The current agonists and positive allosteric modulators of α7 nAChR for CNS indications in clinical trials

Abstract

The alpha-7 nicotinic acetylcholine receptor (α7 nAChR), consisting of homomeric α7 subunits, is a ligand-gated Ca²⁺-permeable ion channel implicated in cognition and neuropsychiatric disorders. Enhancement of α7 nAChR function is considered to be a potential therapeutic strategy aiming at ameliorating cognitive deficits of neuropsychiatric disorders such as Alzheimer's disease (AD) and schizophrenia. Currently, a number of α7 nAChR modulators have been reported and several of them have advanced into clinical trials. In this brief review, we outline recent progress made in understanding the role of the α7 nAChR in multiple neuropsychiatric disorders and the pharmacological effects of α7 nAChR modulators used in clinical trials.

Graphical abstract

The α7 nAChR is emerging as an attractive target for neuropsychiatric disorders such as schizophrenia and Alzheimer's disease. There has been much effort in recent years toward the development of α7 modulators for many indications. Here, we update the progress made in understanding the neuropharmacological functions of α7 and its modulators in clinical trials.

1. Structure and function of α7 nAChRs in the brain

Nicotinic acetylcholine receptors (nAChRs) are ligand-gated ion channels that are activated by the neurotransmitter acetylcholine (ACh) for signaling, and they also respond to drugs including the nicotinic receptor agonist nicotine. The nAChRs can be classified into 5 muscle nAChR subtypes (α1, β1, γ/ε, δ) and 12 neuronal nAChR subtypes (α2–10, β2–4)1., 2.. Among the neuronal nAChR subtypes, the α7 nAChR (also known as α7 receptor) that was first isolated and evaluated in 1990s from avian and rodents are homomeric pentamers widely distributed in the central nervous system (CNS) and periphery organs such as spleen and lymph nodes3., 4., 5., 6., 7.. The five identical α7 nAChR subunits are symmetrically organized around the central pore, and each subunit consists of a large amino-terminal extracellular domain (ECD), four transmembrane domains (TMD, TM1–TM4) and a cytoplasmic domain⁸. In each homomeric α7 nAChR, there are five ACh binding sites within the ECD, which are located at the interface of every two subunits8., 9..

Compared with other subtypes of nAChRs, the α7 nAChR exhibits unique functional properties including: 1) fast activation and desensitization by agonists (on a millisecond scale); 2) high calcium permeability (P_Ca/P_Na ≈ 10); and 3) selective inhibition by α-bungarotoxin (α-Btx) and methyllycaconitine (MLA)3., 4., 10., 11., 12.. In the brain, α7 nAChRs are abundantly expressed in the regions underlying cognition and memory, such as the hippocampus and frontal cortex8., 13.. In neurons, the presynaptically localized α7 nAChRs are physiologically more important although they are widely localized in the synapses (both pre- and postsynaptically) and extrasynaptically9., 14.. Presynaptic α7 nAChRs play a major role in facilitating glutamate release in the cerebellum, auditory cortex, hippocampus and many other brain areas15., 16., 17., 18., 19., 20.. Together with α4β2 nAChRs, presynaptic α7 nAChRs also stimulate γ-aminobutyric acid (GABA) release in the hippocampus²¹. Postsynaptic and extrasynaptic α7 nAChRs are also capable of modulating neuronal activity and neurotransmission²². In addition, the α7 nAChRs are also expressed in non-neuronal cells in the brain, including astrocytes, microglia, microvascular endothelial cells, and lymphocytes, playing a role in immunity, inflammation and neuroprotection9., 23., 24., 25., 26., 27., 28..

2. The relevance of α7 nAChR in CNS diseases and therapy

The function of α7 nAChRs is critical for cognition, sensory processing, attention, working memory, and reward. On the contrary, dysfunctional α7 nAChRs are associated with multiple psychiatric and neurologic diseases including schizophrenia, AD, attention deficit hyperactivity disorder (ADHD), addiction, pain and Parkinson's disease (PD). Thus, modulation of α7 nAChR function is an attractive strategy for potential therapy of CNS diseases.

