Abstract
Phagocytosis mediated by FcγR plays an important role in host defense. The molecular events involved in this process have not been completely defined. The adapter protein Cbl has been implicated in FcγR signaling, but the function of Cbl in phagocytosis is unknown. Here we show that overexpression of the transforming mutants of Cbl, Cbl-70Z, and v-Cbl, but not wild-type (wt) Cbl, enhance phagocytosis mediated by FcγR in COS cells. Cbl-70Z, but not Cbl-wt, also enhanced FcγR-mediated phagocytosis in P388D1 murine macrophage cells. Cbl-70Z did not affect tyrosine phosphorylation or in vitro kinase activity of Syk, indicating that Syk may not be the direct target of Cbl-70Z in the enhancement of phagocytosis. A point mutation (G306E) in the phosphotyrosine domain of Cbl-70Z, as well as a C-terminal 67-aa deletion, partially abolished the enhancing effect on FcγR-mediated phagocytosis. A double mutant of Cbl-70Z containing both the G306E mutation and the C-terminal deletion completely lacked the ability to enhance phagocytosis. Thus, both the phosphotyrosine binding domain and the carboxyl-terminal tail were required for optimal enhancement of phagocytosis by Cbl-70Z. Functional phosphatidylinositol 3-kinase was required for Cbl-70Z to enhance phagocytosis, since wortmannin, a phosphatidylinositol 3-kinase inhibitor, inhibited FcγR-mediated phagocytosis in the presence of Cbl-70Z. These studies demonstrate that mutants of Cbl can modulate the phagocytic pathway mediated by FcγR and imply a functional involvement of c-Cbl in Fcγ receptor-mediated phagocytosis.
FcγR, which bind the constant region (Fc) of IgG molecules, play an important role in immune reactions (1, 2). One of their functions in cells of the monocyte/macrophage lineage is the induction of phagocytosis of foreign substances coated with IgG. The pathway(s) through which FcγR mediate phagocytosis are not completely defined. Recent evidence indicates that Syk protein tyrosine kinase is essential for phagocytosis by FcγR. We have observed that overexpression of Syk kinase enhances phagocytosis by FcγRI and FcγRIIIA in COS cells and that the inhibition of Syk kinase by antisense oligodeoxynucleotides abrogates FcγR phagocytosis of Ab-coated erythrocytes (EA)3 by monocytes/macrophages (3, 4, 5). Furthermore, studies using macrophages from Syk-deficient mice confirm that Syk is essential for FcγR-mediated phagocytosis (6, 7). The importance of Syk in phagocytosis and its role as a protein kinase prompted us to hypothesize that some of the substrates that are phosphorylated by Syk may also play a role in phagocytosis by FcγR.
There is evidence that the adapter protein Cbl, the protein product of the protooncogene c-cbl, is a substrate for the Syk family protein tyrosine kinases (8, 9). The c-cbl gene was first described as a cellular counterpart of a transforming gene, v-cbl, of a murine Cas NS-1 retrovirus (10). c-Cbl has been shown to be tyrosine phosphorylated following the cross-linking of several cell surface receptors including the platelet-derived growth factor receptor, the TCR, the B cell receptors, and the FcR (9, 11, 12, 13). Recent evidence suggests that c-Cbl may be a regulator of protein tyrosine kinase(s) downstream of these receptors. For example, SLI-1, a Caenorhabditis elegans Cbl homologue, functions as a negative regulator of the LET-23 receptor protein tyrosine kinase (14, 15). The demonstration (16) that c-Cbl negatively regulates Syk kinase in signaling by FcεRI in mast cells supports this thesis. Also relevant is evidence that Cbl binds to ZAP-70, another member of the Syk family of tyrosine kinases (17, 18, 19). However, functional consequences of the binding of Cbl to ZAP-70 have not been reported. Interestingly, transforming mutants of Cbl, the 70Z/3 form of Cbl (Cbl-70Z) and the N-terminal domain of Cbl (v-Cbl form), induce tyrosine phosphorylation in NIH-3T3 cells (19), suggesting that c-Cbl and its transforming mutants can have opposing effects on protein tyrosine kinase pathways. For FcγR, receptor cross-linking induces tyrosine phosphorylation of c-Cbl (9, 13) as well as the association of c-Cbl with phosphatidylinositol 3-kinase (PI-3K), Grb2, Shc, Syk, and Lyn (9, 12, 20, 21). However, the role of Cbl in FcγR-mediated phagocytosis has not been studied.
To define the functional role of Cbl, we examined the effects of Cbl and its mutant forms on phagocytosis mediated by FcγR in transfected COS-7 cells. We observed that transforming mutants of Cbl, but not wild-type Cbl (Cbl-wt), enhance phagocytosis by FcγR in COS cells. For optimal enhancement of phagocytosis by Cbl-70Z, both the N-terminal phosphotyrosine binding (PTB) domain and the C-terminal 67-aa stretch were important. The enhancing effect of Cbl-70Z was dependent on an intact PI-3K pathway. Our study demonstrates the modulation of FcγR function by Cbl mutants and suggests the involvement of Cbl in the phagocytic pathway(s) mediated by FcγR.
