Muscarinic Acetylcholine Receptor Subtypes as Potential Drug Targets for the Treatment of Schizophrenia, Drug Abuse, and Parkinson’s Disease

Abstract

The neurotransmitter dopamine plays important roles in modulating cognitive, affective, and motor functions. Dysregulation of dopaminergic neurotransmission is thought to be involved in the pathophysiology of several psychiatric and neurological disorders, including schizophrenia, Parkinson’s disease and drug abuse. Dopaminergic systems are regulated by cholinergic, especially muscarinic, input. Not surprisingly, increasing evidence implicates muscarinic acetylcholine receptor-mediated pathways as potential targets for the treatment of these disorders classically viewed as “dopamine based”. There are five known muscarinic receptor subtypes (M₁ to M₅). Due to their overlapping expression patterns and the lack of receptor subtype-specific ligands, the roles of the individual muscarinic receptors have long remained elusive. During the past decade, studies with knockout mice lacking specific muscarinic receptor subtypes have greatly advanced our knowledge of the physiological roles of the M₁–M₅ receptors. Recently, new ligands have been developed that can interact with allosteric sites on different muscarinic receptor subtypes, rather than the conventional (orthosteric) acetylcholine binding site. Such agents may lead to the development of novel classes of drugs useful for the treatment of psychosis, drug abuse, and Parkinson’s disease. The present review highlights recent studies carried out using muscarinic receptor knockout mice and new subtype-selective allosteric ligands to assess the roles of M₁, M₄, and M₅ receptors in various central processes that are under strong dopaminergic control. The outcome of these studies opens new perspectives for the use of novel muscarinic drugs for several severe disorders of the central nervous system.

Acetylcholine activates two families of receptors: nicotinic receptors, which are ligand-gated cation channels and participate in rapid postsynaptic neurotransmission, and muscarinic receptors, which are G-protein coupled receptors and play roles in modulating the activity of many circuits within the central nervous system (CNS).¹ Due to the wide distribution of muscarinic receptors in the CNS and their involvement in many important neuronal functions, these receptors have long been viewed as possible targets for the treatment of various conditions such as Alzheimer’s disease, schizophrenia, Parkinson’s disease, and drug abuse.²⁻⁴

Preclinical and clinical evidence suggests that the cholinergic and dopaminergic systems operate in a dynamic balance and that a disruption of this balance may lead to neurological and psychiatric disorders.^5,6 The first observation supporting this concept was made in 1867 when Jean-Martin Charcot found that the muscarinic receptor antagonist scopolamine could improve symptoms in Parkinson’s disease patients.⁷ It was later discovered that muscarinic receptor antagonists could also alleviate parkinsonian symptoms induced by dopamine D₂ receptor antagonists used as antipsychotics.⁸ However, antimuscarinic agents have also been reported to induce psychotic symptoms.⁹⁻¹² Thus, the cholinergic and dopaminergic systems have long been regarded as balancing, or opposing, each other. More recent findings revealed that muscarinic receptor stimulation can both interfere with and enhance dopamine signaling, depending on the receptor subtype and brain region under investigation (as described below).

Muscarinic Acetylcholine Receptor Subtypes

Five different muscarinic receptor subtypes (M₁ to M₅) have been cloned (for reviews, see Langmead et al.,³ Wess¹³). The amino acids lining the conventional acetylcholine binding site appear to be identical in the five receptor subtypes. M₁, M₃, and M₅ receptors preferentially couple to Gα_q proteins, resulting in the activation of phospholipase Cβ and the subsequent release of calcium from intracellular stores and the stimulation of protein kinase C. M₂ and M₄ receptors couple predominantly to Gα_i/o proteins to inhibit adenylate cyclase, causing a decrease in intracellular cAMP levels. Activation of Gi_βγ subunits through M₂ and M₄ receptor stimulation also modulates various ion channels including voltage-gated calcium channels as well as inwardly rectifying potassium channels.^14,15

Different experimental approaches have shown that muscarinic receptors are present in many regions of the CNS.^13,16−19 The M₁, M₄ and M₅ receptors are predominantly expressed in the CNS, while the M₂ and M₃ receptor subtypes are widely distributed in both the CNS and peripheral tissues.^2,15 In the forebrain, including the striatum, the M₁ and M₄ receptors are the most abundantly expressed muscarinic receptors, whereas the expression of M₂ and M₃ receptors is moderate²⁰⁻²⁴ and the density of M₅ receptors is low.^19,25

In the present review, we will summarize recent work suggesting that central M₁, M₄, and M₅ receptors represent promising new targets for the treatment of various CNS disorders including schizophrenia, drug abuse, and Parkinson’s disease.

