THE ~JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 251, No. X, Issue of Aprd 25, pp. 2374-2384, 1976

Printed in U.S.A.

Solubilization and Characterization of the ,&Adrenergic Receptor Binding Sites of Frog Erythrocytes”

(Received for publication, September 30, 1975)

MARC G. CARON+ AND ROBERT J. LEFKOWITZ§

From the Departments of Medicine and Biochemistry, Duke University Medical Center, Durham, North Carolina 27710

Specific P-adrenergic receptors present in membrane preparations of frog erythrocytes were identified by binding of (-)- [3H]dihydroalprenolol, a potent competitive fi-adrenergic antagonist. The (-)- [3H]dihydroalprenolol binding sites could be solubilized by treatment of a purified erythrocyte membrane fraction with the plant glycoside digitonin but not by treatment with a wide variety of other detergents. The binding sites appeared to be soluble by several independent experimental criteria including (a) failure to sediment at 105,000 x g for 2 hours; (b) passage through 0.22-p Millipore filters; (c) chromatography on Sepharose 6B gels; and (d) electron microscopy. The soluble receptor sites retained all of the essential characteristics of the membrane-bound sites, namely rapid and reversible binding of p-adrenergic agonists and antagonists; strict stereospecificity toward both p-adrenergic agonists and antagonists; appropriate structure-activity relationships; saturability of the sites at low concentrations of ligand; no affinity for cu-adrenergic drugs, nonphysiologically active catechol com- pounds, and catecholamine metabolites. Based on gel chromatography in the presence of detergent, the molecular weight of the soluble receptor is estimated to be no greater than 130,000 to 150,000. Equilibrium binding studies indicated a K. for the soluble receptor of 2 nM. Hill coefficients (nH) of 0.77 and curved Scatchard plots suggested the presence of negatively cooperative interactions among the solubilized receptors in agreement with previous findings with the membrane-bound sites. Kinetic studies indicated an association rate constant k 1 = 3.8 x lo6 Mm1 min-’ and a reverse rate constant k, = 2.3 x 10m3 min- 1 at 4”. The kinetically derived KD (k,/k J of 0.6 nM is in reasonable agreement with that determined by equilibrium studies.

The soluble receptors were labile at temperature > 4” but could be stabilized with high concentrations of EDTA. Guanidine hydrochloride and urea produced concentration-dependent losses of binding activity which were partially reversible upon dialysis. Trypsin and phospholipase A both degraded the soluble receptors but a variety of other proteases and phospholipases as well as DNase and RNase were without effect. Experiments with group-specific reagents indicated that free lysine, tryptophan, serine, and sulfhydryl groups may be important for receptor binding. These studies suggest that the receptor is probably a protein which requires lipids for functional integrity. Data obtained with the solubilized binding sites are consistent with the contention that these sites represent the physiologically relevant @-adrenergic receptors which have been extracted from the membranes with full retention of their properties.

p-Adrenergic receptors are those components of certain catecholamine-sensitive cells which recognize and bind these drugs. Drug binding leads to generation of a signal which activates a biological process (1, 2). In many tissues the /J-adrenergic receptors are linked to the enzyme adenylate cyclase (3). This is the case in avian (4-7) and amphibian erythrocytes (8-11). Until recently, the fl-adrenergic receptors had not been directly identified. In the last year, however,

* This work was supported by Grant HL16037 from the Department of Health, Education, and Welfare and a grant-in-aid from the American Heart Association with funds contributed in part by The North Carolina Heart Association.

$ Fellow, Medical Research Council of Canada and the government of the province of Quebec.

5 Established Investigator of the American Heart Association.

several groups have demonstrated the feasibility of studying directly the P-adrenergic receptors with radioactively labeled P-adrenergic antagonists (12-17). We have used ( -)-[3H]dihy- droalprenolol, a potent competitive P-adrenergic antagonist, labeled to a specific activity of 17 to 33 Ci/mmol to identify binding sites in frog erythrocyte membranes which possess all of the essential characteristics to be expected of the physiologi- cal P-adrenergic receptors. The membrane-bound receptor sites bind fi-adrenergic agonists and antagonists rapidly and reversibly with affinity constants which are directly parallel to the known fi-adrenergic potency of these agents (13, 16, 18). The binding of /3-adrenergic agents moreover displays marked stereospecificity. There are approximately 1500 receptor sites per frog erythrocyte (16). The structure-activity relationships

2374

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

Soluble /3-Adrenergic Receptors 2375

which determine the binding affinity of p-adrenergic agents for these sites have been extensively examined (18).

As with other adenylate cyclase-coupled receptors, the fl-adrenergic receptors are membrane-bound. Thus, in order to further characterize the nature of these receptors it was necessary to develop procedures for their solubilization from membranes. In this paper, we report (a) the successful solubili- zation of the P-receptors (( -)- [3H]dihydroalprenolo1 binding sites) from a purified membrane fraction derived from frog erythrocytes by treatment with digitonin; (b) a convenient chromatographic assay method which permits study of these sites in the soluble state; (c) data which suggest the identity of the soluble sites with the membrane-bound P-adrenergic receptors; and (d) kinetic, specificity, and other properties of the /3-adrenergic binding sites in the solubilized state.

EXPERIMENTAL PROCEDURE

Materials

(-)- and (+)-Alprenolol hydrochloride were from Hassle. (~ )- and (+)-Propranolol hydrochloride were from Ayerst. The (+)-isomers of isoproterenol bitartrate, epinephrine bitartrate, and norepinephrine bitartrate were from Winthrop. The (-)-isomers of isoproterenol hydrochloride, epinephrine bitartrate, norepinephrine bitartrate, as well as dihydroxymandelic acid, (+)-dihydroxyphenylalanine, (+)-propranolol, cyclic AMP,’ ATP, phosphoenolpyruvate, myoki- nase, and digitonin (lots D43B-335, D61B-069, and 87B-3220) were from Sigma. Digitonin (lot 745499) was also obtained from Fisher; lysolecithin was from General Biochemicals. Pyruvate kinase was from Calbiochem. [3H]Cyc1ic AMP (1 to 5 Ci/mmol) and [a-ms2P]ATP (10 to 20 Ci/mmol) were from New England Nuclear. Alumina, neutral grade was from ICN and Dowex 5OW-X8 was from Bio-Rad. Sepharose 6B and Sephadex G-50 were from Pharmacia. Sources and specific activities of enzymes used were as fellows: phospholipase A (bee venom, 1180 units/mg); phospholipase D (cabbage, 22 units/mg); DNase (beef pancreas, 2100 kilounits/mg); RNase A (bovine pancreas, 100 kilounits/mg); trypsin (type II, 2000 units/mg), and pepsin (hog stomach, 4500 units/mg) were from Sigma. Thermolysin (6530 picou- nits/mg) was from Calbiochem. Southern grass frogs (Rana pipiens) were obtained from Nasco-St,einhilber.

(-)-Alprenolol was tritiated at New England Nuclear by catalytic reduction with tritium gas using palladium as the catalyst, to a specific activity of 17 to 33 Ci/mmol. (-)-Alprenolol was also reduced at New England Nuclear with hydrogen gas using identical procedures. Both materials were purified so that they were homogeneous in six different thin layer chromatography systems (18). As described elsewhere, in each of these systems the tritiated, hydrogenated, and native al- prenolol had identical R, values (18).

When chromatography is performed on silica gel plates impregnated with 5% silver nitrate (“argentation” chromatography), compounds differing by only a single unsaturated bond may be separated (50). When labeled material was chromatographed on such plates and compared with native (-)-alprenolol the following results were ob- tained: (a) acetone/benzene/acetic acid, 70/25/5; (~ i-alprenolol R,, 0.65; tritiated alprenolol and hydrogenated alprenolol R,, 0.83; (b) methanol/benzene/water, 15/2/3; (-)-alprenolol R,, 0.4; tritiated al- prenolol and hydrogenated alprenolol RF, 0.83.

