Inhibition of ADP-induced platelet responses by covalent modification of aggregin, a putative ADP receptor, by 8-(4-bromo-2,3-dioxobutylthio)ADP.

ADP is an important platelet agonist which initiates platelet shape change, aggregation, exposure of fibrinogen receptors, and calcium mobilization. Because of the limitations of previously used affinity analogs and photolabeling studies as well as controversies surrounding the identity of an ADP receptor on platelets, we have used an affinity label capable of alkylating a putative exofacial receptor on platelets. We now report that 8-(4-bromo-2,3-dioxobutylthio)adenosine-5′-diphosphate (8-BDB-TADP), which is an analog of the natural ligand ADP, blocked ADP-induced platelet shape change, aggregation, exposure of fibrinogen-binding sites, secretion, and calcium mobilization. Following modification by 8-BDB-TADP, the rates of aggregation of platelets induced by thrombin, a calcium ionophore (A23187) or a stimulator of protein kinase C (phorbol myristate acetate) were minimally affected. However, the 8-BDB-TADP-modified platelets exhibited decreased rates of aggregation in response to ADP, as well as collagen and a thromboxane mimetic (U46619), both of which partially require ADP. Autoradiograms of the gels obtained by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of solubilized platelets modified by either [β-32P]8-BDB-TADP, or 8-BDB-TADP and NaB[3H]4 showed the presence of a single radiolabeled protein band at 100 kDa. The intensity of this band was reduced when platelets were preincubated with ADP, ATP, and 8-bromo-ADP prior to labeling by the radioactive 8-BDB-TADP. The results show that 8-BDB-TADP selectively and covalently labeled aggregin (100 kDa), a putative ADP receptor, resulting in a loss of ADP-induced platelet responses.

ADP is an important platelet agonist which initiates platelet shape change, aggregation, exposure of fibrinogen receptors, and calcium mobilization. Because of the limitations of previously used affinity analogs and photolabeling studies as well as controversies surrounding the identity of an ADP receptor on platelets, we have used an affinity label capable of alkylating a putative exofacial receptor on platelets. We now report that 8-(4-bromo-2,3dioxobutylthio)adenosine-5-diphosphate (8-BDB-TADP), which is an analog of the natural ligand ADP, blocked ADP-induced platelet shape change, aggregation, exposure of fibrinogen-binding sites, secretion, and calcium mobilization. Following modification by 8-BDB-TADP, the rates of aggregation of platelets induced by thrombin, a calcium ionophore (A23187) or a stimulator of protein kinase C (phorbol myristate acetate) were minimally affected. However, the 8-BDB-TADP-modified platelets exhibited decreased rates of aggregation in response to ADP, as well as collagen and a thromboxane mimetic (U46619), both of which partially require ADP. Autoradiograms of the gels obtained by sodium dodecyl sulfatepolyacrylamide gel electrophoresis of solubilized platelets modified by either [␤- 32 4 showed the presence of a single radiolabeled protein band at 100 kDa. The intensity of this band was reduced when platelets were preincubated with ADP, ATP, and 8-bromo-ADP prior to labeling by the radioactive 8-BDB-TADP. The results show that 8-BDB-TADP selectively and covalently labeled aggregin (100 kDa), a putative ADP receptor, resulting in a loss of ADPinduced platelet responses.

P]8-BDB-TADP, or 8-BDB-TADP and NaB[ 3 H]
Although ADP (Fig. 1A) is one of the earliest known agonists for platelet activation (1,2), the identity of the receptor is still uncertain. ADP receptors on platelets and magakaryocytes are unique (3); they constitute a subtype P 2T of a class of P 2 purinergic receptors where ADP is the strongest agonist and ATP is an antagonist (4). The presence of a P 2T receptor on human erythroleukemia cells has recently been demonstrated (5). Previous work from our laboratory demonstrated that 5Ј-p-fluorosulfonylbenzoyladenosine (FSBA) 1 (Fig. 1B), an ADP affinity label, blocked ADP-induced platelet shape change (6), aggregation, and exposure of fibrinogen-binding sites (7) with concomitant covalent modification of a single surface protein, aggregin (100 kDa), an ADP receptor on platelet surface (8,9). Covalent modification of platelets by FSBA was shown to block platelet aggregation induced by U46619 (a thromboxane mimetic) (10) and collagen (11), suggesting that aggregation induced by these agonists, in part, depends on interaction of ADP with aggregin.
