Clopidogrel Inhibits the Binding of ADP Analogues to the Receptor Mediating Inhibition of Platelet Adenylate Cyclase

Page created by Fred Schneider
 
CONTINUE READING
430

                      Clopidogrel Inhibits the Binding of
                   ADP Analogues to the Receptor Mediating
                    Inhibition of Platelet Adenylate Cyclase
                           D.C.B. Mills, R. Puri, C-J. Hu, C. Minniti, G. Grana, M.D. Freedman,
                                              R.F. Colman, and R.W. Colman

           Clopidogrel, like the homologous thienopyridine derivative tidopidine, selectively inhibits platelet aggregation
           induced by ADP. We have previously described two nucleotide-binding sites on platelets related to ADP-
           mediated platelet responses. The first is a high-affinity binding she for 2-methylthio-ADP (2-MeSADP) that
           is linked to the inhibition of stimulated adenylate cyclase. The second is the 100-kd exofacial membrane
           protein aggregin, which is labeled by the reactive ADP analogue 5'-p-fluorosulfonyibenzoyl adenosine (FSBA)
           that is related to shape change and aggregation. We set out to determine if either of these sites is blocked in
           vivo by clopidogrel or its active metabolite. Six subjects were given clopidogrel (75 mg/day for 10 days) in a
           double-blind crossover experiment All of the subjects developed prolonged bleeding times while taking the
           drug. The rate of onset of the effect on bleeding time varied among subjects. Platelet aggregation induced by
           ADP or thrombin was significantly impaired by the drug treatment, but no effect was detected on shape
           change. The incorporation of [3H]FSBA into aggregin was also unaffected. Inhibition of adenylate cyclase by
           ADP or by 2-MeSADP was greatly reduced in all subjects, and in the case of 2-MeSADP, there was evidence
           for a noncompetitive effect Inhibition of adenylate cyclase by epinephrine was unaffected. In the three subjects
           for whom binding measurements were made, the number of binding sites for [32P]2-MeSADP was reduced
           from 534±44 molecules per platelet during control and placebo periods (11 determinations) to 199±78
           molecules per platelet during drug treatment (three determinations). There was no consistent change in the
           binding affinity. The inhibition of platelet function by clopidogrel is associated with a selective reduction in the
           number of functional receptors mediating the inhibition of stimulated adenylate cyclase activity by ADP.
           (Arteriosclerosis and Thrombosis 1992;12:430-436)
              KEY WORDS • ADP receptor • adenylate cyclase • platelets • thienopyridine • aggregation •
                   n • clopidogrel

P      latelets from humans and other mammalian spe-
         cies aggregate in response to exposure to ADP
         at micromolar concentrations.1 Aggregation is
contingent on the presence of fibrinogen and calcium
ions in the medium, the metabolic activity of the
                                                                        diacylglycerol and inositol-l,4,5-trisphosphate and mo-
                                                                        bilization of Ca2+ from its intracellular depots (for
                                                                        recent reviews, see References 2 and 4).
                                                                           Like several other aggregating agents, ADP inhibits
                                                                        platelet adenylate cyclase, also presumably by a recep-
platelets, and exposure of the functional integrin GPIIb/               tor-mediated action.5 In intact cells this is demonstrated
Ilia complex that acts as a fibrinogen-binding site on                  as a reduction of the increase in platelet adenosine 3',
activated platelets.2 In contrast, ADP-induced shape                    5'-cyclic monophosphate (cAMP) concentration that
change, which is associated with a transient increase in                follows exposure to prostacyclin (prostaglandin I2
cytoplasmic Ca2+ and activation of calmodulin-depen-                    [PGI2]) or other stimulators of adenylate cyclase. Inhi-
dent myosin light-chain kinase,3 does not require fibrin-               bition by ADP of a particulate preparation of adenylate
ogen, integrins, or extracellular Ca2+. ADP, which does                 cyclase from platelet membranes has been observed,6
not normally cross the cell membrane, is assumed to                     with evidence for a requirement for guanine nucleo-
raise cytoplasmic Ca2+ by activating a specific trans-                  tides. A receptor for ADP that mediates this effect has
membrane receptor coupled to Ca2+ gating. ADP may                       been identified by the binding and displacement of
also activate phospholipase C, leading to formation of                  2-azido-ADP (2-NjADP)? and 2-methylthio-ADP (2-
                                                                        MeSADP).8 Normal human platelets have between 500
                                                                        and 1,000 of these receptors per cell, with a binding
   From the Thrombosis Research Center (D.C.B.M., R.P., C.J.H.,         affinity for 2-MeSADP on the order of 10 nM, and a
R.W.C.) and the Departments of Pharmacology (D.C.B.M.) and
Medicine (CM, G.G., R.W.C), Temple University School of Med-            half-maximal displacement of 2-MeSADP binding by 10
icine, Philadelphia, Pa.; Sanofi Pharmaceuticals (M.D.F.), New York;    /AMADP.
and the Department of Chemistry and Biochemistry (R.F.C), Uni-             Aggregation of platelets by ADP is inhibited by the
versity of Delaware, Newark, Del.
   Address for correspondence: David C.B. Mills, PhD, Thrombo-          reactive analogue 5'-p-fluorosulfonylbenzoyladenosine
sis Research Center, Temple University School of Medicine,              (FSBA).9 Under conditions that lead to inhibition of
Philadelphia, PA 19140.                                                 shape change (incubation of washed platelets for 30
   Received October 4, 1991; revision accepted January 13, 1992.        minutes with 40 mM FSBA), tritiated FSBA (pH]FSBA)

