Meeting Report for: The 4th International Symposium on Subtypes of Muscarinic Receptors

The regulation of cellular signal transduction and growth by four human muscarinic acetylcholine receptor (mAChR) subtypes has been studied comparatively. The four mAChRs fall into two functional sub-groups, based on their primary effects on second messenger formation; two of the receptors strongly inhibit adenylyl cyclase activity, whereas the other two strongly stimulate PI hydrolysis. Studies on mAChR regulation of two cellular events involved in cellular growth regulation, the transcription of proto-oncogene c-fos and DNA synthesis, indicate that these events are efficiently activated by those mAChRs which couple primarily to phospholipase C.

Ionizable groups on the cardiac M2 muscarinic receptor which regulate the binding of ligands have been examined by studying the pH dependence of the ligand affinity constants. The presence of three titratable residues (approximate pK values, 5.4, 6.8 and 7.5) whose protonation modulates antagonist binding has been demonstrated. Cardioselective antagonists are selectively affected by the protonation state of the pK 6.8 residue, whereas the binding of antagonists having differing selectivities is more strongly affected by protonation of the pK 5.4 residue on cardiac receptors. Methoctramine is capable of binding to both the pK 5.4 and 6.8 residues simultaneously. Protonation of the residue of highest pK produces a conformational change at the receptor which can affect both agonist and antagonist binding. It is now possible to demonstrate differences both in the way ligands bind to a given receptor subtype and in the way a given ligand binds to different subtypes.

Human and rat genes and/or cDNAs for five different mAChRs have been cloned. The m1, m2 and m3 receptors correspond most closely to the pharmacologically defined M1, M2 and M3 receptors and are expressed in both brain and peripheral tissues. The m4 and m5 receptors are previously unrecognized pharmacological subtypes whose mRNAs are found predominantly in brain. Other less related but uncharacterized genes could represent additional subtypes. The properties of the five receptors and their genes are reviewed and their implications for future research are discussed.

Since the amino acid sequences for several muscarinic and adrenergic receptors are known, an attempt was made to correlate side chain hydrophobicity surrounding the aspartate anionic groups with binding structure-activity relationships. No positive correlation was found, suggesting that secondary binding effects of head group substituents are non-local. Displacement of aspartate ionization is unlikely to be due to such neighbour effects.

A wide variety of receptors for hormones, neurotransmitters and growth factors control cell function via the GTP-dependent activation of phosphoinositide specific phospholipase C (PIC). At least two distinct GTP dependent proteins (G proteins) have been implicated in coupling different receptor populations to the activation of PIC and five immunologically distinct isozymes of PIC have been purified to homogeneity, prompting speculation about the potential for multiple modes of organization of the participants in this signal transduction pathway. The mechanism of hormone and G protein-dependent regulation of PIC has been studied in detail using [3H]inositol labelled turkey erythrocyte membranes and these experiments have provided strong support for the involvement of a heterotrimeric G protein. Further progress requires the development of reconstitution assays in which the regulation of isolated PICs by defined G proteins can be demonstrated.

The discovery of the M1-selective receptor antagonist pirenzepine was the impetus for a research project directed towards the development of selective muscarinic antagonists. In the pursuit of this objective, compounds with different selectivity profiles have been found. AF-DX 116 was the first cardioselective antagonist synthesized. Subsequently novel M2 receptor antagonists have been discovered with higher potency and selectivity. Moreover, a pirenzepine-type compound UH-AH 37 has been identified that, in contrast to pirenzepine, shows a higher affinity for ileal than for atrial muscarinic receptors. Among tricyclic muscarinic receptor antagonists three different selectivity profiles have been identified, namely: M1 greater than M3 greater than M2, Msm for pirenzepine; M2 greater than M1 greater than M3, Msm for AF-DX 116, AF-DX 384, AQ-RA 741; and Msm congruent to M1 greater than M2, M3 for UH-AH 37 and its (+) enantiomer.