Other investigators have proposed different candidates for a putative ADP rceceptor on platelet surface. Greco et al. (12) suggested that there is an ADP-binding site on platelet glycoprotein IIb (GPIIb) based on results obtained by photolabeling of platelets by [ 35 S][adenosine-5Ј-(1-thiotriphosphate)], ATP-␣-S. The same group has recently proposed that an ADP-binding site does not reside on GPIIb but is in close proximity to it (13). However, only ATP (but not ADP analogs) inhibits binding of fibrinogen to its receptor, the GPIIb-IIIa complex (12,14). Patients with thromboasthenia, lacking this receptor, exhibit normal ADP-induced platelet shape change and mobilization of intracellular Ca 2ϩ (15). Although 2-azido-ADP (2-N 3 -ADP) was shown to block ADP-induced aggregation of platelets (16), no covalently labeled ADP-binding protein was detected by gel electrophoresis. Photolabeling by 2-(p-azidophenyl)-ethylthioadenosine-5Ј-diphosphate (AzPET-ADP) was shown to label several proteins on the platelet surface (17). The fact that labeling of one of these proteins (43 kDa) was reduced in the presence of ADP led the authors to conclude that the 43-kDa protein might be an ADP receptor on platelet surface.
Although photoaffinity labeling has provided useful information concerning the structure-function relationship of both purified enzymes (18) and functional proteins in intact cellular systems (14,19), it suffers from a number of drawbacks. The nature of the chemical reaction and the product(s) formed during "photoaffinity" labeling (13,17) remain uncertain. Photoaffinity labeling often yields multiple labeling patterns, thus complicating the process of identification of labeled proteins (13,17). UV radiation can mimic the action of small ligands in activating signal transduction pathways that activate and/regulate mammalian cell functions (20 -22).
FSBA previously used to probe the ADP receptor also has certain limitations. Although FSBA contains a carbonyl group in a position sterically equivalent to the ␣-phosphoryl group of * This investigation was supported by National Heart, Lung, and Blood Institute Grant HL 46431 (to R. W. C. and R. N. P.) and by National Science Foundation Grant MCB-94 -23108 (to R. F. C.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  ADP, it has a benzoyl group and thus lacks the hydrophilicity of ADP. Since the reagent includes an ester bond (Fig. 1B), hydrolysis of FSBA yields adenosine, which is an inhibitor of platelet functions and thus adenosine deaminase has to be included in the incubation mixtures (23). Furthermore, the available method of synthesis of [ 3 H]FSBA yields a product of relatively low specific radioactivity. These properties have frustrated attempts to purify the ADP receptor using [ 3 H]FSBA as a covalent label.
In light of the limitations on the use of FSBA and the controversy surrounding the identity of an ADP receptor on the platelet surface, we decided to investigate the effect of other nucleotide-based affinity labels with the potential for alkylation of a putative receptor. 8-(4-Bromo-2,3-dioxobutylthio)ADP (8-BDB-TADP) (Fig. 1C) has been previously used as an ADP affinity label to probe the structure-function relationship of ADP-requiring enzymes such as pyruvate kinase (24) and glutamate dehydrogenase (25,26). 8-BDB-TADP has the characteristic diphosphate of the natural ligand ADP, and is both hydrophilic and negatively charged at neutral pH. Furthermore, it can be synthesized as the 32 P-labeled reagent with high specific radioactivity. In order to further investigate and test our hypothesis that aggregin is a putative ADP receptor, we chose 8-BDB-TADP as a true ADP affinity ligand to label the ADP receptor and ascertain its effect on ADP-induced platelet responses. In this case the nature of the chemical reaction is well understood and chemically defined products have been isolated (26,27). This report describes in detail the effect of 8-BDB-TADP on ADP-induced platelet responses and the protein(s) modified on the surface of human platelets. A preliminary account of the work has previously appeared (28).
Shape Change and Aggregation of Platelets-Shape change and aggregation of washed platelets were performed in a Lumi-Aggregometer (Chronolog, Havertown, PA) under constant stirring conditions at a concentration of 1 ϫ 10 8 /ml and 5 ϫ 10 8 /ml, respectively. Washed platelets (1 ϫ 10 8 /ml) and suspension buffer were used in reference cuvettes in shape change and aggregation experiments, respectively. ADP-induced platelet shape change was performed in the presence of 2 mM EGTA while platelet aggregation was performed in the presence of 1 mM Ca 2ϩ and 1 mg/ml fibrinogen.
