If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
To whom correspondence should be addressed: Division of Infectious Diseases, Campus Box 8051, Washington University School of Medicine, 660 S. Euclid Ave., St. Louis, MO 63110. Tel.: 314-362-2125; Fax: 314-362-9230
Division of Infectious Diseases, Washington University School of Medicine, St. Louis, Missouri 63110
∗ This work was supported by Grants AI35811, AI24674, and GM38330 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Two tyrosine kinase-dependent pathways exist for activation of the respiratory burst by polymorphonuclear leukocyte (PMN) immunoglobulin G Fc receptors. Direct ligation of FcγRII activates the respiratory burst, but ligation of the glycan phosphoinositol-linked FcγRIIIB does not. Instead, this receptor and the integrin complement receptor CR3 synergize in activation of the respiratory burst (Zhou, M.-J., and Brown, E. J. (1994) J. Cell Biol. 125, 1407-1416). Here we show that direct ligation of FcγRII leads to activation and Triton X-100 insolubility of the Src family kinase Fgr, without effect on the related myeloid Src family member Hck. In contrast, adhesion of PMN via FcγRIIIB leads to activation and Triton X-100 insolubility of Hck but not Fgr. The exclusive association of FcγRIIIB with Hck activation and Triton insolubility is not solely a result of its glycan phosphoinositol anchor, since decay accelerating factor (CD55), another prominent glycan phosphoinositol-anchored PMN protein, is associated with Fgr insolubility to a greater extent than Hck. Ligation of decay accelerating factor, with or without coligation of CR3, does not activate the PMN respiratory burst. Coligation of FcγRIIIB with FcγRII overcomes the pertussis toxin inhibition of H2O2 production in response to direct ligation of FcγRII. These data support the hypothesis that activation of Hck upon FcγRIIIB ligation has a role in generation of the synergistic respiratory burst.
Binding of immune complexes to cell receptors for immunoglobulin is a powerful stimulus to activation of phagocytes. Immune complex binding to these cells leads to activation of effector functions of host defense such as phagocytosis, secretion, cytokine synthesis, and the production of toxic oxygen metabolites, which occurs as a result of NADPH oxidase assembly (
). The cloning of several members of the immunoglobulin Fc receptor family over the past several years has tremendously enhanced understanding of the molecular mechanisms for the functions of these immunologically important receptors. Identification of receptors within the family which associate with a second transmembrane protein, called the γ chain, and realization that the γ chain is itself a member of a family of proteins known to be involved in tyrosine kinase-mediated signal transduction has further enhanced understanding of Fc receptor-mediated cell activation. The two Fcγ receptors expressed on polymorphonuclear leukocytes (PMN),1(
FcγRII and the glycan phosphoinositol (GPI)-linked form of FcγRIII (FcγRIIIB) are distinct within this family for their failure to associate with the γ chain, suggesting alternative mechanisms for cell activation by these receptors. Nonetheless, even FcγRII appears to activate tyrosine kinases and to be phosphorylated on tyrosine during immune complex-mediated cell activation (
In contrast, little is known about the molecular mechanisms of signal transduction from the PMN FcγRIIIB. This receptor is unique among Fc receptors, since it is expressed on the plasma membrane by a GPI anchor, rather than as a transmembrane protein. It has no association with the γ chain, which associates with the transmembrane form of FcγRIII expressed in NK cells and macrophages, and has no intracytoplasmic domain for direct association with cytosolic signal transduction cascades. Indeed, whether FcγRIIIB mediates signal transduction at all remains controversial. Many effects of immune complexes on PMN appear to be mediated exclusively by FcγRII (
). Thus, the role for FcγRIIIB in PMN activation remains uncertain.
