa could inhibit IgE-induced mediator TNF Secretion—RBL transfectants, incubated for1hat37 °C with release and cytokine secretion by mast cells. The experimental IgE anti-DNP and 0 or 3 mg/ml 2.4G2 F(ab9)2, were challenged for 3 h model used made possible the biochemical analysis of intracel- at 37 °C with 10 mg/ml TNP-MAR F(ab9)2. Cell-free supernatants were harvested and titrated for TNFa on L929 cells as described (33). lular events associated with inhibition. Inhibition was corre- Measurement of Intracellular Ca Concentration—RBL transfec- lated with the tyrosyl phosphorylation of the SIRPa chimera, tants, previously sensitized with IgE anti-DNP and incubated with or with the recruitment of SHP-1 and SHP-2 by the phosphoryl- without 2.4G2 F(ab9)2, were loaded with 5 mM Fluo-3-AM in the pres- ated chimera, with a decreased phosphorylation of FceRI ence of 0.2% Pluronic F-127 (Molecular Probes, Eugene, OR) for 30 min ITAMs, with attenuated Ca responses, and with an abolition at room temperature. Cells were resuspended in medium in which of MAP kinase activation. calcium had been buffered to 60 nM (equivalent to the intracellular Ca concentration in resting cells) with EGTA. One min later, they were EXPERIMENTAL PROCEDURES challenged with TNP-MAR F(ab9)2 for 100 s, after which the extracel- cDNA Constructs—-cDNA sequences encoding the transmembrane lular Ca concentration was raised to 1.3 mM with CaCl . Intracellular (TM) and IC domains of SIRPa and of a human KIR were amplified free Ca concentration was monitored by flow cytometry with a FAC- from the human melanoma cell line HT-144 (from the ATCC) and from Scalibur using the software FCS assistant 1.2.9b (Becton Dickinson). the p58.183 KIR cDNA (24) by RT-PCR and PCR, respectively, with the Immunoprecipitation and Western Blot Analysis—RBL transfectants following oligonucleotide primers: for SIRPa, sense, 59-TCTAAGGTAC- were incubated with IgE anti-DNP and 0 or 3 mg/ml 2.4G2 F(ab9)2, CAAACATCTATATTGTGGTG-39, and antisense, 59-AGCAAACCGAG- washed, and challenged for various periods of time at 37 °C with 10 CTCCCATTCACTTCCTCGGGACCTG-39; for KIR, sense, 59-CCCA- mg/ml TNP-MAR F(ab9)2. Aliquots of 3 3 10 cells were lysed for 10 min GAC AGGTACCTGTTCTGATTGGGACC-39, and antisense, 59-CTGA- at 0 °C. For immunoprecipitation of FcgRIIB, lysis buffer contained 10 CTGTGGAGCTCATGGGCAGG-39. KpnI (GGTACC) and SacI (GAGC- mM Tris, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 1 mM Na VO ,5mM 3 4 TC) sites are in boldface. PCR products were inserted into an expression NaF, 5 mM sodium pyrophosphate, 0.4 mM EDTA, 10 mg/ml aprotinin, cassette under the control of the SRa promoter in pBR322 (25), con- 10 mg/ml leupeptin, 10 mg/ml pepstatin, and 1 mM phenylmethylsulfo- taining a resistance gene to neomycin (NT-neo) and containing the nyl fluoride. For immunoprecipitation of FceRI, lysis buffer contained extracellular domain of FcgRIIB. 50 mM Tris, pH 8.0, 1% Nonidet P-40, 1 mM Na VO ,20mM EDTA, 10 3 4 Cells—The human basophil-like KU812 (26), the human mast cells mg/ml aprotinin, 10 mg/ml leupeptin, and 1 mM phenylmethylsulfonyl HMC-1 (27), and the rat mast cells RBL-2H3 (28) were cultured in fluoride. Lysates were centrifuged at 12,000 rpm for 10 min at 4 °C, and Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf supernatants were incubated with immunoadsorbents for1hat4 °C. serum, 100 IU/ml penicillin, 100 mg/ml streptomycin, and 2 mML- Protein G-Sepharose or 2.4G2 coupled to protein G-Sepharose was used glutamine. Culture reagents were from Life Technologies, Inc. Human to precipitate FcgRIIB from cells preincubated with 2.4G2 or not, re- normal mast cells, obtained by culture of cord blood mononuclear cells spectively. Anti-mouse IgE antibodies coupled to protein A-Sepharose as described (29), were kindly donated by Dr. M. Arock (Faculte´de (Sigma) were used to precipitate FceRI from cells sensitized with mouse Pharmacie, Paris, France). More than 99% cells were mast cells as IgE. Immunoprecipitates were fractionated by SDS-PAGE and trans- judged by morphology, the presence of metachromatic granules, and ferred onto Immobilon-P membranes (Millipore, Bedford, MA). Mem- positivity for tryptase. cDNAs encoding FcgRIIB-SIRPa and FcgRIIB- branes were saturated either with 5% BSA (Sigma) or with 5% skimmed KIR were stably transfected in RBL-2H3 cells; transfectants were se- milk (Re ´ gilait, Saint-Martin-Belle-Roche, France) and Western-blotted lected with 2.5 mg/ml G418 (Life Technologies, Inc.) and cloned as with HRP-conjugated anti-phosphotyrosine antibodies or with rabbit described (30). anti-FcgRIIB, mouse anti-SHP-1, mouse anti-SHP-2, mouse anti-FcRb, Antibodies—The mouse IgE anti-DNP mAb 2682-I (31) was used as or rabbit anti-FcRg antibodies, followed by HRP-conjugated GAR or culture supernatant. IgG and F(ab9)2 fragments of the rat anti-mouse GAM. Anti-SHP-2 antibodies detected two species of SHP-2. Anti- FcgRIIB 2.4G2 mAb (32) were obtained as described (30). Rabbit anti- SHP-1 and anti-SHP-2 antibodies recognized specifically SHP-1 and mouse FcgRIIB were raised by Dr. C. Saute ` s (Institut Curie, Paris, SHP-2, respectively. HRP-conjugated antibodies were detected using an France) against recombinant extracellular domains of FcgRIIB. Mouse enhanced chemiluminescence (ECL) kit (Amersham Pharmacia Bio- mAb anti-FcRb were purified with protein G-Sepharose (Amersham tech). When blotted sequentially with different antibodies, filters were Pharmacia Biotech) from culture supernatants of the JRK hybridoma stripped following revelation by being incubated for 30 min at 50 °C in cells. Rabbit anti-FcRg antibodies and rabbit anti-mouse IgE antibodies buffer containing 62.5 mM Tris, pH 6.7, 100 mM b-mercaptoethanol, and were generous gifts of Dr. J.-P. Kinet (Beth Israel Hospital, Boston). 2% SDS. IgG and F(ab9)2 fragments of mouse anti-rat Ig (MAR) and FITC- Western Blot Analysis of MAP Kinases—RBL transfectants were conjugated MAR F(ab9)2 fragments (FITC-MAR F(ab9)2) were from incubated for1hat37 °C with IgE anti-DNP and with 0 or 3 mg/ml Jackson ImmunoResearch Laboratories (West Grove, PA). MAR F(ab9)2 2.4G2 F(ab9)2, washed, and challenged with 10 mg/ml TNP-MAR were trinitrophenylated with trinitrobenzene sulfonic acid (Eastman- F(ab9)2. Equal volumes of cell lysates, obtained as described for immu- Kodak Co.) and purified on Sephadex G-25 (Amersham Pharmacia noprecipitation of FcgRIIB, were fractionated by SDS-PAGE, trans- Biotech). TNP -MAR F(ab9)2 were obtained. Horseradish peroxidase ferred onto Immobilon-P membranes, and blotted with anti-phospho- 10 Inhibition of Cell Activation by Signal Regulatory Proteins 32495 FIG.2. Inhibition of IgE-induced mast cell activation by FcgRIIB-SIRPa. A, inhibition of serotonin release. Cells expressing FIG.1. SIRPa transcripts in human mast cells and structure FcgRIIB-SIRPa,FcgRIIB-KIR, or FcgRIIB(IC1) were sensitized with and expression of recombinant molecules expressed in RBL indicated dilutions of IgE anti-DNP and with 0 (open circles)or3 mg/ml cells. A, SIRPa transcripts in human mast cells. cDNAs from a human (closed circles) 2.4G2 F(ab9)2. Cells were challenged with 10 mg/ml basophil-like cell line (KU812), a human mast cell line (HMC-1), and TNP-MAR F(ab9)2. The figure shows the percentage of serotonin re- cord blood-derived human mast cells (CB-HMC) were analyzed by RT- leased as a function of IgE dilutions. B, requirement for the coaggrega- PCR using oligonucleotides that could amplify sequences encompassing tion of FceRI with FcgRIIB-SIRPa for inhibition of serotonin release. the TM and IC domains of SIRPa. PCR products were analyzed by Cells expressing FcgRIIB-SIRPa were sensitized with indicated dilu- agarose gel electrophoresis. bp, base pair. B, structure of recombinant tions of IgE anti-DNP and with 0 (open symbols)or3 mg/ml (closed molecules. The figure schematizes the structure of the IC domain- symbols) 2.4G2 F(ab9)2. Cells were challenged either with 10 mg/ml deleted FcgRIIB(IC1), of the FcgRIIB-SIRPa chimera, and of the TNP-MAR F(ab9)2 (circles) or with 3 mg/ml DNP-BSA and 10 mg/ml FcgRIIB-KIR chimera expressed in RBL cells. ITIM-like sequences are MAR F(ab9)2 (squares). The figure shows the percentage of serotonin indicated at their respective positions. C, expression of recombinant released as a function of IgE dilutions. C, inhibition of TNFa produc- molecules in RBL-2H3 cells. Clones of stable transfectants were exam- tion. Cells expressing FcgRIIB-SIRPa were sensitized with IgE anti- ined by immunofluorescence for the expression of FcgRIIB-based mol- DNP (supernatant 1/10) and incubated with 0 (open circles)or3 mg/ml ecules. Filled histograms, cells incubated with FITC-MAR F(ab9)2 only; (closed circles) 2.4G2 F(ab9)2. They were challenged with 10 mg/ml open histograms, cells incubated with 2.4G2 and FITC-MAR F(ab9)2. TNP-MAR F(ab9)2for3hat37 °C. Cell-free supernatants were har- vested, and serial 2-fold dilutions were tested for cytotoxicity on L929 cells. The figure represents the percentage of cytotoxicity as a function Erk1/2 or with anti-Erk1/2 antibodies, followed by HRP-conjugated of the dilution of supernatants. Experiments shown in A–C were re- GAR. HRP-conjugated antibodies were detected using the ECL kit. peated three times with the same results. RESULTS Human Mast Cells Express SIRPa—We investigated first gregated with FceRI in RBL transfectants sensitized with the presence of SIRPa transcripts in RNA from human mast mouse IgE anti-DNP, incubated with F(ab9)2 fragments of the cells by RT-PCR using oligonucleotide primers that could am- rat anti-mouse FcgRIIB mAb 2.4G2, and challenged with TNP- plify a 435-base pair long fragment corresponding to the TM MAR F(ab9)2 as described previously (34). Serotonin release and IC domains of human SIRPa. A single PCR product having induced under these conditions was compared with serotonin the expected size was amplified from cDNAs from the human release induced by aggregating FceRI with TNP-MAR F(ab9)2 mast cell line HMC-1 and from cord blood-derived human mast in the same three transfectants sensitized with mouse IgE cells but not from the basophil-like cell line KU812 (Fig. 1A). anti-DNP but not incubated with 2.4G2 F(ab9)2. When coag- PCR products amplified from HMC-1 and from cord blood- gregated with FceRI, FcgRIIB-SIRPa inhibited serotonin re- derived mast cells were sequenced. They had the same nucle- lease as efficiently as FcgRIIB-KIR, but not FcgRIIB(IC1) otide sequence as cDNA encoding the TM and IC domains of (Fig. 2A). human SIRPa (3). To determine the conditions required for FcgRIIB-SIRPa to A SIRPa Chimera