Cooperative integrin/ITAM signaling in platelets enhances thrombus formation in vitro and in vivo

Among the extracellular cues that cells continuously sample and respond to, adhesion is among the most potent, eliciting a wide range of cell biological processes, including but not limited to changes in cytosolic calcium, protein and lipid phosphorylation, cytoskeletal architecture, gene transcription, and cell migration.¹  Members of the integrin family, composed of 24 αβ heterodimer transmembrane receptors,²  are particularly adept at transmitting adhesion-initiated signals into the cell in a process sometimes referred to as outside-in signaling.¹  The platelet-specific integrin αIIbβ3 (CD41/CD61, glycoprotein [GP] IIb-IIIa in the platelet literature) is among the best-studied members of the integrin family, and is thought to be particularly responsive to intracellular and extracellular stimuli. Thus, binding of adhesive ligands to αIIbβ3 has been shown to result in dramatic conformational changes within the αIIbβ3 extracellular domain³  that, in a process likely involving ligand-binding–induced swing out of the β3 subunit hybrid domain and separation of integrin α and β subunit stalk domains,⁴  become propagated across the plasma membrane to the cytoplasmic face of the integrin. Binding of multivalent ligands like fibrinogen, fibronectin, and von Willebrand factor additionally results in clustering of integrin receptors.⁵  These events converge to initiate a series of biochemical and cell biological events at the cytosolic face of the integrin that ultimately lead to changes in cytoskeletal architecture, granule secretion, and further integrin activation that together serve to stabilize the platelet-platelet interactions that occur at sites of a growing thrombus. Outside-in signaling, therefore, contributes importantly to both thrombosis and hemostasis.

An extensive number of adaptor proteins, kinases, and phosphatases have been found to participate in αIIbβ3-mediated, adhesion-initiated signaling. In some cases, ligand binding induces association of the heterotrimeric guanine nucleotide-binding protein (G protein), Gα₁₃, with the β3 cytoplasmic domain,⁶  where it activates integrin-associated Src-family kinases (SFKs) that go on to phosphorylate and activate a plethora of molecular targets.⁷  Of these, 2 tyrosine residues that reside within the immunoreceptor tyrosine-based activation motif (ITAM) of the cytoplasmic domain of Fc receptor γ-chain IIa (FcγRIIa) have been identified as SFK targets⁸  that, once phosphorylated, create a docking site for the tandem Src homology 2 (SH2) domains of the tyrosine kinase Syk.⁹  Activated Syk, in turn, goes on to amplify multiple pathways involved in integrin and platelet activation.¹⁰ 

The precise relationship between αIIbβ3 and Syk has been a subject of much interest in recent years. Early studies using overexpression systems and cultured cells found that the N-terminal SH2 domains of Syk could associate with the integrin β3 cytoplasmic domain in a phosphotyrosine-independent manner.¹¹  More recent studies using mutant forms of Syk expressed in reconstituted murine hematopoietic cells, however, showed that the SH2 domains of Syk are indispensable for integrin-mediated outside-in signaling,¹²  implying that an ITAM-bearing protein might be involved in targeting Syk to the inner face of the plasma membrane, where it could both receive and send adhesion-initiated signals into the cell.

The observation¹³  that a monoclonal antibody (mAb) specific for FcγRIIa is able to inhibit adhesion-initiated tyrosine phosphorylation of FcγRIIa, Syk, and phospholipase Cγ2 (PLCγ2), as well as platelet spreading on immobilized fibrinogen, strongly suggested that a functional interplay between αIIbβ3 and FcγRIIa is required for optimal platelet function during thrombosis and hemostasis. The physiological significance of coupling between this integrin/ITAM pair, however, remains unknown. To explore the importance of FcγRIIa in platelet function, we compared the relative abilities of wild-type (WT) FcγRIIa-negative (FcγRIIa^neg) and transgenic FcγRIIa-positive (FcγRIIa^TGN = FcγRIIa^pos) murine platelets to support thrombosis and hemostasis in a number of well-accepted models of platelet function. The findings described herein establish FcγRIIa as a physiologically important functional conduit for αIIbβ3-mediated outside-in signaling, and suggest that modulating the activity of this novel integrin/ITAM pair might be effective in controlling thrombosis.

