Investigating Drivers for M1 Muscarinic Acetylcholine Receptor‐Mediated Adverse Events by M1 Positive Allosteric Modulators

Abstract

IntroductionTargeting the allosteric site of the M1 muscarinic acetylcholine receptor (M1 mAChR) is a promising strategy to develop selective drugs to treat cognitive disorders. By binding to the allosteric site, positive allosteric modulators of the M1 mAChRs (M1 PAMs) can enhance the binding (α) and/or the efficacy (β) of the endogenous ligand acetylcholine (ACh) (Fig 1), therefore improving impaired cholinergic transmission that is evident in Alzheimer’s disease and Schizophrenia. In addition, M1 PAMs may also display direct allosteric agonism (tB) (Fig 1). To date, many highly selective M1 PAMs with diverse structures have been developed, unfortunately, the majority of them still caused M1‐mediated cholinergic adverse effects (AEs), including gastrointestinal (GI) disturbances, salivation, increased heart rate and convulsions. This results from the ability of the M1 mAChR to interact with multiple intracellular partners, leading to different functional outcomes, both the beneficial (i.e, improved cognition), and the detrimental (i.e, GI AEs) effects. Excitingly, M1 PAMs can bias the M1 mAChR towards specific signalling outcomes versus others, a phenomenon called biased modulation. However, to date, it is unknown which intracellular partner is the most predominant in M1 PAM activity, and how this may drive the preclinically observed AEs.AimTo determine the role(s) Gα protein(s) and β‐arrestins exert on the allosteric activity of 5 structurally distinct M1 PAMs and link these to their reported AEs.MethodsThe M1 PAMs including BQCA, MK7622, PF06767832, MIPS1780 which showed AEs in in vivo studies, and VU0486846 without observed AEs, were assessed against ACh, in parental or CRISPR HEK293A cells with deletion of specific G proteins or β‐arrestins (ΔGαq/11, ΔGα12/13or Δβ‐arrestin 1/2) expressing the human M1 mAChR. [3H]‐NMS binding assays were performed to determine the binding affinity of ACh (KA), and of the 5 M1 PAMs (KB) and their binding cooperativity (α) with ACh. IP1 accumulation and b‐arrestin 2 recruitment were performed to quantify the efficacy (τB) of the M1 PAMs and their functional cooperativity (αβ) with ACh.ResultsΔGαq/11, ΔGα12/13 or Δb‐arrestin 1/2 significantly reduced the binding affinity of ACh, and for 3 M1‐PAMs, MK7622, PF06767832 and MIPS1780. Interestingly, all these deletions significantly increased the binding cooperativity of these M1‐PAMs with ACh. Unsurprisingly, ΔGαq/11 completely abolished IP1 responses from both orthosteric and allosteric ligands. Notably, ΔGα12/13 or Δβ‐arrestin 1/2 both increased ACh efficacy (τA). While ΔG12/13 caused no change in efficacy of tested PAMs but increased functional cooperativity of PF06767832 and MIPS1780 with ACh, Δβ‐arrestin 1/2 affected the functional properties (τB and αβ) of MK7622 in IP1 pathway. In β‐arrestin 2 pathway, ΔGαq/11 increased MIPS1780 efficacy and ΔGα12/13 increased functional cooperativity of BQCA.ConclusionGαq/11, Gα12/13 and b‐arrestin 1/2 influence binding affinity and functional properties for both ACh and M1 PAMs. These transductors affect M1 PAMs with observed AEs more than the M1 PAM without AEs, VU0486846, and thus playing a potential role in driving AEs.