Materials and Methods
Cell culture and transfection
COS-7 cells were obtained from American Tissue Culture Collection (Manassas, VA). Cells were cultured in DMEM (Life Technologies, Gaithersburg, MD) supplemented with l-glutamine (3 mM), penicillin (100 U/ml), streptomycin (100 μg/ml), and 10% FCS at 37°C with 5% CO2. Transient transfections were conducted using the DEAE-dextran method as described previously (4). Briefly, COS-7 cells were seeded to tissue culture plates the day before transfection. Cells were washed once with DMEM, then incubated with 4 μg/ml of plasmid DNA in DMEM containing 10% FCS, 1 mg/ml of DEAE-dextran, and 100 μM chloroquine for 4 h at 37°C. Following osmotic shock with 10% DMSO (Sigma, St. Louis, MO) in PBS for 90 s, cells were washed twice with DMEM, then cultured in fresh medium. Cells were subjected to analysis 48 h after transfection.
The murine macrophage cell line P388D1 was maintained in DMEM supplemented with l-glutamine (3 mM), penicillin (100 U/ml), streptomycin (100 μg/ml), 2-ME (50 μM), and 10% FCS. P388D1 cells were transfected with the aid of FuGENE 6 transfection reagent (Roche Molecular Biochemicals, Indianapolis, IN). Bulk transfectants were subjected to analysis following a selection with 800 μg/ml of G418 (Mediatech, Herndon, VA) for about 3 wk.
Preparation of Ab-coated EA and determination of phagocytosis
EA were prepared as previously described (3, 4, 5). Sheep RBC (Rockland, Gilbertsville, PA) were suspended in calcium- and magnesium-free PBS at 109/ml, then incubated at 37°C for 30 min after the addition of an equal volume of PBS containing the highest subagglutinating titer of rabbit anti-sheep RBC Ab (Cappel Laboratories, Cochranville, PA). COS-7 cells were incubated with 108 EA for 30 min at 37°C. For P388D1 cells, incubation was conducted in DMEM for 1 h. After vigorous washing to remove unbound EA, surface-bound EA were lysed by 30-s exposure to hypotonic buffer. Cells were stained with Wright-Giemsa and examined for internalized RBC by light microscopy. Results were expressed as a phagocytic index, the number of ingested erythrocytes per 100 cells. Statistical analysis used the Newman-Keuls multiple comparison test. A value of p < 0.05 was considered significant.
Flow cytometry
Cells were gently removed from the tissue culture plates and incubated with anti-FcγRI mAb 32.2, anti-FcγRII mAb IV.3, or anti-FcγRIII mAb 3G8 for 30 min at 4°C. After two washes, cells were incubated with FITC-conjugated goat anti-mouse F(ab′)2 IgG (TAGO, Burlingame, CA) for 30 min at 4°C, then washed twice. Isotype controls were used for all reactions. For the analysis of intracellular protein, cells were fixed with 1% paraformaldehyde in PBS and permeabilized in 0.1% Triton X-100 and 1% paraformaldehyde in PBS at 4°C for 30 min before staining with Ab. Fluorescence was measured on a FACStar or a FACScan cytometer (Becton Dickinson, Mountain View, CA).
Construction of recombinant plasmids
Expression vectors for FcγRI, FcγRIIA, FcγRIIIAα, and the FcRγ subunit were described previously (22, 23, 24). The cDNA for the single chain chimeric Fcγ receptor, FcγRI-γ-γ, was constructed by PCR. FcγRI-γ-γ is comprised of the extracellular domain of FcγRI (residues 1–292) and the transmembrane and cytoplasmic domains of the human FcRγ subunit (residues 24–86). The cDNA for Syk kinase was a gift from Dr. C. Couture (McGill University, Montreal, Canada). The cDNA for Cbl-wt and Cbl-70Z with a N-terminal hemagglutinin (HA) epitope were provided by Dr. L. E. Samelson (National Institutes of Health, Bethesda, MD) and were in the pSXSRα expression vector (25). The cDNA for v-Cbl and the G306E mutants of c-Cbl, Cbl-70Z, and v-Cbl were prepared by PCR and cloned into pSXSRα. C-terminal truncation mutants of Cbl-70Z were constructed using internal restriction enzyme sites. All mutants of Cbl contained an N-terminal HA epitope. For expression in P388D1 cells, c-Cbl and Cbl-70Z were subcloned into LK440 (26) along with the internal ribosomal entry site and the neomycin phosphotransferase gene derived from pIRESneo (Clontech, Palo Alto, CA). All constructs were confirmed by DNA sequencing.
Antibodies
Anti-Syk rabbit antiserum was provided by Drs. D. H. Chu and A. Weiss (University of California, San Francisco, CA). Anti-Syk mAb 4D10 (IgG2a) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-influenza HA epitope tag mAb 3F10 (rat IgG1) was obtained from Roche Molecular Biochemicals. Anti-phosphotyrosine mAb 4G10 (mouse IgG2a) was obtained from Upstate Biotechnology (Lake Placid, NY). Anti-phosphotyrosine polyclonal Ab was purchased from PharMingen (San Diego, CA).
Cell lysis and immunoprecipitation
Cells were washed twice with cold PBS and solubilized with buffer (25 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% Brij-96, 1 mM EGTA, 1 mM sodium orthovanadate, 10 μg/ml of aprotinin, 20 μg/ml of leupeptin, and 1 mM PMSF) on ice for 30 min. After centrifugation at 13,000 × g for 15 min at 4°C, clarified lysates were precleared with either Pansorbin (Calbiochem, La Jolla, CA) or protein G-agarose (Santa Cruz Biotechnology) for 1 h at 4°C. For immunoprecipitation, cell lysates were incubated with appropriate Abs for 2 h at 4°C. Immune complexes were captured with protein A-agarose or protein G-agarose. The immunoprecipitates were washed three times with PBS containing 1 mM sodium orthovanadate and 1 mM EDTA and used for either immunoblot or in vitro immune complex kinase assay.
