HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Liénard, H. Articles by Daëron, M. Search for Related Content PubMed PubMed Citation Articles by Liénard, H. Articles by Daëron, M. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 45, 32493-32499, November 5, 1999

Signal Regulatory Proteins Negatively Regulate Immunoreceptor-dependent Cell Activation^*

Hélène Liénard, Pierre Bruhns^§, Odile Malbec, Wolf H. Fridman, and Marc Daëron^¶

From the Laboratoire d'Immunologie Cellulaire et Clinique, INSERM U.255, Institut Curie, 75005 Paris, France

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Signal regulatory proteins of the  subtype (SIRP) are ubiquitous molecules of the immunoglobulin superfamily that negatively regulate protein tyrosine kinase receptor-dependent cell proliferation. Their intracytoplasmic domain contains four motifs that resemble immunoreceptor tyrosine-based inhibition motifs (ITIMs) and that, when tyrosyl-phosphorylated, recruit cytoplasmic SH2 domain-bearing protein tyrosine phosphatases (SHPs). ITIMs are borne by molecules that negatively regulate cell activation induced by receptors bearing immunoreceptor tyrosine-based activation motifs (ITAMs). Because SIRP are coexpressed with ITAM-bearing receptors in hematopoietic cells, we investigated whether SIRP could negatively regulate ITAM-dependent cell activation. We found SIRP transcripts in human mast cells, and we show that a chimeric molecule having the transmembrane and intracytoplasmic domains of SIRP could inhibit IgE-induced mediator secretion and cytokine synthesis by mast cells. Inhibition required that the SIRP chimera was coaggregated with ITAM-bearing high affinity IgE receptors (FcRI). It was correlated with the tyrosyl phosphorylation of the SIRP chimera and the recruitment of SHP-1 and SHP-2. The phosphorylation of FcRI ITAMs was decreased; the mobilization of intracellular Ca²⁺ and the influx of extracellular Ca²⁺ were reduced, and the activation of the mitogen-activated protein kinases Erk1 and Erk2 was abolished. SIRP can therefore negatively regulate not only receptor tyrosine kinase-dependent cell proliferation but also ITAM-dependent cell activation.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Signal regulatory proteins (SIRPs)¹ were described as phosphoproteins that coprecipitated with SH2 domain-bearing protein tyrosine phosphatases (SHPs) (1-3). SIRP molecules were first identified as SHP substrate 1 (SHPS-1) (2) and brain immunoglobulin-like molecules with tyrosine-based activation motifs (BIT) (4). When cloned, these molecules were found to be the members of a multigene family of at least 15 transmembrane immunoglobulin superfamily molecules named collectively SIRPs, in which two types,  and , were recognized, differing by the presence (in SIRP) or the absence (in SIRP) of an intracytoplasmic (IC) domain containing four tyrosine-based regulatory motifs (3). A neuronal adhesion molecule, previously described as P84, was found to belong to the SIRP family, and the widely expressed integrin-associated protein CD47 was recently identified as a ligand of P84 (5).

Interestingly, SIRP were shown to regulate negatively cell proliferation induced by growth factors via protein tyrosine kinase receptors (RTKs) and oncogene products. Little is known of the mechanism of inhibition by SIRP, except that negative regulation was correlated with the tyrosyl phosphorylation of SIRP and the recruitment of SHP-2 (3, 6). That SIRP had inhibitory properties indicated that these molecules bear inhibition motifs rather than activation motifs. SHP-2-binding motifs found in SIRP are indeed reminiscent of immunoreceptor tyrosine-based inhibition motifs (ITIMs). ITIMs are present in a large group of molecules (7) that negatively regulate cell activation induced by receptors bearing immunoreceptor tyrosine-based activation motifs (ITAMs) (8). Negative regulation by ITIM-bearing molecules is correlated with the recruitment of SH2 domain-bearing phosphatases by phosphorylated 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 cells (11). Their IC domain contains two ITIMs that, when tyrosyl-phosphorylated, recruit SHP-1 and SHP-2 (12, 13). Likewise, FcRIIB, a family of low affinity receptors for IgG, negatively regulate cell activation via ITAM-bearing receptors when coaggregated with the latter by immune complexes (8). FcRIIB bear a single ITIM that, when tyrosyl-phosphorylated, recruits selectively the SH2 domain-bearing inositol-5-phosphatase SHIP (14). We found recently that FcRIIB are also capable of inhibiting RTK-dependent cell proliferation (15). This finding opened the possibility that it might be a general property of ITIM-bearing molecules to regulate negatively not only ITAM-dependent cell activation but also RTK-mediated cell proliferation. If so, one could hypothesize that SIRP might inhibit ITAM-dependent cell activation.

By contrast with KIRs whose expression is restricted to NK cells and T cells (11), SIRP were found to be expressed by all human tissues examined, including hematopoietic cells (3), by neural and myeloid cells in rats (16), and by myeloid cells, especially macrophages, but not lymphoid cells in mice (17). Hematopoietic cells also express ITAM-bearing receptors. Thus, myeloid cells express ITAM-bearing receptors for the Fc portion of immunoglobulins (FcR) (18, 19). These comprise high affinity IgE receptors (FcRI) (20), high affinity IgG receptors (FcRI) (21), high affinity IgA receptors (FcRI) (22), and low affinity IgG receptors (FcRIIIA) (23), which all share the ITAM-bearing signal transduction subunit FcR (18). They also include the human-restricted, single chain, low affinity IgG receptors FcRIIA/C that bear one ITAM in their own IC domain (18). If they could interfere with ITAM-dependent cell activation, SIRP would negatively regulate not only the growth but also the many biological functions of hematopoietic cells triggered by Fc receptors.

