Selectivity of Dobutamine for Receptor Subtypes: IN VITRO ANALYSIS BY RADIOLIGAND BINDING

R. Sanders Williams, Timothy Bishop

J Clin Invest. 1981;67(6):1703-1711. https://doi.org/10.1172/JCI110208.

The cardiovascular responses elicited by dobutamine are distinctly different from those produced by other adrenergic or agonists. To test the hypothesis that dobutamine could have differential affinities for subtypes, and that such subtype selectivity could be related to its relatively unique pharmacologic properties, we assessed the ability of dobutamine to displace adrenergic radioligands from membrane receptors in a number of tissues of previously characterized adrenergic receptor subtype. For beta adrenergic receptors identified by (−) 3 [ H] (DHA), dobutamine had significantly greater affinity for the β1 subtype (KD = 2.5 μM in rat and 2.6 μM in turkey erythrocyte) than for the β2 subtype (KD = 14.8 μM in frog heart and 25.4 μM in rat lung) (P < 0.001). For 3 alpha adrenergic receptors, dobutamine had markedly greater affinity for the α1-subtype identified by [ H] (KD = 3 0.09 μM in rat heart and 0.14 μM in rabbit uterus) than for the α2-subtype identified by [ H]dihydroergocryptine (DHE) (KD 3 = 9.3 μM in human platelet) or by [ H] (KD = 5.7 μM in rabbit uterus) (P < 0.001).

Like other β1-agonists, in the absence of guanine nucleotide, dobutamine competition curves for DHA binding in rat heart demonstrated two classes of binding sites, with one site of significantly higher affinity (KD = 0.5 μM, P = 0.008) than the single class […]

Find the latest version: https://jci.me/110208/pdf Selectivity of Dobutamine for Adrenergic Receptor Subtypes

IN VITRO ANALYSIS BY RADIOLIGAND BINDING

R. SANDERS WILLIAMS and TIMOTHY BISHOP, Department of Medicine, Cardiovascular Division, Duke University Medical Center, Durham, North Carolina 27710

A B S T R A C T The cardiovascular responses elicited We conclude that dobutamine is selective for f3 as by dobutamine are distinctly different from those pro- opposed to ,32, and for a, as opposed to a2 adrenergic duced by other adrenergic or dopaminergic agonists. receptors. Furthermore, guanine nucleotide effects on To test the hypothesis that dobutamine could have dif- dobutamine binding, and biochemical response data in ferential affinities for adrenergic receptor subtypes, and vitro suggest that dobutamine is a 8,3-agonist, but has that such subtype selectivity could be related to its rela- little intrinsic activity at ,/2 and a2-receptors. This selec- tively unique pharmacologic properties, we assessed tivity for adrenergic receptor subtypes may be part of the ability of dobutamine to displace adrenergic radio- the basis for dobutamine's distinctive pharmacologic ligands from membrane receptors in a number of tis- properties in vivo. sues of previously characterized adrenergic receptor subtype. For beta adrenergic receptors identified by INTRODUCTION (-) [3H]dihydroalprenolol (DHA), dobutamine had significantly greater affinity for the ,f3 subtype (KD = 2.5 The pharmacologic effects produced by infusions ofthe MM in rat heart and 2.6 MM in turkey erythrocyte) than synthetic catecholamine dobutamine differ from those for the (82 subtype (KD = 14.8 AM in frog heart and 25.4 elicited by other adrenergic agonists. At doses ,uM in rat lung) (P < 0.001). For alpha adrenergic re- producing equivalent enhancement of cardiac inotropy, ceptors, dobutamine had markedly greater affinity for dobutamine induces a smaller positive chronotropic the a,-subtype identified by [3H]prazosin (KD = 0.09 effect than isoproterenol. In addition, dobutamine ,uM in rat heart and 0.14 AM in rabbit uterus) than for induces neither the vasodilatory effects of isoprotere- the a2-subtype identified by [3H]dihydroergocryptine nol or epinephrine, which are mediated through (DHE) (KD = 9.3 MuM in human platelet) or by [3H]- vascular beta adrenergic receptors, nor the vasocon- yohimbine (KD = 5.7 MM in rabbit uterus) (P < 0.001). strictive effects of norepinephrine on arteriolar resist- Like other 3B,-agonists, in the absence of guanine nu- ance vessels, mediated through vascular alpha adren- cleotide, dobutamine competition curves for DHA ergic receptors. Furthermore, dobutamine is report- binding in rat heart demonstrated two classes of edly devoid of the presynaptic effects on endoge- binding sites, with one site of significantly higher nous release exhibited by its parent affinity (KD = 0.5 MM, P = 0.008) than the single class of compound and by other adrenergic ago- binding sites (KD = 5.2 MuM) identified in the presence nists (1-4). of guanine nucleotide. However, unlike /32- or a2-ago- The pharmacologic properties of dobutamine that nists, dobutamine displacement of DHA binding in rat account for its unique spectrum of hemodynamic ef- lung or of DHE binding in human platelets demon- fects have not been previously elucidated. One tenable strated only a single class of binding sites, and guanine hypothesis is that dobutamine may bind selectively to nucleotide had only minimal effects. specific subtypes of alpha and beta adrenergic recep- tors, and that such selective binding could be causally Dr. Williams is a recipient of a Young Investigator's Grant related to its distinctive pharmacologic effects. The from the National Heart, Lung and Blood Institute. Address original division of biologic responses to catechola- correspondence to Dr. Williams, P.O. Box 3945, Duke Uni- versity Medical Center, Durham, N. C. 27710. mines into two categories, alpha and beta, based on the Receivedfor publication 2 October 1980 and in revisedform relative potencies of adrenergic compounds in eliciting 17 February 1981. (or blocking) specific physiologic effects (5) has been