Immunoblot
Immunoprecipitated proteins were eluted into Laemmli sample buffer with 100 mM DTT, boiled, separated on 7.5% SDS-PAGE gels, and electrophoretically transferred to polyvinylidene difluoride membranes (Immobilon P, Millipore, Bedford, MA). HRP-conjugated goat anti-mouse IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) or HRP-conjugated anti-rat IgG (Santa Cruz Biotechnology) was used as a secondary reagent. Detection by enhanced chemiluminescence was performed according to the supplier’s recommendation (ECL, Amersham, Aylesbury, U.K.). Densitometric quantitation was performed using Personal Densitometer SI and ImageQuant (Molecular Dynamics, Sunnyvale, CA).
Immune complex kinase assay
The in vitro immune complex kinase assay was performed as described previously with slight modification (27). Briefly, immunoprecipitates were additionally washed once with low salt buffer containing 100 mM NaCl, 25 mM HEPES NaOH, and 5 mM MnCl2 (pH 7.4), then incubated for 10 min at room temperature in 30 μl of kinase buffer containing 25 mM HEPES NaOH, 5 mM MnCl2, 5 mM p-nitrophenylphosphate, 1 μM ATP (Roche Molecular Biochemicals), and 5 μCi [γ-32P]ATP (3000 Ci/mmol; DuPont-NEN, Boston, MA) with 6 μg of-GST-band 3 as an exogenous substrate (28). Reactions were stopped by the addition of sample buffer and heating to 100°C for 5 min. Samples were analyzed by 12% SDS-PAGE. Gels were dried and subjected to autoradiography.
Results
Cbl-70Z, but not Cbl-wt, enhances phagocytosis by FcγR in COS cells
The Fc receptors FcγRI and FcγRIIIA require coexpression of the FcRγ subunit for phagocytic signaling (24, 29). To determine whether Cbl is involved in phagocytosis by FcγRI/γ and FcγRIIIA/γ, we cotransfected Cbl-wt or Cbl-70Z with the FcRγ subunit and FcγRI or FcγRIIIA in COS cells and examined their effects on phagocytosis. Cbl-70Z, which was cloned from the 70Z/3 pre-B lymphoma cell line, is a transforming mutant of Cbl in which the internal 17 aa (residues 366–382) are deleted (30) (see Fig. 1⇓). Coexpression of Cbl-70Z enhanced phagocytosis by FcγRI/γ 2-fold and that by FcγRIIIA/γ 3-fold (Table I⇓). This enhancement of phagocytosis by Cbl-70Z was similar to that induced by overexpression of Syk, which was used as a positive control. In contrast, the overexpression of Cbl-wt did not affect phagocytosis by these receptors.
Schematic illustration of the constructs of Cbl and its mutants. All the constructs contain the N-terminal HA epitope tag. The proposed functional domains shown are the N-terminal PTB domain, the ring finger domain, the proline-rich domain (PRO), and the C-terminal leucine zipper-like domain (LZ). Cbl-70Z has the internal 17-aa deletion near the ring finger domain as indicated. The v-Cbl construct consists of the N-terminal 357 residues of human c-Cbl, which corresponds to the sequence present in viral Gag-v-Cbl fusion protein.
Effect of Cbl on Fcγ receptor-mediated phagocytosisa
Phagocytosis by FcγRI-γ-γ, a chimeric Fcγ receptor comprised of the extracellular domain of FcγRI and the transmembrane and cytoplasmic domains of the FcRγ subunit, was also enhanced by the overexpression of Cbl-70Z, but not by Cbl-wt (Table I⇑). Subsequently, we employed the chimeric Fcγ receptor, FcγRI-γ-γ, in our experimental model for phagocytosis mediated by FcγRI/γ.
Unlike FcγRI and FcγRIIIA, FcγRIIA does not require coexpression of the FcRγ subunit for transmission of a phagocytic signal (22). Phagocytosis by FcγRIIA, which mediates phagocytosis more efficiently in COS cells than FcγRI/γ or FcγRIIIA/γ, was also enhanced by the expression of Cbl-70Z, but the enhancement was not as great (30% increase) as that observed for FcγRI/γ, FcγRIIIA/γ, or FcγRI-γ-γ.
Expression of Cbl-70Z enhances phagocytosis in P388D1 macrophages
To examine whether Cbl and its mutants have any effect on FcγR-mediated phagocytosis in professional phagocytes, we established P388D1 murine macrophage cells that express HA epitope-tagged Cbl-wt or Cbl-70Z (Fig. 2⇓A). Flow cytometric analysis of permeabilized cells stained with anti-HA mAb indicated that the expression of the protein is homogeneous, although bulk transfectants are analyzed without subcloning (Fig. 2⇓B). Two different batches of bulk transfectants expressing either Cbl-wt or Cbl-70Z and two control batches transfected with empty vector were analyzed for their ability to phagocytose EA. As shown in Fig. 2⇓C, P388D1 cells expressing Cbl-70Z had a 2- to 3-fold higher phagocytic index than control cells transfected with empty vector. The overexpression of Cbl-wt, on the other hand, did not affect phagocytosis of EA by P388D1 macrophages.