To investigate the hypothesis that SIRP might negatively regulate cell activation induced by ITAM-bearing receptors, we constructed an experimental model in mast cells. Mast cells were chosen as an assay system because we found SIRP transcripts in human mast cells. We report here that, when coaggregated with FcRI, a chimeric molecule whose IC domain was that of human SIRP could inhibit IgE-induced mediator release and cytokine secretion by mast cells. The experimental model used made possible the biochemical analysis of intracellular events associated with inhibition. Inhibition was correlated with the tyrosyl phosphorylation of the SIRP chimera, with the recruitment of SHP-1 and SHP-2 by the phosphorylated chimera, with a decreased phosphorylation of FcRI ITAMs, with attenuated Ca²⁺ responses, and with an abolition of MAP kinase activation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

cDNA Constructs-- cDNA sequences encoding the transmembrane (TM) and IC domains of SIRP and of a human KIR were amplified from the human melanoma cell line HT-144 (from the ATCC) and from the p58.183 KIR cDNA (24) by RT-PCR and PCR, respectively, with the following oligonucleotide primers: for SIRP, sense, 5'-TCTAAGGTACCAAACATCTATATTGTGGTG-3', and antisense, 5'-AGCAAACCGAGCTCCCATTCACTTCCTCGGGACCTG-3'; for KIR, sense, 5'-CCCAGAC AGGTACCTGTTCTGATTGGGACC-3', and antisense, 5'-CTGACTGTGGAGCTCATGGGCAGG-3'. KpnI (GGTACC) and SacI (GAGCTC) sites are in boldface. PCR products were inserted into an expression cassette under the control of the SR promoter in pBR322 (25), containing a resistance gene to neomycin (NT-neo) and containing the extracellular domain of FcRIIB.

Cells-- The human basophil-like KU812 (26), the human mast cells HMC-1 (27), and the rat mast cells RBL-2H3 (28) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, 100 IU/ml penicillin, 100 µg/ml streptomycin, and 2 mM L-glutamine. Culture reagents were from Life Technologies, Inc. Human normal mast cells, obtained by culture of cord blood mononuclear cells as described (29), were kindly donated by Dr. M. Arock (Faculté de Pharmacie, Paris, France). More than 99% cells were mast cells as judged by morphology, the presence of metachromatic granules, and positivity for tryptase. cDNAs encoding FcRIIB-SIRP and FcRIIB-KIR were stably transfected in RBL-2H3 cells; transfectants were selected with 2.5 mg/ml G418 (Life Technologies, Inc.) and cloned as described (30).

Antibodies-- The mouse IgE anti-DNP mAb 2682-I (31) was used as culture supernatant. IgG and F(ab')2 fragments of the rat anti-mouse FcRIIB 2.4G2 mAb (32) were obtained as described (30). Rabbit anti-mouse FcRIIB were raised by Dr. C. Sautès (Institut Curie, Paris, France) against recombinant extracellular domains of FcRIIB. Mouse mAb anti-FcR were purified with protein G-Sepharose (Amersham Pharmacia Biotech) from culture supernatants of the JRK hybridoma cells. Rabbit anti-FcR antibodies and rabbit anti-mouse IgE antibodies were generous gifts of Dr. J.-P. Kinet (Beth Israel Hospital, Boston). IgG and F(ab')2 fragments of mouse anti-rat Ig (MAR) and FITC-conjugated MAR F(ab')2 fragments (FITC-MAR F(ab')2) were from Jackson ImmunoResearch Laboratories (West Grove, PA). MAR F(ab')2 were trinitrophenylated with trinitrobenzene sulfonic acid (Eastman-Kodak Co.) and purified on Sephadex G-25 (Amersham Pharmacia Biotech). TNP₁₀-MAR F(ab')2 were obtained. Horseradish peroxidase (HRP)-conjugated mouse mAb anti-phosphotyrosine (PY20) were purchased from Chemicon (Temecula, CA); mouse mAbs anti-SHP-1 and anti-SHP-2 were from Transduction Laboratories (Lexington, KY); mouse mAbs anti-Erk1/2 and anti-phospho-Erk1/2 were from New England Biolabs (Beverly, MA); HRP-conjugated polyclonal goat anti-mouse Ig (GAM) antibodies and polyclonal goat anti-rabbit Ig (GAR) antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA).

Immunofluorescence-- RBL transfectants were incubated for 1 h at 0 °C with 10 µg/ml 2.4G2 mAb in balanced salt solution containing 5% fetal calf serum. Cells were washed and stained by being incubated for 30 min at 0 °C with 50 µg/ml FITC-MAR F(ab')2. Fluorescence was analyzed by flow cytometry using a FACScalibur (Becton-Dickinson, Mountain View, CA).