Recently, we have developed an assay system which allows a more detailed investigation of the contribution of individual PMN Fcγ receptors to signal transduction leading to generation of a respiratory burst (
). We have used PMN adhesion to surfaces coated with monoclonal antibodies (mAb) to individual receptors to assess their contribution to cell activation. We have found that direct ligation of FcγRII leads to a respiratory burst, whereas direct ligation of FcγRIIIB does not. Instead, FcγRIIIB cooperates with the PMN integrin CR3 (Mac-1, CD11b/CD18) to generate what we have termed a synergistic respiratory burst. In the synergistic respiratory burst, the two membrane receptors have distinct roles. Ligation of CR3 immobilizes FcγRII to the adherent plasma membrane by a cytoskeleton-dependent mechanism, and ligation of FcγRIIIB induces appropriate tyrosine kinase activation in proximity to the immobilized FcγRII. Thus, FcγRII is required in addition to CR3 and FcγRIIIB for the synergistic respiratory burst. In the current work, we have examined the nature of the Src family kinases activated by ligation of the two Fcγ receptors. We have found that FcγRII and FcγRIIIB immobilized on the adherent PMN surface by direct ligation lead to the activation and Triton X-100 insolubility of different Src family kinases. FcγRII is associated with activation and translocation of Fgr to the Triton-insoluble cell fraction; and FcγRIIIB is associated with Hck activation and translocation. The exclusive association of FcγRIIIB with Hck activation is not a property of all GPI-linked proteins in PMN, since immobilization of decay accelerating factor (DAF, CD55) leads primarily to Fgr activation at the adherent membrane. Moreover, DAF cannot substitute for FcγRIIIB in synergistic activation of the respiratory burst. From these data we conclude that ligation of FcγRII and FcγRIIIB activate and translocate distinct Src family members in PMN. We hypothesize that the functional consequence of the activation and translocation of distinct kinases is the existence of two separate signal transduction pathways for activation of the PMN respiratory burst by immune complexes.
MATERIALS AND METHODS
The following mAb were used in these studies: IB4 (anti-CD18) (
). W6/32 and B6H12 IgG were prepared using an Amicon Bioreactor (Amicon Inc., Danvers, MA) according to manufacturer's instructions. SDS-PAGE of all purified IgG preparations showed them to be >90% IgG. PY20 (anti-phosphotyrosine) was purchased from Transduction Laboratories (Lexington, KY). Polyclonal anti-Fgr was as described (
) or was purchased from Santa Cruz Biotechnology (Santa Cruz, CA), as was anti-Hck.
Buffers and Other Reagents
Phosphate-buffered saline was from Biowhittaker, Walkersville, MD. Krebs-Ringer buffer (KRPG) was 145 mM NaCl, 4.86 mM KCl, 1.22 mM MgSO4, 5.7 mM Na2HPO4, 0.54 mM CaCl2, and 5.5 mM glucose, pH 7.4. Reaction mixture (RM) consisted of 37.5 μM scopoletin, 1.25 mM NaN3, 1.25 units/ml HPO in KRPG. Kinase buffer was 40 mM Hepes, pH 7.5, 10 mM MgCl2, 3 mM MnCl2, 0.1 mM Na3VO4, 10% glycerol, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride. MBS buffer was 25 mM Mes, pH 6.5, 0.5% Triton X-100, 150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride, 5 mM NaF, 5 mM diisopropyl fluorophosphate, 1 mM Na3VO4, 1 mM iodoacetamide, 10 μg/ml each of aprotinin, leupeptin, and pepstatin A. Other reagents were obtained from standard sources as described previously (
). PMN were obtained from a β2 integrin-deficient (leukocyte adhesion-deficient, LAD) patient followed at Baylor College of Medicine. This patient has been characterized as having the complete deficiency phenotype. Expression of FcγRII and FcRγIII on this patient's cells was normal, whereas CR3 was undetectable (not shown). The patient's blood and a normal control were transported and the PMN prepared as described (
). In each experiment, data were averaged from triplicate wells, which generally varied from each other by ≤10%. Unless otherwise stated, H2O2 accumulation was measured after 60 min of PMN incubation in wells.