Establishment of P388D1 cells expressing Cbl-wt or Cbl-70Z and their ability to phagocytose EA. A, Two different batches of P388D1 cells transfected with empty vector (vect 4.2 and vect 4.3), Cbl-wt (Cbl-811 and Cbl-822), or Cbl-70Z (70Z-811 and 70Z-821) were analyzed for expression by immunoblot with anti-HA mAb as indicated. Equal amounts of protein were loaded (5 μg/lane). B, After fixation and permeabilization as described in Materials and Methods, cells were incubated with 1 μg/ml of anti-HA mAb, washed, and incubated with FITC-labeled anti-rat Ab. Fluorescence was measured by a FACScan flow cytometer (Becton Dickinson). C, Cells were examined for their ability to phagocytose EA. Phagocytic indices were determined as described in Materials and Methods. The mean and SD of four separate experiments are shown. Asterisks indicate statistical differences from vect-4.2 (p < 0.05). Daggers indicate statistical differences from vect-4.3 (p < 0.05).
Activity of Syk is not altered by Cbl-70Z
The observation that Syk kinase is an essential mediator of phagocytosis by Fcγ receptors (5, 6, 7) and that Cbl-wt binds to and negatively regulates Syk in FcεRI signaling in rat mast cells (16) led us to investigate whether Cbl interacts with Syk in FcγR phagocytic signaling. In our study of phagocytosis, it was Cbl-70Z that had a profound effect on this biological response, while in FcεRI signaling, Cbl-wt was effective. In mast cells, in contrast to Cbl-wt, Cbl-70Z was shown to bind to, but not to inhibit, Syk (16). We first speculated that COS cells might express higher amounts of Cbl or Cbl-like inhibitory molecules that constitutively inhibit Syk, and that Cbl-70Z, which lacked the inhibitory action, might activate, or disinhibit, Syk by displacing such inhibitors. If that was the case, we might expect the activation of Syk with overexpression of Cbl-70Z. To examine this possibility, we analyzed the tyrosine phosphorylation of Syk, which reflects Syk activation, and its kinase activity in vitro.
We first analyzed the small amount of Syk endogenously expressed in COS cells. As shown in Fig. 3⇓A, cross-linking of FcγRI-γ-γ induced tyrosine phosphorylation of endogenous Syk in COS cells. However, there was little difference in Syk tyrosine phosphorylation among cells transfected with an empty vector, Cbl-wt or Cbl-70Z. Densitometric analysis showed that the extent of tyrosine phosphorylation was as follows when normalized to the amount of Syk detected by reblotting; vector (EA−), 1.0; vector (EA+), 5.0; Cbl-wt (EA−), 0.4; Cbl-wt (EA+), 3.7; Cbl-70Z (EA−) 1.2; and Cbl-70Z (EA+), 3.9 (mean of three independent immunoblots, relative values with vector (EA−) taken as 1). The kinase activity of Syk, examined by in vitro kinase assay of anti-Syk immunoprecipitates, was also unaffected by the overexpression of Cbl-wt or Cbl-70Z (Fig. 3⇓A, bottom). We also analyzed tyrosine phosphorylation and in vitro kinase activity of Syk in COS cells transfected with FcγR, Syk, and Cbl (Fig. 3⇓B). Expression of Cbl-70Z also failed to modify both tyrosine phosphorylation and in vitro kinase activity of Syk.
Tyrosine phosphorylation and in vitro kinase activity of Syk in COS cells transfected with FcγRI-γ-γ and Cbl. COS cells (6 × 106) were transfected with FcγRI-γ-γ and Cbl-wt or Cbl-70Z as indicated (A). Cells (2 × 106) were transfected with FcγRI-γ-γ and Syk along with Cbl-wt or Cbl-70Z (B). Two days after transfection, cells were incubated with (+) or without (−) EA at 37°C for 30 min to cross-link FcγR. The extracellularly attached erythrocytes were lysed by hypotonic shock. Cells were lysed with 1% Brij-96 as described in Materials and Methods. The lysates were immunoprecipitated with anti-Syk polyclonal Ab. Each immunoprecipitate was split into two samples. One was directly subjected to 7.5% SDS-PAGE, and immunoblot analysis was performed with anti-phosphotyrosine Ab 4G10 (top), then with anti-Syk mAb 4D10 (middle) after stripping. The second sample was analyzed by in vitro immune complex kinase assay with GST-band 3 as a substrate (bottom).
We also examined by functional assay the possible interaction of Cbl and Syk in the phagocytic pathway. We overexpressed both Cbl and Syk with FcγRI-γ-γ in COS cells and determined the efficiency of phagocytosis. Cbl-70Z in combination with Syk enhanced phagocytosis further (Fig. 4⇓). The effects of overexpression of Syk and Cbl-70Z on phagocytosis were largely additive. In contrast, Cbl-wt did not affect phagocytosis by FcγRI-γ-γ even when Syk was overexpressed. These observations suggest that activation of Syk is probably not the mechanism by which Cbl-70Z enhances phagocytosis by Fcγ receptors.