Schizophrenia, with a lifetime prevalence of approximately 1%, chronically and severely afflicts patients all over the world29., 30.. There are at least three distinct symptoms of schizophrenia, including positive symptoms (hallucinations, delusions, thought disorder, and paranoia), negative symptoms (anhedonia, social withdrawal, and thought poverty), and cognitive dysfunction (loss of intellectual abilities such as perception, understanding, working memory, and executive function)²⁹. Almost all the first and second line drugs, including but not limited to chlorpromazine, clozapine, risperidone, olanzapine, and quetiapine, markedly improve positive symptoms for many patients with schizophrenia. However, they show very limited therapeutic effect on negative symptoms and cognitive dysfunction³¹. Genetic studies show that CHRNA7, the gene encoding α7 nAChR protein, and a partial duplication of CHRNA7, CHRFAM7A, are associated with inhibitory sensory gating deficit in schizophrenia patients32., 33.. It has also been reported that there is diminished mRNA of CHRNA7 and decreased α-Btx binding in post mortem brain tissue samples from patients with schizophrenia34., 35.. It has been reported that exposure to the non-selective nAChR agonist, nicotine, shows the effect of improving or normalizing sensory deficits in schizophrenia³⁶.

Alzheimer's disease (AD) is a chronic neurodegenerative disorder characterized by a slow onset of memory loss and a late development of disorientation, mood swings and behavioral problems³⁷. The cause for AD is still mostly unknown except for less than 10% of cases in which genetic variations have been identified³⁸. One of the most convincing theories is that aberrant extracellular amyloid-β peptide (Aβ) deposits are the fundamental cause of AD39., 40.. Aβ is a peptide of 36–43 amino acids crucially involved in AD as the main component of the amyloid plaques found in the brain neurons of AD patients. Aβ exhibits relatively high binding affinity with α7 nAChRs, and they are co-localized in cortical regions and the hippocampus in the brains of AD patients41., 42.. It is controversial as to whether Aβ and its oligomers, Aβ_1–42, are weak agonists or antagonists, but in either role, they are capable of inhibiting endogenous ACh from activating α7 nAChRs by desensitization or non-competitive antagonism43., 44.. The Aβ–α7 nAChR interaction influences neurotransmission, synaptic plasticity, learning and memory45., 46., 47., 48., 49.. Directly or indirectly, the Aβ–α7 nAChR interaction is an important aspect of AD⁵⁰. From 1993 to 2001, several acetylcholinesterase inhibitors (AChEI) including tacrine (approved in 1993), donepezil (1996), rivastigmine (2000) and galanthamine (2001) which non-selectively enhance nAChR function have been approved for treatment of mild to moderate AD51., 52.. However, there are no AChEIs approved since then. A number of AChEIs such as eptastigmine, phenserine, huperzine A, and dimebon have failed or were discontinued in clinical trials due to adverse effects or insignificant benefits53., 54., 55., 56..

α7 nAChR is also reported to be relevant to other multiple CNS disorders including cigarette addiction, PD and pain57., 58., 59.. The opioid antagonist naltrexone, which inhibits the activity of α7 nAChR, was indicated for potential application in tobacco-use cessation⁶⁰. Application of the α7 nAChR selective agonist PNU-282987 has been shown to decrease motivation for nicotine use in rats57., 61.. In the temporal cortices of post-mortem PD patients' brains, α7-expressing neurons are significantly less abundant than in the control group⁶². Accumulating evidence also shows that activation of α7 nAChR can alleviate PD symptoms in animal models58., 63., 64., 65.. Modulation of α7 nAChR function by agonists and positive allosteric modulators (PAMs) exhibits antinociceptive effects in acute and persistent pain66., 67., 68., 69., 70., 71., 72.. Genetic silencing of α7 reveals phenotypes of hyperalgesia and allodynia in mice, whereas α7-hypersensitive mice display decreased pain sensitivity⁵⁹. Altogether, these studies indicate that α7 nAChR serves as a potential therapeutic target for indications such as schizophrenia, AD, ADHD, addiction, pain, PD and other related CNS disorders.

3. α7 nAChR modulators

Over the past two decades, medicinal chemists and biologists have carried out extensive studies in identification and evaluation of α7 nAChR modulators. The major focus was in finding potent and selective compounds and bringing them into therapeutic applications. As summarized in Fig. 1 and Table 173., 74., 75., 76., 77., 78., 79., 80., 81., 82., 83., 84., 85., 86., 87., 88., 89., 90., 91., 92., 93., 94., 95., 96., 97., 98., 99., 100., 101., 102., 103., 104., 105., 106., 107., 108., 109., 110., 111., 112., 113., 114., 115., 116., 117., 118., 119., 120., twelve α7 nAChR modulators were tested in clinical trials since 2006.

Figure 1.

Current α7 nAChR agonists and PAMs in clinical trials for different indications. There are 11 drug candidates, of which ten agonists and one PAM are currently being tested for treatment of schizophrenia, nine agonists for AD, three agonists for nicotinic addiction, two agonists for ADHD, and one agonist each for PD and pain.

Table 1.

α7 nAChR agonists and PAM in clinical trials.

