Integrin αIIbβ3 mediates platelet adhesion and aggregation and plays a crucial role in thrombosis and hemostasis. αIIbβ3 is expressed in a low affinity state on resting platelets. Upon platelet activation, αIIbβ3 shifts to a high affinity conformation that efficiently binds its ligands. On human platelets, the high affinity conformation of αIIbβ3 is detected by the monoclonal antibody (mAb), PAC-1. However, a reagent with binding specificity to high affinity mouse αIIbβ3 has not been described so far.
A novel rat mAb directed against mouse αIIbβ3 (JON/A) was generated and characterized. JON/A was conjugated with fluorescein isothiocyanate (JON/AFITC) or with R-phycoerythrin (JON/APE) and used for flow cytometric analysis of mouse platelets.
Although JON/AFITC bound to resting and activated platelets, virtually no binding of the larger JON/APE to resting platelets was detectable. However, strong binding of JON/APE occurred on platelet activation in a dose-dependent manner. Binding of JON/APE required extracellular free calcium and was irreversible, thereby stabilizing the high affinity conformation of αIIbβ3.
Integrin αIIbβ3 mediates platelet adhesion and aggregation and plays a central role in thrombosis and hemostasis. In resting platelets, αIIbβ3 is expressed on the surface membrane and in intracellular compartments (1, 2). Inside-out signaling triggered by agonists such as thrombin, collagen, collagen-related peptide (CRP), or adenosine diphosphate (ADP) induces a cation-dependent transformation of receptors from a low affinity to a high affinity state (3). Additionally, the number of surface-expressed receptors is increased by the translocation of internal pools to the plasma membrane (1, 4). Soluble, multivalent ligands, most importantly fibrinogen (Fg) and von Willebrand factor (vWF), bind to activated αIIbβ3 and crosslink adjacent platelets leading to thrombus formation (5). The activation of integrin αIIbβ3 is the final common pathway of platelet activation. Therefore, the direct assessment of integrin αIIbβ3 activation is critical for accurate studies on platelet function.
On human platelets, the high-affinity conformation of integrin αIIbβ3 is detected specifically by the monoclonal antibody (mAb), PAC-1. PAC-1 is a ligand-mimetic antibody, i.e., it competes with the ligand for binding to the receptor. The binding of PAC-1 is cation dependent and is blocked by RGD-containing peptides. In flow cytometry, PAC-1 discriminated between the resting and the activated form of the receptor as platelet activation is a prerequisite for its binding (6, 7). PAC-1 has become a widely used tool for studies on αIIbβ3 activation on human platelets.
Recent advances in the manipulation of the mouse genome have contributed enormously to a better understanding of platelet function in hemostasis and thrombosis. Mouse integrin αIIbβ3 shares high homology with its human counterpart and also plays a crucial role in platelet adhesion and aggregation (8, 9). However, there are no reagents that allow direct assessment of the activated mouse αIIbβ3 integrin. Therefore, the amount of surface-bound fluorophore-labeled Fg is used most frequently as a measure of αIIbβ3 activation. This method, however, has some limitations. First, the labeled Fg competes with plasma and platelet-derived Fg for binding to αIIbβ3. Second, activation of αIIbβ3 occurs transiently (as in the case of ADP-induced platelet activation), leading to the loss of bound Fg before analysis. Third, fibrin(ogen) may polymerize on the surface of activated platelets, particularly in the presence of thrombin, yielding inaccurate results.
We describe the development of a fluorescent derivative of a novel mAb against mouse αIIbβ3 (JON/APE) that allows the detection of the activated integrin in flow cytometry. Binding of JON/APE to activated mouse αIIbβ3 is irreversible and the conjugate can be used for studies on whole blood. The use of JON/APE in flow cytometry provides a rapid and easy-to-handle method for studies on activation of this major platelet receptor in mice.
Materials and Methods
Specific pathogen-free mice (NMRI, BALB/c), 6–10 weeks of age, were obtained from Charles River (Sulzfeld, Germany) and kept in our animal facilities.