RT-PCR Analysis of SIRP Transcripts-- RNA was extracted from 1 × 10⁷ KU812, HMC-1, and cord blood-derived human mast cells. cDNAs were prepared and used as templates to amplify sequences corresponding to the TM and IC domains of SIRP using the two oligonucleotides SIRP sense and antisense. Amplified fragments were sequenced using the same two oligonucleotides.

Serotonin Release-- RBL transfectants, loaded with [³H]serotonin (Amersham Pharmacia Biotech), were incubated for 1 h at 37 °C with IgE anti-DNP and with 0 or 3 µg/ml 2.4G2 F(ab')2, washed, and challenged for 30 min at 37 °C with TNP-MAR F(ab')2 or with MAR F(ab'2) and DNP-BSA. The percentage of [³H]serotonin released was measured as described (30).

TNF Secretion-- RBL transfectants, incubated for 1 h at 37 °C with IgE anti-DNP and 0 or 3 µg/ml 2.4G2 F(ab')2, were challenged for 3 h at 37 °C with 10 µg/ml TNP-MAR F(ab')2. Cell-free supernatants were harvested and titrated for TNF on L929 cells as described (33).

Measurement of Intracellular Ca²⁺ Concentration-- RBL transfectants, previously sensitized with IgE anti-DNP and incubated with or without 2.4G2 F(ab')2, were loaded with 5 mM Fluo-3-AM in the presence of 0.2% Pluronic F-127 (Molecular Probes, Eugene, OR) for 30 min at room temperature. Cells were resuspended in medium in which calcium had been buffered to 60 nM (equivalent to the intracellular Ca²⁺ concentration in resting cells) with EGTA. One min later, they were challenged with TNP-MAR F(ab')2 for 100 s, after which the extracellular Ca²⁺ concentration was raised to 1.3 mM with CaCl₂. Intracellular free Ca²⁺ concentration was monitored by flow cytometry with a FACScalibur using the software FCS assistant 1.2.9 (Becton Dickinson).

Immunoprecipitation and Western Blot Analysis-- RBL transfectants were incubated with IgE anti-DNP and 0 or 3 µg/ml 2.4G2 F(ab')2, washed, and challenged for various periods of time at 37 °C with 10 µg/ml TNP-MAR F(ab')2. Aliquots of 3 × 10⁷ cells were lysed for 10 min at 0 °C. For immunoprecipitation of FcRIIB, lysis buffer contained 10 mM Tris, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 1 mM Na₃VO₄, 5 mM NaF, 5 mM sodium pyrophosphate, 0.4 mM EDTA, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 10 µg/ml pepstatin, and 1 mM phenylmethylsulfonyl fluoride. For immunoprecipitation of FcRI, lysis buffer contained 50 mM Tris, pH 8.0, 1% Nonidet P-40, 1 mM Na₃VO₄, 20 mM EDTA, 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride. Lysates were centrifuged at 12,000 rpm for 10 min at 4 °C, and supernatants were incubated with immunoadsorbents for 1 h at 4 °C. Protein G-Sepharose or 2.4G2 coupled to protein G-Sepharose was used to precipitate FcRIIB from cells preincubated with 2.4G2 or not, respectively. Anti-mouse IgE antibodies coupled to protein A-Sepharose (Sigma) were used to precipitate FcRI from cells sensitized with mouse IgE. Immunoprecipitates were fractionated by SDS-PAGE and transferred onto Immobilon-P membranes (Millipore, Bedford, MA). Membranes were saturated either with 5% BSA (Sigma) or with 5% skimmed milk (Régilait, Saint-Martin-Belle-Roche, France) and Western-blotted with HRP-conjugated anti-phosphotyrosine antibodies or with rabbit anti-FcRIIB, mouse anti-SHP-1, mouse anti-SHP-2, mouse anti-FcR, or rabbit anti-FcR antibodies, followed by HRP-conjugated GAR or GAM. Anti-SHP-2 antibodies detected two species of SHP-2. Anti-SHP-1 and anti-SHP-2 antibodies recognized specifically SHP-1 and SHP-2, respectively. HRP-conjugated antibodies were detected using an enhanced chemiluminescence (ECL) kit (Amersham Pharmacia Biotech). When blotted sequentially with different antibodies, filters were stripped following revelation by being incubated for 30 min at 50 °C in buffer containing 62.5 mM Tris, pH 6.7, 100 mM -mercaptoethanol, and 2% SDS.

Western Blot Analysis of MAP Kinases-- RBL transfectants were incubated for 1 h at 37 °C with IgE anti-DNP and with 0 or 3 µg/ml 2.4G2 F(ab')2, washed, and challenged with 10 µg/ml TNP-MAR F(ab')2. Equal volumes of cell lysates, obtained as described for immunoprecipitation of FcRIIB, were fractionated by SDS-PAGE, transferred onto Immobilon-P membranes, and blotted with anti-phospho-Erk1/2 or with anti-Erk1/2 antibodies, followed by HRP-conjugated GAR. HRP-conjugated antibodies were detected using the ECL kit.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Human Mast Cells Express SIRP-- We investigated first the presence of SIRP transcripts in RNA from human mast cells by RT-PCR using oligonucleotide primers that could amplify a 435-base pair long fragment corresponding to the TM and IC domains of human SIRP. A single PCR product having the expected size was amplified from cDNAs from the human mast cell line HMC-1 and from cord blood-derived human mast cells but not from the basophil-like cell line KU812 (Fig. 1A). PCR products amplified from HMC-1 and from cord blood-derived mast cells were sequenced. They had the same nucleotide sequence as cDNA encoding the TM and IC domains of human SIRP (3).

