The Relationship between al-Adrenergic Receptor Occupation and the Mobilization of Intracellular Calcium”

We have simultaneously quantitated al-adrenergic receptor occupation and agonist-elicited Ca2+ mobilization monitored as unidirectional 4aCa2+ efflux from intact BC3H-1 muscle cells in order to examine the relationship between the number of surface receptors occupied and the functional response. [SH]Prazosin has been used to measure receptor number as well as the binding kinetics with surface receptors, and the ob- served equilibrium and kinetic constants are in close accord with values obtained previously in cellular ho- mogenates. Since al-agonist-elicited 4aCa2+ efflux can be monitored over intervals of 3 min or less and pra- zosin dissociation from its receptor has a t1/2 of 44 min, prazosin can be employed to produce a pseudoirrever- sible inactivation of receptors. A comparison of the remaining receptors and residual response reveals an inverse linear relationship between receptors inactivated by prazosin and 45Caz+ efflux. A similar result is obtained following fractional receptor inactivation with the irreversible alkylating agent phenoxybenzamine. Parameters of receptor occupation and response also correlate well for the agonist phenylephrine and for the competitive antagonist phentolamine. The uni- tary relationship between sites available for occupation and response indicates that the al receptor does not function as an oligomer where fewer bound antagonist molecules are required to block the receptor than sites of agonist occupation necessary for activation. Moreover,

The Relationship between al-Adrenergic Receptor Occupation and the Mobilization of Intracellular Calcium" (Received for publication, May 7, 1984) Gabriel AmitaiS, R. Dale Brown, and Palmer Taylor We have simultaneously quantitated al-adrenergic receptor occupation and agonist-elicited Ca2+ mobilization monitored as unidirectional 4aCa2+ efflux from intact BC3H-1 muscle cells in order to examine the relationship between the number of surface receptors occupied and the functional response.
[SH]Prazosin has been used to measure receptor number as well as the binding kinetics with surface receptors, and the observed equilibrium and kinetic constants are in close accord with values obtained previously in cellular homogenates. Since al-agonist-elicited 4aCa2+ efflux can be monitored over intervals of 3 min or less and prazosin dissociation from its receptor has a t 1 / 2 of 44 min, prazosin can be employed to produce a pseudoirreversible inactivation of receptors. A comparison of the remaining receptors and residual response reveals an inverse linear relationship between receptors inactivated by prazosin and 45Caz+ efflux. A similar result is obtained following fractional receptor inactivation with the irreversible alkylating agent phenoxybenzamine. Parameters of receptor occupation and response also correlate well for the agonist phenylephrine and for the competitive antagonist phentolamine. The unitary relationship between sites available for occupation and response indicates that the al receptor does not function as an oligomer where fewer bound antagonist molecules are required to block the receptor than sites of agonist occupation necessary for activation. Moreover, substantial evidence has accrued in intact smooth muscle for a receptor reserve or nonlinear coupling between al receptor occupation and contraction in smooth muscle. Our findings demonstrate that such behavior does not exist for al receptor-elicited mobilization of Ca2+ in the BC3H-1 muscle cell. a-Adrenergic receptors have been classified as a1 and aZ subtypes based on their specificity for antagonists and certain agonists (1). These receptors can also be distinguished in terms of their functional responses where al receptors have been shown in certain systems to stimulate phosphatidylinositol turnover and promote the release of intracellular Ca2+, while stimulation of a2 receptors inhibits adenylate cyclase (2, 3). In recent years coupling proteins which inhibit the * This work was supported in part by United States Public Health Service Grant HL-25457. G. A. was a fellow of the Weizmann Foundation and R. D. B. is Penrose Stout Advanced Postdoctoral Fellow of the California Heart Association. The costs of publication of this article were defrayed in part by the payment of page charges, This article must therefore be hereby marked "aduerti,vemnt" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
$ Present address, Israel Institute for Biological Research, Ness Ziona 70450, Israel. cyclase have been identified, extending our knowledge of how az receptors might transduce their response (4), but much less is known about the mechanism of signal transduction of a1 receptor occupation. Whether phosphatidylinositol hydrolysis and Ca2+ mobilization are proximal or distal events of a single sequence of reactions following receptor occupation or whether they exist as separate parallel events remains unclear. The recent finding that inositol 1,4,5-trisphosphate will release cellular Caz+ suggests that at least in some systems it may serve as a mediator of Ca2+ release (5).
