Guanine Nucleotide and Cation Regulation of μ, δ, and k Opioid Receptor Binding: Evidence for Differential Postnatal Development in Rat Brain

A study of the onset of cation and guanine nucleotide regulation of delta, mu, and kappa rat brain opioid receptors during postnatal development was undertaken. Site-specific binding assays were utilized for each receptor type and the effects of 0.5 mM MnCl2, 100 mM NaCl, and/or 50 microM guanosine-5'-(beta, gamma-imido) triphosphate [Gpp(NH)p] were assessed. The most pronounced changes of opioid binding were seen in the presence of Mn2+. In adults, agonist binding to delta sites was stimulated by Mn2+, whereas that to mu sites was not affected and kappa binding was inhibited. The postnatal development of Mn2+ regulation for the three receptor subtypes was distinctly different. The largest effects were seen on delta sites detected in the early neonatal period, Mn2+ eliciting a 68% stimulation of binding over controls at day 1. Significant inhibition of kappa site binding by Mn2+ was detected only after the third postnatal week. Mn2+ caused a significant reversal of Gpp(NH)p inhibition of delta binding in the early neonatal period, exceeding that in the absence of regulators. Inhibition of mu and delta receptor binding by Na+ was greater, and the Mn2+ reversal of this effect was smaller, in the first 2 postnatal weeks than in adults. Gpp(NH)p + Na+ regulation did not change appreciably during the postnatal period. However, Mn2+ reversal of the considerable inhibition elicited by the combination of Na+ and Gpp(HN)p was developmental time-dependent. The data are discussed in terms of multiple sites of interaction for guanine nucleotides and cations.(ABSTRACT TRUNCATED AT 250 WORDS)     

Solubilization in high yield of opioid receptors retaining high-affinity delta, mu and kappa binding sites

The binding sites for opiates (agonist and antagonist) and opioid peptides can be solubilized from rat brain membranes with digitonin in the presence of Mg2+ (10 mM). High affinity and high capacity binding to the soluble delta, mu, and kappa receptors is obtainable when the membranes are treated in Mg2+ (30 degrees C, 60 min) prior to solubilization. The yields of solubilized binding sites extracted with digitonin, 40-90%, are higher than those obtained from Mg2+-pretreated membranes with other detergents commonly used for receptor solubilization. The stability of the digitonin-soluble opioid receptor at room temperature makes it useful for purification and characterization.     

Differential Sensitivity of Basal and Opioid-Stimulated Low Km GTPase to Guanine Nucleotide Analogs

In membranes derived from NG108-15 cells, the opioid peptide [D-Ala2,D-Leu5]enkephalin (DADLE) stimulates a low Km GTPase. The nucleotide analogs guanosine 5'-O-(2-thio)diphosphate (GDP beta S), guanosine 5'-(beta,gamma-imido)triphosphate [Gpp(NH)p] and guanosine 5'-O-(3-thio)-triphosphate (GTP gamma S) inhibit the basal enzymatic activity with the order of potency GTP gamma S greater than Gpp (NH)p greater than GDP beta S. In the presence of DADLE, the inhibition isotherms of GDP beta S and Gpp(NH)p are shifted to the right five- and fourfold, respectively, compared to the inhibition observed in the absence of DADLE. In contrast, the IC50 of GTP gamma S for inhibiting the enzyme is reduced by 55% in the presence of the opioid. Both Gpp(NH)p and GTP gamma S produce a concentration-dependent increase in the Km(app) of GTPase, without affecting its Vmax, indicating a competitive inhibition. However, the replots of Km(app) versus inhibitor concentration are hyperbolic, suggesting a partial type of inhibition. Both Gpp(NH)p and GTP gamma S, but not GTP, induce an increase in the EC50 of DADLE for stimulating GTPase. These findings indicate that the basal and the opioid-stimulated low Km GTPase differ in their respective sensitivities to inhibition by guanine nucleotide analogs.     