Inhibits IgE-induced Mast Cell Activation inhibit serotonin release, transfectants expressing FcgRIIB- When Coligated with FceRI—In order to examine whether SIRPa were sensitized with mouse IgE anti-DNP and incu- SIRPa could inhibit IgE-induced mast cell activation, we con- bated with 2.4G2 F(ab9)2. FceRI and FcgRIIB-SIRPa were then structed a cDNA encoding a chimeric molecule having the either coaggregated by TNP-MAR F(ab9)2 as above or aggre- extracellular domain of murine FcgRIIB and the TM and IC gated simultaneously, but independently, by DNP-BSA and domains of human SIRPa (FcgRIIB-SIRPa). As a positive con- MAR F(ab9)2, respectively. Cells sensitized with IgE but not trol, a cDNA encoding a chimeric molecule having the extra- incubated with 2.4G2 F(ab9)2 served as positive controls. Sero- cellular domain of murine FcgRIIB and the TM and IC domains tonin release was inhibited when FceRI and FcgRIIB-SIRPa of a human KIR (FcgRIIB-KIR) was also constructed (Fig. 1B). were coaggregated by TNP-MAR F(ab9)2 but not when they Chimeric cDNAs were stably transfected in the rat mast cells were simultaneously aggregated by DNP-BSA and MAR RBL-2H3, which constitutively express FceRI. Clones of trans- F(ab9)2 (Fig. 2B). fectants expressing comparable amounts of FcgRIIB-SIRPa or To determine whether FcgRIIB-SIRPa could inhibit an IgE- FcgRIIB-KIR (Fig. 1C) and, as negative controls, clones ex- induced secretion of cytokine, FceRI were either aggregated or pressing tail-less FcgRIIB (FcgRIIB(IC1)) (30, 33) were used coaggregated with FcgRIIB-SIRPa in transfectants expressing for experiments. Several clones expressing each molecule gave FcgRIIB-SIRPa, using the same ligands as above, and the comparable results. amount of TNFa released in culture supernatants during the FcgRIIB-SIRPa,FcgRIIB-KIR, or FcgRIIB(IC1) were coag- following 3 h was titrated using a bioassay. When coaggregated 32496 Inhibition of Cell Activation by Signal Regulatory Proteins coaggregated with FceRI for various periods, its phosphoryla- tion was apparent at 1 min, maximum between 3 and 10 min, and still visible at 30 min (Fig. 3B). When coaggregated with FceRI, tail-less FcgRIIB failed to be phosphorylated (data not shown). Four tyrosine residues are present in the IC domain of SIRPa, which are each constitutive of an ITIM-like motif. To identify SH2 domain-bearing phosphatases possibly re- cruited in vivo by FcgRIIB-SIRPa, the SIRPa chimera was immunoprecipitated after it had or had not been coaggregated with FceRI, and phosphatases coprecipitated with FcgRIIB- SIRPa were identified by Western blotting. Some SHP-2 copre- cipitated with weakly phosphorylated FcgRIIB-SIRPa in rest- ing cells, and high amounts of SHP-2 coprecipitated with FcgRIIB-SIRPa that became heavily phosphorylated upon co- aggregation with FceRI. SHP-1 was coprecipitated with FcgRIIB-SIRPa following coaggregation of the chimera with FceRI only. Under the same conditions, FcgRIIB-KIR, which was also phosphorylated upon coaggregation with FceRI, re- cruited similar amounts of SHP-1 as FcgRIIB-SIRPa but much less SHP-2. SHIP was recruited neither by FcgRIIB-SIRPa nor by FcgRIIB-KIR (Fig. 3C). The SIRPa Chimera Blocks Signal Transduction by FceRI—An early step in IgE-induced signaling in mast cells is the tyrosyl phosphorylation of FceRI ITAMs (35) which enables the recruitment of SH2 domain-bearing cytoplasmic kinases, FIG.3. Phosphorylation of FcgRIIB-SIRPa upon coaggrega- leading to an increased intracellular Ca tion with FceRI and coprecipitation of SHP-1 and SHP-2 with concentration, and phosphorylated FcgRIIB-SIRPa. RBL transfectants were preincu- adapter molecules that connect phosphorylated receptors with bated with or without 3 mg/ml 2.4G2 F(ab9)2, sensitized or not with IgE the Ras pathway, leading to the transcription of cytokine anti-DNP (supernatant 1/10), and challenged with or without 10 mg/ml genes. In RBL cells, FceRI are associated with two ITAM- TNP-MAR F(ab9)2. A, phosphorylation of FcgRIIB-SIRPa upon coag- bearing subunits, FcRb and FcRg (20). gregation with FceRI. Aliquots of 1 3 10 cells were lysed 3 min after stimulation. FcgRIIB-SIRPa immunoprecipitates were fractionated by The phosphorylation of FceRI ITAMs was examined by West- SDS-PAGE (10% acrylamide), transferred onto Immobilon, and West- ern blot analysis with anti-phosphotyrosine antibodies in RBL ern-blotted with anti-phosphotyrosine antibodies. The filter was transfectants sensitized with IgE anti-DNP, after FceRI were stripped and reblotted with anti-FcgRIIB antibodies. Comparable re- aggregated or coaggregated with the SIRPa chimera for 1, 3, or sults were obtained in three separate experiments. B, kinetics of FcgRIIB-SIRPa phosphorylation. Aliquots of 3 3 10 cells were lysed at 10 min at 37 °C and immunoprecipitated with anti-IgE anti- indicated times. FcgRIIB-SIRPa immunoprecipitates were fractionated bodies. FcRb and FcRg were identified by Western blot analysis by SDS-PAGE (10% acrylamide), transferred onto Immobilon, and of the same filter with corresponding specific antibodies. The Western-blotted with anti-phosphotyrosine antibodies. The filter was phosphorylation of both FcRb and FcRg induced upon FceRI stripped and reblotted with anti-FcgRIIB antibodies. Comparable re- sults were obtained in two separate experiments. C, coprecipitation of aggregation was decreased upon coaggregation of FceRI with SHPs with phosphorylated FcgRIIB-SIRPa. Aliquots of 3 3 10 cells the SIRPa chimera (Fig. 4A). were lysed 3 min after stimulation, and FcgRIIB-SIRPa was immuno- 21 Ca responses were monitored in RBL transfectants follow- precipitated. Left, aliquots of immunoprecipitates corresponding to 3 3 6 eRI aggregation or following the coaggregation of FceRI ing Fc 10 RBL cells were fractionated by SDS-PAGE (10% acrylamide), trans- ferred onto Immobilon, and Western-blotted with anti-phosphotyrosine with the SIRPa chimera and, as a positive control, with the antibodies, stripped and reblotted with anti-FcgRIIB antibodies. Right, KIR chimera, under conditions that permitted us to measure remaining immunoprecipitates were fractionated by SDS-PAGE (8% 21 the mobilization of intracellular Ca and the influx of extra- acrylamide), transferred onto Immobilon, and Western-blotted with cellular Ca separately. FcgRIIB-SIRPa and FcgRIIB-KIR anti-SHIP (upper panel) or with anti-SHP-2 and then anti-SHP-1 with- similarly reduced both the intracellular Ca mobilization and out stripping (lower panel). The same blot was stripped and reblotted with anti-FcgRIIB antibodies. Whole cell lysates (WCL) were used as the extracellular Ca influx (Fig. 4B). positive controls for Western blotting (only lysates from transfectants The activation of the Ras pathway was assessed by examin- expressing FcgRIIB-SIRPa are shown). Comparable results were ob- ing the phosphorylation of the MAP kinases Erk1 and Erk2. tained in two separate experiments. When coaggregated with FceRI, both FcgRIIB-SIRPa and FcgRIIB-KIR abolished IgE-induced Erk1/2 phosphorylation with FceRI, FcgRIIB-SIRPa inhibited about 90% of the TNFa induced by aggregating FceRI (Fig. 4C). secretion (Fig. 2C). DISCUSSION The SIRPa Chimera Becomes Tyrosyl-phosphorylated upon Coligation with FceRI and Recruits Protein Tyrosine Phos- SIRPa were described as negative regulators of growth phatases—We examined next the phosphorylation of FcgRIIB- factor-induced, RTK-mediated cell proliferation (3). RTKs SIRPa when coaggregated with FceRI or not. FcgRIIB-SIRPa transphosphorylate each other when dimerized by growth fac- was immunoprecipitated from transfectants expressing tors (36). They induce cells to enter the cell cycle and to divide. FcgRIIB-SIRPa that had been incubated with medium alone, ITAM-bearing receptors have no intrinsic protein kinase activ- IgE anti-DNP and/or 2.4G2 F(ab9)2, and challenged with or ity, and they do not induce cells to proliferate. When aggre- without TNP-MAR F(ab9)2 for 3 min at 37 °C. Western blot gated by multivalent ligands, they are phosphorylated by Src analysis with anti-phosphotyrosine antibodies showed that the family protein tyrosine kinases, and phosphorylated ITAMs basal phosphorylation of the chimera slightly increased when serve as docking sites for cytoplasmic SH2 domain-bearing FcgRIIB-SIRPa was aggregated but not when FceRI were ag- protein tyrosine kinases and adapter molecules that lead to cell gregated. It increased dramatically when FcgRIIB-SIRPa was activation (37). Although both early and late signals delivered coaggregated with FceRI (Fig. 3A). When FcgRIIB-SIRPa was by RTKs and ITAM-bearing receptors are different, these two Inhibition of Cell Activation by Signal Regulatory Proteins 32497 FIG.4. FcgRIIB-SIRPa inhibits intracellular signaling by FceRI. A, inhibition of the phosphorylation of FceRI ITAMs. Aliquots of 5 3 10 RBL transfectants expressing FcgRIIB-SIRPa were sensitized with IgE anti-DNP (supernatant 1/10), preincubated with or without 3 mg/ml 2.4G2 F(ab9)2, and challenged with or without 10 mg/ml TNP-MAR F(ab9)2 for 1, 3, or 10 min at 37 °C. FceRI immunoprecipitates were fractionated by SDS-PAGE (12.5% acrylamide), transferred onto Immobilon, and Western-blotted with anti-phosphotyrosine antibodies. The blot was then cut into two parts for blotting with anti-FcRb or anti-FcRg antibodies, respectively. Comparable results were obtained in three separate experiments. B, inhibition of Ca mobilization. RBL transfectants expressing FcgRIIB-SIRPa or FcgRIIB-KIR were sensitized with IgE anti-DNP (supernatant 1/10) and preincubated with (gray) or without (black) 3 mg/ml 2.4G2 F(ab9)2. They were loaded with Fluo-3-AM, resuspended in medium containing 21 21 60 nM Ca , and stimulated with 10 mg/ml TNP-MAR F(ab9)2 (arrows). One hundred seconds later, the extracellular Ca concentration was raised to 1.3 mM with CaCl . The figure represents the mean fluorescence of cells as a function of time. Comparable results were obtained in three separate experiments. C, inhibition of Erk1/2 activation. RBL transfectants expressing FcgRIIB-SIRPa or FcgRIIB-KIR were preincubated with or without 3 mg/ml 2.4G2 F(ab9)2, sensitized or not with IgE anti-DNP (supernatant 1/10), challenged with or without TNP-MAR F(ab9)2 for 10 min, and lysed. Whole cell lysates were fractionated by SDS-PAGE (10% acrylamide), transferred onto Immobilon, and Western-blotted with anti-phospho-Erk1/2. The blot was then reblotted with anti-Erk1/2 antibodies without stripping. Comparable results were obtained in two separate experiments. types of receptors utilize common intracellular effectors such as FceRI and FcgRIIB-SIRPa were