View larger version (33K): [in this window] [in a new window]   Fig. 1.   SIRP transcripts in human mast cells and structure and expression of recombinant molecules expressed in RBL cells. A, SIRP transcripts in human mast cells. cDNAs from a human basophil-like cell line (KU812), a human mast cell line (HMC-1), and cord blood-derived human mast cells (CB-HMC) were analyzed by RT-PCR using oligonucleotides that could amplify sequences encompassing the TM and IC domains of SIRP. PCR products were analyzed by agarose gel electrophoresis. bp, base pair. B, structure of recombinant molecules. The figure schematizes the structure of the IC domain-deleted FcRIIB(IC1), of the FcRIIB-SIRP chimera, and of the FcRIIB-KIR chimera expressed in RBL cells. ITIM-like sequences are indicated at their respective positions. C, expression of recombinant molecules in RBL-2H3 cells. Clones of stable transfectants were examined by immunofluorescence for the expression of FcRIIB-based molecules. Filled histograms, cells incubated with FITC-MAR F(ab')2 only; open histograms, cells incubated with 2.4G2 and FITC-MAR F(ab')2.

A SIRP Chimera Inhibits IgE-induced Mast Cell Activation When Coligated with FcRI-- In order to examine whether SIRP could inhibit IgE-induced mast cell activation, we constructed a cDNA encoding a chimeric molecule having the extracellular domain of murine FcRIIB and the TM and IC domains of human SIRP (FcRIIB-SIRP). As a positive control, a cDNA encoding a chimeric molecule having the extracellular domain of murine FcRIIB and the TM and IC domains of a human KIR (FcRIIB-KIR) was also constructed (Fig. 1B). Chimeric cDNAs were stably transfected in the rat mast cells RBL-2H3, which constitutively express FcRI. Clones of transfectants expressing comparable amounts of FcRIIB-SIRP or FcRIIB-KIR (Fig. 1C) and, as negative controls, clones expressing tail-less FcRIIB (FcRIIB(IC1)) (30, 33) were used for experiments. Several clones expressing each molecule gave comparable results.

FcRIIB-SIRP, FcRIIB-KIR, or FcRIIB(IC1) were coaggregated with FcRI in RBL transfectants sensitized with mouse IgE anti-DNP, incubated with F(ab')2 fragments of the rat anti-mouse FcRIIB mAb 2.4G2, and challenged with TNP-MAR F(ab')2 as described previously (34). Serotonin release induced under these conditions was compared with serotonin release induced by aggregating FcRI with TNP-MAR F(ab')2 in the same three transfectants sensitized with mouse IgE anti-DNP but not incubated with 2.4G2 F(ab')2. When coaggregated with FcRI, FcRIIB-SIRP inhibited serotonin release as efficiently as FcRIIB-KIR, but not FcRIIB(IC1) (Fig. 2A).

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Inhibition of IgE-induced mast cell activation by FcRIIB-SIRP. A, inhibition of serotonin release. Cells expressing FcRIIB-SIRP, FcRIIB-KIR, or FcRIIB(IC1) were sensitized with indicated dilutions of IgE anti-DNP and with 0 (open circles) or 3 µg/ml (closed circles) 2.4G2 F(ab')2. Cells were challenged with 10 µg/ml TNP-MAR F(ab')2. The figure shows the percentage of serotonin released as a function of IgE dilutions. B, requirement for the coaggregation of FcRI with FcRIIB-SIRP for inhibition of serotonin release. Cells expressing FcRIIB-SIRP were sensitized with indicated dilutions of IgE anti-DNP and with 0 (open symbols) or 3 µg/ml (closed symbols) 2.4G2 F(ab')2. Cells were challenged either with 10 µg/ml TNP-MAR F(ab')2 (circles) or with 3 µg/ml DNP-BSA and 10 µg/ml MAR F(ab')2 (squares). The figure shows the percentage of serotonin released as a function of IgE dilutions. C, inhibition of TNF production. Cells expressing FcRIIB-SIRP were sensitized with IgE anti-DNP (supernatant 1/10) and incubated with 0 (open circles) or 3 µg/ml (closed circles) 2.4G2 F(ab')2. They were challenged with 10 µg/ml TNP-MAR F(ab')2 for 3 h at 37 °C. Cell-free supernatants were harvested, and serial 2-fold dilutions were tested for cytotoxicity on L929 cells. The figure represents the percentage of cytotoxicity as a function of the dilution of supernatants. Experiments shown in A-C were repeated three times with the same results.

To determine the conditions required for FcRIIB-SIRP to inhibit serotonin release, transfectants expressing FcRIIB-SIRP were sensitized with mouse IgE anti-DNP and incubated with 2.4G2 F(ab')2. FcRI and FcRIIB-SIRP were then either coaggregated by TNP-MAR F(ab')2 as above or aggregated simultaneously, but independently, by DNP-BSA and MAR F(ab')2, respectively. Cells sensitized with IgE but not incubated with 2.4G2 F(ab')2 served as positive controls. Serotonin release was inhibited when FcRI and FcRIIB-SIRP were coaggregated by TNP-MAR F(ab')2 but not when they were simultaneously aggregated by DNP-BSA and MAR F(ab')2 (Fig. 2B).

