Src homology 2 domain-containing protein tyrosine phosphatase (SHP) substrate-1 (SHPS-1) is a transmembrane protein that is expressed predominantly in macrophages. Its extracellular region interacts with the transmembrane ligand CD47 expressed on the surface of adjacent cells, and its cytoplasmic region binds the protein tyrosine phosphatases SHP-1 and SHP-2. Phagocytosis of IgG- or complement-opsonized RBCs by peritoneal macrophages derived from mice that express a mutant SHPS-1 protein that lacks most of the cytoplasmic region was markedly enhanced compared with that apparent with wild-type macrophages. This effect was not observed either with CD47-deficient RBCs as the phagocytic target or in the presence of blocking Abs to SHPS-1. Depletion of SHPS-1 from wild-type macrophages by RNA interference also promoted FcγR-mediated phagocytosis of wild-type RBCs. Ligation of SHPS-1 on macrophages by CD47 on RBCs promoted tyrosine phosphorylation of SHPS-1 and its association with SHP-1, whereas tyrosine phosphorylation of SHPS-1 was markedly reduced in response to cross-linking of FcγRs. Treatment with inhibitors of PI3K or of Syk, but not with those of MEK or Src family kinases, abolished the enhancement of FcγR-mediated phagocytosis apparent in macrophages from SHPS-1 mutant mice. In contrast, FcγR-mediated tyrosine phosphorylation of Syk, Cbl, or the γ subunit of FcR was similar in macrophages from wild-type and SHPS-1 mutant mice. These results suggest that ligation of SHPS-1 on macrophages by CD47 promotes the tyrosine phosphorylation of SHPS-1 and thereby prevents the FcγR-mediated disruption of the SHPS-1-SHP-1 complex, resulting in inhibition of phagocytosis. The inhibition of phagocytosis by the SHPS-1-SHP-1 complex may be mediated at the level of Syk or PI3K signaling.
Macrophages are professional phagocytes that play an important role in innate and acquired immunity as a result of their ability to internalize and degrade pathogens (1, 2, 3). These cells also contribute to preservation of tissue integrity and function by engulfing apoptotic bodies and toxic particles (1, 2, 4). Phagocytosis occurs by extension of the plasma membrane around an extracellular object and subsequent internalization of the latter within a membrane-bounded intracellular phagosome (5). The mechanisms by which phagocytosis is triggered by the interaction of specific receptors on the macrophage surface, including FcγRs (for IgG-coated particles), complement receptors (CRs)3 (for C3bi-opsonized particles), scavenger receptors, and phosphatidylserine receptors (for apoptotic cells), with their ligands have been relatively well characterized (2, 4, 5, 6). In contrast, the molecular mechanisms for negative regulation of phagocytosis in macrophages remain largely unknown.
Src homology 2 domain-containing protein tyrosine phosphatase (SHP) substrate-1 (SHPS-1) (7, 8), also known as signal regulatory protein α (9), brain Ig-like molecule with tyrosine-based activation motifs (10), macrophage fusion receptor (11), and p84 neural adhesion molecule (12), is a member of the Ig superfamily of proteins. It was initially discovered as a tyrosine-phosphorylated transmembrane protein that binds and serves as a substrate for SHP-1 or SHP-2 (7, 13), both of which are SHPs. The putative extracellular region of SHPS-1 comprises three Ig-like domains with multiple N-linked glycosylation sites, whereas its cytoplasmic region contains four YXX(L/V/I) motifs, which are putative tyrosine phosphorylation sites and binding sites for the Src homology 2 domains of SHP-1 and SHP-2 (Fig. 1 A) (7, 9, 14). SHPS-1 is especially abundant in macrophages and neurons (15, 16, 17), although it is also expressed in other cell types, including fibroblasts (7).