                            Downloaded from http://atvb.ahajournals.org/ by guest on November 7, 2015
Mills et al Clopidogrel Blocks ADP Receptors on Platelets        431

is incorporated into a platelet membrane protein of about       Platelet Preparation
100 kd, a reaction inhibited by ADP and ATP.10 This                Blood was collected by free-flow venipuncture into
protein, a putative ADP receptor, has been given the            50-ml tubes containing 5 ml acid-citrate-dextrose anti-
name aggregin.11 As FSBA neither blocks the effect of           coagulant. Platelet-rich plasma (PRP) was prepared by
ADP on adenylate cyclase nor inhibits the binding of            centrifugation for 15 minutes at 320g at room tempera-
2-MeSADP, it is clear that aggregin and the high-affinity       ture. PRP was centrifuged for 15 minutes at l,100g at
2-MeSADP receptor previously identified are functionally        room temperature in the presence of 1.7 mM P G E L
distinct.12                                                     0.02 unit/ml hirudin, and 0.1 unit/ml apyrase,12 and the
   Thienopyridine compounds, including the antithrom-           pellet was resuspended in a buffer containing 136 mM
botic drugs ticlopidine and clopidogrel (SR 25990C),            NaCl, 11.9 mM D-glucose, 11.9 mM NaH2CO3, 2 mM
have been found to selectively inhibit platelet aggrega-        MgClj, 2 mM CaCl2, 0.42 mM NaH2PO4, and 3 5 mg/ml
tion induced by ADP.13 To elucidate the relations               bovine serum albumin, pH 7.35 (buffer A), with the
between aggregin, the 2-MeSADP receptor, and the                addition of 0.05 unit/ml hirudin and 0.1 unit/ml apyrase.
various responses of platelets to ADP, we have studied          After incubation at 37°C for 10 minutes, the centrifuga-
platelet responses before and after administration of           tion was repeated and the pellet resuspended in buffer
clopidogrel or a placebo to human volunteers.                   A with apyrase, 0.1 unit/ml. After a further 10-minute
                                                                incubation at 37°C, the centrifugation was repeated and
                       Methods                                  the pellet finally resuspended in buffer A containing
Materials                                                       18.5 mM Af-2-hydroxylethylpiperazine-.W-2-ethane-
                                                                sulfonic acid (HEPES) at pH 7.35.
   Human thrombin (3,203 National Institutes of Health
units/mg protein) was kindly supplied by Dr. John W.
Fenton II, Division of Laboratories and Research, New           Platelet Aggregation and Shape Change
York State Department of Health, Albany, N.Y. Human                These measurements were performed at 37°C in a
fibrinogen was from Kabi, Franklin, Ohio. [2,8-3H]ade-          Chronolog Lumiaggregometer. Shape change was mea-
nine (1.06 TBq/mmol) and [32P]phosphate (carrier free)          sured after dilution to 1 x 108 platelets/ml and addition
were from ICN Biomedicals, Costa Mesa, Calif., and              of 4 mM ethylene glycol-bisO-aminoethyl ether)-
[U-14C]adenine (10 GBq/mmol) was from Amersham                  N,N,N',N'-tetraacetic acid to prevent aggregation. Ag-
Corp., Arlington Heights, 111. Dowex AG50, urea, and            gregation was measured at a standard platelet count of
acrylamide were from Bio-Rad, Richmond, Calif.                  5 x 108 platelets/ml after addition of 1 mg/ml fibrinogen.
pirjFSBA (500 GBq/mol) was prepared as reported.11              Aggregation and shape change are expressed as the rate
                                                                of increase or decrease, respectively, of light transmis-
Design of the Experiment                                        sion in millivolts per minute during 1 minute after the
   Six healthy male volunteers, aged 22-35, gave in-            addition of an agonist.
formed written consent to a protocol approved by the
Temple University School of Medicine Institutional              Labeling of Aggregin
Review Board. They were studied for a period of 73                 Washed platelets (6 ml, 5 x 10s platelets/ml) were incu-
days divided into three segments. In the first period           bated with 100 ILM pHJFSBA for 40 minutes at 23°C. The
each subject was given either placebo or clopidogrel, 75        labeled platelets were sedimented and washed. Mem-
mg daily by mouth for 10 days. After a washout period           branes were prepared by the grycerol osmotic lysis
of 50 days without treatment, in the third period each          method; solubilized in a solution of sodium dodecyl sulfate
subject received drug or placebo, depending on then-            (SDS), urea, and dithiothreitol; and diaryzed; and the
exposure during period 1. The subjects were randomly            retentate was assayed for radioactivity. With each volun-
assigned to drug-first or placebo-first groups. Neither         teer, a portion of the retentate obtained from at least one
the subject nor the experimenters were informed of the          experiment was subjected to SDS-polyacrylamide gel
treatment (double-blind). Subjects A and D received             electrophoresis (PAGE) on 15% gels to confirm the
the drug in period 1, the remaining four individuals in         labeling pattern. The gels were sliced and assayed for