coaggregated by the same phospholipase C-g, phosphatidylinositol 3-kinase, and mole- extracellular ligand. A possible explanation for the different cules of the Ras/MAP kinase pathway (37, 38). We found SIRPa requirements for SIRPa to inhibit RTK-dependent cell prolif- transcripts in human mast cells, and we investigated the ef- eration and for FcgRIIB-SIRPa to inhibit FceRI-dependent cell fects of a SIRPa chimera on the secretion of inflammatory activation might be that SIRPa constitutively associate with mediators triggered by FceRI, a typical ITAM-bearing Fc re- RTKs (and possibly with FceRI) via their extracellular domains ceptor, in the rat mast cell line RBL-2H3. We provide here the which were removed in the FcgRIIB-SIRPa chimera. Specific first evidence that SIRPa can behave as ITIM-bearing mole- conditions under which FceRI and SIRPa could be coligated, if cules that negatively regulate ITAM-dependent cell activation, not constitutively associated, on mast cells remain to be deter- and we document the mechanism of inhibition by SIRPa. mined, and it was not an aim of this investigation to address The experimental model used to demonstrate the inhibitory that issue. Our work indicates, however, that a previously properties of SIRPa was validated by our finding that human undemonstrated negative regulation may operate in hemato- mast cells contain SIRPa transcripts. This conclusion could be poietic cells that coexpress SIRPa and ITAM-bearing receptors, drawn from results obtained with both cord blood-derived hu- provided the two receptors are maintained close to each other, man mast cells and with the human mastocytoma cells HMC-1. either constitutively or inducibly by common extracellular li- Cord blood-derived mast cells present the advantage of being gands. One SIRP member, the neuronal adhesion molecule normal, non-transformed cells but have the disadvantage of P84, was recently assigned an extracellular ligand. The inte- being potentially contaminated by (less than 1%) other cells, grin-associated protein CD47 was indeed found to bind to P84, possibly including macrophages that express high levels of and both molecules are colocalized in synapse-rich structures of SIRPs. HMC-1 cells maintained in culture for years have the the cerebellum and the retina where their interactions were disadvantage of being transformed cells, but they have the proposed to control synaptic functions (5). Whether CD47 is advantage of excluding any cell contamination. The combined also a ligand for other SIRP family members and/or whether results obtained in the two cell types make it reasonable to other ligands bind to the polymorphic extracellular domains of conclude that human mast cells do express the SIRPa gene, SIRPa is unknown. CD47 is widely expressed on T cells (40), on and although the expression of SIRPa proteins was not for- myeloid cells including neutrophils (41) and mast cells (42), on mally demonstrated here, they justify that mast cell secretory epithelial cells (41), on spleen, liver and bone marrow stromal responses were chosen as readouts to assess the ability of cells (43), on neurones (5, 16), and on red blood cells (44). SIRPa to control cell activation. The mechanism underlying the inhibitory properties of The overexpression of SIRPa was reported to be sufficient to SIRPs is poorly understood. One reason is the constitutive inhibit epidermal growth factor-, platelet-derived growth fac- association of RTKs with SIRPs that made it difficult to ana- tor-, or insulin-induced cell proliferation in NIH-3T3 fibro- lyze the respective roles of interacting molecules. The experi- blasts and ligand-independent proliferation of the same cells mental model used here enabled us to dissect various stages of infected by a retrovirus carrying an oncogenic form of RTK (3). the process and to get some insight in the mechanism of inhi- SIRPa are therefore likely to be constitutively associated with bition by SIRPa. One consequence of the coaggregation of RTKs, and indeed, BIT coprecipitated with CSF-1R in a macro- FceRI with FcgRIIB-SIRPa was a dramatic tyrosyl phosphoryl- phage cell line (39). By contrast, the overexpression of the ation of the chimera. A faint basal phosphorylation of FcgRIIB- FcgRIIB-SIRPa chimera in RBL-2H3 cells affected neither the SIRPa was observed in resting cells. This phosphorylation growth of transfectants (data not shown) nor IgE-induced re- slightly increased upon FcgRIIB-SIRPa aggregation, suggest- sponses. Mediator release was inhibited when and only when ing that low levels of protein tyrosine kinase may associate 32498 Inhibition of Cell Activation by Signal Regulatory Proteins with the chimera in mast cells. It did not increase upon stim- act in concert, and their complementary effects might explain the deep inhibition of the biological responses observed. What- ulation by IgE, confirming that FcgRIIB-SIRPa was not asso- ciated with FceRI. The increased phosphorylation of FcgRIIB- ever the respective roles of the two phosphatases, our work documents a reciprocal cross-talk between SIRPa and ITAM- SIRPa, when coaggregated with FceRI, is likely to depend on bearing receptors during inhibition of cell activation. The co- Src protein tyrosine kinases that are recruited by aggregated aggregation of FcgRIIB-SIRPa with FceRI may indeed enable FceRI, as it was previously demonstrated for the tyrosyl phos- both FceRI to provide protein tyrosine kinases that could phos- phorylation of FcgRIIB (34). phorylate SIRPa ITIMs and FcgRIIB-SIRPa to provide protein Phosphorylated FcgRIIB-SIRPa was found to recruit SH2 tyrosine phosphatases that could dephosphorylate FceRI domain-bearing cytoplasmic protein tyrosine phosphatases. ITAMs and sequester phosphatases possibly required for cell The respective roles of the four ITIM-like motifs in phospha- activation. tase recruitment is not known as no mutational analysis of the Myeloid cells, such as macrophages and mast cells, coexpress four tyrosine residues contained in the SIRPa IC domain has not only ITAM-bearing FcR and SIRPa but also RTKs and been made. Some SHP-2 coprecipitated with lightly phospho- FcgRIIB. We found recently that FcgRIIB, which were known rylated FcgRIIB-SIRPa in resting cells and much greater to regulate negatively ITAM-dependent cell activation (8), can amounts with heavily phosphorylated FcgRIIB-SIRPa, follow- also negatively regulate RTK-dependent cell proliferation and ing its coaggregation with FceRI. Significant amounts of SHP-1 that inhibition is correlated with the recruitment of the SH2 also coprecipitated with the phosphorylated chimera. Associa- domain-bearing inositol-5-phosphatase SHIP (15). We show tion with SHP-2 was instrumental for the identification of here that SIRPa, which were known to regulate negatively SIRPs (1–3). SHP-1 associated with SIRPa when overex- RTK-dependent cell proliferation, can also negatively regulate pressed in fibroblasts (2) and with SHPS-1 (17) and BIT (39) in ITAM-dependent cell activation and that inhibition is corre- macrophages. Noticeably, comparable amounts of SHP-1 copre- lated with the recruitment of SHP-2 and SHP-1. It follows 1) cipitated with FcgRIIB-SIRPa and with FcgRIIB-KIR, whereas that SIRPa bear typical ITIMs which endow these molecules much more SHP-2 coprecipitated with FcgRIIB-SIRPa than with a wide array of previously unsuspected regulatory prop- with FcgRIIB-KIR. This indicates that FcgRIIB-SIRPa prefer- erties, and 2) that ITIMs can negatively regulate both cell entially recruited SHP-2, whereas FcgRIIB-KIR preferentially activation and cell proliferation via the recruitment of either recruited SHP-1, when coligated with the same receptors in SHPs or SHIP. RTK-dependent cell proliferation and ITAM-de- RBL cells. In apparent contradistinction with our observation, pendent cell activation can therefore be negatively regulated, SHPS-1 was reported to associate preferentially with SHP-1 in through common mechanisms, by the same receptors which a mouse macrophage cell line, when phosphorylated following may coordinately control the development and the functions of treatment with an analog of pervanadate, and in resting mouse hematopoietic cells. spleen cells (17). Whether the discrepancy can be explained by differences in the cell types and/or the experimental conditions Acknowledgments—We thank Dr. M. Arock (Faculte ´ de Pharmacie, Paris, France) for cord blood-derived human mast cells; Dr. C. Saute `s needs to be clarified. (Institut Curie, Paris, France) for rabbit anti-FcgRIIB antibodies; Dr. We found that FcgRIIB-SIRPa turned off signals transduced J.-P. Kinet (Beth Israel Hospital, Boston, MA) for rabbit anti-FcRg and by FceRI. The inhibition of serotonin release and of TNFa anti-mouse IgE antibodies; and Dr. E. Tartour (Institut Curie, Paris, synthesis was correlated with attenuated Ca responses, af- France) for the melanoma cell line used to amplify SIRPa cDNA. fecting both the mobilization of intracellular Ca stores and REFERENCES the influx of extracellular Ca and with an abolition of the 1. Matozaki, T., Uchida, Y., Fujioka, Y., and Kasuga, M. (1994) Biochem. Bio- activation of Erk1/2, as assessed by their phosphorylation. Our phys. Res. Commun. 204, 874 – 881 data neither support nor exclude the possibility that inhibition 2. Fujioka, Y., Matozaki, T., Noguchi, T., Iwamatsu, A., Yamao, T., Takahashi, 21 N., Tsuda, M., Takada, T., and Kasuga, M. (1996) Mol. Cell. Biol. 16, directly affected Ca responses or the Ras pathway. The ob- 6887– 6899 served reduced phosphorylation of FceRI ITAMs, however, in- 3. Kharitonenkov, A., Chen, Z., Sures, I., Wang, H., Schilling, J., and Ullrich, A. (1997) Nature 386, 181–186 dicates that inhibition affected signaling events that stand 4. Ohnishi, H., Kubota, M., Ohtake, A., Sato, K., and Sano, S.-I. (1996) J. Biol. upstream of these two responses. One likely explanation would Chem. 271, 25569 –25574 5. Jiang, P., Lagenaur, C. F., and Narayanan, V. (1999) J. Bol. Chem. 274, b and FcRg ITAMs or, possibly, the Src protein be that the FcR 559 –562 tyrosine kinase which phosphorylates FceRI ITAMs, were the 6. Stofega, M. R., Wang, H., Ullrich, A., and Carter-Su, C. (1998) J. Biol. Chem. substrates of protein tyrosine phosphatases recruited by phos- 273, 7112–7117 7. Vivier, E., and Dae ¨ ron, M. (1997) Immunol. Today 18, 286 –291 phorylated FcgRIIB-SIRPa. This mechanism has been pro- 8. Dae ¨ ron, M., Latour, S., Malbec, O., Espinosa, E., Pina, P., Pasmans, S., and posed to account for SHP-1-dependent inhibition of FcgRIIIA- Fridman, W. H. (1995) Immunity 3, 635– 646 9. Scharenberg, A. M., and Kinet, J.