Platelets by [3H]Dihydroergocryptine Binding

Identification of a-Adrenergic Receptors in Human Platelets by [3H]Dihydroergocryptine Binding

KURT D. NEWMAN, LEWIS T. WILLIAMS, N. HAHR BISHOPRIC, and ROBERT J. LEFKOWITZ, Departments of Medicine and Biochemistry, Duke University Medical Center, Durham, North Carolina 27710

A B ST RA CT Binding of [3H]dihydroergocryptine to INTRODUCTION platelet lysates appears to have all the characteristics of binding to a-adrenergic receptors. At 25°C binding The endogenous catecholamines epinephrine and nor- reaches equilibrium within 20 min and is reversible epinephrine exert a variety of regulatory functions on upon addition of excess phentolamine. Binding is satu- human physiological processes. The initial sites of ac- rable with 183+22 fmol of [3H]dihydroergocryptine tion of catecholamines seem to be at membrane adren- bound per mg of protein at saturation, corresponding ergic receptors. Based on the relative order of potencies to 220+26 sites per platelet. Kinetic and equilibrium of a series of adrenergic agonists, Ahlquist proposed studies indicate the dissociation constant of [3H]dihy- that the effects of catecholamines were mediated by droergocryptine for the receptors is 1-3 nM. The spec- two distinct types of adrenergic receptors, termed a- ificity of the binding sites is typical of an a-adrenergic and ,3-adrenergic receptors (1). /-Adrenergic responses receptor. Catecholamine agonists compete for occu- (isoproterenol> epinephrine> norepinephrine) which pancy of the [3H]dihydroergocryptine binding sites are specifically antagonized by propranolol are typified with an order of potency (-)epinephrine> (-)norepi- by relaxation of smooth muscle and stimulation of car- nephrine> (-)isoproterenol. Stereospecificity was dem- diac contractility. a-Adrenergic responses (epineph- onstrated inasmuch as the (+)isomers of epinephrine rine> norepinephrine> isoproterenol) are potently an- and norepinephrine were 10-20-fold less potent than tagonized by phentolamine and are typified by con- the (-)isomers. The potent a-adrenergic antagonists traction of vascular and uterine smooth muscle. phentolamine, phenoxybenzamine, and yohimbine Recent studies have shown that epinephrine and competed potently for the sites, whereas 8-antagonists norepinephrine induce human platelet aggregation such as propranolol and dichlorisoproterenol were whereas isoproterenol does not (2-4). This aggregation quite weak. Dopamine and serotonin competed only at is blocked by the a-antagonist phentolamine (2-4). high concentrations (0.1 mM). Epinephrine and norepinephrine also cause a rapid The [3H]dihydroergocryptine binding sites could decrease in prostaglandin E,-(PGE,)1- enhanced cyclic also be demonstrated in intact platelets where they AMP (cAMP) levels in platelets and produce immedi- displayed comparable specificity, stereospecificity, ate decreases in platelet adenylate cyclase activity (5- and saturability. Saturation binding studies with the 11). These responses are also blocked by the a-adrener- intact platelets indicated 220+±45 receptors per plate- gic blocker phentolamine. Thus, there is evidence that let, in good agreement with the value derived from these adrenergic effects on platelets are mediated by studies with platelet lysates. Ability of a-adrenergic a-adrenergic receptors. agonists to inhibit adenylate cyclase and of a-adrener- The development of radioligand binding methods gic antagonists to antagonize this inhibitory effect di- for the direct study of hormone and drug receptors over rectly paralleled ability to interact with the [3H]dihy- the past few years has facilitated the investigation of droergocryptine binding sites. These data demonstrate the properties and regulation of these receptors. Ap- the feasibility of directly studying a-adrenergic recep- plied to human cell types, these methods provide an tor binding sites in human platelets with [3H]dihydro- approach to the direct study of abnormalities of recep- ergocryptine. tor function in various disease states (12, 13). Recently, Williams et al. described a ligand, [3H] Dr. Lefkowitz is an Investigator of the Howard Hughes dihydroergocryptine, which could be used to identify Medical Institute. Received for publication 4 April 1977 and in revised 1Abbreviations used in this paper: cAMP, cyclic AMP; form 29 September 1977. PGE, prostaglandin E,.

