characterization of the humanplatelet a-adrenergic receptor€¦ · gic, p-adrenergic,...

Characterization of the Human Platelet

a-Adrenergic Receptor

CORRELATIONOF [3H]DIHYDROERGOCRYPTINEBINDING WITH

AGGREGATIONANDADENYLATECYCLASEINHIBITION

R. WN'AYNE ALEXANDER, BARRYCOOPER,and ROBERTI. HANDIN, Cardiouascullarand Hematology Divisions, Peter Bent Brigham Hospital, and the Departmenit ofMedicine, Harvard Medical School, Boston, Massachusetts 02115

A B S T R AC T Human platelets aggregate and un-dergo a release reaction wvhen inetubated wvith catechol-amines. Indirect evidence indicates that these eventsare mediated through a-adrenergic receptors. Weused[3 H]dihydroergocryptine, an a-adrenergic antagonist,to identify binding sites on platelets that have the char-acteristics of a-adrenergic receptors. Catecholaminescompete for the binding sites in a stereo-specific man-ner with the potency series of (-) epinephrine > (-)norepinephrine > (+) phenylephrine > (-) isoproter-enol. The dissociation constant (Kd) of (-) epinephrineis 0.34 AM. Binding is saturable using a platelet par-ticulate fraction but not with intact platelets. There are0.130 pmol binding sites per milligram protein in freshplatelet membranes. This number represents approxi-mately 100 binding sites per platelet. The Kd for [3H]-dihydroergocryptine was 0.003-0.01 AuM. The a-ad-renergic antagonist phentolamine (Kd = 0.0069 ,uM)was much more potent than the j-adrenergic antagonist(+) propranolol (Kd = 27 uM) in competing for thebinding sites. The binding data were correlated withcatecholamine-induced platelet aggregation and in-hibition of basal and prostaglandin El-stimulated ade-nylate cyclase. (-) Epinephrine was more potent than(-) norepinephrine in producing aggregation whereas(-) isoproterenol was ineffective. The threshold dosefor inducing aggregation by (-) epiniephrine was 0.46

This work appeared in abstract form in 1977: Clih. Res.25: 473A. (Abstr.)

Dr. Alexander is the recipient of a Career DevelopmentAward (lK04HL 0332), and Dr. Handin is an EstablishedInvestigator of the American Heart Association. Dr. Cooperwas supported by U. S. Public Health Service train-inig grant60-096-8393.

Receivedfor publication 25 April 1977 and in revisedform16 December 1977.

,M. Phentolamine and dihydroergocyrptine blockedthis response, whereas (±) propranolol had no effect.(-) Epinephrine and (-) norepiniephrine inhibitedbasal and prostaglandin El-stimulated adenylate cy-clase in a dose-related manner. The conicentrationi of( -)epiinephrinie inhibiting adenvlate cyclase 50% was 0.7,uM. This inhibitioni was also blocked by phentol-amiiine and dihydroergocryptine btut not by (±)propranolol. The specificity of binding and theclose correlationi with a-adrenergic receptor-mediatedbiochemical and physiological responses suggest thatthe [3H]dihydroergocryptine binding site represents,or is closely related to, the human platelet a-adrenergicreceptor. The ability to assay this receptor directly andto correlate these data with independently measuredse(uelae of receptor activation should facilitate in-creased understandinig of the physiology and patho-physiology of the hulmlani platelet a-adrenergic receptor.

INTRODUCTIONHuman platelets will aggregate and unidergo a releasereactioll when incuibated with catecholamines. Cate-cholamine receptors have been classified as a-adrener-gic, p-adrenergic, or dopaminergic accordinig to the po-tency of various catecholamiiines in inducing a responise.Thus, 8-adrenergic receptors have the potenicy series(-) isoproterenol > (-) epinepirine 2 (-) norepiinephi-rine, whereas a-adrenergic receptors respond to cate-cholamines with the order of potency (-) epinephrine> (-) norepinephrine > (-) isoproterenol (1). Indirectevidence suggests that catecholamiiine-iniduced aggre-gation of human platelets is mediated through an a-adrenergic receptor: (a) (-) Epinephrine is most potentwhereas (-) isoproterenol is virtually ineffective in in-ducing aggregation (2); (b) aggregation by epinephrineor norepinephrine is blocked by the a-adrenergic an-

J. Clin. Intvest. © The American Society for Clinical Itnvestigation, Inc., ()021-9738/7810501-1136 $1.001136

tagonists phentolamine or dihydroergotamine (2, 3);(c) inhibition of adenylate cyclase activity in plateletlysates by epinephrine is blocked by a-adrenergic an-tagonists (4); (d) the ability of epinephrine to preventthe rise in platelet cyclic AMP induced by prosta-glandin E1, an inhibitor of aggregation, is also blockedby a-adrenergic antagonists (5, 6).

Recent technical advances have made possible thedirect study of a-adrenergic receptors using radioactiveadrenergic antagonists which bind to the receptor site(7-9). In most previous studies, correlation of bindingdata with physiologic data has been achieved indirectlyby comparing affinity constants for binding to mem-branes with published affinity constants for inhibitionor stimulation of contraction in rabbit uterus (7) or ratvas deferens (9) by the same compound. Recently, adirect correlation betveen the potency of (-) epineph-rine in displacing [3 H]dihydroergocryptine (DHEC)1from dissociated rat parotid cells and in stimulatingK+ release from these cells has been noted (10). Wehave characterized the human platelet a-adrenergic re-ceptor by direct radioactive ligand binding assay usingthe a-adrenergic antagonist [3H]DHEC as the ligandand have correlated the binding data with physiologicand biochemical manifestations of platelet a-adrener-gic receptor activation.