The BCSH-1 cell line expresses both a1-and &-adrenergic receptors (6)(7)(8)(9) and shows a number of features in common with vascular smooth muscle in vivo (10). Stimulation of a1 receptors in these cells mobilizes sequestered Ca2+ stores, substantially increasing unidirectional 45CaZ+ efflux over short time intervals (11). Since the cells grow in monolayers with minimal extracellular matrix, equivalent exposure of surface receptors can be achieved. As part of a larger endeavor to examine the linkage between a1 receptor activation and Ca2' mobilization, we have utilized the radiolabeled antagonist [3H]prazosin to quantitate available receptors on intact cells in concert with the functional response of 45Ca2' efflux. By inactivating receptors irreversibly we compare here the residual response with the extent of receptor inactivation. In addition we examine parameters of receptor occupation and response for the agonist phenylephrine and the reversible antagonist phentolamine.

cq-Adrenergic Receptors a d Intracellular Calcium
3.0-ml volume of the same buffer and incubated for 2.5 h at 37 "C. Alternatively, aliquots of a concentrated ['Hlprazosin stock solution were added directly to the experimental growth medium, and the cultures were returned to the incubator for a 2.5-h incubation at 37 "C in an atmosphere of 90% air/lO% COz. The reaction was stopped by washing the monolayers with four 3-ml aliquots of icecold PB. Monolayers were solubilized from the plates with two 0.5ml washes of 3% (w/v) Triton X-100 in 10 mM EGTA, 5 mM HEPES, pH 7.4. ['HIPrazosin associated with the cells was counted in a liquid scintillation spectrometer at 25-30% efficiency. Nonspecific binding was assessed in parallel cultures by adding 10 p~ phentolamine HC1 to the specified ('H]prazosin concentrations. Total cellular protein content of replicate cultures was analyzed by the method of Lowry et d. (12) and averaged -0.5 mg per 35-mm dish. These data were used to calculate specifically bound uersus free ['Hlprazosin concentrations. Results were analyzed either by Scatchard analysis to determine KO and B,, or by the empirical Hill formula (13),  ) or absence (total binding) of 10 phi phentolamine. Dissociation of ['Hlprazosin at 37 "C was initiated by replacing the conditioning solution with a fresh solution containing 1 phi unlabeled prazosin in the presence or absence of 10 p~ phentolamine. At specified time intervals plates were washed rapidly with four 3-ml volumes of icecold PB. Specific binding was computed from the difference between total and nonspecific radioactivity bound in the presence of 10 pM phentolamine. Data were analyzed as a first-order reaction.
Association Kinetics of [3H]Pratosin-Cultures were prepared as above, and ['Hlprazosin at 0.2 or 0.4 nM was allowed to react for specified time intervals in the presence (nonspecific binding) and absence (total binding) of 10 p~ phentolamine. The reaction was stopped by four 3-ml rapid washings with ice-cold PB. Equilibrium binding was estimated from the specifically bound radioligand after 2.5 h of incubation. Specific ['Hlprazosin binding data were analyzed as a pseudo first-order approach to equilibrium plotted according to the following integrated rate equation, where P, is the concentration of bound prazosin at equilibrium, P, is the concentration of bound prazosin at time t, and k h is the pseudo first-order association rate constant of kl(Po) + k-1. Here k, is the bimolecular association rate constant, k-l the dissociation rate constant, and P o the initial prazosin concentration. This analysis assumes that PO is sufficiently in excess of P, to consider the free prazosin concentration effectively constant throughout the binding reaction.
Agonist-elicited "Caz+ Efflux from BC3H-1 Cells-The basic procedures for equilibrating cells with *CaClZ and measuring agonistelicited '%a2+ unidirectional efflux were described in our previous study (11). Initial rates of 'Ta2+ efflux were measured over a 3-min interval of agonist exposure. Efflux was quenched by rapid washing with PB containing 5 mM LaC13 but no MgClz to avoid L a 3 + precipitation. Efflux was treated as a first-order process, and the rate constant was estimated from the "Ca*+ retained after 3 min relative to "Ca2+ in cultures which received the wash protocol without intervening efflux interval. The ,fractional response was then calculated from Equation 3, where kE, k$, and kE-are the observed, basal, and maximal rate constants (rnin") for efflux of "CaZ+. In certain experiments (Figs. 6 and lOA), rapidly exchanging "Ca" was removed during an initial 2min efflux into PB. Agonist-stimulated "Ca2+ efflux was then initiated by rapidly aspirating the PB and replacing it with a fresh 2-ml aliquot of PB containing the specified agonist concentration for a 3min flux interval. Efflux was terminated as above and the rate constant calculated relative to cellular T a 2 + following the 2-min washout interval. The concentration dependence for agonist was similar for both procedures, but the latter technique yielded more precise data at low agonist concentrations.