-P. (1996) Cell 87, 961–964 mediated antibody-dependent cell-mediated cytotoxicity by 10. Dae ¨ ron, M. (1997) The Immunologist 5, 79–86 KIRs (45). Whether SHP-2 can have the same effect as SHP-1 11. Moretta, A., Biassoni, R., Bottino, C., Pende, D., Vitale, M., Poggi, A., Mingari, M. C., and Moretta, L. (1997) Immunol. Rev. 155, 105–117 is unclear. SHP-2 was indeed assigned both positive (46) and 12. Burshtyn, D. N., Scharenberg, A. M., Wagtmann, N., Rajogopalan, S., Berrada, negative effects (13, 47). One may notice that, apparently, K., Yi, T., Kinet, J.-P., and Long, E. O. (1996) Immunity 4, 77– 85 FcgRIIB-SIRPa was not much more inhibitory than FcgRIIB- 13. Bruhns, P., Marchetti, P., Fridman, W. H., Vivier, E., and Dae ¨ ron, M. (1999) J. Immunol. 162, 3168 –3175 KIR, although comparable amounts of SHP-1 coprecipitated 14. Fong, D. C., Malbec, O., Arock, M., Cambier, J. C., Fridman, W. H., and with the two chimeras but much more SHP-2 with the SIRP Dae ¨ ron, M. (1996) Immunol. Lett. 54, 83–91 15. Malbec, O., Fridman, W. H., and Dae ¨ ron, M. (1999) J. Immunol. 162, chimera. This makes the role of SHP-2 in inhibition unclear. 4424 – 4429 One interesting possibility would be that SHP-1 and SHP-2 16. Adams, S., van der Laan, L. J. V., Vernon-Wilson, E., Renardel de Lavalette, C., Do ¨ pp, E. A., Dijkstra, C. D., Simmons, D. L., and van der Berg, T. K. have different substrates that may not fulfill the same func- (1998) J. Immunol. 161, 1853–1859 tions. It has been suggested that the inhibitory properties of 17. Veillette, A., Thibaudeau, E., and Latour, S. (1998) J. Biol. Chem. 273, receptors that recruit SHP-2 could result from the sequestra- 22719 –22728 18. Reth, M. G. (1989) Nature 338, 383–384 tion of that phosphatase, preventing it from exerting positive 19. Dae ¨ ron, M. (1997) Annu. Rev. Immunol. 15, 203–234 effects on activating receptors (16). If so, inhibition by SIRPa 20. Blank, U., Ra, C., Miller, L., White, K., Metzger, H., and Kinet, J. P. (1989) Nature 337, 187–189 might result from the combined dephosphorylation of ITAMs 21. Scholl, D. A., and Geha, R. S. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, and kinases by SHP-1 and the retention of SHP-2. Under these 8847– 8850 conditions, when recruited by SIRPa, SHP-1 and SHP-2 would 22. Pfefferkorn, L. C., and Yeaman, G. R. (1994) J. Immunol. 153, 3228 –3236 Inhibition of Cell Activation by Signal Regulatory Proteins 32499 23. Ra, C., Jouvin, M. H. E., Blank, U., and Kinet, J. P. (1989) Nature 341, 35. Paolini, R., Jouvin, M. H., and Kinet, J. P. (1991) Nature 353, 855– 858 752–754 36. Heldin, C.-H., and Ostman, A. (1996) Cytokines & Growth Factor Rev. 7, 3–10 24. Wagtmann, N., Biassoni, R., Cantoni, C., Verdiani, S., Malnati, M. S., Vitale, 37. Cambier, J. C. (1995) J. Immunol. 155, 3281–3285 M., Bottino, C., Moretta, L., Moretta, A., and Long, E. (1995) Immunity 2, 38. Rottapel, R., Reedijk, M., Williams, D. E., Lyman, S. D., Anderson, D. M., 439 – 449 Powson, T., and Bernstein, A. (1991) Mol. Cell. Biol. 11, 3043–3051 25. Takebe, Y., Seiki, M., Fujisawa, J. I., Hoy, P., Yokota, K., Arai, K. I., Yoshida, 39. Timms, J. F., Carlberg, K., Gu, H., Chen, H., Kamatkar, S., Nadler, M. J., M., and Arai, N. (1988) Mol. Cell. Biol. 8, 466 – 472 Rohrschneider, L. R., and Neel, B. G. (1998) Mol. Cell. Biol. 18, 3838 –3850 26. Kishi, K. (1985) Leuk. Res. 9, 381–390 40. Reinhold, M. I., Lindberg, F. P., Kersh, G. J., Allen, P. M., and Brown, E. J. 27. Butterfield, J. H., Weiler, D., Dewald, G., and Gleich, G. J. (1988) Leuk. Res. (1997) J. Exp. Med. 185, 1–11 12, 345–355 41. Parkos, C. A., Colgan, S. P., Liang, T. W., Nusrat, A., Bacarra, A. E., Carnes, 28. Barsumian, E. L., Isersky, C., Petrino, M. G., and Siraganian, R. P. (1981) Eur. D. K., and Madara, J. L. (1996) J. Cell Biol. 132, 437– 450 J. Immunol. 11, 317–323 42. Ghannadan, M., Baghestanian, M., Wimazal, F., Eisenmenger, M., D., L., 29. Saito, H., Ebisawa, M., Tachimoto, H., Shichijo, M., Fukagawa, K., Kargul, G., Walchshofer, S., Sillaber, C., Lechner, K., and Valent, P. (1998) Matsumoto, K., Iikura, Y., Awaji, T., Tsujimoto, G., Yanagida, M., J. Invest. Dermatol. 111, 689 – 695 Uzumaki, H., Takahashi, G., Tsuji, K., and Nakahata, T. (1996) 43. Furuzawa, T., Yanai, N., Hara, T., Miyajima, A., and Obinata, M. (1998) J. Immunol. 157, 343–350 J. Biochem. (Toyko) 123, 101–106 30. Dae ¨ ron, M., Bonnerot, C., Latour, S., and Fridman, W. H. (1992) J. Immunol. 44. Cherif-Zahar, B., Matassi, G., Raynal, V., Gane, P., Delaunay, J., 149, 1365–1373 Arrizabalaga, B., and Cartron, J. P. (1998) Blood 92, 2535–2540 31. Liu, T. T., Bohn, J. W., Ferry, E. L., Yamamoto, H., and Molinaro, C. A. (1980) 45. Binstadt, B. A., Brumbaugh, K. M., Dick, C. J., Scharenberg, A. M., Williams, J. Immunol. 124, 2728 –2737 B. L., Colonna, M., Lanier, L. L., Kinet, J.-P., Abraham, R. T., and Leibson, 32. Unkeless, J. C. (1979) J. Exp. Med. 150, 580 –596 P. J. (1996) Immunity 5, 629 – 638 33. Latour, S., Bonnerot, C., Fridman, W. H., and Dae ¨ ron, M. (1992) J. Immunol. 46. Gadina, M., Stancato, L. M., Bacon, C. M., Larner, A. C., and O’Shea, J. J. 149, 2155–2162 (1998) J. Immunol. 160, 4657– 4661 34. Malbec, O., Fong, D., Turner, M., Tybulewicz, V. L. J., Cambier, J. C., 47. Waterhouse, P., Marenge ` re, L. E. M., Mittru ¨ cker, H.-W., and Mak, T. W. Fridman, W. H., and Dae ¨ ron, M. (1998) J. Immunol. 160, 1647–1658 (1996) Immunol. Rev. 153, 183–207 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Signal Regulatory Proteins Negatively Regulate Immunoreceptor-dependent Cell Activation

Liénard, Hélène; Bruhns, Pierre; Malbec, Odile; Fridman, Wolf H.; Daëron, Marc
Journal of Biological Chemistry – Nov 1, 1999
Download PDF
7 pages

Loading next page...
 
/lp/unpaywall/signal-regulatory-proteins-negatively-regulate-immunoreceptor-okaFPiEwbB

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.274.45.32493
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 45, Issue of November 5, pp. 32493–32499, 1999 © 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Signal Regulatory Proteins Negatively Regulate Immunoreceptor-dependent Cell Activation* (Received for publication, May 18, 1999, and in revised form, September 3, 1999) He ´le ` ne Lie ´ nard‡, Pierre Bruhns§, Odile Malbec, Wolf H. Fridman, and Marc Dae ¨ ron¶ From the Laboratoire d’Immunologie Cellulaire et Clinique, INSERM U.255, Institut Curie, 75005 Paris, France Signal regulatory proteins of the a subtype (SIRPa) phoproteins that coprecipitated with SH2 domain-bearing pro- tein tyrosine phosphatases (SHPs) (1–3). SIRP molecules were are ubiquitous molecules of the immunoglobulin super- family that negatively regulate protein tyrosine kinase first identified as SHP substrate 1 (SHPS-1) (2) and brain receptor-dependent cell proliferation. Their intracyto- immunoglobulin-like molecules with tyrosine-based activation plasmic domain contains four motifs that resemble im- motifs (BIT) (4). When cloned, these molecules were found to be munoreceptor tyrosine-based inhibition motifs (ITIMs) the members of a multigene family of at least 15 transmem- and that, when tyrosyl-phosphorylated, recruit cyto- brane immunoglobulin superfamily molecules named collec- plasmic SH2 domain-bearing protein tyrosine phos- tively SIRPs, in which two types, a and b, were recognized, phatases (SHPs). ITIMs are borne by molecules that neg- differing by the presence (in SIRPa) or the absence (in SIRPb) atively regulate cell activation induced by receptors of an intracytoplasmic (IC) domain containing four tyrosine- bearing immunoreceptor tyrosine-based activation mo- based regulatory motifs (3). A neuronal adhesion molecule, a are coexpressed with tifs (ITAMs). Because SIRP previously described as P84, was found to belong to the SIRP ITAM-bearing receptors in hematopoietic cells, we in- family, and the widely expressed integrin-associated protein a could negatively regulate vestigated whether SIRP CD47 was recently identified as a ligand of P84 (5). a tran- ITAM-dependent cell activation. We found SIRP Interestingly, SIRPa were shown to regulate negatively cell scripts in human mast cells, and we show that a chi- proliferation induced by growth factors via protein tyrosine meric molecule having the transmembrane and intracy- kinase receptors (RTKs) and oncogene products. Little is a could inhibit IgE-induced toplasmic domains of SIRP known of the mechanism of inhibition by SIRPa, except that mediator secretion and cytokine synthesis by mast cells. negative regulation was correlated with the tyrosyl phospho- a chimera was coag- Inhibition required that the SIRP rylation of SIRPa and the recruitment of SHP-2 (3, 6). That gregated with ITAM-bearing high affinity IgE receptors eRI). It was correlated with the tyrosyl phosphoryla- SIRPa had inhibitory properties indicated that these molecules (Fc a chimera and the recruitment of SHP-1 bear inhibition motifs rather than activation motifs. SHP-2- tion of the SIRP eRI ITAMs was and SHP-2. The phosphorylation of Fc binding motifs found in SIRPa are indeed reminiscent of im- decreased; the mobilization of intracellular Ca and munoreceptor tyrosine-based inhibition motifs (ITIMs). ITIMs the influx of extracellular Ca were reduced, and the are present in a large group of molecules (7) that negatively activation of the mitogen-activated protein kinases regulate cell activation induced by receptors bearing immuno- Erk1 and Erk2 was abolished. SIRPa can therefore neg- receptor tyrosine-based activation motifs (ITAMs) (8). Negative atively regulate not only receptor tyrosine kinase-de- regulation by ITIM-bearing molecules is correlated with the pendent cell proliferation but also ITAM-dependent cell recruitment of SH2 domain-bearing phosphatases by phospho- activation. rylated ITIMs (9, 10). Thus, killer cell inhibitory receptors (KIRs) inhibit cell-mediated cytotoxicity when they bind to major histocompatibility complex class I molecules on target Signal regulatory proteins (SIRPs) were described as phos- cells (11). Their IC domain contains two ITIMs that, when tyrosyl-phosphorylated, recruit SHP-1 and SHP-2 (12, 13). Likewise, FcgRIIB, a family of low affinity receptors for IgG, * This work was supported by the INSERM, the Association pour la Recherche sur le Cancer, and the Institut Curie. The costs of publica- negatively regulate cell activation via ITAM-bearing receptors tion of this article were defrayed in part by the payment of page when coaggregated with the latter by immune complexes (8). charges. This article must therefore be hereby marked “advertisement” FcgRIIB bear a single ITIM that, when