Allosteric modulation of the muscarinic M4 receptor as an approach to treating schizophrenia.

Current antipsychotics provide symptomatic relief for patients suffering from schizophrenia and related psychoses; however, their effectiveness is variable and many patients discontinue treatment due to side effects. Although the etiology of schizophrenia is still unclear, a leading hypothesis implicates an imbalanced dopaminergic system. Muscarinic acetylcholine (ACh) receptors regulate dopamine levels in key areas of the brain involved in psychosis, with the M(4) subtype emerging as a key regulator of dopaminergic hyperactivity. Unfortunately, no selective small molecule tools exist to provide pharmacological validation of this hypothesis. Here, we describe the discovery of a small molecule modulator, LY2033298, that is highly selective for human M(4) receptors by virtue of targeting an allosteric site on this receptor. Pharmacological assays confirmed the selectivity of LY2033298 for the M(4) receptor and revealed the highest degree of positive allosteric enhancement of ACh potency thus far identified. Radioligand binding assays also show this compound to directly potentiate agonist binding while having minimal effects on antagonist binding. Mutational analysis identified a key amino acid (D(432)) in the third extracellular loop of the human M(4) receptor to be critical for selectivity and agonist potentiation by LY2033298. Importantly, LY2033298 was active in animal models predictive of clinical antipsychotic drug efficacy indicating its potential use as a first-in-class, selective, allosteric muscarinic antipsychotic agent.

Estimation of antagonist Kb from inhibition curves in functional experiments: alternatives to the Cheng-Prusoff equation.

Recent Principles articles have considered procedures for the functional estimation of an antagonist's dissociation constant from the results of an antagonist inhibition curve in the presence of a fixed concentration of agonist. Leff and Dougall (TiPS 14, 110-112) have derived an equation, analogous to the (generally invalid) Cheng-Prusoff equation, which requires that the agonist concentration--effect curves have the form of a logistic function. In the final article of the series, Sebastian Lazareno and Nigel Birdsall consider this equation in the context of two related methods for analysing functional inhibition curves-a null method that makes no assumptions about the shapes of curves, and a curve-fitting method that provides an estimate of the antagonist Schild slope.

Effects of contamination on radioligand binding parameters.

Radioligand binding studies are used to provide quantitative estimates of parameters such as the receptor density of a tissue and the affinity values of labelled and unlabelled ligands. The presence of an unlabelled competing contaminant, which might be present because of actual contamination, inadequate radioligand purification or the breakdown of the radioligand to an active species, has surprising effects on these estimates: the apparent affinity of the radioligand is increased but the Ki values of unlabelled ligands are unaffected. The most striking and sensitive effects are on radioligand association kinetics, which become independent of radioligand concentration at high radioligand concentrations.

Allosterism at muscarinic receptors: ligands and mechanisms.

The evaluation of allosteric ligands at muscarinic receptors is discussed in terms of the ability of the experimental data to be interpreted by the allosteric ternary complex model. The compilation of useful SAR information of allosteric ligands is not simple, especially for muscarinic receptors, where there are multiple allosteric sites and complex interactions.

Pirenzepine indicates heterogeneity of muscarinic receptors linked to cerebral inositol phospholipid metabolism.

Muscarinic receptor-mediated inositol phospholipid metabolism was examined in rat cerebral cortex and medulla-pons slices prelabelled with 3H-inositol. Carbachol stimulated accumulation of 3H-inositol phosphate in both regions with identical EC50 values though the maximal stimulation was considerably greater in cortex. The carbachol response was potently suppressed by atropine with very similar affinities in both regions. However, the M1 antagonist pirenzepine displayed a 10-fold greater affinity for the cortical carbachol response. The data supports the concept of heterogeneity of cerebral muscarinic receptors and suggests that both M1 and M2 sites may be linked to inositol phospholipid metabolism in brain.

Allosteric interactions of quaternary strychnine and brucine derivatives with muscarinic acetylcholine receptors.