In experiments where inhibition of the functional response by antagonist was assesed, the antagonist was added during the final 1-2.5 h of exposure to "CaZ+. Antagonist was maintained at the same concentration throughout subsequent washes and the unidirectional efflux interval. Antagonists did not affect the total accumulation of '%a2+ or its basal efflux after removal of "Caz+ from the medium.
Occupationand Inhibitionof Receptors by Irreversible Antagonists-After equilibration with '%a2+ for 18-21 h and the specified concentrations of ['Hlprazosin during the final 2.5 h under standard growth conditions, cultures were rapidly washed three times with 3-ml aliquots of PB and incubated with 6 p~ phenylephrine in the presence of the specified concentration of ['Hlprazosin for 3 min. Efflux of "Ca" was terminated by washing with LaCl3, and cultures were analyzed for retained %a2+ as described previously. Receptor occupancy was measured by incubating sister cultures containing no '%a2+ for 2.5 h with specified ['Hlprazosin concentrations. As before, nonspecific binding at each concentration of ['Hlprazosin was ascertained using 10 phi phentolamine.
Irreversible inactivation by phenoxybenzamine was carried out in PB to prevent scavenging by nucleophilic components of the growth medium. Phenoxybenzamine stock solutions were prepared in cold double-distilled water and kept on ice (1-2 h) before final dilution into PB. Solutions were used within 5-10 min following dilution into the buffer. To measure receptor occupancy, sets of cultures were washed free of growth medium with PB and then exposed to 2-ml aliquots of PB containing specified phenoxybenzamine concentrations (0.1-100 nM) for 10 min at 37 "C. Each culture was then washed with three 3-ml aliquots of PB and equilibrated for 1 h at 37°C with 2 ml of PB containing a saturating concentration of ['H]prazosin (290 phi) in the presence or absence of 10 p~ phentolamine to determine residual specific ['Hlprazosin binding. Phenoxybenzamine inhibition of al-receptor response was measured in sister cultures equilibrated with '%az+. Each culture was washed once with 2 ml of PB containing "Ca2+ at the same specific radioactivity used in loading (5 &i/ml) and then incubated with a second aliquot of PB containing "Caz+ plus the specified concentration of phenoxybenzamine for 10 min at 37 "C. The culture was then washed with three 5-ml aliquots of PB, and unidirectional "Ca2+ efflux was measured over a 3-min interval in the presence of 6 pM phenylephrine.
Agonist and Antagonist Competition with Equilibrium fH]Prazosin Binding-Cultures were washed three times with 3-ml aliquots of PB and then incubated for 30-45 min at 37 "C with 2 ml of PB containing specifiedconcentrations of phentolamine (10-8-10-6 M) or phenylephrine (lO-'-lO-' hi). The incubation solution was aspirated and replaced with 2 ml of PB containing the same concentration of agonist or antagonist plus 2.9 X 10"' M ['Hlprazosin. Nonspecific binding was determined using 10% phentolamine. The reaction was stopped after 1 h and specific ['Hlprazosin binding data were analyzed by the Hill formulation described above. Equivalent results were obtained without prior exposure to the competing drug by simultaneously adding radioligand and competing agonist or antagonist followed by 2.5 h of equilibration at 37 "C.