CD47 has been implicated as a ligand for SHPS-1 (Fig. 1 A) (16, 18). This protein, which was originally identified in association with αvβ3 integrin (19), is also a member of the Ig superfamily, possessing an Ig-V-like extracellular domain, five putative membrane-spanning segments, and a short cytoplasmic tail (20). CD47 and SHPS-1 appear to constitute a cell-cell communication system (the CD47-SHPS-1 system) that plays important roles in hemopoietic cells and other cell types. Indeed, the rate of clearance of CD47-deficient RBCs from the bloodstream was found to be markedly increased compared with that apparent for wild-type cells (21, 22). Furthermore, the phagocytosis of CD47-deficient RBCs by splenic or bone marrow-derived macrophages was markedly enhanced in an in vitro assay (21, 22). The binding of CD47 on RBCs to SHPS-1 on macrophages is thus thought to inhibit the phagocytosis of the former cells by the latter. In addition, SHP-1, which forms a complex with SHPS-1, has been suggested to participate in such regulation, given that the phagocytic response is enhanced in motheaten variable mice, in which the activity of SHP-1 is reduced, compared with that apparent in wild-type mice (22). It remains unclear, however, whether SHPS-1 and SHP-1 indeed participate in such negative regulation of phagocytosis in macrophages. Moreover, the molecular mechanism whereby the binding of CD47 to SHPS-1 might inhibit phagocytosis by these cells is unknown.
We previously generated mice that express a mutant version of SHPS-1 that lacks most of the cytoplasmic region (Fig. 1 A) (23, 24). This mutant protein is not tyrosine phosphorylated, nor does it form a complex with SHP-1 or SHP-2. The mutant mice manifest thrombocytopenia as a result of a reduced circulation time of platelets in the bloodstream (24). Moreover, the preliminary observations indicated that peritoneal macrophages (PEMs) from the mutant mice phagocytose RBCs more effectively than do those from wild-type animals. With the use of PEMs derived from these SHPS-1 mutant mice, we have now investigated the role of SHPS-1, through its interaction with CD47, in the regulation of FcγR- or CR-mediated phagocytosis.
Materials and Methods
Abs and reagents
Rat (anti-p84) and hamster (FG2, 15D9) mAbs to SHPS-1 were kindly provided by C. Lagenaur (University of Pittsburgh, Pittsburgh, PA) and S. Nagata (Osaka University, Osaka, Japan), respectively. Rabbit polyclonal Abs (pAbs) to the γ subunit of FcR and to Gab-2 were kindly provided by T. Takai (Tohoku University, Miyagi, Japan) and T. Hirano (Osaka University, Osaka, Japan), respectively. Rabbit pAbs to SHPS-1 were obtained from ProSci. Rabbit pAbs to SHP-1, to Syk (N-19), and to Cbl; an HRP-conjugated mouse mAb (PY20) to phosphotyrosine; and a mouse mAb to SHP-2 were from Santa Cruz Biotechnology. A mouse mAb to phosphotyrosine (PY100) was from Cell Signaling Technology. Normal rat, mouse, and hamster IgG as well as HRP-conjugated secondary Abs were from Jackson ImmunoResearch Laboratories. Streptavidin-PE was from BD Pharmingen, and rabbit pAbs to mouse RBCs were from Cedarlane Laboratories. A rat IgM mAb to mouse CD24 (J11d) and a rat mAb to FcγRIII/II (2.4G2) were prepared from the culture supernatants of hybridoma cells. The anti-p84 mAb was conjugated to sulfosuccinimidyl-6-(biotinamido)hexanoate (sulfo-NHS-LC biotin) from Pierce. C5-deficient mouse serum was obtained from Charles River Laboratories Japan. Wortmannin, LY294002, and piceatannol were from Sigma-Aldrich; PD98059, PP2, and PP3 were from Calbiochem; and U0126 was from Promega. Glutaraldehyde-stabilized IgG-opsonized SRBCs were obtained from InterCell Technologies. A mouse SHPS-1-Fc fusion protein was prepared, as described previously (25).
The generation of mice that express a mutant version of SHPS-1 that lacks most of the cytoplasmic region was described previously (23). Mice were bred and maintained in the Institute of Experimental Animal Research of Gunma University under specific pathogen-free conditions. The mice were backcrossed onto the C57BL/6 background for five generations. Genotyping of mice was performed by PCR, as described (23). Mice were handled in accordance with the animal care guidelines of Gunma University.
