Eur J . Biochem. 88, 543-554 (1978)

Interaction of Human Blood Platelets with the 2’,3’-Dialdehyde and 2’,3’-Dialcohol Derivatives of Adenosine 5’-Diphosphate and Adenosine 5’-Triphosphate P. Helen PEARCE, Judith M. WRIGHT, Christopher M . EGAN, and Michael C. SCRUTTON

Department of Biochemistry, University of London King’s Collcge (Rcceived February 23, 1978)

1. The 2‘,3‘-dialdehyde derivative of ADP (oADP) at concentrations approaching the millimolar range induces human blood platelets to undergo the transition from discoid to globular morphology (the ‘shape change’) but is incapable of inducing aggregation. 2. When incubated with platelets for 1 min before addition of the agonist, oADP acts as a competitive inhibitor of shape change and aggregation induced by ADP. Under these conditions secretion and hence aggregation induced by low concentrations of collagen ; and secretion and hence secondary aggregation induced by adrenaline, thrombin and vasopressin are also inhibited by this analogue. In addition, oADP stimulates the rate of primary aggregation induced by adrenaline and causes partial inhibition of primary aggregation induced by thrombin or vasopressin. When longer preincubation times are employed the extent of inhibition with respect to all agonists, except for high concentrations of collagen, is increased and the competitive character of the inhibition with respect to ADP is no longer apparent. 3. Incubation of human platelets with the 2’,3’-dialdehyde derivative of ATP (oATP) causes effects similar to those described for oADP except that the analogue neither induces platelet shape change, nor stimulates the rate of primary aggregation induced by adrenaline. In addition oATP fails to cause significant inhibition of platelet shape change induced by serotonin. The extent and character of inhibition caused by addition of oATP is not a function of the time of incubation. 4. The 2’,3’-dialcohol derivatives of A DP and ATP (orADP and orATP) effect the aggregation properties of human blood platelets in a manner generally resembling those observed for the 2’,3‘-dialdehyde analogues. However, orADP is only weakly effective in causing platelet shape change and stimulating the rate of primary aggregation induced by adrenaline and does not inhibit secretion induced by adrenaline, collagen, thrombin and vasopressin. The extent of inhibition by OI-ADP increases only slightly with increased time of incubation. 5. The data suggest that oADP acts as a partial agonist, and oATP as an antagonist, at the platelet ADP receptor, but that platelet membrane stabilisation also results from interaction with these dialdehyde analogues. Such membrane stabilisation does not complicate the interaction of platelets with orADP, which appears to act as a classical antagonist for the ADP receptor.

Many agonists are capable of inducing platelet aggregation including adrenaline, thrombin, vasopressin and serotonin. In citrated plasma and under suitable conditions these agonists induce a biphasic response in which only the first, or reversible, phase is the direct result of platelet-agonist interaction. This initial interaction causes secretion of platelet constituents some of which are themselves inducers of aggregation and hence cause a further response associated ~~

-~

Ahhwviu/iorw. oATP and oADP. the 2’,3‘-dialdehyde derivatibe of ATP and ADP; orATP and orADP, the 2’,3‘-dialcohol derivative of ATP and ADP.

with secretion of the contents of the dense granules and for some agonists also of the lysosomes (cf. [I]). In addition platelets will adhere to certain particulate macromolecules, notably to collagen fibres. Such adhesion causes secretion of platelet constituents, which then promote aggregation and further secretion, thus enhancing growth of the platelet aggregate. Several lines of evidence support the postulate that ADP is one of the platelet constituents which has an important role in the aggregation and secretion response. Addition of a system which utilises ADP, e.g. creatine phosphate + creatine kinase, abolishes aggregation induced by collagen and inhibits the sec-

