Only part of the ,B-adrenergic receptors can undergo functional coupling to the adenylate cyclase regulatory unit. This receptor subpopulation shows an increased affinity for agonists in the presence of Mg2' and undergoes rapid "inactivation" (locking-in of the agonist) by the alkylating reagent N-ethylmaleimide in the presence of agonists. Several experimental conditions, known to modify the total receptor concentration without alteration of the other components of the acenylate cyclase system, do not affect the percentage of receptors that can undergo functional coupling: (i) homologous regulation of B13 receptors in rat brain by noradrenaline (through antidepressive drug or reserpine injections); (it) upand down-regulation of the 182 receptors in Friend erythroleukemia cells by, respectively, sodium butyrate and cinnarizine treatment; and (iiW) dithiothreitol-mediated inactivation of receptors in turkey erythrocytes, Friend erythroleukemia cells, and rat brain. Our findings argue against a stoichiometric limitation in the number of regulatory components, genetically different receptor subpopulations, bound guanine nucleotides, or reduced accessibility of part of the receptors to the agonists as the cause for functional receptor heterogeneity. Differences in either the receptor conformation or its membrane microenvironment are more plausible explanations. dence for the view that the action of both Mg2' and reagent requires the coupling between the receptor and the guanine nucleotide-binding regulatory component of the adenylate cyclase system, designated Ns (8, 12). In this context, current models concerning the MalNEt action mechanism are based on the fact that coupling of the agonist-bound /3-adrenergic receptor to Ns is accompanied by the exposure of sulfhydryl groups, probably at the surface of Ns itself (unpublished data; ref. 10). Alkylation of these groups by MalNEt results in the freezing of the receptors in an active, slow agonist-dissociating conformation (i.e., locking-in of the agonist) and, hence, in their apparent inactivation (9, 10). To investigate the possible basis for the restriction in receptor-Ns coupling, we have tested whether various factors known to alter the total receptor number could also alter the proportion of coupling-prone receptors. Using the agonist/ MalNEt reaction, we show here that changes in the total receptor number by various factors (i.e., chemical inactivation, homologous hormonal regulation, and butyrate and cinnarizine treatment) do not necessarily imply an alteration of the Ns-coupled receptor fraction. Our data suggest that the limited receptor-Ns coupling is probably related to limitations in the membrane or receptor structure. Several recent studies have shed light on marked structural and functional differences between /-adrenergic agonists and antagonists in their interaction with their receptors (17). In particular, the receptors behave as a homogeneous population of sites with regard to antagonist binding, but two subpopulations can be discriminated for agonist binding. This has been established by two types of observations. First, magnesium ions cause an increase in agonist but not in antagonist affinity for binding to a well-defined proportion of the 8-adrenergic receptors in several tissues (5, 6). Second, agonist but not antagonist binding causes a conformational modification of only part of the B3-adrenergic receptors. This conformational modification can be monitored by the ability of a combination of agonist and the group-specific reagent N-ethylmaleimide (MalNEt) to impair subsequent radioligand binding (7-10). Interestingly, both Mg2+ and MalNEt affect the same receptor subpopulation (unpublished data; ref. 8). Moreover, the basis of receptor heterogeneity is functional rather than pharmacological. Experiments on turkey erythrocyte membranes, human adipose cell membranes (,81-adrenergic receptors), and S49 lymphoma cell membranes (132-adrenergic receptors) have shown that the agonist/MalNEt-sensitive and -resistant receptors have the same pharmacological specificity (7, 8, 11). In contrast, there is