Isolation of PEMs
PEMs were isolated essentially as described (26). In brief, the peritoneum of mice was flushed with ice-cold PBS 3 days after i.p. injection with 3 ml of 3% thioglycolate broth (Nissui Pharmaceutical). The exudate cells were separated by centrifugation at 400 × g for 5 min at 4°C, washed with ice-cold RPMI 1640 medium (Sigma-Aldrich), and then resuspended in RPMI 1640 supplemented with 10% FBS. The cells were transferred to culture dishes and were maintained at 37°C under a humidified atmosphere of 5% CO2 in air. After incubation for 24 h, nonadherent cells, which include neutrophils, B cells, and T cells, were washed away. The remaining adherent cells (∼99%) were identified as PEMs (26).
The surface expression of SHPS-1 on PEMs was examined by flow cytometry, as described (27). In brief, cells were detached from culture dishes by treatment with 0.01% EDTA and then washed with ice-cold PBS. They (0.5 × 106 to 1 × 106) were then incubated with rat mAb 2.4G2 to FcγR (1 μg/ml) for 30 min on ice, washed twice with ice-cold PBS, and incubated for 30 min on ice with a biotinylated mAb (anti-p84) to SHPS-1 (2 μg/ml). After two washes with ice-cold PBS, the cells were incubated with streptavidin-PE (1/2000 dilution), washed twice with PBS, suspended in 1 ml of PBS, and analyzed with a FACSCalibur flow cytometer (BD Biosciences). Data were processed with CellQuest software (BD Biosciences).
Opsonization of mouse RBCs
IgG-opsonized mouse RBCs were prepared by incubation of washed RBCs with rabbit polyclonal IgG to mouse RBCs (3 μg/ml) for 20 min at 37°C. Complement-opsonized mouse RBCs were prepared, as described (22). In brief, washed RBCs were incubated first for 30 min at 37°C with a subagglutination dilution of a rat IgM mAb to CD24 and then for 60 min at 37°C with 80% C5-deficient mouse serum. Opsonized cells were finally washed with and resuspended in Ringer’s solution (155 mM NaCl, 5 mM KCl, 1 mM MgCl2, 2 mM Na2HPO4, 10 mM glucose, 10 mM HEPES-NaOH (pH 7.2), 0.5 mg/ml BSA) before assays.
In vitro phagocytosis assays
Isolated PEMs were plated at a density of 1.25 × 105 cells/well in 24-well plates and cultured for 1–2 days. Opsonized or nonopsonized RBCs from wild-type or CD47-deficient mice, or IgG-opsonized SRBCs (InterCell Technologies) were then overlaid on the adherent cells at a density of 1.25 × 107 cells in 200 μl of Ringer’s solution per well. After incubation for the indicated times, the cells were washed with PBS and then incubated for 5 min at room temperature with hemolysis buffer (154 mM NH4Cl (pH 7.3), 10 mM KHCO3, 0.1 mM EDTA) to remove attached, but not phagocytosed, RBCs. After fixation of the cells with 4% paraformaldehyde in PBS, phagocytosis was detected by phase-contrast light microscopy and quantified as a phagocytic index (the number of engulfed RBCs per 100 macrophages). For treatment of PEMs with mAbs, the cells were incubated for 30 min at 4°C with mAbs (100 μg/ml) to SHPS-1 (anti-p84, FG2, or 15D9) or with corresponding control Abs before phagocytosis assays.
Immunoprecipitation and immunoblot analysis
PEMs cultured were incubated in serum-free medium for 12 h before stimulation. For FcγR cross-linking, cells (2 × 106) were placed on ice and incubated for 30 min with rat mAb 2.4G2 to FcγR (10 μg/ml). After washing with PBS, the cells were exposed to secondary cross-linking Abs (15 μg/ml) in PBS and immediately placed at 37°C. For stimulation with IgG-opsonized RBCs, PEMs (∼2 × 106) were incubated with the RBCs (2.0 × 108 cells in 200 μl) for the indicated times at 37°C. Stimulated cells were washed with ice-cold PBS and then lysed on ice in 400 μl of lysis buffer (20 mM Tris-HCl (pH 7.6), 140 mM NaCl, 1 mM EDTA, 1% Nonidet P-40) containing 1 mM PMSF, aprotinin (10 μg/ml), and 1 mM sodium vanadate. The lysates were centrifuged at 10,000 × g for 15 min at 4°C, and the resulting supernatants were subjected to immunoprecipitation and immunoblot analysis.