Pretreatment of PMN
PMN at 2.5 × 106 cells/ml in KRPG were pretreated with 5 μg/ml of Fab or 2.5 μg/ml of F(ab′)2 of various mAb at 4°C for 15 min. Without washing, PMN were added to antibody coated plates containing RM (
). For treatment with pharmacologic agents, PMN were preincubated with herbimycin at 10 μg/ml and then added to RM containing the same concentration of the indicated agent. To pretreat PMN with pertussis toxin, cells in KRPG were incubated with 2 μg/ml of pertussis toxin at 37°C for 75 min. In experiments examining the effects of cytochalasin, PMN were pretreated with 5 μg/ml cytochalasin D and added to RM containing no additional drug. In all experiments, control PMN were incubated with an identical concentration of nonaqueous diluent as the PMN receiving any drug.
“In Situ” Tyrosine Phosphorylation
5 × 106 PMN in KRPG with 0.1 mM Na3VO4 were allowed to adhere to mAb-coated six-well plates at 4°C for 10 min and then at 37°C for 15 min. The nonadherent cells were removed by washing with cold MBS buffer. The adherent cells were extracted with 0.5% Triton X-100 in MBS buffer on ice for 8 min, following which the plate was rinsed with kinase buffer containing 0.1 mM Na3VO4 and 0.2% Triton X-100 once. In situ tyrosine phosphorylation was performed by adding 50 μCi of [γ-32P]ATP and 1 μM cold ATP in 500 μl of kinase buffer at room temperature for 15 min. For immunoprecipitation, following washing with the kinase buffer three times, the cell residuals were solubilized with solubilization buffer (20 mM Tris, pH 8.0, 150 mM NaCl, 1% Triton X-100, 1% deoxycholate, 0.1% SDS, 10 μg/ml each of aprotinin, leupeptin, and pepstatin A, 1 mM phenylmethylsulfonyl fluoride, 2 mM diisopropyl fluorophosphate, 1 mM EDTA, 1 mM Na3VO4) at 4°C for 60 min with shaking. The lysates were diluted with solubilization buffer without SDS and then precleared with protein A-Sepharose at 4°C for 2 h. The precleared lysate was incubated with anti-Fgr, -Hck, or -phosphotyrosine IgG captured by protein A-Sepharose at 4°C for 2 h. The Sepharose mixture was washed once with solubilization buffer and then six times in the same buffer without deoxycholate, SDS, and protease inhibitors. The Sepharose-bound proteins were analyzed by SDS-PAGE as described (
To determine whether Fgr and Hck were activated by adhesion, PMN were allowed to adhere to surfaces coated with various mAb for 15 min at 37°C. After MBS buffer extraction, supernatant was collected and the Triton X-100-insoluble fraction solubilized in solubilization buffer as above. Both the Triton X-100-soluble and -insoluble fractions were immunoprecipitated with anti-Fgr or anti-Hck. Following immunoprecipitation, the immune complexes were incubated with 5 μCi of [γ-32P]ATP in 30 μl of kinase buffer at room temperature containing 20 μg of poly(Glu,Tyr) (Glu:Tyr ratio 4:1, Sigma) for 15 min. The reaction was stopped by adding 6 μl of 10% phosphoric acid and centrifugation. The supernatants were spotted onto Whatman 3MM cellulose filter paper (1.5 × 1.5 cm), and the filters were washed with 10% trichloroacetic acid four times, 5 mM phosphoric acid, and 80% acetone once each. After air-drying the filters, the 32P incorporation was measured by scintillation counting. Alternatively, the supernatants were loaded into the sample bucket of a phosphocellulose unit (Pierce) and washed according to the manufacturer's instructions. In experiments in which the Triton X-100-insoluble cell fraction was immunoprecipitated with PY20, cell residuals were incubated with kinase buffer containing 10 mM unlabeled ATP at room temperature for 15 min prior to solubilization and immunoprecipitation.