Compound	Classification	Potency & efficacy	Animal model on CNS disorders	Indication	Clinical status (Sponsor)
Tropisetron	Partial agonist	Binding affinity:	Mice: phencyclidine-induced cognitive deficits75.	Pain	Phase IV (completed in 2009)
		Ki: 6.9 nmol/L (in rα7)73
					(University Hospital, Clermont-Ferrand)
		Electrophysiology activity:	Rats: young and aged rats76; naloxone-induced place aversion77.
		Hα7 in oocytes: EC50 = 0.6 μmol/L; Emax = 25%74
				Smoking cessation; schizophrenia	Phase III (completed in 2011)
		Mα7 in oocytes: EC50 = 1.3 μmol/L; Emax = 36%73
					(Baylor College of Medicine)
GTS-21/DMXB-A	Partial agonist	Binding affinity:	Rats: normal or isoflurane-induced cognitive impairment aged rats80., 81., 82.; ibotenic acid-induced dementia83; mecamylamine-caused learning impairment84; auditory gating in isolation-reared rats85; apomorphine/MK-801-elicited PPI deficits86., 87..	Schizophrenia	Phase II (completed in 2015)
		Ki: 2000 nmol/L (in hα7)78
		Electrophysiology activity:
		Hα7 in oocytes: EC50 = 11.0 μmol/L; Emax = 9%79			(University of Colorado)
		Rα7 in oocytes: EC50 = 5.2 μmol/L; Emax = 32%79		AD; ADHD	Phase II (completed in 2008)
					(CoMentis)
			Mice: Aβ-induced cognitive deficits45; deficient sensory inhibition88; aggressive behavior in mouse models89.
				Tobacco use disorder	Phase II (not yet recruiting)
			Rabits: aged rabbits90., 91..
					(University of Florida)
			Monkeys: normal monkeys in DMTS task78; Ketamine-induced cognitive deficit92.
ABT-126	Agonist	Binding affinity:	Monkeys: Parkinsonian monkeys95.	AD	Phase II (terminated in 2014)
		Ki: 12–14 nmol/L (in hα7, rα7 and mα7)93., 94.
					(AbbVie)
				Schizophrenia	Phase II (terminated in 2014)
					(AbbVie)
AZD0328	Partial agonist	Binding affinity:	Mice: NOR in normal mice96., 97..	AD	Phase I (completed in 2008)
		Ki: 3.0 and 4.7 nmol/L (in hα7 and rα7)96
		Electrophysiology activity:	Monkeys: normal monkeys in delayed response task98.
		Hα7 in oocytes: EC50 = 338 nmol/L; Emax = 64.7%96			(AstraZeneca)
				Schizophrenia	Phase II (terminated in 2008)
		Rα7 in oocytes: EC50 = 150 nmol/L; Emax = 61.0%96
					(AstraZeneca)
BMS-933043	Partial agonist	Binding affinity:	Rats: MK-801-induced cognitive deficits73; S(+)ketamine-induced sensory gating deficits73.	Schizophrenia	Phase I (completed in 2013)
		Ki: 8.1 and 3.3 nmol/L (in hα7 and rα7)73
		Ca2+ flux assays:
		Hα7 in HEK293 cell line: EC50 = 23.4 nmol/L73			(Bristol-Myers Squibb)
		Electrophysiology activity:
		Hα7 in oocytes: EC50 = 0.29 μmol/L;Emax = 24%73
		Rα7 in oocytes: EC50 = 0.14 μmol/L; Emax = 27%73
			Mice: MK-801-induced cognitive deficits73.
EVP-6124/ Encenicline	Partial agonist	Binding affinity:	Rats: scopolamine-induced deficit99; delay-dependent forgetting in the NOR100; low attentive rats101.	AD; dementia	Phase III (terminated in 2017)
		Ki: 9.98 nmol/L (in rα7)99
					(FORUM)
		Electrophysiology activity:
		Hα7 in oocytes: EC50 = 0.39 μmol/L; Emax = 42%99		Schizophrenia; impaired cognition	Phase III (completed in 2016)
					(FORUM)
				Nicotine dependence; smoking cessation	Phase II (terminated in 2015)
					(A. Eden Evins)
MEM3454/RG3487	Partial agonist	Binding affinity:	Rats: attentional performance in normal rats103; aged rats102; apomorphine-induced deficits in sensorimotor gating102.	AD	Phase II (completed in 2007)
		Ki: 6 nmol/L (in rα7)102
		Electrophysiology activity:			(Memory)
		Hα7 in oocytes: EC50 = 0.8 μmol/L; Emax = 63%102
				Schizophrenia	Phase II (unknown)
		Hα7 in QM cell line: EC50 = 7.7 μmol/L; Emax = 69%102			(Memory)
AQW051	Partial agonist	Binding affinity:	Rats: aged rats05105	Schizophrenia	Phase II (completed in 2013)
		Ki: 27 nmol/L104
		Ca2+ flux assays:	Mice: NOR in normal mice105
		Hα7: EC50 = 7.4 μmol/L; Emax = 73%105	Monkeys: Parkinsonian monkeys106		(Novartis)
		Electrophysiology activity:
		Hα7 in oocytes: EC50 = 7.5 μmol/L; Emax = 75%105		Levodopa-induced dyskinesia in PD	Phase II (completed in 2013)
					(Novartis)
				AD	Phase II (terminated in 2009)
					(Novartis)
TC-5619	Full agonist	Binding affinity:	Mice: th(tk–)/th(tk–) mice108; apomorphine-induced PPI deficits108; NOR in normal mice108.	Schizophrenia	Phase II (completed in 2013)
		Ki: 1 and 1.4 nmol/L (in hα7 and rα7)107., 108.
		Electrophysiology activity:			(Targacept)
		Hα7 in oocytes: EC50 = 33 nmol/L; Emax = 100%108., 109.		AD	Phase I (completed in 2011)
		Rα7 in GH4C1 cell line: EC50 = 17 nmol/L; Emax = 76%107			(Targacept)
				ADHD	Phase II (completed in 2012)
					(Targacept)
SSR-180711	Partial agonist	Binding affinity:	Rats: MK-801/PCP-induced cognitive deficits111; depressive disorders rates111; neurodevelopmental latent inhibition models of schizophrenia112.	AD	Phase II (terminated in 2008)
		Ki: 14 and 22 nmol/L (in hα7 and rα7)110
					(Sanofi)
		Electrophysiology activity:
		Hα7 in oocytes: EC50 = 4.4 μmol/L; Emax = 51%110	Mice: chronic mild stress model113; Aβ-induced memory deficits114; phencyclidine-induced cognitive deficits115; forced swim and tail suspension tests116
		Hα7 in GH4C1 cell line: EC50 = 0.9 μmol/L; Emax = 36%110
APN1125	Partial agonist	Electrophysiology activity:	Rats: NOR in normal rats117	Schizophrenia	Phase I / Phase II (suspended in 2016)
(Structure Undisclosed)		Hα7 in oocytes: EC50 = 1.16 μmol/L; Emax = 41%117
					(CoMentis)
AVL-3288/XY4083/CCMI	Type I PAM	Electrophysiology activity:	Mice: DBA/2 mouse model of sensory-gating deficit118; MK-801-induced hyperlocomotion mode eight-arm radial maze in normal mice118; NOR in normal mice119.	Schizophrenia; schizoaffective disorder	Phase I (recruiting)
		Hα7 in oocytes: EC50 = 0.7 μmol/L; Emax = 9 folds118			(New York State Psychiatric Institute; University of Colorado)
			Rats: 5-CSRTT in normal rats119; ketamine-induced cognitive deficits and social withdrawal120.