Determination of the sizes of the opioid receptors in their membrane environment by radiation inactivation

Opioids, like other drugs, are thought to initiate their effects by association with their specific receptors. However, very little is known about the opioid receptor as a molecular entity. The binding components have been solubilized in detergent and purified by different approaches, but the molecular size of soluble opioid receptor complexes reported by different groups varied from 23,000 to 750,000. In this study, the technique of radiation inactivation by gamma rays was used to investigate the apparent size of the opioid receptor in rat brain membranes under different conditions. The molecular sizes of opioid receptor complexes were estimated as 313,000 +/- 13,500 in the presence of [D-Ala2, D-Leu5] enkephalin, NaCl and Gpp (NH)p; as 165,000 +/- 8,500 in the presence of NaCl only, or of both NaCl and Gpp (NH)p; as 217,000 +/- 6,600 in the presence of Gpp (NH)p only; and as 286,000 +/- 60,900 in the presence of MgCl2 only. A simple model has been proposed to explain these different apparent target sizes of opioid receptors obtained under different conditions.     

G protein regulation of phospholipase C activity in a membrane-solubilized system occurs through a Mg2(+)- and time-dependent mechanism

GTP-binding proteins have been implicated to function as key transducing elements in the mechanism underlying receptor activation of a membrane-associated phospholipase C activity. In the present study, the regulation of phospholipase C activity by GTP-binding proteins has been characterized in a detergent-solubilized system derived from bovine brain membranes. Guanosine-5'-(3-O-thio)triphosphate (GTP-gamma-S) and guanyl-5'-yl imidodiphosphate (Gpp(NH)p) stimulated a dose-dependent increase in phospholipase C activity with half-maximal activation at 0.6 microM and 10 microM, respectively. The maximal degree of stimulation due to Gpp(NH)p or GTP-gamma-S was comparable. 100 microM GTP had only a slight stimulatory effect on phospholipase C activity. Adenine nucleotides, 100 microM adenylyl-imidodiphosphate and ATP, did not stimulate phospholipase C activity, indicating that specific guanine nucleotide-dependent regulation of phospholipase C activity was preserved in the solubilized state. Gpp(NH)p or GTP-gamma-S stimulation of phospholipase C activity was time-dependent and required Mg2+.Mg2+ regulated the time course for activation of phospholipase C by guanine nucleotides and the ability of guanine nucleotides to promote an increase in the Ca2+ sensitivity of phospholipase C. 200 microM GDP-beta-S or 5 mM EDTA rapidly reversed the activation due to GTP-gamma-S or Gpp(NH)p. These findings demonstrate that G protein regulation of phospholipase C activity in a bovine brain membrane- solubilized system occurs through a Mg2+ and time-dependent mechanism. Activation is readily reversible upon addition of excess GDP-beta-S or removal of Mg2+.     

Solubilization and separation of the glucagon receptor and adenylate cyclase in guanine nucleotide-sensitive states.

Adenylate cyclase in liver membranes was solubilized with Lubrol PX and partially purified by gel filtration. The partially purified enzyme was susceptible to activation by guanyl-5'-yl imidodiphosphate (Gpp(NH)p). Studies on the binding of [3H]Gpp(NH)p to various fractions eluted from the gels revealed that an upper limit of 1% of the Gpp(NH)p binding sites is associated with adenylate cyclase activity stimulated by the nucleotide. The glucagon receptor, pretagged with 125I-glucagon in the membranes, solubilized with Lubrol PX, and fractionated on the same gel columns, eluted in a peak fraction that overlaps with, but is separate from, adenylate cyclase in its Gpp(NH)p-stimulated form. Addition of GTP to the solubilized glucagon-receptor complex caused complete dissociation of the complex, as has been shown with the membrane-bound form of the complex. Since the GTP-sensitive form of the glucagon receptor complex separates from the Gpp(NH)p-sensitive form of adenylate cyclase, it is concluded that the receptor and the enzyme are separate molecules, each associated with a distinct nucleotide regulatory site or component. These findings are discussed in terms of the possible structure of the hormone-sensitive state of adenylate cyclase.     

