Sustained unilateral blockade of rat striatal M(1) muscarinic receptors with m1-toxin1 in vivo.

m1-Toxin1 is a trace component of the venom of the green mamba that antagonizes M(1) muscarinic receptors specifically and irreversibly in vitro. It was injected into the right caudatoputamen of Wistar rats to occlude M(1) receptors specifically for several days ("M(1) knockdown"), and to begin to establish the unknown effects of striatal M(1)-neurotransmission on movement. The extent and duration of M(1)-blockade were evaluated with receptor binding assays and light microscopic autoradiography, using 1-2 nM [3H]pirenzepine to assess unoccupied M(1) receptors. An injection of 27 pmol m1-toxin1 blocked almost all of the M(1) receptors in one caudatoputamen (4 pmol), followed by their slow recovery to supranormal levels (115%) during a week. During maximum M(1)-blockade, the binding of 1 nM [3H]N-methylscopolamine to striatal membranes was reduced by only half, indicating no blockade of M(4) receptors, which comprise 51% of striatal muscarinic receptors. It is concluded that m1-toxin1 can produce acute, focal, selective, unilateral and long-lasting M(1)-blockade for investigations of the effects of M(1)-neurotransmission in vivo. Rats with extensive unilateral striatal M(1)-blockade appeared normal, and did not show unilateral deficits in spontaneous forearm use, in contrast to rats with unilateral 6-hydroxydopamine lesions of the substantia nigra. Since M(1)-blockade should decrease the activity of both striatonigral and striatopallidal projection neurons, and the relative activity of these neurons is believed to control movement, the results suggest that a balanced decrease in the activity of these neurons on one side of the brain does not produce a right/left imbalance in spontaneous movement.

Site-directed mutagenesis of m1-toxin1: two amino acids responsible for stable toxin binding to M(1) muscarinic receptors.

m1-Toxin1 binds specifically and irreversibly to M(1) muscarinic receptors and can slow the dissociation of [(3)H]N-methylscopolamine ([(3)H]NMS) from these receptors. Yet only 7 of its 65 amino acids are not conserved in six other mamba toxins that bind reversibly to M(2)-M(5) muscarinic receptors. Two of these seven residues (Phe(38), Lys(65)) were mutated to corresponding residues of the other toxins (Ile(38), Glu(65)), to evaluate amino acids in m1-toxin1 that confer its remarkable affinity and specificity. The cDNA for m1-toxin1 was cloned from venom gland mRNA using polymerase chain reaction (PCR)-based techniques. Its nucleotide sequence is remarkably similar to those of other short-chain neurotoxins. The cDNAs for mutant toxins Phe(38) to Ile(38) (F38I) and Lys(65) to Glu(65) (K65E) were constructed by PCR-based techniques. Each cDNA was expressed in yeast, and the toxins were purified from yeast media by cation-exchange and reversed phase chromatography. Recoveries were 40 to 152 microg/l. Recombinant m1-toxin1 was identical to the native toxin (observed mass: 7471 Da; irreversible blockade of [(3)H]NMS binding to cloned M(1) receptors at 25 degrees C; no blockade of M(2)-M(5) receptors; 6-fold slowing of [(3)H]NMS dissociation at 37 degrees C). F38I also bound specifically to M(1) receptors, but reversibly and without effect on NMS dissociation. Thus, Phe(38) contributes to the stability of toxin-receptor complexes, but not to M(1)-selectivity. K65E bound selectively and irreversibly to unliganded M(1) receptors but did not slow NMS dissociation. It is suggested that the C-terminal Lys(65) of m1-toxin1 may contact an outer loop of the M(1) receptor.

Snake toxins that bind specifically to individual subtypes of muscarinic receptors.

This paper discusses the properties of the three most specific ligands found for the extracellular faces of M1, M2 and M4 muscarinic receptors (m1-toxin1, m2-toxin and m4-toxin, respectively). The primary goal of this paper is to show the known and potential usefulness of these toxins and their biotinylated, radioactive, fluorescent and mutated derivatives.

Biotinylated m4-toxin demonstrates more M4 muscarinic receptor protein on direct than indirect striatal projection neurons.