These data indicate that tritiated alprenolol is equivalent to the hydrogenated alprenolol rather than native (~ )-alprenolol. The struc- ture of this material has now been determined by mass spectroscopy (Environmental Protection Agency, Durham, N. C.) and NMR spec- troscopy and is dihydroalprenol. Thus, “( -)- [‘H]alprenolol” is in fact (-)- [3H]dihydroalprenolo1, as had been anticipated.

When the tritiated material is chromatographed alone on the silver nitrate plates, and the plates are examined by ultraviolet light, only a single spot is seen corresponding to the RF of the radioactivity which in turn corresponds to the R, of dihydroalprenolol. Thus, no native unreacted alprenolol is found contaminating the tritiated material, which is pure dihydroalprenolol.

The abbreviations used are: cyclic AMP, adenosine 3’:5’-mono- phosphate; GuHCl, guanidine hydrochloride.

The fact that the specific activity of the labeled material is less than “theoretical” (i.e. -60 Ci/mmol) is presumably due to tritium ex- change with solvent during the labeling procedure which leads to reduction of some double bonds with hydrogen rather than tritium atoms. It is not due to the presence of unreduced double bonds.

We have tested the biological activity of (-)-dihydroalprenolol, (-)-alprenolol, and several lots of the radiolabeled material with specific radioactivities ranging from 10 to 33 Ci/mmol. When these materials were tested as antagonists of isoproterenol-stimulated aden- ylate cyclase by previously described procedures (16, 18) the K, of all of these materials was identical, -5 nM. Since the tritiated material contains no unreacted native (-)-alprenolol, these assays represent valid potency estimates of the tritiated compound. The results also indicate the biological equivalence of (-)-alprenolol and (-)-dihy- droalprenolol.

This equivalence was further tested by comparing the ability of (-)-dihydroalprenolol and (-)-alprenolol to compete with “( -)- [3H]dihydroalprenolol” for binding sites in frog erythrocytes. The two materials gave identical displacement curves, further indicating their identical biological activity.

Methods

Membrane Preparations-Heparinized blood from grass frogs main- tained at room temperature was collected by cardiac puncture and erythrocytes were washed three times with a solution of 110 mM NaCl/lO mM Tris-HCl, pH 7.4. The buffy coat was removed during this repeated washing procedure. Cells were lysed in water (1 ml of packed erythrocytes/lO ml of water) and homogenized with a glass-Teflon homogenizer; the lysate was adjusted to 5 mM Tris-HCl, pH 8.1/2 mM MgCl, and centrifuged at 30,000 x g for 15 min. A crude erythrocyte membrane preparation was obtained by washing this pellet twice and resuspending the material in 75 mM Tris-HCl, pH 8.1/25 rnM MgCl,. To obtain the purified membrane preparation, the pellet of the first centrifugation was resuspended in 10 mM Tris-HCl, pH 8.1/10 mM MgCl, by homogenization and centrifuged at 2,000 x g for 10 min over a cushion of the same buffer containing 50% sucrose. The material which sedimented through sucrose was discarded and the supernatant was centrifuged at 30,000 x g for 15 min. The pellet was washed once with 5 mM Tris-HCl, pH 8.1/2 rnM MgCl, for adenylate cyclase and (-)- [3H]dihydroalprenolo1 binding assays. For solubilization experi- ments the final pellet was resuspended in 50 mM Tris-HCl, pH 7.4, con- taining digitonin or other detergents. The membrane preparation procedure could be performed at pH 7.4 throughout without any change in the extent of binding or solubilization. Membranes were usually pre- pared fresh prior to each experiment. All procedllres were performed at O-4”.

Solubilization Procedures-Purified erythrocyte membranes (pre- pared as described above) were suspended in 50 mM Tris-HCl, pH 7.4, generally containing 1% digitonin (4.5 to 5.5 mg of membrane protein/3 ml of 1% digitonin buffer). The suspension was allowed to stand in ice for 30 to 40 min with periodic agitation and was then centrifuged in a Sorvall RCZ-B centrifuge at 30,000 x g for 30 min. The clear yellowish supernatant constituted the solubilized preparations which have been used throughout the studies reported here. This centrifugation ‘proce- dure was routinely used since more prolonged centrifugation at higher g forces did not sediment additional material (see “Results”).

Adenylate Cyclase Assays-Adenylate cyclase assays were per- formed as described previously (11) and cyclic [32P]AMP was isolated according to the method of Salomon et al. (19).

Binding Assays-( -)- [SH]Dihydroalprenolol binding in membrane fractions was assayed by a centrifugal assay as described previously (16, 18). ( ~)-[3H]Dihydroalprenolol, -25 no, was incubated with 0.3 to 0.4 mg of membrane protein for 10 min at 37” in a medium containing 50 mM Tris-HCl, pH 7.4, and 15 mM MgCl, in a volume of 150 ~1. At the completion of the incubations, duplicate 50.~1 aliquots were placed over 300 ~1 of the incubation buffer in small polyethylene centrifuge tubes and centrifuged for 1 min in a Beckman Microfuge 152. The membranes and bound (-)- [3H]dihydroalprenolo1 were pelleted almost immediately. The surface of the pellet was washed once with the same buffer, and the pellet was solubilized overnight by shaking with 0.5 ml of 10% sodium dodecyl sulfate and 10 mM EDTA and then counted in a liquid scintillation spectrometer after addition of a Triton X-100 toluene-based fluor. In all experiments the amount of (-)- [3H]dihydroalprenolo1 “nonspecifically” bound and/or trapped in the membrane pellets was determined by incubating membranes and (-)- [3H]dihydroalprenolo1 in the presence of 10 pM (A)-alprenolol or

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

2376 Soluble P-Adrenergic Receptors

propranolol which blocked all P-adrenergic receptor binding sites (16, 18). Although the choice of concentration of unlabeled ligand used to determine nonspecific binding is always somewhat arbitrary, we selected 10 PM propranolol for several reasons. Higher concentrations of ligand do not displace additional radioactivity. Using this concentra- tion of ligand, calculated K, values from binding experiments for numerous ligands are in excellent agreement with K, values deter- mined from adenylate cyclase experiments (18). “Nonspecific” binding was generally 10 to 15% of total binding and was subtracted from all experimental values. “Specific binding” in purified membrane prepa- rations ranged from 2 to 2.5 pmol of ( ~)-[SH]dihydroalprenolol bound/mg of protein, at saturating concentrations of radioligand.

This binding method has been validated by equilibrium dialysis. In three experiments, with (-)- [3H]dihydroalprenolol present at 1 x lOma M, specific binding as assessed by the centrifugation method was 1.30 + 0.2 pmol/mg of protein, and by equilibrium dialysis (18 hours, 4”) 1.37 * 0.14 pmol/mg of protein (not statistically significant). Thus, binding as assessed by this simple centrifugal assay is a true reflection of equilibrium binding. It should be noted that the half-time for dissociation of (-)- [3H]dihydroalprenolo1 from the receptors (-2 min at 37”; -5 min at room temperature at which centrifugation is performed) is sufficiently slow relative to the time needed to pellet the membranes (a few seconds) so that insignificant dissociation occurs during the procedure (18, 51).

(-)- [SH]Dihydroalprenolol binding to solubilized preparations was assayed by equilibrium dialysis or by a chromatographic procedure as described below.

Equilibrium Dialysis-Membrane preparations solubilized with digitonin (500 to 700 pg of protein) were dialyzed in the presence or absence of various drugs and hormones against digitonin solutions containing 50 mM Tris-HCl, pH 7.4, at 4”, and ( p)-[SH]dihydroal- prenolol at a concentration of 10 nM in a total volume of 2 ml. The dialysis cells were rotated for 16 to 20 hours at 4”. Quadruplicate 200-~1 samples were removed from both sides of the dialysis cells and counted in toluene-Triton X-100 scintillation fluid. The difference in radioac- tivity between the sample side and the medium side of cells was taken as “total” binding. “Specific” binding was determined in each experiment as the difference between total binding and the binding which occurred in the presence of 10 @M (+)-propranolol or unlabeled (-)-alprenolol (“nonspecific” binding). Nonspecific binding in the preparations obtained with 1% digitonin varied from 20 to 45% of the total binding.