To determine whether FcRIIB-SIRP could inhibit an IgE-induced secretion of cytokine, FcRI were either aggregated or coaggregated with FcRIIB-SIRP in transfectants expressing FcRIIB-SIRP, using the same ligands as above, and the amount of TNF released in culture supernatants during the following 3 h was titrated using a bioassay. When coaggregated with FcRI, FcRIIB-SIRP inhibited about 90% of the TNF secretion (Fig. 2C).

The SIRP Chimera Becomes Tyrosyl-phosphorylated upon Coligation with FcRI and Recruits Protein Tyrosine Phosphatases-- We examined next the phosphorylation of FcRIIB-SIRP when coaggregated with FcRI or not. FcRIIB-SIRP was immunoprecipitated from transfectants expressing FcRIIB-SIRP that had been incubated with medium alone, IgE anti-DNP and/or 2.4G2 F(ab')2, and challenged with or without TNP-MAR F(ab')2 for 3 min at 37 °C. Western blot analysis with anti-phosphotyrosine antibodies showed that the basal phosphorylation of the chimera slightly increased when FcRIIB-SIRP was aggregated but not when FcRI were aggregated. It increased dramatically when FcRIIB-SIRP was coaggregated with FcRI (Fig. 3A). When FcRIIB-SIRP was coaggregated with FcRI for various periods, its phosphorylation was apparent at 1 min, maximum between 3 and 10 min, and still visible at 30 min (Fig. 3B). When coaggregated with FcRI, tail-less FcRIIB failed to be phosphorylated (data not shown). Four tyrosine residues are present in the IC domain of SIRP, which are each constitutive of an ITIM-like motif.

View larger version (60K): [in this window] [in a new window]   Fig. 3.   Phosphorylation of FcRIIB-SIRP upon coaggregation with FcRI and coprecipitation of SHP-1 and SHP-2 with phosphorylated FcRIIB-SIRP. RBL transfectants were preincubated with or without 3 µg/ml 2.4G2 F(ab')2, sensitized or not with IgE anti-DNP (supernatant 1/10), and challenged with or without 10 µg/ml TNP-MAR F(ab')2. A, phosphorylation of FcRIIB-SIRP upon coaggregation with FcRI. Aliquots of 1 × 107 cells were lysed 3 min after stimulation. FcRIIB-SIRP immunoprecipitates were fractionated by SDS-PAGE (10% acrylamide), transferred onto Immobilon, and Western-blotted with anti-phosphotyrosine antibodies. The filter was stripped and reblotted with anti-FcRIIB antibodies. Comparable results were obtained in three separate experiments. B, kinetics of FcRIIB-SIRP phosphorylation. Aliquots of 3 × 106 cells were lysed at indicated times. FcRIIB-SIRP immunoprecipitates were fractionated by SDS-PAGE (10% acrylamide), transferred onto Immobilon, and Western-blotted with anti-phosphotyrosine antibodies. The filter was stripped and reblotted with anti-FcRIIB antibodies. Comparable results were obtained in two separate experiments. C, coprecipitation of SHPs with phosphorylated FcRIIB-SIRP. Aliquots of 3 × 107 cells were lysed 3 min after stimulation, and FcRIIB-SIRP was immunoprecipitated. Left, aliquots of immunoprecipitates corresponding to 3 × 106 RBL cells were fractionated by SDS-PAGE (10% acrylamide), transferred onto Immobilon, and Western-blotted with anti-phosphotyrosine antibodies, stripped and reblotted with anti-FcRIIB antibodies. Right, remaining immunoprecipitates were fractionated by SDS-PAGE (8% acrylamide), transferred onto Immobilon, and Western-blotted with anti-SHIP (upper panel) or with anti-SHP-2 and then anti-SHP-1 without stripping (lower panel). The same blot was stripped and reblotted with anti-FcRIIB antibodies. Whole cell lysates (WCL) were used as positive controls for Western blotting (only lysates from transfectants expressing FcRIIB-SIRP are shown). Comparable results were obtained in two separate experiments.

To identify SH2 domain-bearing phosphatases possibly recruited in vivo by FcRIIB-SIRP, the SIRP chimera was immunoprecipitated after it had or had not been coaggregated with FcRI, and phosphatases coprecipitated with FcRIIB-SIRP were identified by Western blotting. Some SHP-2 coprecipitated with weakly phosphorylated FcRIIB-SIRP in resting cells, and high amounts of SHP-2 coprecipitated with FcRIIB-SIRP that became heavily phosphorylated upon coaggregation with FcRI. SHP-1 was coprecipitated with FcRIIB-SIRP following coaggregation of the chimera with FcRI only. Under the same conditions, FcRIIB-KIR, which was also phosphorylated upon coaggregation with FcRI, recruited similar amounts of SHP-1 as FcRIIB-SIRP but much less SHP-2. SHIP was recruited neither by FcRIIB-SIRP nor by FcRIIB-KIR (Fig. 3C).