544

ADP Analogues and Human Blood Platelets

that platelet-rich and platelet-poor plasma gave 5 :d and 100% full scale deflection respectively with the recorder set for full scale deflection to a 10-mV signal. For measurement of shape change the channels were calibrated so that with a recorder setting of 3 mV full scale deflection, the difference in absorbance for maximum shape change represented approximately 50% of maximal recorder deflection. EDTA (4 mM) was added to the platelet-rich plasma 30 s before other reagents to prevent aggregation. The extent of aggregation was determined as the total decrease in absorbance, measured between the midpoints of the oscillations in the recorder trace, after addition of the agonist. The extent of shape change was measured similarly, but as the maximal increase in absorbance. The rate of aggregation was measured as the maximal rate of decrease in absorbance after the addition of the agonist. In a biphasic aggregation response the maximal rate of each phase of aggregation was measured. Secretion was measured by a modification of the procedure of David and Hirion [14] using plateletrich plasma labelled with [3H]serotonin as described above. Before and after addition of agonists platelets were sedimented by centrifugation at 12000 x g for 1 min using the Eppendorf microcentrifuge. The supernatant fraction was immediately removed and dupliMATERIALS AND METHODS cate aliquots (0.1 ml) taken for estimation of 3H content using a Triton X-lOO/toluene (1/2, v/v) counting Venous blood was obtained from healthy donors solvent containing 4.0 g PPO per I. who had taken no drugs during the previous two oADP (oATP) was prepared by periodate oxidation weeks. Blood (9 vol.) was immediately mixed with of ADP (ATP) respectively as described by Easter1 vol. acid citrate/dextrose (0.007 M citric acid, brook-Smith et al. [15] ADP (ATP) (0.2 minol) was 0.093 M sodium citrate, 0.139 M glucose pH 6.5), dissolved in 10 nil water and the solution adjusted to [13] containing, in cases where secretion was also measured, 0.005 - 0.01 $3 5 '-hydr~xy-[G-~H]trypt- pH 7.0. Solid sodium periodate (0.22 mmol) was added and the solution was stirred at 2 "C in the dark for 1 h. amine creatinine sulphate and centrifuged at 20 - 25 'C The reaction was stopped by the addition of 0.10 mmol and at 200 x g for 15 min. For studies in which seroglycerol. The solution was concentrated to 1.5 ml tonin or vasopressin were used as agonists blood was on a rotary evaporator and then applied to a Sephadex mixed with 0.01 vol. heparin (500 U/ml in isotonic G-10 column (35 x 2.2 cm) previously equilibrated saline) and centrifuged as above. Platelet-rich plasina with 1 mM sodium phosphate pH 7.4. The column was was removed with a siliconised Pasteur pipette and eluted with the same buffer and fractions containing stored at 20-25 "C in a tightly covered tube after nucleotide but free of iodate were pooled, concentrated flushing with 95 air/5 % CO2 to prevent an increase on a rotary evaporator and stored at - 70 "C. in plasma pH. For studies of secretion incubation was The 2',3'-dialcohol derivatives of ADP (ATP) were continued to give a total exposure time of 60- 70 min prepared by incubating oADP (oATP) with a 10-fold such that at least 60% of the added L3H]serotonin excess of sodium borohydride at pH 8.2 for 30 min at was taken up by the cells. Platelet-poor plasma was 2 T . The disappearance of aldehyde groups was obtained by centrifugation of some of the remaining confirmed by incubating an aliquot of this system blood for 2 min at 12000 rev./min in an Eppendorf with 0.4 0 4 2,4-dinitrophenylhydrazine in 2 M HCI 3200 microcentrifuge. followed by addition of lo:( NaOH [16]. The soluAggregation and shape change were measured at tion was then adjusted to pH 7.4 with 6 M HCI to 3 7 ' C in a Payton dual-channel aggregometer condestroy excess borohydride, applied to a Sephadexnected to a Fisher Recordall recorder (series 5000). G10 column and chromatographed as described above. Platelet-rich plasma (0.25 ml) was stirred at 900 rev./ The concentrations of the nucleotides were estimin in 1-ml cuvettes and additions made as indicated mated from the absorbance at 258 nm, using c after equilibration to temperature. For measurement = 14900 cm-' M-' [17]. Iodate was estimated by of aggregation the instrument was calibrated such

ond phase of aggregation induced by agonists such as adrenaline and vasopressin. However, the initial reversible phase of the response to these latter agonists is not affected by addition of the ADP-utilising system [2 - 41. A similar pattern of inhibition is observed for antagonists which are believed to interact specifically with the ADP receptor, e.g. ATP [5]. Thus the plateletADP interaction has presumed importance to the initiation of haemostasis in vivo. The parameters describing binding of ADP to intact human platelets or to platelet membrane fractions have been reported by several workers [6 - 91 and attempts have been madeto identify themembrane component(s) which serve as receptor(s) for this nucleotide. However until recently the affinity labelling technique [lo] has not been widely applied to this problem [ l l ] . We have therefore prepared the 2',3'-dialdehyde derivatives of ADP (oADP) and AT€' (oATP) with a view to the possibility of using these analogues as affinity labels for the ADP receptor(s). The studies reported here, which have previously been the subject of a short communication [12], define the properties and specificity of interaction of these and related analogues with human blood platelets.

545

P. H Pearce, J . M. Wright, C . M . Egan. and M . C. Scrutton

adding 2 p1 of each fraction to 5.5 ml water containing 0.04% starch, 9 mM KI and 1 drop of 1 M HCl. Preparations of oADP and orADP were assayed for ADP using pyruvate kinase and lactate dehydrogenase [18]. The maximal extent of contamination observed was 0.04% and most preparations had an apparent content of 0.005-0.01 ADP. Addition of authentic ADP as an internal standard at similar concentrations demonstrated that the presence of oADP or orADP did not interfere with the estimation. Preparations of oATP and orATP were assayed for ATP using luciferase essentially as described by Strehler [19]. No ATP was detectable under conditions in which 0.01 "/, contamination would have been observed as indicated by the response to authentic ATP added as an internal standard. The purity of the 2',3'-dialdehyde and 2',3'-dialcohol derivatives was checked by chromatography on polyethyleneimine thin-layer sheets in 0.8 M NH4HC03 (11) and cellulose thin-layer sheets in isobutyric acid/animonia/water (66/1/33, by vol.) (I). Although extensive streaking of the 2',3'-dialdehyde derivatives is observed in system I the nucleotide spot as defined by ultraviolet absorbance coincided exactly with the area which stained with the dinitrophenylhydrazine reagent. Furthermore, after reduction to the 2',3'dialcohol derivatives single coherent spots were obtained in both systems which did not stain for aldehyde and which had an R,=significantly different from that of the parent nucleotide. These data establish the purity of the preparations employed in the subsequent studies. Stock solutions of ADP, serotonin, thrombin and vasopressin were prepared in 0.154 M NaCl and stored at - 15 "C. The stock adrenaline solution (1.25 mM in 0.154 M NaCl) was prepared on the day of use. Collagen was prepared by homogenization in 0.154 M NaC1, followed by centrifugation at 200 x g for 15 min. The concentration of collagen suspended in the supernatant fraction was estimated by the Biuret method after precipitation of the protein with trichloroacetic acid [20]. It should be noted that due to the high proline and hydroxyproline content of collagen, this procedure underestimates the true protein concentration by a factor of approximately 22.5-fold.