Sephader G-50 Chromatogruphic Procedure-Aliquots (200 ~1) of solubilized preparations (50 mM Tris-HCl, pH 7.4, at 4’, 1% digitonin) were incubated at 4’ for 90 min with 30 to 40 nM (-)- [3H]dihydroal- prenolol in a total volume of 250 ~1. At the end of the incubation, samples were diluted to 500 pl with 50 mM Tris-HCl, pH 7.4, and chromatographed at once on Sephadex G-50 columns (0.6 x 12 cm) at 4”. Samples were loaded on top of the pre-equilibrated columns which were eluted with 50 mM Tris-HCl, pH 7.4, by gravity. Buffer, 0.5 ml, was applied to the column and the eluate was discarded. An additional 1.2 ml of buffer was applied and the eluate was collected directly in a scintillation vial. This 1.2.ml fraction contained the “void volume” of the column as determined with blue dextran 2000 as a marker and contained the “bound” (-)- [SH]dihydroalprenolol. At the completion of the chromatography the columns were washed with 20 ml of the same buffer and stored in buffer containing sodium azide (0.05%). The same columns can be reused for 10 to 15 runs. (&)-[3H]Dihydroal- prenolol specific binding was determined from the difference between total binding and the binding obtained in the presence of 10 PM (+)-propranolol or alprenolol (nonspecific binding). Specific binding in solubilized preparations as assessed by chromatography was greater

than 95% of the total binding. As shown under “Results” (Fig. 4), dissociation of bound (-)- [SH]dihydroalprenolol from the sites was negligible during the 15 to 20 min required to complete the chromatog- raphy.

Both methods of assessing binding of (&)-[SH]dihydroalprenolol gave very comparable results. Table I shows a comparison of the two methods. (~ )- [SH]Dihydroalprenolol binding assessed in solubilized preparations by equilibrium dialysis was slightly lower than that measured by the chromatographic procedure. This difference may be attributable to the relative instability of the solubilized binding sites during the prolonged incubations necessary for the equilibrium dialysis procedure. In addition to being a much more rapid method, the chromatography on Sephadex G-50 gave a higher percentage of “specific binding” (95 to 100%).

Protein-Protein determinations were carried out by the method of Lowry et al. (20) using bovine serum albumin as the standard. Estimations of samples containing digitonin were corrected for the presence of digitonin.

RESULTS

Preparation of Purified Erythrocyte Membranes-Lysates prepared from amphibian erythrocytes contain large amounts

of nuclear material as well as other elements. In order to facilitate solubilization of the (-)- [3H]dihydroalprenolo1 bind- ing sites, a purified membrane fraction was prepared. As shown

in Table II, centrifugation of a crude erythrocyte preparation over a 50% sucrose cushion removed a large portion of the protein while affording a considerable increase in the specific activity of the membrane-bound adenylate cyclase and the (-)- [3H]dihydroalprenolo1 binding activity. Basal adenylate

cyclase activity showed a 4.3-fold increase in specific activity

TABLE I

Comparison of equilibrium dialysis with chromatography on Sephader G-50 for assessment of (-)- [3H]dihydroalprenolo1 binding to

solubilized preparations

Binding of (-)- [SH]dihydroalprenolol to a solubilized preparation (0.64 mg/ml) was assessed by both procedures as described under “Methods.” Binding was measured in both cases in the presence of 14 nM (-)- [3H]dihydroalprenolo1. For the chromatographic method sam- ples were incubated at 4” for 90 min and chromatographed.immedi- ately on Sephadex G-50 whereas for equilibrium dialysis, aliquots of the preparation were dialyzed at, 4” for 20 hours before being sampled. Specific (-)- [SH]dihydroalprenolol binding was calculated, as de- scribed, by subtracting the binding observed in the presence of 10 MM (+)-propranolol (nonspecific binding) from the total binding. Results shown here are mean of triplicate determinations * S.E.

Method Specific binding

NonspecJfic binding

pmol ( )- [3H]dih,ydroalpren- % total 0101 boundl200~1 allquut binding

Equilibrium dialysis 0.210 * 0.045 44 i 4.7

Sephadex G-50 chroma- 0.233 i 0.005 5.5 * 0 tography

TABLE II

Purification of a membrane fraction from frog erythrocytes

The various fractions were prepared as described under “Methods.” (-)- [3H]Dihydroalprenolo1 binding was assayed by a centrifugation method. n = the number of preparations tested. All results were determined in duplicate. Values shown are mean * S.E.M. Numbers in parentheses indicate the percentage recovery of the adenylate cyclase and (-)- [3H]dihydroalprenolo1 binding activities.

Fractions n Protein Adenylate cyclase ( ~)-[3H]Dihydroalprenolol hound

mglml erythrocytes pmollmgll5 min pmollmg

Crude erythrocyte preparation 4 11.9 * 0.2 709 + 240 (100) 0.185 i 0.044 (100) 50% sucrose pellet 9 11.6 * 1.3 37.8 * 11.2 (5) 0.123 * 0.030 (64) Purified membrane 9 1.9 * 0.4 3070 * 660 (69) 1.76 i 0.47 (152)

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

Soluble fl-Adrenergic Receptors 2377

and retained catecholamine responsiveness. ( -)- [3H]Dihy- droalprenolol binding activity in the final membrane fraction was purified 9.5fold. (-)- [3H]Dihydroalprenolo1 binding sites in this purified fraction possessed the same characteristics (18,

21, 22) as the sites studied in unfractionated preparations (13, 16). The fact that recovery of binding activity is greater than 100% in the purified membrane fractions may be due to loss of some inhibitory factor during the preparation.

Solubilization of (-)- [3H]Dihydroalprenolol Binding Sites

-Treatment of the purified erythrocyte membrane frac- tions with the plant glycoside digitonin released the mem- brane-bound (-)- [3H]dihydroalprenolo1 binding sites in a

solubilized form to a degree dependent on the concentration of digitonin (Table III). Solubilization of erythrocyte membranes with 1% digitonin was used routinely throughout the studies reported here. The solubilization of the (-)- [3H]dihydroal- prenolol binding sites by digitonin appeared to be quite specific for this detergent since a number of other detergents were ineffective in solubilizing the binding sites in an active form. The detergents tested were Lubrol PX, 0.25, 1%; Lubrol WX, 1%; Triton X-100, X-305, N-101, CF.54 at 1%; sodium deoxy- cholate and octyl sodium sulfate at 0.1, 0.25, 0.5, and 1%. Lithium diiodosalicylate, lysolecithin, phospholipase A, and EDTA were also tested and found to be ineffective in releasing the binding sites in an active form. These data do not rule out the possibility that certain detergents did solubilize the sites but also interfered with the binding assay.

To verify the soluble state of the (-)- [3H]dihydroalprenolo1

binding sites released by treatment with digitonin, we sub- jected solubilized preparations to a number of experimental tests, namely, ultracentrifugation, Millipore filtration, gel filtration, and electron microscopy. When 1% digitonin- solubilized preparations obtained by centrifugation at 30,000 x

g for 30 min were subjected to further ultracentrifugation at 105,000 x g for 2 hours, no sedimented material was observable and (-)- [3H]dihydroalprenolol binding remained unchanged. In addition, after passage of the preparations through 0.45-p and 0.22.~ filters no decrease in binding was observed.

When solubilized preparations were fractioned by molecular sieving on Sepharose 6B, the (-)- [3H]dihydroalprenolol bind- ing activity was well included in the gel with a K,” of about 0.6 (Fig. 5). In addition, electron microscopy indicated the total absence of any recognizable membrane structure in the solubi- lized preparations.