The SIRP Chimera Blocks Signal Transduction by FcRI-- An early step in IgE-induced signaling in mast cells is the tyrosyl phosphorylation of FcRI ITAMs (35) which enables the recruitment of SH2 domain-bearing cytoplasmic kinases, leading to an increased intracellular Ca²⁺ concentration, and adapter molecules that connect phosphorylated receptors with the Ras pathway, leading to the transcription of cytokine genes. In RBL cells, FcRI are associated with two ITAM-bearing subunits, FcR and FcR (20).

The phosphorylation of FcRI ITAMs was examined by Western blot analysis with anti-phosphotyrosine antibodies in RBL transfectants sensitized with IgE anti-DNP, after FcRI were aggregated or coaggregated with the SIRP chimera for 1, 3, or 10 min at 37 °C and immunoprecipitated with anti-IgE antibodies. FcR and FcR were identified by Western blot analysis of the same filter with corresponding specific antibodies. The phosphorylation of both FcR and FcR induced upon FcRI aggregation was decreased upon coaggregation of FcRI with the SIRP chimera (Fig. 4A).

View larger version (30K): [in this window] [in a new window]   Fig. 4.   FcRIIB-SIRP inhibits intracellular signaling by FcRI. A, inhibition of the phosphorylation of FcRI ITAMs. Aliquots of 5 × 107 RBL transfectants expressing FcRIIB-SIRP were sensitized with IgE anti-DNP (supernatant 1/10), preincubated with or without 3 µg/ml 2.4G2 F(ab')2, and challenged with or without 10 µg/ml TNP-MAR F(ab')2 for 1, 3, or 10 min at 37 °C. FcRI 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-FcR or anti-FcR antibodies, respectively. Comparable results were obtained in three separate experiments. B, inhibition of Ca2+ mobilization. RBL transfectants expressing FcRIIB-SIRP or FcRIIB-KIR were sensitized with IgE anti-DNP (supernatant 1/10) and preincubated with (gray) or without (black) 3 µg/ml 2.4G2 F(ab')2. They were loaded with Fluo-3-AM, resuspended in medium containing 60 nM Ca2+, and stimulated with 10 µg/ml TNP-MAR F(ab')2 (arrows). One hundred seconds later, the extracellular Ca2+ concentration was raised to 1.3 mM with CaCl2. 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 FcRIIB-SIRP or FcRIIB-KIR were preincubated with or without 3 µg/ml 2.4G2 F(ab')2, sensitized or not with IgE anti-DNP (supernatant 1/10), challenged with or without TNP-MAR F(ab')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.

Ca²⁺ responses were monitored in RBL transfectants following FcRI aggregation or following the coaggregation of FcRI with the SIRP chimera and, as a positive control, with the KIR chimera, under conditions that permitted us to measure the mobilization of intracellular Ca²⁺ and the influx of extracellular Ca²⁺ separately. FcRIIB-SIRP and FcRIIB-KIR similarly reduced both the intracellular Ca²⁺ mobilization and the extracellular Ca²⁺ influx (Fig. 4B).

The activation of the Ras pathway was assessed by examining the phosphorylation of the MAP kinases Erk1 and Erk2. When coaggregated with FcRI, both FcRIIB-SIRP and FcRIIB-KIR abolished IgE-induced Erk1/2 phosphorylation induced by aggregating FcRI (Fig. 4C).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

SIRP were described as negative regulators of growth factor-induced, RTK-mediated cell proliferation (3). RTKs transphosphorylate each other when dimerized by growth factors (36). They induce cells to enter the cell cycle and to divide. ITAM-bearing receptors have no intrinsic protein kinase activity, and they do not induce cells to proliferate. When aggregated by multivalent ligands, they are phosphorylated by Src family protein tyrosine kinases, and phosphorylated ITAMs serve as docking sites for cytoplasmic SH2 domain-bearing protein tyrosine kinases and adapter molecules that lead to cell activation (37). Although both early and late signals delivered by RTKs and ITAM-bearing receptors are different, these two types of receptors utilize common intracellular effectors such as phospholipase C-, phosphatidylinositol 3-kinase, and molecules of the Ras/MAP kinase pathway (37, 38). We found SIRP transcripts in human mast cells, and we investigated the effects of a SIRP chimera on the secretion of inflammatory mediators triggered by FcRI, a typical ITAM-bearing Fc receptor, in the rat mast cell line RBL-2H3. We provide here the first evidence that SIRP can behave as ITIM-bearing molecules that negatively regulate ITAM-dependent cell activation, and we document the mechanism of inhibition by SIRP.

The experimental model used to demonstrate the inhibitory properties of SIRP was validated by our finding that human mast cells contain SIRP transcripts. This conclusion could be drawn from results obtained with both cord blood-derived human mast cells and with the human mastocytoma cells HMC-1. Cord blood-derived mast cells present the advantage of being normal, non-transformed cells but have the disadvantage of being potentially contaminated by (less than 1%) other cells, possibly including macrophages that express high levels of SIRPs. HMC-1 cells maintained in culture for years have the disadvantage of being transformed cells, but they have the advantage of excluding any cell contamination. The combined results obtained in the two cell types make it reasonable to conclude that human mast cells do express the SIRP gene, and although the expression of SIRP proteins was not formally demonstrated here, they justify that mast cell secretory responses were chosen as readouts to assess the ability of SIRP to control cell activation.