METHODS

Materials. [3H]DHEC (spec act 24.1 Ci/mmol) was ob-tained from New England Nuclear, Boston, Massachusetts.This product was greater than 98%pure on thin-layer chroma-tography in two solvent systems (chloroform:ethanol:glacialacetic acid 9:5:1 and chloroform:benzene:ethanol:ammoniumhydroxide 4:2:1:0.1). [3H]DHEC was stored in the dark inethanol at -10°C. [a32P]ATP (10-30 Ci/mmol) and [3H]cAMP(3'5'-cyclic adenosine monophosphate; 25 Ci/mmol) were alsoobtained from New England Nuclear.

Other compounds used included (-) epinephrine bitartrate,(-) norepinephrine bitartrate, (-) isoproterenol bitartrate, (±)phenylephrine HCl, dopamine HCI, serotonin creatinine sul-fate, normetanephrine, catechol, vanillyl-mandelic acid, ade-nosine triphosphate (ATP), cAMP (Sigma Chemical Co., St.Louis, Mo.); (+) epinephrine bitartrate, (+) norepinephrinebitartrate, (+) isoproterenol bitartrate, and (±) ethyl-norepi-nephrine (Sterling-Winthrop, New York); phentolamine HCI(CIBA Pharmaceutical Company, Summit, N. J.); dihydroergo-cryptine, dihydroergocornine, and dihydroergocristine (San-doz Pharmaceuticals, East Hanover, N. J.); phenoxybenz-amine and chlorpromazine (Smith, Kline & French Labora-tories, Philadelphia, Pa.); azapatine (Hoffmann-LaRoche Inc.,Nutley, N. J.); (±) propranolol (Ayerst Laboratories, NewYork); clonidine HCI (Boehringer Ingelheim Ltd., Elmsford,N. Y.); mephentermine (Wyeth Laboratories, Philadelphia,Pa.); methoxamine (Burroughs Wellcome Co., Research Tri-angle Park, N. C.); imidazole (Eastman Kodak Co., Rochester,

I Abbreviations used in this paper: cAMP, 3',5'-cyclicadenosine monophosphate; DHEC, dihydroergocryptine;EGTA, ethylene glycol-bis (a,-amino-ethyl ether) N,N'-tetra-acetic acid, IC,,, half-maximal inhibition; Kd, dissociationconstant; PGE,, prostaglandin El; PRP, platelet-rich plasma.

N. Y.); L-tryptophan (Calbiochemii, Sani Diego, Calif:); andvohimbine (Aldrich Chemical Co., Inc., Milwaukee, Wis.).Prostaglandin E, (PGE,) was f'rom The Upjohn Comiipany(Kalamazoo, Mich.).

Platelet isolation and preparation of thfe particnlate frac-ti()n. 250 ml of blood was obtained f'rom each normal volun-teer studied. 200 ml of blood was mixed with sufficient acidcitrate dextrose anticoagulant (ACD; National Institutes of'Health Formula A) to lower the pH to 6.5. The remaining 50ml of blood was mixed with 5 ml 3.8% soli0(1m citrate to achievea final concentration of' 15 mMcitrate. Platelet-rich plasma(PRP) was prepared by centrif'ugation at 160 g for 10 mill at25°C in a Sorvall RC-3B centrifoige (Ivan Sorvall Ine., Nor-walk, Conn.). The PRP collected in sodium citrate was as-pirated and kept at room temperatture for up to 1 h for aggrega-tion studies outlined below. The remaining platelets collectedin ACDanticoagulant were sedimented at 1,600 g in the Sor-vall RC-3B centrifuge at 250C and washed twice with a buffercontaining 138 mMNaCl, 5 mMKCI, 8 mnMNa2HPO4, 2 mMNaH2PO4, 10 mMEDTA, pH 7.2 (phosphate-saline). Half' of'the platelet pellet was resuspended in a buffer containing138 mMNaCl, 5 mM MgCl2, 1 mMI etlhVlelle glyco-bis(,8-aminoethyl ether) N,N'-tetraacetic acid (EGTA), 25 mMTris HCl pH 7.55 (Tris-saline) at a proteini concentrationi of'3-5mg/ml.

The remaining half of the washed platelet pellet xvas sub-jected to two cycles of freezing on acetone-dry ice and thawingat room temperature followed by sonication f'or 10 s in a Black-stone BP-4 sonicator (Blackstone Ultrasonics, Inc., James-town, N. Y.). The particulate fraction was then collected bycentrifugation at 230,000 g for 30 min in a Beckman L5-65Bultracentrifuge e(quipped with an SW50.1 rotor (BeckmanInstruments Inc., Spinco Div., Palo Alto, Calif). The super-nate was discarded and the particulate fraction resuspendedin the Tris-saline buffer at a concentration of 3-5 mg/ml. Thewashed platelets and platelet particulate fraction derived fromthese same platelets were kept on melting ice for up to 2 h be-fore measurement of [3H]DHEC binding. Adenylate cyclaseactivity was measured on these same membranes immediatelyafter the freeze-thawing step.

In some experiments, platelet concentrates were obtainedwithin 72-96 h of collection from the Massachusetts Divi-sion of the American Red Cross. 20-40 concentrates werepooled and sufficient ACDadded to lower the pH to 6.5. Theplatelets were then processed as described above, except thatthe platelet particulate fraction was sedimented in a Beckmantype 40 rotor. These particulate fractions were stored at -80°Cfor up to 4 wk before use.

Binding assay. Immediately before assay, the washedplatelets or the two different platelet particulate fractions weregently suspended with a vortex mixer. 100-,ul aliquots of eachpreparation and about 40,000 cpm [3 H]DHECwere incubatedin the dark with and without agonists or antagonists for 17 minat 25°C in a total volume of 2 ml Tris-saline buffer unless other-wise specified.

The final concentration of [3H]DHEC was approximately0.8 nM. At the end of the incubation, 1.8 ml was added to 3 mlTris-saline buffer at room temperature and rapidly filteredthrough a Whatman GF/C glass filter (Whatman, Inc., Clifton,N. J.) under reduced pressure. The assay tube and filter werethen washed with four 5-ml portions of Tris-saline buffer. Thefilter was then dried and counted in 10 ml of Packard Instagelin a Packard liquid scintillation counter (Packard InstrumentCo. Inc., Downers Grove, Ill.) with an efficiency of 45%. Alldrug solutions were freshly prepared in distilled water im-mediately before use. The catecholamine and serotonin solu-tions also contained 0.1% ascorbic acid. Ergot alkaloids weredissolved in 50%ethanol before dilution in distilled H2O. The

Platelet a-Adrenergic Receptor 1137

highest concentration of ethanol (0.5%) did not affect binding.The [3H]DHEC in ethanol was added to assay buffer immedi-ately before use.