J. Clin. Invest. © The American Society for Clinical Investigation, Inc. * 0021-9738181/06/1703/09 $1.00 1703 Volume 67 June 1981 1703-1711 extended in recent years by further subclassification of The non-subtype selective antagonist (-) [3H]dihydroal- both beta adrenergic (6) and alpha adrenergic (7) re- prenolol (DHA)l (New England Nuclear, Boston, Mass.; specific activity 48 Ci/mmol) was utilized to identify /8- sponses. Since current concepts of adrenergic receptors in membranes derived from rat heart, and antagonist action suggest that the specificity of frog heart, turkey erythrocyte, and rat lung. The non-subtype biologic responses to adrenergic agents resides in the selective antagonist [3H]dihydroergocryptine (DHE) (New nature of the sarcolemmal adrenergic receptors, England Nuclear; 33 Ci/mmol sp act) was used to identify a- radioligand binding studies have proven to be a useful adrenergic receptors in human platelets and rat heart. We also used the a,-specific antagonist [3H]prazosin (Pfizer, Sand- complement to physiologic measurements in determin- wich, Kent, England; 33 Ci/mmol sp act) to label a,-receptors ing the subclassification of adrenergic receptors (8-11). in rat heart and rabbit uterus, and the a2-specific antagonist Furthermore, recent insights into the important regula- [3H]yohimbine (New England Nuclear; 81.4 Ci/mmol sp act) to tory function of guanine nucleotides in the coupling of label a2-receptors in rabbit uterus. Incubations were performed in the following respective adrenergic agonist binding to the subsequent bio- concentrations of Tris HCI and MgCl2: 50 mM, 16.7 mM, chemical and physiologic sequelae of receptor activa- pH 7.4 (rat heart, frog heart); 60 mM, 20 mM, pH 7.5 (turkey tion have expanded the role of radioligand binding erythrocyte); 40 mM, 5 mM, pH 8.0 (rat lung); 33.3 mM, 6.7 methods in assessing the molecular pharmacology of mM, pH 7.5 (rabbit uterus); 40 mM, 11 mM, pH 7.4 plus EDTA 2.5 mM (human platelet). Incubations were conducted for 20 adrenergic compounds. At least for 8l-, 12-, and a2- min at 25°C, with two exceptions: turkey erythrocyte mem- receptors, the ability to form a high affinity ligand-re- branes (20 min at 300C) and human platelet membranes (30 ceptor complex which is shifted to a lower affinity state min at 25QC). The effects of guanine nucleotides were assessed by guanine nucleotides is directly correlated with the by adding 0.1 mM (final concentration) guanyl-5'-yl imido- intrinsic activity (i.e., the maximum response produced diphosphate (GPP). We concluded the incubations by rapid vacuum filtration of the entire assay mixture over Whatman by saturating concentrations of agonist) of adrenergic GFC glass fiber filters (Whatman, Inc., Clifton, N. J.) followed agonists in eliciting biochemical or physiologic by rapid washing ofthe filters with 15-20 ml incubation buffer responses. Guanine nucleotides produce a large shift in at 0°C. Filters were placed in a Triton X-100-toluene based apparent affinity for full agonists and a smaller shift for fluor, and radioactivity was determined in a Packard liquid for scintillation counter (Packard Instrument Co., Inc., Downers partial agonists. Competitive binding curves Grove, Ill.). Other incubation conditions and binding char- antagonist compounds are unaffected by guanine nu- acteristics of each membrane-radioligand system we employed cleotides (12-14). are summarized in Table I. To test the hypothesis that dobutamine may bind The saturability, reversibility, stereospecificity, and selectively to specific subtypes of alpha and beta adrenergic receptor subtype specificity of radioligand bind- ing to each of the receptor populations we studied have been adrenergic receptors, we have assessed the ability of previously reported (8-11, 15-17, 19-22), with one excep- dobutamine to displace tritiated adrenergic ligands tion: the use of [3H]prazosin and [3H]yohimbine to label from both alpha and beta adrenergic receptors in a num- selectively the respective a1- and a2-receptor populations of ber of tissues in which the adrenergic receptor subtype rabbit uterus has'been recently characterized by T. N. Lavin In as and R. J. Lefkowitz.2 specificity has been previously defined. addition, The sources of the other reagents used include the follow- an indirect assessment of the specific activity of ing: dobutamine (Eli Lilly and Co., Indianapolis, Ind.), dobutamine for activation of responses mediated (Ciba Pharmaceutical Co., Summit, N. J.), through specific adrenergic receptor subtypes, we have (-) epinephrine, (-) isoproterenol, and (+) assessed the effect ofguanine nucleotides upon dobuta- (Sigma Chemical Co., St. Louis, Mo.). Protein was determined by the method of Lowry et al. (23). mine binding in these selected tissues. Adenylate cyclase assay. We prepared human platelet ly- sates and assayed adenylate cyclase activity as previously de- METHODS scribed by Tsai and Lefkowitz (24), with two exceptions: we utilized hypotonic lysis and mechanical homogenization for Preparation of membrane homogenates. Rat cardiac mem- preparation of platelet lysates without freezing the mem- branes were prepared as described in previous publications branes; and incubations were carried out for 10 min at 37°C. from our laboratory (15), and frog heart membranes were pre- We assessed the intrinsic activities of (-) epinephrine, a com- pared in an identical fashion. Membrane fractions from pound known to be a full agonist at a2-receptors (24), and of human platelet, rabbit uterus, rat lung, and turkey erythrocyte dobutamine by their respective abilities to inhibit the stimu- were prepared with minor modifications of previously pub- lation of adenylate cyclase elicited by 10 ,LM prostaglandin lished techniques (10, 16-18). (PG)E,. Triplicate determinations of adenylate cyclase Radioligand binding experiments. In each radioligand activity were assessed at each drug concentration in three binding experiment, aliquots of a given membrane prepara- separate membrane preparations. tion were incubated with a fixed concentration of radioligand in the presence of varying concentrations of dobutamine rang- ing from 1 nM to 1 mM. In addition, in each experiment, IAbbreviations used in this paper: DHA, [3H]dihydroal- radioligand binding in the absence of added dobutamine, and prenolol; DHE, [3H]dihydroergocryptine; GPP, guanyl-5'- in the presence of either 10 ,LM (t) propranolol (/3-receptors yl imidodiphosphate; KD, dissociation constant; PGE,, prosta- except lung), 10 tLM (+) propranolol (lung ,8-receptors), or 10 glandin El. ,aM phentolamine (a-receptors) was used to define maximal 2 Lavin, T. N., and R. J. Lefkowitz. Manuscript submitted binding and nonspecific binding, respectively. for publication.