Effect of coexpression of Syk and Cbl on phagocytosis mediated by FcγRI-γ-γ. COS cells were cotransfected with FcγRI-γ-γ, Syk, and Cbl-wt or Cbl-70Z as indicated. An empty vector (vector) was used as a control to adjust DNA loading between each treatment. Phagocytic indices were determined as described in Table I⇑. The mean and SD of three separate experiments are shown. Asterisks indicate statistical differences from vector control (p < 0.05). Daggers indicate statistical differences from Syk plus Cbl-70Z (p < 0.05).
v-Cbl enhances phagocytosis by FcγRI-γ-γ
To further characterize the enhancement of phagocytosis by Cbl mutants, we examined the effect of v-Cbl on phagocytosis. v-Cbl consists of the N-terminal residues of c-Cbl (1–357 in human) corresponding to the sequence present in viral gag-v-Cbl fusion protein (see Fig. 1⇑) and possesses transforming activity in vitro (31, 32). As shown in Fig. 5⇓, overexpression of v-Cbl also enhanced phagocytosis by the chimeric FcγR FcγRI-γ-γ 2-fold compared with that of control cells transfected with FcγR alone. The phagocytic index of the cells transfected with v-Cbl was reproducibly less than cells transfected with Cbl-70Z (Fig. 5⇓), although the expression of v-Cbl, as determined by immunoblot with anti-HA Ab, was equal to or greater than that of Cbl-70Z (data not shown).
The effects of v-Cbl and the G306E mutants of Cbl on phagocytosis mediated by FcγRI-γ-γ. COS cells were transfected with an empty vector (control), Cbl-wt, Cbl/G306E, Cbl-70Z, Cbl-70Z/G306E, v-Cbl, or v-Cbl/G306E along with FcγRI-γ-γ as indicated. Phagocytic indexes were determined as described in Table I⇑. The mean and SD of three separate experiments are shown. Asterisks indicate statistical differences from vector control (p < 0.05). Daggers indicate statistical differences from the corresponding G306E mutant (p < 0.05). The difference between Cbl-70Z and v-Cbl was also statistically significant (p < 0.05).
The G306E mutation of transforming mutants of Cbl partially abolishes the ability to enhance phagocytosis
A point mutation (G306E) of Cbl, which corresponds to a loss of function mutation of the C. elegans Cbl homologue SLI-1, abrogates the binding of Cbl to ZAP-70 as well as the transforming ability of v-cbl (18, 31). It is thus proposed that the N-terminal region of Cbl contains a PTB domain. Our observation that v-Cbl as well as Cbl-70Z enhance phagocytosis suggested that the PTB domain might be essential for the effect on FcγR-mediated phagocytosis. We therefore examined the impact of the G306E mutation on enhancement of phagocytosis by Cbl-70Z and v-Cbl. As shown in Fig. 5⇑, the ability of Cbl-70Z and v-Cbl to enhance phagocytosis was markedly decreased by the G306E mutation. Enhancement of phagocytosis by Cbl-70Z and v-Cbl thus appears to be dependent on a functional PTB domain. It should be noted, however, that the G306E mutant of Cbl-70Z was not completely inactive in enhancing phagocytosis. Cbl-70Z/G306E still enhanced phagocytosis 2-fold compared to the control, which was transfected with FcγRI-γ-γ and an empty vector. Taken together with the observation that v-Cbl is less active than Cbl-70Z in enhancing phagocytosis, the ability of Cbl-70Z/G306E to enhance phagocytosis suggests that the C-terminal region of Cbl-70Z is also necessary for optimal enhancement of phagocytosis by Cbl-70Z.
The C-terminal deletion of Cbl-70Z results in reduced ability to enhance phagocytosis
The C-terminal region of Cbl, which is absent in v-Cbl, contains several functional motifs, including the ring finger, proline-rich, and leucine zipper-like domains (30). This C-terminal region also contains the binding sites for Vav, the p85 subunit of PI-3K, and for Nck and Crk (33, 34, 35, 36, 37, 38, 39). To determine the region of Cbl-70Z that is required for optimal enhancement of phagocytosis, we constructed several C-terminal deletion mutants of Cbl-70Z and studied their effects on phagocytosis. As shown in Fig. 6⇓A, deletion of the C-terminal 67 aa (Cbl-70Z/839) resulted in ∼50% decrease in the ability to enhance phagocytosis compared with that of full-length Cbl-70Z. Further deletions (Cbl-70Z/730, Cbl-70Z/541, and Cbl-70Z/480) did not further decrease the enhancing effect on phagocytosis.
Effects of the C-terminal truncation and both the C-terminal truncation and the G306E mutation on the ability of Cbl-70Z to enhance phagocytosis by FcγRI-γ-γ. A, COS cells were transfected with an empty vector (control), full-length Cbl-70Z, C-terminal truncation mutants (70Z/839, 70Z/730, 70Z/541, 70Z/481), or v-Cbl along with FcγRI-γ-γ. B, COS cells were transfected with Cbl-70Z, Cbl-70Z/G306E, 70Z/839, or a double mutant Cbl-70Z/G306E/839 along with FcγRI-γ-γ. Phagocytic indices were determined as described in Table I⇑. The mean and SD of four separate experiments are shown. Asterisks indicate statistical differences from vector control (p < 0.05). Daggers indicate statistical differences from Cbl-70Z (p < 0.05).
These results indicate that the C-terminal 67 aa as well as the PTB domain play an important role in the enhancement of phagocytosis by Cbl-70Z. To further verify these results, we constructed a double mutant of Cbl-70Z containing both the G306E mutation and the C-terminal 67-aa deletion and tested its effect on phagocytosis. As expected, this double mutant completely lacked the ability to enhance phagocytosis by FcγRI-γ-γ in COS cells (Fig. 6⇑B). Thus, the data indicate that both the functional PTB domain and the 67 aa of the C-terminal tail of Cbl-70Z are required for optimal enhancement of phagocytosis by FcγRI-γ-γ in COS cells.