Clrenzicuh ATP, L-adrenaline bitartrate, collagen (insoluble type V, bovine Achilles tendon), synthetic lysine vasopressin, serotonin and heparin were obtained from Sigma Chemical Co. ADP was obtained from PL Biochemicals Inc. Cellulose thin-layer sheets containing a fluorescent indicator were obtained from Eastman-Kodak Co.

and polyethyleneimine sheets from CamLab Ltd. 5-Hydroxy [G-3H]tryptamine creatinine sulphate (500 Ci/mol) (TRA. 223) was obtained from the Radiochemical Centre. Human thrombin (MRC reagent 66/305) was the kind gift of the National Institute of Biological Standards and Control.

RESULTS

Eflects of the 2',3'-Dialdehyde and 2',3'-Dialcohol Derivatives of A D P and ATP on Human Platelets. Interaction of These Analogues with ADP When microinolar concentrations of ADP are added to citrated human platelet-rich plasma the discshaped native platelets are rapidly transformed into spheres from which pseudopodia protrude ('shape change') [21]. This shape change of platelets is indicated in the aggregometer recorder tracing by a decrease in light transmission and a loss of amplitude of the oscillations which are caused by the movement of the stirred platelets in the light path. Similar changes are observed after the addition of millimolar concentrations of oADP to platelet-rich plasma, but no such effect is observed on addition of these concentrations of oATP. The maximal shape change inducible by oADP is 60 - 70 7; of that observed with a saturating concentration of ADP while a half-maximal response to this analogue is obtained at a concentration of 0.5 mM (Fig. 1 A). It is important to note that although some ADP appeared to be present in the oADP preparations, the maximal amount which could have been added (0.2 pM in 1 m M oADP) is insufficient to account for the full extent of shape change observed. At most approximately one quarter of the response to oADP could in principle be attributed to the contaminant ADP based on the data of Fig. 1 B. However, this calculation does not accord with the observation that orADP, which appears to be contaminated with ADP to the same extent as the dialdehyde derivative, causes a barely detectable ( < 5 %) shape change response. Since orADP is a less effective inhibitor of shape change induced by ADP than is oADP (Table 1) it appears that calculations based on the extent of apparent contamination overestimate the contribution of residual ADP to the shape change response induced by oADP (Fig. 1 A). It is conceivable that this discrepancy may be due to slight reaction of oADP in the pyruvate kinase reaction hence leading to an overestimate of the contamination of preparations of this nucleotide by ADP. oADP also causes inhibition of the shape change induced by ADP. As shown in Fig. 1 A, the prior response to oADP added 1-2 min before ADP diminishes the extent of response to the latter agonist. At

546

A D P Analogues and Human Blood Platelets

1x)

.

100

-p

-

+ C

Table 1. hhihition hq' A D P and A T P analogues of aggregation and shape change induced hq' 5 p M ADP Platelet-rich plasma (prepared using the anticoagulant as indicated) was incubated at 37 "C with increasing concentrations of the inhibitors for 1 min before the addition of 5 1tM ADP. The total extent of aggregation or of shape change was measured in separate experiments with 4 mM EDTA included in the system for measurement of shape change. The extent of the response was plotted against the concentration of inhibitor and the concentration of inhibitor causing 50% inhibition was estimated. The figures in parentheses indicate the range of the values and the number of determinations

A

-\

80 0

0

. e

\ Analogue

Anticoagulant Concentrations causing SOU,; inhibition ([I],).))of employcd _ _ . aggregation shape change

;MI m r 0

~

Q W

-

m

r

*

mM

40

~-

~~

oADP _ _

. ~ ~~.~ .

20

~

oATP ~~

0

citrate heparin

~-

~

orADP 0

0

2 3 [ o A O P ] (mM)

1

5

10

15 20 [ADPI (CLM)

4

25

5

100

Fig. 1. Shupeclrangc, indncetlhj i l D P o r r d < i / l D P .( A ) Increasing concentrations of oADP were incubated at 37 C with citraled platcletrich plasma containing 4 mM EDTA Ihr 2 min before the addition of 5 pM ADP. Shape changes induccd by oADP ( 0 )and by suhsequcnt addition ol' 5 pM ADP (0) were measured as a perccnlagc of the shapc change induced by 5 pM ADP in thc absence of inhibitor. Total shape change (m)represents the sum of the maximum extent of shape change induced by oADP at the concentrations indicated and then by 5 LLMADP. (B) Shape change induccd by increasing concentrations of ADP in citrated platelet rich plasma containing 4 m M EDTA was measured in the absence of oADP ( 0 ) or 2 min after the addition of 1.6 mM oADP (0). The results are expressed as a percentage of the value obtained with the saturating concentrations of ADP in the absence of inhibitor

~~

~

orATP

~

~

~

citrate heparin ~. ..

. .

0.8 (0.7-0.9) (3) 1.4 (0.9- 1.8) (4) .~ _ _ - 0.4 (0.3-0.6) (4) O.q(O.7-1.0)(3) - - -

citrate ~

~

.~.

..

citrate

~

-~ ~.