1704 R. S. Williams and T. Bishop TABLE I Experimental Conditions and Radioligand Binding Characteristics for each Membrane Preparation under Study Concenitration Concentration Radioligand Membrane preparation membrane Radioligand radioligand KD Total binding Specific binding

mg protein/ml nM nM pM* % Rat heart 1.8 DHA 4.0 1.9 53+1 69 1.7 DHE 2.0 1.7 58±4 50 1.7 [3H]Prazosin 0.42 0.36 57+4 85 Frog heart 1.3 DHA 6.0 1.2 63+6 81 Turkey erythrocyte 1.6 DHA 10.5 5.9 328+21 87 Rat lung 0.35 DHA 3.7 2.0 164+34 71 Rabbit uterus 5.1 [3H]Prazosin 1.3 0.50 129±39 71 5.1 [3H]Yohimbine 13.3 10.0 349+97 56 Human platelet 1.1 DHE 4.2 7.8 300+56 75 * Mean-+SE.

Data analysis. In each experiment analyzing the dose-re- tion to an equivalent degree to the maximal inhibition sponse relationship of dobutamine competition for DHA, produced by 10 ,uM or 1 ,uM propranolol for DHA bind- DHE, [3H]prazosin or [3H]yohimbine binding sites in the membrane preparation under study, the binding of radio- ing or for 10 ,uM phentolamine for DHE, [3H]prazosin, ligand in the presence of each concentration of dobutamine or [3H]yohimbine binding. This indicates that do- was first normalized on a scale where binding in the presence butamine was indeed competing for the specific bind- of no added dobutamine was set at 100% and the binding in ing sites that represent physiologically relevant the presence of either 1 ,uM or 10 AM propranolol (X3-receptors) adrenergic receptors. or 10 ,uM phentolamine (a-receptors) was set at 0%. The nor- malized binding values at each concentration of dobutamine Beta adrenergic subtype specificity. Dobutamine from separate experiments were meaned, and the pooled dose- exhibited greater affinity for beta adrenergic receptors response curves obtained in this fashion were utilized for all of the 8,3-than of the 182-subtype. The dissociation con- subsequent analyses. stants for dobutamine in membranes derived from rat Each competition curve was analyzed by a nonlinear least- squares curve-fitting procedure employing a generalized heart and turkey erythrocytes, which are predominantly model for complex ligand-receptor interactions and per- of the f31-subtype (8, 22), were significantly lower than formed using an iterative program in PL/1 on a PDP 11/45 com- those observed in membranes derived from frog heart puter. Fitted parameter estimates were thus obtained for each and rat lung, which contain predominantly ,32-receptors competition curve. The details of this analytical method have (8, 10) (Table II; Figs. 1 and 2). In addition, dobutamine been previously described (25). To determine the significance between parameter estimates (such as KD) fronm two separate competition curves in rat cardiac membranes quanti- competition curves, each curve was first analyzed inde- tated in the absence of guanine nucleotide were best pendently determining the KD estimates which provided the characterized by two classes of binding sites: a high "best fit" of each separate dose-response curve. The curves affinity site (KD = 0.5 ,uM) representing 44% of the were then analyzed simultaneously constraining the KD esti- = mate for dobutamine competition for radioligand binding in receptor population, and a lower affinity site (KD 5 curve 1 to be equivalent to the KD estimate for curve 2. The ,uM) representing the remaining 56% of the receptors goodness of fit obtained by analyzing the data first inde- present. In the presence of 0.1 mM GPP, only a single pendently, and then with this constraint, was assessed by class ofbinding sites of the lower affinity (KD = 5.2 ,uM) measuring the residual variance between the data and each was observed (Fig. 1). The formation ofthis high affinity of the fitted curves, and comparing these variances by an F ratio test. For example, the KD estimated from dobutamine state, which is reversible by guanine nucleotide, is competition for radioligand binding in one tissue homogenate characteristic ofbeta adrenergic agonists in rat heart (8). (A) would be considered significantly different from On the other hand, dobutamine competition curves to the KD estimated from dobutamine competition for radioligand ,32-receptors of rat lung in the absence of guanine binding in a different homogenate (B) when the residual vari- ance of curves constraining KD (A) = KD (B) was increased nucleotide were described best by only a single class of significantly (by F test) from the residual variance observed binding sites, and were shifted only minimally by 0.1 when KD (A) and KD (B) were estimated independently. mM GPP. Shifts in agonist affinity for beta2 receptors of rat lung induced by guanine nucleotides have been RESULTS previously well characterized (13), and we observed We observed dobutamine to displace adrenergic radio- similar effects of 0.1 mM GPP on (-) isoproterenol ligands from binding sites in each membrane prepara- competition curves for DHA binding in our rat lung