Cbl-70Z, but not Cbl-wt, is constitutively tyrosine phosphorylated in COS-7 cells
In normal cells, c-Cbl becomes phosphorylated on tyrosine following cellular stimulation. In transfected NIH-3T3 cells, Cbl-70Z is constitutively hyperphosphorylated, whereas v-Cbl is reported to lack any major tyrosine phosphorylation sites (40). To examine whether tyrosine phosphorylation of Cbl is involved in its ability to enhance phagocytosis, we examined the tyrosine phosphorylation state of Cbl in COS cells. Immunoprecipitation with anti-HA Ab followed by immunoblot with anti-phosphotyrosine Ab demonstrated that Cbl-70Z was constitutively hyperphosphorylated in COS cells compared with Cbl-wt (Fig. 7⇓). We could not detect tyrosine phosphorylation of v-Cbl with or without receptor cross-linking (data not shown).
Tyrosine phosphorylation of transfected Cbl. COS cells (2 × 106) were transfected with FcγRI-γ-γ and Cbl-wt, Cbl-70Z, or mutants of Cbl-70Z as indicated. Two days after transfection, cells were incubated with (+) or without (−) EA at 37°C for 30 min to cross-link FcγR. Cells were washed, and the extracellularly attached erythrocytes were lysed by hypotonic shock. Cells were lysed with 1% Brij-96 as described in Materials and Methods. The lysates were immunoprecipitated with anti-HA mAb 3F10. Immunoprecipitates were separated by 7.5% SDS-PAGE and analyzed by immunoblot with anti-phosphotyrosine mAb 4G10 (top), then with anti-HA Ab (bottom) after stripping.
We also examined the tyrosine phosphorylation of Cbl-70Z mutants Cbl-70Z/G306E and Cbl-70Z/839, which are partially active in the enhancement of phagocytosis, and the double mutant Cbl-70Z/G306E/839, which is completely inactive in phagocytosis. As shown in Fig. 7⇑, these mutants of Cbl-70Z were tyrosine phosphorylated to a lesser extent than Cbl-70Z.
The PI-3K inhibitor, wortmannin, inhibits enhancement of phagocytosis by Cbl-70Z and v-Cbl
Previous studies have shown that the inhibitor of PI-3K, wortmannin, inhibits phagocytosis by FcγR (41, 42, 43), implying that PI-3K is essential for phagocytosis. We examined whether wortmannin inhibits the enhancement of FcγR-mediated phagocytosis by Cbl-70Z or v-Cbl. As shown in Fig. 8⇓, phagocytosis was inhibited by wortmannin in the presence of Cbl-70Z, v-Cbl, or Syk.
Inhibitory effect of wortmannin on phagocytosis mediated by FcγRI-γ-γ in the presence of Cbl-70Z, v-Cbl, or Syk. COS cells were transfected with FcγRI-γ-γ and an empty vector (control), Cbl-70Z, v-Cbl, or Syk as indicated. Two days after transfection, cells were incubated with the indicated concentration of wortmannin in PBS for 15 min at 37°C. Cells were then further incubated with 108 EA for 30 min. The phagocytic indices were normalized with the phagocytic indices of samples without wortmannin being taken as 100%. The actual phagocytic indices without wortmannin were 27 (control), 81 (Cbl-70Z), 54 (v-Cbl), and 86 (Syk). Means of three separate measurements are shown.
To examine the potential interaction of Cbl-70Z with PI-3K, we conducted a series of immunoprecipitation studies. Cbl-wt or Cbl-70Z was overexpressed in COS cells with FcγRI-γ-γ, which was either cross-linked with EA or left unstimulated. Immunoblotting with anti-p85 PI-3K showed that little of the p85 subunit of PI-3K coimmunoprecipitated with Cbl-wt or Cbl-70Z regardless of cross-linking of FcγRI-γ-γ (data not shown). In addition, immunoprecipitation with anti-p85 did not specifically coimmunoprecipitate Cbl-wt or Cbl-70Z. The inhibition of phagocytosis by wortmannin suggests that intact PI-3K activity is required for the enhancement of phagocytosis by Cbl-70Z and v-Cbl. However, a direct interaction of Cbl and PI-3K may not be necessary for Cbl-70Z to enhance phagocytosis by Fcγ receptors.
Discussion
In this study we have demonstrated that the transforming mutants of Cbl, Cbl-70Z and v-Cbl, enhance phagocytosis in COS-7 cells heterologously expressing FcγR. Expression of Cbl-70Z was also shown to enhance FcγR-dependent phagocytosis in the P388D1 macrophage cell line. The enhancing effect of Cbl-70Z on phagocytosis was mapped to both its N-terminal PTB domain and the C-terminal 67-aa residues. The effects of Cbl-70Z and v-Cbl appear to require an intact PI-3K activity. However, Syk kinase may not be the direct target of the action of the Cbl mutants.