Kinetics of Prazosin Association and Dissociation with cq
Receptors on Intact BC3H-I Cells-Kinetics of dissociation of [3H]prazosin from intact BC3H-1 cells in the presence of excess unlabeled prazosin are shown in Fig. 1. The dissociation of both total and nonspecifically bound prazosin (initial binding carried out in the presence of 10 p~ phentolamine) shows deviations from a first-order process. Dissociation of prazosin presumed to be specifically bound to a1 receptors is ascertained by subtracting the values obtained in 10 pM phentolamine from the total binding. The subtraction yields kinetics of dissociation which approximate a first-order reaction with a rate constant ( L 1 ) of 0.016 min". This value compares well with our previous studies of [3H]prazosin dissociation from a crude membrane fraction prepared from these cells = 0.018 min-') (9). The kinetics of total and nonspecific association of [3H]   prazosin (0.4 nM) are shown in Fig. 2A. The difference between the two conditions also yields a kinetic profile for specific binding of [3H]prazosin. Analysis of the data according to the integrated rate expression given in Equation 2 for a bimolecular process, ['Hlprazosin + receptor .prazosin-receptor yields an association rate constant (k,) of 3.1 X 10s M" min". This value is also in good agreement with the rate constant found for prazosin binding to crude membranes (9) of 2.4 X 108 M" min". The dissociation constant determined from the Concentration Dependence of Phenylephrine-elicited 'Ta2+ Efix-In previous studies with BCSH-1 cells we demonstrated that a1 receptor activation mobilizes Ca2+ from sequestered cellular stores, resulting in increased unidirectional T a 2 + efflux and depletion of intracellular Ca2+ (11). Fig. 4 shows time courses of unidirectional %aZ+ efflux obtained at increasing concentrations of the agonist phenylephrine. Although the efflux kinetics are clearly multiphasic as expected for a compartmentalized distribution of cellular Ca2+ (14), the predominant effect of agonist stimulation is to augment the initial phase of '%a2+ efflux in concentration-dependent fashion. This response forms the basis for quantitative assay of a1 receptor function as shown in Fig. 5. In order to determine the concentration dependence of agonist activation, a true rate constant of efflux rather than a quantity depleted from the cell should be ascertained. Accordingly we have estimated the initial rate of efflux from the fraction of 'Ta2+ retained at 3 min, assuming a first-order process. The fractional agonist-stimulated response was then calculated as described in Equation 3 by correcting for basal efflux and normalizing to the response obtained at a maximally effective agonist concentration. Hill analysis of the data obtained using phenyl- used to monitor initial rates of 45Ca2+ efflux, this antagonist could be employed to act as a pseudoirreversible inhibitor of receptor response. This argument is established by the data in Fig. 6, where prior blockade of a1 receptors with increasing concentrations of prazosin results in apparent noncompetitive inhibition of phenylephrine-stimulated efflux of 45Ca2+. That is, increasing prazosin concentrations depress the response obtained at maximal phenylephrine concentrations without effect on the agonist affinity of the remaining unoccupied receptors. In a separate experiment, we verified that loT5 M phenylephrine does not affect the rate of prazosin dissociation (data not shown).

al-Adrenergic Receptors
The specificity and kinetics of [3H]prazosin binding can be exploited in parallel with the functional assay to inactivate temporarily varying fractions of the receptor population and thus quantitate the relationship between available receptors and residual functional responsiveness. The results of such an experiment are shown in Fig. 7. Fig. 7A shows the influence of prior occupation by increasing concentrations of [3H]prazosin upon the 45Ca2+ efflux response to a near-saturating phenylephrine concentration (6 pM). In sister cultures [3H] prazosin binding is measured following similar conditions of incubation, and data similar to that described in Fig. 3 are generated. Plotted in Fig. 7B is the relationship between available binding sites and the residual functional response following increasing [3H]prazosin occupation. An inverse linear plot with a slope of -0.88 is found. In duplicate experiments, half-maximal inhibition of phenylephrine-stimulated allow complete formation of the reactive aziridinium ion prior to addition to the cells (15). Phenoxybenzamine has long been known to inactivate irreversibly al-adrenergic receptors (16), and in some cases it appears to possess the requisite specificity to identify labeled receptors (17). Following exposure to various concentrations of phenoxybenzamine in sister cultures, we have examined the residual al-adrenergic binding sites ascertained from specific [3H]prazosin binding (Fig. 8A) and the remaining response to phenylephrine (Fig. 8B). In both cases, we note a steep concentration dependence for phenoxybenzamine inactivation with apparent Hill slopes greater than 1.0. Since phenoxybenzamine is an irreversible agent, one can only measure the extent of inactivation rather than the true concentration dependence for receptor binding. There is also no assurance that the concentration of phenoxybenzamine in the vicinity of the receptor is equal to that in the bulk medium, and the scavenging of a finite amount of the reactive nucleophile would be sufficient to yield the narrow range of apparent concentration dependence. Hence the unusually large Hill coefficient does not necessarily imply a cooperative reaction of phenoxybenzamine with multiple sites on the receptor. Correlation of the data in Fig. 8, A and B, again reveals a unitary relationship between the number of available receptors and the remaining functional response (Fig. 8C). The slope of the line is -1.04. In duplicate experiments halfmaximal loss of available a1 receptor binding and response occurred at 7.6 and 5.0 nM phenoxybenzamine, respectively. Hence the relationship between receptor number as determined by [3H]prazosin binding and the residual response appears to be the same whether inactivation is a result of the pseudoirreversible block by prazosin or alkylation by phenoxybenzamine.