For immunoprecipitation, the supernatants were incubated for 3 h at 4°C with Ab-coupled protein G-Sepharose beads (20 μl of beads) (Amersham Biosciences). The beads were then washed three times with 1 ml of wash buffer (50 mM HEPES-NaOH (pH 7.6), 150 mM NaCl, 0.1% Triton X-100), suspended in Laemmli sample buffer, boiled for 5 min, and subjected to SDS-PAGE, followed by immunoblot analysis with various Abs and an ECL detection system (Amersham Biosciences).
A short interfering RNA (siRNA) specific for mouse SHPS-1 mRNA was designed according to the criteria of Elbashir et al. (28). Oligonucleotides corresponding to residues 105–123 (relative to the translation start codon) of mouse SHPS-1 cDNA (5′-GGUGACUCAGCCUGAGAAAdTdT-3′) and to the complementary sequence (5′-UUUCUCAGGCUGAGUCACCdTdT-3′) were synthesized (Nippon Bioservice) and annealed to form a short dsRNA with 3′-dithymidine overhangs. An irrelevant siRNA duplex with a random nucleotide sequence (Qiagen) was used as a control. RNA interference (RNAi) was performed, as described (28), with minor modifications. In brief, PEMs at 50% confluency (5.0 × 105 cells per 35-mm dish) were transfected for 24 h with 100 pmol siRNA with the use of OligoFectamine (Invitrogen Life Technologies) in complete medium without antibiotics. The cells were then transfected again for 24 h with another batch of siRNA before assays.
Data are presented as means ± SE. Statistical analysis was performed by Student’s t test with the use of Stat View 5.0 software (SAS Institute). A p value of <0.05 was considered statistically significant.
Expression of a mutant SHPS-1 protein in mouse PEMs
To investigate the role of SHPS-1 in regulation of phagocytosis in macrophages, we backcrossed mice of mixed C57BL/6 × 129Sv genetic background that were heterozygous for a mutant SHPS-1 allele (23) to the C57BL/6 background for five generations. Wild-type and homozygous mutant littermates were then used in the present study.
The mutant protein lacks most of the cytoplasmic region of SHPS-1, and its expression was first examined in lysates of PEMs by immunoblot analysis with either a mAb that recognizes the extracellular region of SHPS-1 (anti-p84) or pAbs that recognize the cytoplasmic region of the protein (anti-SHPS-1) (Fig. 1,A). Immunoblot analysis with anti-p84 revealed the presence of full-length SHPS-1 (∼120 kDa) in lysates of PEMs from wild-type mice, but only the mutant form of SHPS-1 (∼100 kDa) was detected in the lysates of PEMs from homozygous mutant animals; the abundance of the mutant protein in these latter cells was also markedly reduced compared with that of the full-length protein in wild-type PEMs (Fig. 1,B). In contrast, immunoblot analysis with the pAbs to SHPS-1 yielded no signal with PEMs from the homozygous mutant mice. The amounts of SHP-1 and SHP-2 in PEMs from the SHPS-1 mutant mice were similar to those in PEMs from wild-type animals (Fig. 1 B).
Consistent with these observations, flow cytometric analysis with anti-p84 revealed the presence of SHPS-1 immunoreactivity on the surface of PEMs from homozygous mutant mice, although its abundance was substantially reduced compared with that on the surface of wild-type cells (Fig. 1,C). Exposure of PEMs from wild-type mice to pervanadate induced a marked increase in the extent of tyrosine phosphorylation of SHPS-1 as well as in that of its binding to SHP-1 or SHP-2, whereas the mutant SHPS-1 neither underwent tyrosine phosphorylation nor associated with SHP-1 or SHP-2 in response to pervanadate (Fig. 1 D).