Characteristics of an adenosine A1 binding site in human placental membranes

Binding sites were solubilized from human placental membrane using 1.5% sodium cholate and were assayed using polyethylene glycol precipitation. These soluble binding sites had properties of an adenosine A1 binding site. 2-[3H]Chloroadenosine and N-[3H]-ethylcarboxamidoadenosine (NECA) binding were time dependent and reversible. Scatchard plots indicate two classes of binding sites with Kd values of 6 and 357 nM for 2-chloro[8-3H]adenosine and 0.1 and 26 nM with [3H]NECA. The specificity of [3H]NECA binding was assessed by the ability of adenosine analogs to complete for binding sites. Using this approach the estimated IC50 values were 60 nM for (R-PIA), 160 nM for S-PIA, 80 nM for NECA, and 20 nM for 2-chloroadenosine. Binding of [3H]NECA to the soluble sites is inhibited to 48% of the control value by 100 microM guanylyl-5'-imidodiphosphate (Gpp(NH)p). The IC50 value for NECA binding to the soluble binding site was increased from 80 nM to 1500 by Gpp(NH)p. There was a shift of binding affinity from a mixture of high and low affinity to only low affinity with 100 microM Gpp(NH)p. Despite these alterations a NECA prelabeled molecular species of 150 kDa did not decrease in molecular weight upon the addition of 100 microM Gpp(NH)p during high-performance liquid chromatography on a Superose 12 column. Other evidence to support the concept of preferential solubilization and assay of a small population of A1 binding sites was obtained. Following solubilization adenosine A2-like binding sites could be detected only in reconstituted vesicles. The existence of small amounts of A1 binding sites in intact human placental membranes was directly demonstrated using the A1 agonist ligand N6-[3H]cyclohexyladenosine and the A1 antagonist ligand 8-[3H]cyclopentyl-1,3-dipropylxanthine. JAR choriocarcinoma cells have "A2-like" membrane binding sites. In contrast to placental membranes, only A2-like binding sites could be solubilized from JAR choriocarcinoma cells. These observations indicate that human placental membranes contain adenosine A1 binding sites in addition to A2-like binding sites. These sites are guanine nucleotide sensitive, but do not shift to a lower molecular weight form upon assumption of a low affinity state.     

Interaction with lectin of kappa opioid binding sites solubilized from human placenta

Kappa opioid binding sites from human placenta, prelabeled with 3H-etorphine and solubilized, were retained on wheat germ agglutinin (WGA) agarose and specifically eluted with N-acetylglucosamine. No significant retention was found with other immobilized lectins, including Concanavalin A (Con A), soybean seed lectin (SBA), Pisum sativum lectin (PsA), Lens culinaris Medik. lectin (LcA), and Lathyrus tingitanus lectin(LtA). About 23% of applied kappa sites were specifically eluted from WGA agarose, less than half of the proportion of rat brain opioid binding sites eluted from the same lectin (55%). Receptors from placental extracts were compared with those from other tissues enriched in either kappa or mu sites. The proportion of applied kappa sites from guinea pig cerebellum eluted specifically from WGA agarose was 36%, whereas elution of binding sites from rat thalamus and rabbit cerebellum (enriched in mu sites) was at a level of 55%. This difference in the level of retention on and elution from WGA may reflect differences in the sugar composition of the glycoproteins of the two types of receptors. Succinylation of WGA abolished its ability to retain opioid binding sites, consistent with involvement of sialic acid. However, currently available evidence suggests that differences in retention on WGA between kappa and mu sites may be due to differences in either sialic acid or N-acetylglucosamine content or both.     

Effect of urea-treatment on agonist binding affinity of the muscarinic acetylcholine receptor

Urea-treatment of the microsome fraction of the heart of guinea-pigs caused selective reduction in the apparent affinity of an agonist (carbachol), but not an antagonist (atropine), to muscarinic acetylcholine receptors (mAChR), measured as inhibition of binding of ³H-quinuclidinyl benzilate (³H-QNB). This effect was similar to that of Gpp(NH)p. The effects of urea-treatment and Gpp(NH)p were not additive. On the other hand, treatment of the microsome fraction with 5,5′-dithiobis (2-nitrobenzoic acid) (DTNB) increased the apparent affinity of agonist, but not antagonist. The effect of DTNB predominated over those of urea-treatment and Gpp(NH)p, when these treatments were combined with DTNB.     