The affinity and allosteric properties of 22 quaternary derivatives of strychnine and brucine at the m1-m4 subtypes of muscarinic receptors have been analyzed and compared. The subtype selectivity, in terms of affinity, was in general m2 > m4 > m1 > m3. The highest affinities were found for N-benzyl, N-2-naphthylmethyl, and N-4-biphenylylmethyl strychnine (13, 14, and 18, respectively). All the strychnine and brucine derivatives were positively cooperative with the antagonist, N-methylscopolamine, at m2 receptors and, in the case of the strychnine analogues, were positively cooperative with N-methylscopolamine at least at one other subtype. The strychnine analogues were negatively cooperative with the neurotransmitter, acetylcholine, at all subtypes whereas brucine and five of the six derivatives examined were positively cooperative with acetylcholine at one or more subtypes (m1-m5) and exhibited different patterns of subtype selectivity. The ability to generate subtype-selective allosteric enhancers of acetylcholine binding and function may be of use in the development of drugs for the treatment of Alzheimer's disease.

Estimation of competitive antagonist affinity from functional inhibition curves using the Gaddum, Schild and Cheng-Prusoff equations.

1. The estimation of antagonist affinity from functional experiments in which the effect of a fixed agonist concentration is reduced by a range of antagonist concentrations ('functional inhibition curves') has been considered from both a theoretical and experimental viewpoint. 2. Theoretical predictions are compared with results obtained from the stimulation of [35S]-GTP gamma S binding by acetylcholine to membranes of Chinese hamster ovary (CHO) cells stably transfected with human m1-m4 muscarinic receptors, and inhibition of the stimulated binding by pirenzepine and AQ-RA 741. 3. The usual procedure of applying the Cheng-Prusoff correction is shown to be theoretically invalid, and predictions are made of the size and distribution of errors associated with this procedure. 4. A different procedure for estimating antagonist affinity, using the principles of dose-ratio analysis and analogous to use of the Gaddum equation, is found to be accurate and theoretically valid. 5. A novel method of analysis allows accurate estimation of both antagonist affinity and Schild slope, by fitting the combined data from an antagonist inhibition curve and an agonist activation curve directly to a form of the Schild equation (derived by Waud) using non-linear regression analysis. 6. It is shown that the conventional Schild analysis can be enhanced by treating part of the data as a family of inhibition curves and including in the Schild plot dose-ratios estimated from the inhibition curves.

Pharmacological characterization of acetylcholine-stimulated [35S]-GTP gamma S binding mediated by human muscarinic m1-m4 receptors: antagonist studies.

1. We have used dose-ratio analysis to estimate functionally the affinity constants (pKb) and Schild slope factors of a range of selective or atypical antagonists at human muscarinic m1-m4 receptors. 2. The functional response was the stimulation by acetylcholine of [35S]-GTP gamma S binding to membranes from Chinese hamster ovary (CHO) cells stably expressing individual receptor subtypes. 3. A novel experimental design and analysis was used which allowed the estimation of affinity and Schild slope factor from a single antagonist inhibition curve, and the results were compared with other methods of analysis, both theoretically valid and invalid. 4. In general, the affinity estimates were very similar to previously reported values obtained in binding studies with animal tissues and cloned human receptors and the Schild slope factors were close to unity. 5. These results demonstrate the validity of the assay and provide no evidence for species differences in antagonist affinity for muscarinic receptor subtypes. 6. The results confirm both the utility of himbacine in distinguishing between m1 and m4 receptors and a previously reported modest m4-selectivity for tropicamide and secoverine. 7. The cholinesterase inhibitor, tacrine (THA), had a potency profile similar to that of gallamine but with less selectivity. Its affinity could not be determined since it had Schild slope factors of about 2 at all subtypes. 8. o-Methoxy-sila-hexocyclium had only a modest selectivity for the m1 subtype.

Functional and binding studies with muscarinic M2-subtype selective antagonists.