The overexpression of SIRP was reported to be sufficient to inhibit epidermal growth factor-, platelet-derived growth factor-, or insulin-induced cell proliferation in NIH-3T3 fibroblasts and ligand-independent proliferation of the same cells infected by a retrovirus carrying an oncogenic form of RTK (3). SIRP are therefore likely to be constitutively associated with RTKs, and indeed, BIT coprecipitated with CSF-1R in a macrophage cell line (39). By contrast, the overexpression of the FcRIIB-SIRP chimera in RBL-2H3 cells affected neither the growth of transfectants (data not shown) nor IgE-induced responses. Mediator release was inhibited when and only when FcRI and FcRIIB-SIRP were coaggregated by the same extracellular ligand. A possible explanation for the different requirements for SIRP to inhibit RTK-dependent cell proliferation and for FcRIIB-SIRP to inhibit FcRI-dependent cell activation might be that SIRP constitutively associate with RTKs (and possibly with FcRI) via their extracellular domains which were removed in the FcRIIB-SIRP chimera. Specific conditions under which FcRI and SIRP could be coligated, if not constitutively associated, on mast cells remain to be determined, and it was not an aim of this investigation to address that issue. Our work indicates, however, that a previously undemonstrated negative regulation may operate in hematopoietic cells that coexpress SIRP and ITAM-bearing receptors, provided the two receptors are maintained close to each other, either constitutively or inducibly by common extracellular ligands. One SIRP member, the neuronal adhesion molecule P84, was recently assigned an extracellular ligand. The integrin-associated protein CD47 was indeed found to bind to P84, and both molecules are colocalized in synapse-rich structures of the cerebellum and the retina where their interactions were proposed to control synaptic functions (5). Whether CD47 is also a ligand for other SIRP family members and/or whether other ligands bind to the polymorphic extracellular domains of SIRP is unknown. CD47 is widely expressed on T cells (40), on myeloid cells including neutrophils (41) and mast cells (42), on epithelial cells (41), on spleen, liver and bone marrow stromal cells (43), on neurones (5, 16), and on red blood cells (44).

The mechanism underlying the inhibitory properties of SIRPs is poorly understood. One reason is the constitutive association of RTKs with SIRPs that made it difficult to analyze the respective roles of interacting molecules. The experimental model used here enabled us to dissect various stages of the process and to get some insight in the mechanism of inhibition by SIRP. One consequence of the coaggregation of FcRI with FcRIIB-SIRP was a dramatic tyrosyl phosphorylation of the chimera. A faint basal phosphorylation of FcRIIB-SIRP was observed in resting cells. This phosphorylation slightly increased upon FcRIIB-SIRP aggregation, suggesting that low levels of protein tyrosine kinase may associate with the chimera in mast cells. It did not increase upon stimulation by IgE, confirming that FcRIIB-SIRP was not associated with FcRI. The increased phosphorylation of FcRIIB-SIRP, when coaggregated with FcRI, is likely to depend on Src protein tyrosine kinases that are recruited by aggregated FcRI, as it was previously demonstrated for the tyrosyl phosphorylation of FcRIIB (34).

Phosphorylated FcRIIB-SIRP was found to recruit SH2 domain-bearing cytoplasmic protein tyrosine phosphatases. The respective roles of the four ITIM-like motifs in phosphatase recruitment is not known as no mutational analysis of the four tyrosine residues contained in the SIRP IC domain has been made. Some SHP-2 coprecipitated with lightly phosphorylated FcRIIB-SIRP in resting cells and much greater amounts with heavily phosphorylated FcRIIB-SIRP, following its coaggregation with FcRI. Significant amounts of SHP-1 also coprecipitated with the phosphorylated chimera. Association with SHP-2 was instrumental for the identification of SIRPs (1-3). SHP-1 associated with SIRP when overexpressed in fibroblasts (2) and with SHPS-1 (17) and BIT (39) in macrophages. Noticeably, comparable amounts of SHP-1 coprecipitated with FcRIIB-SIRP and with FcRIIB-KIR, whereas much more SHP-2 coprecipitated with FcRIIB-SIRP than with FcRIIB-KIR. This indicates that FcRIIB-SIRP preferentially recruited SHP-2, whereas FcRIIB-KIR preferentially recruited SHP-1, when coligated with the same receptors in RBL cells. In apparent contradistinction with our observation, SHPS-1 was reported to associate preferentially with SHP-1 in a mouse macrophage cell line, when phosphorylated following treatment with an analog of pervanadate, and in resting mouse spleen cells (17). Whether the discrepancy can be explained by differences in the cell types and/or the experimental conditions needs to be clarified.