"Specific binding" was defined as the difference betweenradioactivity bound in the presence and absence of 100 ,uMphentolamine and is referred to in all figures and tables. Spe-cific binding amounted to 65-80% of total bound counts. Totalbound counts were approximately 3-4% of added radioactiv-ity. The filter blank was 0.5% of added radioactivity. Separateexperiments demonstrated that the agonists and antagonistsused did not affect the radioactivity bound to the filters.

Adenylate cyclase assay. Enzyme activity was measuredby the method of Salomon et al. (11) as previously modified(12) using a-[32P]ATP as substrate and directly measuring thea[32P]cAMP product. Assay mixtures contained 1.5 x 106 dpma[32P]ATP; 1.0 mMATP; 25 mMTris-HCl, pH 7.4; 5 mMMgCl2; 2 mMcAMP; 0.1% albumin; 10 mMtheophylline; 1mMEGTA; and an ATP-regenerating system consisting of 20mMcreatine phosphate and 1 mg/ml creatine kinase. The ad-dition of EGTAincreased basal and prostaglandin-sensitiveenzyme activity by 20%. Reactions were initiated by the addi-tion of 100 ul of platelet suspension containing 50-80 iLg pro-tein and incubated for 10 min at 37°C. Enzyme activity waslinear for at least 20 min at protein concentrations up to 80,ug. Data were expressed as nanomoles cAMPper milligramsprotein per 10 min.

Platelet aggregation. Platelet aggregation was carried outusing the standard nephelometric technique (13). Platelet ag-gregation was initiated by adding various catecholamines in0.1 ml buffer made up of 15 mMTris-HCl and 138 mMNaCl,pH 7.6, to 0.4 ml stirred PRPusing a chrono-log aggregometer(Chrono-Log Corp., Havertown, Pa.). The aggregometer wasadjusted so that PRP sample allowed 10% light transmissionand platelet-poor plasma from the same donor permitted 90%light transmission. The PRPwas stirred for 2 min at 37°C be-fore addition of catecholamine and, in some cases, the plateletswere incubated for an additional 5-20 min at room tempera-ture with a- or ,/-adrenergic antagonists before the addition ofthe catecholamines. The aggregation tracings were analyzedfor the presence of primary and secondary waves of aggrega-tion. The threshold dose of a given aggregating agent is de-fined as the lowest dose that produces a discernible secondarywave of aggregation. A change in light transmission of at least10% after the addition of a catecholamine was arbitrarily de-fined as a primary wave.

Platelet counts were performed on PRPusing an electronicparticle counter (Coulter model F, Coulter Electronics Inc.,Hialeah, Fla.). Protein determinations were carried out by thetechnique of Lowry et al. (14) using bovine serum albumin asa standard. The experimental protocol has been reviewed andapproved by the HumanSubjects Committee of the Peter BentBrigham Hospital.

Calculations. The interaction of [3H]DHEC with plateletsand platelet membrane particles was directly analyzed by themethod of Scatchard (15). The Kd for the nonradioactive ago-nists and antagonists was calculated from the concentrationrequired to inhibit [3H]DHEC binding by 50% (IC50) by themethod of Cheng and Prusoff (16) according to the relation-ship: Kd [I1 + [ligand]/Kd ligand] = IC50. The Kd of the ligand([3H]DHEC) is determined independently. This assumescompetitive inhibition of a first order ligand substrate inter-action. Each Kd value was determined from the average of twoor three experiments in which triplicate dose-response curveswere generated.

The Kd for DHECwas also calculated from kinetic data (7).The binding reaction was considered to be pseudo-first orderbecause it is reversible and because ligand concentration (1nM) was much greater than receptor concentration (0.05 nM).

The second order rate constant, k,, was calculated from therelationship k1 = (K0b -K2)/[DHEC]. The observed forwardrate constant, k,b, is determined from kob * t = (ln Bmaxl[Bmax - B]), where Bmax equals total specific binding at equi-librium and B equals binding at time t. The first order rateconstant, k2, is calculated from the relationship k2 * t = ln(BIBmax). Then Kd = k2/k1.

RESULTS

Saturability and affinity of [3H]DHEC binding sites.In these experiments, increasing concentrations of[3H]DHEC were incubated with intact platelets, theparticulate fraction derived from these platelets, or aparticulate fraction prepared from platelets 72-96 hafter collection. Fig. 1 depicts the binding of [3H]DHECto intact, washed platelets. Phentolamine-displace-able binding was linear and nonsaturable over the range1-40 nM [3 H]DHEC.

The pattern of [3H]DHEC binding by the particulatefraction derived from these same platelets is shownin Fig. 2. In contrast with the whole platelets, satura-tion occurred with 0.13 pmol ligand bound/mg protein.At a ligand concentration of approximately 4 nM, halfthe binding sites were occupied. This value gives anestimate of the dissociation constant (Kd) of [3H]DHECfor the binding sites. Analysis of the data by Scatchardplot (Fig. 2, inset) demonstrated a single class of bind-ing sites with a Kd of 10.3 nM. This analysis gave anestimate of 0.173 pmol ligand bound/mg protein whichrepresents approximately 100 binding sites per platelet.

A similar study performed with the particulate frac-

a 0.6z

n0m 0.5

z-ZP-0 0.4oh .

o E 0.3C9

o7tE 0.2

I0.1

0

[3 H] DIHYDROERGOCRYPTINE(M xlO9)

FIGURE 1 The binding of [3H]DHEC to intact platelets isdepicted. Mean values of triplicate determinations from fiveseparate experiments are shown. In these experiments in-creasing quantities of [3H]DHEC were added to a fixed num-ber of platelets (approximately 5 x 108). The samples wereincubated for 20 min at 25°C. Specific binding is plotted anddefined as the radioactivity displaceable by 100 ,uM phentol-amine.