Following PMN adhesion on mAbs and bovine serum albumin-coated plates, the adherent cells were extracted with 0.5% Triton X-100 in KRPG buffer with protease inhibitors on ice for 5 min. The cell residuals were lysed by solubilization buffer with 0.2% SDS. The recovered lysates were precipitated by adding 5 volumes of cold acetone and kept at −20°C for 30 min. After centrifugation at top speed in an Eppendorf centrifuge at 4°C for 10 min, the pellets were washed with 80% acetone twice and allowed to air-dry. The precipitated proteins were dissolved by SDS-sample buffer with 5 mM β-mercaptoethanol and 2 mM EDTA and separated by SDS-PAGE, then transferred to polyvinylidene difluoride (Millipore, Bedford, MA). The membranes were incubated overnight in blocking buffer (PBS Tween and 3% bovine serum albumin) and then with anti-Hck and anti-Fgr polyclonal IgG at 1:1000 at room temperature for 4 h. The membranes finally were incubated with horseradish peroxidase-conjugated protein A and developed using the ECL chemiluminescence kit (Amersham Corp.).
FcγRII or FcγIII Ligation in PMN Leads to Activation of Distinct Src Family Tyrosine Kinases
Our previous studies have shown distinct roles for the two PMN Fcγ receptors in activation of the respiratory burst. PMN adhesion to an FcγRII ligand leads to a respiratory burst, whereas the GPI-linked FcγRIIIB is required for the synergistic respiratory burst, but is incapable of activation of the NADPH oxidase on its own. Nonetheless, our previous data demonstrate that both pathways are inhibited by tyrosine kinase inhibitors and suggest that FcγRIIIB ligation is associated with tyrosine kinase activity (
). To begin to understand tyrosine kinase function in the two pathways, we examined activation of two predominant PMN Src family kinases, Fgr and Hck, upon ligation of PMN Fcγ receptors. As assessed by phosphorylation of poly(Glu,Tyr), ligation of FcγRII led to Fgr activation, whereas ligation of FcγRIIIB led to Hck activation (Fig. 1A). Fgr activation has been associated previously with FcγRII ligation in PMN (
). Ligation of either HLA or CR3 did not lead to a detectable increase in the activity of either kinase. Total Fgr activity increased approximately 6.5-fold (average of two determinations) in cells adherent via FcγRII compared with ligation of HLA, and total Hck activity increased more than 20-fold upon FcγRIII ligation (Fig. 1A). A small amount of Fgr kinase activity could be detected in the Triton X-100-soluble fraction in PMN incubated on the control surfaces. However, neither Fgr nor Hck kinase activity was detected in the Triton X-100-insoluble fraction from PMN adherent via HLA or CR3. After either FcγRII or FcγRIIIB ligation, kinase activity was increased in both Triton X-100 soluble and insoluble cell fractions. Thus, Fgr kinase activity was stimulated by ligation of FcγRII; Hck activity was stimulated by ligation of FcγRIIIB.
FcγRII or FcγRIII Ligation Leads to Association of Distinct Src Family Kinases with the Triton X-100-insoluble Cell Fraction
Src family kinases often associate with the Triton X-100-insoluble cytoskeleton upon activation, and Src-induced oncogenic transformation may require cytoskeletal association (
), suggesting that important kinase substrates are encountered through translocation to the cytoskeleton. Since Triton X-100 insolubility may reflect this translocation to the cytoskeleton, we examined the differences in the Triton X-100 solubility of Fgr and Hck upon receptor ligation. PMN were adhered to mAb-coated surfaces, briefly solubilized with Triton X-100, and the cell residue analyzed by SDS-PAGE and Western blotting (Figs. 1B and 2). Hck was present in the Triton X-100- insoluble cell fraction if, and only if, PMN were adherent to anti-FcγRIIIB (Fig. 2A). Whether or not anti-CR3 also was present on the adherent surface made no difference for Hck localization, even though respiratory burst activation only occurs when both antibodies are present. Fgr was present in the Triton X-100-insoluble fraction only when FcγRII was ligated on the adherent surface (Fig. 2B). Unlike FcγRIIIB, ligation of a different GPI-linked protein, DAF, led predominantly to Fgr localization in the Triton X-100-insoluble residue. When cells were adherent to anti-CR3 alone, no Fgr or Hck protein was detectable. Specific activity of Fgr and Hck was estimated in both the Triton X-100-soluble and -insoluble fractions after FcγR ligation by comparing kinase activity (Fig. 1A) to protein concentration as estimated from the Western blots (Fig. 1B). Interestingly, although active Fgr or Hck was present in both Triton X-100-soluble and -insoluble compartments, the specific activity of both kinases was 6-7-fold greater in the Triton X-100-insoluble fraction after appropriate receptor ligation (Fig. 1C). Presumably, this reflects the fact that not all Triton X-100-soluble kinase has been activated by receptor ligation.