Receptor Subtype Selectivity of Dobutamine 1705 TABLE II _. 100( - _ RO** hOMItA1) (-GPP) Inhibition of Radioligand Binding to Adrenergic 0 - E Ra_*I t hort (At) (*GPP) Receptor Subtypes by Dobutamine \§ b\ O Frog owt(V2)( GPP) x Tissue Radioligand Subtype Dobtutamine KD* Pt \ (D60 .\ \ JM z 0 \ \ Rat heart§ DHA PI, 5.2+0.6 Z 40 (4) <0.001 Frog heart DHA 32 14.8+±2.8 I I~~~~~~' (3) -'0 Turkey DHA A3l 2.6±0.4 8 7 6 5 4 3 2 erythro- <0.001 (-) Log [DOBUTAMINE] (M) cyte (2) Rat lung DIJA (2 25.4+3.1 FIGURE 1 Dobutamine competition for DHA binding in rat (3) heart and frog heart membranes. Each point is the mean of single or duplicate determinations in four (rat heart) or three Rat heart DHE a, 0.17+0.05 (frog heart) separate experiments. 100% maximal binding (3) refers to DHA binding in the absence of dobutamine and 0% Rat heart [3H]Prazosin a, 0.09±0.01 <0.001" binding refers to DHA binding to the presence of 10 AM (+) (4) propranolol. GPP 0.1 mM had no effect on dobutamine dis- Human DHE a2 9.3± 1.4 placement of DJIA binding in frog heart, so the binding values platelet for curves obtained in the presence and absence of GPP (2) were pooled. The lines represent computer-derived "best fits" Rabbit [3H]Prazosin a, 0.14±0.04 of the data points for each curve. uterus <0.001 (3) Rabbit [3H]Yohimbine a2 5.7±1.4 It should be noted that if one considers only the high uterus affinity site for dobutamine binding to rat heart recep- (3) tors observed in the absence of guanine nucleotides, the ,31-selectivity of dobutamine assessed by compar- * Except for turkey erythrocyte DHA binding, rat heart DHE ing its affinity for beta receptors of rat heart vs. rat lung binding, and rat heart [3H]prazosin binding, the values ex- pressed here are derived from experiments performed in the or frog heart increases to >10-fold, from the 3-fold re- presence of 0.1 mM guanyl-5'-yl imidodiphosphate (mean ported in Table II. ±SE). The statistical comparisons were made on competition Since (-) isoproterenol binds equally well to both curves obtained in the presence of quanine nucleotide be- ,f1- and,82-receptors, we compared the relative affinities cause such curves demonstrate only a single class of binding and in competing for sites, and the derived parameter estimates are more readily of dobutamine (-) isoproterenol compared than those derived from complex competition DHA binding in each of our beta receptor populations. curves obtained in the absence of guanine nucleotide, which, The KD for (-) isoproterenol in each tissue was either in some cases, demonstrated two classes of binding sites (e.g., dobutamine competition for DHA binding in rat heart; - Fig. 1). 100 0 o t F ratio test. See text for description of statistical methods. § Numbers in parentheses refer to the number of experiments from which data are derived. "Significance of rat heart (3) vs. human platelet and of rat heart (4) vs. platelet. z 40 - \ membrane preparation (Fig. 2). In a manner analogous x 20 - to the observations in rat lung membranes, we observed in a single experiment no effect of 0.1 mM GPP on dobutamine competition curves for DHA binding in 9 8 7 6 5 4 3 2 frog heart membranes (data not shown). Presumably oc(-) Log ADRENERGIC AGONIST (M) because the preparation contains endogenous guanine FIGURE 2 Dobutamine (open symbols) and (-) iso- nucleotide, the 813-receptors of crude (as opposed to proterenol (filled symbols) competition for DHA binding in highly purified) turkey erythrocyte membranes do not rat lung membranes in the presence (squares) and absence exhibit this shift in of (circles) of 0.1 mM GPP. Each point is the mean of single de- agonist affinities with the addition terminations in three separate experiments. 100% maximal guanine nucleotides (26), and we did not test guanine binding refers to DHA binding in the absence of dobutamine nucleotide effects upon dobutamine binding in this or (-) isoproterenol, and 0% binding refers to DHA binding tissue. in the presence of 1 ,uM (±) propranolol.

1706 R. S. Williams and T. Bishop derived from published data (rat heart [8], frog heart [8], turkey erythrocyte [27]), or determined in sepa- rate experiments (rat lung). The ratio ofKD dobutamine to KD (-) isoproterenol (in the presence of guanine nu- cleotide) in rat heart was 2.4 and was 2.8 in turkey erythrocyte, whereas this ratio was 7.1 in frog heart and 13.6 in rat lung. Alpha adrenergic subtype selectivity. Dobutamine exhibited greater affinity for alpha adrenergic receptors of the a,-subtype than of the a2-subtype. The dissocia- tion constant for dobutamine in membranes derived 8 7 6 5 4 3 2 from rat heart, which are predominantly et, (19), was sig- (-) Log [DOBUTAMINE] (M) nificantly lower than that observed in human platelet membranes, which are exclusively a2 (20) (Table II, FIGURE 4 Dobutamine competition for [3H]prazosin 0*) and uterine contain a mix- [3H]yohimbine (0) binding in rabbit uterine membranes. Fig. 3). Since rabbit membranes Each point is the mean of single or duplicate determinations ture of al- and a2-receptors (20), we used the subtype- in three separate experiments. Both curves were obtained in specific radioligand [3H]prazosin and [3H]yohimbine the presence of 0.1 mM GPP. For an explanation of the units to identify selectively each respective receptor poputa- on the ordinate, see legend to Fig. 3. tion. The dissociation constant derived from dobuta- mine competition for [3H]prazosin binding to al-re- ceptors was significantly lower than that observed for is best characterized by the assumption oftwo classes of dobutamine competition for [3H]yohimbine binding to binding sites, of which the high affinity site is not ob- a2-receptors (Table II, Fig. 4). served in the presence of GPP. This pattern is typical for a2-agonist compounds. On the other hand, guanine Since guanine nucleotides influence agonist affinity nucleotide had no effects on dobutamine competition at a2-receptors in an analogous manner to the changes for DHE binding sites. observed at /,3- and /32-receptors (14), we studied the These competitive binding data at a2-receptors were effects of0.1 mM GPP on dobutamine and (-) epineph- extended by our analysis of the effects of (-) epineph- rine binding to a2-receptors of human platelets. As rine and dobutamine upon adenylate seen in Fig. 5, (-) epinephrine competition for DHE PGE,-stimulated in membranes in the absence of GPP cyclase activity in human platelet lysates. In three ex- binding platelet periments, 10 ,uM PGE1 increased adenylate cyclase activity 26.5-, 20.2-, and 24.0-fold over basal values. We observed a dose-dependent inhibition of PGE -stimu- giOlo o^ lated adenylate cyclase activity by (-) epinephrine E (which could be blocked by 10 ,uM phentolamine, but E loo. 0 V not by 10 ,uM [+] propranolol), whereas dobutamine