Fibrinogen Binding-Human 125 I-fibrinogen (specific radioactivity 2.12 mCi/mg protein) was prepared by the iodogen method as described previously by Puri et al. (34) except that the equilibration and elution buffer contained 184 mM HEPES, 136 mM NaCl, 2.7 mM KCl, and 0.35 mg/ml bovine serum albumin, pH 7.35. The preparation of radioiodinated fibrinogen was found to be 93 Ϯ 5% (n ϭ 4) clottable. The radioiodinated fibrinogen was diluted with unlabeled fibrinogen (20 mg/ml) in a 1:1 ratio before using it in binding experiments. Binding of 125 I-fibrinogen to washed platelets was measured as reported previ- ously (34). For total binding of 125 I-fibrinogen, platelets (1 ϫ 10 8 /200 l) were placed in an Eppendorf tube at 25°C and incubated with increasing amount of 125 I-fibrinogen in the presence of 1 mM Ca 2ϩ for 1 min. The reaction mixture was then incubated with 30 M ADP for another 5 min. The reaction was terminated by removing a 50-l aliquot from each sample and layering over 200 l of silicon oil mixture in a microtube (Sarsdet, Princeton, NJ) (34) and centrifuging in a Beckman model E microcentrifuge for 3 min. The pellet at the bottom of the microtube was excised and radioactivity in the pellet measured by counting in an automatic gamma counter (model 1470 WIZARD, Wallac Oy, Turku, Finland) equipped with RIaCalc WIZ software 2.2. Nonspecific binding of 125 I-fibrinogen to platelets was determined identically in the presence of 2 mM EDTA and 10-fold molar excess of unlabeled fibrinogen. Binding in the presence of 8-BDB-TADP was determined similarly by preincubating the platelets in the dark with 8-BDB-TADP at 25°C for 30 min.
Platelet Secretion-Secretion of ADP by platelets (5 ϫ 10 8 /ml) following exposure to ADP in the absence and presence of 8-BDB-TADP, was estimated, simultaneously with aggregation, by measuring the intensity of chemiluminescence of commercial luciferase-luciferin reagent (30 l/ml of platelet suspension; freshly constituted as described in the protocol provided by the suppliers) in a dual-channel Lumi Aggregometer (Havertown, PA). The ATP release in the incubation mixtures was computed from a standard curve prepared by estimating chemiluminescence from solutions of known concentration of ATP in the absence and presence of 8-BDB-TADP as described above. 8-BDB-TADP, as high as 0.6 mM, had no effect on the assay.
Measurement of Intracellular Calcium Levels-Platelets were loaded with fluorescent indicators by incubating platelet-rich plasma with Fura-2/AM (5 M) or Quin-2/AM (20 M) at 37°C for 20 min. Stock solutions of the indicators were prepared in Me 2 SO. Maximum fluorescence (F max ), reflecting total content of Ca 2ϩ in storage granules, was estimated by treating the Fura-2-( ex and em , 340 and 510 nm, respectively) loaded platelets (2 ϫ 10 8 /ml) with 2 M ionomycin (in Me 2 SO), and autofluorescence (AF) was determined by quenching the above incubation mixture with 10 mM MnCl 2 . Platelets loaded with the fluorophores were then incubated with 1 mM CaCl 2 at 37°C for 1 min and then stirred with an increasing concentration of ADP to monitor the fluorescence emission intensity (F obs ). The intracellular level of Ca 2ϩ , [Ca 2ϩ ] i , was calculated (35,36) by using the equation [F obs -F min /F max -F obs ] ϫ K d (where F min ϭ AF ϩ (F max -AF)/6, and K d ϭ 228 for Fura-2/ AM). The loaded platelets were then incubated with 8-BDB-TADP at 25°C in the dark and ADP-induced release of Ca 2ϩ determined as described above. When Quin-2/AM was used as the fluorophore, the protocol used for determining [Ca 2ϩ ] i was similar to the one described above except that the ex and em used were 339 and 492 nm, respectively, and a value of 114 was used for K d (37).
Measurement of Intracellular cAMP Levels-Intracellular levels of cAMP in washed and 8-BDB-TADP-modified platelets (1 ϫ 10 8 /ml) were determined by the commercial cAMP-binding assay as outlined by the manufacturers and described previously by Puri et al. (34).