tyrosyl-phosphorylated, in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. recruits selectively the SH2 domain-bearing inositol-5-phos- ‡ Recipient of a Rho ˆ ne-Poulenc Rorer CIFRE contract. § Recipient of a fellowship from the Ministe ` re de l’Enseignement phatase SHIP (14). We found recently that FcgRIIB are also Supe ´ rieur et de la Recherche. capable of inhibiting RTK-dependent cell proliferation (15). To whom correspondence should be addressed: Laboratoire This finding opened the possibility that it might be a general d’Immunologie Cellulaire et Clinique, INSERM U.255, Institut Curie, property of ITIM-bearing molecules to regulate negatively not 26 rue d’Ulm, 75005 Paris, France. Tel.: 33-1-4432-4223; Fax: 33-1- 4051-0420; E-mail: [email protected]. only ITAM-dependent cell activation but also RTK-mediated The abbreviations used are: BIT, brain immunoglobulin-like mole- cell proliferation. If so, one could hypothesize that SIRPa might eRI, high affinity IgE cule with tyrosine-based activation motifs; Fc inhibit ITAM-dependent cell activation. receptors; FcgRIIB and FcgRIIIA, low affinity IgG Receptors; GAM, By contrast with KIRs whose expression is restricted to NK goat anti-mouse Ig; GAR, goat anti-rabbit Ig; HRP, horseradish perox- cells and T cells (11), SIRPa were found to be expressed by all idase; IC, intracytoplasmic; ITAM, immunoreceptor tyrosine-based ac- tivation motif; ITIM, immunoreceptor tyrosine-based inhibition motif; human tissues examined, including hematopoietic cells (3), by KIRs, killer cell inhibitory receptors; MAR, mouse anti-rat Ig; RTK, receptor tyrosine kinase; SH2, Src homology domain 2; SHIP, SH2 domain-bearing inositol-phosphate phosphatase; SHP, SH2 domain- BSA, bovine serum albumin; mAb, monoclonal antibody; FITC, fluores- bearing protein tyrosine phosphatase; SHPS-1, SH2 domain-bearing cein isothiocyanate; PAGE, polyacrylamide gel electrophoresis; RT- phosphatase substrate 1; SIRPs, signal regulatory proteins; TNP, tri- PCR, reverse transcriptase-polymerase chain reaction; TNF, tumor ne- nitrophenyl; MAP, mitogen-activated protein; DNP, 2,4-dinitrophenol; crosis factor; TM, transmembrane. This paper is available on line at http://www.jbc.org 32493 This is an Open Access article under the CC BY license. 32494 Inhibition of Cell Activation by Signal Regulatory Proteins (HRP)-conjugated mouse mAb anti-phosphotyrosine (PY20) were pur- neural and myeloid cells in rats (16), and by myeloid cells, chased from Chemicon (Temecula, CA); mouse mAbs anti-SHP-1 and especially macrophages, but not lymphoid cells in mice (17). anti-SHP-2 were from Transduction Laboratories (Lexington, KY); Hematopoietic cells also express ITAM-bearing receptors. mouse mAbs anti-Erk1/2 and anti-phospho-Erk1/2 were from New Eng- Thus, myeloid cells express ITAM-bearing receptors for the Fc land Biolabs (Beverly, MA); HRP-conjugated polyclonal goat anti- portion of immunoglobulins (FcR) (18, 19). These comprise high mouse Ig (GAM) antibodies and polyclonal goat anti-rabbit Ig (GAR) affinity IgE receptors (FceRI) (20), high affinity IgG receptors antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Immunofluorescence—RBL transfectants were incubated for1hat (FcgRI) (21), high affinity IgA receptors (FcaRI) (22), and low 0 °C with 10 mg/ml 2.4G2 mAb in balanced salt solution containing 5% affinity IgG receptors (FcgRIIIA) (23), which all share the fetal calf serum. Cells were washed and stained by being incubated for ITAM-bearing signal transduction subunit FcRg (18). They 30 min at 0 °C with 50 mg/ml FITC-MAR F(ab9)2. Fluorescence was also include the human-restricted, single chain, low affinity analyzed by flow cytometry using a FACScalibur (Becton-Dickinson, IgG receptors FcgRIIA/C that bear one ITAM in their own IC Mountain View, CA). domain (18). If they could interfere with ITAM-dependent cell RT-PCR Analysis of SIRPa Transcripts—RNA was extracted from 1 3 10 KU812, HMC-1, and cord blood-derived human mast cells. activation, SIRPa would negatively regulate not only the cDNAs were prepared and used as templates to amplify sequences growth but also the many biological functions of hematopoietic corresponding to the TM and IC domains of SIRPa using the two cells triggered by Fc receptors. oligonucleotides SIRPa sense and antisense. Amplified fragments were To investigate the hypothesis that SIRPa might negatively sequenced using the same two oligonucleotides. regulate cell activation induced by ITAM-bearing receptors, we Serotonin Release—RBL transfectants, loaded with [ H]serotonin constructed an experimental model in mast cells. Mast cells (Amersham Pharmacia Biotech), were incubated for1hat37 °C with IgE anti-DNP and with 0 or 3 mg/ml 2.4G2 F(ab9)2, washed, and chal- were chosen as an assay system because we found SIRPa lenged for 30 min at 37 °C with TNP-MAR F(ab9)2 or with MAR F(ab92) transcripts in human mast cells. We report here that, when and DNP-BSA. The percentage of [ H]serotonin released was measured coaggregated with FceRI, a chimeric molecule whose IC domain as described (30). was that of human SIRPa could inhibit IgE-induced mediator TNF Secretion—RBL transfectants, incubated for1hat37 °C with release and cytokine secretion by mast cells. The experimental IgE anti-DNP and 0 or 3 mg/ml 2.4G2 F(ab9)2, were challenged for 3 h model used made possible the biochemical analysis of intracel- at 37 °C with 10 mg/ml TNP-MAR F(ab9)2. Cell-free supernatants were harvested and titrated for TNFa on L929 cells as described (33). lular events associated with inhibition. Inhibition was corre- Measurement of Intracellular Ca Concentration—RBL transfec- lated with the tyrosyl phosphorylation of the SIRPa chimera, tants, previously sensitized with IgE anti-DNP and incubated with or with the recruitment of SHP-1 and SHP-2 by the phosphoryl- without 2.4G2 F(ab9)2, were loaded with 5 mM Fluo-3-AM in the pres- ated chimera, with a decreased phosphorylation of FceRI ence of 0.2% Pluronic F-127 (Molecular Probes, Eugene, OR) for 30 min ITAMs, with attenuated Ca responses, and with an abolition at room temperature. Cells were resuspended in medium in which of MAP kinase activation. calcium had been buffered to 60 nM (equivalent to the intracellular Ca concentration in resting cells) with EGTA. One min later, they were EXPERIMENTAL PROCEDURES challenged with TNP-MAR F(ab9)2 for 100 s, after which the extracel- cDNA Constructs—-cDNA sequences encoding the transmembrane lular Ca concentration was raised to 1.3 mM with CaCl . Intracellular (TM) and IC domains of SIRPa and of a human KIR were amplified free Ca concentration was monitored by flow cytometry with a FAC- from the human melanoma cell line HT-144 (from the ATCC) and from Scalibur using the software FCS assistant 1.2.9b (Becton Dickinson). the p58.183 KIR cDNA (24) by RT-PCR and PCR, respectively, with the Immunoprecipitation and Western Blot Analysis—RBL transfectants following oligonucleotide primers: for SIRPa, sense, 59-TCTAAGGTAC- were incubated with IgE anti-DNP and 0 or 3 mg/ml 2.4G2 F(ab9)2, CAAACATCTATATTGTGGTG-39, and antisense, 59-AGCAAACCGAG- washed, and challenged for various periods of time at 37 °C with 10 CTCCCATTCACTTCCTCGGGACCTG-39; for KIR, sense, 59-CCCA- mg/ml TNP-MAR F(ab9)2. Aliquots of 3 3 10 cells were lysed for 10 min GAC AGGTACCTGTTCTGATTGGGACC-39, and antisense, 59-CTGA- at 0 °C. For immunoprecipitation of FcgRIIB, lysis buffer contained 10 CTGTGGAGCTCATGGGCAGG-39. KpnI (GGTACC) and SacI (GAGC- mM Tris, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 1 mM Na VO ,5mM 3 4 TC) sites are in boldface. PCR products were inserted into an expression NaF, 5 mM sodium pyrophosphate, 0.4 mM EDTA, 10 mg/ml aprotinin, cassette under the control of the SRa promoter in pBR322 (25), con- 10 mg/ml leupeptin, 10 mg/ml pepstatin, and 1 mM phenylmethylsulfo- taining a resistance gene to neomycin (NT-neo) and containing the nyl fluoride. For immunoprecipitation of FceRI, lysis buffer contained extracellular domain of FcgRIIB. 50 mM Tris, pH 8.0, 1% Nonidet P-40, 1 mM Na VO ,20mM EDTA, 10 3 4 Cells—The human basophil-like KU812 (26), the human mast cells mg/ml aprotinin, 10 mg/ml leupeptin, and 1 mM phenylmethylsulfonyl HMC-1 (27), and the rat mast cells RBL-2H3 (28) were cultured in fluoride. Lysates were centrifuged at 12,000 rpm for 10 min at 4 °C, and Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf supernatants were incubated with immunoadsorbents for1hat4 °C. serum, 100 IU/ml penicillin, 100 mg/ml streptomycin, and 2 mML- Protein G-Sepharose or 2.4G2 coupled to protein G-Sepharose was used glutamine. Culture reagents were from Life Technologies, Inc. Human to precipitate FcgRIIB from cells preincubated with 2.4G2 or not, re- normal mast cells, obtained by culture of cord blood mononuclear cells spectively. Anti-mouse IgE antibodies coupled to protein A-Sepharose as described (29), were kindly donated by Dr. M. Arock (Faculte´de (Sigma) were used to precipitate FceRI from cells sensitized with mouse Pharmacie, Paris, France). More than 99% cells were mast cells as IgE. Immunoprecipitates were fractionated by SDS-PAGE and trans- judged by morphology, the presence of metachromatic granules, and ferred onto Immobilon-P membranes (Millipore, Bedford, MA). Mem- positivity for tryptase. cDNAs encoding FcgRIIB-SIRPa and FcgRIIB- branes were saturated either with 5% BSA (Sigma) or with 5% skimmed KIR were stably transfected in RBL-2H3 cells; transfectants were se- milk (Re ´ gilait, Saint-Martin-Belle-Roche, France) and Western-blotted lected with 2.5 mg/ml G418 (Life Technologies, Inc.) and cloned as with HRP-conjugated anti-phosphotyrosine antibodies or with rabbit described (30). anti-FcgRIIB, mouse anti-SHP-1, mouse anti-SHP-2, mouse anti-FcRb, Antibodies—The mouse IgE anti-DNP mAb 2682-I (31) was used as or rabbit anti-FcRg antibodies, followed by HRP-conjugated GAR or culture supernatant. IgG and F(ab9)2 fragments of the rat anti-mouse GAM. Anti-SHP-2 antibodies detected two species of SHP-2. Anti- FcgRIIB 2.4G2 mAb (32) were obtained as described (30). Rabbit anti- SHP-1 and anti-SHP-2 antibodies recognized specifically SHP-1 and mouse FcgRIIB were raised by Dr. C. Saute ` s (Institut Curie, Paris, SHP-2, respectively. HRP-conjugated antibodies were detected using an France) against recombinant extracellular domains of FcgRIIB. Mouse enhanced chemiluminescence (ECL) kit (Amersham Pharmacia Bio- mAb anti-FcRb were purified with protein G-Sepharose (Amersham tech). When blotted sequentially with different antibodies, filters were Pharmacia Biotech) from culture supernatants of the JRK hybridoma stripped following revelation by being incubated for 30 min at 50 °C in cells. Rabbit anti-FcRg antibodies and rabbit anti-mouse IgE antibodies buffer containing 62.5 mM Tris, pH 6.7, 100 mM b-mercaptoethanol, and were generous gifts of Dr. J.