period 3. All subjects were screened for abnormalities of       radioactivity. The radioactivity peak at 100 kd corresponds
blood chemistry, hematology, and bleeding time before           to [3H]FSBA-labeled aggregin.1011
admittance to the study, and these tests were repeated
on days 1, 3, 6, 11, 15, 52, 54, 57, 62, and 66. Platelet       Adenylate Cyclase Inhibition
responses were measured on days 1, 11, 52, and 62,                 PRP was incubated for 1 hour with 1 yM [U-HC]ad-
corresponding to the beginning and end of each treat-           enine to label intracellular ATP. Duplicate samples of
ment period. For simplicity the results are all presented       this labeled PRP (0.45 ml) were incubated for 2 minutes
in the following sequence: pre 1 (first measurement),           with 50 fi\ saline containing 50 mM EDTA and five
placebo, pre 2 (after the washout period), and drug,            concentrations of ADP (final concentration, 0.1-10
although for subjects A and D the actual order for drug         fiM), five concentrations of 2-MeSADP (1-100 nM),
and placebo was reversed.                                       and three concentrations of epinephrine (0.01-1 JAM).
                                                                PGI2 (0.1 fjM) was then added, and after a further
Ivy Bleeding Time                                               2-minute incubation the reaction was stopped with
   The forearm bleeding time was measured on several            trichloroacetic acid; the radioactivity of cAMP was then
occasions before and during the drug and placebo                measured after purification by chromatography on
periods using a Simplate device at a constant venous            Dowex AG50 and alumina. An internal standard of
pressure of 40 mm Hg. Bleeding times of 30 minutes or           pHjcAMP was used to measure recovery. Hill plots of
longer were recorded as 30 minutes.                             the data were used to interpolate, where possible, the

                         Downloaded from http://atvb.ahajournals.org/ by guest on November 7, 2015
432     Arteriosclerosis and Thrombosis     Vol 12, No 4 April 1992

                                                                      30
                                                                                                                •                                        B
                                                                                                                                       l\    **•
                                                                      25
                                                                                                            •
                                                                                                                        /
                                                                                                            •
                                                                                                                        •          r        \\
                                                                      20
                                                                                                        •
                                                                                                                        i

                                                                                                                                  4/ \\
                                                                                                        •           j       9
                                                                  o                                    :            /             f     \
                                                                      15 •
                                                                                                     / Pf                         —              V
                                                                      10
                                                                                                   Iff?*
                                                                                                   M l /                                             V
                                                                                                                                                   Mb        ;
                                                                                               «V
                                                                               • * • " • • •
                                                                       5                                                                             •
                                                                               •               '

                                                                                                   i                    Drug
      • 5                                         15                    • 5                                             5          10              15
                                                                                                                            Day