1. The potency of a series of selective muscarinic antagonists has been measured on two functional isolated tissue preparations (rat ileum and atria) and these compared with their potency on a range of binding preparations in order to determine whether the subtypes of M2 receptor measured functionally are the same as those measured in binding studies. 2. On the functional preparations pirenzepine, hexahydrosiladiphenidol (HSD) and 4-diphenylacetoxy-N-methylpiperidine (4-DAMP) were more potent on the ileum than on the atrium (3 fold, 29 fold and 5 fold respectively), whereas himbacine, AF-DX 116 and methoctramine showed the opposite selectivity (5 fold, 3 fold and 56 fold respectively). Atropine had a similar potency on the ileum and atrium. 3. [3H]-N-methyl scopolamine was used to study M2 binding sites on membranes from rat heart and rat submandibular gland. Each preparation appeared to contain a homogeneous binding site population. The potencies of the five M2 selective antagonists (and pirenzepine) in binding studies to heart membranes were very similar to those observed in functional studies of rat atria (correlation coefficient = 0.98). Similarly the binding to submandibular gland membranes was very similar to that observed in functional studies on rat ileum (correlation coefficient = 0.97). 4. [3H]-pirenzepine was used to examine the binding of these antagonists to M1 binding sites on membranes from rat cerebral cortex. The affinities of 4-DAMP, HSD, AF-DX116 and himbacine at M1 sites were similar to their affinities on the gland. Only pirenzepine and methoctramine had higher affinity on M1 sites than on the gland. 5. Himbacine had a 20 fold lower affinity at M1 binding sites than at heart sites, and it should therefore be an important tool in identifying M1 sites. 6. Inhibition of [3H]-N-methyl scopolamine binding to rat ileum and rat brainstem by M2-selective antagonists was best described by a two-site model. In both cases the major population of sites (70-90%) appeared to be similar to sites found on the heart (correlation coefficients = 0.95 and 0.97). The other site appeared to be similar to that on the submandibular gland (correlation coefficients = 0.96 and 1.00). 7. The correlations observed in these studies in which a range of selective muscarinic antagonists was used lend weight to previous studies indicating the presence of three functionally important muscarinic receptor subtypes, typified by the binding sites studied in the cerebral cortex, submandibular gland and heart. 8. We propose that the sub-classification of the M2 muscarinic receptor into M2 and M3 subtypes on the basis of ligand binding studies should be extended to cover functionally-defined receptors as well.

The effects of saponin on the binding and functional properties of the human adenosine A1 receptor.

1. Experiments with adenosine deaminase suggest that adenosine is present in membrane preparations from CHO cells bearing adenosine A1 receptors. 2. Pretreatment of the membranes (ca 0.6 mg protein ml-1) with the permeabilizing agent saponin (100 micrograms ml-1) or addition of saponin (10 micrograms ml-1) to the membranes (0.02-0.08 mg protein ml-1) in the assay, generates homogeneous low affinity agonist binding curves in the presence of GTP and an increased function, assessed by agonist stimulation of [35S]-GTP gamma S binding. The affinity constants for the binding of an agonist and an antagonist are not affected by this saponin treatment. Saponin facilitates the interaction of guanine nucleotides with receptor G-protein complexes, possibly by removing a permeability barrier to access of G-proteins by GTP. However, adenosine is still present in the binding assays after saponin treatment. 3. The agonist binding properties of the human A1 receptor have been characterized. In saponin pretreated membranes, 80-90% of the A1 receptors are capable of forming agonist-receptor-G protein complexes in the absence of GTP. These complexes have a 300-600 fold higher affinity than uncoupled receptors for N6-cyclohexyladenosine. 4. A very slow component is observed in the association and dissociation kinetics of the agonist [3H]-N6-cyclohexyladenosine ([3H]-CHA) and in the association but not dissociation kinetics of the antagonist [3H]-8-cyclopentyl-1,3-dipropylxanthine ([3H]-DPCPX). The slow association component of [3H]-DPCPX is essentially absent when incubations are carried out in the presence of GTP. The slow dissociation component of [3H]-CHA binding is rapidly disrupted by GTP. 5. It is hypothesized that long-lasting adenosine-receptor-G protein complexes are present in the CHO membrane preparations. The existence of these complexes, resistant to the action of adenosine deaminase but sensitive to GTP, may rationalize the observed kinetics and the increase in 3H-antagonist binding produced by GTP which has been observed in essentially all studies of A1 receptors and has been ascribed previously to precoupling of A1 receptors to G-proteins in the absence of agonists.

Stereoselective recognition of the enantiomers of phenglutarimide and of six related compounds by four muscarinic receptor subtypes.