-P. Kinet (Beth Israel Hospital, Boston). 2% SDS. IgG and F(ab9)2 fragments of mouse anti-rat Ig (MAR) and FITC- Western Blot Analysis of MAP Kinases—RBL transfectants were conjugated MAR F(ab9)2 fragments (FITC-MAR F(ab9)2) were from incubated for1hat37 °C with IgE anti-DNP and with 0 or 3 mg/ml Jackson ImmunoResearch Laboratories (West Grove, PA). MAR F(ab9)2 2.4G2 F(ab9)2, washed, and challenged with 10 mg/ml TNP-MAR were trinitrophenylated with trinitrobenzene sulfonic acid (Eastman- F(ab9)2. Equal volumes of cell lysates, obtained as described for immu- Kodak Co.) and purified on Sephadex G-25 (Amersham Pharmacia noprecipitation of FcgRIIB, were fractionated by SDS-PAGE, trans- Biotech). TNP -MAR F(ab9)2 were obtained. Horseradish peroxidase ferred onto Immobilon-P membranes, and blotted with anti-phospho- 10 Inhibition of Cell Activation by Signal Regulatory Proteins 32495 FIG.2. Inhibition of IgE-induced mast cell activation by FcgRIIB-SIRPa. A, inhibition of serotonin release. Cells expressing FIG.1. SIRPa transcripts in human mast cells and structure FcgRIIB-SIRPa,FcgRIIB-KIR, or FcgRIIB(IC1) were sensitized with and expression of recombinant molecules expressed in RBL indicated dilutions of IgE anti-DNP and with 0 (open circles)or3 mg/ml cells. A, SIRPa transcripts in human mast cells. cDNAs from a human (closed circles) 2.4G2 F(ab9)2. Cells were challenged with 10 mg/ml basophil-like cell line (KU812), a human mast cell line (HMC-1), and TNP-MAR F(ab9)2. The figure shows the percentage of serotonin re- cord blood-derived human mast cells (CB-HMC) were analyzed by RT- leased as a function of IgE dilutions. B, requirement for the coaggrega- PCR using oligonucleotides that could amplify sequences encompassing tion of FceRI with FcgRIIB-SIRPa for inhibition of serotonin release. the TM and IC domains of SIRPa. PCR products were analyzed by Cells expressing FcgRIIB-SIRPa were sensitized with indicated dilu- agarose gel electrophoresis. bp, base pair. B, structure of recombinant tions of IgE anti-DNP and with 0 (open symbols)or3 mg/ml (closed molecules. The figure schematizes the structure of the IC domain- symbols) 2.4G2 F(ab9)2. Cells were challenged either with 10 mg/ml deleted FcgRIIB(IC1), of the FcgRIIB-SIRPa chimera, and of the TNP-MAR F(ab9)2 (circles) or with 3 mg/ml DNP-BSA and 10 mg/ml FcgRIIB-KIR chimera expressed in RBL cells. ITIM-like sequences are MAR F(ab9)2 (squares). The figure shows the percentage of serotonin indicated at their respective positions. C, expression of recombinant released as a function of IgE dilutions. C, inhibition of TNFa produc- molecules in RBL-2H3 cells. Clones of stable transfectants were exam- tion. Cells expressing FcgRIIB-SIRPa were sensitized with IgE anti- ined by immunofluorescence for the expression of FcgRIIB-based mol- DNP (supernatant 1/10) and incubated with 0 (open circles)or3 mg/ml ecules. Filled histograms, cells incubated with FITC-MAR F(ab9)2 only; (closed circles) 2.4G2 F(ab9)2. They were challenged with 10 mg/ml open histograms, cells incubated with 2.4G2 and FITC-MAR F(ab9)2. TNP-MAR F(ab9)2for3hat37 °C. Cell-free supernatants were har- vested, and serial 2-fold dilutions were tested for cytotoxicity on L929 cells. The figure represents the percentage of cytotoxicity as a function Erk1/2 or with anti-Erk1/2 antibodies, followed by HRP-conjugated of the dilution of supernatants. Experiments shown in A–C were re- GAR. HRP-conjugated antibodies were detected using the ECL kit. peated three times with the same results. RESULTS Human Mast Cells Express SIRPa—We investigated first gregated with FceRI in RBL transfectants sensitized with the presence of SIRPa transcripts in RNA from human mast mouse IgE anti-DNP, incubated with F(ab9)2 fragments of the cells by RT-PCR using oligonucleotide primers that could am- rat anti-mouse FcgRIIB mAb 2.4G2, and challenged with TNP- plify a 435-base pair long fragment corresponding to the TM MAR F(ab9)2 as described previously (34). Serotonin release and IC domains of human SIRPa. A single PCR product having induced under these conditions was compared with serotonin the expected size was amplified from cDNAs from the human release induced by aggregating FceRI with TNP-MAR F(ab9)2 mast cell line HMC-1 and from cord blood-derived human mast in the same three transfectants sensitized with mouse IgE cells but not from the basophil-like cell line KU812 (Fig. 1A). anti-DNP but not incubated with 2.4G2 F(ab9)2. When coag- PCR products amplified from HMC-1 and from cord blood- gregated with FceRI, FcgRIIB-SIRPa inhibited serotonin re- derived mast cells were sequenced. They had the same nucle- lease as efficiently as FcgRIIB-KIR, but not FcgRIIB(IC1) otide sequence as cDNA encoding the TM and IC domains of (Fig. 2A). human SIRPa (3). To determine the conditions required for FcgRIIB-SIRPa to A SIRPa Chimera Inhibits IgE-induced Mast Cell Activation inhibit serotonin release, transfectants expressing FcgRIIB- When Coligated with FceRI—In order to examine whether SIRPa were sensitized with mouse IgE anti-DNP and incu- SIRPa could inhibit IgE-induced mast cell activation, we con- bated with 2.4G2 F(ab9)2. FceRI and FcgRIIB-SIRPa were then structed a cDNA encoding a chimeric molecule having the either coaggregated by TNP-MAR F(ab9)2 as above or aggre- extracellular domain of murine FcgRIIB and the TM and IC gated simultaneously, but independently, by DNP-BSA and domains of human SIRPa (FcgRIIB-SIRPa). As a positive con- MAR F(ab9)2, respectively. Cells sensitized with IgE but not trol, a cDNA encoding a chimeric molecule having the extra- incubated with 2.4G2 F(ab9)2 served as positive controls. Sero- cellular domain of murine FcgRIIB and the TM and IC domains tonin release was inhibited when FceRI and FcgRIIB-SIRPa of a human KIR (FcgRIIB-KIR) was also constructed (Fig. 1B). were coaggregated by TNP-MAR F(ab9)2 but not when they Chimeric cDNAs were stably transfected in the rat mast cells were simultaneously aggregated by DNP-BSA and MAR RBL-2H3, which constitutively express FceRI. Clones of trans- F(ab9)2 (Fig. 2B). fectants expressing comparable amounts of FcgRIIB-SIRPa or To determine whether FcgRIIB-SIRPa could inhibit an IgE- FcgRIIB-KIR (Fig. 1C) and, as negative controls, clones ex- induced secretion of cytokine, FceRI were either aggregated or pressing tail-less FcgRIIB (FcgRIIB(IC1)) (30, 33) were used coaggregated with FcgRIIB-SIRPa in transfectants expressing for experiments. Several clones expressing each molecule gave FcgRIIB-SIRPa, using the same ligands as above, and the comparable results. amount of TNFa released in culture supernatants during the FcgRIIB-SIRPa,FcgRIIB-KIR, or FcgRIIB(IC1) were coag- following 3 h was titrated using a bioassay. When coaggregated 32496 Inhibition of Cell Activation by Signal Regulatory Proteins coaggregated with FceRI for various periods, its phosphoryla- tion was apparent at 1 min, maximum between 3 and 10 min, and still visible at 30 min (Fig. 3B). When coaggregated with FceRI, tail-less FcgRIIB failed to be phosphorylated (data not shown). Four tyrosine residues are present in the IC domain of SIRPa, which are each constitutive of an ITIM-like motif. To identify SH2 domain-bearing phosphatases possibly re- cruited in vivo by FcgRIIB-SIRPa, the SIRPa chimera was immunoprecipitated after it had or had not been coaggregated with FceRI, and phosphatases coprecipitated with FcgRIIB- SIRPa were identified by Western blotting. Some SHP-2 copre- cipitated with weakly phosphorylated FcgRIIB-SIRPa in rest- ing cells, and high amounts of SHP-2 coprecipitated with FcgRIIB-SIRPa that became heavily phosphorylated upon co- aggregation with FceRI. SHP-1 was coprecipitated with FcgRIIB-SIRPa following coaggregation of the chimera with FceRI only. Under the same conditions, FcgRIIB-KIR, which was also phosphorylated upon coaggregation with FceRI, re- cruited similar amounts of SHP-1 as FcgRIIB-SIRPa but much less SHP-2. SHIP was recruited neither by FcgRIIB-SIRPa nor by FcgRIIB-KIR (Fig. 3C). The SIRPa Chimera Blocks Signal Transduction by FceRI—An early step in IgE-induced signaling in mast cells is the tyrosyl phosphorylation of FceRI ITAMs (35) which enables the recruitment of SH2 domain-bearing cytoplasmic kinases, FIG.3. Phosphorylation of FcgRIIB-SIRPa upon coaggrega- leading to an increased intracellular Ca tion with FceRI and coprecipitation of SHP-1 and SHP-2 with concentration, and phosphorylated FcgRIIB-SIRPa. RBL transfectants were preincu- adapter molecules that connect phosphorylated receptors with bated with or without 3 mg/ml 2.4G2 F(ab9)2, sensitized or not with IgE the Ras pathway, leading to the transcription of cytokine anti-DNP (supernatant 1/10), and challenged with or without 10 mg/ml genes. In RBL cells, FceRI are associated with two ITAM- TNP-MAR F(ab9)2. A, phosphorylation of FcgRIIB-SIRPa upon coag- bearing subunits, FcRb and FcRg (20). gregation with FceRI. Aliquots of 1 3 10 cells were lysed 3 min after stimulation. FcgRIIB-SIRPa immunoprecipitates were fractionated by The phosphorylation of FceRI ITAMs was examined by West- SDS-PAGE (10% acrylamide), transferred onto Immobilon, and West- ern blot analysis with anti-phosphotyrosine antibodies in RBL ern-blotted with anti-phosphotyrosine antibodies. The filter was transfectants sensitized with IgE anti-DNP, after FceRI were stripped and reblotted with anti-FcgRIIB antibodies. Comparable re- aggregated or coaggregated with the SIRPa chimera for 1, 3, or sults were obtained in three separate experiments. B, kinetics of FcgRIIB-SIRPa phosphorylation. Aliquots of 3 3 10 cells were lysed at 10 min at 37 °C and immunoprecipitated with anti-IgE anti- indicated times. FcgRIIB-SIRPa immunoprecipitates were fractionated bodies. FcRb and FcRg were identified by Western blot analysis by SDS-PAGE (10% acrylamide), transferred onto Immobilon, and of the same filter with corresponding specific antibodies. The Western-blotted with anti-phosphotyrosine antibodies. The filter was phosphorylation of both FcRb and FcRg induced upon FceRI stripped and reblotted with anti-FcgRIIB antibodies. Comparable re- sults were obtained in two separate experiments. C, coprecipitation of aggregation was decreased upon coaggregation of FceRI with SHPs with phosphorylated FcgRIIB-SIRPa. Aliquots of 3 3 10 cells the SIRPa chimera (Fig. 4A). were lysed 3 min after stimulation, and