DMTS, delayed matching-to-sample; NOR, novel object recognition; PCP, neonatal phencyclidine; 5-CSRTT, five-choice serial reaction time task.

Indications and clinical status of α7 nAChR modulators above are obtained from https://clinicaltrials.gov/.

3.1. α7 nAChR agonists

Currently, most developed α7 nAChR agonists are partial agonists. Unlike full agonists such as endogenous ACh, α7 nAChR partial agonists are orthosteric ligands that can only produce a small maximal current even at concentrations where all receptors occupied¹²¹.

Tropisetron ([(1R,5S)-8-methyl-8-azabicyclo[3.2.1]octan-3-yl] 1H-indole-3-carboxylate), firstly identified as 5-HT₃ receptor antagonist (K_i=5.3 nmol/L), is used clinically in preventing and treating nausea and vomiting after cancer therapy122., 123.. In 2001, Macor et al.¹²³ evaluated activity of several 5-HT₃ receptor antagonists on α7 nAChRs and found that tropisetron acted as a selective α7 nAChR partial agonist (K_i=6.9 nmol/L; EC₅₀ =0.6 μmol/L; E_max=25%). Researchers showed that tropisetron could attenuate or improve cognitive deficits in animal models75., 76., 77.. However, tropisetron has not been shown to be effective in improving cognitive deficits in clinical trials. In a phase II clinical trial of tropisetron in patients with schizophrenia, administration of tropisetron significantly improved auditory sensory gating P50 deficits and sustained visual attention, which supports the safety and efficacy of adjunctive tropisetron for treatment of cognitive deficits in schizophrenia¹²⁴.