Succinimidyl 4-(N-maleimidomethyl) cyclohexane-1-carboxylate (SMCC) and EZ-Link sulfo-NHS-LC-biotin (both from Pierce, United Kingdom), ECL (Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom), R-phycoerythrin (PE; Europa, United Kingdom), ADP, mouse Fg, fluorescein isothiocyanate (FITC), phorbol 12-myristate 13-acetate (PMA), N-ethylmaleimide (NEM; all from Sigma, Deisenhofen, Germany), thrombin (Roche Diagnostics, Mannheim, Germany), and streptavidin-horseradish peroxidase (HRP; Dako, Glostrup, Denmark) were purchased.
The rat anti-mouse P-selectin mAb, RB40.34, was provided by D. Vestweber (Muenster, Germany) and modified in our laboratories. MWReg30 and p0p6 antibodies were generated, produced, and modified in our laboratories: MWReg30 (anti- αIIbβ3, IgG1; ref. 10) and p0p6 (anti-GPIX, IgG1; ref. 11).
Mice were bled under ether anesthesia from the retroorbital plexus. The blood was collected in a tube containing 10% (v/v) 7.5 U/ml heparin. Platelet-rich plasma (PRP) was obtained by centrifugation at 300 × g for 10 min at room temperature (RT). PRP was centrifuged at 1,000 × g in the presence of prostacyclin (0.1 μg/ml) for 7 min at RT. After two washing steps, pelleted platelets were resuspended in modified Tyrodes-HEPES buffer (137 mM NaCl, 0.3 mM Na2HPO4, 2 mM KCl, 12 mM NaHCO3, 5 mM HEPES, 5 mM glucose, 1 mM CaCl2, pH 7.3) containing 0.35% bovine serum albumin (BSA) at a density of 2 × 105 platelets per microliter in the same buffer. Platelet preparations were incubated for 30 min at 37°C in the presence of 0.02 U/ml of the ADP scavenger apyrase, a concentration sufficient to prevent desensitization of platelet ADP receptors during storage.
Production of mAbs
Female Wistar rats, 6–8 weeks of age, were immunized repeatedly with affinity-purified mouse αIIbβ3. Purification was performed according to standard methods. Briefly, MWReg30 (anti-αIIbβ3; 10) was coupled covalently to N-hydroxysuccinimide (NHS)–activated Sepharose. With this affinity column, αIIbβ3 was purified from an NP-40 lysate of mouse platelets. mAbs were produced as described previously (12). The positive hybridoma (JON/A) was subcloned twice prior to large-scale production. JON/A bound to mouse platelets, but not to human platelets. Labeling of purified mAbs and Fg with FITC or R-PE was performed as described previously (12).
Modification of Antibodies
Affinity-purified antibodies were fluoresceinated to a fluorescein-to-protein ratio of approximately 3:1 by standard methods with FITC and separated from free FITC by gel filtration on a PD-10 column (Pharmacia, Uppsala, Sweden). Labeling of purified mAbs with R-PE (molecular weight, 240 kD) at a ratio of 1:2 was performed according to standard protocols, resulting in conjugates of about 630 kD. Briefly, 330 μg of SMCC was added to 3.0 mg PE while vortexing and incubated for 60 min at RT. PE-SMCC conjugates were separated from free compounds on a PD-10 column. Purified mAbs (1.5 mg) were reduced by adding 20 mM dithioerythritol (DTE) while vortexing. After a 30-min incubation at RT, the reduced mAb was passed over a PD-10 column and the protein concentrations were determined colorimetrically. Pooled antibody was added to PE-SMCC and incubated for 60 min at RT, followed by addition of 50 μg NEM to block free sulfhydryls. The product was exchanged over a PD-10 column into storage buffer and the purity was tested by flow cytometry. The molecular weights of the conjugates were checked by gel electrophoresis.
Immunoprecipitation was performed as described previously (13). Briefly, 10 μg mAb was added with 25 μl of protein G-Sepharose to a lysate of biotinylated glycoproteins and precipitation took place overnight at 4°C. Samples were separated by sodium dodecylsulfate polyacrylamide gel electrophoresis on a 9–15% gradient (SDS-PAGE) under nonreducing conditions and transferred to a polyvinylidene difluoride (PVDF) membrane. Biotinylated proteins were visualized by streptavidin-HRP/ECL. The αIIbβ3-depleted lysate was generated by two preclearing cycles with MWReg30 (10 μg) and 25 μl protein G-Sepharose for 6 h each at 4°C.