now ample eviMATERIALS AND METHODS Materials. The following were obtained as kind gifts: (-)isoproterenol bitartrate from Sterling Winthrop; fenoterol hydrobromide from Boehringer-Ingelheim; practolol hydrochloride from ICI; (+)-alprenolol hydrochloride, phentolamine hydrochloride, desipramine, and reserpine from CIBA-Geigy; doxepin from Pfizer; mianserin from Organon; pargyline from Abbott; and nisoxetine from Eli Lilly. Nialamide, MalNEt, and dithiothreitol were purchased from Sigma. GTP and GMP were from Boehringer Mannheim. (-)-[3H]Dihydroalprenolol hydrochloride ([3H]H2Alp, 91 Ci/mmol; 1 Ci = 37 GBq) was obtained from New England Nuclear. [3H]CGP 12177 (42 Ci/mmol) was from Amersham. All other chemicals were of analytical grade. Source and Preparation of Membranes. Turkey erythrocyte membranes were prepared as described (7). Male Wistar rats (-250 g) underwent either 14 daily intraperitoneal injections with saline solutions of antidepressant drugs (i.e., desipramine, doxepin, nialamide, nisoxetine, and mianserin at 10 mg/kg and pargyline at 25 mg/kg) or 4 daily injections with a reserpine solution at 2.5 mg/kg; they were then decapitated. All subsequent steps were performed at 4°C. The cerebral cortex was homogenized in 10 volumes of 10 mM Tris-HCl, pH 7.4/0.25 M sucrose in an Ultraturax Abbreviations: MalNEt, N-ethylmaleimide; [3H]H2Alp, (-)-[3H]dihydroalprenolol. tDeceased on May 25, 1983. 4637 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 4638 Biochemistry: Severne et al. homogenizer and subsequently in a glass/Teflon homogenizer (five strokes). The homogenate was centrifuged subsequently at 900, 10,000, and 40,000 x g for 15 min. The final pellet of cortical membranes was retained. Friend erythroleukemia cell culture and membrane preparation are described in ref. 13. Receptor induction was obtained by incubation of the culture with 2 mM sodium butyrate for 24 hr. Incubation with 10 ,M cinnarizine lasted 24 hr. All membrane preparations were suspended in 10 mM Tris HCl, pH 7.4, 10% (vol/vol) glycerol and stored in liquid nitrogen at a protein concentration of -10 mg/ml. Protein concentrations were determined according to Lowry et al. (14). Membrane Pretreatments. Membranes (1-4 mg of protein per ml) were preincubated for 10 min at 30'C in 75 mM Tris HCl, pH 7.4/25 mM MgCl2 containing 0.5 AuM (-)-isoproterenol/0.2 mM MalNEt in a final volume of 0.5 ml or 1 ml (for rat brain membranes). Preincubations were terminated as follows: turkey erythrocytes and rat brain, centrifugation (2 min at 12,000 rpm in an Eppendorf centrifuge at room temperature) and resuspension of the membranes in 1 ml of fresh buffer (three times); Friend erythroleukemia cells, centrifugation as above but in the presence of4% (vol/vol) polyethylene glycol 6000 (final concentration). Polyethylene glycol 6000 allowed fast and quantitative precipitation of the membranes and affected neither [3H]H2Alp binding nor receptor-Ns coupling. Preincubation of various membranes with 5 mM dithiothreitol was performed as described (15). Radioligand Binding. Binding of [3H]H2Alp and [3H]CGP 12177 to the membrane preparations was assayed by filtration on glass fiber filters. Membrane protein (0.75-2 mg/ml) was incubated with the indicated concentrations of radioligand for 10 min at 30°C in 50 mM Tris HCl, pH 7.4/25 mM MgCl2 containing 10 AM phentolamine (to prevent radioligand binding to a-adrenergic receptors) in a final volume of 200-500-,l. Membranes were then filtered as described (7), and the radioactivity on the filters was assayed in a Packard liquid scintillation spectrometer. Specific binding was obtained by subtracting nonspecific binding [i.e., binding in the presence of 10 ,M (-)-isoproterenol] from total binding. In all figures and tables, bound radioligand refers to specific binding as defined above. The Scatchard plots (16) of the saturation binding curves were rectilinear for all of the membrane preparations, so that the total receptor number and equilibrium dissociation constant (Kd) for [ H]H2Alp and for [3H]CGP 12177 binding could be calculated by linear regression analysis.