Labeling of Platelets by [␤- 32  , and benzamidine (2 mM). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed on 10% gels according to the method of Laemmli (38) with the modification that the gels were prepared from a solution of acrylamide-bisacrylamide mixture containing leupeptin (10 M), p-APMSF (30 M), and benzamidine (1 mM). Gels were calibrated with prestained molecular weight standards. The dried gels were exposed to CRONEX x-ray film (NEN-DuPont) at Ϫ70°C for 7-14 days before developing. The labeled platelets were resuspended, dissolved, and electrophoresed as described above. Gels were carefully shaken with ENLIGHTNER for 20 min at 25°C and dried immediately. Gels were exposed to the x-ray film at Ϫ70°C for 1-2 months.

RESULTS
Shape Change-8-BDB-TADP inhibited ADP-induced shape change in a concentration-dependent manner with an IC 50 of 75 M (Fig. 2A). Complete inhibition of ADP-induced shape change required 400 M of the 8-BDB analog. The inhibition of shape change by the 8-BDB analog followed biphasic kinetics. The time course of inhibition of ADP-induced platelet shape change by 200 M 8-BDB-TADP showed complete inhibition of shape change in 70 min (Fig. 2B). However, loss of 50% of the rate of shape change occurred in a relatively short time, 10 min. 8-BDB-TADP, at concentrations as high as 400 M, did not induce platelet shape change. The results show that 8-BDB-TADP was a potent inhibitor of ADP-induced shape change.
Aggregation-8-BDB-TADP blocked ADP-induced platelet aggregation in a concentration-dependent manner with an IC 50 of about 75 M (Fig. 3A). Inhibition of ADP-induced aggregation was also time-dependent, and the 8-BDB analog (150 M) inhibited the rate of ADP-induced aggregation by 50% in 10 min and 90% in 40 min (Fig. 3B). 8-BDB-TADP, as high as 400 M, did not act as an agonist of platelet aggregation. The results demonstrate that 8-BDB-TADP is an effective inhibitor of ADP-induced platelet aggregation.
Fibrinogen Binding-When platelets were exposed to 30 M ADP, total binding of 125 I-fibrinogen was concentration dependent (Fig. 4). Nonspecific binding of 125 I-fibrinogen to platelets remained at about 10% level over the 16-fold concentration range of the radiolabeled fibrinogen. When 8-BDB-TADP-modified platelets were incubated with an increasing concentration of the radiolabeled fibrinogen followed by exposure to 30 M ADP, the total binding of fibrinogen was diminished to a mean (ϮS.E.) value of 21 Ϯ 5% compared with that of unmodified platelets. The results show that specific binding of 125 I-fibrinogen to platelets (or exposure of GPIIb-IIIa complex) induced by ADP was blocked by 80% when platelets were exposed to the 8-BDB-TADP.
Aggregation Induced by Various Agonists-When platelets were modified by 8-BDB-TADP, centrifuged, washed, and resuspended in Tyrode's buffer, pH 7.35, only 9% of the ADPinduced aggregation was detectable ( Table I). Rates of collagenand U46619-induced aggregation of platelets modified by 8-BDB-TADP were significantly lower compared with the corresponding controls (Table I). In contrast, thrombin-induced aggregation of 8-BDB-TADP-modified platelets was comparable to that of the unmodified platelets. Rates of aggregation of 8-BDB-TADP-modified platelets by A23187, PMA, and A23187ϩPMA were minimally affected. The results show that only rates of aggregation of platelets by ADP, U46619, and collagen agonists that completely or partially utilize ADP-dependent mechanism (7, 10, 11) were almost completely or significantly affected as a result of covalent modification of the ADP-binding sites on the platelet surface.
Secretion-Thrombin and ADP both caused secretion of 5 and 2 M ATP, respectively, from the storage granules of platelets (Fig. 5). Whereas thrombin-induced secretion of nucleotides and aggregation are synchronous events, both reaching maximum within a minute, ADP-induced secretion only starts after maximum aggregation has occurred. This lag may be due to the presence of a higher concentration of ADP (some released from the platelets) in the localized environment of platelets.