Localization and Function of Central M₁ Receptors

The M₁ receptor subtype is expressed throughout the forebrain, including the neocortex, dorsal striatum, nucleus accumbens (NAcc), and hippocampus.^{17,18,20−22,26} M₁ receptors have been implicated in many functions of the CNS. For example, pharmacological and genetic studies support a role of M₁ receptors in cognitive functions like learning and memory, especially in the acquisition phase (see Robinson et al.²⁷ for review). In the striatum, M₁ receptors are coexpressed with D₂ dopamine receptors by GABAergic projection neurons,^20,21 suggesting that activation of M₁ receptors may oppose D₂ receptor-mediated neuronal inhibition.⁵

Localization and Function of Central M₄ Receptors

In rodents, the M₄ receptor is highly expressed in the dorsal striatum, NAcc, neocortex, and hippocampus.^17,26 The expression levels decrease caudally toward the diencephalon and mesencephalon and are lowest in the metencephalon, that is, pons and cerebellum.²⁵ The M₄ receptor is the most highly expressed muscarinic receptor in the striatum, where it is present on medium spiny GABAergic output neurons (MSNs).^22,28 The M₄ receptor subtype is also localized on cholinergic interneurons, where it acts as an autoreceptor.²⁹ M₄ receptor-expressing MSNs usually coexpress dopamine D₁ receptors.^20,21,30 Yan et al.²⁸ have shown that the M₄ receptor is five times more abundant on D₁ receptor-expressing MSNs in the striatonigral direct pathway than on D₂ receptor-expressing MSNs in the striatopallidal indirect pathway. Since M₄ receptor activation inhibits D₁ receptor-stimulated cAMP formation,³¹ it is likely that this interaction is of physiological relevance for the regulation of striatal function.

Localization and Function of Central M₅ Receptors

The M₅ receptor is expressed at relatively low levels in the CNS but has been detected in the cerebral cortex, striatum, hippocampus, thalamus-hypothalamus, midbrain, pons, medulla, and cerebellum.^19,25 M₅ receptor mRNA is the only muscarinic receptor mRNA that has been identified in dopaminergic neurons of the ventral tegmental area (VTA) and substantia nigra pars compacta (SNc).^20,32 In these regions, it is coexpressed with D₂ dopamine receptor mRNA, which has led to the suggestion that the M₅ receptor might play a role in modulating dopaminergic neurotransmission.²⁰ Studies with M₅ receptor-deficient mice (M₅^–/– mice) also revealed a role of this receptor subtype in cerebrovascular function.^33,34

Muscarinic Receptor Knockout Mice

For many years, studies of the roles of the central muscarinic cholinergic system have been complicated by the overlapping expression patterns of the M₁–M₅ receptors and the lack of receptor subtype-specific ligands. To overcome these obstacles, we, as well as other laboratories, have developed M₁–M₅ receptor knockout mice.² More recently, we employed Cre-lox technology to delete specific muscarinic receptors in a cell type- or brain region-specific fashion. For example, we recently generated mutant mice (D₁-M₄^–/–mice) lacking M₄ receptors only in D₁ receptor-expressing neurons.³⁵ In the following, we will review recent behavioral studies carried out with M₁^–/–, M₄^–/–, M₅^–/–, and D₁-M₄^–/– mice.

The use of constitutive knockout mice as tools to study normal muscarinic receptor function may be complicated by compensatory changes in the expression patterns of other muscarinic receptor subtypes or downstream signaling molecules. However, many studies have shown that the inactivation of one particular muscarinic receptor species usually has little effect on the expression levels of the remaining muscarinic receptor subtypes.¹⁸