beta 1-Adrenergic receptors of turkey and beta 2-receptors of rat erythrocyte membranes were identified by (-)-[3H]-dihydroalprenolol ([3H]-DHA) binding. Treatment of these membranes with the reagent cyanogen bromide causes a concentration dependent decrease in the number of binding sites, without affecting the affinity of the tracer. Whereas inactivation of the beta 1-receptors is not affected by the pH (between pH 6 and 9), substantial inactivation of the beta 2-receptors takes place only under mild acidic conditions. The ability of CNBr (cyanogen bromide) to discriminate between both receptor subclasses at basic pH is also confirmed for rat heart membranes (beta 1-receptors), Friend erythroleukemia and S49 lymphoma cell membranes (beta 2-receptors).

The beta 2-adrenergic receptors of bovine trapezius muscle membranes demonstrate tight agonist binding as a result of the formation of complexes between agonists, receptors, and Ns. Preincubation of the membranes with (-)-isoproterenol (followed by washing) causes a time- and concentration-dependent decrease in the number of radioligand-binding sites to a plateau value of 41.5 +/- 4%. The affinity of the remaining sites for the radioligand (-)-[3H]dihydroalprenolol is unchanged. This decrease is stable under radioligand binding conditions but is readily reversed in the presence of GTP. The isoproterenol/(-)-[3H]dihydroalprenolol competition binding curves are shallow. Such a phenomenon is usually interpreted in terms of two interconvertible affinity states of the receptor: the high affinity state reflecting the coupling of the agonist X receptor complex to Ns and the low affinity state not interacting with Ns. However, the competition curves undergo time-dependent shifts to the left. This apparent non-equilibrium can be explained by a model in which tight agonist binding to part of the receptor population is included. The usual computerized interpretation of the competition binding curves do not allow the correct evaluation of agonist binding parameters in the presence of tight agonist binding.

Both alpha- and beta-adrenergic receptors have been identified in human myometrium by radioligand binding. Both types of receptors mediate uterine contractility: alpha-adrenergic agonists cause uterine contraction, whereas beta-adrenergic agonists induce relaxation. We studied the possible regulatory effects of gonadal steroids on the affinity, concentration, subtype distribution, and linkage to adenylate cyclase of beta-adrenergic receptors in human myometrium removed during different phases of the menstrual cycle, from postmenopausal women and during depo-progestin (medroxyprogesterone acetate) therapy. We identified beta-adrenergic receptors in human myometrial membranes using the radiolabeled antagonist (--)-[3H]-dihydroalprenolol (DHA). The binding of this radioligand was rapid, reversible, of high affinity (KD = 0.71 nM) and stereo-selective. Total beta-receptor concentration was determined by Scatchard analysis of DHA saturation binding and the ratio of receptor subtypes determined by computer-assisted analysis of beta 2 selective antagonist ICI 118 551/DHA competition binding curves. The fraction of receptors functionally coupled to adenylate cyclase was determined by the agonist/N-ethylmaleimide inactivation method. The affinity of DHA and the fraction of receptors undergoing functional coupling was similar under all hormonal conditions. However, whereas the net concentration of beta-receptors was the same in all groups, beta 1-adrenoreceptors could only be detected in myometrial particulate fractions from uteri obtained in the midfollicular phase, indicating the importance of considering adrenoreceptor subtypes as separately regulatable receptors.

Mammalian β2-adrenergic receptors (R) have been shown to be structurally heterogenous with respect to glycosylation (Stiles et al. J. biol. Chem.259, 8655 (1984)). They are also heterogenous with respect to functional coupling to Ns. The ternary H.R.Ns complex can be frozen in the presence of the alkylating reagent N-ethylmaleimide. In hamster lung membranes 45% of the receptors are agonist/N-ethylmaleimide sensitive (i.e. coupling-prone receptors). β-Receptors in both native and isoproterenol/N-ethylmaleimide pretreated membrane preparations are retained by affinity chromatography on concanavalin A and wheat germ agglutinin and are equally sensitive to neuraminidase treatment. This is exhibited by the increase in mobility of the 125I-iodocyanopindolol-azide photoaffinity labeled receptor peptide in SDS-polyacrylamide gel electrophoresis. These observations suggest that there is no link between the structural and functional heterogeneity of the receptors. Moreover, both partial (using neuraminidase) and near total (using endoglycosidase F) deglycoslation of membrane-bound receptors does not affect the H.R.-Ns coupling capacity as compared to native receptors.

The beta-adrenergic receptor in membranes from cat, rat, and guinea pig hearts appears to undergo spontaneous coupling to the stimulatory nucleotide-regulatory protein, Ns. This was demonstrated by incubating cardiac membranes from acutely reserpinized animals with the alkylating reagent N-ethylmaleimide (NEM), which is known to "freeze" the beta-receptor Ns-complex. The concentration of norepinephrine in these membrane preparations was less than 0.1 nM. Cat heart membranes were preincubated for 10 min at 30 degrees with NEM (10(-7)-10(-3) M) and subsequently incubated with (-)-[125I]iodopindolol (IPIN) (18 min, 30 degrees). NEM caused a concentration-dependent decrease in specific IPIN binding with a maximal reduction of about 20% at 0.1 mM. This decrease occurred even in the absence of MgCl2 (0.5 mM EDTA added as well). A comparable reduction was also observed in myocardial membranes from rat and guinea pig. This fall reflects a decrease in the number of beta-adrenergic receptor sites, as demonstrated by saturation binding experiments with IPIN. The equilibrium dissociation constant of the radioligand for the remaining receptors was not affected. When increasing concentrations of GTP were included in the preincubation mixture, it resulted in a dose-dependent protection of NEM-induced decrease in IPIN binding. The protection was complete at 0.1 mM GTP. In addition, GTP reversed the NEM effect when added to the incubation mixture 10 min after NEM. The apparent reduction in cardiac beta-adrenergic receptor number by NEM (in the absence of beta-receptor agonist) is compatible with a model in which part of the receptor population is able to undergo spontaneous coupling to Ns.