Diluted whole blood (1/20 in modified Tyrodes-HEPES buffer) or washed platelets were stimulated as indicated in the figure legends. Incubation with fluorophore-conjugated mAbs (5 μg/ml) or FgFITC was performed for 10 min at 37°C and the samples were analyzed immediately. For all measurements, a FACSCalibur flow cytometer (BD-Biosciences, Heidelberg, Germany) equipped with an argon laser, a diode laser, and four photomultipliers (PMT) for fluorescence detection was used. In our experiments, a single laser excitation at 488 nm was applied to excite FITC or R-PE–labeled antibodies. The optics for green fluorescence FITC detection and for red fluorescence R-PE detection included a 530/30 (FL1) and a standard 585/42 (FL2) band pass filter, respectively. All measurements were performed with CellQuest Software (BD-Biosciences) and data from 10,000 cells per sample were stored in list mode format on a Macintosh G3 computer. Data evaluation was done with the same software using multiparameter gating. The gated data were used to define the levels of FITC or R-PE fluorescence and were illustrated in a one-parameter histogram. A minimum of six independent measurements were run for each experimental setup. As negative control, platelets were stained with FITC or PE-conjugated isotype-matched control antibodies.
To determine platelet aggregation, light transmission was measured using PRP (200 μl with 0.5 × 106 platelets per microliter). Transmission was recorded on a Fibrintimer four-channel aggregometer (APACT Laborgeräte und Analysensysteme, Hamburg, Germany) and expressed as arbitrary units with 100% transmission adjusted with plasma. Platelet aggregation was induced by the addition of ADP (10 μM) or CRP (2 μg/ml).
JON/A Binds to and Functionally Block sMurine αIIbβ3
A newly generated mAb against mouse αIIbβ3 (JON/A, IgG2b) was used in the current study. JON/A, like the well-established anti-αIIbβ3 mAb MWReg30 (10), precipitated αIIbβ3 from the lysate of surface-biotinylated platelets (Fig. 1a). JON/A inhibited platelet aggregation induced by the strong GPVI-specific agonist CRP (2 μg/ml; 14) and the weak agonist ADP (10 μM), demonstrating that the mAb binds to an epitope critical for αIIbβ3 function (Fig. 1b). Flow cytometric analysis demonstrated that JON/A inhibited Fg binding to CRP-activated platelets in a dose-dependent manner, but it did not displace receptor-bound Fg (Fig. 1c). Bound Fg, on the other hand, did not influence binding of JON/A (Fig. 1d), demonstrating that JON/A and Fg do not compete for the same epitope on αIIbβ3.
JON/APE Binds to the High-Affinity Conformation of αIIbβ3
For flow cytometric studies on JON/A binding to αIIbβ3, the mAb was conjugated with FITC and R-PE. The sizes of the two fluorophore conjugates differ significantly: JON/AFITC (about 150 kD) and JON/APE (about 630 kD). As a control, MWReg30 was labeled with FITC and PE at the same mAb-to-fluorophore ratio as JON/A. On resting washed platelets, JON/AFITC, MWReg30FITC, and MWReg30PE displayed maximal binding at concentrations higher than 3 μg/ml (5 min, 37°C, not shown). To definitively reach saturating concentrations, all conjugates were used at 5 μg/ml (10 min, 37°C) in the current study. Due to the translocation of internal receptor pools from the surface-connected canalicular system (SCCS) and the α-granules (determined by surface expression of P-selectin, not shown) to the outer membrane, an increase of surface-expressed αIIbβ3 was detectable, which reached a maximum at 10 μM ADP (about 1.3-fold), 1 μg/ml CRP (about 1.8-fold; Fig. 2), and 0.05 U/ml thrombin (about 1.9-fold, not shown).
In contrast to the other conjugates, virtually no binding of JON/APE to resting platelets was detected, suggesting that the access of JON/A was impeded sterically by the large R-PE molecules. Upon platelet activation, however, a marked increase in JON/APE binding was detectable (Fig. 2). Binding of JON/APE occurred in a dose-dependent manner and reached a maximum at agonist concentrations of 1 μg/ml CRP (Fig. 3a), 10 μM ADP (Fig. 3b), or 0.05 U/ml thrombin (not shown). These data strongly suggested that conformational changes of the receptor required for binding of Fg (5, 15) also allow binding of JON/APE.