The hybridoma cell line for the anti-FcγRIIa mAb, IV.3, was obtained from the American Type Culture Collection (Manassas, VA). Antibodies specific for Syk were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies specific for tyrosine-phosphorylated Syk tyrosine residues 525 to 526 and tyrosine-phosphorylated PLCγ2 tyrosine residue 759 were purchased from Cell Signaling Technology (Danvers, MA). The antiphosphotyrosine mAb 4G10 was purchased from Millipore (Billerica, MA). Protease and phosphatase inhibitor cocktails were purchased from EMD Chemicals (San Diego, CA). Monovalent Fab fragments were prepared from mAb IV.3 and anti–platelet endothelial cell adhesion molecule 1 (anti–PECAM-1) mAb 235.1 using a Fab preparation kit from Pierce Biotechnology (Rockford, IL). The synthetic peptide collagen-related peptide (CRP) was synthesized by the Protein Chemistry Core Laboratory at the Blood Research Institute, BloodCenter of Wisconsin.

Mice were maintained in a facility free of known pathogens under the supervision of the Biomedical Resource Center at the Medical College of Wisconsin. Animal protocols were approved by the Medical College of Wisconsin Institutional Animal Care and Use Committee. FcγRIIa transgene mice were generated in the laboratory of S.E.M.¹⁴  All of the FcγRIIa mice studied were littermates on a C57BL/6J background, and were compared with WT C57BL/6J littermates. Mouse genotypes were determined by polymerase chain reaction amplification of genomic tail DNA, and presence of FcγRIIa expression was confirmed by flow cytometry and western blot analysis of platelet lysates.

All human blood samples were donated as approved by the institutional review board of the BloodCenter of Wisconsin, and informed consent was obtained in accordance with the Declaration of Helsinki. Blood from healthy volunteers free from medication for 2 weeks was collected into 90µM PPACK (D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone; EMD Chemicals, Philadelphia, PA). Mouse blood was drawn from the inferior vena cava of anesthetized mice into a syringe containing 3.8% sodium citrate (1/10 volume) or PPACK and heparin as anticoagulants.

Whole blood drawn into 3.8% sodium citrate (1/10 volume) was diluted 1:1 with modified Tyrode-HEPES (Tyrode–N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid) buffer (10mM HEPES [pH 7.4], 12mM NaHCO₃, 137mM NaCl, 2.7mM KCl, 5mM glucose, 0.25% bovine serum albumin [BSA]). Diluted whole blood was supplemented with 50 ng/mL prostaglandin E1 (PGE₁) and spun at 200g for 10 minutes. Platelet-rich plasma was collected, and after the addition of 50 ng/mL PGE₁, platelets were pelleted at 750g for 10 minutes. Platelets were washed once in Tyrode-HEPES buffer containing 50 ng/mL PGE₁ and 1mM EDTA, pH 7.4, and finally resuspended in Tyrode-HEPES to a final concentration of 3 × 10⁸/mL.

Eight-chamber glass tissue-culture slides (Becton Dickinson, Franklin Lakes, NJ) were coated overnight at 4°C with human fibrinogen (25 μg/mL), rinsed with phosphate-buffered saline, and blocked with 1% BSA (precleared using protein G beads) for 1 hour at room temperature. Washed platelets (200 µL) at a concentration of 2.5 × 10⁷/mL in Tyrode buffer supplemented with 1mM CaCl₂, 0.25 U/mL apyrase, and 10μM indomethacin were allowed to adhere to the immobilized fibrinogen for the indicated period of time. Nonadherent platelets were removed by gently rinsing 3 times. The remaining adherent platelets were fixed with 2% paraformaldehyde for 20 minutes, permeabilized for 5 minutes at room temperature with 0.1% Triton X-100 in 100mM Tris-Cl (pH 7.4), 150mM NaCl, 10mM EGTA, 5mM MgCl₂, containing 1× protease inhibitor cocktail, and stained with phalloidin tetramethylrhodamine isothiocyanate (1 µg/mL) for 60 minutes at room temperature in the dark. Samples were mounted in Vectashield mounting medium (Vector Laboratories, Burlingame, CA). Images were acquired with a SenSys camera (Photometrics, Tucson, AZ) using an Axioscop microscope (Carl Zeiss, Oberkochen, Germany) with a 100× oil-immersion lens (1.3 numeric aperture; Carl Zeiss) and analyzed using Metamorph software (Universal Imaging, Downingtown, PA). Statistical analysis of the area occupied by spread platelets was performed using a 2-tailed Student t test for unpaired samples. For biochemical analysis, platelets were incubated at 37°C for 15, 30, and 45 minutes on 10-cm tissue-culture dishes and lysed directly with 2× lysis buffer (30mM HEPES [pH 7.4], 300mM NaCl, 20mM EGTA, 0.2mM MgCl₂, 2% Triton X-100) containing 2× protease and phosphatase inhibitor cocktails, and subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis immunoblot analysis.