2.7 (1.7-3.5) (4) 1.1 (0.8-1.3)(5)

~-~-

~~~~

1.2(1.0-1.5)(4) ~- -

~

-

~

~~

- - - -

0.7 (0.6-0.8) (3) 1.8 (1.3-2.1) (4)

~~

~~

1.3(1,2-1.5)(4)

.

~

~

3.9 (3.5-4.2)(3) ~

~

~

-

~

-

~

-~

1.2(2.0-1.3)(3)

concentrations of oADP below 1.5 mM the total response to oADP + ADP approximates 100% inindicating that only a finite effect can be induced and that the effects of the two agonists are not additive. In the presence of concentrations of oADP in excess of 2 niM an actual decrease in total response is observed since the extent of response to 5 pM ADP decreases without further significant increase in the response to the 2',3'-dialdehyde analogue (Fig. 1A). This suggestion of inhibition by the less effective agonist is confirmed by measurement of the concentration dependence of the response to ADP in the presence of a fixed concentration of oADP (Fig. 1 B). The concentration of ADP required for half-maximal response is increased from 0.5 pM to 0.9 pM in the presence of 1.6 mM oADP. The response to ADP observed in the presence of oADP does not approach 100% at high concentrations of ADP (Fig. 1 B) since the addition of this analogue itself causes a partial shape change (cf. Fig. 1A). Incubation of platelet-rich plasma with oADP at inillimolar concentrations does not cause the platelets to aggregate. However, it leads to severe inhibition of aggregation induced by the subsequent addition of 5 pM ADP, a concentration sufficient to elicit a maximal response in the absence of the inhibitor. Approximately 50 % inhibition of the response to this concentration of ADP is observed in the presence of 0.8 m M oADP (Fig. 2A). The apparently biphasic nature of the relationship between extent of aggregation and oADP concentration reflects the marked sensitivity

.

~

541

P. H. Pearce. J . M. Wright, C. M. Egan, and M. C. Scrutton

I 100

80

-

-z

I

0 c

"

5 c ._

60

0

c

m

0

E0 40 P 20

0 0

2

4

3 [oADP) (mM)

1

5

0

1

2

3

4

5

6

( o A T P ] (mM)

Fig. 3. Ejfrct o f o A TP o/i .drupe c/iun~r,uiid uggwgu/ion induc,c.t/ by A D P in citrutrd plate1i.t-rich plasma. Citrated platelet-rich plasma was incubated at 37 'C with oATP at the concentrations indicated for 2 min before the addition of 5 pM ADP. For the measurement of shape change (0)4 mM EDTA was included in the system. The total extent of aggregation (A)was measured in a separate experiment. The values are expressed as a percentage of the corresponding value obtained with 5 UM ADP in the absence of inhibitor

I

0

20

40

I

I

60 80 [ADPI ( W )

100

200

Fig. 2. Iriliihition l>j. o A D P of qqgrc,gu/irm i d u c d h j 5 jtM A D P . ( A ) The total extent of aggregation ( 0 )and rate of primary aggregainduced by 5 pM ADP were measured 1 min after the tion (0) addition of increasing concentrations of oADP to citrated plateletrich plasma. (B) The rate of primary aggregation induced in citrated platelet-rich plasma by increasing concentrations of ADP was measured in the absence of oADP (0) or l min after the addition of 0.65 (W), 1.3 (0) and 2.8 (0) mM oADP. All values are expressed as a percentage of the corresponding value obtained with 5 p M ADP in the absence of inhibitor

of secondary aggregation which is completely inhibited by 0.9 mM oADP. Inhibition caused by oADP appears competitive with respect to ADP, since as shown in Fig.2 B in the presence of various concentrations of oADP in the range 0.7 - 2.8 mM, a maximal aggrega-

tion response can be obtained if the concentration of ADP is increased to 100-200 pM. The concentration of ADP required to give 50 maximal response is increased from 1.5 pM to 22 pM under these conditions (Fig.2B). Data similar to those shown in Fig.2B are obtained if the extent of aggregation is measured as a function of ADP concentration in the presence of various concentrations of oATP or of orADP. Furthermore these latter analogues also cause inhibition of shape change induced by ADP as shown for oATP in Fig. 3 . In contrast to the effect of oADP (Fig. 1 B), oATP causes essentially complete inhibition of shape change induced by ADP although when citrate is used as the anticoagulant the concentration required to give 50 '%, inhibition ([I]o.s) of shape change is nearly an order of magnitude greater than that which gives 50% inhibition of aggregation (Fig. 3 ) (Table 1). This difference in [I]o.~is not however observed when heparin is employed as anticoagulant. A similar, although less marked, difference in [I]o.s for aggregation and shape change induced by ADP is observed when orADP is used as the inhibitor in citrated platelet-rich plasma but no significant difference is observed for oADP (Table 1). Table 1 also demonstrates that although all the derivatives tested are inhibitory in the millimolar

ADP Analogues and Human Blood Platelets

\

0

m

40

60

Time (min)

Fig. 4. Aggrogoiiori iiidwcd / I ) , A D P u/tc i' r i i c d x i r i o i i of p/utc+r-rdf plusmu f b r varj'/ri,q rinws with oADP or OA' TP. Platelet-rich plasma was incubated at 37 'C with 0.6 mM oATP or 1.6 mM oADP and ADP was added at the times indicated. The total extent of aggregation was measured as a percentage or the corresponding value obtained in the absence of inhibilor. Similar results were obtained whether or not the platelet-rich plasma was stirred throughout the S FM ADP; ( 0 ) 0 . 6 inM period ofincubation with the inhibitor. (0) oATP. S pM A D P ; (A) 1.6 mM oADP. 5 pM ADP; (m)0.6 m M oATP, 50 pM ADP; (A)3 . 6 mM oADP, 50 pM ADI'