GTS-21 (3-(2,4-dimethoxybenzylidene)-anabaseine), also named DMXB-A, is a derivative of the natural product anabaseine identified as an α7 nAChR agonist and brought into clinical trials¹²⁵. It has been extensively characterized in vitro and in vivo. This compound acts as a partial agonist in α7 nAChRs and displays better potency and efficacy on rat α7 nAChRs (EC₅₀ =5.2 μmol/L; E_max=32%) than with human nAChRs (EC₅₀ =11 μmol/L; E_max=9%) in Xenopus oocytes⁷⁹. Selectivity of GTS-21 is not favorable in ion flux studies as it inhibits α4β2 nAChRs (IC₅₀=17 μmol/L) and activates α3β4 nAChRs (EC₅₀=21 μmol/L)⁷⁸. However, in electrophysiological recordings in Xenopus oocytes, 100 μmol/L GTS-21 barely evoked current from α4β2 and α3β4 nAChRs⁸³. Extensive in vivo studies were carried out to confirm the pharmacological effect of GTS-21 on cognitive deficits and sensory gating models of rodents and primates (Table 1). Scientists from Abbot and the University of South Florida found that intraperitoneally injecting GTS-21 significantly enhanced the learning and memory ability of aged rats in a water maze, 17-arm radical maze, and Lashley III maze tests80., 81.. When the cognition of aged rats was further impaired by isoflurane, GTS-21 still could mitigate such cognitive deficits⁸². Moreover, acquisition, retention and relearning abilities in eyeblink classical conditioning are much improved in GTS-21-treated aged rabbits than in the vehicle group90., 91.. Cognitive deficits or dementia in rodents and primates as induced by chemical impairment could also be attenuated or normalized by treatment with GTS-21. For instance, Chen et al.⁴⁵ reported that treatment with GTS-21 (1 mg/kg) perfectly prevented Aβ_25–35 induced depression of the α7 nAChR response, which further led to cognitive deficits in mice. These results indicate that GTS-21 may have substantive therapeutic value in the treatment of cognitive deficit in age-associated memory impairment, AD and schizophrenia. Furthermore, sensory gating deficits in rodents could be improved with GTS-21. This compound improved deficient sensory inhibition in DBA/2 mice, and normalized auditory gating in isolation-reared rats, and also ameliorated prepulse inhibition (PPI) deficits induced by apomorphine or MK-80185., 86., 87., 88.. These data show that GTS-21 might have a therapeutic potential for schizophrenia. In 2014, GTS-21 was in phase II clinical studies for treatment of schizophrenia, AD and ADHD. Though GTS-21 failed in improving cognition in schizophrenia patients, high dose of GTS-21 significantly improved negative symptoms in schizophrenia¹²⁶. However, GTS-21 is not a prototypical α7 nAChR agonist due to its relatively higher affinity for α4β2 nAChRs (K_i=20 nmol/L at human and 19 nmol/L at rat) compared with α7 nAChRs (K_i=2000 nmol/L at human and 650 nmol/L at rat)⁷⁸. Thus, the clinical benefits of GTS-21 cannot be simply attributed to α7 nAChR pharmacology.

The most explored structure of α7 nAChR agonists to date is quinuclidine derivatives such as spirooxazolidinones and quinuclidine carbamates, amides, and ethers. The first spirooxazolidinone, AR-R17779 ((–)-spiro[1-azabicyclo[2.2.2]octane-3,5ʹ-oxazolidin-2ʹ-one]) was identified and evaluated in vitro and in vivo127., 128.. However, the cross reactivity with 5-HT₃ receptors and poor penetration of AR-R17779 into the CNS remains a great challenge for clinical development²². AZD0328 ((29 R)-spiro-[1-azabicyclo[2.2.2]octane-3,29(39H)-furo[2,3-b]pyridine] d-tartrate) is an optimized molecule identified as α7 nAChR agonist by AstraZeneca from the spirooxazolidinone series compounds based on AR-R17779 through structure–activity relationship (SAR) studies⁹⁶. AZD0328 acts as a partial α7 nAChR agonist exhibiting an EC₅₀ of 338 nmol/L and an efficacy of 65% on Xenopus oocytes expressing human α7 nAChRs⁹⁶. Compared with the maximal current elicited by serotonin (5-HT) on human 5-HT_3A receptors and ACh on human nAChRs, maximal activity of AZD0328 was only about 12% for human 5-HT_3A receptors, 4% for α4β2 nAChRs and no activity on α3β4 nAChRs⁹⁶. Studies showed that AZD0328 is very stable and has favorable pharmacokinetic (PK) properties, which suggests that this compound is acceptable for clinical trials98., 129.. Through activation of α7 nAChRs, AZD0328 is able to enhance cortical dopamine release in rats and improve novel object recognition (NOR) in mice96., 97.. AZD0328 also displays efficacy in improving working memory in a spatial delayed response task in Rhesus macaques⁹⁸. In 2008, AstraZeneca terminated AZD0328 for a phase II clinical trial for being “unlikely to meet the current target product profile”¹³⁰.