The concentration of 8-BDB-TADP (50 M) which minimally affected ADP-induced platelet aggregation completely blocked ADP-induced secretion. The results show that (a) thrombinand ADP-induced secretion of nucleotides from dense granules proceed by two different mechanisms, and (b) ADP-induced secretion does require binding of the agonist to the receptor, and chemical modification of this receptor by 8-BDB-TADP blocks this response.
Intracellular Calcium Mobilization-Intracellular levels of calcium in platelets exposed to ADP rose from a basal value of 100 to 703 nM and 683 nM as measured by using Fura-2/AM and Quin-2/AM, respectively (Fig. 6, data for measurements by using Quin-2/AM not shown). Preincubation of platelets with 230 M 8-BDB-TADP for 30 min at 25°C completely blocked ADP-induced elevation of [Ca 2ϩ ] i . 8-BDB-TADP itself did not cause any change in the basal levels of Ca 2ϩ in platelets. were withdrawn from each incubation mixture and layered over a mixture of silicon oils as described under "Experimental Procedures" and centrifuged in a microfuge for 3 min. The pellet at the bottom of the centrifuge tube was excised and assayed for radioactivity. Nonspecific binding of 125 I-fibrinogen to washed platelets was determined as described above except that the incubation mixtures contained 2 mM EDTA and 10-fold molar excess of unlabeled fibrinogen. Binding to platelets in the presence of 8-BDB-TADP was performed as described in the case of total binding except that the washed platelets were preincubated in the dark with 300 M 8-BDB-TADP at 25°C for 30 min. Specific binding of 125 I-fibrinogen to platelets in the absence (q) and presence (E) of 8-BDB-TADP were computed by subtracting nonspecific binding from the total binding in each case. The data are expressed as mean of the molecules of 125 I-fibrinogen bound Ϯ S.E./platelet versus concentration of 125 I-fibrinogen.  Intracellular Levels of cAMP-There is little or no intracellular cAMP present in resting platelets (Table II, 1) (39,40). Iloprost (a stable carbocyclic derivative of prostaglandin I 2 ) and papaverine (a phosphodiesterase inhibitor) in combination raise and sustain intracellular levels of cAMP by stimulating membranous adenylate cyclase which converts ATP to cAMP (Table II, 2). Thrombin (40,41) to a larger extent and ADP (Table II, 5 and 6) (42) to a lesser extent antagonize the increase in intracellular levels of cAMP in platelets induced by prostaglandins. 8-BDB-TADP neither raised (Table II, 3) nor affected the ability of iloprost to raise intracellular levels of cAMP (Table II, 4). However, the 8-BDB analog impaired the ability of ADP to antagonize the increase in cAMP levels in platelets exposed to iloprost (Table II, 8). These results show that covalent modification of ADP-binding sites on platelet surface results in the loss of the ability of ADP to modulate stimulated adenylate cyclase activity.

TABLE I Effect of chemical modification by 8-BDB-TADP on platelet aggregation induced by various agonists
Identification of 8-BDB-TADP-labeled Protein on Platelet Surface-We have previously demonstrated that a single peak of radioactivity at 100 kDa in the radioactivity distribution profile of slices of gels obtained by SDS-PAGE of solubilized [ 3 H]FSBA-labeled membranes obtained from [ 3 H]FSBA-labeled platelets corresponds to an ADP receptor, aggregin, on the platelet surface (Ref. 9, and references cited therein). However, because of the relatively low specific radioactivity of the available [ 3 H]FSBA and the low emitting threshhold of tritium, it is cumbersome and time consuming to obtain good radiograms by autoradiography of gels obtained by SDS-PAGE of solubilized [ 3 H]FSBA-labeled platelets. Nevertheless, an autoradiograph of the gel obtained by SDS-PAGE of solubilized [ 3 H]FSBA-labeled platelets shows the presence of a single protein band corresponding to radiolabeled aggregin (100 kDa) (data not shown). [␤-32 P]8-BDB-TADP also labeled a single protein (100 kDa) on the platelet surface; the intensity of this band was reduced when labeling was carried out with platelets preincubated with 2 mM ADP (Fig. 7a). When FSBA-modified platelets or platelets exposed to 10 mM ADP or 10 mM ATP were incubated with [␤-32 P]8-BDB-TADP, no label was incorporated into the protein corresponding to aggregin (Fig. 7b); whereas 2 mM ADP and 8-bromo-ADP (Fig. 7b, lanes B and D) considerably reduced the intensity of the [␤-32 P]8-BDB-TADP-labeled aggregin, ATP-␣-S, adenosine, AMP, and GDP (Fig. 7c, lanes  E-H), at the same concentration, minimally protected aggregin from labeling by [␤-32 P]8-BDB-TADP (Fig. 7c). Platelets previously exposed to 2 mM p-chloromercuribenzenesulfonate (a sulfhydryl modifying reagent) were not labeled by [␤-32 P]8-BDB-TADP (Fig. 7c, lane I). This result is consistent with our previous demonstration that aggregin contains lysine and cysteine residues at the ADP-binding domain in aggregin and that FIG. 5. Effect of 8-BDB-TADP on ADP-induced platelet secretion. ATP release following exposure of platelets (5 ϫ 10 8 /ml) to thrombin (2 nM) and ADP (30 M) was measured by the commercial luciferaseluciferin reagent as described under "Experimental Procedures." Platelets (5 ϫ 10 8 /ml) were preincubated in the dark with 50 M of 8-BDB-TADP for 15 min at 25°C followed by treatment with 30 M ADP. 8-BDB-TADP had no effect on the intensity of chemiluminiscence produced by solutions of known concentration of ATP (used for calibration of the assay) treated with luciferase-luciferin reagent under identical conditions. Thrombin-and ADP-induced secretion were computed from two separate calibrations employing standard solutions of ATP. Similar results were obtained with platelets from the blood of two different donors.

TABLE II
Effect of 8-BDB-TADP on ADP-mediated antagonism of increase of intracellular levels of cAMP induced by iloprost Intracellular levels of cAMP in washed platelets in the absence and presence of 8-BDB-TADP were estimated by using commercial protein binding assay described previously (34). cAMP level in unmodified platelets following exposure to iloprost was set at 100% (control). Experiments were performed in triplicate and data are expressed as the mean Ϯ S.E. In 3, 4, 7, and 8 platelets were preincubated in the dark with 8-BDB-TADP for 30 min at 25°C before subsequent treatments.  4 covalently and selectively label the same ADP-binding protein, aggregin, on the platelet surface. DISCUSSION 8-BDB-TADP inhibited ADP-induced platelet shape change and aggregation in a concentration-and time-dependent manner signifying that the reagent covalently modifies ADP-binding sites on platelet surface. Preincubation of the platelets with 8-BDB-TADP effectively blocked ADP-induced binding of 125 Ifibrinogen to platelets. These results are consistent with the fact that the 8-BDB analog inhibits ADP-induced platelet aggregation. When 8-BDB-TADP-modified platelets were examined for their ability to aggregate by exposure to various agonists, only ADP-induced aggregation was almost completely blocked. The results that the rates of collagen-and U46619induced platelet aggregation were significantly reduced in 8-BDB-TADP-modified platelets are in accord with the previous findings that platelet aggregation induced by these two agonists proceeds, at least in part, by ADP-dependent mechanisms (10,11). The rates of aggregation induced by thrombin A23187, PMA, and A23187ϩPMA were minimally affected by chemical modification of ADP-binding sites. These results are in accord with the fact that aggregation of platelets induced by the above agonist follow ADP-independent mechanisms (43)(44)(45). 8-BDB-TADP also blocked ADP-induced secretion of nucleotides by platelets.
ADP-induced release of Ca 2ϩ from the dense tubular system plays an important role in the platelet responses mediated by the binding of ADP to its receptor (46). Our results show that 8-BDB-TADP itself was not an agonist of intracellular mobilization of Ca 2ϩ , but it effectively blocked similar mobilization by ADP. ADP is a non-penetrating reagent, and platelet responses elicited by this agonist are mediated through its interaction with specific receptors on platelet surface (9,47). In addition to inducing intracellular increase in Ca 2ϩ , ADP also has another important function: it antagonizes elevation of intracellular levels of cAMP induced by prostaglandins (42). Our observation that covalent modification of platelets by 8-BDB-TADP blocked the ability of ADP to inhibit stimulated adenylate cyclase activity is consistent with the previous studies.  4 . Platelets (2 ϫ 10 8 /0.2 ml) were labeled by 8-BDB-TADP, and the modified platelets were then exposed to NaB[ 3 H] 4 . The reaction mixtures were centrifuged, pellets solubilized, and subjected to SDS-PAGE as described under "Experimental Procedures." Lanes A-C, correspond to platelets labeled in the absence of any other reagent, 10 mM ATP, and 10 mM ADP, respectively. NaB[ 3 H] 4 alone did not label platelets. The gel was calibrated with prestained molecular mass standards (Bio-Rad) each of which was covalently attached to a dye of different color and consisted of the following: myosin, 217 kDa; ␤-galactosidase, 135 kDa; bovine serum albumin, 72 kDa; carbonic anhydrase, 42 kDa, and soybean trypsin inhibitor, 31 kDa. The values of the molecular weight correspond to those supplied by Bio-Rad.