Overexpression of Cbl-70Z in COS cells enhanced the phagocytosis mediated by FcγRI plus the FcRγ subunit, FcγRIIIA plus the FcRγ subunit, and FcγRIIA (Table I⇑). The enhancement was observed to a lesser degree with FcγRIIA compared with FcγRI or FcγRIIIA, which rely on the coexpressed FcγR γ subunit for transduction of the phagocytic signal. The low level of enhancement with FcγRIIA may be due to the high basal level of phagocytosis observed with FcγRIIA in the absence of coexpression of Cbl-70Z. FcγRIIA, whose activation motif differs from the classic immunoreceptor tyrosine-based activation motif, mediates phagocytosis more efficiently than FcγRI and FcγRIIIA, as described previously (3, 4). Although the mechanism by which FcγRIIA mediates phagocytosis more efficiently in COS cells than the FcRγ subunit is unknown, the high baseline level of phagocytosis with FcγRIIA may have limited the extent of enhancement by Cbl-70Z. Qualitatively, phagocytosis mediated by all three classes of FcγR were enhanced by coexpression of Cbl-70Z, indicating that Cbl-70Z modulates the pathway of phagocytosis that is shared by the FcRγ subunit and FcγRIIA.
In contrast to Cbl-70Z, Cbl-wt did not modulate phagocytosis in COS cells. Previous studies have shown different results on whether c-Cbl or its transforming mutants have dominant biological and biochemical effects, suggesting that the effect is dependent on the cell type studied (14, 16, 18, 44). However, we observed a similar positive effect with expression of Cbl-70Z and little effect with Cbl-wt on FcγR-mediated phagocytosis in P388D1 macrophages (Fig. 2⇑). The reason why expression of Cbl-70Z rather than that of Cbl-wt is effective in modulating FcγR-mediated phagocytosis is unclear. One possibility is that Cbl-70Z, due to its 17-aa deletion, behaves as a dysregulated or an activated form of c-Cbl, as suggested by previous studies (14, 16, 18, 44). On the other hand, endogenous expression of c-Cbl or Cbl-like molecules may obscure the effect of overexpression of Cbl-wt. Although it may be dysregulated, Cbl-70Z would still be expected to retain some characteristics of c-Cbl and to modulate cellular processes normally regulated by c-Cbl. Thus, our observation implies that c-Cbl may be functionally involved in the phagocytic pathway mediated by FcγR.
In addressing the question of how Cbl-70Z enhances phagocytosis by FcγR, we particularly focused on the effect of Cbl-70Z on Syk kinase. This was because Syk is essential for phagocytosis mediated by FcγR (6, 7, 12) and because negative regulation of Syk by Cbl-wt in FcεRI signaling suggests that Syk is a potential target of Cbl-70Z (16). Anti-phosphotyrosine immunoblots and in vitro kinase assays of Syk, either expressed endogenously in COS cells or overexpressed by cotransfection, failed to demonstrate that Cbl-70Z activates Syk (Fig. 3⇑). These results were surprising, in that our initial hypothesis was that the effect of Cbl-70Z was mediated by activation of Syk kinase. Functional analyses also showed that the overexpression of Syk and Cbl-70Z has an additive effect on phagocytosis mediated by FcγRI-γ-γ (Fig. 4⇑), suggesting that these two molecules enhance different pathways or steps in phagocytosis. Taken together, these findings suggest that Syk is not the direct target of Cbl-70Z in the enhancement of phagocytosis by FcγR.
A series of experiments employing mutants of Cbl-70Z demonstrated that both the PTB domain, which may be structurally a SH2 domain (45), and the C-terminal 67 aa of Cbl-70Z are critical for the optimal enhancement of phagocytosis mediated by FcγR. Our observation that v-Cbl as well as Cbl-70Z enhance phagocytosis by FcγR suggests that the N-terminal region of Cbl, including the PTB domain, plays a role in the enhancement of phagocytosis. Furthermore, the G306E mutation of Cbl-70Z and v-Cbl results in reduced enhancement of phagocytosis, thus confirming the involvement of the PTB domain in the enhancement of phagocytosis by Cbl-70Z and v-Cbl. However, the ability to enhance phagocytosis is not completely dependent on the PTB domain, since Cbl-70Z with the G306E mutation still retained some ability to enhance phagocytosis (Fig. 5⇑). The observation that the ability of v-Cbl to enhance phagocytosis is weaker than that of Cbl-70Z also suggests that the C-terminal region absent in v-Cbl (see Fig. 1⇑) is involved in the action of Cbl-70Z in FcγR-mediated phagocytosis. The dampening effect on phagocytosis of several C-terminal truncation mutants further demonstrated that the C-terminal 67 aa contribute to the ability of Cbl-70Z to enhance FcγR-mediated phagocytosis. The observation that a Cbl-70Z mutant bearing both the G306E point mutation and the C-terminal 67-aa deletion was totally lacking in ability to enhance FcγR phagocytosis further confirmed that both the PTB domain and the C-terminal tail region play roles in the enhancement of phagocytosis by Cbl-70Z.