range the 2',3'-dialdehydes are generally somewhat more effective than the 2',3'-dialcohol analogues. It should be noted that inhibition caused by oADP or orADP cannot be attributed to induction of a refractory condition [22] by the small amount of ADP which may be present in the preparations of these nucleotides. The extent of inhibition observed is independent of the time of preincubation with these analogues provided that this time does not exceed 1 min at 37 'C. Furthermore the same extent of inhibition is observed whether or not the system is stirred during the preincubation period. Control experiments demonstrated that the refractory condition only developed on exposure of unstirred platelets to ADP (cf. L221). If the tiine of incubation of the platelet-rich plasma with oADP prior to addition of ADP is increased beyond 1 - 2 min, the extent of the response to either low ( 5 pM) or high (50 pM) ADP is then decreased (Fig. 4). The decrease in response observed on incubation with this analogue occurs in parallel for both the low ( 5 pM) and high (50 pM) concentrations of ADP with a half-time of approximately 12 min. This effect there-

fore differs from the competitive relationship observed when ADP is added 1 min after oADP (Fig. 2B). A similar, but much less marked, increase in inhibition is observed on prolonged incubation with orADP but no such effect is observed on incubation with oATP (Fig. 4). The effect illustrated for oADP in Fig.4 appears similar to that reported previously by Salzmann et al. [23]for 5'-AMP. In the case of 5'-AMP the increased inhibition can be blocked by addition either of CN[23] or of adenosine deaminase [24] indicating that the effect is due to generation of adenosine which is believed to cause inhibition of platelet function primarily by increasing the concentration of intracellular 3': 5'-AMP (adenosine 3': 5-monophosphate) [25]. We have therefore examined the effect of CN- on the timedependent increase in the extent of inhibition observed when platelet-rich plasma is incubated with AMP or oADP. In accord with the findings of Salzmann et al. 1231 addition of 5 mM CN- decreased the extent of inhibition caused by 0.05 mM AMP provided that the system is stirred at 37 "C during the 10-min preincubation period. If an unstirred system is employed CNis ineffective or in some cases tends to enhance the effect of AMP. In a stirred system 5 mM C N - also decreased the increase in the extent of inhibition caused by 10 min incubation of the platelet-rich plasma with 1.1 mM oADP (Fig.4). However this observation is not definitive since CN- would be expected to react with the aldehyde groups of this analogue to form a cyanohydrin and hence the observed effect could be attributable to removal of the inhibitory agent. Indeed incubation of 23 mM oADP with 104 mM CN- for 10 min at 37 "C before addition of platelet-rich plasma markedly decreases the extent of inhibition.

Phase Contrast hlicroscopy Since observation of shape change by changes in light transmission can give misleading results, we have confirmed qualitatively the conclusions described above using phase contrast microscopy. When the agent under examination is diffused into platelet-rich plasma, the discoid, translucent cells become spherical and opaque if the agonist induces shape change. Such an effect is observed on addition of ADP, oADP and N-ethylmaleimide but little, if any, such response can be observed on addition of orADP, and no response occurs on addition of oATP. Effict of'oADP, oATP, and o r A D P on the Properties of Aggrcgatiou Induced hy Otlier Agonkts

Further insight into the mechanism of inhibition by oADP, oATP and orADPwas obtained by examination of the effects of these analogues on aggregation

P. H. Pearce, J. M. Wright, C. M . Egan, aiid M. C. Scrutton

induced by agonists other than ADP. In these studies we have measured both aggregation and secretion in order to discriminate between effects on the aggregation response which result from blockade of the ADP receptor with consequent impairment of the ability to respond to secreted ADP and those which result from prevention of the secretion response itself. The data obtained in such studies are shown as aggregation traces in Fig. 5 for adrenaline, collagen, vasopressin and thrombin as agonists with the extent of secretion expressed as a percentage of platelet-bound 3H released by the agonist shown in parentheses next to each trace. It is apparent that for all four agonists the 2',3'-dialdehyde analogues cause marked inhibition of secretion. Hence the inhibition of overall (collagen) or secondary (adrenaline, vasopressin, thrombin) aggregation (Fig. 5) is a secondary consequence of the diminished extent of secretion which results from addition of these analogues. This interpretation is supported by studies (data not shown) in which the aggregation and secretion responses were measured as a function of the concentration of oADP or oATP. Such studies demonstrate a reasonable correlation for a given inhibitor concentration between the extent of inhibition of secretion and the extent of inhibition of the rate and extent of aggregation induced by collagen or of the rate of secondary aggregation induced by adrenaline, vasopressin and thrombin, except that the aggregation response is somewhat more effectively inhibited in the case of collagen and thrombin. Furthermore the concentrations of the 2',3'-dialdehyde analogues required to observe 50 inhibition of the aggregation and secretion responses to these agonists are consistent with those which cause 50 '%, inhibition of aggregation induced by ADP (Table 1). In contrast orADP which is an effective inhibitor of aggregation induced by collagen and of secondary aggregation induced by thrombin and adrenaline, has little if any effect on the extent of secretion as measured by release of [3H]serotonin (Fig. 5). Secondary aggregation induced by vasopressin is also inhibited by orADP without concomitant inhibition of secretion although the effect is less marked than that observed for the other agonists (Fig. 5C). Studies of the concentration dependence of inhibition by orADP (data not shown) have shown that [1]0.5 measured for aggregation induced by collagen and secondary aggregation induced by adrenaline and thrombin is in reasonable agreement with that obtained for inhibition of aggregation induced by ADP (Table 1). It is also apparent from Fig. 5 that effects are observed on the primary phase of aggregation induced by adrenaline, vasopressin and thrombin on addition of the 2',3'-dialdehyde and -dialcoho1 analogues. For example (Fig. 5A) oADP causes a marked stimulation of the rate of primary aggregation induced by adrena-