Equilibrium Binding Competition between [3HJPrazosin and Reversible Agonists or Antagonists-Equilibrium competition binding experiments between [3H]prazosin and unlabeled ligands should provide a means to measure fractional occupancy by specified concentrations of a-adrenergic agonists or antagonists. By using Equation 2 to determine the concentration of competing drug which produces half-maximal displacement of specifically bound [3H]prazosin the KD for the drug can be calculated using Equation 4 (18).  Hlprazosin (2.9 X 1 0 " ' M) in the presence or absence of 10 p~ phentolamine was added for 1 h at 37 "C to ascertain the number of specific ['H]prazosin binding sites. Data are expressed relative to controls which received no phenoxybenzamine pretreatment. Specific binding was determined from the difference between total binding and nonspecific binding measured in the presence of 10 p~ phentolamine. E, concentration dependence of phenoxybenzamine inhibition of a1 receptor-mediated '%a2+ efflux. Sister culture dishes equilibrated with 'Ta2' were exposed to the same phenoxybenzamine concentrations as in A with external '%a*+ maintained at constant specific radioactivity during the exposure interval. After rapid washing the initial rate of efflux was  (19). The value for KI estimated in this manner was 1.66 2 0.15 x lo-' M. This value was independently confirmed in the experiment shown in Fig. 10B. Here the response to a test concentration of phenylephrine (6 p~) was measured in the presence of increasing concentrations of phentolamine. The IC, value for phentolamine inhibition was calculated from Equation 1, and the phentolamine KI was estimated using Equation 4 to correct for the concentration of competingphenylephrine relative to its K&. This procedure gave values of KI = 1.87 * 0.22 X 10" M and nH = 1.12 f 0.04, in good agreement with the experiment in Fig. 1OA where agonist concentrations were varied.

DISCUSSION
Antagonist and Agonist Association with a1 Receptors on Intact BC3H-1 Cells-We have modified a previous prazosin binding assay (9) for measurements of a-adrenergic receptors on monolayer cultures of intact cells. Respectable ratios of phentolamine-dissociable to nondissociable binding have been achieved (Fig. 3), and we have defined the binding dissociable by 10 p~ phentolamine as associating specifically with sites on the a,-adrenergic receptor. Since nonselective association of prazosin is also relatively slow on intact cells, measurement of the kinetics of prazosin association with the receptor requires simultaneous monitoring of total and nonspecific binding. Specific association behaves as a bimolecular reaction while specific dissociation can be described by a simple unimolecular process. The K D value determined by kinetic techniques (52 PM) agrees with the value determined in equilibrium measurements (20 PM). Equilibrium binding also reveals no evidence for substantial cooperativity in association, although some variability in the average Hill coefficient was observed. Similar findings were also obtained when binding was measured at 21 "C (not shown). Thus in these cells [3H] prazosin binds to a population of equivalent noninteracting sites. The [3H]prazosin binding assay, therefore, may be used to assess binding constants of competing reversible ligands and to quantitate available receptors following irreversible blockade.
Receptor number and dissociation constants determined here on intact cells ( B , = 106 fmol/mg; KD = 20 PM) are in close accord with those found previously when [3H]prazosin binding measurements were performed on cellular homogenates (Emax = 137 f 32 fmol/mg; K D = 86 f 38 PM (9)). These results indicate an absence of appreciable internal receptors with capacity to bind the radioligand. We also note that phenylephrine and the hydrophobic ligand phentolamine will dissociate prazosin from an equal number of sites. Since the phenylethylamine exists in ionized form at physiological pH values, ita penetration to the interior of the cell should be minimal. However, we cannot exclude the possibility that homogenization destroys a fraction of surface receptors, and this fraction is compensated by the exposure of receptors normally internal to the cell.