               FIGURE 1. Line plots of forearm Ivy Simplate bleeding times of six subjects during and after placebo (panel A)
               and drug (panel B) treatment periods.

concentration of each inhibitor required for 50% inhi-           the paired t test, with the average results for all other
bition (ED,,,). In one experiment in which adenylate             periods for the same donor.
cyclase was measured in isolated platelets, the platelets
were labeled by incubation for 1 hour with 0.2 nM                                           Results
[2,8-3H]adenine, acidified to pH 6.4, centrifuged at                Treatment with 75 mg clopidogrel/day for 10 days
l,030g for 15 minutes at room temperature, and resus-            caused a sharp increase in the Ivy bleeding time in all of
pended in isotonic saline buffered to pH 7.4 with 10 mM          the six subjects, although the pattern of increase was
HEPES. [14C]cAMP was used as the recovery standard.              diverse (Figure 1). In two subjects (C and F) the increase
                                                                 in bleeding time was rapid but small. In two others (A and
Equilibrium Binding of p-[32P]2-MeSADP                           B) it was slower but more pronounced. In the remaining
   2-MeSAMP was converted to the imidazolate and                 two subjects (D and E) the increase was rapid and large.
phosphorylated with tri-n-butylammonium phosphate                Infiveof the six subjects the bleeding time had returned to
(37 GBq/mmol). The diphosphate was isolated by paper             near normal within 4 days of withdrawing the drug. In the
electrophoresis followed by anion-exchange high-per-             sixth (D) it was still abnormal 4 days after withdrawing the
formance liquid chromatography. Samples of PRP (0.45             drug. The mean bleeding time (±SEM) for all observa-
ml) were incubated for 10 minutes at 37°C with 50 ftl            tions in the pre 1, pre 2, and placebo periods was 5.4±0.2
                                                                 minutes («=42). The mean bleeding time was increased in
saline containing 50 mM EDTA and the radioligand at
                                                                 the drug period to 18.1 ±2.3 minutes (n = 18), recording
10 concentrations in the range 2-120 nM (final). Dupli-
                                                                 "30 minutes" for bleeding times greater than or equal to
cate aliquots (0.2 ml) were layered over a cushion of            30 minutes. There was no significant difference between
20% sucrose in tris(hydroxymethyl)aminomethane-                  the bleeding times measured in pre 1 and those in pre 2 or
buffered saline containing 10 mM EDTA in Sarstedt                placebo. No other hematological abnormalities were de-
microsediment collection tubes and centrifuged for 2             tected, and there was no evidence of blood loss or other
minutes at full speed in a modified Fisher Model 59              consequences of this marked effect on cutaneous hemo-
microcentrifuge. The tubes were frozen in solid CO2              stasis. All of the other routine blood chemistry tests,
and cut 1 mm above the pellet. Pellet and supernatant            including serum electrolytes, liver and kidney function,
were assayed for radioactivity by Cerenkov counting.             blood glucose, hematocrit and hemoglobin, and whole-
Nonspecific binding was determined at two concentra-             blood platelet and leukocyte counts, remained within
tions of the radioligand in the presence of 1 mM ADP.            normal limits throughout the study.
Statistical Analysis                                                Platelet aggregation and shape change in response to
                                                                 ADP and thrombin (Figure 2) were studied in washed
   In each case the mean of observations on all subjects         platelet suspensions so that the results would be directly
made on a given day, placebo, pre 2, or drug, was                comparable to those of the aggregin labeling experi-
compared with the mean for day 1 (pre 1) by the                  ments. No significant difference in the shape change
unpaired t test. Alternatively, the results obtained for         response to either 1 fiM ADP or 1 nM thrombin was
each subject for the drug period were compared, using            seen when the drug period was compared with the

                         Downloaded from http://atvb.ahajournals.org/ by guest on November 7, 2015
Mills et al   Clopidogrel Blocks ADP Receptors on Platelets        433