549

line. This effect is mare clearly documented in Fig.6 which shows that it occurs over a much lower range of concentration than that over which oADP causes inhibition of secondary aggregation. A much less marked stimulation of the rate of primary aggregation is observed in the presence of orADP, thus excluding the possibility that the effect illustrated in Fig. 5A and 6 could be due to ADP present as a contaminant in the oADP preparation. Neither oATP nor orATP have a significant effect on the rate of primary aggregation induced by adrenaline. Additionally both oADP and oATP decrease the rate and extent of primary aggregation induced by vasopressin and thrombin (Fig. 5 C and D). Studies of the extent of inhibition as a function of oADP or oATP concentration (data not shown) demonstrate that the effect occurs over a range of concentration similar to that for which inhibition of secondary aggregation is obtained but that complete inhibition of primary aggregation induced by these agonists is not observed. orADP causes some inhibition of the rate of primary aggregation induced by vasopressin but has no significant effect on that induced by thrombin (Fig. 5 C and D). In the experiments illustrated in Fig.5 and 6, the agonists were added after the platelet-rich plasma had been incubated for 1 min at 37 C with the nucleotide analogue. When the time of preincubation with oADP is increased prior to addition of adrenaline, thrombin or a low concentration (50 - 60 &ml) of collagen, the extent of inhibition becomes more marked as observed for ADP (Fig.4) and the specificity of the effects are diminished. Addition of higher concentrations of adrenaline and thrombin have no effect on the increase in inhibition caused by prolonged incubation with oADP but this inhibition is reversed on addition of a high collagen concentration (300 pg/ml). We have also examined the effect of oATP on the extent of shape change induced by serotonin as a further test of specificity. The data obtained (not shown) demonstrate that this analogue has little effect on the response to serotonin under conditions where marked inhibition of shape change induced by ADP is observed.

DISCUSSION The data presented provide a comprehensive account of the properties of interaction of human platelets with ADP and ATP modified at the 2' and 3' positions of the ribofuranosyl ring. orADP has the properties of a weak, but specific antagonist at the ADP receptor with slight inhibition of primary aggregation induced by vasopressin as the only evidence of non-specificity (Fig. 5). The effects of the 2',3'-dialdehyde analogues arc more complex. The data of

ADP Analogues and Human Blood Platelets

550

A

2 rnin

d

e

2 min

Fig. 5. Eflects ojoADP, oA TP, undorADP 011 uggrcgclrion undsecrc~iioni n d i ~ e d b yvnrious ugonisIs. In all cases the figures in parentheses indicate the extent of secretion expressed as a percentage of'ttie platelet-bound [3H]serotoninwhich is found in the supernatant fraction after exposure to the agonist in the presence or absence of the ADP analogue. The platelets took up 6 0 - 7 8 x of the 'H added during the initial incubation. (A) Aggregation and secretion induced by adrenaline. The additions made at the arrows as indicated were (a) 1 pM adrenaline, (b) 1.4 mM oADP, (c) 3.8 mM orADP, (d) 1.1 mM oATP. (B) Aggregation and secretion induced by collagen. The additions made at the arrows as

P. H Pearce, J . M. Wright, C. M. Egdn. and M. C. Scrutton

C

f

b

551

f

c

f

2 min

e-----,

2 min

indicated wcre (e) 53 pg collagcn/ml; (b) 1.1 niM oADP, (c) 5.5 m M orADP, (d) 1.0 mM oATP. (C) Aggregation and secretion induced by vasopressin. The additions made at the arrows as indicated were (f) 2 mU/nil vasopressin; (b) 1.5 mM oADP, (c) 3.8 m M orADP, (d) 1.5 m M oATP. (D) AggreyatIon and secretion induced by thrombin. The additions made at the arrows as indicated were (g) 400 m U Ihroinbiniml; (b) 1.1 m M oADP, (c) 5.5 mM orADP, (d) 1.0 niM oATP

ADP Analogues and Human Blood Platelets

552 200

-e -

0

150 /

c

s

I

.+-

/

0

d .- 100 0

m m

E

m 4

50

A

\

.

... Q

-..-..

0

- -* I

0

1

2 (oADP] (mM)

-

3

Fig. 6. .Fffrc,/ of iiic.rco.ting u m c ’ w / i ~ t i c ~ / i .of \ o t l D P 011 aggwgarion in(liic.ecf hy u d ~ w d i n cPlatelet-rich ~. plasma was incubated at 37 ‘C with oADP at the concentrations indicated for 1 inin before the addition of adrenaline. The total extents of aggregation and rates of the primary and secondary phases are expressed as a percentage of the corresponding values obtained with 5 pM adrenaline in the absence ofoADP. (m) 5 pM adrenalinc, total extent of aggregation; ( 0 )5 pM adrenaline, rate of primary phase; (A) 5 pM adrenaline, ritte of secondary phase; (0)1 pM adrenaline, total extent of aggre1 pM adrenaline. rate of primary phase; (A) 1 pM adregation; (0) naline, rate of secondary phase