Enhanced phagocytosis of IgG- or C3bi-opsonized RBCs by PEMs from SHPS-1 mutant mice
We next examined the effect of the loss of wild-type SHPS-1 expression on phagocytosis by PEMs in an in vitro assay. Little phagocytosis of nonopsonized RBCs from wild-type donor mice was apparent with PEMs from either wild-type or SHPS-1 mutant mice (Fig. 2,A). In contrast, IgG-opsonized mouse RBCs underwent phagocytosis to a marked extent by PEMs from wild-type mice. Furthermore, the phagocytic response of PEMs from SHPS-1 mutant mice was even greater than that apparent with PEMs from wild-type animals. Similar results were obtained with C3bi-opsonized mouse RBCs (Fig. 2,B), although the extent of the phagocytic response in both wild-type and mutant PEMs was greatly reduced compared with that for FcγR-mediated phagocytosis (Fig. 2,A). We also examined the phagocytosis of IgG-opsonized SRBCs by PEMs. RBCs exhibit a high level of surface expression of CD47 (21, 29). Indeed, a mouse SHPS-1-Fc fusion protein bound to a marked extent to the surface of mouse RBCs. This fusion protein failed to bind to SRBCs, however (data not shown), suggesting that mouse SHPS-1 is not able to bind to CD47 expressed on the surface of SRBCs. The phagocytic response of mutant PEMs with IgG-opsonized SRBCs was still greater than that apparent for wild-type PEMs (Fig. 2,C), but the difference in these responses was greatly reduced compared with that observed with IgG-opsonized mouse RBCs (Fig. 2 A). These results suggested that SHPS-1 on the surface of PEMs negatively regulates FcγR- or CR-mediated phagocytosis of opsonized RBCs.
Effects of CD47 ablation in RBCs or of mAbs to SHPS-1 on FcγR-mediated phagocytosis by PEMs
The enhanced phagocytosis observed in PEMs from SHPS-1 mutant mice might have been attributable either to a lack of functional interaction of SHPS-1 (on PEMs) with CD47 (on RBCs) or to a loss of SHPS-1 function independent of CD47. To distinguish between these possibilities, we performed in vitro phagocytosis assays with RBCs from CD47-deficient mice (21, 30). Nonopsonized CD47−/− RBCs underwent only a low level of phagocytosis by PEMs from either wild-type or SHPS-1 mutant mice even after incubation for 120 min (Fig. 3,A). The phagocytosis of IgG-opsonized CD47−/− RBCs by PEMs from wild-type mice was markedly increased, however, compared with that of IgG-opsonized CD47+/+ RBCs (Fig. 3 B), consistent with previous observations with bone marrow-derived macrophages (22). In contrast, the phagocytosis of IgG-opsonized CD47−/− RBCs by PEMs from SHPS-1 mutant mice was similar to that of IgG-opsonized CD47+/+ RBCs by the mutant PEMs as well as to that of IgG-opsonized CD47−/− RBCs by wild-type PEMs. These results thus support the notion that the inhibition by SHPS-1 of phagocytosis by macrophages is almost entirely dependent on interaction of SHPS-1 on PEMs with CD47 on RBCs.
We also examined the effects of mAbs to SHPS-1 (anti-p84, FG2, 15D9) on FcγR-mediated phagocytosis of RBCs by PEMs. These mAbs block the interaction between CD47 and SHPS-1 (Fig. 1,A) (18, 31). Treatment of PEMs from either wild-type or SHPS-1 mutant mice with these mAbs did not affect the phagocytosis of nonopsonized mouse RBCs (Fig. 3 C). In contrast, treatment with either of the three mAbs to SHPS-1 markedly enhanced the phagocytosis of IgG-opsonized mouse RBCs by wild-type PEMs. This increased level of phagocytosis in wild-type PEMs was similar to that apparent in mutant PEMs treated with either the SHPS-1-specific mAbs or control IgG.
Effect of RNAi-mediated SHPS-1 depletion on FcγR-mediated phagocytosis in PEMs
It was possible that expression of the SHPS-1 mutant protein enhanced the phagocytic response of PEMs as a result of an unexpected gain of function rather than of a loss of function. To exclude this possibility, we examined the effect of SHPS-1 depletion by RNAi on FcγR-mediated phagocytosis. Transfection with an siRNA specific for SHPS-1 mRNA resulted in a marked decrease in the amount of SHPS-1 protein in PEMs from wild-type mice, without an effect on the abundance of SHP-1 or SHP-2 (Fig. 4,A). Transfection of wild-type PEMs with the SHPS-1 siRNA also promoted FcγR-mediated phagocytosis of RBCs (Fig. 4 B), suggesting that the observed phenotype of the SHPS-1 mutant cells was due to a loss of function of SHPS-1 and not to a gain of function of the mutant.