                                                                        Incorporation of the reactive ADP analogue
                                                                     pHJFSBA into platelet membranes was unaffected by
                                                                     drug treatment. The mean extent of incorporation dur-
                                                                     ing the drug period (53.0±3.2 cpm/108 platelets) was
                                                                     essentially the same as the mean in the other three
                                                                     periods (51.1 ±1.8). This value corresponds to the incor-
                                                                     poration of about 31.8X103 molecules of FSBA per
                                                                     platelet, without allowance for possible losses during
                                                                     membrane isolation. Samples of labeled membranes
                                                                     analyzed on nonreduced SDS-PAGE confirmed that the
                                                                     radioactivity was predominantly in the 100-kd protein.
                                                                        When platelets are incubated in PRP with radioactive
                                                                     adenine, the label is rapidly incorporated into the
       Control Drug                           Control Drug           cytoplasmic pool of rapidly equilibrating adenine nucle-
                                                                     otides, predominantly as ATP; stored nucleotides in the
                         Shape Chang*                                dense granules and actin-bound ADP are labeled much
                                                                     more slowly. When prelabeled platelets are incubated
                                                                     with 0.1 fiM PGI2, within 2 minutes the radioactivity of
                                                                     cAMP in the cells increases from the basal level
                                                                     (
434     Arteriosclerosis and Thrombosis Vol 12, No 4 April 1992

                           Subject E                   Subject F                     Normal Donor

                                                                               1.5

                                                                                                             Control
                                                                               1.0                           50%
                                                                                                       •o—   ADP
                                                                                                      —a—    BbSAOP

                                                                               O5

                                                                                         ••-a

                                                                               0J0
                      - 9 -8 -7 -S - S            - 9 - 8 - 7 - 8 -5             •9 - 8 - 7 - 6 - 5
                                                  Log Concentration (U)

                         Subject    E          Subject    F        Normal Donor

                      - 0 - 8 - 7 - 6 - 5   - 9 - 8 - 7 - 6 - 5    - 9 - 8 - 7 - 6 - 5

                                        Log Concentration (M)

FIGURE 4. Plots of inhibition ofadenosine 3',5'-cyclic monophosphate accumulation (cyclic AMP) in platelets from two subjects
(E and F) measured at the end of the drug treatment period and from a normal volunteer. Panel A; Measured in platelet-rich
plasma. Panel B: Measured in platelets isolated from plasma by centrifugation and resuspended in buffered saline. 2MeSADP,
2-methylthio-ADP.

   In one experiment with blood from subjects E and F                nM) and of the number of binding sites (534 ±44 per
and a third normal donor who was not included in the                 platelet) from 11 separate determinations in four sub-
trial, we compared the inhibition of adenylate cyclase in            jects during control or placebo phases of the trial were
PRP (Figure 4A) and in the same platelets isolated by                in agreement with published figures for normal donors.8
centrifugation and resuspended in HEPES-buffered saline              Only three measurements were possible (on subjects A,
(Figure 4B). With all three donors, the isolated platelets           E, and F) during the drug treatment phase. In all three
showed the same or greater sensitivity to the inhibitory             of these subjects the number of sites was considerably
effect of epinephrine as platelets in PRP; their sensitivity
                                                                     lower than estimates made on the same subject during
to ADP and 2-MeSADP was somewhat reduced. The two
experimental subjects, E and F, were both receiving                  the control phase, and the mean of the treatment
clopidogrel on the day this experiment was done. Subject             measurements (199±78 sites per cell) was significantly
F showed a very strong response to the drug so that no               lower than the control mean (/>
Mills et al   Clopidogrel Blocks ADP Receptors on Platelets    435