After treatment of membranes with digitonin, adenylate

cyclase activity is present in the solubilized preparations at a specific activity comparable to that observed in the membrane fraction (membrane: 280 pmol/min/mg; soluble: 245 pmol/

min/mg, comparison of four experiments). The solubilized enzyme, however, no longer responds to catecholamine stimu- lation (data not shown).

The (-)- [3H]dihydroalprenolol binding activity present in

solubilized preparations was stable to storage at -93” for at least 1 month. Binding activity assessed either by equilibrium dialysis or chromatographic techniques was linearly related to the amount of protein over the range tested (0.05 to 1 mg/ml).

Stereospecific Binding Interaction of Adrenergic Agents and Other Agents with the Solubilized Sites-Fig. 1 demonstrates the ability of several adrenergic antagonists to compete with (-)- [3H]dihydroalprenolol for binding to solubilized sites. As in the intact membrane (13, 16), the unlabeled p-adrenergic

antagonists (-)-alprenolol and (-)-propranolol appear to be equipotent in inhibiting the binding of (-)- [3H]dihydroal- prenolol. The data also indicate that the solubilized sites retain their stereospecificity in binding /3-adrenergic antagonists. The (-)-isomers of alprenolol and propranolol are about 2 orders of magnitude more potent than the respective (+)-isomers. The racemic compound (h)-dichlorisoproterenol, as expected, com- peted for the binding sites with a lesser potency than either (- )-alprenolol or (-)-propranolol. The solubilized binding sites show no affinity for the cu-adrenergic antagonist phentola- mine. The dissociation constants (K,) calculated from compe-

tition experiments (Fig. 1) (23) correlate very well with those previously delineated for the binding sites in intact frog erythrocyte membranes (18) (Table IV) even though the two sets of experiments were performed at quite different tempera- tures (4” soluble, 37” particulate). Table IV lists the calculated dissociation constants (K,) of several antagonists for the

solubilized receptor sites in comparison to those obtained for the membrane-bound binding sites.

Fig. 2 demonstrates the ability of various adrenergic agonists to compete with (-)- [3H]dihydroalprenolo1 for binding, to the solubilized sites. These data indicate that strict stereospecific- ity is also apparent in the binding interaction of fi-adrenergic agonists with these sites. The (-)-isomers of isoproterenol, epinephrine and norepinephrine, are considerably more potent than the (+)-isomers of these compounds. Table V summarizes the stereoisomeric potency ratios for several agonists and antagonists obtained in the solubilized preparations and com- pares them with those obtained with the membrane-bound

TABLE III

Solubilization of( -)- [3H]dihydroalprenolo1 binding sites from purified frog erythrocyte membranes

The purified membrane fraction was prepared as described under “Methods.” Membrane protein, 4.5 to 5.5 mg, was suspended in 3 ml of 50 mM Tris-HCl buffer, pH 7.4, containing the indicated concentrations of digitonin. (-)- [3H]Dihydroalprenolo1 binding in solubilized prep- arations was assessed by equilibrium dialysis. The results shown are the mean of quadruplicate determinations from three experiments f S.E.

( )- [3H]Dihydn,alprenolol bound Preparation Protein

Yield -

i.kdml plWJlld pmollmg “‘c binding sites soiubilized

Purified membrane fraction 1700 + 184 2.47 i 0.37 1.45 * 0.21

Solubilized preparations 0.1% digitonin 105 * 4.9 0.029 i 0.028 0.276 i 0.275 1.8 L 1.5 0.25% digitonin 325 zt 37.8 0.778 + 0.420 2.39 * 1:39 29.9 * 12.9 0.5% digitonin 561 i 39.8 2.02 * 0.20 3.60 i 0.36 87.8 i 22.0 1.0% digitonin 846 + 66.3 2.09 * 0.20 2.47 + 0.24 89.6 i 19.6

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

2378 Soluble /3-Adrenergic Receptors

IOO-

90 - + (-)Alprcnolol

-z e(+) II

.2 *(-)Propranolol

,= so- *(+) ”

5 +(*)MJ-7963-I

f it (*)Dichlor-

Phentolomine

c H (-) lsoproterenol

:; ao- e*(+) $1

e H(-)Eptnephrlne

AZ A-r(+) 41

F 7D - c-c (-)Noreplnephrlne 0\9

B-x(+) (1

P

m t

a 0 60

Dlhydroxy-

5 mandelbc Acid /

0 (+)DOPA

/

0 50- x

5

h 40-

:: G r” 30- .- n

7 r& 20-

J

0

’ [A~+ogoni~~] M-L;,0

FIG. 1. Competition of adrenergic antagonists for the binding of (- )- [3H]dihydroalprenolo1 to solubilized erythrocyte membrane prep- arations. (-)- [3H]Dihydroalprenolo1 (10 IIM) binding was determined by equilibrium dialysis and gel chromatography in the presence and absence of the indicated concentrations of the various antagonists as described under “Methods.” One hundred per cent or control binding was 2.46 + 0.18 pmol/mg. The results shown are the mean of quadruplicate determinations from two experiments.

[Agonist](M)-Log10

FIG. 2. Competition of P-adrenergic agonists and other agents for the binding of (- )- [3H]dihydroalprenolo1 to solubilized erythrocyte membrane preparations. (-)- [3H]DihydroalprenoloI binding was deter- mined by equilibrium dialysis and gel chromatography in the presence and absence of various concentrations of these agents. (-)- [3H]Dihy- droalprenolol was present in the assays at 10 no, One hundred per cent or control binding was 1.42 + 0.04 pmol/mg. Results shown are the mean of quadruplicate determinations from two experiments.

TABLE IV that ( P)-isoproterenol is 2 to 3 orders of magnitude more

Dissociation constants (K,) ofpotent /%xhmrgic antagonists for the potent than (- )-norepinephrine in competing for the binding

solubilized ( -)-[3H]dihydroaIprenolol binding sites sites. In general, the affinities of agonists (( -)- and (+)-

Values for K, were calculated according to Cheng and Prusoff (23) isomers) for the solubilized receptor sites correlate reasonably

for competition experiments shown in Fig. 1. Data for the membrane- with the affinity of the binding sites for the same compounds in

bound receptors were taken from Mukherjee et al. (18). membrane preparations (Table VI). It is also apparent from the data shown that all of the

structure-activity relations which characterize the membrane- bound P-adrenergic receptors have been preserved in the solubilized preparations. In addition to the (- )-stereoisomeric configuration of the p-carbon hydroxyl the size of the substitu- ent on the amino nitrogen is an important determinant of affinity for the receptor. (*)-C-34, which has an aromatic substituent on the amino nitrogen, is several-fold more potent than isoproterenol in inhibiting (-)- [3H]dihydroalprenolol

binding. The same observation can be made by comparing (h)-soterenol and MJ796331. ( P)-Isoproterenol and (h)- dichlorisoproterenol had very comparable affinity. As noted elsewhere (18) the ring substituents are more important in determining the intrinsic activity of adrenergic agents for stimulation of adenylate cyclase than the affinity for the receptors. The solubilized (-)- [3H]dihydroalprenolo1 binding sites did not interact with the physiologically inert catechol compounds (5).dihydroxyphenylalanine and dihydroxyman- delic acid, even at concentrations as high as 100 FM.

receptor sites (16, 18) and those from intact tissues obtained from the literature (24). A reasonable correlation exists be- tween the three sets of data.