Reported in 2016, a new spirooxazolidinone named BMS-933043 ((2 R)-N-(6-(1H-imidazol-1-yl)-4-pyrimidinyl)-4ʹH-spiro[4-azabicyclo[2.2.2]octane-2,5ʹ-[1,3]oxazol]-2ʹ-amine) was identified by Bristol-Myers Squibb as a selective partial agonist for the α7 nAChR (K_i=8.1 nmol/L at human α7 nAChRs)⁷³. Preclinical studies showed cognition enhancement and sensory gating improvement in rodents⁷³. This compound was advanced into a phase I clinical trial for schizophrenia in 2012.

Analogs with quinuiclidine, aromatic moieties, and functional linkers such as amides and ethers have been substantially explored. EVP-6124 (((R)-7-chloro-N-quinuclidin-3-yl)benzo[b] thiophene-2-carboxamide) is a representative quinuclidine amide analog developed by FORUM (formerly EnVivo) that acts as a potent partial agonist at α7 nAChR (EC₅₀=0.39 μmol/L, E_max=42%) and an antagonist at 5-HT₃ receptors (IC₅₀<10 nmol/L)⁹⁹. It is also reported that EVP-6124 enhanced dopamine, acetylcholine, and glutamate efflux in the rat cortex and nucleus accumbens131., 132.. In vivo, EVP-6124 reversed scopolamine-induced deficit and improved natural forgetting and low attention in rats99., 101.. Treatment with EVP-6124 in phase I and II trials for mild-to-moderate AD was well tolerated and showed statistically significant improvements compared with placebo on cognitive and functional measures133., 134.. A phase II, a double-blind, randomized, placebo-controlled, parallel-design clinical trial conducted for schizophrenia showed statistical significant cognition improvement in schizophrenia patients¹³⁵. However, the results of phase III clinical trial from 2012 to 2016 for schizophrenia did not meet the primary clinical end point as high efficacy in placebo group. Consequently, the other two Phase III trials for AD and dementia were suspended in 2017. Like EVP-6124, MEM3454 ((R)-3-(6-p-tolyl-pyridin-3-yloxy)-1-aza-bicyclo[2.2.2]octane)developed by Memory Pharmaceuticals also exhibited antagonism at 5-HT₃ receptors and procognitive effects in normal and aged rodents102., 103.. Similarly, MEM3454 enhanced dopamine efflux by nAChR stimulation and ACh efflux primarily mediated via 5-HT₃ receptor antagonism¹³⁶. In a phase II clinical trial, MEM3454 failed to improve cognitive deficits in patients with schizophrenia, but moderate negative symptoms in patients were significantly improved¹³⁷. Through homology modeling, molecular docking, and pharmacophore elucidation techniques, Targacept designed and synthesized a series of amide quinuclidine compounds, among which TC-5619 (N-[2-(pyridin-3-ylmethyl)-1-azabicyclo[2.2.2]oct-3-yl]-1-benzofuran-2-carboxamide) exhibits excellent activity and selectivity on α7 nAChR (K_i=1 nmol/L at human α7 nAChRs, 2800 nmol/L at human α4β2 nAChRs, IC₅₀> 10 μmol/L at human 5-HT₃ receptors)107., 108.. This compound acted as an α7 nAChR full agonist with an EC₅₀ of 33 nmol/L in Xenopus oocytes expressing human α7 nAChRs108., 109.. In vivo studies showed adequate properties of TC-5619, including PK profiles, rapid CNS permeability and procognitive effect in rodents108., 109.. However, after two phase II clinical trials were conducted, it was confirmed that TC-5619 did not improve cognitive deficits and negative symptoms in schizophrenia138., 139., 140..