How can the PTB domain and C-terminal region of Cbl-70Z potentiate phagocytosis? Cbl has been observed to be tyrosine phosphorylated upon cross-linking of cell surface receptors (46). Although it is not apparent from Fig. 7⇑, longer exposure of our immunoblots showed that tyrosine phosphorylation of Cbl-wt was induced by cross-linking of coexpressed FcγR (data not shown). On the other hand, Cbl-70Z was hyperphosphorylated even without cross-linking of FcγR. Cross-linking of FcγR induced a slight increase in tyrosine phosphorylation of Cbl-70Z in Fig. 7⇑, but the result was not consistent among different experiments. The internal 17-aa deletion of Cbl-70Z has been suggested to induce a conformational change that leads to increased tyrosine phosphorylation of the protein (32, 40). Phosphotyrosine residues may, in turn, recruit other molecules to enhance phagocytosis. Minimal degrees of tyrosine phosphorylation of Cbl-wt may not be sufficient to enhance the phagocytic pathway. On the other hand, v-Cbl, which also enhances phagocytosis, was not tyrosine phosphorylated in the presence or the absence of FcγR cross-linking, suggesting that the enhancement of phagocytosis by the N-terminal PTB domain is apparently independent of tyrosine phosphorylation. The contribution to phagocytosis of the C-terminal region of Cbl-70Z, rather than the N-terminal PTB domain, may, however, be dependent on tyrosine phosphorylation. The truncation mutant of Cbl-70Z (70Z/839) is partially active, and a double mutant 70Z/G306E/839 is completely inactive in phagocytosis. Both are tyrosine phosphorylated to a lesser degree than Cbl-70Z, suggesting the involvement of Cbl-70Z tyrosine phosphorylation in enhancement of phagocytosis. The C-terminal 67-aa stretch (residues 840–906), to which the activity in enhancing phagocytosis was partially mapped, contains two tyrosine residues (residues 869 and 871). However, they are not the major tyrosine phosphorylation sites of Cbl-70Z reported to date (47). The deletion of the C-terminal tail may indirectly affect tyrosine phosphorylation of the proximal C-terminal region, which reportedly contains binding sites to PI-3K, Vav, Crk, and Nck, and thus may affect the ability to enhance phagocytosis.
The PTB domain in the N-terminal region of Cbl has been implicated in the regulation of tyrosine kinases such as LET-23 in C. elegans, platelet-derived growth factor, and ZAP-70 (14, 18, 32). However, we could not detect an increase in overall tyrosine phosphorylation in COS cells expressing Cbl-70Z or v-Cbl (data not shown). This observation suggests that mechanisms other than activation of tyrosine kinases are involved in the ability of the PTB domain to enhance phagocytosis. Alternatively, an increase in tyrosine phosphorylation undetectable in our transient expression system in COS cells may be responsible for the activity of the PTB domain.
Inhibition of the stimulatory effect of Cbl-70Z or v-Cbl on phagocytosis by wortmannin, a PI-3K inhibitor, indicates that PI-3K is required for Cbl-70Z or v-Cbl to enhance phagocytosis. However, immunoprecipitation studies suggest that the direct interaction of Cbl-70Z and the p85 subunit of PI-3K is not involved in the enhancement of FcγR-mediated phagocytosis in COS cells. The study with C-terminal deletion mutants of Cbl-70Z (Fig. 6⇑) also showed that the deletion encompassing Tyr731, which is reportedly the major binding site to the p85 subunit of PI-3K (44, 48), did not change the ability of Cbl-70Z to enhance phagocytosis, implying that the physical interaction of Cbl-70Z and the p85 subunit may not play a role in the enhancement of phagocytosis. Cbl may indirectly modulate a molecule(s) upstream of PI-3K. Alternatively, PI-3K may be in a pathway parallel to that modified by Cbl-70Z, playing a permissive role for cells to perform phagocytosis.
We have studied the mechanism of phagocytosis by FcγR in a heterologous expression system in COS cells (3, 4). This approach provides the opportunity to study the role of a single class of FcγR in the absence of any other Fc receptor and allows the coexpression of proteins of interest. COS cells possess the machinery for phagocytosis, evidenced by their ability to phagocytose IgG opsonized particles when transfected with appropriate FcγR. Phagocytosis mediated by FcγR in COS cells is also demonstrated morphologically to be quite similar to that in professional phagocytes (41, 47, 49) indicating that COS cells are an appropriate experimental model to study the molecular events involved in FcγR-mediated phagocytosis. However, it is not certain whether the mechanism of phagocytosis in COS cells is precisely the same as that in such myeloid cells as monocytes/macrophages. Although we have shown that the expression of Cbl-70Z enhances phagocytosis in both COS cells and murine macrophages, the molecular mechanism of the enhancement may or may not be identical.
In summary, we have observed the stimulatory effects of mutants of Cbl on phagocytosis induced by FcγR in COS cells and the macrophage cell line P388D1. Our data indicate that transforming mutants of Cbl can modulate phagocytic signaling mediated by FcγR and imply a functional involvement of Cbl in FcγR-mediated phagocytosis. Our observation that both the PTB domain and the C-terminal tail contribute to the action of Cbl-70Z is novel and provides an insight into the structural basis of the action of Cbl and its oncogenic mutants.
Acknowledgments
We thank Drs. L. E. Samelson, C. Couture, D. H. Chu, and A. Weiss for cDNAs and Abs; Dr. Z. K. Indik for helpful comments; and Dr. P. Chien for technical assistance.
Footnotes
-
↵1 This work was supported by National Institutes of Health Grants HL28207 and AI22193 and a grant from Tsumura & Co. (Tokyo, Japan).
-
↵2 Address correspondence and reprint requests to Dr. Alan D. Schreiber, Department of Medicine, University of Pennsylvania School of Medicine, Room 705 Biomedical Research Building II/III, 421 Curie Boulevard, Philadelphia, PA 19104. E-mail address: schreibr{at}mail.med.upenn.edu
-
↵3 Abbreviations used in this paper: EA, Ab-coated erythrocytes; PI-3K, phosphatidylinositol 3-kinase; Cbl-wt, wild-type Cbl; PTB, phophotyrosine binding; HA, influenza hemagglutinin.
- Received December 30, 1998.
- Accepted September 21, 1999.
- Copyright © 1999 by The American Association of Immunologists