Attempts have been made by means of recombinant DNA technology to understand the molecular basis of the functional heterogeneity of the muscarinic acetylcholine receptor (mAChR). Molecularly defined mAChR subtypes have been produced from the cloned DNAs in Xenopus oocytes and NG108-15 neuroblastoma-glioma hybrid cells as transient and stable expression systems, respectively, and agonist-induced cellular responses have been examined. The results obtained provide evidence that mAChR subtypes are selectively coupled with different effector systems, albeit not exclusively.

Clinical experience with muscarinic agonists in the symptomatic treatment of Alzheimer's disease includes studies of the effects of pilocarpine, arecoline, bethanechol, oxotremorine and RS 86. Although the results are somewhat conflicting, there is evidence that a subgroup of patients may respond with an improvement of cognitive and/or behavioural function. The existing agents tend to induce adverse effects due to the stimulation of peripheral muscarinic receptors. Furthermore they reduce (at least in vitro) acetylcholine release by an action on presynaptic receptors. Strategies to overcome these problems include the development of potent agonists with high blood-brain barrier penetration, the search for agents selective for muscarinic receptor subtypes (using cloned receptors as tools) and the identification of agents acting as presynaptic receptor antagonists, to increase acetylcholine release.

There is now substantial evidence that acetylcholinesterase inhibitors and muscarinic receptor agonists increase the pain threshold after both systemic and spinal administration. In rats, physostigmine gave a significant dose-dependent increase in latency times in the tail immersion test following intrathecal administration. The effect was antagonized with atropine. Neostigmine gave more prolonged latencies as did the muscarinic receptor agonist carbachol. Spinal cholinergic pathways for antinociception interacted with the spinal opioid and adrenergic nerve tracts. No cross-tolerance to the selective alpha 2-adrenoreceptor agonist guanfacine or to morphine was seen in rats tolerant of spinal carbachol antinociception. The mechanism of spinal cholinergic antinociception is not known but a muscarinic interneuron may explain the interactions with other neurotransmitters. Clinically, the centrally active cholinesterase inhibitor physostigmine has been shown to give postoperative pain relief although of short duration. Severe neurogenic pain has been successfully treated with physostigmine or distigmine.

A conserved aspartic acid residue in transmembrane helix 3 of the muscarinic acetylcholine receptors is important in binding the headgroup of muscarinic ligands. This acidic amino acid probably points into a relatively hydrophilic cavity whose walls are formed by the amphipathic transmembrane helices of the receptor. Amino acid side chains within this cavity contribute to ligand binding.

A series of hexahydro-difenidol (HHD) and hexahydro-sila-difenidol (HHSiD) analogues modified in the amino group, the phenyl ring and in the alkylene chain were investigated for their binding and functional properties at muscarinic M1, M2 and M3 receptors. Novel muscarinic receptor antagonists were obtained which exhibited different receptor selectivity profiles from the parent compounds HHD and HHSiD (M1 congruent to M3 greater than M2), e.g. HHD and HHSiD methiodides, M1 greater than M2 congruent to M3; p-fluoro-HHSiD, M3 greater than M1 greater than M2; trans-hexbutenol, M1 greater than M3 greater than M2; and (s)-p-fluoro-hexbutinol, M3 greater than M2 congruent to M1. Stereoselectivity ratios [(R)/(S)] for the enantiomers of HHD, hexbutinol and p-fluoro-hexbutinol were highest at M1, intermediate at M3 and lowest at M2 receptors.

Muscarinic acetylcholine receptor subtypes m1, m3 and m5 couple strongly to phosphatidylinositol turnover and hence to intracellular Ca2+ concentration via pertussis toxin (PTX) sensitive and insensitive G proteins. The m2 and m4 muscarinic receptor subtypes strongly inhibit adenylyl cyclase production via PTX sensitive G proteins. Additionally, the cardiac M2 receptor is closely coupled to a K+ current (IK.ACh). To characterize this functional diversity more completely, we measured the ACh-induced Ca2+ responses of cells transfected with the muscarinic receptor subtypes m1, m2, m3 and m4. As expected, cells transfected with m1 or m3 receptors exhibited large dose-dependent increases in Ca2+ in response to ACh application. Unexpectedly, cells transfected with m2 or m4 receptors also exhibited increases in Ca2+ in response to agonist application. The m2- or m4-coupled responses were smaller in amplitude, required higher concentrations of agonist and were much more sensitive to PTX treatment when compared to m1- or m3-coupled responses. We discuss this remarkable diversity of function in terms of the receptor subtype's coupling to G proteins.