The potency series for the various agonists in inhibiting binding of (~ )- [3H]dihydroalprenolol was (-)-isoproterenol > (-)-epinephrine > (-)-norepinephrine, which is typical of the P-adrenergic receptor. It is also apparent from these data that the solubilized receptor sites retain the characteristics of a Pz-adrenergic receptor (25). This is demonstrated by the fact

Equilibrium Studies of (-)- [3H]Dihydroalprenolol Binding in Soluble Preparations-Binding of (-)- [3H]dihydroal-

prenolol to the solubilized sites was a saturable process (Fig. 3). (Nonspecific binding was only 1 to 2% of the total binding and

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

Soluble p-Adrenergic Receptors 2379

TABLE V

Comparison of the stereoisomeric potency ratios of (-)- and (+)-P-adrenergic agonists and antagonists for P-adrenergic receptors

The values for the physiological responses in intact tissues are from the data of Figs. 1 and 2. Value for propranolol marked with an taken from the literature (24). Values for stereoisomeric potency ratios asterisk (*) refers to antagonism of isoproterenol-induced tachycardia in frog erythrocyte membrane fractions are taken from our previous in the cat. ~ indicates that (+)-isomer of norepinephrine is too weak to work (18). Values for the solubilized preparations were determined calculate a valid potency ratio.

Potency ratio = [(~)-isomery[(+)-isomer]:

Compounds

Alprenolol Propranolol Isoproterenol Epinephrine Norepinephrine

Intact tissues: guinea pig atrium

calf tracheal muscle (24)

4om50* 300-500

40-50 20

Frog erythrocyte

Membrane fractions Soluhilized

Adenylate (~ )- [3H]D~~d~~~alpren~lol ( ~)-(3H]Dihydroalprenolol C>dW L 1 binding

30 44 55 45 60 146

700-2300 450 325 53 30 8

4 3

TABLE VI

Dissociation constants (K,) of fi-adrenergic agonists for binding to solubilized receptor sites

K, values were calculated (23) from data shown in Fig. 2. Data for the membrane-bound receptors were taken from Mukherjee et al. (18).

(-)[3H]D,hydroalprenO101 BIndIn

(-) Noreplnephrlne OH OH H 49’2

(+) Noreplnephrlne OH OH H 200

c-1 Eplnephrlne OH OH CH3 46502

(+) Eptnephrlne OH OH CH3 137t4

C-J ,soproterenol

(+) lsopro+erenol

(Z) Solerenol

(‘) cc34

OH OH -HC p3

\ 04ot-0005

CH3

OH OH -HC /w \

1.33:06 CH3

OH CH3S02NH -CH p3

2 4’046 \ CH3

OH OH

132!28

37’07

28’14

021’005

68:22

0084~00l

OOlZ*-0001

approximated blank values.) In five experiments a saturation

value of 2.0 & 0.3 pmol/mg was obtained. Half-maximal saturation, which provides an estimate of the equilibrium dissociation constant (K,) of (-)- [sH]dihydroalprenolol for the soluble binding sites, was 2.2 f 0.2 nM (n = 5). Data obtained by equilibrium binding were analyzed graphically by Hill plots (Fig. 3) and Scatchard plots (Fig. 3). From the Hill plot an

estimation of the K, for ( -)- [3H]dihydroalprenolol can also be obtained (intercept with the abscissa at (BIB,,, - B) = 1. This value was 2.6 + 0.28 nM (n = 5). The slope of the Hill plot in the region corresponding to the K, of the ligand, i.e. Hill

coefficient, was n2H = 0.77 & 0.05 with a linear regression correlation coefficient, r = 0.99 * 0.006 (n = 5). Hill coeffi- cients were significantly lower than unity. These data are consistent with a retention in the solubilized preparations of

negatively cooperative interactions among the binding sites documented to exist in frog erythrocyte membrane prepara- tions (26). Fig. 3 also shows a Scatchard plot for ( p)-[3H]dihy- droalprenolol binding to solubilized sites. This shows a slight

curvature with an upward concavity, consistent with the presence of negative cooperativity (27, 28) or more than one order of sites.

Kinetic Studies of (-)- [3H]Dihydroalprenolol Binding in Soluble Preparations-Binding of (-)- [3H]dihydroalprenolo1 to the solubilized sites was time-dependent. With (-)- [3H]dihydroalprenolol present at 37.4 nM specific binding at 4” reached equilibrium between 40 and 90 min (Fig. 4A). The

calculated second order association rate constant determined from the time course of binding of (-)- [3H]dihydroalprenolol was k, = 3.8 x lo6 Mm’ min-’ (27). Fig. 4B shows the time course of dissociation of (-)- [3H]dihydroalprenolol from the soluble sites at 4”. The dissociation of ( p)-[3H]dihydroal- prenolol is relatively slow at 4”, tK about 5 to 6 hours. As pointed out earlier, this permits the Sephadex G-50 chromato- graphic method of determining specific binding .of (-)-

[3H]dihydroalprenolo1 under equilibrium conditions. From the slope of the line in Fig. 4B a dissociation rate constant (k,) of 2.3 x 10m3 min-’ can be calculated. An estimate of the dis- sociation constant (K, = k,/K,) of 0.6 nM can be obtained from the kinetic rate constants. This value is in reasonable agreement with the estimate of K,] obtained by other methods of analysis (see above).

It seemed of interest to compare the kinetic characteristics of the soluble and particulate receptors. Lability of the soluble

receptors at 37” precluded analysis at this temperature. When kinetic studies were performed at 4” with the membrane receptors, these results were obtained: k, = 6.6 x lo6 Mm’ min-‘, k, = 6.6 x 10e3 min-‘, K, = (k,/k,) = 1 nM.

Thus, at 4” the kinetics of the soluble and particulate receptors are fairly comparable.

Gel Filtration of Solubilized Receptors-Solubilized prepa- rations containing 1% digitonin were chromatographed on a Sepharose 6B column which was equilibrated with 50 mM Tris-HCl, pH 7.4, at 4”. The elution profile of a typical run is shown in Fig. 5. The (-)- [3H]dihydroalprenolo1 binding activ-

ity was eluted as a major peak with a K,, of 0.58 to 0.60. A small peak of binding activity also appeared to be present in the void volume. The elution pattern of the solubilized receptor activity was the same whether the preparations were chromato- graphed on columns equilibrated with a buffer containing 0.1%

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

2380 Soluble @-Adrenergic Receptors

09 r 08

04-

OS-

02-

01 -

[FREE DI HYDROALPRENOLOL] ntd

FIG. 3. Binding of (-)- [3H]dihydroalprenolo1 to solubilized erythro- cyte preparations as a function of increasing concentration of ligand. Solubilized preparations (515 to 705 pg/ml) were incubated at 4” with various concentrations of (-)- (SH]dihydroalprenolol (33 Ci/mmol) (20 to 200 nM) for 180 min. Bound and free (-)- [3H]dihydroalprenolo1 were separated by Sephadex G-50 column chromatography as described under “Methods.” Specific (-)- [3H]dihydroalprenolo1 binding was calculated from the difference between total binding and binding in the presence of 10 pM (*) propranolol (nonspecific binding). Satura- tion binding was 2.0 + 0.3 pmol/mg (n = 5). Results shown are mean of duplicate determinations. This experiment was replicated five times. Upper inset, Hill plot of ( -)-[‘H]dihydroalprenolol binding to solubi-

digitonin or no detergent. It should be underscored that even when the columns are eluted with buffers which do not contain digitonin, digitonin remains bound to the soluble protein and

can be detected in the peak fractions. This protein-bound detergent is apparently sufficient to maintain the receptors in a soluble state. Adenylate cyclase activity in these fractions could not be readily measured, presumably due to the lability

of this enzyme activity during the time required for chromatog- raphy. Proteins with known molecular weights were also fractionated on the same Sepharose 6B column. The arrows in Fig. 5 indicate the position of their respective elution on Sepharose 6B. The inset in Fig. 5 is a plot of the relative position of elution of each protein as a function of its molecular weight (29). From this relation, an apparent molecular weight of 130,000 to 150,000 can be obtained for the soluble (-)-

[3H]dihydroalprenolol binding sites. It should be noted that this estimate represents a maximum value for the molecular weight of the receptor sites because the presence of detergent (digitonin) can markedly increase the apparent molecular weight of such detergent-solubilized proteins (30-32).