1. We have compared the binding properties of the enantiomers of phenglutarimide (1) and of six related compounds to M1 receptors in NB-OK-1 cells, M2 receptors in rat heart, M3 receptors in rat pancreas and the M4 receptors of rat striatum, with their functional (antimuscarinic) properties in rabbit vas deferens (M1/M4-like), guinea-pig atria (M2) and guinea-pig ileum (M3) receptors. The binding properties of the enantiomers of three of the compounds were also measured on cloned human m1-m4 receptors expressed by CHO cells, using [3H]-N-methylscopolamine ([3H]-NMS) as radioligand. 2. The high affinity enantiomers behaved as competitive antagonists in binding and pharmacological studies. (S)-phenglutarimide (pKi-M1 = 9.0/9.3) and (R)-thienglutarimide (pKi-M1 = 8.6/9.2) recognized selectively the native M1 > M4 > M3 > M2 receptors in tissues as well as the respective cloned receptors. 3. The pA2 values at the inhibitory heteroreceptors in the rabbit vas deferens, and at the guinea-pig atria and ileum for the seven more potent enantiomers were compatible with the previous classification of these receptors as M1/M4-like, M2 and M3, respectively. 4. Replacement of the phenyl by a thienyl ring or of the diethylamino by a piperidino group in the phenglutarimide molecule did not affect markedly the potencies of the high affinity enantiomer. In contrast, replacement of the phenyl by a cyclohexyl ring decreased 20 fold the active enantiomers potency. Methylation of the piperidine-2,6-dione nitrogen also reduced markedly the eutomers' affinities, more on the M1 than on the other subtypes. 5. The selectivity profiles (recognition of four receptor subtypes) of six of the seven less active enantiomers were different from the corresponding more active enantiomers selectivity profiles, suggesting that the preparations used in this study were pure. However, we cannot not exclude the hypothesis that the batch of (S)-thienglutarimide used in this study was contaminated by less than 0.02% of the eutomer. 6. In contrast with the eutomer binding site, replacement of the phenyl ring by a thienyl or cyclohexyl ring did not affect binding of the low affinity enantiomers to the muscarinic receptor or the [3H]-NMS-receptor complex. The replacement of the diethylamino group by a piperidine ring, and N-methylation of the piperidine-2,6 dione moiety increased slightly these enantiomers' potencies. 7. The muscarinic receptors were extremely stereoselective, and had up to 20000 fold lower affinity for the less active enantiomers. However, the stereochemical requirements of the muscarinic receptor subtypes were different for the enantiomers of compounds 1-7, being most stringent at M1 receptors. 8. The weaker enantiomers behaved as competitive antagonists in pharmacological studies, at least in the concentration-range investigated.

d-Amphetamine and punished responding: the role of catecholamines and anorexia.

Rats were trained to press a lever for food on a schedule in which components of variable interval reinforcement (V12') alternated with conflict components in which every response resulted in food delivery and footshock. Low doses of d-amphetamine selectively suppressed responding in the conflict component in a dose-dependent manner, whereas prefeeding suppressed responding in both components. Pretreatment with noradrenergic blocking agents (propranolol, phentolamine and phenoxybenzamine) did not diminish the suppressant effect of d-amphetamine, but this effect was reduced by pretreatment with alpha-methyl-para-tyrosine methylester and dopamine blockers (spiroperidol, haloperidol and clozapine) indicating that d-amphetamine was exerting its selective suppressant effect via the release of dopamine. It is suggested that the effects of low doses d-amphetamine on behaviour in conflict situations may provide a useful model for investigating the mode of action of neuroleptic drugs.

Differential effects of selective and non-selective neuroleptics on intracellular and extracellular cyclic AMP accumulation in rat striatal slices.