Characterization of the alpha adrenergic receptor population in hippocampus up regulated by serotonergic raphe deafferentiation

Serotonergic raphe deafferentiation elicits an up regulation of a nM (3H)WB-4101 binding site in rat hippocampus for which norepinephrine displays high affinity and prazosin displays low affinity. Guanine nucleotide affects the nM binding to hippocampal alpha-1 adrenergic receptors. Firstly, Gpp(NH)p, a nonhydrolyzable analog of GTP, inhibits (3H)WB-4101 binding at 3 nM concentration of the radioligand, the ligand concentration labelling the lower affinity, nM, binding site. Secondly, the addition of Gpp(NH)p causes recovery of the heterogeneity of binding sites lost upon preincubation of the membranes with 100 microM epinephrine, apparently by decreasing the affinity of the nM (3H)WB-4101 binding site for the adrenergic receptors. The phenomenon was still observed in the presence of saturating concentrations of the alpha-2 antagonist, yohimbine, and the beta antagonist, propranolol. The results imply that Gpp(NH)p regulates ligand binding to hippocampal alpha-1 agonist sites. It is likely that agonist and antagonist binding sites for the alpha-1 receptor exist in hippocampus with the agonist site being modulated by serotonin.     

Effects of trypsin-treatment on agonist binding affinity to the muscarinic acetylcholine receptor

Trypsin-treatment of the microsome fraction of the ileum and the synaptic membrane fraction of the cerebral cortex of guinea-pig caused selective reduction in the apparent affinity of an agonist (carbachol), but not an antagonist (atropine), to muscarinic acetylcholine receptors (mAChR), measured as inhibition of binding of ³H-quinuclidinyl benzilate (³H-QNB). This effect was similar to that of Gpp(NH)p. The effects of trypsin and Gpp(NH)p were not additive. On the other hand, treatment of these fractions with 5,5'-dithiobis (2-nitrobenzoic acid) (DTNB) increased the apparent affinity of agonist, but not antagonist. The effect of DTNB predominated over those of trypsin and Gpp(NH)p, when the fractions were treated with two reagents simultaneously.     

Kinetics and Physical Parameters of Rat Brain Opioid Receptors Solubilized by Digitonin and CHAPS

Rat brain opioid receptors were solubilized with digitonin and a zwitterionic detergent, 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate (CHAPS). The yield of solubilization was 70-75% with digitonin and 30-35% with CHAPS. Kinetic and equilibrium studies performed from digitonin extracts resulted in KD values comparable with those of the membrane fractions. Two [3H]naloxone binding sites were obtained in the extracts similarly to membrane fractions. The rank order potency of drugs used in the competition experiments did not change during solubilization. The distributions of mu, delta, and kappa opioid receptor binding sites were similar in membrane and digitonin-solubilized fractions (48-50% mu, 35-37% kappa, and 13-17% delta subtypes). The hydrodynamic properties of digitonin- and CHAPS-solubilized preparations were studied by sucrose density gradient centrifugation and Sepharose-6B chromatography. In all cases, two receptor populations were identified with the following parameters: sedimentation coefficients for the digitonin extracts were 9.2S and 13.2S and for CHAPS extract 8S and 15.6S; the Stokes radii were 45 A and 65A for the digitonin extract and 31A and 76A for the CHAPS-solubilized preparation.     

A simple test for enhanced guanyl nucleotide exchange in brain adenylate cyclase systems activated by neurotransmitters

The mechanism of receptor-induced activation of adenylate cyclase has been proposed to involve an enhanced exchange of GDP for GTP. The kinetics of this process have not been investigated so far in the brain due to a spontaneous activation of the enzyme by guanyl nucleotides, which precludes the ability to follow receptor-dependent events. We show that it is possible to investigate the mechanism of receptor action in such systems by using a combination of guanosine 5'-(beta-gamma-imino)triphosphate (Gpp(NH)p) and guanosine 5'-(2-O-thio)diphosphate (GDP beta S). In pineal membranes, beta-adrenergic agonists increase the rate of adenylate cyclase activation by 10 or 100 microM Gpp(NH)p about 40-fold (0.023-0.9 min-1 kact) and decrease the inhibitory potency of GDP beta S nearly 1000-fold. As a result, 100 microM GDP beta S which blocks 90% of the activation by 10 microM Gpp(NH)p has no inhibitory effect in the presence of 10 microM Gpp(NH)p and 10 microM noradrenaline or isoproterenol. In caudate nucleus, dopamine does not appear to increase the rate of activation of adenylate cyclase by 10 microM Gpp(NH)p. Nevertheless, 100 microM GDP beta S blocks 90% of the activation by 10 microM Gpp(NH)p but has no inhibitory effects in the presence of dopamine. Thus, one can demonstrate that even weakly activating receptors have the capacity to facilitate a functional exchange of GDP beta S for Gpp(NH)p and measure the efficacy of the interaction between the receptor and the functionally linked guanyl nucleotide subunit.     