Z 60 b z

az 40\ b\0 O 20 00

8 7 6 5 4 3 2 () Log [DOBUTAMINE] (M) FIGuRE 3 Dobutamine competition for [3H]prazosin (0) binding in rat cardiac membranes and for DHE (0) binding in human platelet membranes. Ea-h point is the mean of single or duplicate determinations in four (heart) or two (plate- (-) Log [ADRENERGIC AGONIST] (M) let) separate experiments. The platelet curve shown here demonstrates the binding in the presence of 0.1 mM GPP. FIGuRE 5 Dobutamine (open symbols) and (-) epinephrine Because guanine nucleotides do not alter agonist competition (filled symbols) competition for DHE binding to human curves for [3H]prazosin or DHE binding in rat cardiac mem- platelet membranes in the presence (squares) and absence branes (see text), the data for rat heart binding shown here (circles) of guanine nucleotide (0.1 mM GPP). Each point were obtained only in the absence of GPP. 100% maximal represents the mean of duplicate determinations in either binding refers to radioligand binding in the absence of five (epinephrine) or two (dobutamine) separate experiments. dobutamine, and 0% binding refers to binding in the presence For an explanation of the units on the ordinate, see legend to of 10 ,tM phentolamine. Fig. 3.

Receptor Subtype Selectivity of Dobutamine 1707 lacked this effect (Table III). The EC50 for (-) epineph- The relative lack of positive chronotropic effects of rine inhibition of PGE,-stimulated adenylate cyclase dobutamine compared with those produced by isopro- activity was 1.9 ,uM, paralleling its potency in the terenol at doses of each compound eliciting equivalent competitive binding assay (Fig. 5). positive inotropic effects has been demonstrated in clinical studies (2, 38, 39), in both conscious and anes- DISCUSSION thetized dog models (1-4) and in some isolated heart The widely accepted view (28) that both chronotropic preparations (1, 40). Although other investigators have and inotropic responses to catecholamines are medi- failed to observe an inotropic selectivity in isolated ated exclusively through cardiac adrenergic receptors myocardial preparations (41, 42), there is agreement of the ,63-subtype has recently been challenged on two that the inotropic selectivity of dobutamine in vivo is fronts. Persuasive evidence supporting the existence of not fully attributable to baroreceptor influences, or to functional postsynaptic cardiac alpha adrenergic re- effects on endogenous norepinephrine release, degra- ceptors influencing both chronotropy and inotropy has dation, or reuptake (1-3). been presented (15, 29-31); and some investigators By what mechanism, then, are the pharmacologic re- have advanced the hypothesis that cardiac receptors of sponses to dobutamine mediated? The relative selec- the 832-subtype may play a role in mediating chrono- tivity of dobutamine for inotropic as opposed to chrono- tropic, but not'inotropic, responses to catecholamines. tropic stimulation, when compared with (-) iso- For example, Carlsson et al. (32) observed in a reser- proterenol, renders untenable the hypothesis that both pinized anesthetized cat model that the ratio ofthe drug agents exert their direct cardiac effects exclusively dose required to produce a 50% maximal chronotropic through identical ,B,-receptors. An alternative hypothe- response to the dose required to produce a 50% maxi- sis is that dobutamine is relatively selective for stimu- mal inotropic response was 0.93 for the 8,-selective lation of /3- as opposed to /32-adrenergic receptors, and agonist (-) H80/60, 0.5 for the non-subtype selective that the chronotropic responses to catecholamines are agonist (-) isoproterenol, and 0.16 for the p32-selective indeed mediated more by agonist binding to ,/2 than to agonist . Other authors have reported pro- ,B,-receptors. Our current data demonstrate a greater portionately greater effects upon than upon affinity of dobutamine for ,B,- than for ,82-receptors in contractility from 82-selective, as opposed to non-sub- vitro, and, by virtue of the minimal ability of dobuta- type selective or /31-selective agents in other experi- mine to form the high affinity agonist-receptor com- mental preparations (33-37). plex which is reversed by guanine nucleotide at 182- receptors in vitro, predict that dobutamine should have TABLE III minimal stimulatory effects upon physiologic or bio- Inhibition of Prostaglandin-stimulated Adenylate chemical responses mediated through ,82-receptors. Cyclase Activity in Human Platelets Previous pharmacologic data confirm this prediction: dobutamine has only minimal effects upon peripheral Adenylate cyclase activity B32-receptors mediating relaxation of vascular smooth Drug concentration (-) Epinephrine Dobutamine muscle (1-4), and previous studies of biochemical re- sponses to dobutamine have also suggested that this % PGE, response compound has little agonist activity at ,82-receptors. For 0 100 100 example, Pike and Lefkowitz (43) found dobutamine to 0.1iM 99+3* 108±1t be a partial agonist for stimulation of adenylate cyclase 1lM 76±10 117±4 or of GTPase activity in turkey erythrocyte membranes 3 ,uM 63±14 120±4 (i83) with peak stimulatory activity 24 and 19%, respec- 10 AM 51±10 118±4 30,uM 43±12 122±7 tively, of that observed for (-) isoproterenol, whereas 0.1 mM 41±7 116±10 dobutamine had only minimal activity for stimulation of 0.3 mM 38±8 116±14 adenylate cyclase (8% of [-] isoproterenol response) 1 mM 34±6 84+19 or for GTPase activity (<5% of [-] isoproterenol response) in frog erythrocyte membranes (682). * Mean+SD of triplicate determinations from three ex- The surprisingly high affinity (greater than [-] epi- periments. nephrine) of dobutamine that we observed for alpha, t The increases in adenylate cyclase activity above that adrenergic receptors in rat cardiac and rabbit uterus induced by PGE1, which we observed in the presence of membrane homogenates suggests an alternative dobutamine, may reflect beta adrenergic agonist effects of dobutamine that are unopposed by a2-agonist effects. These hypothesis regarding the inotropic vs. chronotropic small increases were blunted by 10 ,uM (±) propranolol. selectivity of dobutamine's action. Since stimulation Propranplol had no effect on the inhibition of PGE,-stimulated of cardiac a,-receptors has been reported to augment adenylate cyclase activity induced by (-) epinephrine, where- inotropy while reducing the rate of phase 4 depolar- as this response was abolished by 10 ,uM phentolamine. ization (reduce chronotropy) in cardiac tissue (29-31),