Effect of Temperature on Soluble (-)- [3H]Dihydroal- prenolol Binding Sites-Treatment of the erythrocyte mem- brane fractions with digitonin at temperatures of 25”, 30”, and 37” instead of O-4” afforded progressively less soluble binding activity. Similarly, incubation of solubilized preparations (obtained at 4”) at these temperatures for short periods >;oUuces loss of activity. EDTA, however, at high concentra- tions (5 to 50 mM) seemed to protect the binding activity. The high concentrations necessary for the effect suggest that some mechanism other than chelation of heavy metals may be involved.

Despite this relative instability of the soluble binding sites to

temperatures in the physiological range, the binding activity of the solubilized preparations is fairly stable at O-4” (half-life 2

lized receptor sites. B indicates the amount of specific (-)- [3H]dihy- droalprenolol binding at various concentrations of “free dihydroal- prenolol.” B,,, represents the amount of dihydroalprenolol specifically bound to sites at saturation (Fig. 3). Data shown are the means of duplicate determinations. The experiment shown is typical of five such experiments. The values of the slopes (nH = 0.77 * 0.05, n = 5) and the correlation coefficient r = 0.98 * 0.006 (n = 5) were calculated by linear regression analysis. Lower inset, Scatchard plot of (~ )- PH]dihy- droalprenolol binding to solubilized erythrocyte membrane prepara- tions. Data shown are means of duplicate determinations and this plot is typical of five such experiments.

to 3 days). When kept frozen, solubilized preparations retain their full binding activity for extended periods of time (months).

Effects of Denaturants on Binding Activity of Solubilized Preparations-Treatment of solubilized preparations with urea and guanidine hydrochloride produced a concentration- dependent loss of binding activity. The effects occurred ab- ruptly at concentrations greater than 0.5 M guanidine hydro- chloride and 2 M urea, suggesting that protein denaturation is likely the cause of loss of binding. Concentrations of 2 M GuHCl and 6 M urea were sufficient to abolish binding completely. Removal of the denaturing agents (2 M GuHC1/5 M urea) by dialysis for 6 to 7 hours partially restored the binding activity (10 to 15% of the total binding was restored above binding determined in the presence of denaturants).

Effects of Enzymic Treatments and Group-specific Reagents on Solubilized (-)- [3H]Dihydroalprenolol Binding Sites-As shown in Table VII, digestion of solubilized preparations with phospholipase A and trypsin at high concentrations decreased the (-)- [3H]dihydroalprenolol binding activity of these prepa- rations. The combined effect of trypsin and phospholipase A on the ( -)-[3H]dihydroalprenolol binding was more than either enzyme alone. Other phospholipases and proteases even at high concentrations did not affect the binding of (-)- [3H]dihy- droalprenolol. Lysolecithin, the product of phospholipase A digestion of lipids, showed no effect on binding. DNase and

RNase were also without effect. The fact that high concentra- tions of enzymes were required for the effects may be due to the detergent which is likely bound to the receptor sites. The (-)- [3H]dihydroalprenolo1 binding sites present in intact membrane preparations show the same susceptibility to phos- pholipases and proteases but at lower enzyme concentrations

(33). The reagent 2,4,6-trinitrobenzenesulfonic acid, which is

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

Soluble /3-Adrenergic Receptors 2381

50;~ ,p, ?;;:’ , ~

0 IO 20 30 40 50 60 70 60 90 Timefmin)

FIG. 4. A, time course of’ ( ~)-[3H]dihydroalprenolol binding to solubilized preparations of erythrocyte membranes. A solubilized preparation (6.0 ml, 50 rn~ Tris-HCl, pH 7.4/l% digitonin) was incubated at 4’ with 37.4 no (-)-[‘H]dihydroalprenolol in a total volume of 7.5 ml. The reaction was started by addition of the labeled l&and. At the indicated times 250.~1 aliquots were sampled and immediately added to 250 ~1 of 20 PM (_t)-propranolol to prevent further (-)- [SH]dihydroalprenolol binding, during the time required for the separation of bound and free (-)- [3H]dihydroalprenolo1 on Sephadex G-50 columns. Samples were then chromatographed on Sephadex G-50 columns as described under “Methods.” As shown subsequently (B), addition of 10 PM propranolol is followed by no significant dissociation of labeled ligand during the 15 to 20 min necessary for the chromatographic procedure but it effectively stopped the binding reaction (see nonspecific binding). Specific binding was determined as described under “Methods.” Nonspecific binding (10 PM

specific for the t-amino group of lysine residues of proteins (34), decreased the binding markedly. Treatment of solubilized

preparations with a reagent (2-hydroxy-5-nitrobenzylbromide) directed primarily at tryptophan residues (35-37) decreased binding of (-)- [3H]dihydroalprenolo1. These reagents can also react with sulfhydryl groups and their effects therefore cannot unequivocally be ascribed to specific modification of trypto- phan residues (36). Phenylmethylsulfonylfluoride, which has been found to modify reactive serine residues on proteolytic enzymes (38-40), decreased binding of (-)- [3H]dihydroal- prenolol to the soluble sites. p-Chloromercuribenzoate and p-hydroxymercuribenzoate, which modify sulfhydryl groups (41), drastically reduced binding. These results are summa- rized in Table VII. l-Ethyl-3-(3-dimethylaminopropyl)car-

bodiimide, a reagent which can react with free carboxyl groups (42), and n-acetylimidazole, which reacts with the phenolic moiety of tyrosine residues (43), had no or minimal effects on binding.

DISCUSSION

Solubilization of the P-adrenergic receptors (( -)- [3H Jdihy-

droalprenolol binding sites) from a purified membrane fraction of frog erythrocytes was achieved by treatment with digitonin. Although digitonin is not a widely used detergent for the solubilization of membrane proteins, it was used for this purpose long ago in studies with rhodopsin (44) and more recently with the muscarinic cholinergic receptor (45). The

Time (hr)

(+)-propranolol) approximated blank values which were determined by incubation of buffer (50 rnM Tris-HCl, pH 7.4, containing 1% digitonin) with (-)- [3H]dihydroalprenolo1 followed by gel chromatog- raphy. Data shown are means of duplicate determinations, and brackets represent the range. B, dissociation of (~ )- [3H]dihydroal- prenolol from solubilized preparations of frog erythrocyte membranes. A solubilized preparation was equilibrated at 4’ with (~ )- [3H]dihy- droalprenolol (35 IIM) for 90 to 120 min. At time zero, (i)-propranolol was added to the incubation mixture to a final concentration of 10 PM. At the indicated times, 250.~1 aliquots were removed and chromato- graphed on Sephadex G-50 to determine bound (~ )- [3H]dihydroal- prenolol as described under “Methods.” Initial binding refers to the amount of (-)- [3H]dihydroalprenolo1 bound at equilibrium prior to the addition of (+)-propranolol. The slope of the line provides an estimate of the dissociation rate constant k, at 4”. Results shown are from duplicate determinations, and brackets represent the range.

truly soluble state of the digitonin-extracted receptor prepara-

tions was defined by the usual experimental criteria of gel filtration, electron microscopy, ultrafiltration, and ultracen- trifugation. A variety of other detergents were quite ineffective in solubilizing the receptors in an active form.

Of the two methods used to assay ( -)-[3H]dihydroalprenolol binding to the soluble receptors, equilibrium dialysis and gel filtration, the filtration method proved more satisfactory for several reasons. Thus, equilibrium dialysis routinely gave nonspecific binding values which ranged from 20 to 45%. By comparison, nonspecific binding with the gel filtration method was 0 to 5% of total binding and approximated the blank. Several other advantages of the gel filtration assay are that it is rapid; it is highly reproducible and variability between dupli-

ca,es is usually <5%; it eliminates possibly important loss of receptor activity due to prolonged incubations necessary for equilibrium dialysis; and more importantly, as shown in Fig. 4 it gives a close estimate of the true equilibrium binding. It is worth pointing out here that the level of nonspecific binding observed in a preparation may not be a reflection of properties of the preparation but rather, as we have shown here, it may be a function of the method of assessment of binding used.