Regulation of thyrotropin-releasing hormone binding by monovalent cations and guanyl nucleotides

Binding of thyrotropin-releasing hormone (TRH) to specific receptors on membranes isolated from GH4C1 pituitary cells was inhibited by monovalent cations and guanyl nucleotides. NaCl and LiCl inhibited TRH binding by 70%, with half-maximal inhibition at 30 mM; RbCl and KCl inhibited only 10% at concentrations up to 150 mM. NaCl decreased both the apparent number and the affinity of TRH receptors and increased the rate of dissociation of TRH from both membrane and Triton X-100-solubilized receptors. Guanyl nucleotides inhibited TRH binding up to 80%, with guanyl-5'-yl imidodiphosphate (Gpp(NH)p) approximately GTP much greater than GDP approximately ATP greater than GMP. GTP and Gpp(NH)p exerted half-maximal effects at 0.3 microM and decreased receptor affinity to one-third of control but did not change receptor number. Gpp(NH)p accelerated the dissociation of TRH from membranes but not from solubilized receptors. The effects of NaCl were independent of temperature, while GTP and Gpp(NH)p were much more inhibitory at 22 degrees C (70%) than at 0 degrees C (10%). Inhibition by NaCl could be reversed by washing the membranes, and inhibition by GTP was reversed if membranes were chilled to 0 degrees C. The inhibitory effects of low concentrations of NaCl and Gpp(NH)p were additive. Neither monovalent cations nor GTP prevented the TRH-receptor complex from undergoing transformation from a state with rapid dissociation kinetics to a slower dissociating form. The results suggest that sodium ion regulates TRH binding by interacting with a site on the receptor, while guanyl nucleotides regulate TRH binding indirectly.     

Unique guanine nucleotide binding properties of the human placental GTP-binding protein Gp

Gp is a major GTP-binding protein of human placenta and platelets [Evans, T., Brown, M. L., Fraser, E. D., & Northup, J. K. (1986) J. Biol. Chem. 261, 7052-7059]. High-affinity guanine nucleotide binding is associated with a polypeptide migrating identically with H-ras on SDS-PAGE. We have characterized the interactions of preparations of purified human placental Gp with guanine nucleotides in detergent solution. Equilibrium binding studies with [35S]GTP gamma S, [3H]Gpp(NH)p, and [3H]GTP identified a single class of sites with a dissociation constant of 10 +/- 1, 153 +/- 61, and 125 +/- 77 nM for the ligands, respectively. These three ligands were mutually competitive with Ki values consistent with the Kd values from direct binding experiments. Competition for the binding of [3H]Gpp(NH)p was used to determine the specificity of the site. Ki values determined from this assay were 14 nM for GTP gamma S, 143 nM for Gpp(NH)p, 3.3 microM for GDP beta S, 69 nM for GTP, and 64 nM for GDP. ATP, ADP, cAMP, cGMP, and NAD+ had no detectable affinity for this site. While the equilibrium binding data fit well to a single class of sites, association kinetics of these ligands were better fit to two rate constants. Dissociation kinetics, however, were not clearly resolved into two rates.(ABSTRACT TRUNCATED AT 250 WORDS)     

Differential regulation of high-affinity agonist binding to muscarinic sites in the rat heart, cerebellum, and cerebral cortex

The muscarinic agonist [3H]cismethyldioxolane ([3H]CD) was used to characterize the effects of regulators upon high-affinity agonist binding sites of the rat heart, cerebral cortex and cerebellum. Comparative studies with sodium ions (Na+), magnesium ions (Mg++), N-ethylmaleimide (NEM) and the guanine nucleotide Gpp(NH)p revealed tissue-specific effects. Mg++ preferentially enhanced while Gpp(NH)p and NEM reduced high-affinity [3H]CD binding in the heart and cerebellum. By comparison NEM enhanced high-affinity agonist binding in the cerebral cortex while Gpp(NH)p and Mg++ had little or no effect. Kinetic studies support an allosteric mechanism for these effects and provide further evidence for muscarinic receptor subtypes in mammalian tissues.     