1138 R. W. Alexander, B. Cooper, and R. I. Handin

Cz0mLU 0.20

0H- 1-x a 0.15O ED'-

OE OICQ~

cr0

0-

' 0.05

Li-:

r 075

- Nv 0.50 *0

-D w0.25O LL-m

O l0 20 30 40 50 60[3H] DIHYDROERGOCRYPTINEBOUND

(M x 1012)

0~~~~~~

I_

0 2 4 6 8 10 12[3H] DIHYDROERGOCRYPTINE(M x 109)

.50

120 2z E 0.90-:D0

w 0.60

_ 0.08 0.30

0 10 20 30 40 50

0.06- [3H] DIHYDROERGOCRYPTINEBOUND

oE

0I 4

14 16

FIGURE 2 The binding of [3H ] DHECto a particulate fractionderived from fresh platelets is depicted. Mean values fromiithree separate experiments with each point run in triplicateare shown. Assay conditions are identical to those describedin Fig. 1. Protein concentrations averaged 0.35 mg/ml. Theinset shows the Scatchard analysis of this data. The line repre-sents the least squares fit of the data points and has a slope of-9.7 x 107 M (r = 0.88). The reciprocal of the slope, theKd, equals 10.3 nM. The intercept of the line with the ordinategives an estimate of 0.173 pmol receptor/mg protein.

tion derived from stored platelets is shown in Fig. 3.The binding sites for [3H]DHEC were saturable withhalf the sites occupied at about 2 nM. The Scatchardplot again shows a single class of binding sites with aKd of 3.3 nMand 60 fmol ligand bound/mg protein. Thisvalue represents 33 binding sites per platelet, stuggest-ing that a loss of binding sites may occur during storage.

Specificity of [3H]DHEC bindinig. The ability ofvarious catecholamines to compete f{r binding sites onlthe intact platelets or their membrane fraction is illus-trated in Fig. 4. The potency series for the three (-)stereoisomers tested was (-) epinephrine > (-) nor-epinephrine > (-) isoproterenol which, as previouslynoted, is characteristic of an a-adrenergic response.The dose-response curves for intact platelets wereshifted to the right, and the maximum [3H]DHEC dis-placement by catecholamines was decreased comparedwith the platelet particulate fraction. (-) Epinephrine-displaceable counts in intact platelets were 30-60% ofthat displaceable by 100 uM phentolamine over therange 3-40 nM [3H]DHEC (data not shown). The Kdsfor (-) epinephrine, (-) norepinephrine, and (-) iso-proterenol were 0.41 ,tM, 2.5 ,uM, and 25.4 ,M, respec-tively, in the particulate fraction. The apparent differ-ences in the affinities of catecholamines for the binid-ing sites in the two preparations probably reflect thediscrepancy between the radioactivity displaceable by(-) epinephrine and by phentolamine in the intactplatelet. Thus, if specific binding is redefined in the in-tact platelet as that displaceable by 100 ,uM (-) epi-nephrine, then the IC50 is approximately 1 AuM, which

[ 2 3 4 5 6

[3H] DIHYDROERGOCRYPTINE(M x 109)

FIGURE 3 The binding of[3H] DHECto a platelet particulatefraction derived from stored platelets is depicted. Conditionsare identical to those described in Figs. 1 and 2. Values fromtwvo experiments are shown. Protein concentration was 0.4 mg/ml. The slope of the Scatchard plot is -3.0 x 101 M(r = 0.97)and the Kd e(quals 3.3 nM. There are 0.06 pmol receptor/miigprotein.

is not strikingly dissimilar from that in the particulatefraction.

Although the total number of [3H]DHEC bindingsites was diminished wheni a particuilate fraction wasprepared from stored platelets, the affinity for catechol-amines was not affected. The Kds for (-) epinephrine,

Lu 100 oZ o(-)E

o H-NE

azG0c

crz

~ 0

M40 ~ ~ ~ LO1 (GNIT

FIGURE 4 The inhibition Of [3H]DHEC binding by (-) epi-nephrinle (0), (-) norepinephrinle (O), and (-) isoproterenlol(V) is depicted. The left-hand panel depicts the response ofintact platelets and the right-hand panel a platelet particulatefraction prepared from the same platelets. W\hen studying theintact platelets, 1 pAlI pargylinle and 0).3 ,urM imipramline wsercincluded in the assay mixture to minimize uptake and metabo-lism of added catecholamines. At this concentration, neithercompound reduced specific binding. In these experiments,either 5 x 108 platelets or 300 ,ug of particle protein wereincubated 4vith -0.8 nM [3H]DHEC and varying concentra-tionls of the three agonists. The reactions s~ere carried out inTris-saline buffer for 17 minl at i5°C and terminated ly filtra-ti()n OntO glass fiber filters. The percent displacement of[3H]DHEC vs. the concetrationd of each agonist is plotted.Each point from three experiments was determined in tripli-cate and the actual Kd ealcilated from the IC50 as deseribed inthe text.


5.

5.

7 8

(-) norepinephrine, and (-) isoproterenol (0.34, 2.5,and 51.0 MM, respectively) were virtually identicalwith the values obtained with freshly prepared plateletmembranes. Competition for the binding sites is stereo-specific because the (+) isomers are approximately 10-fold less potent than are the (-) isomers. The KdS forthe (+) isomers of epinephrine, norepinephrine, andisoproterenol were 4.5, 18.0, and 430.0 uM, respec-tively.