The ability of P-adrenergic agonists to compete for the binding of (~ )- [3]dihydroalprenolol to solubilized sites was typical of a P-adrenergic receptor, i.e. isoproterenol > epineph- rine >> norepinephrine. fi-Adrenergic antagonists also com- peted for the binding sites with appropriate affinities (Table

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

2382 Soluble /3-Adrenergic Receptors

05.

Thyroglobulln Apoferrltln ,(466.000) &6rjd%,';~7~~;ferr'" I

d * A ' ' > > ' ' ' . b-----+-r-\

k

0 0 5 30 35 40 45 50 55 60 65 70 75 80 85 90 95 -loo- 105 110 115 ;20 I25

VOLUMEtml)

i

0.

i

0.

JO

FIG. 5. Sepharose 6B chromatography of solubilized (~ )- 13H]dihy- droalprenolol binding protein. A solubilized preparation (2.0 ml) concentrated to 4.2 mg/ml by ultrafiltration (Amicon PM-IO) was chromatographed on a Sepharose 6B column (1.5 x 62 cm) at a constant flow rate of 10 ml/hour. The column was equilibrated with 40 rn~ Tris-HCl, pH 7.4, at 4”. Fractions of 2 ml were collected. Specific (&)- [3H]dihydroalprenolo1 binding (0- - -0) was assayed (35 nM (-)- [3H]dihydroalprenolol) on the various fractions as described under “Methods” for the Sephadex G-50 chromatographic assay. The protein elution profile (04) was determined spectrophotometrically

280nm). Arrows indicate the position of the peak protein elution 280n,,,) when various marker proteins (5 to 15 mg/2 ml) were

chromatographed on the same column under identical conditions. The results shown are representative of three experiments. Inset, estima- tion of the apparent molecular weight of solubilized ( -)-[3H]dihy- droalprenolol binding sites by gel filtration on Sepharose 6B (29). The curve was constructed by plotting the known molecular weights of the various marker proteins uersua their relative elution volume (K,,) on Sepharose 6B.

TABLE VII

Effect of enzymic treatments and protein-modifying reagents on soluble (-)- [3H]dihydroalprenolo1 binding sites

Solubilized preparations were treated with the various enzymes for ing was determined chromatographically on Sephadex G-50. The mean 10 min at 25’. After the treatment, samples were immediately cooled to control (-)- [W]dihydroalprenolol binding in these experiments was 4” and assayed for (-)- [3H]dihydroalprenolo1 binding. Aliquots (200 1.78 * 0.04 pmol/mg (n = 5). All reagents were dissolved in water ~1) of soluble preparations (50 rn~ Tris-HCl, pH 7.4, at 4’/1% and the pH was adjusted at 7.4 except for p-chloromercuribenzoate, digitonin) were incubated in the presence of the various reagents which was dissolved in 1 M glycylglycine, pH 7.6 n represents the (200 ~1) at 23” for 20 min. At the end of the incubation, binding number of experiments performed in duplicate for each reagent or was assayed at 4’ for 90 min in a total volume of 500 ~1 in the presence enzyme. Concentrations of enzymes are given in micrograms per ml; of 23 nM (-)- [3H]dihydroalprenolo1. ( -)- [3H]dihydroalprenolo1 bind- concentration of reagents are millimolar.

Treatment Concentration R ( ~)- [3H]Dihydroalprenolol specific binding

Enzyme Phospholipase A Phospholipase A Trypsin Trypsin Trypsin plus Phospholipase A

Reagent 2,4,6Trinitrobenzenesulfonic acid 2.Hydroxy-5nitrobenzylbromide Phenylmethylsulfonylfluoride p-Chloromercuribenzoate p-Hydroxymercuribenzoate

10 1.0

500 100 500

10

10

5 5 5 5

2 7.5 -c 0.24

58.0 + 6.6 95.1 * 4.9 31.2 + 2.6 52.2 i 2.0

15.4 * 7.8 33.0 A 2.7 43.7 * 6.8 21.8 + 10

12.8 + 1.0

IV). A striking property of the soluble receptor sites was their conserved. Thus, a (-)-configuration of the p-carbon hydroxyl

retention of the strict stereospecificity of binding. As shown in was essential for high affinity and increasing size of the

Table V the stereoisomeric potency ratios of (-)- and (+)-iso- substituent on the amino nitrogen markedly increased affinity

mers of P-adrenergic agents in solubilized preparations cor- of adrenergic compounds. Solubilization of certain membrane-

related well with those determined in membranes (16, 18) and bound receptors has been occasionally complicated by changes

in intact tissues (24). The principal structure-activity features in specificity and affinity (45-47).

of the binding of adrenergic agents to the sites were also As shown in Fig. 3 (-)- [3H]dihydroalprenolol binding in

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

Soluble p-Adrenergic Receptors 2383

solubilized preparations was saturable. Analysis of these equi- librium data yielded estimates of the dissociation constant (K,) for (-)- [3H]dihydroalprenolol binding (2.6 * 0.28 nM by Hill plot, 2.2 * 0.2 nM by saturation analysis) which are in good agreement with the value 4.6 nM derived from competition experiments with unlabeled alprenolol (Fig. 1 and Table IV). Scatchard (curvilinear upward) and Hill plot (nH < 1) analysis of equilibrium data both suggested that site-site interactions among fl-adrenergic receptors may have been retained in solubilized preparations as they exist in intact membranes (26). The retention of negatively cooperative interactions

among receptor sites after solubilization has been demon- strated recently for the binding of thyrotropin-stimulating

hormone to its receptor in bovine thyroid (48). Hill coefficients less than one are also consistent with the

existence of multiple binding sites with differing affinities. Steady state binding data cannot distinguish between the presence of heterogeneous sites, negatively cooperative interac- tions, or the simultaneous presence of both. Therefore, direct kinetic techniques similar to those applied in particulate preparations (26, 28, 49) will need to be designed and applied to confidently assess the presence of negative cooperativity among the solubilized receptor sites.

Kinetic analysis of (- )- [3H]dihydroalprenolol binding to

solubilized sites determined at 4” provided estimates of the rate constants of association and dissociation (k, = 3.8 x lo6 Mm’mind’, k, = 2.3 x 10m3 min-I). The K, calculated from rate data was 0.6 nM, which is in reasonable agreement with K. values determined by other methods.

It should be noted that throughout this manuscript we have

referred to estimates of the equilibrium dissociation constant (K,) of (- )- [3H]dihydroalprenolol for the soluble fi-adrenergic

receptor binding sites. Since it is apparent that negatively cooperative site-site interactions may exist in these solubilized preparations, these K, values might more appropriately be described as apparent or average K, values.

Chromatography of solubilized preparations on Sepharose 6B showed that the (-)- [3H]dihydroalprenolol binding activity was well included in the gel (K,, -0.58 to 0.6), further verifying the soluble nature of the receptors. The estimates of molecular weight (130,000 to 150,000) were obtained in the presence of detergent and therefore should be regarded as a highest estimate of the molecular weight since detergents bound to the membrane protein might increase the estimate of molecular weight (30-32).

The effect of denaturants and enzymic treatments on the binding activity of solubilized preparations suggests that the

receptor sites are protein in nature. In all experiments where perturbation of binding activity was studied, whether by reagents, denaturants, or enzymes, the effects on specific binding observed were reflected always by a decrease in total binding with no change in nonspecific binding. With denatu- rants (urea and GuHCl) typical sharp denaturation curves

were obtained, although denaturation was not assessed by other means than loss of (-)- [3H]dihydroalprenolo1 binding activity. Enzymic treatment studies showed that high concen-

trations of enzymes were required to affect binding activity. As suggested, this relative insensitivity might be explained by a coating of the binding sites with the detergent. The fact that phospholipase A in addition to trypsin affected the binding suggests that lipids may be functionally required for optimal binding activity.