Allosteric Modulators of Muscarinic Receptors

For reasons already outlined above, the development of orthosteric muscarinic ligands endowed with a high degree of muscarinic receptor subtype selectivity remains a very challenging task.³⁶ To circumvent this problem, medicinal chemists have redirected their efforts toward developing compounds that act at less conserved allosteric binding sites present on the extracellular surface of different muscarinic receptor subtypes. “First-generation” allosteric ligands targeting the M₁ or M₄ receptor lacked efficacy and the physiochemical properties required for in vivo use.³⁷⁻⁴⁰ More recently, M₁, M₄, and M₅ receptor-selective allosteric ligands useful for in vivo studies have been developed (see Bridges et al.⁴¹ for review). The emergence of these new highly subtype-selective ligands has intensified interest in developing new classes of muscarinic drugs for clinical use.^{3,36,40,42,43}

Centrally active allosteric (or “ectopic”) agonists and positive allosteric modulators (PAMs) with high selectivity for the M₁ receptor have been generated recently. These agents include the allosteric M₁ agonists TBPB, VU0184670, and VU0357017 and the M₁ PAMs VU0405652 and BQCA.⁴³⁻⁴⁸ TBPB was found to attenuate amphetamine-induced locomotor activity in rats.⁴⁶ In addition, we recently tested TBPB, VU0357017, and BQCA in a cocaine discrimination procedure and TBPB in a chronic intravenous cocaine self-administration procedure. In these studies, all agents attenuated cocaine’s effects.^49,50 While some gastrointestinal side effects were observed after treatment of mice with BQCA, which is poorly brain-penetrant, no adverse effects were observed with the more brain-penetrant ligands TBPB and VU0357017.^49,50

Centrally active PAMs of the M₄ receptor currently include LY2033298,⁴² VU0152099, VU0152100,⁵¹ VU0152129 (initially named 13k,) and VU0359509 (initially named 21o).^41,52 Consistent with findings in M₄^–/– mice, LY2033298 dose-dependently attenuated apomorphine-induced deficits in prepulse inhibition of the startle response.⁴² Moreover, VU0152099 and VU0152100 potently attenuated amphetamine-induced hyperlocomotion in rats.⁵¹ To obtain M₄ receptor PAMs with increased metabolic stability and improved physiochemical properties, Kennedy et al.⁵² developed VU0152129 and VU0359509.⁴¹ However, these compounds displayed a more moderate attenuation of amphetamine-induced hyperlocomotion in rats, most likely due to low potency at rat M₄ receptors.⁵² Since the action of M₄ receptor PAMs requires the presence of endogenous acetylcholine it is likely that such agents may not cause major motor side effects in clinical use. Consistent with this notion, VU0152100 did not affect performance in the rotarod test adversely.⁵¹ Thus, data obtained with both M₄^–/– mice and M₄ receptor PAMs support the concept that central M₄ receptors represent an attractive drug target for the treatment of various CNS disorders.^2,3,50,53

Recently, selective allosteric ligands targeting the M₅ receptor have also been developed.^54,55 Future in vivo studies with this new class of compounds should reveal whether drug-induced modulation of M₅ receptor activity may have therapeutic potential.

Muscarinic Receptors and Schizophrenia

Post mortem studies have consistently shown widespread decreases in the levels of muscarinic receptors in brains from patients suffering from schizophrenia.⁵⁶⁻⁵⁹ Specifically, it has been reported that M₁ and M₄ receptor levels are decreased in striatal areas, hippocampus and prefrontal cortex of schizophrenic patients.⁶⁰⁻⁶⁴ Altered M₁ receptor function has also been described in a subset of schizophrenic patients suffering from “muscarinic receptor-deficit schizophrenia”.⁶⁵ The detection of anti-M₁ antibodies in schizophrenic patients suggests that an autoimmune response may contribute to muscarinic receptor dysfunction in schizophrenia.^66,67 These findings, together with the known localization of M₁ and M₄ receptors in brain areas relevant for psychosis, suggested that the M₁ and/or M₄ receptor may represent a new target for the treatment of psychosis including schizophrenia. This notion is supported by a considerable body of preclinical evidence.⁶⁸⁻⁷³

Receptor localization, pharmacological and genetic studies converge to support a role for M₁ receptors in cognitive functions (see Robinson et al.²⁷ for review). Interestingly, Bymaster et al.⁷⁵ speculated that M₁ (and/or M₄) agonists may improve cognitive function in schizophrenia and other CNS disorders, largely based on data with the M₁/M₄ receptor-preferring muscarinic agonist xanomeline (see below).