FcgRIIB-SIRPa was immuno- 21 Ca responses were monitored in RBL transfectants follow- precipitated. Left, aliquots of immunoprecipitates corresponding to 3 3 6 eRI aggregation or following the coaggregation of FceRI ing Fc 10 RBL cells were fractionated by SDS-PAGE (10% acrylamide), trans- ferred onto Immobilon, and Western-blotted with anti-phosphotyrosine with the SIRPa chimera and, as a positive control, with the antibodies, stripped and reblotted with anti-FcgRIIB antibodies. Right, KIR chimera, under conditions that permitted us to measure remaining immunoprecipitates were fractionated by SDS-PAGE (8% 21 the mobilization of intracellular Ca and the influx of extra- acrylamide), transferred onto Immobilon, and Western-blotted with cellular Ca separately. FcgRIIB-SIRPa and FcgRIIB-KIR anti-SHIP (upper panel) or with anti-SHP-2 and then anti-SHP-1 with- similarly reduced both the intracellular Ca mobilization and out stripping (lower panel). The same blot was stripped and reblotted with anti-FcgRIIB antibodies. Whole cell lysates (WCL) were used as the extracellular Ca influx (Fig. 4B). positive controls for Western blotting (only lysates from transfectants The activation of the Ras pathway was assessed by examin- expressing FcgRIIB-SIRPa are shown). Comparable results were ob- ing the phosphorylation of the MAP kinases Erk1 and Erk2. tained in two separate experiments. When coaggregated with FceRI, both FcgRIIB-SIRPa and FcgRIIB-KIR abolished IgE-induced Erk1/2 phosphorylation with FceRI, FcgRIIB-SIRPa inhibited about 90% of the TNFa induced by aggregating FceRI (Fig. 4C). secretion (Fig. 2C). DISCUSSION The SIRPa Chimera Becomes Tyrosyl-phosphorylated upon Coligation with FceRI and Recruits Protein Tyrosine Phos- SIRPa were described as negative regulators of growth phatases—We examined next the phosphorylation of FcgRIIB- factor-induced, RTK-mediated cell proliferation (3). RTKs SIRPa when coaggregated with FceRI or not. FcgRIIB-SIRPa transphosphorylate each other when dimerized by growth fac- was immunoprecipitated from transfectants expressing tors (36). They induce cells to enter the cell cycle and to divide. FcgRIIB-SIRPa that had been incubated with medium alone, ITAM-bearing receptors have no intrinsic protein kinase activ- IgE anti-DNP and/or 2.4G2 F(ab9)2, and challenged with or ity, and they do not induce cells to proliferate. When aggre- without TNP-MAR F(ab9)2 for 3 min at 37 °C. Western blot gated by multivalent ligands, they are phosphorylated by Src analysis with anti-phosphotyrosine antibodies showed that the family protein tyrosine kinases, and phosphorylated ITAMs basal phosphorylation of the chimera slightly increased when serve as docking sites for cytoplasmic SH2 domain-bearing FcgRIIB-SIRPa was aggregated but not when FceRI were ag- protein tyrosine kinases and adapter molecules that lead to cell gregated. It increased dramatically when FcgRIIB-SIRPa was activation (37). Although both early and late signals delivered coaggregated with FceRI (Fig. 3A). When FcgRIIB-SIRPa was by RTKs and ITAM-bearing receptors are different, these two Inhibition of Cell Activation by Signal Regulatory Proteins 32497 FIG.4. FcgRIIB-SIRPa inhibits intracellular signaling by FceRI. A, inhibition of the phosphorylation of FceRI ITAMs. Aliquots of 5 3 10 RBL transfectants expressing FcgRIIB-SIRPa were sensitized with IgE anti-DNP (supernatant 1/10), preincubated with or without 3 mg/ml 2.4G2 F(ab9)2, and challenged with or without 10 mg/ml TNP-MAR F(ab9)2 for 1, 3, or 10 min at 37 °C. FceRI immunoprecipitates were fractionated by SDS-PAGE (12.5% acrylamide), transferred onto Immobilon, and Western-blotted with anti-phosphotyrosine antibodies. The blot was then cut into two parts for blotting with anti-FcRb or anti-FcRg antibodies, respectively. Comparable results were obtained in three separate experiments. B, inhibition of Ca mobilization. RBL transfectants expressing FcgRIIB-SIRPa or FcgRIIB-KIR were sensitized with IgE anti-DNP (supernatant 1/10) and preincubated with (gray) or without (black) 3 mg/ml 2.4G2 F(ab9)2. They were loaded with Fluo-3-AM, resuspended in medium containing 21 21 60 nM Ca , and stimulated with 10 mg/ml TNP-MAR F(ab9)2 (arrows). One hundred seconds later, the extracellular Ca concentration was raised to 1.3 mM with CaCl . The figure represents the mean fluorescence of cells as a function of time. Comparable results were obtained in three separate experiments. C, inhibition of Erk1/2 activation. RBL transfectants expressing FcgRIIB-SIRPa or FcgRIIB-KIR were preincubated with or without 3 mg/ml 2.4G2 F(ab9)2, sensitized or not with IgE anti-DNP (supernatant 1/10), challenged with or without TNP-MAR F(ab9)2 for 10 min, and lysed. Whole cell lysates were fractionated by SDS-PAGE (10% acrylamide), transferred onto Immobilon, and Western-blotted with anti-phospho-Erk1/2. The blot was then reblotted with anti-Erk1/2 antibodies without stripping. Comparable results were obtained in two separate experiments. types of receptors utilize common intracellular effectors such as FceRI and FcgRIIB-SIRPa were coaggregated by the same phospholipase C-g, phosphatidylinositol 3-kinase, and mole- extracellular ligand. A possible explanation for the different cules of the Ras/MAP kinase pathway (37, 38). We found SIRPa requirements for SIRPa to inhibit RTK-dependent cell prolif- transcripts in human mast cells, and we investigated the ef- eration and for FcgRIIB-SIRPa to inhibit FceRI-dependent cell fects of a SIRPa chimera on the secretion of inflammatory activation might be that SIRPa constitutively associate with mediators triggered by FceRI, a typical ITAM-bearing Fc re- RTKs (and possibly with FceRI) via their extracellular domains ceptor, in the rat mast cell line RBL-2H3. We provide here the which were removed in the FcgRIIB-SIRPa chimera. Specific first evidence that SIRPa can behave as ITIM-bearing mole- conditions under which FceRI and SIRPa could be coligated, if cules that negatively regulate ITAM-dependent cell activation, not constitutively associated, on mast cells remain to be deter- and we document the mechanism of inhibition by SIRPa. mined, and it was not an aim of this investigation to address The experimental model used to demonstrate the inhibitory that issue. Our work indicates, however, that a previously properties of SIRPa was validated by our finding that human undemonstrated negative regulation may operate in hemato- mast cells contain SIRPa transcripts. This conclusion could be poietic cells that coexpress SIRPa and ITAM-bearing receptors, drawn from results obtained with both cord blood-derived hu- provided the two receptors are maintained close to each other, man mast cells and with the human mastocytoma cells HMC-1. either constitutively or inducibly by common extracellular li- Cord blood-derived mast cells present the advantage of being gands. One SIRP member, the neuronal adhesion molecule normal, non-transformed cells but have the disadvantage of P84, was recently assigned an extracellular ligand. The inte- being potentially contaminated by (less than 1%) other cells, grin-associated protein CD47 was indeed found to bind to P84, possibly including macrophages that express high levels of and both molecules are colocalized in synapse-rich structures of SIRPs. HMC-1 cells maintained in culture for years have the the cerebellum and the retina where their interactions were disadvantage of being transformed cells, but they have the proposed to control synaptic functions (5). Whether CD47 is advantage of excluding any cell contamination. The combined also a ligand for other SIRP family members and/or whether results obtained in the two cell types make it reasonable to other ligands bind to the polymorphic extracellular domains of conclude that human mast cells do express the SIRPa gene, SIRPa is unknown. CD47 is widely expressed on T cells (40), on and although the expression of SIRPa proteins was not for- myeloid cells including neutrophils (41) and mast cells (42), on mally demonstrated here, they justify that mast cell secretory epithelial cells (41), on spleen, liver and bone marrow stromal responses were chosen as readouts to assess the ability of cells (43), on neurones (5, 16), and on red blood cells (44). SIRPa to control cell activation. The mechanism underlying the inhibitory properties of The overexpression of SIRPa was reported to be sufficient to SIRPs is poorly understood. One reason is the constitutive inhibit epidermal growth factor-, platelet-derived growth fac- association of RTKs with SIRPs that made it difficult to ana- tor-, or insulin-induced cell proliferation in NIH-3T3 fibro- lyze the respective roles of interacting molecules. The experi- blasts and ligand-independent proliferation of the same cells mental model used here enabled us to dissect various stages of infected by a retrovirus carrying an oncogenic form of RTK (3). the process and to get some insight in the mechanism of inhi- SIRPa are therefore likely to be constitutively associated with bition by SIRPa. One consequence of the coaggregation of RTKs, and indeed, BIT coprecipitated with CSF-1R in a macro- FceRI with FcgRIIB-SIRPa was a dramatic tyrosyl phosphoryl- phage cell line (39). By contrast, the overexpression of the ation of the chimera. A faint basal phosphorylation of FcgRIIB- FcgRIIB-SIRPa chimera in RBL-2H3 cells affected neither the SIRPa was observed in resting cells. This phosphorylation growth of transfectants (data not shown) nor IgE-induced re- slightly increased upon FcgRIIB-SIRPa aggregation, suggest- sponses. Mediator release was inhibited when and only when ing that low levels of protein tyrosine kinase may associate 32498 Inhibition of Cell Activation by Signal Regulatory Proteins with the chimera in mast cells. It did not increase upon stim- act in concert, and their complementary effects might explain the deep inhibition of the biological responses observed. What- ulation by IgE, confirming that FcgRIIB-SIRPa was not asso- ciated with FceRI. The increased phosphorylation of FcgRIIB- ever the respective roles of the two phosphatases, our work documents a reciprocal cross-talk between SIRPa and ITAM- SIRPa, when coaggregated with FceRI, is likely to depend on bearing receptors during inhibition of cell activation. The co- Src protein tyrosine kinases that are recruited by aggregated aggregation of FcgRIIB-SIRPa with FceRI may indeed enable FceRI, as it was previously demonstrated for the tyrosyl phos- both FceRI to provide protein tyrosine kinases that could phos- phorylation of FcgRIIB (34). phorylate SIRPa ITIMs and FcgRIIB-SIRPa to provide