                                                                         high-affinity binding site for ADP and certain 2-substi-
                                                                         tuted ADP analogues. This site was identified by equi-
                                                                         librium binding to platelets of 2-NjADP labeled with "P
                                                                         in the /J-phosphate7 and was confirmed with the tighter
                                                                        binding analogue, 2-MeSADP, that was similarly la-
                                                                        beled.8 The binding affinity of 2-MeSADP in control
                                                                        platelets estimated from binding isotherms performed
                                                                        in PRP was 20.8 nM compared with an EDJO for
                                                                        inhibition of adenylate cyclase, also in PRP, of 4.4 nM,
                                                                        values that are close to published figures.58 The ED.*,
                                                                        for ADP for inhibition of adenylate cyclase was 2.7 ±0.3
                                                                        fiM (n = 18) in control platelets, whereas the KD for
                                                                        displacement of bound 2-MeSADP by ADP was 3.5
              0        50     100        150                200         fiM.s 2-N3ADP, with intermediate affinity, binds to
                   2-UaSADP Concentration (nil)                         platelets with an apparent KD of 140 nM, displaces
                                                                        bound 2-MeSADP with a KD of 120 nM, and inhibits
                                                                        adenylate cyclase with an ED50 of 88 nM.7 These figures
                                                                        suggest that the site identified by 2-MeSADP binding is
                                                                        the receptor through which ADP inhibits adenylate
                                                                        cyclase. This conclusion is supported by the observation
                                           o    PRE1                    that the non-cell-penetrating thiol reagent /?-chlo-
                                           *    PLACEBO                 romercuribenzene sulfonate blocks the binding of both
                                           o    PRE2
                                           •    DRUG
                                                                       2-NjADP7 and 2-MeSADP8 to platelets. This reagent
                                                                       blocks the inhibitory effect of ADP and its analogues on
                                                                       adenylate cyclase without inhibiting ADP-induced
                                                                       shape change.
                                                                           The reactive nucleotide analogue FSBA has been
                                                                       used to identify the binding sites for adenine nucleo-
                                                                       tides on a large number of purified proteins.16 FSBA
                                                                       inhibits platelet aggregation and shape change induced
                                                                       by ADP, although it has no effect on adenylate cyclase
                   200    400     600       800 1000 1200
                                                                       or on the inhibition of adenylate cyclase by ADP. 12
                   Binding    (fmoles/1ba
                                               Platelets)              Inhibition of shape change by FSBA is progressive,
 FIGURE 5. Panel A: Binding isotherms for 2-methylthio-
                                                                       corresponding to the covalent incorporation of label
                                                                       from pHJFSBA into a platelet membrane protein, ag-
ADP (2-MeSADP) with the platelets of one subject (F) in
                                                                       gregin, of Mr 100 kd.17 No other proteins become
platelet-rich plasma, snowing three measurements made dur-
                                                                       labeled under these conditions. ADP prevents the in-
 ing control periods (pre 1, placebo, pre 2) and one measure-
                                                                       corporation of FSBA, and at high concentrations FSBA
 ment at the end of the drug treatment period (drug). Results          causes a rapid shape change,18 supporting the conclu-
 are corrected for nondisplaceable binding measured in the             sion that FSBA acts at the same site as ADP.
presence of 1 mMADP, equivalent to 0.1-0.3% of the added                   Our results confirm that clopidogrel has a dramatic
 isotope, or less than 5% of the specific binding at half              effect on the bleeding time when given orally to human
saturation in controls. Panel B: Scatchard plots of the data          volunteers and that platelet aggregation induced by
from panel A The lines were drawn by eye.                             ADP and thrombin is markedly inhibited. We saw no
                                                                      effect of the drug on shape change induced by either
                                                                       agent, although wide variations in the responsiveness of
 Ticlopidine has been approved for use in the United                  control samples of washed platelets made accurate
 States for the prevention of strokes in patients sensitive           comparisons difficult. The incorporation of label from
 to aspirin. Both drugs act on blood platelets by inhibit-            pH]FSBA into platelet membrane proteins was entirely
 ing their response to ADP. This effect has been dem-                 unaffected by treatment with clopidogrel, indicating
 onstrated in humans and animals in ex vivo experiments,              that the drug has no effect on the availability or the
 in which platelet aggregation was measured in vitro in               nucleotide-binding properties of aggregin.
blood taken from volunteers or animals given the drug.13                   Adenylate cyclase inhibition by ADP and by 2-
 Recently, it has also been shown that the inhibitory                 MeSADP was severely curtailed by clopidogrel in all of
effect of ADP on the accumulation of cAMP in platelets                the six volunteers studied, in agreement with recent
after stimulation of adenylate cyclase by PGEj is also                findings in experimental animals that clopidogrel given
blocked during drug treatment.14-15 The specificity of                orally reduces the inhibitory effect of ADP on platelet
the inhibitory effect of these drugs for ADP or for                   adenylate cyclase stimulated by PGs.14 Similar results
agents like thrombin that act in part by releasing ADP                have been seen in human volunteers given ticlopidine.15
from the platelet dense granules suggests that they                       The similarity between the results obtained in PRP
might act at the level of the expression or function of               and after isolation of the platelets indicates that the
ADP receptors on the platelet membrane rather than,                   effects of clopidogrel are not easily reversed during
for instance, by interfering with the fibrinogen receptor.            short-term in vitro experiments and suggest either a
   We have demonstrated two distinct nucleotide-bind-                 permanent modification of the receptor or a defect in its
ing sites on platelet membranes. The first of these is a              biosynthesis. The effect is apparently indirect, as neither