Inasmuch as the binding characteristics of the partic-ulate fraction from these platelets were quite similarto those of the fresh material and were available inmuch larger quantities, more extensive analysis of [3H]-DHECbinding was carried out exclusively with thismaterial. The inhibition of [3H]DHEC binding by avariety of a-adrenergic antagonists is illustrated in Fig.5. The indolealkylamine alkaloid yohimbine is mostpotent (Kd = 0.0015 MuM) and the dibenzazepine de-rivative, azapetine, the least potent inhibitor (Kd = 0.95,uM). The dihydrogenated ergotoxine alkaloids (Kd= 0.014 ,uM) and the 2-substituted imidazoline phentol-amine (Kd = 0.0069,uM) are more potent than any of thecatecholamine agonists. For comparison the Kd for (±)propranolol, a well-recognized /8-adrenergic antago-nist, was 27 MM(Table I).

Table I also summarizes the effects on [3H]DHECbinding of some additional agonists and of several non-catecholamine compounds. Serotonin competes forbinding sites at relatively high concentration (Kd = 8.7,uM) whereas catecholamine metabolites like normeta-nephrine or vanillylmandelic acid were less active. Inaddition, very high concentrations of imidazole, cate-chol, and L-tryptophan did not block binding.

Binding kinetics. [3 H]DHEC binding to platelet

z

801-~~~~~0~~~~~(DI 60-cZO

wcr_ 40(-Iz

5 0 20-

9 OGI8 (ANTAGONIST) 3

FIGURE 5 The inhibition of [3H]DHEC binding by variousantagonists is depicted. Conditions were identical to those inFig. 4. Each curve represents mean values from two to threeexperiments with each point run in triplicate. Yohimbine(U), Kd 0.0015 ,uM; phentolamine (v), Kd 0.0069 ,tM; phen-oxybenzamine (0), Kd 0.26 ,M; azapetine (O), Kd 0.95 ,uM;dihydroergocryptine (@), Kd 0.014 ,uM; dihydroergocristine(0), Kd 0.014 ,M; dihydroergocornine (V), Kd 0.014 ,uM.

TABLE IEffect of Additional Agonists and Miscellaneous Compounds

on [3H]DHEC Binding

AntagonistsChlorpromazine(+) Propranolol

AgonistsClonidine(±) PhenylephrineDopamineMephentermine(±) Ethyl-norepinephrineMethoxamine

MiscellaneousSerotoninNormetanephrineVanillylmandelic acidImidazoleCatecoholL-tryptophan

Kd(OFM)

0.3027.0

0.0973.84.37.09.5

27.0

8.733.0

*

*

NItNI

* <50% inhibition at 0.1 mM.t NI, no inhibition (<1.O% at 10-4 M).

fragments reached equilibrium at approximately 20min with t12 of 8.5 min, and binding was stable for aslong as 60 min (Fig. 6). The observed forward rateconstant was 0.120 min-' giving a second order rateconstant (k,) of 1.03 x 108 min-' M-1. At equilibrium,addition of the a-adrenergic antagonist phentolamine(0.5 mM) resulted in slow displacement of [3H]DHECfor up to 40 min (Fig. 7). This slow reverse rate is com-patible with the observations that a-adrenergic block-ade with ergot alkaloids, although competitive, is quitepersistent relative to substituted imidazoline antago-nists such as phentolamine (17). Because of evidencethat the [3H]DHEC was not chemically stable forlonger than 60 min under the assay conditions, it wasnot meaningful to extend the incubation. The first orderrate constant (k2) for the reversal of [3H]DHEC bindingwas 0.017 min-'. The Kd calculated from the reverserate constant was 0.2 nMwhich is an order of magnitudeless than the Kd estimated from equilibrium data(2-10 nM).

Platelet aggregation. Based on the study of 10 nor-mal subjects, threshold doses that caused visible aggre-gation were 0.46 MMfor (-) epinephrine and 1.95 AMfor (-) norepinephrine, a fourfold difference in concen-tration. In each case, as the concentration of the twocatecholamines was increased, the magnitude of theprimary aggregation wave increased and the time to theonset of secondary aggregation decreased. Maximal ag-gregation, as measured by the final change in lighttransmission, was achieved with 0.93 MMepinephrineand 3.9 MMnorepinephrine, also a fourfold differencein dose. (-) Isoproterenol and methoxamine did not in-


zD0onLUizF- -

ox0--

oL

I0

C')I

100 -

m 2.0

E 1.5 -

mE kO = 0.12 0 mirnl

1- P/,~c 0 10 20

MINUTES

z0

U,5-.cn

4 60 -

40

-J 20

0

MINUTES

FIGURE 6 Rate of "specific" binding of [3H]DHEC to plate-let particles at 250C is depicted. The protein concentration was0.4 mg/ml. The reaction was initiated by adding membranesto two tulbes containing 1.0 nm [3H]DHEC. One tube con-tained 100 ,uM phentolamine. Each tube was sampled at thetimes indicated. Each point represents the mean of two experi-ments with duplicate determinations at each point. The insetshows the plot for calculating the observed forward rate con-stant. Bmax refers to maximum binding at e(luilibrium.

duce aggregation at any concentration studied up to amaximum of 0.35 mM.

(-) Epinephrine- and (-) norepinephrine-inducedaggregation were blocked by preincubation with phen-tolamine but not with (±) propranolol. As shown inFig. 8, increasing concentrations of DHECcould alsoinhibit (-) epinephrine-induced aggregation. Aggrega-

o-E 033 ke0.017min-

co0.6 * *0z 80 0.[qz 8 0 10 20 30 40

> : 60 MINUTES

0 X0<ctE 40o 0

0>-

>_ 20_

0 8 16 24 32 40MINUTES

FIGURE 7 Rate of dissociation of [3H]DHEC from plateletparticles at 250C. Platelet particles were incubated with 1.0nNI [3H]DHEC for 20 min, and at time zero 500 ,uM phentol-amine wras added, and aliquots were taken at the times indi-cated. Each point represents the mean of points determinedin duplicate from three separate experiments. The inset showsthe plot for determining the first order rate constant. The linerepresents the least squares fit (r = 0.9).