Xanomeline: An M₁/M₄ Receptor Preferring Muscarinic Agonist

In a large randomized, placebo-controlled, double-blind clinical trial on Alzheimer’s patients, xanomeline, an M₁/M₄ receptor-preferring muscarinic agonist, displayed a robust effect against psychosis-like behaviors.⁷⁵ A smaller clinical trial with xanomeline in schizophrenic patients supported these initial findings.⁷⁶ Xanomeline also improved cognitive performance in some tests (e.g., verbal learning, short-term memory), supporting the notion that M₁ and/or M₄ agonists may prove useful in the treatment of cognitive deficits in schizophrenia that are typically poorly managed by existing medications.^75,76 In both clinical studies, xanomeline administration was associated with significant gastrointestinal and other side effects,^75,76 precluding further development of the drug for clinical use. Recent studies with M₄^–/– and M₁^–/– mice suggest that the antipsychotic effects of xanomeline are mediated primarily through M₄ receptors.^50,77

Interestingly, M₁^–/– mice display a phenotype that is similar to that seen in animal models of psychosis, including hyperactivity, increased striatal dopamine release, certain cognitive deficits, and an elevated response to amphetamine.⁷⁸⁻⁸⁰ Moreover, in several studies M₄^–/– mice displayed increased locomotion following the administration of selective D₁, nonselective, and indirect dopamine receptor agonists.^74,81,82 In vivo microdialysis studies revealed that M₄^–/– mice exhibit elevated basal dopamine release in the NAcc and enhanced dopamine release in response to psychostimulants (amphetamine and phencyclidine).⁸³ In addition, M₄^–/– mice have increased basal acetylcholine tonus in the midbrain, consistent with the established role of M₄ receptors as autoreceptors inhibiting acetylcholine release.⁸³ However, in studies with M₄^–/– mice that had been extensively backcrossed (mouse genetic background: C57BL/6NTac), we did not find any difference in basal dopamine release in the NAcc between M₄^–/– mice and wildtype (WT) littermates.⁸⁴ In line with previous findings, backcrossed M₄^–/– mice showed an increased dopamine efflux in the NAcc in response to cocaine (Figure 1).⁸⁴

Figure 1.

Exaggerated cocaine-induced increases in dopamine in the nucleus accumbens of M₄^–/– mice. Cocaine-induced increases in extracellular dopamine in M₄^–/– (white symbols/bars) and M₄^+/+ (black symbols/bars) mice were measured by in vivo microdialysis in freely moving animals in the nucleus accumbens after s.c. administration of cocaine 10 mg/kg (squares), 30 mg/kg (circles), or vehicle (triangles) Cocaine (30 mg/kg) caused a greater increase in extracellular dopamine in M₄^–/– mice (open circles) compared to WT mice (filled circles). Inset shows dopamine as % baseline, area under the curve (AUC) from 20 to 80 min (***p < 0.001, **p < 0.01, *p < 0.05). [Reprinted with permission from Psychopharmacology.]⁸⁴

To investigate the functional role in regulation of dopaminergic neurotransmission of the subpopulation of M₄ receptors present on D₁ receptor-expressing neurons, we generated mice that lack M₄ receptors only in D₁ receptor-expressing cells (D₁-M₄^–/– mice).³⁵ Similar to the whole-body M₄^–/– mice, these mice also displayed a “dopamine hypersensitivity phenotype” with increased hyperlocomotion in response to dopamine agonists and enhanced sensitization to psychostimulants, a process thought to reflect adaptive changes induced by drug abuse (Figure 2).³⁵ Moreover, we found that the antipsychotic-like effects of xanomeline were also almost completely abolished in D₁-M₄^–/– mice (Figure 3).⁸⁵ Taken together, these data suggest that a subpopulation of M₄ receptors present on D₁ receptor-expressing neurons plays an important role in regulating central dopaminergic neurotransmission and that these receptors are of key importance in mediating the antipsychotic-like effect of xanomeline.

Figure 2.