The Journal of Clinical Investigation Volume 61 February 1978 395-402 395 a-adrenergic receptors in uterine smooth muscle (14). resuspended in "incubation buffer" (Tris-HCl 0.05 M, pH 7.5, In the present studies we report the successful identifi- EDTA 0.005 M) for use in the binding assay. Whole platelet and lysate preparations were examined by cation and characterization of a-adrenergic receptors phase-contrast and interference-contrast microscopy at x 1000 in human platelets with [3H ]dihydroergocryptine. magnification as well as by electron microscopy. The platelet Binding of [3H]dihydroergocryptine to human platelet preparations appeared intact. The lysate preparations show lysates has all the characteristics of specificity, stereo- lysis of some platelets with areas of cell debris including selectivity, saturability, and reversibility which would scattered cytoplasmic granules and granule-free membrane fragments. Some apparently unlysed platelets were also seen, be expected of binding to a physiological a-adrenergic but these were markedly swollen. receptor. a-Receptor binding is also demonstrated in [3H]Dihydroergocryptine binding assay. [3H]Dihydroer- intact platelets. The specificity of the binding data is gocryptine (5 nM unless otherwise specified) and platelet ly- also correlated with an a-adrenergic response in the sates (2.0 mg/ml unless otherwise specified) or intact platelets (1-2 x 109/ml) were incubated at 25°C with shaking for 18 min platelet lysates inhibition of PGE,-stimulated adenyl- in a total volume of 150 ,ul of incubation buffer for platelet ate cyclase. lysates, or buffer I for intact platelets. In competition experiments, varying concentrations of agonists or antagonists were added to the incubation as indicated. Incubations were ter- METHODS minated by rapidly diluting the total volume with 2 ml of incubation buffer or buffer I at 250C followed by rapid filtration Pharmacological agents. [3H] Dihydroergocryptine was through Whatman GFC glass fiber filters (Whatman, Inc., used as the a-adrenergic radioligand for this study. It has a Clifton, N. J.). Filters were rapidly washed with 20 ml of incu- specific activity of 23.0 Ci/mmol and has biologic activity and bation buffer. After drying, filters were counted in a Triton- chromatographic properties identical to native dihydroer- toluene scintillation mixture at an efficiency of 40-50%. gocryptine (14). Other compounds used in this study were "Nonspecific" binding was defined as binding which was from the following sources: Sigma Chemical Co., St. Louis, not displaced by a high concentration of 10 ,uM phentolamine, Mo.-(-)epinephrine bitartrate, (-)norepinephrine bitar- a potent a-adrenergic antagonist, or 1 mM epinephrine, an trate, (- )isoproterenol bitartrate, dopamine hydrochloride, a-adrenergic agonist. In platelet lysates, 1 mM epinephrine dihydroxyphenylalanine, dihydroxymandelic acid, (+)norme- and 10 ,uM phentolamine inhibited binding to the same extent, tanephrine, .dihydroergotamine, ergotamine tartrate, ergocryp- and higher concentration of phentolamine did not inhibit tine, (-)phenylephrine hydrochloride, and (+)propranolol; binding further. In intact platelets, 10 ,M phentolamine in- Winthrop Laboratories, New York-phenylephrine hydro- hibited binding slightly, but consistently to a greater extent chloride (used only in certain experiments); Sandoz Pharma- (- 20%), than 1 mM epinephrine. Accordingly in studies with ceuticals, E. Hanover, N. J.-dihydroergocryptine methane intact platelets we have used 1 mM epinephrine to determine sulfonate; Sterling Drug, Inc., New York-(+)epinephrine nonspecific binding, assuming that all true a-adrenergic re- bitartrate and (+)norepinephrine bitartrate; Ayerst Labora- ceptors should be accessible to epinephrine as well as phento- tories, New York-practolol hydrochloride and (+)butacla- lamine. "Specific" binding, defined as total binding minus mol; Eli Lilly and Co., Indianapolis, Ind. -(+)dichlorisopro- nonspecific binding is shown in all figures and tables. Specific terenol; CIBA Pharmaceutical Co., Div. of CIBA-Geigy Corp., binding was generally 55-70% of the [3H]dihydroergocryp- Summit, N. J. -phentolamine hydrochloride; McNeil Labora- tine bound in the platelet lysates and 30-40% in the intact tories, Inc., Fort Washington, Pa.-dibozane and haloperidol; platelets. In competition experiments with [3H]dihydroergo- ICN Nutritional Biochemicals Div., Cleveland, Ohio-yo- cryptine present at -5 nM, specific binding was generally himbine hydrochloride; Boehringer Ingelheim Ltd., Elms- -110 fmol/mg protein in the platelet lysates. ford, N. Y.-clonidine hydrochloride; Burroughs Wellcome To assess the integrity of the [3H]dihydroergocryptine the Co., Research Triangle Park, N. C.