Fig.1 and 2 suggest that these derivatives interact with the ADP receptor and in the case of oADP also mimic the initial feature of the response to ADP. Their effects on aggregation induced by collagen and secondary aggregation induced by adrenaline, vasopressin and thrombin (Fig. 5 ) are however attributable to inhibition of secretion rather than to blockade of the ADP receptor. Two lines of evidence suggest that inhibition of secretion results from the initial interaction at the ADP receptor. First, a similar concentration dependence is observed for both effects (Table 1, Fig. 6). And second a simple dialdehyde such as glutaraldehyde which lacks any of the groups which might be expected to direct interaction at the ADP receptor shows none of the features of specific inhibition (e.g. Fig.2B, 5 and 6) which characterise the interaction of human platelets with oADP and oATP (P. H. Pearce and M. C. Scrutton, unpublished results). Thus it seems possible that interaction of these analogues at the ADP receptor allows the highly reactive aldehyde groups to react with components of the membrane with a consequent stabilisation which prevents secretion. This latter effect is similar in principle to that

proposed by Mills and Roberts [26] to explain the effect of drugs such as chloropromazine and imipramine on the response of human platelets to ADP and adrenaline although it should be noted that in contrast to oADP and oATP these membrane-stabilising drugs do not affect the primary phase of aggregation induced by A D P but d o decrease that induced by adrenaline. The stimulation by oADP of the rate of primary aggregation induced by adrenaline (Fig. 6) appears to be related to the status of this analogue as a partial agonist at the ADP receptor since no such effect is observed on addition of oATP. Potentiation by adrenaline of the response of human platelets to ADP is well established 1271 while we have recently demonstrated that low concentrations of ADP potentiate the primary rate of aggregation induced by adrenaline (J. A. Grant and M. C. Scrutton, unpublished results). The inhibition of the rate and extent of primary aggregation induced by thrombin and vasopressin (Fig. 5 C and D) which is observed on addition of both oADP and oATP appears to represent a more serious degree of non-specificity than that which characterises the interaction with orADP. However, one should note that other ADP analogues, e.g. 5-fluorosulphonylbenzoyladenosine, also cause inhibition of primary aggregation induced by thrombin [ l l ] and that interaction between the effects of thrombin and ADP in the platelet response is well documented [28 - 301. The data presented here taken together with other reports of the effects of nucleotide analogues on platelet aggregation enables us to begin to outline some features of the nature of the ADP-receptor interaction. Modification of the ribofuranosyl ring either as reported here or by removal of hydroxyl groups on either C-2‘ or C-3’ gives rise to analogues which are weak agonists or partial agonists having association constants 20- 30 times less favourable than that observed for ADP (Fig. 1 and 2, Table 1) 131,321. The integrity of the ribofuranosyl ring appears to be as important as the presence of the hydroxyl groups on C-2’ and C-3‘ since the 2’,3’-dialcohol analogues are if anything less effective than the 2‘,3‘-dialdehyde derivatives (Table 1). Modification of the C-1, C-6, or C-8 on the adenine ring also gives rise to analogues which are either weak agonists or antagonists but substitution at C-2 on this ring enhances the efficiency of the response as compared with that observed for ADP [33 - 361. The pyrophosphate group seems likely to be involved primarily in initiation of the response since the presence of an additional phosphate (as in ATP) or the replacement of this group by a fluorosulphonylbenzoyl moiety leads to production of effective antagonists having association constants approximately ten times less favourable than that for ADP [5,11]. However some role in binding is indicated by the observation that the a,P-phosphate analogue of ADP is a weak agonist having an association constant

553

P. H Pearce, J. M . Wright, C. M. Egan, and M . C. Scrutton

similar to that observed for the derivatives containing modifications to the ribofuranosyl ring [37]. These considerations suggest an important role for the ribofuranosyl ring in the ADP-receptor interaction. Hence although modifications to this ring severely impair the efficiency of platelet-analogue interaction, reactive groups in these analogues, such as the dialdehyde groups in oADP or oATP, are likely to be brought into close proximity with the amino acid residues which comprise the receptor site. Thus the observation that oADP is a partial agonist at the ADP receptor (Fig.1 and 2) could be explained if reaction of the aldehyde groups of this analogue with lysyl or cysteinyl residues in the receptor 'freezes' the platelet in a conformation which is unable to undergo secretion and hence aggregate formation. Although this postulate appears somewhat at variance with the competitive relationship observed between oADP and ADP provided that this agonist is added soon after the dialdehyde analogue (Fig. 2B), the products of reaction of aldehydes with amino or sulphydryl groups are known to be labile unless stabilised by further reaction. Alternatively we may postulate that shape change is not an obligatory discrete step in the aggregation response to ADP as has generally been supposed. This latter suggestion appears consistent with the observation that in citrated platelet-rich plasma oATP is