Functional Responses Resulting from a-Adrenergic Occupation-Two phenomena have been proposed to mediate al receptor activation in various tissues. The first is an enhanced turnover of phosphatidylinositol, which has been measured either as increased catabolism of phosphoinositides and polyphosphoinositides or as resynthesis of phosphatidylinositol following the presumed breakdown reaction (2, 3,20). Although inositol 1,4,5-trisphosphate will effect the release of Caz+ from parotid glands and hepatocytes rendered permeable by saponin treatment (51, it remains unclear how the turnover of these phospholipids is related to the excitatory or inhibitory processes that ensue following receptor stimulation. In other tissues al-receptor stimulation has been shown to increase the efflux of Ca" (2,3,21,22), presumably reflecting mobilization of intracellular Caz+. Other ionic events such as K+ efflux are probably more distal to receptor activation since in liver and in Taenia coli the bee venom apamin blocks the latter response but not Caz+ efflux (21, 23).
Enhanced phosphatidylinositol breakdown and synthesis can be detected in BC3H-1 cells upon agonist stimulation (24). Because of calcium's role in excitation-secretion and excitation-contraction coupling we have preferred to use unidirectional '%a2+ efflux as a marker for the functional response. Moreover, Ca2+ mobilization constitutes a sufficiently large fraction of total internal Ca" to allow adequate quantitation over relatively short time periods. The treatment of initial rate data as a first-order process underestimates the magnitude of efflux rate constants due to the multiphasic kinetics of %!a2+ efflux. However, the relative error of the measurement does not change with the extent of receptor activation. This conclusion is supported by comparison of Fig.   12526

al-Adrenergic Receptors and Intracellular Calcium
10, A versus B, where similar KI values for phentolamine were obtained independently of the extent of receptor activation. Comparison of the agonist activation data in Figs. 5 and 6 (solid circles) further indicates that similar concentration dependences for agonist activation were obtained whether or not rapidly exchanging 45Ca2+ was removed during an initial 2-min incubation in physiological buffer.

The Linkage between Receptor Occupation and Response-
We might consider three general cases that would relate receptor occupation to the functional response. 1) In this case there is the presence of a receptor reserve. These considerations stem from the now classical observations of Stephenson (25) and later Furchgott (26,27) on the presence of spare muscarinic receptors in gastrointestinal smooth muscle. Indirect evidence has accrued that this preparation possesses a muscarinic receptor reserve and means have been developed for its quantitation by measuring the receptor response following progressive degrees of irreversible inactivation. With such a receptor reserve, occupation of only a fraction of the total receptors will give rise to the maximum response. If we let y be the number of receptors inactivated then the fractional response k / k , will be k k,

(64
Both the extent of receptor reserve and the fractional response in the absence of receptor inactivation will govern the value of n in Equation 6a (see "Appendix").
2) The receptor behaves functionally as an oligomeric protein where activation requires simultaneous agonist occupation of more than a single site ( n sites). If antagonist occupation of a single site on the oligomer is sufficient to block the receptor then For example, the nicotinic acetylcholine receptor requires agonist occupation of two sites to activate the ion channel, while antagonist occupation of one site is sufficient to block the response (28,29). Accordingly, the relation of k/k, = (1y)' best fits the data when cobra a-toxin irreversibly blocks the sites (28).
3) The receptor is a monomeric or oligomeric protein and the response is proportional to the number of sites occupied by agonist k -= 1y.

(6c)
Our findings with irreversible inactivation are consistent with the third case. While this is the simplest scheme, it would not necessarily be anticipated from previous data. Initial studies from rabbit aorta (30,31) as well as more recent data from rat vas deferens (32,33) are consistent with al-adrenergic receptor reserves in these tissues. That is, increasing extents of receptor inactivation shift the agonist concentration-response curve to higher agonist concentrations and only at higher concentrations depress the maximum agonist-elicited response (see "Appendix").
There are several reasons which might account for the differences between our results and previous studies measuring contractile responses from intact tissue. First, in other systems phenoxybenzamine has been shown to block Ca2+ channels and influence Ca2+ permeability independently of action at the receptor site (34). Should such an action prevail in BC3H-1 cells, the loss in functional response at each level of receptor inactivation would be overestimated. Hence k m the phenoxybenzamine occupancy-inactivation relationship would be skewed toward the situation described in Case 2. The observation that both prazosin and phenoxybenzamine yield inverse linear relationships between the number of aIadrenergic receptor sites and the mobilization of intracellular Ca2+ suggests that we are not detecting phenoxybenzamine inactivation at sites other than the site of prazosin labeling. The fact that phenoxybenzamine is active at nanomolar concentrations provides additional evidence for selectivity in its site of action. Second, the observed contractile response may not be proportional to receptor occupancy because of the structural and functional complexity inherent to intact smooth muscle. For example, activation of a fraction of cells with greater accessibility to bath-applied agonist may be sufficient to elicit a full response. Alternatively, punctate stimulation of sites on a few cells might activate neighboring cells by electrotonic or mechanical coupling. Studies in cell culture obviate these difficulties since each cell receives an equivalent exposure to agonist, and its contribution to the cumulative 45Ca2+ efflux will reflect the fractional agonist occupation of that cell.