Tyrosine dephosphorylation of SHPS-1 in response to FcγR cross-linking and its prevention by engagement of SHPS-1 by CD47
Our results indicated that the interaction of CD47 on RBCs with SHPS-1 on PEMs regulates FcγR- or CR-meditated phagocytosis by PEMs in a negative manner. To investigate the molecular mechanism of such regulation, we first examined the effects of cross-linking of FcγRs on tyrosine phosphorylation of SHPS-1 and on its formation of a complex with SHP-1. Only a low level of tyrosine phosphorylation of SHPS-1 was apparent in immunoprecipitates prepared from serum-deprived wild-type PEMs with the mAb anti-p84 (Fig. 1,C). In contrast, tyrosine phosphorylation of a ∼120-kDa protein was prominent in immunoprecipitates prepared from serum-deprived PEMs with pAbs to SHP-1 (Fig. 5,A). Reprobing of the same blot with anti-p84 showed that this ∼120-kDa protein was indeed SHPS-1. Cross-linking of FcγRs with the mAb 2.4G2 in wild-type PEMs resulted in an initial small increase (apparent at 1 min) in the extent of tyrosine phosphorylation of SHPS-1 associated with SHP-1, but the amount of tyrosine-phosphorylated SHPS-1 associated with SHP-1 decreased thereafter (Fig. 5 A).
Incubation of wild-type PEMs with IgG-opsonized RBCs from wild-type mice resulted in a transient increase apparent between 5 and 20 min in the amount of tyrosine-phosphorylated SHPS-1 associated with SHP-1 (Fig. 5 B). In contrast, no such effect was apparent on incubation of wild-type PEMs with IgG-opsonized RBCs from CD47−/− mice. These results suggest that engagement of surface SHPS-1 on PEMs by CD47 on RBCs promotes the tyrosine phosphorylation of SHPS-1 and prevents the dephosphorylation of SHPS-1 induced by FcγR cross-linking.
Effects of inhibitors of Src family kinases, Syk, PI3K, or MEK on the enhancement of FcγR-mediated phagocytosis in PEMs from SHPS-1 mutant mice
The cross-linking of FcγRs induces tyrosine phosphorylation by a Src family kinase of the receptors themselves and of associated proteins that contain an ITAM (32, 33, 34). The phosphorylated ITAMs then serve as docking sites for the tyrosine kinase Syk. Downstream signaling mediated by PI3K, MAPK, and small GTPases of the Rho family eventually triggers phagocytosis of IgG-opsonized particles (32, 33, 34, 35, 36). We investigated the effect of an inhibitor of Src family kinases, PP2 (37), on the enhancement of FcγR-mediated phagocytosis apparent in PEMs from SHPS-1 mutant mice. Treatment with PP2 resulted in slight inhibition of FcγR-mediated phagocytosis in PEMs from either wild-type or SHPS-1 mutant mice, but it did not abolish the enhancement of this process induced by loss of SHPS-1 function (Fig. 6,A). The inactive analog PP3 had no effect on FcγR-mediated phagocytosis in wild-type or mutant PEMs. In contrast, the Syk inhibitor piceatannol (26, 38) inhibited FcγR-mediated phagocytosis in wild-type PEMs and abolished the enhancement of this process apparent in PEMs from SHPS-1 mutant mice (Fig. 6,B). The PI3K inhibitors wortmannin and LY294002 (39, 40) exerted effects similar to those of piceatannol (Fig. 6,C). Two different inhibitors of MAPK kinase MEK, PD98059 (41) and U0126 (42), slightly inhibited FcγR-mediated phagocytosis by PEMs from wild-type mice, but they had minimal effects on the enhancement of this process apparent in PEMs from SHPS-1 mutant mice (Fig. 6 D).