The striatum has nearly equal numbers of striatonigral and striatopallidal projection neurons. All are GABAergic and inhibitory, but they lie in separate neuronal circuits ('direct' and 'indirect', respectively) that appear to exert opposite effects on movement. Methods are needed to evaluate the function of each circuit. A potential way to control striatonigral neurons selectively is via M4 muscarinic receptors. The striatum has many more M4 receptors than other tissues, they are located on approximately half of all projection neurons, and mRNA for M4 receptors is prevalent only in striatonigral neurons. In order to more rigorously compare the distribution of M4 receptors on rat neurons in these pathways a toxin that binds with very high specificity to M4 receptors (m4-toxin) was biotinylated for use as a selective probe for M4 receptor protein. Pooled biotin-toxin complexes were found to retain high M4-specificity and affinity. Neurons were first labeled by retrograde transport of fluorescent microbeads (FluoSpheres) injected into the substantia nigra and globus pallidus. Coincident labeling of only 4% of the cells confirmed the validity of the retrograde labeling technique. Labeled neurons were probed for M4 receptor protein using biotinylated m4-toxin and fluorescent avidin. M4 receptors were found on 14% of indirect and 86% of direct neurons. It may be concluded that there is a relative abundance of M4 receptors controlling the direct pathway. This work supports the hypothesis that M4-selective drugs will prove useful to control the function of striatonigral neurons in the direct projection pathway.

m1-toxin isotoxins from the green mamba (Dendroaspis angusticeps) that selectively block m1 muscarinic receptors.

The venom of the green mamba, Dendroaspis angusticeps, was found to have at least six trace m1-specific toxins that block the binding of 3H-N-methylscopolamine to cloned m1 muscarinic receptors. Four were isolated by gel filtration, cation exchange HPLC and reversed-phase HPLC and named m1-toxins1-4. Recovery was 180, 90, 20 and 10 microg/g dry venom, respectively. m1-Toxin1 (the original m1-toxin) was found to have the sequence, L T C V K S N S I W F P T S E D C P D G Q N LC F K R W Q Y I S P R M Y D F T R G C A A T C P K A E Y R D V I N C C G T D K C N K, calculated mass = 7473 Da and calculated pI = 8.2. This sequence had been predicted previously from a cDNA cloned from the venom glands of this snake. The binding of m1-toxin1 was irreversible, so its Kd could not be determined. m1-Toxin2 differed only in proline-19, mass = 7455 and pI = 8.5. Partial sequence data for m1-toxin3 showed aspartate-7 and m1-toxin4 showed isoleucine-12, asparagine-16 and alanine-19. m1-Toxins1-4 have seven conserved amino acids not found in homologous mamba toxins that bind to other muscarinic receptors (MT1, MT2, m4-toxin = MT3, MT4, MT5, MTalpha and MTbeta). Some of these residues may be essential for m1-specificity. Since m1-toxin1 binds irreversibly in artificial cerebrospinal fluid at 37 degrees C, it is a particularly attractive antagonist for studies in vivo.

Muscarinic-induced modulation of potassium conductances is unchanged in mouse hippocampal pyramidal cells that lack functional M1 receptors.

Activation of muscarinic acetylcholine (ACh) receptors (mAChRs) increases excitability of pyramidal cells by inhibiting several K+ conductances, including the after-hyperpolarization current (Iahp), the M-current (Im), and a leak K+ conductance (Ileak). Based on pharmacological evidence and the abundant localization of M1 receptors in pyramidal cells, it has been assumed that the M1 receptor is responsible for mediating these effects. However, given the poor selectivity of the pharmacological agents used to characterize these mAChR responses, rigorous characterization of the receptor subtypes that mediate these actions has not been possible. Surprisingly, patch clamp recording from CA1 pyramidal cells in M1 knockout mice revealed no significant difference in the degree of inhibition of Iahp, Im, or Ileak by the mAChR agonist, carbachol (CCh), as compared with wildtype controls. In addition, the M1-toxin was not able to block CCh's inhibition of the Iahp, Im, or Ileak These data demonstrate that the M1 receptor is not involved in increasing CA1 pyramidal cell excitability by mediating ACh effects on these K+ conductances.

m2-toxin: A selective ligand for M2 muscarinic receptors.