Characterization of radiolabeled agonist binding to beta-adrenergic receptors in mammalian tissues

Recent evidence suggests that the molecular interactions of agonists with beta-adrenergic receptors differ from those of antagonists. Most of this evidence has come from studies of agonist inhibition of radiolabeled antagonist binding. We have examined agonist binding directly in rat lung membranes using radiolabeled hydroxybenzylisoproterenol (3H-HBI). Specific binding of 3H-HBI was stereoselective and was inhibited by catecholamines with a potency order characteristic of beta 2-adrenergic receptors. Gpp(NH)p increased the rates of association and dissociation of 3H-HBI from the receptor. In the absence of Gpp(NH)p, Scatchard plots were curvilinear suggesting a complex interaction of the agonist with the receptor. The total number of 3H-HBI binding sites was similar to that of 125I-IHYP binding sites. In the presence of increasing concentrations of Gpp(NH)p, the affinity of 3H-HBI was decreased and Scatchard plots became linear. Sodium chloride mimicked the effect of Gpp(NH)p in lowering the affinity of the receptor for 3H-HBI. Magnesium chloride had the opposite effect in that it promoted high affinity binding. The effect of sodium chloride was largely overcome by the presence of magnesium chloride.     

Properties of detergent-dispersed adenylate cyclase from cerebral cortex. Presence of an inhibitor protein

Adenylate cyclase was solubilized from washed particulate fraction of rabbit cerebral cortex with the nonionic detergent Lubrol 12A9 and subjected to either gel filtration on Ultrogel AcA 34 or chromatography on DEAE Bio-Gel A. By both procedures the enzyme was resolved into two components, one insensitive to guanyl 5'-yl imidodiphosphate [Gpp(NH)p] and NaF but stimulated by Ca2+ and calmodulin, and another that was sensitive to Gpp(NH)p and NaF but relatively insensitive to Ca2+ and calmodulin. The data support the possibility that two independent forms of adenylate cyclase exist in cerebral cortex, one regulated by guanine nucleotide regulatory protein and another by Ca2+-calmodulin. Fractions containing the guanylnucleotide-sensitive activity were found to contain a factor that inhibited basal and Ca2+-stimulated adenylate cyclase in the Ca2+-sensitive fraction. The inhibitor was inactivated by heating at 60 degrees C and by incubation with trypsin. Inhibition was not time-dependent, and it was not due to destruction of cAMP by phosphodiesterase or of ATP by ATPase. Inhibitory action was not reversed by calmodulin and therefore it does not appear to be a calmodulin binding protein. Sucrose density gradient sedimentation indicated a sedimentation coefficient of 4S for the inhibitor; by this technique it co-sedimented with the adenylate cyclase sensitive to Gpp(NH)p and NaF.     

5'-Guanylylimidodiphosphate decreases affinity for agonists and apparent molecular size of a frog brain opioid receptor in digitonin solution

When assayed for specific opiate binding in the presence of 120 mM NaCl, digitonin extracts from frog (Rana ridibunda) brain membranes were found to contain about the same quantity (0.5 pmol/mg of protein) of high (Kdh = 0.4 nM) and of lower (Kdl = 15-20 nM) affinity sites for the opiate agonist [3H]etorphine. The two classes of [3H]etorphine binding sites displayed equally high (Kd = 0.3 nM) affinity for the opiate antagonist [3H]diprenorphine. 5'-Guanylylimidodiphosphate (GppNHp) selectively and potently (IC50 = 0.1 microM) inhibited high affinity binding of the tritiated agonist, and this inhibition resulted from the GppNHp-induced conversion of the high into the lower affinity sites for [3H]etorphine. Following centrifugation of the digitonin extract in sucrose gradients, opioid binding activity was found to be associated with two clearly separated macromolecular components of apparent sedimentation coefficients 11.5 and 9.7 S, respectively. The two components bound [3H]diprenorphine equally well, whereas the fast sedimented component bound [3H]etorphine better than did the slower sedimented one. In addition, labeling of the component of bigger apparent size with [3H]etorphine was considerably reduced in the presence of 50 microM GppNHp. Finally, in soluble extracts which had been (i) preincubated with and (ii) centrifuged in the presence of GppNHp, the fast sedimented component was no longer observed while there was about twice as much of the component of smaller apparent size as in control (no GppNHp) extracts. Together, these results demonstrated the existence of an opioid receptor-G protein complex which, in digitonin solution, was still amenable to regulation (dissociation) by guanine nucleotides.