-methoxamine hydro- following experiment was performed. Binding incubations chloride; and Smith Kline & French Laboratories, Div. of were performed in the usual manner. The membranes were Smith Kline Corp., Philadelphia, Pa.-phenoxy-benzamine then collected by centrifugation. Radioactivity from the super- hydrochloride. nate and that extracted from the pellet with ethanol was then Preparation of platelets and platelet lysates. 100-200 cm3 chromatographed on thin layer plates as previously described of blood were takei} from peripheral veins of healthy human in the solvent system chloroform:ethanol:acetic acid:9:5: 1 volunteers. The first 5 cm3 of blood was discarded, and the (14). Both the bound and free [3H]dihydroergocryptine dis- remainder was collected with 3.2% sodium citrate as antico- played R,s identical to native authentic material. agulant. The blood was centrifuged at 380 g for 10 min. The Protein was determined by the method of Lowry et. al. (15). resulting platelet-rich plasma was centrifuged at 16,000 g for Adenylate cyclase assays. Adenylate cyclase assays of 10 min at 25°C. The resulting platelet pellet was resuspended platelet lysates were performed by methods comparable to in "buffer I" (Tris-HCl, 0.05 M, pH 7.35, NaCl 0.15 M, EDTA those previously described from this laboratory (16). Assay 0.02M) and then recentrifuged at 16,000 g for 10 min, re- volume was 50 ,ul containing 30 mM Tris-HCl (pH 7.4), 10 mM suspended in buffer I, and recentrifuged at 16,000 g for 10 min at MgCl2, 0.1 mM cAMP, 1.5 mM ATP [a-32P]ATP (1-2 x 106 25°C. The washed platelets were resuspended in buffer I at a con- cpm), 5 mM phosphoenolpyruvate, pyruvate kinase (40 ,tg/ centration of 1-2 x 1Q9 platelets/ml for use in binding assays ml), and myokinase (20 gg/ml). Incubations were for 10 min at or were resuspended in "lysing buffer" (Tris HCI 0.005M, pH 37'C and were stopped by addition of 1 ml of a solution con- 7.5, EDTA 0.005M) at 4°C for 1 min for preparation of platelet taining [3H]cAMP (15,000 cpm/ml), 100 ,ug ATP, and 50 ,ug lysates. No significant contaminating leukocytes were found cAMP. [32P]cAMP that was formed was isolated by the in stained preparations. This suspension was homogenized for method of Saloman et al. (17). In separate experiments, it was 20 strokes with a motor-driven teflon-tipped pestle (controlled demonstrated that the assay was linear with respect to protein by a Powerstat rheostat (Fisher Scientific Co., Pittsburgh, Pa.) and time over the time and concentration ranges used in these at a setting of 120 (-) pestle clearance 0.1-0.5 mm) and then studies, and that the stopping solution in fact stopped the centrifuged at 39,000 g for 10 min at 4°C. The final pellet was enzymatic reaction.

396 K. D. Newman, L. T. Williams, N. H. Bishopric, and R. J. Lefkowitz For the routine assay of a-adrenergic effects on platelet E adenylate cyclase, the enzyme was stimulated by 10 ,uM PGE1. The ability of a-adrenergic agonists such as epinephrine to E 100 inhibit the PGE -stimulated enzyme was then measured. Inas- much as has been noted previously (10), platelet lysate adenylate cyclase is stimulated via a ,f-adrenergic receptor, all assays ,0 80 also contained 10 ,M (+)propranolol to block 83-adrenergic receptors. In general, PGE,-stimulated adenylate cyclase ac- cL tivity in the platelet lysates was 869+39 pmol cAMP gener- ated/min per mg protein and was maximally reduced to 462 40- +24 pmol/min per mg by 1 mM epinephrine (n = 10 experiments). 'v 20- Platelet aggregation. Platelet-rich plasma was prepared as described above. Platelet aggregation was initiated by addi- a 6- tion of (-)epinephrine to a final concentration of 30 ,uM in a rT& 4 8 12 16 20 24 28 32 36 40 volume of 0.5 ml of stirred platelet rich plasma at 37°C. Aggre- Time (minutes) gation was followed using a platelet aggregation profiler (PAP- 3, Bio/Data Corp., Willow Grove, Pa.) FIGURE 2 Reversibility of [3H]dihydroergocryptine binding to platelet Ilvsates. Lysates were incubated with [3H]dihydroergocryptine (5.5 nM) at 25°C for 16 min after which a RESULTS large excess of phentolamine (10 ,M) was added. The time of phentolamine addition is defined in this figure as time = 0. Kinetic characteristics of [3H ]dihydroergocryptine At the indicated times, specific [3H ] dihydroergocryptine binding to platelet lysates. The time-course of spe- binding was determined. Maximum binding is defined as the cific binding of [3H]dihydroergocryptine to human amount of binding just before the addition of phentolamine platelet lysates at 25°C is shown in Fig. 1. Binding at time = 0. Each value is the mean of duplicate determina- was reversible upon addition of excess phentolamine tions. (Fig. 2). Assuming the