Platelet aggregation was carried out using a whole-blood lumi-aggregometer (Chrono-Log). Washed platelets (300 µL) in Tyrode-HEPES containing 1mM CaCl₂ were placed in a siliconized glass cuvette at 37°C with constant stirring at 1000 rpm. Platelet activation and secretion was initiated by addition of CRP, and platelets were allowed to aggregate for 10 minutes. For thrombin-induced clot retraction, washed platelets from WT or FcγRIIa transgenic mice were resuspended in plasma at a final concentration of 3 × 10⁸/mL. Thrombin (0.2 U/mL) and CaCl₂ (1mM) were added to initiate clot retraction.

Thrombus formation was evaluated by a whole-blood perfusion assay over fibrillar collagen (Chrono-Log Corp, Havertown, PA) or collagen/fibrinogen (Enzyme Research Laboratories, South Bend, IN) coated microchannels under both arterial and venous shear conditions (shear rates of 2888 s⁻¹ and 444 s⁻¹, respectively). Briefly, Vena8 FLUORO+ Biochips (Cellix Ltd, Dublin, Ireland) were coated overnight at 4°C with either fibrillar collagen (50 μg/mL) or fibrinogen (25 μg/mL) and blocked with Hank balancing salt solution containing 0.1% BSA. Whole blood from WT and FcγRIIa transgenic mice was anticoagulated with heparin and PPACK, labeled with mepacrine (CalBiochem, La Jolla, CA), and perfused over protein-coated microchannels. For function-blocking studies, blood was preincubated with mAb IV.3 or isotype control Fabs before perfusing through the microchannels. Images of platelet adhesion and thrombus formation were acquired by epifluorescence microscopy in real time at a frame rate of one frame per second. Thrombus formation was determined as the mean percentage of total area covered by thrombi and as the mean integrated fluorescence intensity per micrometer squared. Image analysis was performed using Metamorph software (Universal Imaging).

The studies on laser-induced thrombosis in the cremaster were performed as previously described.¹⁵  Briefly, FcγRIIa transgenic mice and WT mice were intraperitoneally injected with sodium pentobarbital (11 mg/kg; Abbott Laboratories, North Chicago, IL), and maintained under anesthesia with the same anesthetic delivered via catheterized jugular vein. Platelet adhesion was studied in arterioles of the cremaster muscle using an Olympus BX61WI microscope (Olympus, Tokyo, Japan or Center Valley, PA) with a 40×/0.8 numeric aperture water-immersion objective lens. Labeled antibodies specific for platelets and fibrin were introduced 5 minutes before injury, and laser injury was induced using an SRS NL100 Nitrogen Laser system (Photonic Instruments, St Charles, IL) as previously described.¹⁵  Data were collected for 2.5 minutes at 5 frames per second and then averaged at each time point. No more than 5 arteriole and 5 vein injuries were performed per mouse. Data were analyzed using Slidebook 5.0.0.17 software (Intelligent Imaging Innovations, Denver, CO).