1708 R. S. Williams and T. Bishop perhaps direct al-agonist activity in the heart could ac- = 15-25 ,uM) and for a2 receptors (KD= 6-9 ,M) in count for dobutamine's distinctive effects. The work of membrane preparations are considerably less than the Tuttle and Mills (1) suggested that dobutamine's ino- affinities of more typical adrenergic antagonists such as tropic effects were only minimally affected by phenoxy- propranolol (KD = 0.0046 ,uM in frog erythrocyte [,B2] benzamine, but the effects of alpha blockade on the membranes [46]), or phentolamine or yohimbine heart rate responses to dobutamine have not, to our (KD = 0.005 and 0.0008 ,uM, respectively, in human knowledge, been reported. Since agonist binding to al- platelet [a2] membranes [20]). Nonetheless, it is plaus- receptors does not appear to be markedly influenced by ible that antagonism of endogenous catecholamine ef- guanine nucleotides in rat heart3 and rabbit uterus (14), fects mediated through ,2- or a2-receptors may influence we cannot distinguish al-agonists from a,-antagonists the pharmacologic actions of dobutamine. Further- by radioligand binding experiments alone, and further more, 32- or a2-antagonist activity by dobutamine could studies will be required to resolve the agonist-antago- alter the effects of 32- or a2-agonist drugs that might be nist nature ofthe high affinity a,-binding ofdobutamine utilized concurrently with dobutamine. These possi- that we observed. bilities have not, to our knowledge, been rigorously Because previous pharmacologic investigations of tested in animal preparations or in man, and seem to dobutamine action have suggested that dobutamine merit further study. functions neither as an alpha adrenergic agonist nor as In a previous report, Minneman et al. (47) have as- an antagonist in vascular smooth muscle, some further sessed the affinity of dobutamine for inhibition of consideration of the functional significance of dobuta- [1251]iodohydroxybenzylpindolol binding to f31-recep- mine's high affinity a,-binding is required. An interest- tors of rat heart homogenates and to 132-receptors of rat ing hypothesis that would reconcile this apparent dis- lung homogenates; they concluded that no subtype crepancy has recently been formulated and supported selectivity of dobutamine was present. The explanation by data from Langer (44) and Yamaguchi and Kopin for the discrepancy between those findings and our (45). Studying blood pressure responses in pithed rats, results is not readily evident; however, the more highly these latter investigators have noted that the a1 selec- detailed nature of our current competition curves, tive antagonists and dibenamine coupled with the prior physiologic and biochemical were considerably more potent in blocking the pressor response data suggesting that dobutamine has only effects of direct sympathetic stimulation than in block- weak activity at 132-receptors, support the conclusion ing the effects of infused exogenous norepinephrine. that dobutamine is indeed a 81-selective ligand. The converse was true for the a2-antagonists We conclude that dobutamine has greater affinity for and . Citing the evidence for postsynaptic re- 3, - than for 132-adrenergic receptors, and for al- than for ceptors of the a2-subtype in other tissues (7), they pro- a2-adrenergic receptors. In addition, analysis of posed that two subpopulations of postsynaptic alpha guanine nucleotide effects on competitive binding as- adrenergic receptors mediating vasoconstriction are says for dobutamine would predict that dobutamine has present in peripheral blood vessels: a,-receptors agonist properties at 813-receptors, but little or no localized to the region of the vascular neuro-effector agonist activity at 132- or a2-receptors, and we have junctions, and separate a2-receptors which are the most confirmed this conclusion directly for a2 receptors. The important sites of action of infused or humorally cir- physiologic importance of findings obtained in studies culating alpha agonists. If this hypothesis is valid, and of cell-free systems must always be interpreted with if such a distinction between "intrasynaptic" a, and caution and, as other authors have suggested (41, 42), "extrasynaptic" (44) a2 postsynaptic receptors of vascu- dobutamine's inotropic selectivity may be unrelated to lar smooth muscle is physiologically relevant, the rela- its receptor binding characteristics. However, our data tively weak effects of infused dobutamine on vascular regarding the adrenergic receptor subtype selectivity of smooth muscle in vivo (1-4) would be predicted from dobutamine binding in vitro are best reconciled to its our radioligand binding studies, which demonstrated a known pharmacologic properties by the two recent much greater affinity of dobutamine for al- as opposed hypotheses suggesting a functional heterogeneity of (a) to a2-receptors, regardless of whether dobutamine is an the subtypes of beta adrenergic receptors mediating al-agonist or antagonist. chronotropy as opposed to inotropy in the myocardium, Since dobutamine clearly binds to both 132- and a2- and (b) the subtypes of postsynaptic alpha adrenergic receptors, yet exhibits only weak agonist effects, it receptors mediating vasoconstriction of vascular smooth would be expected, at appropriate concentrations, to muscle in response to exogenously administered, as function as an antagonist at receptors of these subtypes. opposed to endogenously released, adrenergic agonists. Dobutamine's binding affinities for 182-receptors (KD ACKNOWLE DGMENTS 3Williams, R. S., and R. J. Lefkowitz. Manuscript in prepara- The authors would like to acknowledge Dr. Robert J. Lefko- tion. witz both for his useful suggestions regarding the design of the