binding reaction to be a simple bimolecular process, the forward rate constant (k,) is x by equilibrium studies (see below). It should be noted, calculated to be 1.8 107M-'min-'. The first order re- however, that the reverse kinetic experiment shows an verse rate constant (dissociation rate constant) is 0.018 of -20% of min-'. initial rapid loss the binding followed by a The ratio k2/k, (1 nM) provides a kinetically slower decay. The significance of this initial rapidly derived estimate of the dissociation constant of [3H] is not known. dihydroergocryptine for the binding sites. This is in reversing component of dissociation reasonable agreement with the Kd (3.1 nM) determined Equilibrium characteristics of [3H ] dihydroergocryptine binding to platelet lysates. Binding was a saturable process with 183+22 fmol (mean+SEM) of [3H]dihydroergocryptine bound per mg of protein at 90- saturation (Fig. 3). Half-maximal saturation occurred at ' 80_* 3.1 nM [3H ] dihydroergocryptine providing an estimate so / of the equilibrium dissociation constant, Kd of [3H]di- 70- hydroergocryptine for the binding sites. Inasmuch as we found that 5 x 108 platelets provide 1 mg of lysate 860_ protein, the saturation value of 183 fmol/mg cor- _ / responds to 220±+-26 receptors per platelet. Specificity of the [3H ] dihydroergocryptine binding *'40 - / sites in platelet lysates. Adrenergic agonists competed for the [3H]dihydroergocryptine binding sites with an order of potency, (-)epinephrine > (-)norepinephrine > (-)isoproterenol (Fig. 4). The a-adrenergic agonists had high affinity for the binding sites, with Kd values of 0.26 uM and 0.85 ,tM for (-)epinephrine and (-)norepinephrine, respectively (Table I). The binding sites exhibited stereospecificity with Kd values 0 4 8 12 16 20 24 28 32 for the (+)isomers of epinephrine and norepinephrine Time (minutes) 8.5- and 20-fold higher than the values for the corre- FIGURE 1 [3H]Dihydroergocryptine binding to human plate- sponding (-)isomers. let lysates as a function of time. [3H]Dihvdroergocrvptine (3.9 nM) was incubated with platelet lysates (-2 mg/ml) for a-Adrenergic antagonists potently competed for the the indicated times, and specific binding was determined as [3H]dihydroergocrvptine binding sites. Phentolamine, described in Methods. Each value is the mean of duplicate a potent a-antagonist, competed with a Kd of 0.014 ,M determinations. (Table I). Other a-antagonists such as phenoxybenza-

a-Adrenergic Receptors in Human Platelets 397 220 200 0 180 . 160 0 140 #I 120 100 80 I 60 40 20 1L I I I I 0 10 20 30 40 50 60 [[3H] Dihydroxrgocryptine] (nM) FIGURE 3 Specific binding of [3H]dihydroergocryptine to human platelet lysates as a function of [3H]dihydroergocryptine concentration. Human platelet lysates were incubated with indicated concentrations of [3H]dihydroergocryptine, and binding was determined as described in Methods. Each value is the mean of duplicate determinations from four separate experiments. mine, yohimbine, and dibozane also inhibited the [3H]- Dopamine had a Kd of 3.1 ,M consistent with its dihydroergocryptine binding (Table I). known a-adrenergic activity. Serotonin had a low affin- In contrast, 8-adrenergic antagonists such as pro- ity for the binding sites with a Kd of 24.0 ,uM. When pranolol, dichlorisoproterenol, and practolol competed added to maximally effective concentrations of phentol- only at high concentrations for the binding sites. Com- amine (10 uM), high concentrations of dopamine and pounds that are devoid of a-adrenergic physiologic ef- serotonin did not displace additional [3H]dihydroergo- fects, but which are structurally related to catechola- cryptine. The a-agonist clonidine was about 10 times mines, such as DOPA, dihydroxymandelic acid, nor- more potent than (-)epinephrine in inhibiting [3H]- metanephrine, and catechol, did not compete for the dihydroergocryptine binding (Table I). [3H]dihydroergocryptine binding sites. Dihydroergo- The dopaminergic antagonists haloperidol and (+)- cryptine itself competed potently for the binding sites butaclamol inhibited binding but at considerably and was five times more potent than the parent com- higher concentrations than the classical a-blockers. pound ergocryptine (Table I). Several other ergot alka- This result is consistent with the known weak a-adren- loids such as ergotamine and dihydroergotamine were ergic antagonist activity of these compounds (18). also potent in competing for binding (Table I). Kd values for a number of agents as assessed by [31H]dihydroergocryptine binding are summarized in -0- (-) Epsn.phrine 100 - (+) Epiwphrin Table I. o' -0- (-)N.r.rn.pin C -U- (+)Nropinepirine [3H ]Dihydroergocryptine binding to intact plate- X(-)I soproter nol/ cB8 lets. [3 