Electrolytic-induced vascular venous injury was performed as previously described¹⁶  using 10- to 12-week-old mice. Platelets labeled with rhodamine 6G and fibrin, identified with a fibrin-specific labeled antibody, were co-injected 5 minutes before injury. Electrolytic injury was created in femoral veins using a steel microsurgical needle applied to the outer surface of each vessel, with a 1.5-V positive direct current delivered for 30 seconds. Vessels were illuminated uniformly with beam-expanded green (532 nm) and red (650 nm) laser lights. Fluorescent images were captured over a 60-minute interval with a low-light video camera attached to an operating microscope at ×100 magnification; video images were quantitatively evaluated every 2 minutes for analysis of relative fluorophore intensity (Image J software, Research Services Branch, National Institutes of Health, Bethesda, MD) within the thrombus zone and normalized for interanimal comparisons.¹⁶  Venous thrombosis data were evaluated for statistical significance at the 60-minute time point using the Student t test.

A growing number of blood and vascular cells have been shown to use ITAM-bearing adaptor proteins to transmit adhesion-initiated, integrin-mediated signals into cells.^12,17  In human platelets, integrins α2β1 and α6β1, which mediate adhesion to the extracellular matrix proteins collagen and laminin, respectively, use the GPVI/Fc receptor γ-chain (FcRγ) complex for this purpose.^18-20  The observation that a number of αIIbβ3-mediated outside-in signaling reactions can be inhibited by the anti-FcγRIIa function-blocking antibody, mAb IV.3,¹³  suggests that FcγRIIa might similarly function as an ITAM-bearing protein that transmits fibrinogen binding-induced activation signals into platelets.

To examine the importance of FcγRIIa/αIIbβ3 stimulus-response coupling in platelet function, we took advantage of the fact that, unlike their human counterparts, WT mouse platelets are FcγRIIa^neg,¹⁴  while transgenic mice harboring a human FcγRIIa transgene express this transmembrane receptor on the platelet surface at roughly the same density as exists on human platelets.¹⁴  We then subjected FcγRIIa^pos and FcγRIIa^neg mice to a number of in vitro and in vivo models of thrombosis and hemostasis. As shown in Figure 1A-B, expression of FcγRIIa conferred a small, but significant, improvement in the ability of murine platelets to spread on immobilized fibrinogen–an increase that was abolished by prior addition of monovalent Fab fragments of mAb IV.3. Compared with FcγRIIa^neg platelets, platelet expression of FcγRIIa significantly enhanced ligand binding-induced tyrosine phosphorylation of Syk and PLCγ2, 2 key players in integrin-mediated outside-in signal transduction (Figure 1C-D). As with human platelets,¹³  fibrinogen binding-induced tyrosine phosphorylation of FcγRIIa, Syk, and PLCγ2 in FcγRIIa^pos platelets could be inhibited by mAb IV.3 (Figure 1E-F). These data support the notion of FcγRIIa as an amplifier of adhesion-initiated signaling in platelets.

Previous loss-of-function studies have shown that blocking FcγRIIa with mAb IV.3 inhibits both aggregation of human platelets induced by low-dose platelet agonists²¹  as well as platelet-mediated retraction of a fibrin clot.¹³  To examine the hypothesis that addition of FcγRIIa to the platelet surface might conversely potentiate these reactions, we examined the relative ability of WT FcγRIIa^neg vs littermate transgenic FcγRIIa^pos murine platelets to aggregate in response to low-dose agonists. As shown in Figure 2A-B, expression of FcγRIIa enhanced platelet aggregation in response to low-dose CRP. Similar results were obtained using low-dose thrombin as the agonist (not shown). FcγRIIa^pos murine platelets exhibited markedly enhanced ability to mediate clot retraction (Figure 2C-D ). These gain-of-function studies demonstrate that FcγRIIa functions to amplify αIIbβ3-mediated outside-in signals into cells.

To examine the contribution of FcγRIIa to platelet function under conditions of flow, we analyzed platelet adhesion and thrombus formation on a fibrillar collagen-coated surface under conditions of both arterial and venous shear using a whole-blood microfluidic perfusion system. Platelets in whole blood were labeled with mepacrine, and accumulation of fluorescent platelets on collagen-coated (Figure 3) or collagen/fibrinogen-coated (Figure 4) surfaces was used to quantify adhesion and thrombus generation. As shown in Figure 3, thrombus formation was significantly enhanced in blood obtained from FcγRIIa^pos transgenic, compared with WT FcγRIIa^neg, mice at both high (Figure 3A-B) and low (Figure 3C-D) shear (2888 s⁻¹ and 444 s⁻¹, respectively). That the enhanced thrombus formation was due solely to the presence of FcγRIIa was demonstrated by addition of the FcγRIIa-specific blocking antibody, mAb IV.3, which reduced the level of thrombus formation by FcγRIIa^pos platelets to that observed for FcγRIIa^neg platelets (Figure 4A,C). IV.3 also reduced thrombus formation of human platelets (Figure 4B,D), which are also FcγRIIa^pos. Taken together with the results shown in Figure 1 obtained under static conditions, these data demonstrate that cooperative FcγRIIa/αIIbβ3 signaling plays a positive role in multiple in vitro models of platelet function.