Selective ligands are needed for distinguishing the functional roles of M2 receptors in tissues containing several muscarinic receptor subtypes. Because the venom of the green mamba Dendroaspis angusticeps contains the most specific antagonists known for M1 and M4 receptors (m1-toxin and m4-toxin), it was screened for toxins that inhibit the binding of [(3)H]N-methylscopolamine ([(3)H]NMS) to cloned M2 receptors. Desalted venom had as much anti-M2 as anti-M4 activity. The most active anti-M2 toxin in the venom was isolated by gel filtration, cation-exchange chromatography, and reversed-phase HPLC, and called m2-toxin. Spectrometry yielded a mass of 7095 Da, and N-terminal sequencing of 53 amino acids showed RICHSQMSSQPPTTTFCRVNSCYRRTLRDPHDPRGT-IIVRGCGCPRMKPGTKL. This sequence is more homologous to antinicotinic than antimuscarinic toxins, but it lacks three almost invariant residues of antinicotinic toxins required for their activity. m2-Toxin fully blocked the binding of [(3)H]NMS and [(3)H]oxotremorine-M to M2 receptors with Hill coefficients near 1, and blocked 77% of the binding sites for 0.1 nM [(3)H]NMS in the rat brainstem (K(i) = 11 nM). Concentrations that fully blocked cloned M2 receptors had no effect on M4 receptors, but slightly increased [(3)H]NMS binding to M1 receptors, an allosteric effect. In accord with these results, light microscopic autoradiography of the rat brain showed that m2-toxin decreased [(3)H]NMS binding in regions rich in M2 receptors and increased binding in regions rich in M1 receptors. Thus m2-toxin is a novel M2-selective, short-chain neurotoxin that may prove useful for binding and functional studies.

Muscarinic receptor subtypes involved in hippocampal circuits.

Muscarinic receptors modulate hippocampal activity in two main ways: inhibition of synaptic activity and enhancement of excitability of hippocampal cells. Due to the lack of pharmacological tools, it has not been possible to identify the individual receptor subtypes that mediate the specific physiological actions that underlie these forms of modulation. Light and electron microscopic immunocytochemistry using subtype-specific antibodies was combined with lesioning techniques to examine the pre- and postsynaptic location of m1-m4 mAChR at identified hippocampus synapses. The results revealed striking differences among the subtypes, and suggested different ways that the receptors modulate excitatory and inhibitory transmission in distinct circuits. Complementary physiological studies using m1-toxin investigated the modulatory effects of this subtype on excitatory transmission in more detail. The implications of these data for understanding the functional roles of these subtypes are discussed.

Activation of the genetically defined m1 muscarinic receptor potentiates N-methyl-D-aspartate (NMDA) receptor currents in hippocampal pyramidal cells.

Evidence suggests that cholinergic input to the hippocampus plays an important role in learning and memory and that degeneration of cholinergic terminals in the hippocampus may contribute to the memory loss associated with Alzheimer's disease. One of the more prominent effects of cholinergic agonists on hippocampal physiology is the potentiation of N-methyl-D-aspartate (NMDA)-receptor currents by muscarinic agonists. Here, we employ traditional pharmacological reagents as well as m1-toxin, an m1 antagonist with unprecedented selectivity, to demonstrate that this potentiation of NMDA-receptor currents in hippocampal CA1 pyramidal cells is mediated by the genetically defined m1 muscarinic receptor. Furthermore, we demonstrate the colocalization of the m1 muscarinic receptor and the NR1a NMDA receptor subunit at the electron microscopic level, indicating a spatial relationship that would allow for physiological interactions between these two receptors. This work demonstrates that the m1-muscarinic receptor gene product modulates excitatory synaptic transmission, and it has important implications in the study of learning and memory as well as the design of drugs to treat neurodegenerative diseases such as Alzheimer's.

Use of antimuscarinic toxins to facilitate studies of striatal m4 muscarinic receptors.