H]Dihydroergocryptine also bound to intact 80 platelets. The kinetics of binding were very compara- C o CL ble to those observed with lysates. At 23°C, binding : 60- reached steady state by 20 min. Binding was also fully reversible with a t112 of 30' (cf. Fig. 2). Fig. 5 demon- 40- strates a saturation plot for specific binding of [3H]dihydroergocryptine to intact platelets. The apparent Kd (-4 nM) is in excellent agreement with that determined 20- in the platelet lysates (-3 nM) (Fig. 3). At saturation, 3.66 fmol [3H]dihydroergocryptine were bound/107 -7 -6 -5 -4 -3 platelets. This corresponds to 220+45 receptor sites/ Log1O [Adrenergic Agonist] (M) platelet (n = 13). Thus, the numbers of binding sites determined from experiments with platelet lysates and FIGURE 4 Inhibition of [3H]dihydroergocryptine binding by intact platelets are in excellent agreement (220+26 vs. adrenergic agonists. Platelet lysates were incubated with 220±45). [3H]dihydroergocryptine (5.5 nM) in the absence and presence of the indicated agonists, and specific binding was The specificity of [3H]dihydroergocryptine binding determined. Each value is the mean of duplicate deter- to the intact platelets was also in good agreement with minations from at least two separate experiments. that determined in the Iysates. Fig. 6 shows displace-

398 K. D. Newman, L. T. Williams, N. H. Bishopric, and R. J. Lefkowitz TABLE I 5.0r Inhibition of [3H]Dihydroergocryptine Binding to Platelet Lysates by Adrenergic and Other Drugs c *(5) o 4.0)F

Drug Inhibition of [3H]DHE binding-Kd - c ._ .6 (1)0 M oB 03.0 (5) Adrenergic 9-t (6) (2) Agonists 0 C Clonidine 0.017 V - ;(6)

(-)Epinephrine 0.26 >o 0 (+)Epinephrine 2.3 '-- E 1.0 (-)Norepinephrine 0.85 . - (4) (+)Norepinephrine 17.0 , (-)Phenylephrine 0.86 (3) 1I (-)Isoproterenol 142.0 0 10 20 30 40 50 60 70 [[3H] Dihydroergocryptine] (nM) Antagonists Yohimbine 0.002 FIGURE 5 Specific binding of [3H]dihydroergocryptine to Dibozane 0.004 intact human platelets as a function of [3H ] dihydroergo- Dihydroergocryptine 0.011 cryptine concentration. Intact platelets were incubated with Phentolamine 0.014 the indicated concentrations of [3H ] dihydroergocryptine, Ergotamine 0.014 and binding was determined as described under Methods. Dihydroergotamine 0.023 Each value shown is the mean of duplicate determinations from the number of experiments shown in parenthesis. Ergocryptine 0.057 Comparable data were obtained when platelets were in- (+)Butaclamol 0.1 cubated at 37°C. Haloperidol 2.6 Propranolol 12.0 Dichlorisoproterenol 15.7 in the platelet lysates as a convenient biochemical as- Practolol 140.0 say of a-adrenergic effects which could be used for Others correlation with the [3H]dihydroergocryptine binding Dopamine 3.1 data. Serotonin 24.0 a-Adrenergic catecholamine agonists inhibited en- Dopa N.I. zyme activity with a relative potency order identical Catechol N.I. to that observed for inhibition of [3H]dihydroergocryp- Normetanephrine N.I. tine binding and typical of an a-adrenergic receptor Dihydroxymandelic acid N.I. mediated process, i.e., (-)epinephrine > (-)norepinephrine > (-)isoproterenol (Fig. 7). Inhibition of Dissociation constants were calculated from the concentrations of the agents which caused 50% inhibition of Intact Platelets [3H]dihydroergocryptine binding using the relationship 100 Kd = EC50/1 + (SIKm) (23), where [s] = the concentration of [3H]dihydroergocryptine used in the binding assays (5.5 nM), and Km = the dissociation constant of [3H]dihydroergocryp- cE 80 tine (2 nM). Each drug was tested two to eight times, and C, 80Phentolamine the mean value was recorded. NI indicates no inhibition at a concentration of 1 mM. >*60 (-)Epi Co 40 ment curves for phentolamine, (-) and (+)epinephrine. The displacement curves for phentolamine and (-)epinephrine in intact platelets are virtually superimposa- 20 (+Ep ble on those obtained in platelet lysates although that for (+)epinephrine is displaced somewhat to the right. in -8 -7 -6 -5 -4 -3 Specificity of adenylate cyclase inhibition platelet [d-Adrenergic Agent] (M) lysate. a-Adrenergic catecholamines such as epi- Log1o nephrine inhibit PGEl-stimulated adenylate cyclase FIGURE 6 Inhibition of [3H]dihydroergocryptine binding to activity in (10, 11). The pattern of inhibition intact platelets by adrenergic agents. Intact platelets were platelets incubated with [3H]dihydroergocryptine (-5 nM) in the by catecholamines appears to be that of an a-adrenergic absence and presence of the indicated agents, and specific response. Accordingly, we utilized epinephrine-stimu- binding was determined. Each value is the mean of dupli- lated inhibition of PGE1-stimulated adenylate cyclase cate determinations from five to six experiments.