S OF PAPERS SUBMITTED FOR PRESENTATION TO THIS SOCIETY The following papers have been submitted to the Secretary and the Associate Secretaries of the Society for presentation at meetings of the Society. They are numbered serially throughout this volume. Cross-references to them in the reports of the meetings will give the number of this volume, the number of this issue, and the serial number of the abstract. 192. Dr. W. I. Miller: Fundamental regions for the simple group of order 168 in S±. Fundamental regions have been determined for certain groups in 6*4 (Price, American Journal of Mathematics, vol. 40 (1918), pp. 108-112). In this paper fundamental regions for the ternary G\^ are obtained by the use of 21 forms of degree four in which every term is of degree two in the variables x\, #2, #3, and of degree two in the conjugate imaginary variables. These forms are written as the differences of seven positive definite forms, so that their behavior under the group may be studied by means of the permutation group of degree seven. If we set xi/xz — x+iu, x\ ABSTRACTS OF PAPERS 501 " 3 ," the "relation of implication" of Principia Mathematica. The number of postulates is thus considerably smaller than the number (eight, including an inadvertently omitted existence postulate) in Huntington's set of postulates expressed in terms of " 3 " (Proceedings of the National Academy of Sciences, vol. 18(1932), p. 180). The author also shows that there is a very close relation between the operation "D " and the operation " — " of "exception," by virtue of which relation any set of postulates in terms of " — " is essentially also a set of postulates in terms of "D ," and vice versa. Finally, the author derives from his postulates the "theory of deduction " of Principia Mathematica. This derivation brings out the fact, perhaps more directly than has been done before, that the propositions of the theory of deduction are all propositions in the general logic of classes. (Received May 16, 1933.) 196. Mr. A. M. Tut t le : A system of independent mathematical postulates f or Dirac's quantum mechanics. This paper exhibits a system of postulates which are sufficient to furnish a rigorous formal foundation for a type of mathematics used by Dirac in his quantum mechanics (Dirac, Principles of Quantum Mechanics, Oxford, 1930). The first twelve postulates furnish a calculus for the system. The notion of a general linear process has been introduced instead of the processes of summation and integration which Dirac uses. In the first twelve postulates this process is not so restricted as to exclude an ordinary integration and many of Dirac's results are obtained upon this basis. However, in order to derive all of Dirac's expansion theorems, it has been necessary to add two more postulates which so restrict the general linear process as to exclude an ordinary integration. The postulates are shown to be consistent and independent. (Received May 19, 1933.) 197. Professor H. S. Vandiver: Summary of results and proofs on Fermât's last theoem. Seventh paper. The main result obtained is as follows. If / is an odd prime and x+y-\-z=*0 is satisfied in rational integers, none zero, and prime to the odd prime I, then the second factor of the class number of the algebraic field defined by e is divisible by /. Other criteria are also found and the proofs indicated. (Received May 19, 1933.) 198. Professor A. A. Albert: Integral domains of rational generalized quaternion algebras. In recent years several papers on the integers of generalized quaternion algebras over R have appeared and various special cases have been considered. But the general case has not yet been completed; in fact the question as to whether it would not be possible to eliminate many of these special cases has not been discussed. The present paper utilizes simple algebraic transformations as well as the theory of ternary quadratic null forms to obtain a remarkable canonical form for any generalized quaternion algebra over R. Moreover the author shows that for algebras in the canonical form there is a single domain of integrity containing the basal units, and determines this domain. (Received May 20, 1933.) License or copyright restrictions may apply to redistribution; see https://www.ams.org/journal-terms-of-use 502 ABSTRACTS OF PAPERS [July, 199. Professor H. L. Olson: Bilinear correspondences in two and in three dimensions. Every bilinear form in n-\-n variables xi, • • • , xn, yi, • * * , yn can be interpreted geometrically as determining a correspondence, in terms of homogeneous coordinates, between points and hyper-planes (linear subspaces of n — 2 dimensions) in a flat space of » —1 dimensions. This paper classifies, projectively, such correspondences in 3 + 3 and 4 + 4 variables, and presents a canonical form for each class. (Received May 17, 1933.) 200. Miss A. V. Newton: Consecutive covariant configurations at a point of a space curve. The purpose of this paper is to make some contributions to the projective differential geometry of space curves. There are many configurations covariantly connected with a point P of such a curve, such as tangent line, osculating cone, and Halphen point. As the point P moves along the curve each covariant point describes a curve, each covariant line describes a surface, and each covariant surface has an envelope. On considering a point Q infinitesimally near to the point P and setting up a transformation of coordinates between local coordin* ate systems at P and Q, it becomes possible to obtain considerable information about these curves and envelopes. In particular, one can find the tangent lines of the curves, the characteristics of the envelopes, and the foci of the edges of regression of the envelopes. (Received May 20, 1933.) 201. Professor A. H. Copeland: Consistency of the conditions determining Kollektivs. A rigorous basis for the statistical theory of probability is obtained by the establishment of the consistency of the conditions determining the set of Kollektivs. The author shows that , in a certain sense, the physical situation must be in agreement with this theory. In this respect the theory differs from all other physical theories. Moreover, the Kollektiv establishes a closer relationship between physical situations and their mathematical formulations. The conditions determining Kollektivs are due to von Mises. Consistency can be established by means of the construction of sequences of points satisfying the conditions. The consistency of the conditions for Kollektivs of a very special type can be shown by means of the nombres normaux of Borel. For this type the conditions assume a much simpler form. In this paper it is shown that a slight modification of the conditions for the more general type is essential to achieve consistency. I t is further shown by a method similar to that of Borel that these modified conditions can be satisfied. (Received May 17, 1933.) 202. Mr. T. N. E. Greville: Invariance of the property of admissibility under certain general types of transformations. This paper is concerned with a set of transformations on certain sequences, which include the type of Kollektiv called by von Mises "die einfachste alternative" and the admissible number of Copeland. I t is possible to formulate completely in terms of the transformations of this set a large class of problems arisLicense or copyright restrictions may apply to redistribution; see https://www.ams.org/journal-terms-of-use 1933-] ABSTRACTS OF PAPERS 503 ing in the classical theory of probability for which the operations given by von Mises are not adequate. The consistency of the assumptions of the classical theory is investigated by means of certain properties of the sequences which are invariant under these transformations. The method used in demonstrating the invariance of the properties in question is similar to that employed by Borel in proving the existence of nombres normaux. (Received May 17, 1933.) 203. Dr. Francis Regan: The application of the theory of admissible numbers to time series with variable probability. A time series is a sequence of occurrences which may be represented by a set of points on a time axis. In this paper the author is interested in the time series whose points are determined with relation to a distribution function which assigns a definite probability to every interval of the time axis. The probability varies according to the length and position of the interval. The concept of admissibility is extended to this type of time series by s