Tyrosine Kinase Activity in the Triton X-100-insoluble Fraction after Ligation of FcγRII or FcγRIIIB
To examine the phosphorylation patterns in the Triton X-100-insoluble fractions after FcγRII or FcγRIIIB ligation, an in situ kinase assay was performed after extraction of adherent PMN with Triton X-100. After addition of [32P]ATP to the Triton X-100-insoluble residue adherent to the immune complex-coated surfaces, phosphotyrosine-containing proteins were examined by immunoprecipitation of the products of the in vitro kinase assay with anti-phosphotyrosine mAb, PY20, after solubilization with a more stringent detergent (1% Triton X-100, 1% deoxycholate, 0.1% SDS). As shown in Fig. 3, phosphotyrosine-containing proteins with molecular mass of ~55-60 kDa were the major products of the in vitro kinase assay when PMN were adherent to either an anti-FcγRII- or an anti-FcγRIIIB- coated surface. Under the same conditions, these phosphotyrosine-containing proteins were not detected from PMN adherent to anti-HLA- or anti-CD47-coated surfaces. Importantly, no tyrosine kinase activity was found in PMN adherent to an anti-CR3-coated surface (Fig. 3), even though FcγRII is present in the Triton X-100-insoluble material (
), and CR3 participates in the synergistic respiratory burst. No reproducible kinase activity was found in anti-FcγRII or anti-FcγRIIIB immunoprecipitates from unstimulated, nonadherent PMN (data not shown).
The autophosphorylated Triton X-100-insoluble material also was immunoprecipitated with monospecific antibodies against several Src family kinases. No radiolabeled proteins were immunoprecipitated by antibodies to Yes, Fyn, Src, or Lck from PMN adherent to any surface (data not shown). Phosphorylated proteins from lysates of PMN adherent to an anti-FcγRIIIB coated surface were specifically immunoprecipitated by anti-Hck, but not by anti-Fgr (Fig. 4). As described previously (
), the autophosphorylated Hck appeared as a doublet. Conversely, a phosphorylated band was immunoprecipitated by anti-Fgr, but not by anti-Hck from the lysate from an anti-FcγRII-coated surface. When autophosphorylated lysates from an anti-HLA coated surface were precipitated by both anti-Hck and anti-Fgr, no phosphorylated bands were identified, consistent with the minimal kinase activity associated with the Triton X-100-insoluble material remaining from PMN adherent to this substrate (see Fig. 1A and 3).