While platelets are a well-known component of arterial thrombi, evidence exists that activated platelets also participate importantly in the initiation of fibrin generation in large veins.^16,22,23  To investigate whether the enhanced platelet activation conferred by FcγRIIa might be relevant under either arteriolar or venous conditions in vivo, we compared the extent to which FcγRIIa^neg and FcγRIIa^pos platelets contribute to these processes using 2 distinct murine models of thrombus formation. In the first, FcγRIIa^pos transgenic mice and their WT FcγRIIa^neg littermates were subjected to laser injury of their cremaster arterioles, and the rate and extent of platelet accumulation captured by video microscopy using previously described methods.¹⁵  As shown in Figure 5A, although the initial rate of platelet accumulation was similar in FcγRIIa^neg and FcγRIIa^pos mice, the overall extent of thrombus growth was significantly enhanced in FcγRIIa^pos mice. To examine the contribution of integrin→ITAM signaling in platelets to fibrin generation under low-shear, venous conditions, we subjected these 2 mouse strains to electrolytic injury of the femoral vein.¹⁶  While platelet accumulation was comparable between FcγRIIa^neg and FcγRIIa^pos platelets (not shown) fibrin accumulation was significantly enhanced in FcγRIIa^pos mice (Figure 5B), perhaps due to a greater degree of platelet activation in the FcγRIIa^pos mice leading to increased granule release, phosphatidylserine exposure, etc, in the fibrin-dominated venous thrombosis model. Taken together, these data support a role for FcγRIIa as a physiologically important amplifier of both arterial and venous thrombosis.

Although FcγRIIa binds with high affinity to the Fc region of aggregated immunoglobulin G (IgG), or to the Fc region of antigen-bound IgG immune complexes, it does not interact at all with monomeric IgG,²⁴  the species that is present at 10 mg/mL in normal plasma. Platelets contain an additional pool of IgG in their α-granules,²⁵  however, it is neither antigen-bound nor aggregated (H.Z. and P.J.N., unpublished observations, July 2012). Nevertheless, to eliminate the possibility that IgG/FcγRIIa interactions are contributing to, or responsible for, the amplification effects that we ascribe to integrin/ITAM connections, we crossed generated FcγRIIa^TGN mice with μMT mice that lack B cells and all forms of immunoglobulin.²⁶  As shown in supplemental Figure 1, platelets from these mice continue to both spread on immobilized fibrinogen and form thrombi under conditions of flow significantly better than their FcγRIIa^neg counterparts and comparable with platelets from FcγRIIa^TGN, IgG-containing mice. These data demonstrate that IgG does not contribute to the enhancement conferred by FcγRIIa on αIIbβ3-mediated outside-in signaling.

Thrombus formation is a multistep process that involves GPIb-mediated tethering to von Willebrand factor presented on exposed collagen fibrils, integrin α2β1- and α6β1-mediated adhesion of platelets to collagen and laminin, activation, release, and generation of secondary platelet agonists that activate cell-surface G-protein–coupled receptors, conformational changes in the major platelet integrin αIIbβ3, and finally binding of the platelet-platelet bridging molecule, fibrinogen, to the platelet surface.²⁷  Once the exposed extracellular matrix is coated with the primary layer of platelets, however, continued accumulation of platelets to form a stable thrombus requires sustained agonist-induced stimulation. Although the roles of soluble agonists like adenosine 5′-diphosphate, thromboxane A₂, and thrombin in this process are well-appreciated and understood, the relative importance/contribution to the growth of a platelet thrombus of αIIbβ3-mediated outside-in signal transduction has been a matter of debate.