Receptor Subtype Selectivity of Dobuta mine 1709 study, and for his review of the manuscript. Drs. Brian Hoff- smooth muscle membranes by [3H]dihydroergocryp- man, Thomas Lavin, Michael Tharpe, Thomas Michel, Andre tine binding. J. Biol. Chem. 251: 6915-6923. DeLean, Andrew Wallace, and Judith Swain also contributed 18. Hoffman, B. B., D. Mullikin-Kilpatrick, and R. J. Lefko- valuable assistance. The assistance of Ms. Penny Hodgson and witz. 1979. Desensitization of beta-adrenergic stimulated Ms. Janice James in the editing and preparation of the manu- adenylate cyclase in turkey erythrocytes. J. Cyclic script is also gratefully appreciated. Nucleotide Res. 5: 355-366. 19. Williams, R. S., D. F. Dukes, and R. J. Lefkowitz. 1981. Subtype specificity of alpha adrenergic receptors in rat REFERENCES heart. J. Cardiovasc. Pharmacol. In press. 1. Tuttle, R. R., and J. Mills. 1975. Dobutamine: develop- 20. Hoffman, B. B., A. L. DeLean, C. L. Wood, D. D. ment of a new catecholamine to selectively increase car- Schocken, and R. J. Lefkowitz. 1979. Alpha-adrenergic re- diac contractility. Circ. Res. 36: 185-196. ceptor subtypes: quantitative assessment by ligand bind- 2. Sonnenblick, E. H., W. H. Frishman, and T. H. LeJemtel. ing. Life Sci. 24: 1739-1746. 1979. Dobutamine: a new synthetic cardioactive sym- 21. Wood, C. L., C. D. Arnett, W. R. Clarke, B. S. Tsai, and pathetic amine. N. Engl. J. Med. 300: 17-22. R. J. Lefkowitz. 1979. Subclassification of alpha- 3. Robie, N. W., D. D. Nutter, C. Moody, and J. L. McNay. adrenergic receptors by direct binding studies. Biochem. 1974. In vivo analysis of adrenergic receptor activity of Pharmacol. 28: 1277-1282. dobutamine. Circ. Res. 34: 663-671. 22. Atlas D., M. L. Steer, and A. Levitzki. 1974. Stereospecific 4. Vatner, S. F., R. J. McRitchie, and E. Braunwald. 1974. binding of propranolol and catecholamines to the Effects of dobutamine on left ventricular performance, ,8-adrenergic receptor. Proc. Natl. Acad. Sci. U. S. A. 71: coronary dynamics, and distribution of cardiac output 4246-4248. in conscious dogs. J. Clin. Invest. 53: 1265-1273. 23. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. 5. Ahlquist, R. P. 1948. A study ofthe adrenotropic receptors. Randall. 1951. Protein measurement with the Folin Am. J. Physiol. 153: 586-600. phenol reagent.J. Biol. Chem. 193: 265-275. 6. Lands, A. M., A. Arnold, J. P. McAuliff, F. P. Ludena, and 24. Tsai, B. S., and R. J. Lefkowitz. 1979. Agonist-specific ef- T. G. Brown. 1967. Differentiation of receptor systems by fects of guanine nucleotides on alpha-adrenergic re- sympathomimetic amines. Nature (Lond.). 214: 597-598. ceptors in human platelets. Mol. Pharmacol. 16: 61-68. 7. Berthelsen, S., and W. A. Pettinger. 1977. A functional 25. Kent, R. S., A. DeLean, and R. J. Lefkowitz. 1980. A basis for classification of a-adrenergic receptors. Life quantitative analysis of beta-adrenergic receptor interac- Sci. 21: 595-606. tions: resolution of high and low affinity states of the 8. Hancock, A. A., A. L. DeLean, and R. J. Lefkowitz. 1979. receptor by computer modeling of ligand binding data. Quantitative resolution of beta-adrenergic receptor Mol. Pharmacol. 17: 14-23. subtypes by selective ligand binding: application of a 26. Brown, E. M., S. A. Fedak, C. J. Woodward, G. D. Aurbach, computerized model fitting program. Mol. Pharmacol. 16: and D. Rodbard. 1976. ,8-adrenergic receptor interactions: 1-9. direct comparison of receptor interaction and biological 9. Hoffman, B. B., and R. J. Lefkowitz. 1980. Alpha-adrener- activity. J. Biol. Chem. 251: 1239-1246. gic receptor subtypes. N. Engl. J. Med. 302: 1390-1396. 27. Stadel, J. M., A. DeLean, and R. J. Lefkowitz. 1980. A 10. U'Prichard, D. C., D. B. Bylund, and S. H. Synder. 1978. high affinity agonist ,B-adrenergic receptor complex is an (±) [3H]Epinephrine and (-) [3H]dihydroalprenolol intermediate for catecholamine stimulation of adenylate binding to 81- and 82-noradrenergic receptors in brain, cyclase in turkey and frog erythrocyte membranes. heart, and lung membranes. J. Biol. Chem. 253: 5090- J. Biol. Chem. 255: 1436-1441. 5102. 28. Koelle, G. B. 1975. Neurohumoral transmission and the 11. Minneman, K. P., A. Hedberg, and P. B. Molinoff. 1979. autonomic nervous system. In The Pharmacologic Basis Comparison ofbeta adrenergic receptor subtypes in mam- for Therapeutics. L. S. Goodman and A. Gilman, editors. malian tissues. J. Pharmacol. Exp. Ther. 211: 502-508. McMillan Publishing Co., New York. 404-444. 12. Lefkowitz, R. J., D. Mullikin, and M. C. Caron. 1976. 29. Govier, W. C., N. C. Mosal, P. Whittington, and A. H. Regulation of,8-adrenergic receptors by guanyl-5'-yl imi- Broom. 1966. Myocardial alpha and beta-adrenergic dodiphosphate and other purine nucleotides. J. Biol. receptors as demonstrated by atrial function refractory Chem. 251: 4686-4692. period changes. J. Pharmacol. Exp. Ther. 154: 255-263. 13. Hegstrand, L. R., K. P. Minneman, and P. B. Molinoff. 30. Mary-Rabine, L., A. J. Hordof, F. 0. Bowman, J. R. Malm, 1979. Multiple effects of guanosine triphosphate on and M. R. Rosen. 1978. Alpha and beta-adrenergic effects beta adrenergic receptors and adenylate cyclase ac- on human atrial specialized conducting fibers. Circula- tivity in rat heart, lung and brain. J. Pharmacol. Exp. tion. 57: 84-90. Ther. 210: 215-221. 31. Rabinowitz, B., L. Chuck, M. Kligerman, and W. W. Parm- 14. Hoffman, B. B., D. Mullikin-Kilpatrick, and R. J. Lefko- ley. 1975. Positive inotropic effects of : witz. 1980. Heterogeneity of radioligand binding to a- evidence for alpha-adrenergic receptors in ventricular adrenergic receptors: analysis of guanine nucleotide regu- myocardium. Am. J. Physiol. 229: 582-585. lation of agonist binding in relation to receptor subtypes. 32. Carlsson, E., C. Dahlof, A. Hedberg, H. Persson and B. J. Biol. Chem. 255: 4645-4652. Tangstrand. 1977. Differentiation of cardiac chronotropic 15. Williams, R. S., and R. J. Lefkowitz. 1978. Alpha-adrener- and inotropic effects of f3-adrenergic agonists. Naunyn- gic receptors in rat myocardium: identification by bind- Schmiedeberg's Arch. Pharmacol. 300: 101-105. ing of [3H]dihydroergocryptine. Circ. Res. 43: 721-727. 33. Carlsson, E., B. Ablad, A. Brandstrom, and B. Carlsson. 16. Newman, K., L. T. Williams, N. Bishopric, and R. J. Lef- 1972. Differentiated blockade of the chronotropic kowitz. 1978. Identification of platelet alpha-adrenergic effects of various adrenergic stimuli in the cat heart. receptors by [3H]dihydroergocryptine binding. J. Clin. Life Sci. 11(I): 953-958. Invest. 61: 395-402. 34. Furnival, C. M., R. J. Linden, and H. M. Snow. 1971. The 17. Williams, L. T., D. Mullikin, and R. J. Lefkowitz. 1976. inotropic and chronotropic effects of catecholamines on Identification of a-adrenergic receptors in uterine the dog heart. J. Physiol. (Lond.). 214: 15-28.