Increased amphetamine-induced hyperlocomotion in D₁-M₄^–/– mice. To induce behavioral sensitization repeated injections of amphetamine (amph; 2 mg/kg, s.c.) were paired with exposure of the mice to activity test cages for 1 h per day. After an initial saline injection at day 0, D₁-M₄^–/– mice (black) and control floxed littermates (white) were divided into two groups that received either saline (triangles) or amphetamine (circles) for 6 days. In D₁-M₄^–/– mice, amphetamine-induced hyperlocomotion was significantly greater on days 4 and 5 as compared to day 1 (^#p < 0.05, ^###p < 0.001). In control mice, amphetamine injections resulted in a clear trend toward enhanced locomotor responses on days 2–5; however, this effect did not reach statistical significance. The repeated amphetamine injections generally induced higher levels of hyperlocomotion in D₁-M₄^–/– mice, reaching significance on days 4 and 5 (*p < 0.05, p < 0.01). After a 13-day drug- and test-free period, all mice were injected with amphetamine on day 20 and retested. Amphetamine pretreated D₁-M₄^–/– mice showed a significantly increased locomotor response (^###p < 0.001 versus amphetamine-pretreated control mice). The observed hyperlocomotion could not be ascribed to context conditioning (day 21). [Reprinted with permission Journal of Neuroscience.]³⁵

Figure 3.

Lack of attenuation of amphetamine-induced hyperlocomotion by xanomeline in D₁-M₄^–/– mice. The effect of xanomeline (xan) on amphetamine (amph)-induced hyperlocomotion was measured after coadministration of xanomeline, vehicle (veh), and/or amphetamine (2 mg/kg, s.c.) for 2 h in an open field arena. Amphetamine induced a significant increase in locomotor activity (measured as total distance moved) in both genotypes (*p < 0.05, **p < 0.01 vs vehicle). In floxed control mice, 2 mg/kg xanomeline reversed the amphetamine-induced hyperlocomotion, but had no effect in D₁-M₄^–/– mice (^###p < 0.001 vs WT). [Reprinted with permission from Journal of Neuroscience.]⁸⁵

In summary, the findings reviewed above suggest that selective allosteric M₄ receptor agonists or M₄ receptor PAMs may prove beneficial in the treatment of schizophrenia. It remains to be established whether these agents are endowed with a more favorable side effect profile than xanomeline.

Muscarinic Receptors and Drug Abuse

In M₁^–/– mice, both cocaine- and morphine-conditioned place preference was found to be reduced at low drug doses.⁸⁶ It remains unclear whether this phenotype reflects altered rewarding effects of the drugs or altered acquisition of the context/reward association, that is, changes in cognition. So far, M₁^–/– mice have not been characterized in self-administration models, which would allow to address this question. However, the allosteric M₁ agonist TBPB and the M₁/M₄ agonist xanomeline both reduced cocaine self-administration to saline levels in WT mice, without decreasing food-maintained behavior significantly.⁴⁹ M₁^–/– mice and M₁^–/–M₄^–/– double knockout mice were successfully trained to discriminate cocaine from saline.^49,50 Studies with these mutant mice confirmed that M₁ receptor stimulation was necessary for the anticocaine effects of the allosteric M₁ agonist VU0357017.⁵⁰ Furthermore, we found that the ability of xanomeline to attenuate cocaine discrimination was blunted in the M₁^–/– mice (Figure 4).⁵⁰ These findings suggest that selective M₁ agonists may become useful clinically for treating psychostimulant addiction.

Figure 4.

Reduced attenuation of the discriminative stimulus of cocaine by xanomeline in both M₁^–/– and M₄^–/– mice. WT mice, M₁^–/– mice and M₄^–/– mice were trained to discriminate 10 mg/kg (i.p.) cocaine from saline in a standard drug discrimination procedure. Pretreatment with 1.8 mg/kg xanomeline (s.c.) produced a significant (8-fold) rightward shift in the cocaine dose–effect function in the WT mice. This effect was still significant, but blunted or more variable, in both M₁^–/– mice and M₄^–/– mice. Ordinates: % responses emitted on the cocaine-paired side (top panels); rate of responding (maintained by food), in responses per second (bottom panels). [Reprinted with permission from Psychopharmacology.]⁵⁰

We also tested the potential involvement of M₄ receptors in the reinforcing effects of drugs of abuse. We found that M₄^–/– mice self-administered more cocaine than WT mice, and worked harder to earn a cocaine injection than WT mice (Figure 5), suggesting that M₄ receptors also play a role in modulating the reinforcing properties of cocaine.⁸⁴ In addition, as observed with M₁^–/– mice, xanomeline was less effective in attenuating cocaine discrimination in M₄^–/– mice as compared to WT mice (Figure 4).⁵⁰ These drug discrimination data suggest that combined stimulation of M₁ and M₄ receptors is likely to reduce cocaine’s abuse-related effects more effectively than activation of either of the two receptors alone. However, this hypothesis remains to be tested experimentally.