Striatal m4 muscarinic receptors are important because their blockade controls movement, and they are preferentially located on striatal neurons that project to the internal globus pallidus. The following studies were performed in vitro to provide a basis for using antimuscarinic toxins to study the effects of selective m4 blockade on movement in vivo. Because m4-toxin has limited selectivity alone (102-fold higher affinity for m4 than m1 receptors), m1-toxin was used first to occlude m1 receptors selectively, fully and irreversibly. It blocked 42% of the sites for 1.0 nM 3H-N-methylscopolamine in rat striatal membranes and 43% in sections of cat striatum. m4-Toxin (>500-fold higher affinity for m4 than m2, m3 or m5 receptors) blocked 88% of the residual, non-m1 sites in membranes, showing 64 pmol m4 receptors/g tissue. In comparison, AFDX-116, biperiden, clozapine, gallamine, hexahydrodifenidol, himbacine, R(+)hyoscyamine, methoctramine, pirenzepine, silahexocyclium, trihexyphenidyl and tripitramine did not distinguish m4 from other non-m1 receptors. 3H-Pirenzepine dissociated twice as rapidly from non-m1 as m1 receptors. Autoradiography was used to test the idea that m4 receptors are localized preferentially in the striosomes of the cat striatum. Non-m1 receptors were distributed equally in striosomes and matrix, indicating that striatal neurons with m4 receptors are in both compartments. Thus m1-toxin facilitates studies of m4 receptors by occluding m1 receptors, and m4-toxin is a selective antagonist for residual m4 receptors.

Non-selectivity of the monoclonal antibody M35 for subtypes of muscarinic acetylcholine receptors.

The monoclonal antibody M35, one of the first monoclonal antibodies successfully raised against muscarinic acetylcholine receptors, has been widely used to study the distribution of this protein in a variety of tissues and cell types of different species. It is not fully known, however, to which muscarinic acetylcholine receptor subtypes M35 binds. Knowledge of subtype-selectivity of M35 is a necessary step towards a functional interpretation of the obtained immunocytochemical data. The aim of the present study was to determine the subtype-selectivity of M35 employing transfected CHO-K1 cells stably expressing human m1-m5 muscarinic acetylcholine receptors separately, and to study M35 immunoreactivity in areas of rat central and peripheral tissues known to be specifically enriched in a single muscarinic acetylcholine receptor subtype. The results show that (a) all five transfected cell lines were immunopositive for M35, (b) nontransfected control cells were immunonegative, (c) the number of mAChRs expressed per cell correlated positively with the intensity of M35 immunoreactivity, and (d) cell types in aldehyde-fixed rat tissue enriched in a single m1-m4 subtypes revealed clear M35 immunoreactivity. Taken together, the present results show that M35 does not discriminate between muscarinic acetylcholine receptor subtypes. Evidently, the epitope of M35 on the receptor-protein is preserved on all muscarinic acetylcholine receptor subtypes. The epitope for M35 must, therefore, be localized on a homologous part of each subtype.

Anti-muscarinic toxins from Dendroaspis angusticeps.

Toxins from the venom of the African green mamba, Dendroaspis angusticeps, fulfill a major need for selective ligands for some of the five genetically defined subtypes of muscarinic acetylcholine receptors (m1-m5). Two toxins have been found that are highly selective antagonists for m1 and m4 receptors (m1-toxin and m4-toxin, respectively). Two other toxins (MT1 and MT2) bind with high affinity to both m1 and m4 receptors, and are agonists. Components of the venom also modify the binding of radiolabeled antagonists to m2 receptors, but an m2-selective toxin has not yet been isolated, m1-Toxin can bind to m1 receptors at the same time as typical competitive antagonists, suggesting that this toxin binds to the N-terminal and outer loops of m1 receptor molecules, rather than within the receptor pocket where typical agonists and antagonists bind. The binding of toxins to the outer parts of receptor molecules probably accounts for their much higher specificity for individual receptor subtypes than is seen with smaller ligands. Toxins are useful for identifying, counting, localizing, activating and blocking m1 and m4 receptors with high specificity.

Identification of muscarinic receptor subtypes in mouse parotid gland.

Immunoprecipitation of muscarinic receptors from mouse parotid membranes by specific subtype antisera showed that M3 and M1 receptors represented 75 and 15% of the total number of precipitable receptors, respectively. [N-methyl-3H]methylscopolamine (NMS) labeled a single class of high-affinity binding sites in membranes from parotid glands with a dissociation constant of 0.67 +/- 0.02 nM and a maximum binding capacity of 176 +/- 15 fmol/mg protein. Competition curves for NMS, atropine, 4-diphenylacetoxy-N-methylpiperidine methiodide (4-DAMP) and para-fluoro-hexahydro-sila-difenidol fit best to a one-site binding model, whereas pirenzepine and methoctramine fit best to a two-site binding model, indicating 76-90% M3 receptors. Results from the use of pirenzepine indicated that the second mouse parotid receptor subtype, unlike that of the submandibular gland, has atypical characteristics for an M1 receptor. The rank order of potency of muscarinic antagonists in inhibiting phosphoinositide turnover and biphasic effects of carbachol on isoproterenol-stimulated adenosine 3',5'-cyclic monophosphate (cAMP) accumulation was atropine > or = 4-DAMP >> pirenzepine > AF-DX 116. A specific M1 antagonist, m1-toxin, had no effect on carbachol augmentation or inhibition of isoproterenol responses. Results suggest that M3 receptors couple to both augmentation and inhibition of stimulated cAMP levels.