Cyclic AMP was measured in both striatal slices and in the incubation medium after exposure to dopamine and dopamine antagonist. Dopamine increased cyclic AMP in both tissue and medium. The effect of dopamine was enhanced by sulpiride and domperidone, and to a lesser extent by haloperidol, but alpha-fluphenthixol had only an inhibitory effect. The enhancement by sulpiride was stereoselective and totally suppressed by the D1 antagonist SCH 23390. Cyclic AMP in the medium provided the more sensitive measure of drug effect and increased linearly for up to 20 min., whereas the nucleotide in tissue remained stable or declined after 10 min. It is concluded that: the increase in dopamine-stimulated cyclic AMP efflux caused by D2 antagonists reflects increased intracellular cyclic AMP accumulation rather than an effect on the efflux mechanism; dopamine enhances cyclic AMP accumulation via a D1 receptor, and simultaneously inhibits it through a D2 receptor; and changes in D1 receptor-stimulated cyclic AMP formation in striatum may not be related to the clinical actions of neuroleptics. It remains possible that D2 receptor-mediated inhibition of cyclic AMP accumulation stimulated by a different agonist system may underlie some of the therapeutic actions of dopamine agonists and antagonists.

Selective labelling of dopamine (D2) receptors in rat striatum by [3H]domperidone but not by [3H]spiperone.

Specific binding of [3H]spiperone and [3H]domperidone, displaceable by 1 microM d-butaclamol, was examined in rat striatal membranes. Initial saturation and displacement experiments indicated that [3H]spiperone bound to more sites than [3H]domperidone and that, whilst all displacing drugs were more potent against [3H]domperidone, this difference in potency was greatest for dopamine agonists and specific antagonists and least for 5HT-related drugs. Sulpiride displaced [3H]spiperone biphasically, and was used at a concentration of 50 microM to examine two classes of [3H]spiperone binding: site 1 displaceable by sulpiride, and site 2 displaceable by butaclamol but not by sulpiride. Site 1 had twice the capacity of site 2 and ten times the affinity for [3H]spiperone. Dopaminergic drugs displaced preferentially from site 1, whilst 5HT-related drugs were more potent against site 2. GTP reduced the potency of dopamine, noradrenaline and, to a lesser extent, 5HT at site 1, but had no effect at site 2. [3H]Domperidone sites had the same capacity as [3H]spiperone site 1, and dopamine, noradrenaline and 5HT, in the absence or presence of GTP, and sulpiride had essentially identical affinities for [3H]domperidone sites and [3H]spiperone site 1. It is concluded that [3H]domperidone and [3H]spiperone label an identical population of dopamine (D2) receptors, whilst [3H]spiperone also labels a substantial number of non-dopamine sites, at least some of which are 5HT-related. [3H]spiperone also labels a substantial number of non-dopamine sites, at least some of which are 5HT-related. [3H]Domperidone is the better radioligand for dopamine receptors.

Muscarinic interactions of bisindolylmaleimide analogues.

We have used radioligand binding studies to determine the affinities of seven bisindolylmaleimide analogues, six of which are selective inhibitors of protein kinase C, at human muscarinic M1-M4 receptors. The compounds were most potent at M1 receptors, and Ro-31-8220 was the most potent analogue, with a Kd of 0.6 microM at M1 receptors. The weakest compounds, bisindolylmaleimide IV and bisindolylmaleimide V, had Kd values of 100 microM. If it is necessary to use protein kinase C inhibitors at concentrations of 10 microM or more in studies involving muscarinic receptors then bisindolylmaleimide IV may be the most appropriate inhibitor to use.

The affinity of adenosine for the high- and low-affinity states of the human adenosine A1 receptor.

The affinity of adenosine for the human adenosine A1 receptor expressed on Chinese hamster ovary cell membranes has been measured in the presence and absence of GTP. The competitive effect of endogenous adenosine on the binding properties of adenosine A1 receptors was estimated from differences in the binding of N6-cyclohexyladenosine measured in the absence and presence of adenosine deaminase. From these data, the affinity of adenosine for the high- and low-affinity states of the human adenosine A1 receptor (7 x 10(7) and 1.3 x 10(5) M-1, respectively) was calculated.

Pharmacological characterization of guanine nucleotide exchange reactions in membranes from CHO cells stably transfected with human muscarinic receptors m1-m4.