MINUTES

FIGURE 8 The inhibition of (-) epinephrine-induced plate-let aggregation by increasing concentrations of DHECis de-picted. PRPwas incubated with 0.1-1 ,M DHECfor 20 min atroom temperature in the dark. 0.4-ml portions were then trans-ferred to the aggregometer, stirred at 1,000 rpm at 37°C for2 min and the reaction initiated by addition of 5 ,uM (-) epi-nephrine. The change on the percentage of light transmissionwas continuously recorded for 5 min.

tion induced by collagen or ADPwas not inhibited by1 ,uM DHEC.

Adenylate cyclase. The effect of (-) epinephrineon basal adenylate cyclase activity is depicted in Fig. 9.In these experiments, the fresh platelet lysates from thesame platelets on which binding studies were per-formed were incubated with varying conceintrations ofthe catecholamine, and adenylate cyclase activity wasdetermined. In five experiments, (-) epinephrinelowered basal adenylate cyclase by as much as 30%(P > 0.01). Half-maximal inhibition, the IC50, occurredat 0.7 ,M (-) epinephrine, whereas maximal inhibitionrequired 10 uM (-) epinephrine. To increase totaladenylate cyclase activity, the membrane fractions

H-

C-)

z <OJmoJ

I t>i

z'1J

40F

30

p< 0.0I

p<0.0I

p<00I

20F

10

8 7 6 5 4

-LOGIO (-) EPINEPHRINE (M)

FIGURE 9 Inhibition of basal adenylate cvclase activity by(-) epinephrine is depicted. In these experiments, 60-80 jgof platelet protein was incubated for 10 min at 37°C in thepresence of increasing concentrations of (-) epinephrine.Adenylate cyclase activity was determinled as describedin Methods. Mean values (±SEM) f'rom five experimiienitswith each point determined in duplicate are showvn. TheP values refer to comparison using Student's t test betweenbasal adenylate cyclase activity and experimental poinits wvith(-) epinephrine added.

Platelet a-Adrenergic Receptor 114 1

were incubated with 5 ,uM PGE,. This concentrationof PGE1 stimulated enzyme activity an average of 23-f(Wd over base line. The effects of' several adrenergicagonists and antagonists on this PGE,-stimulated ad-enylate cyclase activity are outlined in Table II. PGE1increased basal activity from 0.28 to 6.17 nmol cAMP/mgprotein per 10 min of' incubation. The addition of 10,uM (-) epinephrine decreased PGE,-stimulated ad-enylate cyclase by 70% to 2.85 nmol/mg protein per10( miin. A similar inhibitory effect was noted after in-cubation with 10 ,uM (-) norepinephrine, whereas 10,uM (-) isoproterenol decreased adenylate cyclase ac-tivity by only 10%. Phentolamine completely blockedthe inhibitory eff'ects of (-) epineplhrine and (-) nor-epinephrine. DHECalso blocked inhibition by (-) epi-nephrine in a dose-dependent manner.

Table III summarizes and compares the effects of'catecholamines on [3H]DHEC binding, adenylate cy-clase inhibition, and aggregation threshold. The con-centrations of (-) epinephrine and (-) norepinephrinere(luired to initiate aggregation and to achieve half-maximal inhibition of adenylate cyclase are closelycomparable to the Kd for inhibiting [3H]DHEC bindingfor each compound. (-) Isoproterenol is much less po-tent than the other two catecholamines in inhibitingbinding and adenylate cyclase and in promoting aggre-gation. Thus, each independently measured manifesta-

TABLE IIEffect of Adrenergic Agonists and Antagonists on

PGE -Stimulated Adenylate Cyclase

Adenylate cyclaseAddition activity

nmol cAMP/mgprotein/1O min

None 0.285,M PGE, plus: 6.17

10lM (-) Epinephrine 2.8510 yM (-) Epinephrine, 2 AM

phentolamiine 6.1710 ,M (-) Norepinephrine 2.9210 ,uM (-) Norepinephrine, 2 ,uM

phentolamine 6.0310lM (-) Isoprotereniol 5.19

5 ,uM PGE, 10 ,uM (-) epinephrinie plus:10-9 MDHEC* 3.9910-8 MDHEC 4.1710-' MDHEC 6.001(-6 MDHEC 6.12

* For these experiments, intact platelets were incubatedwith the indicated concentration of DHEC, then centrifuged.The plasnma was discarded and the pellet disrupted by afreeze-thaw cycle. An aliquot was then added to the adenylatecyclase assay buffer containing 5 AM PGE, and 10 ,Mepinephrine to initiate the reaction.

TABLE IIICom1parison of [3H ]DHEC Binding and Physiologic

anid Biochemical Responses

ICC5, Adenylate AggregationiAgonist KI* cyclaseI threshold (dose§

AM AM AM

(-) Epinephrinie 0.34 0.7 0.5(-) Norepinephrinie 2.5 4.96 1.93() Isoprotereniol 51.0 >100 >300

K, wais caleilated from IC50 for binding usinlg freshly pre-palred memb)rane as (lescribed in the text.I IC50, the concentration of catecholamine inhibiting PGE,(5 /LM)-stintulated platelet adenylate cyclase 50%.§ Threshold dose, lowest concentration of catecholamineprotducing a biphasic aggregation pattern when added tostirred PRP.

tion of catecholamine function exhibited the a-adrener-gic potency series of (-) epinephrine > (-) norepi-nephrine > (-) isoproterenol. Furthermore, the dosesreqtuired to inhibit binding or adenylate cyclase andto initiate aggregation are quite similar.

DISCUSSION

Our data suggest that [3 H]DHEC, an a-adrenergicantagonist, identifies binding sites in human plateletswith characteristics of a-adrenergic receptors. First, thecatecholamine potency series for displacing [3 H]-DHECis appropriate, and a-adrenergic antagonists aremuch more potent than is the ,B-adrenergic antagonistpropranolol. The binding site exhibits stereospecificitywith respect to the competition of catecholamines forbinding, and binding is saturable in disrupted plateletparticulate fractions. Furthermore, inhibition of bind-ing and adenylate cyclase and induction of aggregationrequire the same concentrations of catecholamine. Thecharacteristics of' the binding and the close correlationwith a-adrenergic receptor-mediated biochemical andphysiological responses within the platelet suggest thatthe [3H]DHEC binding site represents, or is closelyrelated to, the a-adrenergic receptor.