Stable allosteric binding of m1-toxin to m1 muscarinic receptors.

m1-Toxin was found to slow the dissociation of [3H]N-methyl-scopolamine (NMS) and [3H]pirenzepine from m1 muscarinic receptors expressed in the membranes of Chinese hamster ovary cells. When toxin-NMS-receptor complexes were formed in membranes and then dissolved in digitonin, or when these complexes were formed in solution, the toxin completely stopped the dissociation of [3H]NMS for 6 hr at 25 degrees C. Toxin-receptor complexes formed in membranes or in solution were also highly stable in solution at 25 degrees, as shown by the ability of the toxin to prevent the binding of [3H]quinuclidinyl benzilate (QNB). [3H] QNB-receptor complexes were equally stable, whereas unliganded soluble receptors lost most of their ability to bind QNB within an hour. Toxin-receptor complexes could be partially dissociated by incubation at 37 degrees in the presence of digitonin and [3H]QNB, and the freed receptors were then labeled. These results demonstrate that m1-toxin binds allosterically and pseudoirreversibly to m1 receptors, and that the toxin can stabilize the outward-facing pocket of m1 receptors which contains and binds competitive antagonists. The allosteric nature of the binding of m1-toxin should prove to be useful for such unusual purposes as stabilizing the binding of readily reversible and/or nonselective ligands specifically to m1 receptors, for purifying labeled or unlabeled receptors by affinity techniques which recognize the toxin, for recognizing receptors with genetically or biochemically altered primary binding sites, and for stabilization of the native conformation of m1 receptors for structural studies.

Use of m1-toxin as a selective antagonist of m1 muscarinic receptors.

m1-Toxin is the only ligand which is known to bind specifically to the extracellular face of genetically defined m1 muscarinic receptors; it binds pseudoirreversibly. A variety of studies were performed to evaluate the usefulness of m1-toxin as a selective antagonist of m1 receptors. Exposure of slices of the rat cerebral cortex to m1-toxin in physiological buffer blocked the subsequent binding of 1.0 nM [3H]pirenzepine to m1 receptors in the slices. The toxin also blocked 70% of carbachol-stimulated turnover of radiolabeled inositol phosphates in hippocampal slices. Autoradiographs showed that m1-toxin bound to sections of once-frozen tissue and blocked the binding of [3H]quinuclidinyl benzilate to regions of the rat brain rich in m1 receptors. The toxin blocked the binding of [3H]antagonists to pure m1 receptors on the surface of living Chinese hamster ovary cells, but did not block intracellular receptors. In membrane preparations from the rat cortex and hippocampus the toxin blocked the binding of [3H] antagonists to m1 receptors quantitatively and selectively, but had no effect on binding sites for [3H]nicotine. Subsaturating amounts of the toxin bound to m1 receptors in membranes at 4 degrees C in less than 30 sec. Low concentrations of m1-toxin blocked m1 receptors in solution in digitonin but had no effect on separate preparations of pure m2, m3, m4 or m5 receptors. Thus m1-toxin appears to be a very useful antagonist for m1 receptors in intact tissue, on isolated cells, in membranes and in solution, in a variety of media.

Purification and properties of m1-toxin, a specific antagonist of m1 muscarinic receptors.