It is well known that extracellular Ca2+ is required for the activation-dependent conformational changes of αIIbβ3 (16). Therefore, we analyzed binding of JON/APE to activated platelets in the presence of varying concentrations of extracellular free calcium. As shown in Figure 3c, half-maximal binding of JON/APE to CRP-activated platelets (1 μg/ml) was observed at concentrations of about 200 μM CaCl2 and binding was maximum at 1 mM CaCl2. Similar results were obtained with ADP and thrombin-activated platelets (not shown). No binding of JON/APE was detectable in the presence of EDTA. However, bound JON/APE resisted dissociation by addition of EDTA (Fig. 3d), suggesting that the conjugate stabilizes the activated state of αIIbβ3.
Detection of Activated αIIbβ3 in Whole Blood With JON/APE
Whole blood flow cytometry is a powerful method to detect activation-dependent changes on the surface of single platelets in very small volumes of blood. This is of particular interest in mice, in which the total blood volume is limited. However, detection of activated αIIbβ3 in whole blood may be affected by the presence of high concentrations of plasma Fg. We tested whether JON/APE can be used for such studies. For that purpose, CRP (1 μg/ml) was added to washed platelets or diluted whole blood for 10 min and the samples were stained subsequently with JON/APE in combination with p0p6FITC (anti-GPIX; 11). Platelets in whole blood were identified by their light scatter characteristics (forward scatter [FSC]) and Fl1 positivity (Fig. 4a). As shown in Figure 4b, activation of platelets in diluted whole blood induced binding of JON/APE to a similar extent as observed on washed platelets, confirming that the presence of plasma Fg does not significantly affect binding of the conjugate.
Detection of Time-Dependent Changes in αIIbβ3 Activation by JON/APE
Activation of αIIbβ3 is a rapid process that can be reversible, particularly when induced by weak agonists such as ADP. We tested whether JON/APE can be used to detect time-dependent changes in the activation state of αIIbβ3 in flow cytometry that correlate with the formation of (ir)reversible platelet aggregates under stirring conditions. Three agonists were used for these studies: ADP (10 μM), which induces rapid but reversible aggregation; ADP in combination with the GPVI-specific mAb, JAQ1 (10 μg/ml), which induces stronger activation of αIIbβ3 and prolonged aggregation compared with ADP alone (17); and the strong agonist CRP (2 μg/ml), which induces the formation of irreversible aggregates. To detect time-dependent changes in the activation state of αIIbβ3, JON/APE was added to platelets at different time points after stimulation. As shown in Figure 5, all agonists caused maximum binding of JON/APE at 1 min. However, marked differences in the extent and reversibility of αIIbβ3 activation were observed. For ADP alone, the staining intensity progressively decreased with time reaching almost basal values at 6 min (Fig. 5a). Activation of αIIbβ3 in response to the combination of ADP and JAQ1 was increased and was prolonged markedly compared with ADP alone (Fig. 5b). In response to CRP, activation was strongest and completely irreversible (Fig. 5c). Thus, (ir)reversibility of αIIbβ3 activation as determined by binding of JON/APE tightly correlates with the stability of aggregates as seen in standard aggregometry.
In the current study, we describe a fluorescent derivative of a novel mAb directed against mouse αIIbβ3 (JON/APE) that specifically binds to the activated receptor and therefore resembles PAC-1 binding to the human receptor. However, in contrast to PAC-1, which is a ligand-mimetic antibody that competes with Fg for binding to αIIbβ3 (6), binding of JON/APE is not affected by the presence of plasma Fg. When platelets are preincubated with JON/A, however, Fg binding and platelet aggregation in response to different agonists are inhibited (Fig. 1). JON/A may affect Fg binding by allosteric inhibition of the receptor as proposed by the model of dual interacting ligand binding sites on this integrin (18). However, because JON/A does not replace receptor-bound Fg, this hypothesis is not favorable. The inhibitory effect of JON/A on Fg binding may be based on steric effects resulting from binding to an epitope located in close vicinity to the Fg-binding pocket on αIIbβ3.