Incubation of turkey erythrocyte and rat lung membranes with the beta-adrenergic agonist isoproterenol in the presence of the group specific reagent N-ethylmaleimide (NEM) causes the apparent inactivation of a well defined fraction of the beta-adrenergic receptors. This inactivation process requires the coupling between the receptors and the guanine nucleotide binding regulatory component of the adenylate cyclase system. Accordingly, only the isoproterenol/NEM sensitive receptor population is susceptible to undergo this functional coupling. Incubation of the treated membranes in alkaline buffer and/or in the presence of GTP reactivates the receptors. The reactivated receptors show the initial binding properties for both isoproterenol and the radiolabelled antagonist [3H]-dihydroalprenolol and are still dithiothreitol sensitive. Moreover the reactivated receptors can again couple to the regulator in the presence of agonists. The present study makes it possible to discriminate between the different models advanced for the N-ethylmaleimide action and is valid for beta 1- as well as beta 2-adrenergic receptors. The reagent appears indeed to freeze the hormone-receptor-regulator complex by alkylating the latter component, as previously proposed by Korner et al. (1982) J. biol. Chem. 257, 3389-3396. It is also suggested that the existence of a limited fraction of coupling-prone receptors results from differences in the membrane microenvironment or from a structural heterogeneity of the receptors rather than from a stoichiometric limitation of the regulator concentration.