Recently, Novartis disclosed a quinuclidine ether α7 nAChR partial agonist, AQW051 ((R)-3-(6-p-tolyl-pyridin-3-yloxy)-1-aza-bicyclo[2.2.2]octane). In vitro characterization with human α7 nAChR expressed on Xenopus oocytes yielded an EC₅₀ of 7.5 μmol/L and an efficacy of 75%¹⁰⁶. Not only did it show a favorable PK profile and procognitive effects in rodents, this compound also displayed potential in the therapy of PD by reducing l-dopa-induced dyskinesias and extending the duration of l-dopa effects in parkinsonian monkeys104., 105., 106.. AQW051 has been advanced in phase II clinical trials for schizophrenia, AD, and l-dopa-induced PD. It was reported in a phase II randomized, double-blind, placebo-controlled study that evaluated the efficacy and safety of AQW051 in patients with PD and levodopa-induced dyskinesia that AQW051 did not significantly reduce dyskinesia or parkinsonian severity¹⁴¹.

Additionally, structural diversity of α7 nAChR agonists is continuously expanding in the literature from the series of quinuiclidine-based moieties. ABT-126 (2-(((1R,3R,4S,5S,7S)-1-azaadamantan-4-yl)oxy)-5-phenyl-1,3,4-thiadiazole), developed by AbbVie (formerly Abbott), is an azaadamantane derivative that acted as an α7 nAChR agonist (K_i=12–14 nmol/L)⁹³. In a phase II clinical trial in patients with mild-to-moderate AD, ABT-126 demonstrated significant improvement compared with placebo in the primary efficacy endpoint93., 94.. A phase II trial of ABT-126 for treatment of cognitive impairment in schizophrenia was also conducted and revealed that this compound demonstrated a procognitive effect in nonsmoking subjects¹⁴². However, in a phase IIb clinical trial, ABT-126 did not demonstrate a consistent effect on cognition in nonsmoking subjects with schizophrenia but a trend toward an effect on negative symptoms¹⁴³.

Researchers from Sanofi described a diazabicyclononane α7 nAChR partial agonist named SSR180711 (4-bromophenyl (1S,5S)-1,4-diazabicyclo[3.2.2]nonane-4-carboxylate, K_i=14 nmol/L at human α7 nAChRs)¹¹⁰. SSR180711 displayed effects of antidepression, procognition and sensory gating improvement in multiple in vivo studies in rodents112., 113., 114., 116.. However, the phase II clinical trial was terminated in 2008 for insufficient expected benefit and risk. APN1125 developed by Comentis with an undisclosed structure also acted as an α7 nAChR partial agonist (EC₅₀=1.16 μmol/L; E_max=41% at human α7 nAChRs)¹¹⁷. It is currently suspended in a phase I/phase II clinical trial for schizophrenia for business reasons¹⁴⁴.

Most of the clinical trials conducted with α7 nAChR agonists showed a paucity of effects. With limited clear reports, we can only assume the lack of sufficient selectivity over 5-HT₃ receptors and improper designation of clinical trials might be the cause of the discontinued compounds. However, the crucial function of α7 nAChRs in the brain and the compelling evidence of preclinical studies still suggest that selective agonists activating α7 nAChRs may be an attractive therapeutic strategy for schizophrenia, AD and other CNS diseases.

3.2. α7 nAChR PAMs and ago-PAMs

A large number of compounds modulate α7 nAChR function by binding to allosteric sites instead of the orthosteric site that binds agonists and antagonists. α7 nAChR-positive allosteric modulators (PAMs) are a category of these compounds that can potentiate α7 currents in the presence of an agonist such as acetylcholine. On the basis of their macroscopic effects, α7 nAChR PAMs have been classified and distinguished as type I and type II. Type I PAMs mainly enhance agonist-evoked peak currents without delaying desensitization and do not reactivate desensitized receptors, whereas type II PAMs can delay desensitization and reactivate desensitized receptors²². When compared with agonists, α7 nAChR PAMs are more promising therapeutic tools because of their maintenance of endogenous activation characteristics, better selectivity profile, higher structural diversity and better final effects with an extra neuroprotection effect¹⁴⁵. α7 nAChR ago-PAMs can activate receptors from non-orthosteric sites while still retaining the properties of PAMs.

AVL-3288 ((E)-N-(4-chlorophenyl)-3-((4-chlorophenyl)amino)-2-(3-methylisoxazol-5-yl)acrylamide), which also named XY4083 or CCMI, is a representative type I α7 nAChR PAM. Screened from a small library of GABA_A receptor PAM analogs, researchers from University of California, Irvine identified a highly selective type I α7 nAChR PAM, AVL-3288¹¹⁸. In rodent models, treatment with AVL-3288 in the presence or absence of agonist both corrected the sensory deficits and improved cognition118., 119., 120., 146.. In 2017, AVL-3288 has advanced into a phase I clinical trial for schizophrenia and schizoaffective disorder, which demonstrated that a type I PAM can be safely administered to humans and that it has potential positive neurocognitive effects in CNS disorders¹⁴⁷.