Figure 5.

Increased intravenous cocaine self-administration in M₄^–/– mice. Intravenous cocaine self-administration was measured under an FR 1 (A) and a PR (B) schedule of reinforcement in M₄^–/– mice (open symbols) and WT littermates (filled symbols). M₄^–/– mice exhibited higher response rates than WT mice at cocaine doses of 0.3 and 1.0 mg/kg/infusion under the FR 1 schedule. Under the PR schedule of reinforcement, M₄^–/– mice reached higher breaking points than WT mice at the 1.0 mg/kg per infusion dose (*p < 0.05, **p < 0.01, and ***p < 0.001 vs WT). [Reprinted with permission from Psychopharmacology.]⁸⁴

The activity of midbrain dopaminergic neurons projecting to the NAcc is believed to mediate the reinforcing effects of drugs of abuse.⁸⁷ In intact animals, electrical stimulation of the laterodorsal tegmental nucleus (LDT) produces both acute and prolonged dopamine release in the NAcc.^88,89 Interestingly, the prolonged release was absent in M₅^–/– mice.⁹⁰ This observation led to the hypothesis that M₅ receptor activity can modulate rewarded behaviors. Consistent with this notion, M₅^–/– mice showed reduced morphine-conditioned place preference and less severe morphine withdrawal symptoms.⁹¹ M₅^–/– mice also displayed less cocaine-conditioned place preference and decreased rates of cocaine self-administration relative to WT mice (Figures 6 and 7).^92,93 These effects appeared to be selective for abuse-related effects of morphine and cocaine, as food-maintained operant behavior and cocaine-induced locomotor activity were not affected by the absence of M₅ receptors.^93,95,97 In contrast to the findings obtained with cocaine, amphetamine-induced hyperlocomotion, sensitization and evoked dopamine release in the NAcc were increased in M₅^–/– mice (Figures 8 and 9(95) see, however, Wang et al.,⁹⁶ who used a different M₅^–/– mouse line). These discrepant results probably reflect the pharmacological differences between a dopamine reuptake inhibitor (cocaine) and a dopamine-releasing agent (amphetamine).

Figure 6.

Reduced cocaine-conditioned place preference (CPP) in M₅^–/– mice. Mice were initially habituated to the two-compartment CPP apparatus to determine the side-preference of each individual mouse. Mice were then administered either cocaine (2.5 mg/kg, i.p.) or saline and placed in either the preferred (saline) or nonpreferred (cocaine) compartment for 30 min for 7 days. On the test day, when side-preference was reassessed, cocaine induced significantly less CPP in M₅^–/– compared to WT mice (‡ p<0.05 vs pretest; ** p < 0.01 vs M₅^–/– mice). [Reprinted with permission from Journal of Neuroscience Research.]⁹²

Figure 7.

Reduced intravenous cocaine self-administration in M₅^–/– mice. Intravenous cocaine (0.03, 0.3, 3.2 mg/kg per infusion) self-administration was measured under a progressive ratio schedule of reinforcement in M₅^–/– (open) and M₅^± (gray) mutant mice and their WT littermates (black). M₅^–/– mice reached lower breaking points than WT mice at doses of 0.03 and 0.32 mg/kg per infusion (^†p < 0.05, ^††p < 0.01 vs WT). [Reprinted with permission from Journal of Neuroscience.]⁹³

Figure 8.

Increased amphetamine-induced hyperlocomotion in M₅^–/– mice. To induce behavioral sensitization repeated injections of amphetamine (amph; 2 mg/kg, s.c.) were paired with exposure of the mice to an open field arena for 45 min per day. After an initial saline injection at day 0, M₅^–/– mice and WT controls were divided into two groups that received either saline or amphetamine for 6 days. The repeated amphetamine administration significantly increased locomotor activity in both genotypes, however this effect was significantly greater in M₅^–/– compared to WT mice (*p < 0.05). After a 6-day, followed by an 8-day drug- and test-free period, all mice were injected with amphetamine on days 12 and 20 and retested. Amphetamine pretreated M₅^–/– mice showed a significantly increased sensitized locomotor response (*p < 0.05 vs amphetamine-pretreated WT mice). [Reprinted with permission from Psychopharmacology.]⁹⁵

Figure 9.