We have demonstrated in this paper that solubilization by digitonin treatment of the P-adrenergic receptor binding sites present in a purified membrane fraction of frog erythrocytes

can be affected with maintenance of all of the essential characteristics expected of the P-adrenergic receptors. Such soluble preparations should be of great value in studies aimed at the ultimate purification of the P-adrenergic receptors as

well as for reconstitution experiments directed at elucidating the relationship of the receptors to adenylate cyclase.

Acknowleclgments-The electron microscopic studies in con- sultation were performed by Dr. J. R. Sommer, Director, Veterans Administration EM Laboratory, Veterans Adminis- tration Hospital, Durham, N.C. We are grateful to Dr. P. Jeffs,

Department of Chemistry, Duke University, for assistance in interpretation of NMR and mass spectroscopic studies and to the Analytical Chemistry Department of New England Nu- clear Co. for assistance in development of chromatographic procedures for alprenolol and dihydroalprenolol.

REFERENCES

1. Rob&n, G. A., Butcher, R. W., and Sutherland, E. W. (1971) Cyclic AMP, Academic Press, New York

2. Roth, J. (1973) Metabolism 22,1059-1073 3. Kahn, C. R. (1975) in Methods inMembrane Biology (Kern, E. D.,

ed) Vol. 3, pp. 177-242, Plenum Press, New York 4. Davoren, P. R., and Sutherland, E. W. (1963) J. Biol. Chem. 238,

3009-3015 5. @ye, I., and Sutherland, E. W. (1966) Biochim. Biophys. Acta 127,

347-354 6. Schramm, M., Feinstein, H., Naim, E., Long, M., and Lasser, M.

(1972) Proc. Natl. Acad. Sci. U. S. A. 69, 523-527 7. Bilezikian, J. P., and Aurbach, G. D. (1973) J. Biol. Chem. 248,

5577-5583 8. Rosen, 0. M., Erlichman, J., and Rosen, S. M. (1970) Mol.

Pharmacol. 6, 524-531 9. Grunfeld, C., Grollman, A. P., and Rosen, 0. M. (1974) Mol.

Pharmacol. 10, 6055614 10. Caron, M. G., and Lefkowitz, R. J. (1974) Nature 249, 258-260 11. Lefkowitz, R. J. (1974) J. Biol. Chem. 249, 6119-6124 12. Levitzki, A., Atlas, D., and Steer, M. L. (1974) Proc. N&l. Acad.

Sci. U. S. A. 71,2773-2776 13. Lefkowitz, R. H., Mukherjee, C., Coverstone, M., and Caron, M.

G. (1974) Biochem. Biophys. Res. Commun. 60, 703-709 14. Atlas, D., Steer, M. L., and Levitzki, A. (1974) Proc. N&l. Acad.

Sci. U. S. A. 71, 4246-4248 15. Aurbach, G. D., Fedak, S. A., Woodard, C. J., Palmer, J. S.,

Hauser, D., and Troxler, F. (1974) Science 186, 1223-1224 16. Mukherjee, C., Caron, M. G., Coverstone, M., and Lefkowitz, R. J.

(1975) J. Biol. Chem. 250, 4869-4876 17. Maguire, M. E., Wiklund, R. A., Anderson, H. J., and Gilman, A.

G. (1976) J. Biol. Chem. 251,1221-1231 18. Mukherjee, C., Caron, M. G., Mullikin, D., and Lefkowitz, R. J.

(1976) Mol. Pharmacol. 12, 16-31 19. Salomon, Y., Londos, C., and Rodbell, M. (1974) Anal. Biochem.

58, 541-548 20. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

(1951) J. Biol. Chem. 193, 265-275 21. Mukhejee, C., Caron, M. G., and Lefkowitz, R. J. (1975) Proc.

Natl. Acad. Sci. U. S. A. 72, 1945-1949 22. Limbird, L. E., and Lefkowitz, R. J. Proceedings of the Nuto

Advanced Study Course on Cell Surface Receptors, in press 23. Cheng, Y., and Prusoff, W. H. (1973) Biochem. Pharmacol. 22,

3099-3108 24. Ariens, E. J. (1967) Ann. N. Y. Acad. Sci. 139, 606-631

25. Lefkowitz, R. J. (1975) Biochem. Pharmacol. 24, 583-590 26. Limbird, L. E., De Meyts, P., and Lefkowitz, R. cJ. (1975) Biochem.

Riophys. Res. Comma. 64, 1160-1168 27. Rodbard, D. (1973) Adu. Exp. Med. Biol. 36, 289-326 28. De Meyts, P. (1976) in Methods in Molecular Biology (Blecher M.,

ed) Marcel Dekker, New York, in press

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

2384 Soluble /3-Adrenergic Receptors

29. Laurent, T. C., and Killander, J. (1964) Biochim. Biophys. Acta 112,346-362

30. Tanford, C. (1973) in The Hydrophobic Effect: Formation of Micelles and Biological Membranes, John Wiley and Sons, New York

31. Fish, W. W., Reynolds, J. A., and Tanford, C. (1970) J. Biol. Chem. 245, 5166-5168

32. Martinez-Carrion, M., Sator, V., and Raftery, M. A. (1975) Biochem. Biophys. Res. Commun. 65, 129-137

33. Limbird, L. E., and Lefkowitz, R. J. (1976) Mol. Pharmacol., in press

34. Burton, P. M., and Josse, J. (1970) J. Biol. Chem. 245,4358-4364 35. Horton, H. R., Kelly, H., and Koshland, D. E., Jr. (1965) J. Biol.

Chem. 240, 722-724

44. Hubbard, R. (1954) J. Gen. Physiol. 37, 381-399 45. Beld, A. J., and Ariens, E. J. Eur. J. Pharmacol. 25, 203-209 46. Razin, S. (1972) Biochim. Biophys. Acta 265, 241-296 47. Coleman, R. (1973) Biochim. Biophys. Acta 300, l-30 48. Tate, R. L., Holmes, J. M., Kohn, I,. D., and Winand, R. J. (1975)

36. Horton. H. R., and Koshland. D. E.. Jr. (1967) Methods Entvmol. 49. 11, 556-565

37. Horton, H. R., and Koshland, D. E., Jr. (1965) J. Am. Chem. Sot. 87, 1126-1132

38. Neet, K. E., Nanci, A., and Koshland, D. E., Jr. (1968) J. Biol. Chem. 243, 6392-6401

39. Weiner, H., White, W. N., Hoare, D. G., and Koshland, D. E., Jr. (1969) J. Am. Chem. Sot. 88, 3851-3859

40.

41.

42.

43.

50.

51.

Neet, K. E., and Koshland, D. E., Jr. (1966) Proc. Natl. Acad. Sci. U. S. A. 56, 1606-1611

Riordan, J. F., and Vallee, B. L. (1967) Methods Enzymol. 11, 541-548

Hoare, D. G., and Koshland, D. E., Jr. (1967) J. Biol. Chem. 242, 2447-2453

Riordan, J. F., and Vallee, B. L. (1967) Methods Enzymol. 11, 570-576

J. Biol. Chem. 250, 6527-6533 De Meyts, P., Roth, J., Neville, D. M., Jr., Gavin, J. R., III, and

Lesniak, M. A. (1973) Biochem. Bioph.ys. Res. Commun. 55, 154-161

Stahl, E. Thin Layer Chromatography (1969) 2nd Ed, pp. 396-400, Springer-Verlag, New York

Lefkowitz, R. J., Mukherjee, C., Caron, M. G., Limbird, L. E., Williams, L. T., Alexander, R. W., Mickey, J. V., and Tate, R. (1976) Recent Prog. Horm. Res., in press

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

M G Caron and R J Lefkowitzfrog erythrocytes.

Solubilization and characterization of the beta-adrenergic receptor binding sites of

1976, 251:2374-2384.J. Biol. Chem. 

  http://www.jbc.org/content/251/8/2374Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/251/8/2374.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from