Ideally, it would be desirable to characterize fullythe a-adrenergic receptor in intact platelets. This wasnot practical because of the lack of saturability of [3H]-DHECbinding sites in whole platelets. These findingsare not unexpected because similar findings have beenreported for intact oxygenated rat parotid cells (10). Up-take and internalization of the triated ligand by intactcells may be the explanation for these observations.In spite of these difficulties, a catecholamine potencyseries appropriate for the a-adrenergic receptors wasobtained even in the intact platelet (Fig. 4). The ap-parent rightward shift of the dose-response curves mayreflect a lack of accessibility of the ligand for displace-


ment by catecholamines under these circumstancesrather than a true change in the affinity.

Because [3H ] DHECbinding is much slower than theonset of a catecholamine-induced physiologic re-sponse, this ligand cannot be used for kinetic studies ofagonist-receptor interactions. In addition, the kineticstudies reported here give a Kd which differs from theKd calculated from equilibrium studies. Three previousstudies have reported kinetically derived Kds that wereconsistently two- to threefold less than the Kd derivedfrom equilibrium data in rabbit uterus (7), rat parotidcells (10), and rat brain (18). In view of these observa-tions and of the present data, the possibility must beconsidered that [3H]DHEC binding is more complexthan a simple bimolecular reaction with the a-adrener-gic receptor.

The potency of the primary catecholamine agonistsin displacing binding, presented in Fig. 4, reflects theknown potency of these compounds in stimulatingplatelet aggregation and release (2, 3). The binding datado not explain why a-adrenergic agonists such asphenylephrine and dopamine do not induce aggrega-tion since they are similar to (-) norepinephrine inpotency. The data do not support the suggestion thatthe human platelet a-adrenergic receptor may beunique in not interacting with such agonists (19). Dopa-mine and phenylephrine potentiate aggregation toother agents through an a-adrenergic mechanism (20).Thus, measurement of catecholamine-induced aggre-gation and adenylate cyclase inhibition may not beappropriate to assess intrinsic activity in this sytem.

Of the antagonists tested, all except the /8-adrenergicantagonist (±+-) propranolol have the pharmacologicproperties of a-adrenergic antagonists (17). The Kds re-ported here are in reasonable agreement with previ-ously published binding data (7, 9) except for yohim-bine which appears to be a more potent antagonist inthe human platelet than in uterine (7) or vas deferensmembranes (9). However, yohimbine has also been re-ported to be more potent than phentolamine in inhibit-ing platelet aggregation (2). (+) Propranolol is almostfour orders of magnitude less potent than is phentol-amine. The Kd of 27 ,uM is identical with that previouslyreported for inhibition of [3H]DHEC binding in rabbituterine membranes (7).

The nature of the interactions between the a-adren-ergic receptor and adenylate cyclase in various tissuesis unclear (1). It has been speculated that there is areciprocal relationship between a- and f3-adrenergicreceptors and the catalytic subunit of adenylate cyclase(21). Although the 8-adrenergic receptor clearly stimu-lates adenylate cyclase and increases intracellularcAMP, the general physiologic significance of a de-crease in cAMP mediated by a-adrenergic agonists isnot clear (1). In fact, a-adrenergic receptor-mediatedmetabolic events in rat liver have been demonstrated

recently which are not accompanied by changes incAMP (22).

The present data and those of another recent reportclearly demonstrate that catecholamines can inhibitbasal platelet adenylate cyclase activity (4). It is possi-ble that this inhibition is due to a direct link betweenthe platelet a-adrenergic receptor and the catalytic sub-unit of adenylate cyclase. It is also tenable that a-adren-ergic receptor activation alters membrane conforma-tion in a more general way and hence leads to inhibitionof this membrane-bound enzyme.

Although the concentration of catecholamines usu-ally found in the circulation may not be sufficienit tocause aggregation, catecholamines mav sensitize theplatelet to the effects of other aggregating substances(23-25). They may be responsible for the increasedaggregation seen wvith cigarette smoking and in animalsand humans subjected to psychologic and physicalstress (26-28). Platelets obtained from patients withhyperbetalipoproteinemia aggregate with very lowdoses of (-) epinephrine (29, 30). This hyperaggrega-bility may be a postreceptor event related to the in-creased cholesterol and increased microviscosity of theplatelet membrane or may be due to alteration in theplatelet a-adrenergic receptor. Thus, the platelet a-ad-renergic receptor may be important in modulatingevents in a number of pathologic states. The ability toassay this receptor directly in human platelets usingbinding techniques and to correlate these data withwell-defined biochemical and physiological correlatesof receptor activation should facilitate our understand-ing of the a-adrenergic receptor and its role in humanplatelet physiology and pathophysiologv.

ACKNOWLEDGCIENTSThe authors would like to acknowledge the excellenit technicalassistance of Vera Mlartin, Mary Ellen Dtugan, and RobertBayer.

This work was supported by grants from the U. S. PublicHealth Service (HL 17513 and HL 20054), from the Massa-chusetts Affiliate American Heart Association, and from Bio-medical Research Support finds (RR(5489).

REFERENCES1. Moran, N. C. 1975. Adrenergic receptors. Handb. Phyjsiol.

6: 447-472.2. Barthel, W., and F. Markwardt. 1974. Aggregation of blood

platelets by biogenic amines and its inhibition by anti-adrenergic and antiserotoninergic agents. Biochem. Phar-macol. 24: 37-46.

3. O'Brien, J. R. 1963. Some effects of adrenaline and anti-adrenaline compounds on platelets in vitro and in vivo.Nature (Lond.). 200: 763-765.