                             Downloaded from http://atvb.ahajournals.org/ by guest on November 7, 2015
436      Arteriosclerosis and Thrombosis           Vol 12, No 4 April 1992

 the drugs themselves nor their principal metabolites                    2. Smith JB, Mills DCB: Chemical agents that inhibit platelet aggre-
 have demonstrable effects on platelets in vitro.13                          gation, in Fisher JW (ed): Biochemical Pharmacology of the Blood
                                                                             and Blood Forming Organs: Handbook of Experimental Pharmacol-
   It has previously been shown that ticlopidine, which is                   ogy Volume 101. Berlin, Springer-Verlag, 1991, pp 353-376
 similar to clopidogrel in structure and pharmacology,                   3. Daniel JL, Molish IR, Holmsen H: Myosin phosphorylation in
blocks the binding of [MC]ADP to a low-affinity site                         intact platelets. J Biol Chan 1981^256:7510-7514
                                                                         4. Hourani SMO, Cusack NJ: Pharmacological receptors on blood
 (KD>100 /iM) on human platelets.19 Experiments with                         platelets. Pharmacol Rev 1991;43:243-298
ring-labeled nucleotides are complicated by the possi-                   5. Macfarlane DE, Mills DCB: Inhibition by ADP of prostaglandin
bility that dephosphorylation by enzymes in plasma or                        induced accumulation of cyclic AMP in intact human platelets.
                                                                            / Cyclic Nucl Res 1981;7:1-11
on the platelet surface would lead to the generation of                  6. Cooper DMF, Rodbell M: ADP is a potent inhibitor of human
labeled adenosine, which is rapidly accumulated into                         platelet plasma membrane adenylate cyclase. Nature 19T7;282:
the cell and phosphorylated to ATP. It is possible that                     517-518
inhibition of this nonsaturable binding may in fact                      7. Macfarlane DE, Mills DCB, Srivastava PC: Binding of 2-azido-
                                                                             adenosine [^"PJdiphosphate to the receptor on intact human
represent the blockade of nucleoside uptake or of ADP                        blood platelets which inhibits adenylate cyclase. Biochemistry
metabolism.                                                                  198221:544-549
   We have found no correlation between the severity of                  8. Macfarlane DE, Srivastava PC, Mills DCB: 2-Methylthioadenosine
                                                                             O^PJdiphosphate: An agonist and radioligand for the receptor
the bleeding time response and the effect on adenylate                      that inhibits the accumulation of cyclic AMP in intact blood
cyclase activity or, indeed, between the cyclase effect and                  platelets. / din Invest 1983;71:420-428
the impairment of 2-MeSADP binding; more precise                         9. Figures WR, Niewiarowski S, Morinelli TA, Colman RF, Colman
measurements on larger numbers of subjects will be                           RW: Affinity labeling of a human platelet membrane protein with
                                                                            5-p-fluorosulfonylbenzoyl adenosine. J Biol Chem 1981;256:
required to examine this question. The variability in the                    7789-7795
speed of onset and the extent of the increase in the                    10. Bennett JS, Colman RF, Colman RW: Identification of adenine
bleeding time suggest that clopidogrel is metabolized at a                  nucleotide binding proteins in human platelet membranes by
variable rate in different subjects; the appearance in four                 affinity labelling with 5-p-fluorosulfonylbenzoyl adenosine. / Biol
                                                                            Chem 1978^53:7346-7354
subjects of prolonged bleeding times within 2 days of drug              11. Colman RW, Figures WR, Wu QX, Chung XY, Morinelli TA,
treatment suggests that this effect, if mediated by platelets,              Tuszynslti GP, Colman RF, Niewiarowski S: Distinction between
is due to an action on the mature cell rather than to a                     glycoprotein Ilia and the lOOkDa membrane protein (aggregin)
failure of normal development of the megakaryocyte.                         mediating ADP-induced platelet activation. Arch Biochem Biophys
                                                                             1988;262:298-306
   These results do not offer a convincing mechanism of                 12. Mills DCB, Figures WR, Scearce LM, Stewart GJ, Colman RF,
action for the antithrombotic thienopyridine com-                           Colman RW: Two mechanisms for inhibition of ADP-induced
pounds, but they do suggest some novel possibilities.                       platelet shape change by S'-p-fluorosulfonylbenzoyl adenosine.
Aggregin appears to be involved in ADP-induced shape                        JBiol Chem 1985;260:8078-8083
                                                                        13. Defreyn G, Bernat A, Delebassee D, Maffrand J-P: Pharmacology
change and exposure of fibrinogen-binding sites, but it                     of ticlopidine: A review. Semin Thromb Hemost 1989;15:159-166
plays no role in the action of clopidogrel. On the other                14. Gachet C, Cazenave J-P, Ohlmann P, Bouloin C, Defreyn G, Driot
hand, the fact that clopidogrel blocks the 2-MeSADP                         F, Maffrand J-P: The thienopyridine ticlopidine selectively pre-
binding site, a receptor through which ADP inhibits                         vents the inhibitory effects of ADP but not of adrenaline on cAMP
                                                                            levels raised by stimulation of the adenylate cyclase of human
adenylate cyclase, is clear; however, it is not known how                   platelets by PGE,. Biochem Pharmacol 1990;40:2683-2687
the blocking of this receptor modulates ADP-induced                     15. Defryn G, Gachet G, Savi P, Driot F, Cazenave J-P, Maffrand J-P:
aggregation. Although many aggregating agents, includ-                      Ticlopidine and clopidogrel (SR 25990C) selectively neutralize
ing platelet-activating factor, epinephrine, thrombin,                      ADP inhibition of PGEractivated platelet adenylate cyclase in rats
                                                                            and rabbits. Thromb Haemost 1991;65:186-190
and vasopressin, share the ability to inhibit adenylate                 16. Colman RF: Site specific modification of enzyme sites, in Boyer
cyclase, inhibition of this enzyme by the intracellular                     PD (ed): The Enzymes, Volume 19, ed 3. New York, Academic
"P"-site agonist 2'5'-dideoxyadenosine does not cause                       Press, Inc, 1990, pp 283-321
or potentiate aggregation.20 2-MeSADP inhibits adenyl-                  17. Colman RW: Aggregin: A platelet ADP receptor mediating acti-
ate cyclase at concentrations 100-200-fold lower than                       vation. FASEB J 1990^:1425-1435
                                                                        18. Figures WR, Scearce L, DeFeo P, Stewart G, Zhou F, Chen J, Daniel
are required to cause aggregation. Our data do, how-                        JL, Colman RF, Colman RW: Direct evidence for the interaction of
ever, suggest that a defect in the receptor mechanism                       the nucleotide affinity analog 5'-p-fluorosulfonyIbenzoyl adenosine
for adenylate cyclase inhibition may be associated with                     with a platelet ADP receptor. Blood 1987;70:796-803
impaired aggregation.                                                   19. Lips JPM, Sixma JJ, Schiphorst ME: Binding of adenosine diphos-
                                                                            phate to human blood platelets and to isolated platelet mem-
                                                                            branes. Biochim Biophys Ada 1980;628:451-467
                           References                                   20. Haslam RJ, Davidson MM, Desjardins JV: Inhibition of adenylate
 1. Gaarder A, Jonsen A, Laland S, Hellem A, Owren PA: Adenosine            cyclase by adenosine analogues in preparations of broken and
    diphosphate in red cells as a factor in the adhesiveness of human       intact platelets: Evidence for the unidirectional control of platelet
    blood platelets. Nature 1961;192:531-532                                function by cyclic AMP. Biochem J 1978;176:83-95