Hck Is the Predominant Triton X-100-insoluble Tyrosine Kinase Associated with FcγRIIIB Ligation, and Fgr Is the Predominant Triton X-100-insoluble Tyrosine Kinase Associated with FcγRII Ligation
To determine whether Hck was the predominant tyrosine kinase associated with the Triton X-100-insoluble material from FcγRIIIB ligation and Fgr the predominant kinase associated with FcγRII ligation, poly(Glu,Tyr) phosphorylation was assayed. When lysates from anti-FcγRIII coated surface were immunoprecipitated with anti-Hck, the anti-Hck immune complexes displayed marked tyrosine kinase activity (Fig. 5). Minimal tyrosine kinase activity could be immunoprecipitated by PY20 after anti-Hck preclearing. In contrast, anti-Fgr precipitated almost no poly(Glu,Tyr) phosphorylating activity from the PMN adherent via FcγRIIIB. When lysates from anti-FcγRII-coated surface were immunoprecipitated with anti-Fgr and anti-Hck, the anti-Fgr immune complexes contained 80% of the tyrosine kinase activity. There was very little detectable Hck tyrosine kinase activity from lysates from anti-FcγRII coated surfaces (Fig. 5). Immunoprecipitates with PY20 after anti-Fgr preclearing contained minimal kinase activity toward poly(Glu,Tyr). Thus, we concluded that Hck accounted for almost all the Triton X-100-insoluble tyrosine kinase activity associated with FcγRIIIB ligation, and Fgr accounted for the kinase activity associated with FcγRII ligation. These data suggest that Hck specifically associates with the Triton-insoluble cell fraction from cells adherent via FcγRIIIB, and Fgr kinase specifically associates with the Triton-insoluble cell fraction following adhesion via FcγRII.
To determine whether the apparent exclusive association of Hck activity with FcγRIIIB ligation was true for all GPI-linked PMN proteins, PMN were allowed to adhere to anti-DAF-coated surfaces and Triton X-100-insoluble kinase activity determined (Fig. 5). Unlike PMN adherent to anti-FcγRIIIB, cells adherent to anti-DAF showed more Fgr than Hck activity.
Requirement for Both FcγRIIIB and FcγRII for Synergistic Respiratory Burst Activation
The signal transduction involved in activation of the respiratory burst by direct ligation of FcγRII or by synergy between FcγRIIIB and CR3 differ in several respects, including sensitivity to pertussis toxin (PT) and cytochalasin D (
). Direct ligation of FcγRII activates H2O2 production, which is inhibited by PT (Fig. 6A); activation of the synergistic respiratory burst is unaffected by PT. Because our previous studies suggested that the role for CR3 in the synergistic respiratory burst was to immobilize FcγRII in proximity to FcγRIIIB, we tested whether direct ligation of FcγRII and FcγRIIIB together could activate the synergistic respiratory burst, by taking advantage of the PT sensitivity of the respiratory burst generated by direct ligation of FcγRII. PT inhibited the respiratory burst generated by ligation of FcγRII, but coligation of FcγRIIIB with FcγRII restored the respiratory burst in the presence of PT (Fig. 6B). Ligation of FcγRIIIB specifically is required for generation of the respiratory burst with FcγRII in the presence of PT, since neither coligation of DAF, another prevalent GPI-linked protein on PMN, nor of HLA, could stimulate respiratory burst in combination with FcγRII in the presence of PT (Fig. 6B). Coligation of FcγRIIIB and DAF also failed to stimulate the respiratory burst (not shown). Addition of Fab or F(ab′)2 fragments of mAbs against either FcγRII or FcγRIIIB specifically inhibited the respiratory burst stimulated by coligation of FcγRIIIB with either FcγRII or CR3 (Fig. 7A). mAb against CR3 inhibited the PMN respiratory burst by coligation of FcγRIIIB with CR3, but had no effect on the respiratory burst generated by coligation of FcγRII and FcγRIIIB in the presence of PT (Fig. 7A). Thus, coligation of FcγRII and FcγRIIIB in the presence of PT is a second form of the synergistic respiratory burst in which the need for CR3 has been eliminated.
These data suggested that the only role for CR3 in the synergistic respiratory burst was to immobilize FcγRII on the adherent surface of the PMN in proximity to the ligated FcγRIIIB. To test this hypothesis, we examined LAD PMN which genetically lack CR3. Coligation of FcγRIIIB with FcγRII activated PT intoxicated LAD PMN normally (Fig. 6C), even though these PMN lacked the synergistic respiratory burst from adhesion to surfaces coated with anti-FcγRIIIB and anti-CR3 (
). As a second test of the hypothesis, we examined the effect of cytochalasin D on the synergistic respiratory burst. Although the synergistic activation by adhesion to anti-CR3 and anti-FcγRIIIB is sensitive to cytochalasin D (
), the PT-insensitive respiratory burst in response to coligation of FcγRII and FcγRIIIB was not (Fig. 7B). Since cytochalasin has been shown to inhibit CR3-mediated immobilization of FcγRII on the adherent PMN membrane (
), this experiment proves that coimmobilization of FcγRII and FcγRIIIB by antibodies overcomes the need for cytoskeletal assembly in the synergistic respiratory burst.