Increased amphetamine-potentiated nucleus accumbens dopamine release in M₅^–/– mice. The effect of amphetamine (2 mg/kg, i.p.) on medial forebrain bundle-stimulated dopamine release in the nucleus accumbens of M₅^–/– and M₅^+/+ mice was measured by fixed potential amperometry. Amphetamine increased dopamine efflux in both genotypes. This effect was significantly enhanced in M₅^–/– compared to WT mice (*p < 0.05, **p < 0.01, ***p < 0.001). [reprinted with permission from Psychopharmacology.]⁹⁵

Muscarinic Receptors and Parkinson’s Disease

In Parkinson’s disease, the loss of dopamine-containing neurons in the substantia nigra pars compacta disturbs the balance between cholinergic and dopaminergic neurotransmission in the striatum. As discussed above, this balance is essential for proper locomotor control. Initially, Karasawa et al.⁹⁷ investigated the cataleptic responses induced by haloperidol in M₄^–/– mice (30 min postinjection). In this study, M₄^–/– mice displayed a small decrease in cataleptic activity in some tests (not statistically significant). When monitoring cataleptic responses at 60 and 90 min after drug administration (haloperidol and risperidone), we found that the cataleptic response to both drugs was greatly attenuated in M₄^–/– mice (Figure 10).⁹⁸ This observation suggested that M₄ receptors may play an important role in mediating antipsychotic-induced motor-side effects.⁹⁸ However, haloperidol and risperidone-induced cataleptic responses were completely abolished by treatment both of WT and M₄^–/– mice with the nonsubtype-selective muscarinic receptor antagonist scopolamine, indicating that other muscarinic receptor subtypes may also be involved in this activity⁹⁸ (see however Karasawa et al.⁹⁷). In agreement with our findings in whole-body M₄^–/– mice, the cataleptic response to antipsychotics was also attenuated in D₁-M₄^–/– mice,³⁵ suggesting that M₄ receptors present on D₁ receptor-containing neurons play an important role in mediating drug-induced catalepsy. These data suggest that M₄ receptor antagonists may prove useful for the treatment of Parkinson’s disease. However, these findings also raise the possibility that the use of M₄ receptor agonists or PAMs for the treatment of schizophrenia may elicit motor side effects. The occurrence of such side effects may be less likely in the case of M₄ receptor PAMs, which are only active when M₄ receptors are occupied by endogenous acetylcholine.

Figure 10.

Reduced cataleptic effect of antipsychotic drugs in M₄^–/– mice. Cataleptic responses were measured as the time spent in the placed position (cutoff time: 60 s), at 30, 60, and 90 min after i.p. drug injection. Catalepsy induced by haloperidol (A, B) or risperidone (C, D) was attenuated in M₄^–/– mice compared to WT mice. (*p < 0.05, ***<0.001). The effect of scopolamine (Scop, 5.0 mg/kg, i.p. after 90 min) was examined 120 min after the initial drug administration. Scopolamine significantly reduced the cataleptic responses in both genotypes (^††p < 0.01, ^†††p < 0.001) [Reprinted with permission from European Journal of Pharmacology.]⁹⁸

Conclusions

Phenotypic analysis of muscarinic receptor mutant mice has been instrumental in elucidating the physiological roles of the different muscarinic receptor subtypes. In this review, we focused on the M₁, M₄, and M₅ receptors and their potential as drug targets for the treatment of schizophrenia, drug abuse, and Parkinson’s disease. Preclinical data suggest that M₁ agonists, M₄ agonists and M₅ antagonists may prove useful for treating psychostimulant addiction. Moreover, both preclinical and clinical data with xanomeline suggest that M₁ and M₄ agonists may show clinical efficacy in the treatment of psychosis, including schizophrenia. In addition, animal data indicate that M₄ antagonists could be beneficial in the treatment of Parkinson’s disease. Hopefully, muscarinic receptor subtype-selective agonists, PAMs, and antagonists will show efficacy and acceptable side effect profiles in future clinical trials.

The authors declare no competing financial interest.

Author Contributions

^⊥ D.D. and M.T. contributed equally to this work.

Funding Statement

National Institutes of Health, United States