protein Phosphorylated FcgRIIB-SIRPa was found to recruit SH2 tyrosine phosphatases that could dephosphorylate FceRI domain-bearing cytoplasmic protein tyrosine phosphatases. ITAMs and sequester phosphatases possibly required for cell The respective roles of the four ITIM-like motifs in phospha- activation. tase recruitment is not known as no mutational analysis of the Myeloid cells, such as macrophages and mast cells, coexpress four tyrosine residues contained in the SIRPa IC domain has not only ITAM-bearing FcR and SIRPa but also RTKs and been made. Some SHP-2 coprecipitated with lightly phospho- FcgRIIB. We found recently that FcgRIIB, which were known rylated FcgRIIB-SIRPa in resting cells and much greater to regulate negatively ITAM-dependent cell activation (8), can amounts with heavily phosphorylated FcgRIIB-SIRPa, follow- also negatively regulate RTK-dependent cell proliferation and ing its coaggregation with FceRI. Significant amounts of SHP-1 that inhibition is correlated with the recruitment of the SH2 also coprecipitated with the phosphorylated chimera. Associa- domain-bearing inositol-5-phosphatase SHIP (15). We show tion with SHP-2 was instrumental for the identification of here that SIRPa, which were known to regulate negatively SIRPs (1–3). SHP-1 associated with SIRPa when overex- RTK-dependent cell proliferation, can also negatively regulate pressed in fibroblasts (2) and with SHPS-1 (17) and BIT (39) in ITAM-dependent cell activation and that inhibition is corre- macrophages. Noticeably, comparable amounts of SHP-1 copre- lated with the recruitment of SHP-2 and SHP-1. It follows 1) cipitated with FcgRIIB-SIRPa and with FcgRIIB-KIR, whereas that SIRPa bear typical ITIMs which endow these molecules much more SHP-2 coprecipitated with FcgRIIB-SIRPa than with a wide array of previously unsuspected regulatory prop- with FcgRIIB-KIR. This indicates that FcgRIIB-SIRPa prefer- erties, and 2) that ITIMs can negatively regulate both cell entially recruited SHP-2, whereas FcgRIIB-KIR preferentially activation and cell proliferation via the recruitment of either recruited SHP-1, when coligated with the same receptors in SHPs or SHIP. RTK-dependent cell proliferation and ITAM-de- RBL cells. In apparent contradistinction with our observation, pendent cell activation can therefore be negatively regulated, SHPS-1 was reported to associate preferentially with SHP-1 in through common mechanisms, by the same receptors which a mouse macrophage cell line, when phosphorylated following may coordinately control the development and the functions of treatment with an analog of pervanadate, and in resting mouse hematopoietic cells. spleen cells (17). Whether the discrepancy can be explained by differences in the cell types and/or the experimental conditions Acknowledgments—We thank Dr. M. Arock (Faculte ´ de Pharmacie, Paris, France) for cord blood-derived human mast cells; Dr. C. Saute `s needs to be clarified. (Institut Curie, Paris, France) for rabbit anti-FcgRIIB antibodies; Dr. We found that FcgRIIB-SIRPa turned off signals transduced J.-P. Kinet (Beth Israel Hospital, Boston, MA) for rabbit anti-FcRg and by FceRI. The inhibition of serotonin release and of TNFa anti-mouse IgE antibodies; and Dr. E. Tartour (Institut Curie, Paris, synthesis was correlated with attenuated Ca responses, af- France) for the melanoma cell line used to amplify SIRPa cDNA. fecting both the mobilization of intracellular Ca stores and REFERENCES the influx of extracellular Ca and with an abolition of the 1. Matozaki, T., Uchida, Y., Fujioka, Y., and Kasuga, M. (1994) Biochem. Bio- activation of Erk1/2, as assessed by their phosphorylation. Our phys. Res. Commun. 204, 874 – 881 data neither support nor exclude the possibility that inhibition 2. Fujioka, Y., Matozaki, T., Noguchi, T., Iwamatsu, A., Yamao, T., Takahashi, 21 N., Tsuda, M., Takada, T., and Kasuga, M. (1996) Mol. Cell. Biol. 16, directly affected Ca responses or the Ras pathway. The ob- 6887– 6899 served reduced phosphorylation of FceRI ITAMs, however, in- 3. Kharitonenkov, A., Chen, Z., Sures, I., Wang, H., Schilling, J., and Ullrich, A. (1997) Nature 386, 181–186 dicates that inhibition affected signaling events that stand 4. Ohnishi, H., Kubota, M., Ohtake, A., Sato, K., and Sano, S.-I. (1996) J. Biol. upstream of these two responses. One likely explanation would Chem. 271, 25569 –25574 5. Jiang, P., Lagenaur, C. F., and Narayanan, V. (1999) J. Bol. Chem. 274, b and FcRg ITAMs or, possibly, the Src protein be that the FcR 559 –562 tyrosine kinase which phosphorylates FceRI ITAMs, were the 6. Stofega, M. R., Wang, H., Ullrich, A., and Carter-Su, C. (1998) J. Biol. Chem. substrates of protein tyrosine phosphatases recruited by phos- 273, 7112–7117 7. Vivier, E., and Dae ¨ ron, M. (1997) Immunol. Today 18, 286 –291 phorylated FcgRIIB-SIRPa. This mechanism has been pro- 8. Dae ¨ ron, M., Latour, S., Malbec, O., Espinosa, E., Pina, P., Pasmans, S., and posed to account for SHP-1-dependent inhibition of FcgRIIIA- Fridman, W. H. (1995) Immunity 3, 635– 646 9. Scharenberg, A. M., and Kinet, J.-P. (1996) Cell 87, 961–964 mediated antibody-dependent cell-mediated cytotoxicity by 10. Dae ¨ ron, M. (1997) The Immunologist 5, 79–86 KIRs (45). Whether SHP-2 can have the same effect as SHP-1 11. Moretta, A., Biassoni, R., Bottino, C., Pende, D., Vitale, M., Poggi, A., Mingari, M. C., and Moretta, L. (1997) Immunol. Rev. 155, 105–117 is unclear. SHP-2 was indeed assigned both positive (46) and 12. Burshtyn, D. N., Scharenberg, A. M., Wagtmann, N., Rajogopalan, S., Berrada, negative effects (13, 47). One may notice that, apparently, K., Yi, T., Kinet, J.-P., and Long, E. O. (1996) Immunity 4, 77– 85 FcgRIIB-SIRPa was not much more inhibitory than FcgRIIB- 13. Bruhns, P., Marchetti, P., Fridman, W. H., Vivier, E., and Dae ¨ ron, M. (1999) J. Immunol. 162, 3168 –3175 KIR, although comparable amounts of SHP-1 coprecipitated 14. Fong, D. C., Malbec, O., Arock, M., Cambier, J. C., Fridman, W. H., and with the two chimeras but much more SHP-2 with the SIRP Dae ¨ ron, M. (1996) Immunol. Lett. 54, 83–91 15. Malbec, O., Fridman, W. H., and Dae ¨ ron, M. (1999) J. Immunol. 162, chimera. This makes the role of SHP-2 in inhibition unclear. 4424 – 4429 One interesting possibility would be that SHP-1 and SHP-2 16. Adams, S., van der Laan, L. J. V., Vernon-Wilson, E., Renardel de Lavalette, C., Do ¨ pp, E. A., Dijkstra, C. D., Simmons, D. L., and van der Berg, T. K. have different substrates that may not fulfill the same func- (1998) J. Immunol. 161, 1853–1859 tions. It has been suggested that the inhibitory properties of 17. Veillette, A., Thibaudeau, E., and Latour, S. (1998) J. Biol. Chem. 273, receptors that recruit SHP-2 could result from the sequestra- 22719 –22728 18. Reth, M. G. (1989) Nature 338, 383–384 tion of that phosphatase, preventing it from exerting positive 19. Dae ¨ ron, M. (1997) Annu. Rev. Immunol. 15, 203–234 effects on activating receptors (16). If so, inhibition by SIRPa 20. Blank, U., Ra, C., Miller, L., White, K., Metzger, H., and Kinet, J. P. (1989) Nature 337, 187–189 might result from the combined dephosphorylation of ITAMs 21. Scholl, D. A., and Geha, R. S. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, and kinases by SHP-1 and the retention of SHP-2. Under these 8847– 8850 conditions, when recruited by SIRPa, SHP-1 and SHP-2 would 22. Pfefferkorn, L. C., and Yeaman, G. R. (1994) J. Immunol. 153, 3228 –3236 Inhibition of Cell Activation by Signal Regulatory Proteins 32499 23. Ra, C., Jouvin, M. H. E., Blank, U., and Kinet, J. P. (1989) Nature 341, 35. Paolini, R., Jouvin, M. H., and Kinet, J. P. (1991) Nature 353, 855– 858 752–754 36. Heldin, C.-H., and Ostman, A. (1996) Cytokines & Growth Factor Rev. 7, 3–10 24. Wagtmann, N., Biassoni, R., Cantoni, C., Verdiani, S., Malnati, M. S., Vitale, 37. Cambier, J. C. (1995) J. Immunol. 155, 3281–3285 M., Bottino, C., Moretta, L., Moretta, A., and Long, E. (1995) Immunity 2, 38. Rottapel, R., Reedijk, M., Williams, D. E., Lyman, S. D., Anderson, D. M., 439 – 449 Powson, T., and Bernstein, A. (1991) Mol. Cell. Biol. 11, 3043–3051 25. Takebe, Y., Seiki, M., Fujisawa, J. I., Hoy, P., Yokota, K., Arai, K. I., Yoshida, 39. Timms, J. F., Carlberg, K., Gu, H., Chen, H., Kamatkar, S., Nadler, M. J., M., and Arai, N. (1988) Mol. Cell. Biol. 8, 466 – 472 Rohrschneider, L. R., and Neel, B. G. (1998) Mol. Cell. Biol. 18, 3838 –3850 26. Kishi, K. (1985) Leuk. Res. 9, 381–390 40. Reinhold, M. I., Lindberg, F. P., Kersh, G. J., Allen, P. M., and Brown, E. J. 27. Butterfield, J. H., Weiler, D., Dewald, G., and Gleich, G. J. (1988) Leuk. Res. (1997) J. Exp. Med. 185, 1–11 12, 345–355 41. Parkos, C. A., Colgan, S. P., Liang, T. W., Nusrat, A., Bacarra, A. E., Carnes, 28. Barsumian, E. L., Isersky, C., Petrino, M. G., and Siraganian, R. P. (1981) Eur. D. K., and Madara, J. L. (1996) J. Cell Biol. 132, 437– 450 J. Immunol. 11, 317–323 42. Ghannadan, M., Baghestanian, M., Wimazal, F., Eisenmenger, M., D., L., 29. Saito, H., Ebisawa, M., Tachimoto, H., Shichijo, M., Fukagawa, K., Kargul, G., Walchshofer, S., Sillaber, C., Lechner, K., and Valent, P. (1998) Matsumoto, K., Iikura, Y., Awaji, T., Tsujimoto, G., Yanagida, M., J. Invest. Dermatol. 111, 689 – 695 Uzumaki, H., Takahashi, G., Tsuji, K., and Nakahata, T. (1996) 43. Furuzawa, T., Yanai, N., Hara, T., Miyajima, A., and Obinata, M. (1998) J. Immunol. 157, 343–350 J. Biochem. (Toyko) 123, 101–106 30. Dae ¨ ron, M., Bonnerot, C., Latour, S., and Fridman, W. H. (1992) J. Immunol. 44. Cherif-Zahar, B., Matassi, G., Raynal, V., Gane, P., Delaunay, J., 149, 1365–1373 Arrizabalaga, B., and Cartron, J. P. (1998) Blood 92, 2535–2540 31. Liu, T. T., Bohn, J. W., Ferry, E. L., Yamamoto, H., and Molinaro, C. A. (1980) 45. Binstadt, B. A., Brumbaugh, K. M., Dick, C. J., Scharenberg, A. M., Williams, J. Immunol. 124, 2728 –2737 B. L., Colonna, M., Lanier, L. L., Kinet, J.-P., Abraham, R. T., and Leibson, 32. Unkeless, J. C. (1979) J. Exp. Med. 150, 580 –596 P. J. (1996) Immunity 5, 629 – 638 33. Latour, S., Bonnerot, C., Fridman, W. H., and Dae ¨ ron, M. (1992) J. Immunol. 46. Gadina, M., Stancato, L. M., Bacon, C. M., Larner, A. C., and O’Shea, J. J. 149, 2155–2162 (1998) J. Immunol. 160, 4657– 4661 34. Malbec, O., Fong, D., Turner, M., Tybulewicz, V. L. J., Cambier, J. C., 47. Waterhouse, P., Marenge ` re, L. E. M., Mittru ¨ cker, H.-W., and Mak, T. W. Fridman, W. H., and Dae ¨ ron, M. (1998) J. Immunol. 160, 1647–1658 (1996) Immunol. Rev. 153, 183–207

Journal

Journal of Biological Chemistry – Unpaywall

Published: Nov 1, 1999

There are no references for this article.