Although binding of soluble fibrinogen to αIIbβ3 requires that the integrin be transformed into an active, ligand-binding competent conformation,²⁸  adhesion of resting platelets to immobilized fibrinogen, first described in the late 1960s,²⁹  is unique in that it does not require activation of platelets^30,31  and involves distinct adhesive sequences on the fibrinogen molecule.³¹  Fibrinogen, immobilized on the surface of already adherent platelets, likely represents one of the more abundant proteins that circulating platelets encounter as they approach a growing thrombus, and the adhesion-initiated signaling processes that ensue are composed of a series of downstream biochemical reactions that can be broken down into a series of temporally and spatially regulated events. Activation of integrin-associated SFKs,^7,32  triggered either by microclustering of integrin receptors following the binding of a multivalent ligand,⁵  or if thrombin is present, catalyzed by β3-associated Gi₁₃,⁶  is among the earliest detectable events. Soon afterwards SFK-mediated phosphorylation of multiple cellular targets, promotes the assembly of a nascent signaling complex at the inner face of the platelet plasma membrane that relays signals to the actin cytoskeleton, thereby driving the dynamic reorganizational events that lead to platelet spreading.³³  The nonreceptor tyrosine kinase Syk is among the earliest targets of integrin-associated SFKs,^10,34,35  and is recruited to the nascent αIIbβ3 signaling complex shortly after ligand binding.³⁴  The molecular nature of the physical and functional linkage to αIIbβ3, however, has only been clarified in the last decade. Thus, although Syk has been shown to associate directly with the integrin β3 tail in transfected cell lines,¹¹  Abtahian et al¹²  found that the SH2 domains of Syk were required for platelets to spread efficiently on immobilized fibrinogen. The finding by Boylan et al¹³  that the ITAM-bearing transmembrane adaptor molecule FcγRIIa efficiently recruits Syk to the nascent αIIbβ3 signaling complex—resulting in robust functional coupling between ligand binding and downstream events like platelet spreading and αIIbβ3–mediated clot retraction—added an important component to the outside-in signaling circuit. The present work extends these observations by demonstrating the physiological advantage imparted by FcγRIIa to platelet function under conditions of flow, both in vitro and in vivo.

The signaling circuit in platelets defined by ligand binding to αIIbβ3→integrin-associated SFKs→FcγRIIa ITAM phosphorylation→Syk recruitment and activation represents only one of a growing number of examples in which a cell-surface integrin enlists an ITAM-bearing protein to serve as a conduit for signal transduction into the cell. Though ITAM-containing adaptors have been known for many years to coordinate assembly and localization of macromolecular complexes following ligation of immunoreceptors in B cells, T cells, natural killer cells, and mast cells,³⁶  there were hints of cross-talk between integrin and ITAM signaling as early as 1996, when it was shown that the ITAMs of the FcRγ chain become phosphorylated when platelets adhere to collagen.¹⁸  That integrins might as a rule, rather than as an exception, cooperate with ITAM-bearing proteins to transmit adhesion-initiated signals into the cell, however, was not appreciated until 7 years ago, when Abtahian et al¹²  found that intact SH2-domains of Syk were required for murine platelets to be able to spread efficiently on immobilized fibrinogen, followed by the report of Mócsai et al¹⁷  that β2 integrins on mouse neutrophils and macrophages use the FcRγ chain and/or the related ITAM-bearing adaptor molecule, DAP-12 for outside-in signaling. Other examples of integrin/ITAM connections continue to accumulate, including a report by Zou et al³⁷  that αvβ3 in murine osteoclasts use the FcRγ-chain and DAP-12 to regulate bone resorption, and a more recent report by March et al³⁸  that the integrin, αLβ2 (LFA-1) on natural killer cells enlists the T-cell receptor ξ-chain to effect granule polarization and degranulation. The entirety of integrin receptors that use ITAMs to co-opt Syk for outside-in signal transduction remains to be defined.