We found that FcRIIB-SIRP turned off signals transduced by FcRI. The inhibition of serotonin release and of TNF synthesis was correlated with attenuated Ca²⁺ responses, affecting both the mobilization of intracellular Ca²⁺ stores and the influx of extracellular Ca²⁺ and with an abolition of the activation of Erk1/2, as assessed by their phosphorylation. Our data neither support nor exclude the possibility that inhibition directly affected Ca²⁺ responses or the Ras pathway. The observed reduced phosphorylation of FcRI ITAMs, however, indicates that inhibition affected signaling events that stand upstream of these two responses. One likely explanation would be that the FcR and FcR ITAMs or, possibly, the Src protein tyrosine kinase which phosphorylates FcRI ITAMs, were the substrates of protein tyrosine phosphatases recruited by phosphorylated FcRIIB-SIRP. This mechanism has been proposed to account for SHP-1-dependent inhibition of FcRIIIA-mediated antibody-dependent cell-mediated cytotoxicity by KIRs (45). Whether SHP-2 can have the same effect as SHP-1 is unclear. SHP-2 was indeed assigned both positive (46) and negative effects (13, 47). One may notice that, apparently, FcRIIB-SIRP was not much more inhibitory than FcRIIB-KIR, although comparable amounts of SHP-1 coprecipitated with the two chimeras but much more SHP-2 with the SIRP chimera. This makes the role of SHP-2 in inhibition unclear. One interesting possibility would be that SHP-1 and SHP-2 have different substrates that may not fulfill the same functions. It has been suggested that the inhibitory properties of receptors that recruit SHP-2 could result from the sequestration of that phosphatase, preventing it from exerting positive effects on activating receptors (16). If so, inhibition by SIRP might result from the combined dephosphorylation of ITAMs and kinases by SHP-1 and the retention of SHP-2. Under these conditions, when recruited by SIRP, SHP-1 and SHP-2 would act in concert, and their complementary effects might explain the deep inhibition of the biological responses observed. Whatever the respective roles of the two phosphatases, our work documents a reciprocal cross-talk between SIRP and ITAM-bearing receptors during inhibition of cell activation. The coaggregation of FcRIIB-SIRP with FcRI may indeed enable both FcRI to provide protein tyrosine kinases that could phosphorylate SIRP ITIMs and FcRIIB-SIRP to provide protein tyrosine phosphatases that could dephosphorylate FcRI ITAMs and sequester phosphatases possibly required for cell activation.

Myeloid cells, such as macrophages and mast cells, coexpress not only ITAM-bearing FcR and SIRP but also RTKs and FcRIIB. We found recently that FcRIIB, which were known to regulate negatively ITAM-dependent cell activation (8), can also negatively regulate RTK-dependent cell proliferation and that inhibition is correlated with the recruitment of the SH2 domain-bearing inositol-5-phosphatase SHIP (15). We show here that SIRP, which were known to regulate negatively RTK-dependent cell proliferation, can also negatively regulate ITAM-dependent cell activation and that inhibition is correlated with the recruitment of SHP-2 and SHP-1. It follows 1) that SIRP bear typical ITIMs which endow these molecules with a wide array of previously unsuspected regulatory properties, and 2) that ITIMs can negatively regulate both cell activation and cell proliferation via the recruitment of either SHPs or SHIP. RTK-dependent cell proliferation and ITAM-dependent cell activation can therefore be negatively regulated, through common mechanisms, by the same receptors which may coordinately control the development and the functions of hematopoietic cells.

	ACKNOWLEDGEMENTS

We thank Dr. M. Arock (Faculté de Pharmacie, Paris, France) for cord blood-derived human mast cells; Dr. C. Sautès (Institut Curie, Paris, France) for rabbit anti-FcRIIB antibodies; Dr. J.-P. Kinet (Beth Israel Hospital, Boston, MA) for rabbit anti-FcR and anti-mouse IgE antibodies; and Dr. E. Tartour (Institut Curie, Paris, France) for the melanoma cell line used to amplify SIRP cDNA.

	FOOTNOTES

^* This work was supported by the INSERM, the Association pour la Recherche sur le Cancer, and the Institut Curie.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of a Rhône-Poulenc Rorer CIFRE contract.

^§ Recipient of a fellowship from the Ministère de l'Enseignement Supérieur et de la Recherche.

^¶ To whom correspondence should be addressed: Laboratoire d'Immunologie Cellulaire et Clinique, INSERM U.255, Institut Curie, 26 rue d'Ulm, 75005 Paris, France. Tel.: 33-1-4432-4223; Fax: 33-1-4051-0420; E-mail: Marc.Daeron@curie.fr.

	ABBREVIATIONS

The abbreviations used are: BIT, brain immunoglobulin-like molecule with tyrosine-based activation motifs; FcRI, high affinity IgE receptors; FcRIIB and FcRIIIA, low affinity IgG Receptors; GAM, goat anti-mouse Ig; GAR, goat anti-rabbit Ig; HRP, horseradish peroxidase; IC, intracytoplasmic; ITAM, immunoreceptor tyrosine-based activation motif; ITIM, immunoreceptor tyrosine-based inhibition motif; 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-bearing protein tyrosine phosphatase; SHPS-1, SH2 domain-bearing phosphatase substrate 1; SIRPs, signal regulatory proteins; TNP, trinitrophenyl; MAP, mitogen-activated protein; DNP, 2,4-dinitrophenol; BSA, bovine serum albumin; mAb, monoclonal antibody; FITC, fluorescein isothiocyanate; PAGE, polyacrylamide gel electrophoresis; RT-PCR, reverse transcriptase-polymerase chain reaction; TNF, tumor necrosis factor; TM, transmembrane.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES