Muscarinic Acetylcholine Receptors (M1–M5) in Schizophrenia: Clinical Relevance and Therapeutic Potential

M1 and M2 mAChRs receptor densities in the frontal and temporal cortex have been reported to correlate with psychotic-related disturbances in patients with Alzheimer’s disease (AD). In contrast, polymorphisms in the genes of these two receptors were also implicated in the cognitive deficit features of schizophrenia. [Yonh and Cohn 2018]; [Paul et al. 2022]

Over the last 25 years, a substantial amount of molecular research has been dedicated to understanding the physiological role mAChRs may play in regulating cognitive ability and their involvement in behavioural disturbances associated with schizophrenia and AD.

• mAChRs (M1 to M5 subtypes) are found in several synaptic locations throughout the central (e.g., hippocampus and the basolateral nuclear complex of the amygdala) and peripheral nervous systems (e.g., myocardial and smooth muscle)
• mAChRs are coupled to G proteins, leading to downstream secondary messenger pathways [Lanzafame et al. 2003] that can activate or inhibit message transduction systems. [Wess et al. 2007]
• mAChR neurotransmission is characterised by allosteric modulator-induced signal bias whereby some mAChRs activate more than one signal transduction pathway, thus resulting in different physiological outcomes. [van der Westhuizen et al 2020]

Seminal work by Levey and colleagues in the 1990s led to the identification of mAChR subtypes in the brain and periphery and advanced our understanding of their roles in cholinergic neurotransmission [Levey et al. 1991]; [Levey 1996]:

M1 mAChR:

• It is predominantly post-synaptic and plays a role in the control of glutamatergic neurotransmission.

M2 mAChR:

• This is a pre-synaptic autoreceptor expressed in the basal forebrain that negatively affects ACh’s release.

M2 and M4 mAChRs:

• They are co-expressed with M1 receptors in forebrain regions; however, their expression level is much lower.

M4 mAChR:

• They are found pre-and post-synaptically in the striatum, caudate, and putamen, where it regulates DA release in the striatum.

M3 and M5 mAChR:

• They are expressed at very low levels in the brain; both receptors share very similar ligand-binding properties and are suggested to mediate cholinergic vasodilation in various peripheral arteries.

VTA neurons release DA at the Nucleus Accumbens.

Cholinergic stimulation of VTA neurons increases DA release.

[Paul et al. 2022]

Role of M4 receptors (M4Rs): 

• The M4Rs modulate ACh release from the cholinergic interneurons (CIN).
• Stimulation of M4 autoreceptors on the CIN decreases ACh release.
• Laterodorsal tegmental nucleus (LDT) ACh release is also modulated by the M4 autoreceptor.
• Stimulation of M4 autoreceptors on the LDT afferent decreases ACh release.
• The net effect, thus, of M4R stimulation is decreased VTA neuron firing and less VTA DA release.
• The decreased DA outflow occurs in the mesolimbic area involved in psychosis, not motor systems.

Role of M5 receptors (M5Rs):

• VTA neurons have M5 receptors on their axons and nicotinic receptors (nAChR) on axonal terminals that receive ACh input from striatal cholinergic interneurons (CIN).
• The M5 receptors also receive ACh input from the midbrain LDT pathway.
• Cholinergic stimulation of these receptors on VTA neurons increases DA release.

Role of M1 receptors (M1Rs):

• GABA-ergic inhibitory interneurons control glutamate release from excitatory PFC pyramidal neurons.
• The pyramidal glutamate neurons have downstream projections to VTA neurons that stimulate dopamine release.
• Stimulation of Prefrontal Cortex M1Rs on GABA-ergic inhibitory interneurons inhibits pyramidal cell glutamate release, which reduces VTA DA release.

Summary: 

M4 + M1 agonists may reduce psychotic symptoms by reducing presynaptic DA release via bottom-up and top-down DA modulation.

They do not pose a risk of D2-related movement disorders and adverse metabolic effects.

Bottom-Up regulation:

• M4 agonism reduces presynaptic dopamine via “bottom-up” regulation from the midbrain to the striatum.

Top-Down regulation: 

• M1 agonism contributes to antipsychotic effects in a “top-down” fashion from the cortex to the striatum.

Xanomeline is an M1/M4-preferring mAChR agonist without directly affecting dopamine receptors.

In a phase 2 clinical trial involving patients with AD, xanomeline was shown to have a dose-dependent antipsychotic effect with potential benefits on cognition at higher doses; however, it was also reported that 52% of patients discontinued treatment due to adverse events (AE). [Bodick et al. 1997]

Since then, xanomeline has demonstrated significant benefits in patients with schizophrenia by reducing positive and negative symptoms as assessed by the Positive and Negative Syndrome Scale (PANSS), the Brief Psychiatric Rating Scale, and the Clinical Global Impression Scale. [Shekhar et al. 2008]

• However, tolerability issues have again limited its use in this patient population due to the excessive peripheral activation of mAChRs causing cholinergic AEs such as gastrointestinal distress and hypersalivation.

Overall, mAChR pharmacology is not fully understood, evidenced by the number of non-selective mAChR agonists discontinued during clinical development.

Their low intrinsic activity, stimulation of other mAChR subtypes (i.e., there is a high degree of homology across all five subtypes), and poor pharmacokinetic properties have prevented their opportunity to be a novel therapeutic option. [Foster et al. 2021]

Antipsychotics differ in their pharmacological profiles but generally reduce dopamine (DA) D2 signalling. As such, most typical and atypical antipsychotics show little to no efficacy in reducing negative or cognitive symptoms. [Purdon et al 2003]; [McCutcheon et al 2019]

Therefore, novel therapeutic strategies are needed to provide efficacy across the multiple symptom domains of schizophrenia.

• The wide array of symptoms associated with schizophrenia suggests a syndrome with a spectrum of underlying aetiologies.
• Of note, clozapine has superior efficacy in treatment-resistant psychosis, which is hypothesised to be related to a complex bidirectional modulation of mAChRs by not only clozapine but also its metabolites.

Despite clozapine’s efficacy, most early mAChR agonist clinical trials were discontinued due to either tolerability issues (i.e., poor selective bioavailability and the activation of mAChR in the periphery) or the failure to meet primary endpoints.

Novel mAChRs Agonists: 

Structure-based drug design and the analysis of structural–activity relationships across mAChR subtypes have enabled a more in-depth characterisation of potential drug candidate profiles that could be used to treat psychosis. [Brown et al. 2021]

Current muscarinic approaches include direct M1/M4 receptor agonism and selective positive allosteric modulation. However, clinical outcomes have varied between compounds, indicating that efficacy cannot be assumed to represent a uniform class effect.

Xanomeline–trospium chloride: Cobenfy, formerly KarXT

• On September 26th 2024, the FDA approved KarXT (Cobenfy), an oral dual M1/M4 muscarinic acetylcholine receptor (mAChR) agonist for schizophrenia. It was the first approved schizophrenia treatment targeting cholinergic receptors rather than directly blocking dopamine D2 receptors. [US Food and Drug Administration, 2024]
• KarXT is a co-formulation of xanomeline (a dual M1 and M4 receptor agonist) and trospium chloride, a pan-mAChR antagonist approved for overactive bladder. [Sanctura Prescribing Information, 2012]
• Xanomeline showed dose-dependent cholinergic adverse events of nausea, vomiting, diarrhoea, sweating, and hypersalivation mediated by stimulation of peripheral muscarinic cholinergic receptors.
• The strong positive charge on the ammonium group makes tropsium too hydrophilic to cross the blood-brain barrier and, therefore, has limited central anticholinergic effects.
• Trospium, being a peripherally restricted muscarinic receptor antagonist, can reduce the peripheral cholinergic effects of xanomeline.
• However, trospium chloride has been reported to have a variety of anticholinergic-related AEs, including dizziness, confusion, hallucinations, and somnolence.

In a randomised placebo-controlled phase 2 clinical trial (EMERGENT-1), KarXT significantly improved least-squares mean PANSS scores (-17.4 [1.8] vs -5.9 [1.7], p<0.001) in patients with schizophrenia compared with placebo.[Brannan et al. 2021]

Cholinergic or anticholinergic AEs were reported to be more frequent in the KarXT group; however, this did not increase trial discontinuations.

• 182 patients were enrolled and randomly assigned 1:1 to receive KarXT or placebo BID for 5 weeks.
• The primary endpoint was the change from baseline to week 5 in PANSS total score.
• Secondary endpoints included changes in PANSS positive and negative symptom subscores and the change from baseline in the CGI-S scale.

The phase 3 EMERGENT-2 and EMERGENT-3 trials demonstrated consistent efficacy in adults with acute exacerbations of schizophrenia. Compared with placebo, xanomeline–trospium produced placebo-adjusted reductions of 9.6 and 8.4 points in PANSS total scores, with effect sizes of 0.61 and 0.60, respectively. [Kaul et al., 2024a] ,  [Kaul et al., 2024b].

A pooled analysis of the EMERGENT-1, EMERGENT-2 and EMERGENT-3 trials included 640 participants. Xanomeline–trospium produced a 9.9-point greater reduction in PANSS total score than placebo, with an effect size of 0.65. A reduction of at least 30% in PANSS total score occurred in 41.4% of the treatment group compared with 20.9% receiving placebo. [Kaul et al., 2024c].

Improvements were observed across positive and negative symptom measures. However, the trials involved acutely unwell patients and were not specifically designed for primary negative symptoms. Apparent negative-symptom benefits therefore require confirmation in longer trials involving patients with persistent, predominant negative symptoms.

Pooled phase 3 data also demonstrated cognitive improvement among participants with clinically significant cognitive impairment at baseline. This effect was not explained by improvement in psychotic symptoms. Nevertheless, cognition was an exploratory outcome in acutely unwell patients, and dedicated trials in clinically stable patients are required before a procognitive effect can be established. [Horan et al., 2025].

The most frequent adverse effects are cholinergic or anticholinergic and predominantly gastrointestinal, including nausea, vomiting, dyspepsia, constipation and diarrhoea. Hypertension, increased heart rate, dry mouth and dizziness may also occur. Across the short-term trials, xanomeline–trospium was not associated with clinically meaningful extrapyramidal symptoms, hyperprolactinaemia, somnolence, weight gain or adverse metabolic changes. [Kaul et al, 2024a], [Kaul et al., 2025]

The 52-week EMERGENT-4 and EMERGENT-5 open-label trials found no new safety signals and suggested sustained symptom improvement. However, only 21.8% of participants enrolled in EMERGENT-4 and 48.9% in EMERGENT-5 completed 52 weeks. These studies lacked placebo or active-comparator groups, limiting conclusions about comparative long-term efficacy and tolerability. [Kaul et al., 2026a], [Kaul et al., 2026b].

The evidence therefore supports efficacy for acute psychosis and a distinct adverse-effect profile. Its role relative to established antipsychotics, clozapine, long-acting injectables and combination treatment remains uncertain because the pivotal trials excluded treatment-resistant schizophrenia and did not include an active comparator.

Emraclidine (Cereval Therapeutics)

• Emraclidine is a selective M4-positive allosteric modulator recently evaluated in a phase 1b study and published in the Lancet. [Krystal et al. 2022] This two-part study aimed to determine the safety, tolerability, pharmacokinetics, and pharmacodynamics of emraclidine compared to placebo (NCT04136873).
• In part 1, 49 patients were enrolled and randomly assigned to receive either ascending doses of emraclidine once daily (OD) or a placebo for 5 weeks.
• In part 2, 81 patients were enrolled to receive emraclidine (30 mg OD or 20 mg BID) or placebo.
• The primary endpoint was the incidence of treatment-emergent AEs, with a focus on suicidality (as assessed by the C-SSRS scale), extrapyramidal symptoms (as assessed by the SAS and BARS scales), and dyskinesia (as assessed by AIMS).

Results from this trial showed that emraclidine had a favourable safety profile, with headaches reported to be the most common AE (emraclidine 30 mg OD, 30%; emraclidine 20 mg BID, 26%; and placebo 26%). Transient increases in blood pressure (BP) and heart rate (HR) were initially reported, but these were asymptomatic and decreased over time. It is important to note that despite these transient increases being described as not clinically meaningful, emraclidine’s effect on BP and HR in patients with schizophrenia is being investigated further (NCT05245539).

However, the subsequent phase 2 EMPOWER-1 and EMPOWER-2 trials did not demonstrate statistically significant improvement in PANSS total scores compared with placebo at six weeks.

Emraclidine remained generally well tolerated, but the negative efficacy findings indicate that successful targeting of the muscarinic system may depend on receptor profile, degree of receptor engagement and pharmacological context rather than M4 modulation alone. [AbbVie, 2024].

ML-007 (Maplight Therapeutics):

• ML-007 is a brain-penetrant M1/M4 muscarinic receptor agonist combined with a peripherally acting muscarinic antagonist as ML-007C-MA. Phase 1 studies supported progression into clinical efficacy trials.

• The phase 2 ZEPHYR trial is evaluating ML-007C-MA in approximately 300 adults with an acute exacerbation of schizophrenia. Until controlled phase 2 findings are reported, its clinical efficacy and comparative tolerability remain undetermined. [MapLight Therapeutics, 2025].

The well-documented tolerability issues of mAChR agonists were initially hypothesised to be due to the peripheral expression of M2 and M3 receptors; as such, targeting M1, M4, and M5 in the CNS was the desired therapeutic goal.

Of note, xanomeline is an M1- and M4-preferring orthosteric agonist; however, its reported activity at M2 and M3 receptors caused severe gastrointestinal distress that resulted in the previously mentioned high discontinuation rate. [Bodick et al. 1997]

More recently, however, preclinical research has highlighted that M1 receptor activation may also contribute to these peripheral AEs, whereby some of the side effects are driven through on-target overstimulation. As such, targeting just the M1 and M4 receptors may not be sufficient to avoid cholinergic AEs. [van der Westuizen et al 2020]

Therefore, to achieve the best therapeutic benefit, the most attractive option is to fine-tune mAChR drug development to create biased orthosteric ligands that target M1 and M4 mAChRs but crucially still have a balanced potentiation profile.

Muscarinic receptor modulation has progressed from a mechanistic hypothesis to a clinically validated treatment strategy for schizophrenia.

Xanomeline–trospium has demonstrated consistent efficacy for acute psychosis across phase 2 and phase 3 trials and became the first US FDA-approved schizophrenia treatment without direct dopamine D2 receptor blockade.

Its principal tolerability burden is gastrointestinal and autonomic rather than extrapyramidal, metabolic or prolactin-related. Nevertheless, cardiovascular, gastrointestinal, hepatic, renal and urinary risks require clinical consideration.

Preliminary findings for negative and cognitive symptoms are promising but remain insufficient to establish efficacy in persistent negative symptoms or cognitive impairment associated with schizophrenia.

Long-term studies suggest sustained benefit without new safety signals, although their open-label designs, high attrition and absence of active comparators limit interpretation.

The failure of the selective M4 positive allosteric modulator emraclidine in phase 2 trials also demonstrates that validation of the muscarinic pathway does not imply that all muscarinic compounds will be effective.

Muscarinic treatments expand the neurobiological framework of schizophrenia beyond direct dopamine receptor blockade. Their eventual position in treatment algorithms will depend on comparative trials, real-world effectiveness, long-term adherence and evidence in treatment-resistant illness.