a-Adrenergic Receptors in Human Platelets 399 mine and dihydroergocryptine itself competitively antagonized the epinephrine-induced inhibition of en- E w -6- I zyme activity (Fig. 8). The comiipetitive nature of this antagonism is indicated by the parallel shifts in the E 80 epinephrine dose-response curve caused by these agents with the same maximutm response achieved. -0< From data such as that shown in Fig. 8, apparenit dis- 0. sociation constants could be calculated for phentola- 0 4 mine and dihydroergocryptine as antagonists of the a- I~~ ~- -6 -S -4 - receptor-mediated enzyme inhibition (see figure leg- end). These values were, for phentolaminie 0.2 ,uM 20 (n = 6), and for dihydroergocryptine 0.02 ,uM (n = 8). These values were somewhat higher than those calcu- C lated from binding studies. Dihydroergocryptine had -7 -6 -5 -4 -3 no agonist effect of its own at concentrations up to Logo [Agonist] (M) 10 ,uM. Yohimbine was also a potent antagonist (Kd FIGURE 7 Inhibition of PGE,-stimulated adenylate cyclase = 0.06 ,M). In contrast, the classical f8-adrenergic an- activity in human platelet lysates by adrenergic agonists. tagonist propranolol was very weak in antagonizing Maximum response refers to the decrease in cAMP genera- the effect with a calculated tion produced by 1 mM (-)epinephrine. This decrease epinephrine Kd from was generally a 40-60% reduction in PGEl-stimulated adenylate cyclase studies of 89 ,M. enzyme activity. Values shown are the means of duplicate Phenylephrine appeared to be a "partial agonist". determinations from at least two separate experiments. The inhibition of enzyme activity by phenylephrine ranged from 20-60% (n = 6) of that caused by epineph- PGE,-stimulated adenylate cyclase activity by these rine at the highest concentration studied (1 mM). catecholamines also displayed stereospecificity com- Phenylephrine also antagonized the reduction in en- parable to that observed in the binding studies (Fig. 7). zyme activity produced by epinephrine as would be Potent a-adrenergic antagonists such as phentola- expected of a partial agonist. At 1 mM dopamine and serotonin caused slight inhibition of PGE -stimulated adenylate cyclase amount- -- no blocker ing to 40 and 16% of the maximal effect of epinephrine, E --- 0.1MM Dihydroergocryptine E 100 -- MM Phentolomine respectively. Effects of dihydroergocryptitne on platelet aggrega- E tion. If the [3H]dihydroergocryptine binding sites E 80o identified in the intact platelets and platelet lysates are in fact equivalent to the a-adrenergic receptors mediat- w O <60- ing the aggregating effects of epinephrine, then dihydroergocryptine binding should lead to antagonism of 0 epinephrine-stimulated platelet aggregation. We found o >, 40- that dihydroergocryptine, 0.1 ,uM, inhibits the ability of epinephrine (30 u.M) to aggregate platelets (data not 20 shown).

DISCUSSION -_ -D - Log1o[(-)Epinephrine] (M) [3H]Dihydroergocryptine binding sites identified in human platelet lysates exhibit the specificity, stereo- FIGURE 8 Antagonism of epinephrine induced inhibition specificity, saturability, and kinetics to be expected of of PGEl-stimulated adenylate cyclase by phentolamine to and dihydroergocryptine. Enzyme assays were performed in binding physiological a-adrenergic receptors. The the presence of the indicated concentrations of epinephrine binding sites were also demonstrated in intact plate- with or without the concurrent presence of phentolamine lets. The binding characteristics agree closely with or dihydroergocryptine. Maximum response refers to the de- those of [3H]dihydroergocryptine binding sites previ- crease in cAMP generation produced by 1 mM (-)epineph- ously studied in smooth muscle membranes derived rine. These data were used to calculate Kds of the antagonists from using the relationship Kd = [antagonist]/(CR-1), where CR rabbit uterus (14). (concentration ratio) refers to the ratio of equiactive con- The ability of a-adrenergic agonists to inhibit PGEj- centrations of (-)epinephrine in the presence and ab- sensitive adenylate cyclase in the platelet lysates di- sence of the stated concentration of antagonist (24). rectly paralleled their ability to inhibit [3H]dihydro-

400 K. D. Newman, L. T. Williams, N. H. Bishopric, and R. J. Lefkowitz ergocryptine binding. Similarly, ability of adrenergic ylephrine is a classical full agonist at peripheral a- antagonists to antagonize enzyme inhibition and to in- adrenergic receptors. However, it appeared to have hibit [3H]dihydroergocryptine binding were parallel. only partial agonist activity in our study of adenylate ,B-Antagonists were very weak both as inhibitors of cyclase and in previous reports was a partial agonist [3H]dihydroergocrvptine binding and as antagonists of for reduction of cAMP levels in intact platelets (6). the epinephrine-induced inhibition of the adenylate Phenylephrine also appears to be devoid of agonist cyclase. For example, propranolol was 450 times activity in aggregating platelets (21). weaker than phentolamine as an antagonist in the en- The ability to study other hormone and drug recep- zyme assays and 850 times weaker in the binding as- tors in circulating cells has provided a useful approach says. Although the relative potencies of the various to the investigation of receptor abnormalities in human adrenergic agents determined by [3H]dihydroergo- disease states (12, 22). The ability to quantitate and cryptine binding and adenylate cyclase studies were characterize [3H]dihydroergocryptine binding sites in comparable, somewhat higher concentrations of agents human platelets now extends this approach to the direct were necessary to produce effects on the adenylate study of a-adrenergic receptors in normal and patho- cyclase than on [3H]dihydroergocryptine binding. This physiologic states in man. is presumably due to a number of factors. (a) Differ- ences in experimental conditions for the two types of ACKN OWLE DGME NT assays: thus, for example, to optimize enzyme activity The authors are grateful to Ms. Debra Mullikin for excel- higher temperatures (37°C) were utilized, and PGE, lent technical assistance, to Dr. Judith Anderson for assistance was included in the enzyme assays; and (b) the pres- with the platelet aggregation studies, and to Nis. Donna Addi- ence of poorly understood "coupling factors" which son for excellent secretarial assistance. The electron micro- scopic studies were performed in consultation with Dr. J. R. affect the ultimate translation of receptor occupancy Sommer, Director, Veterans Administration Electron Micros- into biological response in the platelet lysates. copy Laboratory, Veterans Administration Hospital, Durham, The activity of the PGE,-stimulated adenylate cy- N.C. 27705. clase which we observed in the platelet lysates was This work was supported by grants HL 16037 and HL 20339 several-fold higher than that from the National Institutes of Health and by a grant-in-aid previously reported by from the American Heart Association with funds contributed several other laboratories (10, 19, 20). This may be due in part by the North Carolina Heart Association. to differences in membrane preparation (freeze-thaw or sonication vs. mechanical disruption) or assay conditions (higher ATP and PGE, concentrations in our ex- REFERENCES periments). 1. Ahlquist, R. P. 1948. A study of the adrenotropic receptors. The characteristics of [3H]dihydroergocryptine bind- Am. J. Physiol. 53: 586-600. ing to intact platelets appeared to be quite comparable 2. O'Brien, J. R. 1963. Some effects of' adrenaline and anti- to those observed in the platelet lysates. The number adrenaline compounds on platelets in vitro and in vivo. of receptors per platelet determined from the studies Nature (Lond.). 200: 763-764. 3. Mills, D. C. B., and G. C. K. Roberts. 1967. Effects of with lysates and intact platelets was also identical (220 adrenaline on human blood platelets. J. Physiol. (Lond.). per platelet). The ability of dihydroergoeryptine to 193: 443-453. inhibit epinephrine-induced platelet aggregation lends 4. Bygdeman, S., and 0. Johnson. 1969. Studies oIn the effect further credence to the contention that the sites labeled of adrenergic blocking drugs on catecholamine-induced are in fact equivalent to the platelet aggregation and uptake by noradrenaline and 5- physiological a-adrenergic hydroxytryptamine. Acta Physiol. Scand. 75: 129-138. receptors. 5. Salzman, E. W., and L. L. Neri. 1969. Cyclic 3',5'adeno- The specificity of the [3H]dihydroergocryptine bind- sine monophosphate in human blood platelets. Nature ing sites which we have identified appear to be identi- (Lond.). 224: 609-610. cal to the previously demonstrated specificity of. (a) 6. Marquis, N. R., J. A. Becker, and R. L. Vigdahl. 1970. Platelet aggregation, an epinephrine-induced decrease in epinephrine-induced aggregation of platelets, which cyclic-AMP synthesis. Biochem. Biophys. Res. Commun. is blocked by phentolamine but not by ,8-antagonists 39: 783-789. (2-4); (b) epinephrine-induced reduction of PGE,- 7. Cole, B., G. A. Robison, and R. C. Hartman. 1971. stimulated cAMP levels in intact platelets (5-9); and Studies on the role of cAMP in platelet function. Ann. N. Y. Acad. Sci. 185: 477-487. (c) epinephrine-induced inhibition of PGE,-stimu- 8. Haslam, R. J., and A. Taylor. 1971. Effects of catechola- lated adenylate cyclase in platelet lysates (10). mines on formation of adenosine 3':5' cyclic monophos- All of these phenomena have typical characteristics phate in human blood platelets. Biochemn. J. 125: 377- of an a-adrenergic receptor-mediated process. None- 379. theless, it should be noted that both our data and previ- 9. Harwood, J. P., J. Moskowitz, and G. Krishna. 1972. Dy- namic interaction of prostaglandin and norepinephrine ous reports (10, 21) suggest that the platelet a-adrener- in the formation of adenosine 3-5' monophosphate in gic receptors may differ in certain respects from other human and rabbit platelets. Biochim. Biophys. Acta. 261: peripheral a-adrenergic receptors. For example, phen- 444-456.