The spatial relationship between the FcγRIIa and the αIIbβ3 integrin complex that allows for their functional coupling (Figure 6) remains obscure. Despite there being only 2000 to 4000 FcγRIIa molecules on the surface of a platelet, FcγRIIa has been reported to be proximal to αIIbβ3,³⁹  the GPIb complex,⁴⁰  and PECAM-1.⁴¹  Were it to be highly mobile, FcγRIIa might be able to subserve so many different, and differently abundant, cell-surface receptors. Alternatively, subpopulations of FcγRIIa,^21,42  GPIb,⁴²  and PECAM-1⁴³  have all been reported to be enriched in lipid rafts. There is no evidence, however, that αIIbβ3 ever becomes raft localized. They are likely to need to be proximal to each other at some point for signal amplification to occur. However, the FcγRIIa-mediated amplification of Syk phosphorylation that occurs when platelets encounter immobilized fibrinogen (Figure 1) does not occur when fibrinogen binds in such a way as to minimize multivalent ligand-binding–induced transactivation of integrin-associated SFKs: that is, when fibrinogen binds to activated platelets in the absence of stirring (supplemental Figure 2). Whether FcγRIIa becomes physically associated with αIIbβ3 to amplify αIIbβ3 outside-in signaling and the nature of their physical association, if any, is unknown.

In addition to the functional coupling between αIIbβ3 and FcγRIIa described herein, focal adhesion kinase (FAK) pp125^FAK has also been shown to play a key role in mediating outside-in signaling downstream of integrin engagement. A 125-kDa nonreceptor protein-tyrosine kinase that was first described in 1992, FAK has been shown in platelets to play a prominent role in integrin-mediated signaling and thrombus stability,⁴⁴  functions that have been ascribed (at least in part) to its ability to interact with the Arp2/3 complex and orchestrate actin assembly during the process of filopodial and lamellipodial protrusion.⁴⁵  Ligand binding to αIIbβ3 triggers autophosphorylation of FAK Y₃₉₇³³  in a reaction that some investigators find dependent on costimulation with other agonist receptors. FAK Y₃₉₇, in turn, is able to recruit and activate nearby Src-family kinases, which in turn phosphorylate the kinase domain of FAK, further activating it.⁴⁶  Curiously, the inability of platelets missing components of the integrin/ITAM signaling pathway to spread on fibrinogen is phenocopied in FAK-deficient murine platelets.⁴⁴  Whether the integrin/ITAM and integrin/FAK pathways intersect to mediate outside-in signaling events leading to platelet spreading and thrombus stability, however, is not yet known.

Finally, while the development of effective fibrinogen receptor antagonists has been a major advance in the management of coronary artery diseases,⁴⁷  these agents have been shown to increase the incidence of clinically significant thrombocytopenia and bleeding⁴⁸  and a few reports suggest that thrombosis might also be an additional rare complication of eptifibatide therapy.⁴⁹  Improved understanding of the molecular components that mediate αIIbβ3 outside-in signaling has the potential to lead to identification of new therapeutic targets like FcγRIIa, the intervention of which might be effective in inhibiting the later stages of thrombus formation (ie, recruitment of resting platelets to fibrinogen-coated platelets) without affecting initial platelet adhesion or repair at sites of vascular damage.

The authors thank Dr Jimmy Crockett for development of the microfluidic system during the early stages of this investigation, and Dr Demin Wang for helpful discussions and for providing the µMT mice.

This work was supported by grants HL-44612 (P.J.N.) and P01HL110860 (M.P.) from the Heart, Lung, and Blood Institute of the National Institutes of Health.

National Institutes of Health

Contribution: H.Z. designed and performed research, analyzed results, and wrote the manuscript; L.R., V.H., C.G., B.B., and B.C.C. performed research and analyzed results; D.K.N. analyzed results; S.E.M. contributed transgenic mice lines and analyzed results; M.P. designed research and analyzed results; and P.J.N. designed research, analyzed results, and wrote the manuscript.

Conflict-of-interest disclosure: P.J.N. is a member of the scientific advisory boards of the New York Blood Center, the Puget Sound Blood Center, and Children’s Hospital of Boston. The remaining authors declare no competing financial interests.

Correspondence: Peter J. Newman, PhD, BloodCenter of Wisconsin, Blood Research Institute, 638 North 18th St, Milwaukee, WI 53201; e-mail: peter.newman@bcw.edu.