a more effective inhibitor of aggregation than of shape change induced by ADP (Table 1). A similar effect has been observed by MacIntyre et al. [38] for the 2-amylthio derivative of AMP and by ourselves for thiol reagents such as p-chloroniercuriphenylsulphonate. Prolonged incubation of the platelets with oADP prior to the addition of agonist causes a marked increase in the extent, and a decrease in the specificity, of inhibition (Fig.4). A similar effect is not observed for oATP (Fig.4) in accord with the finding by MacFarlane and Mills [5] that the extent of inhibition by ATP does not increase with time of incubation. Recent studies [39] in which intact platelets have been incubated with [3H]oADP or [3H]oATP may provide an explanation for this effect in the case of the 2',3'dialdehyde analogues. The extent of incorporation of [3H]oADPinto several polypeptides increases with time of incubation. In contrast for [3H]oATPthe extent of incorporation, which appears more specific, shows little time dependence. Although such non-specificity of incorporation with time appears the most likely explanation for the data of Fig.4 we cannot at present exclude the possibility that the effect is due to generation of the 2',3'-dialdehyde derivative of adenosine, or possibly to refractoriness [22] induced by interaction of the platelets with oADP. These studies were supported by a grant rrom the Medical Research Council.

REFERENCES 1. MacIntyre, D. E. (1976) in Platelets in Biology and Pathology (Gordon, J. L., ed.) pp. 61 -85, North Holland, Amsterdam, New York, Oxford. 2. Holmsen, H. (1972) in Clinics in Huenzaiolrigy (O'Bricn, J . R., ed.) pp. 235-266, W. B. Saunders Co.. Philadelphia, London, Sydney. 3. Izrael, V., Zawilska, K., Jaisson, F., Levy-Toledano, S. & C a m , J. (1974) in Platelets: Production, Function, Trun.$u.sion und Storuge(Baldini, M. G. &Ebbe, S.,eds)pp. 187- 196, Grune and Stralton, New York. 4. Packham, M. A , , Kinlough-Rathbone, R. L., Reimers, H. J., Scott, S. & Mustard, J . F. (1978) in Prostuglunriins in Herniltology (Silver, M. J., Smith, J. B. & Kocsis, J. J., eds) Spectrum Publishing Co., Cocheton, N.Y., in the press. 5. MacFarlane, D. E. & Mills, D. C. B. (1975) Blood, 46, 309320. 6. Born, G. V. R . (1965) Nature (Lond.) 206, 1121 - 1122. 7. Feinberg, H. (1973) Br. J . Pharmucd. 47, 65p. 8. Nachman, R. & Ferris, B. (1974) J . Biol. Chem. Z4Y, 704--710. 9. Nachman, R. (1975) CIBA Found. Symp. 35 (new series), 2331. 10. Koshland, D. E. (1960) Adv. Enzymol. 22, 45-97. 11. Bennett, J. S., Colman, R. W., Figures, W. & Colman, R. F. (1975) Pvoc. 10th Int. Congr. Biochem. p. 279. 12. Pearce, P. H. & Scrutton, M. C. (1977) Bioclzrm. Tram. 5, 138 - 139. 13. Aster, R. H. & Jandl. J. H. (1964) J . Clin. Invest. 43, 843-855. 14. David, J. L. & Herion, F. (1972) Adv. Exp. Med. B i d . 34, 335- 339. 15. Easterbrook-Smith, S. B., Wallace, J. C. & Keech, D. B. (1976) Eur. J . Biochem. 62, 125- 130. 16. Newcombe, A. C . & Reid, S. G. (1953) Nuturr (Lond.) 172, 455 - 456. 17. Hansske, F., Sprinzl, M. & Cramer, F. (1974) Biorx. Chenz. 3, 367- 376. IS. Adam. H. (1963) in Methods ?/Enzymatic Aiiulj,.sis (Bergmeyer, H. U . , ed.) pp. 573-577, Academic Prcss, London and New York. 19. Strehler, B. H. (3963) in Methods in Enzyrnutic, Anulysis (Bergmeyer, H. U., ed.) pp. 559-568, Academic Press, London and New York. 20. Green, A. A. & Cori, G . T. (1943) J . B i d . Chem. 151, 21 36. 21. Born, G. V. R. (1970) J . Physiol. (Lond.) 209, 437-451. 22. Holme, S. & Holmsen, H. (1975) Scand. J . Huemutol. 15, 96103. 23. Salzmann, E. W., Chambers, D . A. & Neri, L. L. (1966) Thromb. Diutli. Huemorrh. 15, 52-68. 24. Rozenberg, M. C. & Holmsen, H. (1968) Biochim. Biophyx Aria, 155, 342 - 352. 25. Mills, D. C. B.&Smith, J. B.(1971) Bioc.1zem.J. 121, 185-196. 26. Mills, D. C. B. & Roberts, G. C. K. (1967) Nature iLond.) 213, 35-38. 27. Mills, D. C . B. & Roberts, G. f.K. (1967) J . Piiysiol. 193, 443 - 453. 28. Niewiarowski, S. & Thomas, D. P. (1966) Nuturc. (Lond.) 2 / 2 , 1544-1547. 29. Packham, M. A,, Guccione, M . A , , Cliarig, P. L. & Mustard, J. F. (1973) A m , J . Phj,.siol. 225, 38-47. 30. Han, P. & Ardlie, N . G . (1977) Brit. J . Huenurtol. 26, 357372. 31. Gaarder, A., Jonsen, J., Laland, S.. Hellem, A. & Owren, P. A. (1961) Naturr LO^.) 192, 531 -533. 32. Mills, D. C. B. & MacFarlane, D. E. (1976) in Platelets in Biology and Paihologj. (Gordon, J. L., ed.) pp. 159-202, North Holland, Amsterdam, New York, Oxford. 33. Maguire, M. H. & Michal, F. (1968) Nnturr (Lond.) 217, 572 -573. -

554

P.H. Pearce, J. M. Wright, C. M. Egan, and M. C. Scrutton: ADP Analogues and Human Blood Platelets

34. Gough, G., Maguire, M. H. & Penglis, F. (1972) Mol. PhurmaC O ~ .8, 170 - 177. 35. Kikugawa, K., Suehiro, H. & Ichino, M. (1973) J . Med. Chrm. 16,1389-1391. 36. Stone, J . V., Singh, R. K., Horak, H. & Barton, P. G. (1976) Can. J . Biochenz. 54, 529- 533.

37. Horak, 13. & Barton, P.G . (1974) Biochinz. Bi0phj.s. Acta, 373, 471 - 480. 38. Maclntyre, D. E., Gordon, J . L., Drummond, A. H., Steer, M. & Salzmann, € .W. (1977) Thumb. Huemostasis,38, 6. 39. Pearce, P. H. & Scrutton, M. C. (1977) T h r o d ~ Haemosrasis, . 38. 6.

P. H. Pearce, J . M. Wright, C. M . Egan, and M. C. Scrutton, Department of Biochemistry, King's College, University of London, Strand, London, Great Britain, WC2R 2LS

Interaction of human blood platelets with the 2',3'-dialdehyde and 2',3'-dialcohol derivatives of adenosine 5'-diphosphate and adenosine 5'-triphosphate.

Eur J . Biochem. 88, 543-554 (1978) Interaction of Human Blood Platelets with the 2’,3’-Dialdehyde and 2’,3’-Dialcohol Derivatives of Adenosine 5’-Di...
987KB Sizes 0 Downloads 0 Views