                             Downloaded from http://atvb.ahajournals.org/ by guest on November 7, 2015
Clopidogrel inhibits the binding of ADP analogues to the receptor mediating inhibition of
                                 platelet adenylate cyclase.
 D C Mills, R Puri, C J Hu, C Minniti, G Grana, M D Freedman, R F Colman and R W Colman

                             Arterioscler Thromb Vasc Biol. 1992;12:430-436
                                      doi: 10.1161/01.ATV.12.4.430
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
                                            Avenue, Dallas, TX 75231
                      Copyright © 1992 American Heart Association, Inc. All rights reserved.
                                  Print ISSN: 1079-5642. Online ISSN: 1524-4636

 The online version of this article, along with updated information and services, is located on the
                                        World Wide Web at:
                                     http://atvb.ahajournals.org/content/12/4/430

  Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
  Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the Copyright
  Clearance Center, not the Editorial Office. Once the online version of the published article for which permission
  is being requested is located, click Request Permissions in the middle column of the Web page under Services.
  Further information about this process is available in the Permissions and Rights Question and Answerdocument.

  Reprints: Information about reprints can be found online at:
  http://www.lww.com/reprints

  Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
  at:
  http://atvb.ahajournals.org//subscriptions/

                        Downloaded from http://atvb.ahajournals.org/ by guest on November 7, 2015
You can also read