Muscarinic receptors have been identified in the airways in several species, including humans, located on airway smooth muscle, secreting cells and on the nerves. M1 receptors are found in sympathetic ganglia in the guinea-pig and in parasympathetic ganglia in humans. M2 receptors (inhibitory autoreceptors) are found in cholinergic parasympathetic nerve terminals in many species, including humans, whereas the muscarinic receptors found on airway smooth muscle and mucus glands belong to the M3 subtype. It is possible that a defect in neuronal M2 receptor function may explain beta-blocker-induced asthma. M2 antagonists such as methoctramine are promising tools for elucidating the role of muscarinic receptor subtypes in the lung. However, they can potentially increase acetylcholine release. This property is not shown by drugs with a higher selectivity for M1 and M3 receptors which are likely to be useful clinically in the treatment of airway disease.

The possibility that polymethylene tetraamines act as divalent ligands has been explored. Structure-activity relationship studies among polymethylene tetraamines have shown that four nitrogens are necessary for high affinity binding to M2 receptors while being less important for M3 muscarinic receptors. Replacement of one terminal methoxybenzyl group of the potent and selective muscarinic antagonist methoctramine by different moieties led to weaker antagonists suggesting that the two terminal nitrogens of methoctramine interact with two similar receptor sites. Data are presented which suggest that methoctramine might interact with four acidic residues of the receptor: two residues are buried in the third transmembrane segment whereas the others are located extracellularly on the loop 4-5 which may represent the allosteric site where several antagonists such as gallamine bind. An hypothetical model describing the interaction of methoctramine with the M2 receptor is proposed. It may provide a useful working hypothesis for the design of new selective muscarinic ligands.

Effects of atropine and of the subtype selective mAChR antagonists pirenzepine (PZ) and AF-DX 116 were studied in humans. Dose- or time-response curves were established for heart rate and salivary flow. Plasma samples were drawn in parallel with the effect measurements and analysed for drug concentrations. Subtype-selective radioreceptor assays of the samples served to estimate the respective receptor occupancy in vivo. It is shown that low doses of PZ (M1-selective blockade) cause cholinomimetic effects indicated by bradycardia and increase in salivary flow. After high doses of PZ or atropine, tachycardia and inhibition of salivary flow are observed in parallel with occupancy of both the M2 and M3 subtypes. AF-DX 116 induces a tachycardia together with an increased salivary flow in agreement with its selectivity profile (M2 greater than M1 greater than M3). The diagnostic and therapeutic applications of M1- or M2-selective blockade by low dose PZ or AF-DX 116 respectively are discussed.

Much evidence has clearly revealed the existence of marked cholinergic deficits in cortical and hippocampal areas in Alzheimer's disease. Although not necessarily of etiological origin, these deficits have been associated with learning and memory disabilities observed in this neurogenerative disorder. We report here that in addition to deficits in choline acetyltransferase (ChAT) activity, the maximal densities of high affinity [3H]acetylcholine and [3H]AF-DX 116 (possibly M2), but not M1 muscarinic receptor binding sites are decreased in cortex and hippocampus in Alzheimer's disease. Similar findings are also observed in Parkinson's disease with Alzheimer's type dementia. Additionally, animal studies suggest that a population of M2 receptors is presynaptically located on cholinergic nerve terminals where they can act as negative autoreceptors to decrease acetylcholine release. Interestingly, blockade of these sites facilitates acetylcholine release and learning in rats. This may be relevant for the design of more appropriate therapeutic approaches toward the treatment of certain symptoms of Alzheimer's disease.