Tyrosine phosphorylation of signaling molecules downstream of FcγR in PEMs from SHPS-1 mutant mice
We further investigated the effect of loss of SHPS-1 function on signaling downstream of FcγR by examining the tyrosine phosphorylation of signaling molecules in PEMs from wild-type and SHPS-1 mutant mice. However, neither the tyrosine phosphorylation of the γ subunit of FcR nor that of Syk induced by FcγR cross-linking differed substantially between PEMs from wild-type mice and those from SHPS-1 mutant mice (Fig. 7, A and B). Tyrosine phosphorylation of Cbl is implicated in the FcγR-mediated activation of PI3K and phagocytosis (43). However, tyrosine phosphorylation of Cbl induced by FcγR cross-linking was similar in PEMs from wild-type mice and those from SHPS-1 mutant mice (Fig. 7 C).
We have shown that phagocytosis of IgG- or C3bi-opsonized RBCs by PEMs derived from mice expressing a mutant SHPS-1 protein that lacks most of the cytoplasmic region was markedly increased compared with that apparent with wild-type PEMs. Depletion of SHPS-1 from wild-type PEMs by RNAi also promoted FcγR-mediated phagocytosis. These data thus suggest that SHPS-1 negatively regulates FcγR- or CR-mediated phagocytosis by PEMs, and that the cytoplasmic region of SHPS-1 is required for such negative regulation. SHP-1 is a protein tyrosine phosphatase that negatively regulates multiple functions of hemopoietic cells, including macrophages (44, 45). SHPS-1 is implicated in the recruitment and activation of SHP-1 at sites near the plasma membrane in response to cytokines or integrin-mediated cell adhesion in macrophages (46, 47, 48). It is therefore likely that SHPS-1 negatively regulates phagocytosis in macrophages through its formation of a complex with SHP-1.
The rate of clearance of transfused CD47-deficient RBCs from the bloodstream is greater than that for wild-type cells (21, 22). In addition, the phagocytosis of CD47-deficient RBCs by splenic or bone marrow-derived macrophages was found to be markedly enhanced in an in vitro assay (21, 22). However, it was unclear whether such enhancement was mediated by loss of the interaction of CD47 on RBCs with SHPS-1 on macrophages. We have now shown that the enhancement of the phagocytic response in PEMs from SHPS-1 mutant mice was greatly reduced with IgG-opsonized SRBCs, which failed to bind mouse SHPS-1, compared with that observed with mouse RBCs. In addition, the difference in the phagocytic responses between PEMs from wild-type mice and those from SHPS-1 mutant animals was also abolished with CD47−/− mouse RBCs as the target. Furthermore, the phagocytic response of wild-type PEMs with wild-type mouse RBCs was potentiated by the presence of blocking mAbs to SHPS-1, whereas the difference in the responses between PEMs from wild-type mice and those from SHPS-1 mutant animals was no longer apparent in the presence of these Abs. We have also shown that transfused RBCs from wild-type donor mice were cleared more efficiently from the peripheral circulation in SHPS-1 mutant mice than in wild-type mice (T. Ishikawa, Y. Kaneko, and T. Matozaki, unpublished data). Together, these results thus suggest that negative regulation of phagocytosis by SHPS-1 is attributable predominantly to its interaction with CD47 on target RBCs.
We found that SHPS-1 is tyrosine phosphorylated and present in a complex with SHP-1 in serum-deprived wild-type PEMs. Integrin-mediated cell adhesion most likely contributes to this tyrosine phosphorylation of SHPS-1, as previously shown in fibroblasts and bone marrow macrophages (7, 46, 47, 49). In contrast, the extent of tyrosine phosphorylation of SHPS-1 was markedly reduced by cross-linking of FcγRs with the mAb 2.4G2. Dephosphorylation of SHPS-1 in response to FcγR cross-linking was previously observed in the macrophage cell line BAC1.2F5 (50). These observations suggest that ligation of FcγR on macrophages promotes phagocytosis not only through activation of downstream signaling molecules such as PI3K or small GTPases of the Rho family (5, 34, 51), but also through dephosphorylation of SHPS-1 and subsequent disruption of its interaction with SHP-1.