1710 R. S. Williams and T. Bishop 35. Brittain, R. T., D. Jack, and A. C. Ritchie. 1970. Recent 41. Bodem, R., C. L. Skeleton, and E. H. Sonnenblick. 1974. j3-adrenergic . Adv. Drug Res. 5: 197-253. Inotropic and chronotropic effects of dobutamine on 36. Farmer, J. B., G. P. Levy, and R. J. Marshall. 1970. A com- isolated cardiac muscle. Eur. J. Cardiol. 2: 181-189. parison of the /8-adrenergic properties of 42. Lumley, P., K. J. Broadley, and G. P. Levy. 1977. , and sotemol with those of Analysis of the inotropic: chronotropic selectivity of . J. Pharm. Pharmacol. 22: 945-947. dobutamine and dopamine in anesthetized dogs and 37. Ablad, B., K. 0. Borg, E. Carlsson, L. Ek, G. Johnsson, guinea pig isolated atria. Cardiovasc. Res. 11: 17-25. T. Malmfors, and C. Regardh. 1975. A survey of the 43. Pike, L. J., and R. J. Lefkowitz. 1980. Activation and de- pharmacological properties of in animals and sensitization of 83-adrenergic receptor-coupled GTPase man. Acta Pharmacol. Toxicol. 36(Suppl. V): 7-23. and adenylate cyclase of frog and turkey erythro- cyte membranes. J. Biol. Chem. 255: 6860-6867. 38. Liers, C. V., P. T. Heban, P. Huss, C. A. Bush, and R. P. 44. Langer, S. Z. 1980. Subclassification of a-adrenoceptors Lewis. 1978. Comparative systemic and regional hemo- into a, and a2 types: relevance to cardiovascular effects of dynamic effects of dopamine and dobutamine in patients drugs. Circulation. 62(Suppl. III): 6-7. with cardiomyopathic . Circulation. 58: 466- 45. Yamaguchi, I., and I. J. Kopin. 1980. Differential inhibi- 475. tion of al- and Ca2-adrenoceptor-mediated pressor re- 39. Stephens, J., H. Ead, and R. Spurrell. 1979. Hemodynamic sponses in pithed rats.J. Pharmacol. Exp. Ther. In press. effects of dobutamine with special reference to myo- 46. Mukherjee, C., M. G. Caron, M. Coverstone, and R. J. cardial blood flow: a comparison with dopamine and iso- Lefkowitz. 1974. Identification of adenylate-cyclase prenaline. Br. Heart J. 42: 43-50. coupled beta-adrenergic receptors in frog erythrocytes 40. Tuttle, R. R. C. C. Hillman, and R. E. Toomey. 1976. Dif- with (-) [3H]. J. Biol. Chem. 250: 4869-4876. ferential f8-adrenergic sensitivity of atrial and ventricular 47. Minneman, K. P., L. R. Hegstrand, and P. B. Molinoff. tissue assessed by chronotropic, inotropic, and cyclic 1979. The pharmacological specificity of beta-I and beta-2 AMP responses to isoprenaline and dobutamine. Cardio- adrenergic receptors in rat heart and lung in vitro. Mol. vasc. Res. 10: 452-458. Pharrmacol. 16: 21-33.

Receptor Subtype Selectivity of Dobuta mtitie 1711