The venom of the Eastern green mamba from Africa, Dendroaspis angusticeps, was found to block the binding of 3H-quinuclidinyl benzilate to pure m1 and m4 muscarinic ACh receptors expressed in Chinese hamster ovary cells. The principal toxin in the venom with anti-m1 muscarinic activity was purified by gel filtration and reversed-phase HPLC. This toxin has 64 amino acids, a molecular mass of 7361 Da, and an isoelectric point of 7.04. Its cysteine residues are homologous with those in curare-mimetic alpha-neurotoxins, and with those in fasciculin, which inhibits AChE. At low concentrations the toxin blocked m1 receptors fully and pseudoirreversibly while having no antagonist activity on m2-m5 receptors; the toxin is therefore named "m1-toxin." At higher concentrations m1-toxin interacted reversibly with m4 receptors, and half of the toxin dissociated in 20 min at 25 degrees C. The affinity of m1-toxin is therefore much higher for m1 than for m4 receptors. By comparison with m1-toxin, pirenzepine has sixfold higher affinity for m1 than for m4 receptors. Autoradiographs of muscarinic receptors in the rat brain demonstrated that m1-toxin blocked the binding of 2 nM 3H-pirenzepine only in regions known to bind m1-specific antibodies. Thus, m1-toxin is a much more selective ligand than pirenzepine for functional and binding studies of m1 muscarinic receptors.

m1-toxin.

The venom of the Eastern green mamba from Africa, Dendroaspis angusticeps, contains a number of toxins which block the binding of 3H-antagonists to genetically-defined m1 and m4 muscarinic acetylcholine receptors. Most of the anti-muscarinic activity of the venom is due to the presence of a newly-isolated toxin, "m1-toxin", which has 64 amino acids and a molecular mass of 7361 Daltons. At present m1-toxin is the only ligand which is known to be capable of fully blocking m1 receptors without affecting m2-m5 receptors. It binds very rapidly, specifically and pseudoirreversibly to the extracellular face of m1 receptors on cells, in membranes or in solution, whether or not the primary receptor site is occupied by an antagonist. Bound toxin can either prevent the binding and action of agonists or antagonists, or prevent the dissociation of antagonists. The toxin is useful for identifying m1 receptors during anatomical and functional studies, for recognizing and stabilizing receptor complexes, and for occluding m1 receptors so that other receptors are more readily studied.

Effect of chronic cholinergic denervation on the m1 muscarinic receptor mechanism.

The first three parts of the transduction mechanism for m1 muscarinic receptors (m1 receptors, receptor-G protein coupling, and the activation of phospholipase C) were studied in the rat hippocampus following unilateral or bilateral surgical lesions of the fimbria/fornix. One nM 3H-pirenzepine was used to label m1 receptors selectively. No changes in m1 receptor numbers were found between age 1.7 and 29 months old during normal aging or one year after cholinergic denervation. The interaction between m1 receptors and their associated G protein was examined by competition between 1 nM 3H-pirenzepine and oxotremorine-M in the presence and absence of a guanine nucleotide. The percentage of guanine nucleotide-sensitive high affinity binding sites for the agonist was similar in rats 1.7-29 months old and in rats 1 year after denervation. The ability of oxotremorine-M to activate phospholipase C, via m1, m3, and m5 receptors was also unchanged more than a year after cholinergic denervation of the hippocampus. We concluded that the initial steps in the m1 receptor transduction mechanism remain remarkably stable after denervation.

Evidence of paired M2 muscarinic receptors.

Binding assays involving various antagonists, including N-[3H] methylscopolamine, [3H]quinuclidinyl benzilate, AFDX-116, pirenzepine, and propylbenzilylcholine mustard, disclosed only a single population of M2 muscarinic receptors in membranes from the rat "brainstem" (medulla, pons, and colliculi). However, competition curves between N-[3H]methylscopolamine and various agonists, including oxotremorine, cis-dioxolane, and acetylethylcholine mustard, showed approximately equal numbers of guanine nucleotide-sensitive high affinity (H) sites and guanine nucleotide-insensitive low affinity (L) sites. This 50% H phenomenon persisted in different buffers, at different temperatures, after the number of receptors was halved (and, thus, the remaining receptor to guanine nucleotide-binding protein ratio was doubled), after membrane solubilization with digitonin, and when rabbit cardiac membranes were used instead of rat brainstem membranes. Preferential occupation of H sites with acetylethylcholine mustard, and of L sites with quinuclidinyl benzilate or either mustard, yielded residual free receptor populations showing predominantly L and H sites, respectively. Low concentrations of [3H]-oxotremorine-M labeled only H sites, and the Bmax for these sites was 49% of the Bmax found with [3H]quinuclidinyl benzilate plus guanine nucleotide. These and other results are most consistent with the idea that H and L receptor sites exist on separate but dimeric receptor molecules and with the hypothesis that only the H receptors cycle between high and low affinity, depending upon interactions between this receptor molecule and a guanine nucleotide-binding protein.