Incubation of PEMs from wild-type mice with IgG-opsonized wild-type RBCs increased the level of tyrosine phosphorylation of SHPS-1 and of its association with SHP-1. The dephosphorylation of SHPS-1 induced by ligation of FcγRs was thus prevented under these conditions. In contrast, incubation of PEMs with IgG-opsonized CD47−/− RBCs failed to induce such a response. Exposure of bone marrow-derived macrophages to wild-type (but not CD47-deficient) RBCs was previously shown to promote tyrosine phosphorylation of SHPS-1 in the presence of pervanadate (21). Our results therefore suggest that ligation of SHPS-1 (on macrophages) by CD47 (on RBCs) inhibits FcγR-mediated phagocytosis by promoting tyrosine phosphorylation of SHPS-1 and its interaction with SHP-1 and thereby preventing SHPS-1 dephosphorylation in response to FcγR activation. The mechanism responsible for the tyrosine phosphorylation of SHPS-1 induced by CD47 ligation is unknown. It is possible that its engagement by CD47 induces translocation of SHPS-1 to a cellular compartment that facilitates its interaction with tyrosine kinases (such as Src family kinases) responsible for its tyrosine phosphorylation (49).
We found that PEMs from either wild-type or SHPS-1 mutant mice phagocytosed nonopsonized RBCs at only a low level. The enhancement of phagocytosis in PEMs from SHPS-1 mutant mice was thus observed only with opsonized RBCs. Signaling downstream of the FcγR includes protein tyrosine phosphorylation (5, 6, 43). Ligation of SHPS-1 by CD47 may promote the recruitment and activation of SHP-1 at a site near the plasma membrane, where it has access to proteins that have undergone tyrosine phosphorylation in response to FcγR activation. An inhibitor of Src family kinases, PP2, did not affect the enhancement of FcγR-mediated phagocytosis apparent in PEMs from SHPS-1 mutant mice, however. In addition, the FcγR-induced tyrosine phosphorylation of the γ subunit of FcR, which is thought to be mediated by the Src family kinase Lyn (6, 52), was not enhanced in PEMs from SHPS-1 mutant mice. In contrast, the Syk inhibitor piceatannol abolished the enhancement of FcγR-mediated phagocytosis apparent in PEMs from SHPS-1 mutant mice. Although the FcγR-mediated tyrosine phosphorylation of Syk was not enhanced in these cells, it is possible that the tyrosine phosphorylation by Syk of adapter proteins, such as SH2 domain-containing leukocyte phosphoprotein of 76 kDa or B cell linker protein (26, 53), is regulated by the SHPS-1-SHP-1 complex.
Inhibitors of PI3K, but not those of MEK, also abolished the enhancement of FcγR-mediated phagocytosis in PEMs from SHPS-1 mutant mice. Potentiation of the FcγR-mediated activation of PI3K may thus contribute to the increased level of phagocytosis in PEMs from SHPS-1 mutant mice. In contrast, the FcγR-stimulated tyrosine phosphorylation of Cbl, which binds to the p85 subunit of PI3K in response to FcγR ligation (43), was not enhanced in PEMs from SHPS-1 mutant mice. In addition, the FcγR-mediated tyrosine phosphorylation of Gab-2, which also binds to the p85 subunit of PI3K in response to FcγR ligation (54, 55), was not increased in PEMs from the mutant mice (H. Okazawa, S. Motegi, N. Ohyama, and T. Matozaki, unpublished data). Together, these results thus suggest that the SHPS-1-SHP-1 complex might inhibit PI3K-mediated signaling downstream of the FcγR, presumably through tyrosine dephosphorylation of an unidentified protein (or proteins).
We thank S. Nagata, T. Hirano, C. F. Lagenaur, and T. Takai for reagents, as well as H. Kobayashi, K. Tomizawa, Y. Hayashi, and Y. Niwayama for technical assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by a Grant-in-Aid for Scientific Research on Priority Areas Cancer, a Grant-in-Aid for Scientific Research (B), a Grant-in-Aid for Young Scientists, and a 21st Century Center of Excellence Program grant from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, as well as by grants from the Swedish Research Council (06P-14098, 31X-14286), National Institutes of Health (GM57573-06), the Swedish Society of Medicine, the Faculty of Medicine at Umeå University, the Uehara Foundation, the Nakatomi Foundation, and the Kato Memorial Foundation.
Abbreviations used in this paper: CR, complement receptor; pAb, polyclonal Ab; PEM, peritoneal macrophage; RNAi, RNA interference; SHP, Src homology 2 domain-containing protein tyrosine phosphatase; SHPS-1, SHP substrate-1; siRNA, short interfering RNA.