A third and more intriguing possibility may be that measurements of 45Ca2+ efflux versus muscle contractile response reflect fundamentally different functional consequences of a1 receptor activation. Since both Ca2+ efflux and muscle contraction are presumably distal to the mobilization of intracellular Ca2+ which occurs upon aI receptor activation, differing extents of amplification could intervene between the two responses. Differing Ca'+ concentration dependences for actomyosin activation relative to the plasmalemma1 Ca'+ transport system could also contribute to these results. Moreover, a1 receptor activation may elevate intracellular Ca'+ in a highly nonuniform manner because of specialized compartmental relationships between a1 receptors, the contractile apparatus, and sites of Ca'+ storage or transport. Delineation of these possibilities awaits simultaneous measurement of actomyosin activation and intracellular free Ca'+ in intact smooth muscle cells, which should be feasible in the BC3H-1 system.
Combined measurements of receptor occupancy and functional response should also be useful for studying the actions of reversible agonists and antagonists, providing correction is made for the presence of competing ligands. In general the data obtained with phenylephrine and phentolamine support the receptor inactivation studies and argue against the presence of significant receptor reserve in this system. However, small discrepancies exist between the equilibrium dissociation constants determined for receptor binding to these ligands versus activation constants for excitatory or inhibitory responses. These differences may reflect an alteration occurring in receptor state during the prolonged agonist exposure in the equilibrium binding assays relative to the shorter duration of the functional response measurement. Altered a1 receptor affinity for agonists has been recently described in studies of intact uersus homogenized preparations of BC3H-1 cells (35). It will, therefore, be of interest to examine the affinity of the receptor for agonist over short time intervals by measuring competition with initial rates of [3H]prazosin binding. We would emphasize that the studies with prazosin and phenoxybenzamine studies reported here avoid these complications since receptors are occupied by the antagonist prior to measurement of agonist-elicited response from the remaining unoccupied receptors.
to Kelly Ambler and Paul Culver for helpful discussions.

APPENDIX
Activation of a1 receptors in our system approximates a hyperbolic function (cf. Fig. 5 ) . In the absence of receptor reserve the fractional response k/km will be proportional to fractional occupation. From the law of mass action, Here ( A ) , (AR), and (RT) denote the concentrations of agonist, agonist-receptor complex, and total receptor sites, with KA the equilibrium dissociation constant for agonist. Upon reduction in receptor number by inactivation the modified response becomes where q is the fraction of unmodified receptor sites remaining (26). Accordingly the response observed relative to that before inactivation will be k' k where y is the fraction of receptor sites inactivated. This situation corresponds to the case described in Equation 6c (see text). In systems where evidence for receptor reserve has been documented, the concentration-response relationship for agonist maintains its hyperbolic form (cf. Case 1 and Refs. 26 and 27). Under these conditions fractional response is a function of both agonist occupation and efficacy (e). Efficacy is defined as the relative capacity of a drug to elicit a response upon receptor occupation (25,26). The efficacy of an antagonist equals zero, whereas agonist efficacies may theoretically take on any positive value.
As derived by Furchgott for a hyperbolic dependence of response on agonist concentration (27), and following fractional inactivation, The relative response obtained at a given agonist concentration following fractional inactivation will be the ratio of these last two expressions, which may be simplified to yield Thus Equation A-4 describes the concentration dependence for response to agonist, whereas Equations A-6 and 6a describe the response obtained at fixed agonist concentration following receptor inactivation by increasing concentrations of irreversible antagonist. In the limit where appreciable receptor reserves are absent, e is small and the expression reduces to the linear dependence on agonist occupation de-scribed in Equation A-3 above. When e is substantial, k ' / k will be greater than q, and a parabolic (concave outward) relationship between receptor inactivation and the functional response will be observed. It is noteworthy that the extent of curvature will increase with increasing the ratio A/Ka.