Immune complexes are potent stimuli for activation of PMN. Immune complex deposition and subsequent PMN activation is an important part of the pathogenesis of serum sickness, the Arthus reaction, acute glomerulonephritis, rheumatoid arthritis, and other idiopathic inflammatory diseases. Although these are host-damaging diseases, immune complex-mediated PMN activation also plays an essential role in host defense against bacterial infection. It is clear that in vivo both host defense and host damaging aspects of the PMN-immune complex interaction involve complement activation and deposition onto the immune complexes (
). The major PMN receptor for the complement deposited onto immune complexes is the leukocyte integrin CR3. Therefore, to understand immune complex activation at a molecular level requires understanding the roles for both Fcγ receptors and CR3. Although the role for CR3 in this activation process was originally thought to be passive, merely increasing the interaction between IgG and its cellular receptors (
In the current work, we have studied signal transduction during the synergistic respiratory burst and compared it with the signal transduction cascade activated by direct ligation of FcγRII. We have shown that the role for CR3 in the synergistic respiratory burst is to immobilize FcγRII in proximity to ligated FcγRIIIB. Indeed, activation of Fgr and Hck through Fcγ receptor ligation occurs normally in LAD PMN (data not shown). Thus, unlike the respiratory burst induced by tumor necrosis factor-α stimulation of PMN (
) that complement simply enhances immune complex presentation to phagocyte Fcγ receptors, is correct. However, it is clear that signal transduction in the synergistic respiratory burst is more complex, since it requires two different Fcγ receptors, with distinct roles for each.
By using PT, which suppresses the respiratory burst from direct ligation of FcγRII, we were able to reconstitute the synergistic respiratory burst in the complete absence of CR3. This experiment proves our early hypothesis that FcγRII plays an essential role in the synergistic respiratory burst, even when it is immobilized only by its association with CR3 and not directly ligated (
). In this situation, we found no tyrosine kinase activity associated with the immobilized FcγRII. In contrast, direct ligation of FcγRII led to Fgr tyrosine kinase activation and association of a significant fraction of the active kinase with the Triton-resistant adherent cell membrane. An association between FcγRII and Fgr in PMN has been reported previously (
). In our assay system at least, stable association of Fgr with the Triton X-100-insoluble cytoskeleton requires direct ligation of FcγRII. Even unactivated Fgr protein is not associated with the Triton X-100-insoluble cytoskeleton in the absence of direct FcγRII ligation. Presumably for this reason, FcγRII immobilized by association with ligated CR3 is unable to signal assembly of the NADPH oxidase. Because of the nature of the assay, we do not know if the FcγRII-Fgr association we detect is direct or indirect. The presence of Fgr in the Triton X-100-insoluble cell residue after FcγRII ligation may depend on other Triton-insoluble PMN proteins. It is also possible that active Fgr associates with phosphorylated tyrosine residues in the FcγRII cytoplasmic tail (
) and has no direct association with cytoskeletal proteins.
Although direct evidence that Fgr kinase activity is involved in NADPH oxidase assembly is lacking, we believe it very likely that Fgr has a role in generation of the respiratory burst from direct ligation of FcγRII. The anti-FcγRII-generated respiratory burst is extremely sensitive to tyrosine kinase inhibitors (Fig. 6 and Ref.