4. Jakobs, K. H., W. Saur, and G. Schultz. 1976. Reductionof'adenylate cyclase activity in lysates of human plateletsby the alpha-adrenergic component of epinephrine. J.Cytclic Nucleotide Res. 2: 381-392.

5. Salzman, E. XV., and L. L. Neri. 1969. Cyclic 3',5'-adeno-sine moniophosphate in human blood platelets. Natuire(Lond.). 224: 609-612.


6. Robison, G. A., A. Arnold, and R. C. Hartmann. 1969. Di-vergent effects of epinephrine and prostaglandin El onlevel of cyclic AMPin human blood platelets. Pharmacol.Res. Commun. 1: 325-334.

7. Williams, L. T., D. Mullikin, and R. J. Lefkowitz. 1976.Identification of a-adrenergic receptors in uterine smoothmuscle membranes by [3H]dihydroergocryptine binding.

J. Biol. Chem. 251: 6915-6923.8. Greenberg, D. A., D. C. U'Prichard, and S. H. Snyder.

1976. Alpha-noradrenergic receptor binding in mamma-lian brain: differential labeling of agonist and antagoniststates. Life Sci. 19: 69-76.

9. Ruffalo, R. R., J. W. Fowble, D. D. Miller, and P. Patil.1976. Binding of [3H]dihydroazapetine to alpha-adreno-receptor-related proteins from rat vas deferns. Proc. Natl.Acad. Sci. U. S. A. 73: 2730-2734.

10. Strittmatter, W. J., J. N. Davis, and R. J. Lefkowitz. 1977.a-Adrenergic receptors in rat parotid cells. I. Correlationof [3H]dihyroergocryptine binding and catecholamine-stimulated potassium efflux. J. Biol. Chem. 252: 5472-5477.

11. Salomon, Y., C. Londos, and M. Rodbell. 1974. A highlysensitive adenylate cyclase assay. Anal. Biochem. 58:541-548.

12. Cooper, B., and R. I. Gregerman. 1976. Hormone-sensi-tive fat cell adenylase cyclase in the rat: influences ofgrowth, cell size, and aging.J. Clin. Invest. 57: 161-168.

13. Buchanan, G., V. Martin, P. Levine, K. Scoon, and R. I.Handin. 1977. The effects of "anti-platelet" drugs onbleeding time and platelet aggregation in normal humansubjects. Am. J. Clin. Pathol. 68: 355-359.

14. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J.Randall. 1951. Protein measurement with the Folin phe-nol reagent.J. Biol. Chem. 193: 265-275.

15. Scatchard, G. 1949. The attractions of proteins for smallmolecules and ions. Ann. N. Y. Acad. Sci. 51: 660-672.

16. Cheng, Y. and W. Prusoff. 1973. Relationship between theinhibition constant (K1) and the concentration of inhibitorwhich causes 50 percent inhibition ('50) of an enzymaticreaction. Biochem. Pharmacol. 22: 3099-3108.

17. Nickerson, M., and N. K. Hollenberg. 1967. Blockade ofa-adrenergic receptors. In Physiological Pharmacology.Vol. 4. The Nervous System Part D: Autonomic NervousSystem Drugs. W. S. Root and Z. G. Hoffman, editors.Academic Press, Inc., New York. 243-305.

18. Davis, J. N., W. J. Strittmatter, E. Hoyer, and R. J. Lef-kowitz. 1977. [3H]dihydroergocryptine binding in ratbrain. Brain Res. 132: 327-336.

19. Berry, D. G., and J. Miller. 1974. The effects of sympatho-mimetic agents on ADP aggregation of rat platelets invitro. Eur. J. Pharmacol. 28: 164-171.

20. Drummond, A. 1976. Interactions of blood platelets withbiogenic amines: uptake, stimulation and receptor bind-ing. In Research Monographs in Cell and Tissue Physiol-ogy. I. Platelets in Biology and Pathology. J. Gordon,editor. American Elsevier Publishing Co. Inc., NewYork.203-239.

21. Robison, G. A., R. W. Butcher, and E. W. Sutherland.1967. Adenyl cyclase as an adrenergic receptor. Ann. N. Y.Acad. Sci. 139: 703-723.

22. Cherrington, A. D., F. D. Assimacopoulos, S. Harper, J.Borbin, C. Park, and J. Exton. 1976. Studies on the a-adrenergic activation of hepatic glucose output. II. In-vestigation of the roles of adenosine 3':5'-monophosphateand adenosine 3':5'-monophosphate dependent proteinkinase in the actions of phenylpherine in isolated hepa-tocytes.J. Biol. Chem. 251: 5209-5218.

23. Mills, D. C. B., and G. C. K. Roberts. 1967. Effects ofadrenaline on human blood platelet. J. Physiol. (Lond.).143: 443-453.

24. Niewiarowski, S., and D. P. Thomas. 1966. Platelet aggre-gation by ADPand thrombin. Nature (Lond.). 212: 1544-1547.

25. Thomas, D. P. 1968. The role of catecholamines in the ag-gregation of platelets by collagen and thrombin. Exp. Biol.Med. 3: 129-134.

26. Levine, P. H. 1973. An acute effect of cigarette smokingon platelet function. Circulation. 48: 619-623.

27. Haft, J. I., and R. Levites. 1974. Exercise-induced stressand platelet aggregation. Circulation. 50(Suppl. III):288A. (Abstr.)

28. Haft, J. I., and Y. S. Arkel. 1975. Effect of emotional stresson platelet aggregation in humans. Clin. Res. 28: 186A.(Abstr.)

29. Shattil, S. J., R. Anaya-Galindo, J. Bennett, R. W. Colman,and R. A. Cooper. 1975. Platelet hypersensitivity inducedby cholesterol.J. Clin. Invest. 55: 636-643.

30. Carvalho, A. C. A., R. W. Colman, and R. S. Lees. 1974.Platelet function in hyperlipoproteinemia. N. Engl. J.Med. 290: 434-438.


characterization of the humanplatelet a-adrenergic receptor€¦ · gic, p-adrenergic,...

Documents