In the present study we describe a novel series of oxadiazole based tertiary amines which include the most efficacious and potent muscarinic ligands known. These compounds possess physicochemical characteristics which enable rapid equilibration into the CNS and are able to fully activate cortical muscarinic receptors. Data obtained from this series have allowed us to propose a pharmacophoric model which distinguishes high and low affinity state binding. This in turn has led us to suggest that agonists and antagonists may bind at two independent sites on the receptor protein and to speculate on the steps putatively involved in agonist-induced receptor activation.

The regulation of the number and function of the muscarinic receptors has been investigated in cultured chick cardiac cells and in cells expressing cloned genes encoding mammalian, Drosophila, and chick muscarinic receptors. A serum-free defined medium for the culture of chick embryonic heart cells has been used to study the regulation of mAChR number and function by serum lipoproteins. Addition of rooster high density lipoprotein to the culture medium results in an attenuation of muscarinic receptor-mediated inhibition of cAMP accumulation without a change in the number of receptors or inhibitory G proteins. Clones encoding the mouse m1 receptor and a homologous receptor from Drosophila have been isolated. When expressed in Y1 adrenal cells, both receptors stimulate phosphoinositide hydrolysis but do not inhibit cAMP accumulation. Deletion of 123 out of the 156 amino acids in the third cytoplasmic loop of the mouse m1 receptor does not impair its ability to stimulate phosphoinositide hydrolysis. A genomic clone encoding a muscarinic receptor expressed in chick heart has been isolated. When expressed in Y1 cells, it causes inhibition of cAMP accumulation but does not stimulate phosphoinositide hydrolysis.

The beta-adrenergic receptor (beta AR), which has been extensively characterized pharmacologically, serves as a useful model system for the analysis of the structure-function relationships of G protein-coupled receptors. Genetic and biochemical analysis has revealed that the ligand binding domain of the receptor involves residues within the hydrophobic transmembrane core of the protein. Molecular substitution experiments suggest that adrenergic agonists and antagonists are anchored to the receptor through an ionic interaction between Asp113 in the third hydrophobic region of the receptor and the protonated amine group of the ligand. In addition, catecholamine agonists are bound through hydrogen bonding interactions between two serine residues in the fifth hydrophobic domain of the receptor and the catechol hydroxyl groups of the ligand. Agonist-mediated activation of the G protein Gs requires residues within the cytoplasmic loop linking the fifth and sixth transmembrane helices which are predicted to form amphipathic alpha-helices. The strong structural similarities among G protein-coupled receptors imply that the information gained from genetic analysis of the beta AR should be applicable to other hormone and neurotransmitter receptors of this class.

Muscarinic agonists open potassium-selective K(ACh) channels in cardiac myocytes of pacemaker or atrial origin. Receptor activation is coupled to channel opening by a membrane bound guanine nucleotide-binding protein (GK) through a process that does not require cytoplasmic intermediates. We have used the muscarinic potassium channel and the corresponding macroscopic current, IACh, as rapid, sensitive and specific indicators of the state of activation of GK. This approach, developed here in quantitative detail, allowed us to identify the salient kinetic processes involved in the activation and deactivation of GK in vivo. Agonist was found to act by accelerating the rate of GDP release, and the subsequent GTP uptake by GK, while deactivation was found to occur by a process that requires GTP hydrolysis. Unexpectedly, deactivation in the intact system is much more rapid than the rate of GTP hydrolysis by isolated G proteins, suggesting the presence of a GTPase-stimulating factor in intact cells.

The stereoselectivity of the interaction with muscarinic receptors of enantiomers of a series of chiral antagonists is receptor subtype dependent. There is no overall relationship between stereoselectivity and receptor affinity. Depending on the antagonist studied, receptor stereoselectivity may indeed reflect: (1) the weakening or loss of a single interaction involving one of the four groups bound to the asymmetric carbon; (2) steric hindrance preventing optimum interaction of the low affinity steroisomer with the receptor; and/or (3) the inversion of the relative positions of two moieties of the ligand with similar structural and electronic properties i.e. comparable affinities for the two corresponding subsites in the receptor.