). Since a major difference between direct ligation of FcγRII with antibody (which is associated with H2O2 production) and its indirect immobilization through CR3 ligation (which leads to no respiratory burst) is the presence of Triton X-100-insoluble kinase-active Fgr, it is likely that Fgr is involved in initiation of the respiratory burst after direct FcγRII ligation. This pathway for generation of the respiratory burst is sensitive to PT. However, the association of active Fgr with FcγRII after direct ligation is unaffected by PT, and the in situ kinase assay reveals no different phosphorylated proteins in pertussis-intoxicated cells. Thus, the PT sensitivity of the signal transduction cascade must result from inhibition at a step beyond Fgr association with receptor and its activation.
Immobilized FcγRII is required for a respiratory burst even in pertussis-intoxicated PMN. Although FcγRII is phosphorylated on tyrosine during Fcγ receptor-mediated PMN activation (
These data suggest that there is another, unidentified, protein associated with FcγRII which is needed to signal assembly of the NADPH oxidase. A possible candidate is syk, a kinase known to be involved in immune complex-mediated PMN activation and known to be phosphorylated by FcγRII ligation (
), but its specific role has been unclear. This exclusive association of FcγRIIIB ligation with Hck activation is not a property of all GPI-linked PMN proteins, since DAF ligation is associated predominantly with Fgr rather than Hck. This may explain why DAF cannot substitute for FcγRIIIB in the synergistic respiratory burst. However, other differences, such as the association of extracellular domains of FcγRIIIB with specific membrane proteins (
), may also contribute to the lack of equivalence between DAF and FcγRIIIB in the synergistic respiratory burst. Nonetheless, in PMN, whether a particular Src family kinase is activated and becomes Triton insoluble by ligation of a specific GPI-linked protein depends on more than simply the mechanism of membrane attachment of the receptor.
Thus, we hypothesize that the role for ligation of FcγRIIIB in the synergistic respiratory burst is to bring Hck into proximity with FcγRII, when the FcγRII has no associated kinase or has had its kinase activation pathway blocked by PT (Fig. 8). This appears to be an essential early step in activation of the synergistic respiratory burst. Hck activation of the downstream effector pathway for assembly of the NADPH oxidase is not affected by PT, providing an alternative pathway for cell activation. Since FcγRII immobilization is required for the synergistic respiratory burst in the absence of associated Fgr, it is tempting to speculate that the downstream target for both Hck in the synergistic respiratory burst and Fgr in direct ligation of FcγRII is the same protein physically associated with FcγRII. Activation of this target by either a PT-insensitive Hck-dependent pathway or a PT-sensitive Fgr-dependent pathway could then lead to a final common pathway for NADPH oxidase assembly.
In summary, we have shown that in PMN ligated FcγRII and FcγRIIIB are associated with distinct Src family kinases. FcγRII leads to Fgr activation and FcγRIIIB to Hck activation. The Hck-dependent pathway initiates what we have termed the synergistic respiratory burst. In the absence of pertussis toxin, this pathway for PMN activation apparently requires three different receptors: CR3, FcγRII, and FcγRIIIB. It is possible that immune complexes deposited in tissue, by ligating FcγRII and FcγRIIIB simultaneously, could activate the synergistic pathway in the absence of complement. Indeed, the PMN respiratory burst in response to insoluble immune complexes in vitro is PT-insensitive (
). For this reason and because there are about 10-fold more FcγRIIIB than FcγRII on PMN, the most physiologically relevant interaction of immune complexes with PMN is likely to be via CR3 and FcγRIIIB; direct interaction with the less abundant FcγRII will likely be minimized by complement deposition. These are precisely the conditions which lead to the synergistic respiratory burst. We believe that the Hck-dependent synergistic pathway for PMN activation is likely to be extremely relevant to immune complex activation of PMN in vivo.
We are indebted to Dr. Donald Anderson for provision of blood from a patient with LAD and Dr. Andrey Shaw for antibodies against Fyn, Yes, Lck, and Src. We also thank various members of the Brown laboratory for helpful comments during the execution of these experiments.
Abramson J.S. Wheeler J.G. The Neutrophil. IRL Press,