a-Adrenergic Receptors in Human Platelets 401 10. Jakobs, K. H., W. Saur, and G. Schultz. 1976. Reduc- sensitive adenylate cyclase assay. Anal. Biochem. 58: tion of adenylate cyclase activity in lysates of human 541-548. platelets by the a-adrenergic component of epineph- 18. U'Prichard, D. C., D. A. Greenberg, and S. H. Snyder. rine. J. Cyclic Nucleotide Res. 2: 381-392. 1977. Binding characteristics of a radiolabeled agonist 11. Zieve, P. D., and W. B. Greenough. 1969. Adenyl and antagonist at central nervous system alpha-nor- cyclase in human platelets: activity and responsiveness. adrenergic receptors. Mol. Pharmacol. 13: 454-473. Biochem. Biophys. Res. Commun. 35: 462-466. 19. Haslam, R. J. 1973. Interactions of the pharmacological 12. Kahn, C. R., J. S. Flier, R. S. Bar, J. A. Fisher, P. Gorden, receptors of blood platelets with adenylate cyclase. M. M. Martin, and J. Roth. 1976. The syndromes of Ser. Haematol. 6: 333-350. insulin resistance and acanthosis nigricans, insulin re- 20. Rodman, G. A., and M. B. Feinstein. 1976. Relationships ceptor disorders in man. N. Engl. J. Med. 294: 739-745. between Ca2+ and adenylate and guanylate cyclase in 13. Goldstein, J. L., and M. S. Brown. 1974. Binding and the control of platelet secretion and aggregation. Proc. degradation of low density lipoproteins by cultured Natl. Acad. Sci. U. S. A. 73: 1829-1833. human fibroblasts: comparison of cells from a normal 21. Rossi, E. C., and E. A. Zeller. 1976. On the molecular subject and from a subject with homozygous familial basis for the action of epinephrine upon platelets. Clin. hypercholesterolemia. J. Biol. Chem. 249: 5153-5162. Res. 24: 573A. (Abstr.) 22. Ho, Y. K., M. S. Brown, D. W. Bilheimer, and J. L. Gold- 14. Williams, L. T., D. Mullikin, and R. J. Lefkowitz. 1976. stein. 1976. Regulation of low density lipoprotein re- Identification of alpha-adrenergic receptors in uterine ceptor activity in freshly isolated human lymphocytes. smooth muscle membranes by [3H]dihydroergocryptine J. Clin. Invest. 58: 1465-1474. binding. J. Biol. Chem. 251: 6915-6923. 23. Cheng, Y. C., and W. H. Prusoff. 1973. Relationship 15. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. between the inhibition constant (Ki) and the concentra- Randall. 1951. Protein measurement with the Folin tion of inhibitor which causes 50 percent inhibition phenol reagent. J. Biol. Chem. 193: 265-275. (150) of an enzymatic reaction. Biochemn. Pharmacol. 16. Lefkowitz, R. J. 1974. Stimulation of catecholamine- 22: 3099-3108. sensitive adenylate cyclase by 5'-guanylyl-imidodiphos- 24. Furchgott, R. 1967. The pharmacological differentiation phate.J. Biol. Chem. 249: 6119-6124. of adrenergic receptors. Ann. N. Y. Acad. Sci. 139: 17. Saloman, Y., C. Londos, and M. Rodbell. 1974. A highly 553-570.

402 K. D. Newman, L. T. Williams, N. H. Bishopric, and R. J. Lefkowitz