We have studied muscarinic agonist stimulated [35S]GTP gamma S binding and [gamma 32P]GTP hydrolysis (GTPase) in membranes from CHO cells stably transfected with human muscarinic m1-m4 receptors. 'Full' agonists were at least 10-fold more potent at m2 & m4 receptors than at m1 & m3. This pattern was less marked with 'partial' agonists, which had a greater maximal effect at m2 & m4 than at m1 & m3. McN-A343 uniquely was more potent and efficacious at m4 than at m2 receptors. Antagonist affinity constants were estimated by fitting the data from inhibition curves directly to the Schild model. Antagonist affinity estimates were very similar to those measured earlier in binding studies using animal tissues, and confirmed a small degree of m4 selectivity for tropicamide and secoverine. The receptor subtypes activated more than one G-protein subtype; m2 & m4 receptors activated only pertussis (PTX) sensitive G-proteins, while m1 & m3 coupled to both PTX sensitive and insensitive G-proteins. Acetylcholine (ACh) was more potent in stimulating guanine nucleotide exchange in PTX-treated m1 cells than in controls.

Multiple allosteric sites on muscarinic receptors.

Proteins and small molecules are capable of regulating the agonist binding and function of G-protein coupled receptors by multiple allosteric mechanisms. In the case of muscarinic receptors, there is the well-characterised allosteric site that binds, for example, gallamine and brucine. The protein kinase inhibitor, KT5720, has now been shown to bind to a second allosteric site and to regulate agonist and antagonist binding. The binding of brucine and gallamine does not affect KT5720 binding nor its effects on the dissociation of [3H]-N-methylscopolamine from M1 receptors. Therefore it is possible to have a muscarinic receptor with three small ligands bound simultaneously. A model of the M1 receptor, based on the recently determined structure of rhodopsin, has the residues that have been shown to be important for gallamine binding clustered within and to one side of a cleft in the extracellular face of the receptor. This cleft may represent the access route of acetylcholine to its binding site.

Selective allosteric enhancement of the binding and actions of acetylcholine at muscarinic receptor subtypes.

The ternary allosteric model predicts the possibility of discovering molecules with novel and highly subtype-selective modes of action. This approach has been applied to muscarinic receptors. The alkaloid brucine is capable of selectively enhancing by an allosteric mechanism the effects of low but not high concentrations of acetylcholine at only the m1 subtype of muscarinic receptors. A simple derivative of brucine, N-chloromethylbrucine, enhances acetylcholine actions selectively at only m3 receptors. In addition it binds to, but does not affect, the properties of m4 receptors, thereby demonstrating neutral cooperativity and an 'absolute' selectivity of action at m3 receptors over m4 receptors. Brucine N-oxide enhances acetylcholine binding at m3 and m4 receptors and is neutral at m1 and m5 receptors. These findings allow the possibility of developing muscarinic agents that have a novel and highly targeted mode of action; they may act only on a single muscarinic receptor subtype which is functioning sub-optimally and therefore be of use therapeutically in the early stages of Alzheimer's Disease.

Allosteric effects of four stereoisomers of a fused indole ring system with 3H-N-methylscopolamine and acetylcholine at M1-M4 muscarinic receptors.

We previously demonstrated that brucine and some analogues allosterically enhance the affinity of ACh at muscarinic receptor subtypes M1, M3 or M4. Here we describe allosteric effects at human M1-M4 receptors of four stereoisomers of a pentacyclic structure containing features of the ring structure of brucine. All compounds inhibited 3H-NMS dissociation almost completely at all subtypes with slopes of 1, with similar affinity values at the 3H-NMS-occupied receptor to those estimated from equilibrium assays, consistent with the ternary complex allosteric model. Compound 1a showed positive cooperativity with H-NMS and small negative or neutral cooperativity with ACh at all subtypes. Its stereoisomer, 1b, showed strong negative cooperativity with both 3H-NMS and ACh across the subtypes. Compound 2a was positive with 3H-NMS at M2 and M4 receptors, neutral at M3 and negative at M1 receptors; it was negatively cooperative with ACh at all subtypes. Its stereoisomer, 2b, was neutral with 3H-NMS at M1 receptors and positive at the other subtypes; 2b was negatively cooperative with ACh at M1, M3 and M4 receptors but showed 3-fold positive cooperativity with ACh at M2 receptors. This latter result was confirmed with further 3H-NMS and 3H-ACh radioligand binding assays and with functional assays of ACh-stimulated 35S-GTPgammaS binding. These results provide the first well characterised instance of a positive enhancer of ACh at M2 receptors, and illustrate the difficulty of predicting such an effect.