The ADP-binding site mediating ADP-induced platelet shape change may be different from the one mediating intracellular Ca 2ϩ mobilization (48). Other investigators suggested that that ADP-binding site for platelet shape change might be different from those antagonizing stimulated adenylate cyclase activity (49). Savi et al. (50) recently presented experimental evidence for two types of ADP-binding sites that differ in their affinity for ADP. Unlike 8-BDB-TADP, FSBA was not able to antagonize the stimulated adenylate cyclase activity, but this may be due to its lower affinity for aggregin (23).
Although the two-receptor hypothesis for different ADP-induced platelet responses has been widely discussed, the identity of two distinct ADP-binding proteins on platelet surface has never been established. Results presented in this investigation, for the first time, show that [ 3 H]FSBA and [␤-32 P]8-BDB-TADP label the same protein, aggregin (100 kDa), the ADP receptor on the platelet surface. Covalent modification of aggregin by 8-BDB-TADP not only inhibits ADP-induced shape change, aggregation, and exposure of fibrinogen-binding sites but also blocks the ability of ADP to antagonize stimulated adenylate cyclase activity suggesting that all of these responses are mediated by aggregin. Complete prevention of incorporation of [␤-32 P]8-BDB-TADP into aggregin required either prior covalent modification of aggregin by FSBA or preincubation of platelets with a concentration of ADP or ATP as high as 10 mM. Our experience has shown that those concentrations of 2Ј,3Јdialdehyde derivative of ADP (10 mM) that far exceeded those necessary to inhibit ADP-induced platelet shape change and aggregation, could not completely block labeling of platelets by [ 3 H]FSBA. 2 We have recently shown that 7-chloro-4-nitro-2oxa-1,3-diazole blocks ADP-induced platelet responses by covalently modifying aggregin (51). Preincubation of platelets with 30 mM ADP, 30 mM ATP, or covalent modification of platelets by FSBA was necessary to block completely labeling of aggregin by [ 14 C]7-chloro-4-nitro-2-oxa-1,3-diazole. Other investigators have shown that a concentration of ATP, as high as 30 mM, was needed to block ADP-induced mobilization of Ca 2ϩ in human erythroleukemia cells (5). Cristalli and Mills (17) found that ADP (0.8 mM) reduced the intensity of the 43 kDa radiolabeled protein band and claimed that it was the ADP receptor. Higher concentration of ADP (Ͼ0.8 mM) or its analogs might have reduced the intensity of other protein bands shown in the autoradiograph presented in that report. It is noteworthy that the autoradiograph in the above report contains a dense radiolabeled protein band around 100 kDa (17). Our results, however, strongly suggest that only one ADP-binding protein on platelet surface, aggregin (100 kDa), mediates all of the ADP-induced platelet responses. Our results do not rule out the possibility that more than one class of ADP-binding site may exist on this receptor which differ in their affinity for ADP. Another explanation is that different signal transducing mechanisms exist for ADP-induced platelet shape change and inhibition of stimulated adenylate cyclase activity. Various macrophage cell lines have been reported to express G proteincoupled purinoreceptors (52)(53)(54). We further demonstrated that 8-BDB-TADP and NaB[ 3 H] 4 also label aggregin, and their ability to modify aggregin was blocked by the presence of ADP or ATP. These results are inconsistent with the possibility that a 43-kDa protein on platelet surface is the ADP receptor suggesting that labeling of this protein by AzPET-ADP might be due to nonspecific interactions.
In summary, we have demonstrated that 8-BDB-TADP, an ADP-affinity analog, blocks ADP-induced shape change, aggregation, exposure of fibrinogen-binding sites, secretion, calcium mobilization, and inhibition of stimulated adenylate cyclase activity by ADP in washed platelets by covalently modifying a single surface protein, aggregin (100 kDa), a putative ADP receptor.