The tight binding of noradrenaline (NA) to β-adrenergic receptors was studied in membranes from the left ventricular myocardium of the rat. Addition of GTP (0.1 mM) to membrane preparations from control rats (NA concentration 8.5 ± 2.1 nM) caused a 4–35% (N = 8) increase (P < 0.01) in the number of specific binding sites of [125I](−)pindolol. In contrast, addition of GTP did not cause any changes in the number of β-adrenergic receptors in heart membranes from reserpinized animals (NA concentration < 0.1 nM). In heart membranes from reserpinized animals, preincubation with NA (followed by washing) revealed a time- and concentration-dependent decrease with a maximum of 35–40% in the [125I](−)pindolol-binding sites. This agonist-mediated decrease in the number of receptors was prevented if GTP was also present in the NA-preincubation medium. It is concluded that NA can undergo tight binding to β-adrenergic receptors in rat heart membranes. The heart-membrane preparations contain endogenous NA which, via tight agonist binding, is responsible for masking part of the β-adrenergic receptor population.

Only part of the beta-adrenergic receptors can undergo functional coupling to the adenylate cyclase regulatory unit. This receptor subpopulation shows an increased affinity for agonists in the presence of Mg2+ and undergoes rapid "inactivation" (locking-in of the agonist) by the alkylating reagent N-ethylmaleimide in the presence of agonists. Several experimental conditions, known to modify the total receptor concentration without alteration of the other components of the adenylate cyclase system, do not affect the percentage of receptors that can undergo functional coupling: (i) homologous regulation of beta 1 receptors in rat brain by noradrenaline (through antidepressive drug or reserpine injections); (ii) up- and down-regulation of the beta 2 receptors in Friend erythroleukemia cells by, respectively, sodium butyrate and cinnarizine treatment; and (iii) dithiothreitol-mediated inactivation of receptors in turkey erythrocytes, Friend erythroleukemia cells, and rat brain. Our findings argue against a stoichiometric limitation in the number of regulatory components, genetically different receptor subpopulations, bound guanine nucleotides, or reduced accessibility of part of the receptors to the agonists as the cause for functional receptor heterogeneity. Differences in either the receptor conformation or its membrane microenvironment are more plausible explanations.

We have characterized two rearrangements consisting of inverted repeats of the argE gene. The promoters (p) of argE and of argCBH face each other over an internal operator. The rearrangements were obtained as reactivations of argE in a strain harboring an argEp deletion on a lambda darg prophage. In both cases the repeat included argE and argCBHp on either side of a unique sequence; the result is a divergent operon in which each copy of argCBHp reads into the adjacent argE repeat. In one case, the pair of repeats adjoins the silent parental gene, forming a triplication (comes from leads to comes from). The other rearrangement consists of a single argE palindrome, but the whole prophage is rearranged into an inverted repeat, analogous to certain lambda dv's. Both structures could be explained by breakage of a replication fork passing argE and by inaccurate rejoining of strands. The lambda dv-like rearrangement would result from breakage at both replication forks of a phage or prophage replicating during transient release of immunity. The triplication would imply breaking of a chromosomal replication fork, formation of a cyclic intermediate by recombination between the daughter duplex molecules and reinsertion into the parental argE gene. Formation of a triplication by replication errors involving appropriate strand switchings and branch migrations can not be excluded however.

Beta-Adrenergic agonists cause adenylate cyclase activation via a conformational change of their receptor. This was earlier revealed by the use of N-ethylmaleimide as a structural probe: agonist-bound receptors are rapidly inactivated by 0.1 mM N-ethylmaleimide while free and antagonist-bound receptors remain unaffected. Furthermore, the beta-adrenergic receptors only change conformation when coupled to the guanine nucleotide regulatory component of the adenylate cyclase system. It is shown in this report that treatment of turkey erythrocyte membranes with the sulphydryl-specific agent 2,2'-dinitro-5,5'-dithiodibenzoic acid and elevated concentrations of N-ethylmaleimide (1-10 mM) do not affect the total receptor number, but prevent the ability of beta-adrenergic agonists to mediate conformational changes. This effect is mimicked by GTP. These three compounds cause also a two- to four-fold decrease in agonist affinity. Both phenomena may be explained by the ability of the reagents to prevent, and of GTP to reverse, the functional coupling between the receptors and the regulatory component of the adenylate cyclase system. Removal of Mg2+ from the incubation medium (i.e. presence of 1 mM EDTA) produces a similar decrease in agonist affinity, but does not impair the ability of agonist/N-ethylmaleimide to inactivate the receptors. This suggests that Mg2+ increases the agonist affinity for the receptor-regulatory component complex, but is not required for its formation.