To study JON/A binding to murine αIIbβ3 in flow cytometry, we used two fluorescent derivatives of the mAb that markedly differ in size: the 150-kD (approximately) JON/AFITC and the 630-kD (approximately) JON/APE. JON/AFITC normally bound to resting and activated receptors when compared with staining with the FITC derivative of the well-established anti-αIIbβ3 mAb MWReg30 (10). Unexpectedly, a profound difference in binding was observed between the two PE-coupled mAbs. Although MWReg30PE rapidly bound to resting and activated platelets, virtually no binding of JON/APE to resting platelets was detected (Fig. 2). This finding strongly suggests that the increased size of the PE derivative (compared with unlabeled/ FITC-labeled JON/A) sterically impedes its binding to αIIbβ3 on resting platelets. This hypothesis is supported by the observation that the intermediate-sized PE-labeled Fab fragments of JON/A (about 290 kD) yielded a weak but significant staining of resting platelets (data not shown). On the other hand, it is very unlikely that PE conjugation of JON/A affects the antigen recognition sites of the antibody. This is because coupling occurs between the sulfhydryl-reactive maleimide group of the chemical crosslinker SMCC (coupled to PE) and free sulfhydryl groups in the hinge region of the partially reduced antibody. Therefore, the increase in the molecular size of JON/A directly influences its binding properties to resting αIIbβ3 integrin.
Upon cellular activation, however, a strong, calcium-dependent increase in JON/APE (intact IgG) binding to platelets was detectable. This strongly suggests that the conformational changes of the receptor required for binding of Fg (5, 15, 16) also allow binding of JON/APE. Therefore, JON/AFITC and JON/APE provide the first example of fluorescent antibody derivatives with identical antigenic specificitiy that allow the discrimination between the resting and the activated state of an integrin.
One of the major advantages of mAbs like PAC-1 is that they can be used to detect activation of αIIbβ3 in whole blood flow cytometry (7). This approach requires only microliter volumes of blood and significantly lowers the risk of unintentional platelet activation induced by platelet preparation. However, whole blood flow cytometric detection of activated αIIbβ3 can only be successful if the analysis is not affected by plasma Fg. For PAC-1, which competes with Fg for binding to αIIbβ3, it has been proposed that this may be due to the higher affinity of the mAb compared with the ligand (7). In our experiments with JON/APE, no significant difference in binding to αIIbβ3 was observed when platelets were activated (CRP) in the presence or absence of plasma Fg, even when the antibody derivative was added to platelets 10 min after stimulation of the cells (Fig. 4).
Fg binding to αIIbβ3 is a reversible process. However, the extent of the reversibility depends on the agonist that activates the platelet. JON/APE bound to αIIbβ3 stabilizes the activation-dependent conformation of the receptor and does not dislocate from the receptor, even when EDTA is added (Fig. 3d). The stabilizing effect of JON/APE was confirmed in time course experiments. These experiments tested whether JON/APE can be used to detect changes in the activation state of αIIbβ3 in response to the weak agonist ADP, ADP in combination with the anti-GPVI mAb JAQ1, and in response to the strong agonist CRP (Fig. 5). For all agonists, maximum binding of JON/APE was detected already 1 min upon stimulation. However, although no significant change in JON/APE binding to CRP-activated platelets was detectable for 15 min (Fig. 5c), a rapid decrease was observed on ADP-activated platelets, reaching almost basal values at 6 min (Fig. 5a). For platelets activated with ADP in combination with JAQ1, αIIbβ3 activation was prolonged but partly reversible (Fig. 5b). The observed time-dependent changes in the activation state of αIIbβ3 in response to agonists of different strength correlate well with the stability of aggregates under stirring conditions. These findings demonstrate that JON/APE firmly binds to activated αIIbβ3 and strongly suggest that it stabilizes the high-affinity conformation of the receptor.
In summary, we report the generation of a fluorescent derivative of a mAb against mouse αIIbβ3 (JON/APE) that specifically binds to the activated form of the receptor. JON/APE provides an easy-to-handle reagent for direct assessment of αIIbβ3 function in mice and may become a very important tool for future studies on the regulation of this dominant platelet integrin.
This work was supported by Grant Ni 556/2-1 (to B.N.) from the Deutsche Forschungsgemeinschaft and the BAYER AG. We thank K. Rackebrandt for excellent technical assistance and U. Barnfred for constant support throughout the study.