NS1738 (1-(5-chloro-2-hydroxy-phenyl)-3-(2-chloro-5-trifluoromethyl-phenyl)-urea) developed by NeuroSearch and LY2087101 ([2-[(4-fluorophenyl)amino]-4-methyl-5-thiazolyl]-3-thienylmethanone) by Eli Lilly are also type I α7 nAChR PAMs148., 149.. NS1738 was also reported to enhance agonist potency in rescuing scopolamine-induced cognitive deficits¹⁴⁸. Both of NS1738 and LY2087101 have not brought into clinical trials yet.

The first selective type II PAM PNU-120596 (1-(5-chloro-2,4-dimethoxy-phenyl)-3-(5-methyl-isoxazol-3-yl)-urea) developed by Pfizer was shown to not only potentiate the peak α7 current but also delay desensitization of α7 nAChRs¹⁵⁰. Though this compound augmented the procognitive effects of an acetylcholinesterase inhibitor in rodents and non-human primates, it was not able to advance into clinical trial for its potential toxic effects resulting from excessively high calcium influx118., 151.. A-867744 (4-(5-(4-chlorophenyl)-2-methyl-3-propionyl-1H-pyrrol-1-yl)benzenesulfonamide) is type II PAM with moderate potency and efficacy (EC₅₀=1.12 μmol/L; E_max=733% to ACh-evoked α7 current in Xenopus oocytes) developed by AbbVie¹⁵². Other reported type II PAMs such as TQS (4-naphthalene-1-yl-3a,4,5,9b-tetrahydro-3H-cyclopenta[c]quinoline-8-sulfonic acid amide), JNJ-1930942 (2-[[4-fluoro-3-(trifluoromethyl)phenyl]amino]-4-(4-pyridinyl)-5-thiazolemethanol), and RO5126946 ((5-chloro-N-[(1S,3R)-2,2-dimethyl-3-(4-sulfamoyl-phenyl)-cyclopropyl]-2-methoxy-benzamide) also exhibited α7 potentiation effects in vitro and precognition effects in vivo153., 154., 155..

On the basis of the conventional type II α7 nAChR PAM TQS, researchers from Eli Lilly identified a compound named GAT-107 or 4BP-TQS (4-(4-bromophenyl)-3a,4,5,9b-tetrahydro-3H-cyclopenta[c]quinoline-8-sulfonamide), which exhibited potent allosteric agonism and allosteric potentiation at α7 nAChRs¹⁵⁶. Moreover, with GAT-107 as a tool, it is reported that the direct allosteric activation site is located in the interface of α7 nAChR subunits157., 158..

Exploiting α7 nAChR PAMs and ago-PAMs is still in its early stages, and clinical trials of these compounds are still in their infancy. However, with the property of modulating α7 nAChR activity, α7 nAChR PAMs and ago-PAMs represent an additional therapeutic possibility for CNS diseases.

4. Concluding remarks

Abundant literature has shown us the critical role of α7 nAChRs in cognition, learning, memory, and sensory processing in animal models. Compelling preclinical evidence has shown that α7 nAChR agonists and PAMs could enhance cognition and alleviate sensory gating deficiency.

Most clinical trials of α7 nAChR agonists are terminated or suspended. With the limited data, we are not able to assign the cause of clinical failure. However, almost all of the α7 nAChR agonists show cross-activity with 5-HT₃ receptors. Thus, we assume that the lack of selectivity over 5-HT₃ receptors might be one reason for the failure of α7 nAChR agonists in clinical trials. In phase II clinical trials for cognitive deficits in schizophrenia, GTS-21 and ABT-126 showed significant improvement in negative symptoms but not in ameliorating cognitive deficits. In addition, EVP-6124 failed to reach the primary clinical endpoint because of the unexpected high effect of the placebo. Therefore, improper design of clinical trials might be another reason for the failure of α7 nAChR agonists in clinical trials.

As for α7 nAChR PAMs and ago-PAMs, the cytotoxic effect of PNU-120596 indicates that a too-potent activity of type II PAM is not favorable in drug discovery. However, the reported procognition and sensory gating improvement effects in animal models demonstrates a promising future for α7 nAChR PAMs. Moreover, positive results of AVL-3288 in a phase I clinical trial indicates that an α7 nAChR PAM is a potential new therapy for cognitive deficit in schizophrenia. Pharmacological studies on α7 nAChR ago-PAMs have not been reported yet. However, based on the activity of GAT-107 in enhancing α7 nAChR function, ago-PAMs remain a positive choice in developing therapeutic solution in CNS disorders.

Taken together, α7 nAChR agonists and PAMs (including ago-PAMs) remain a viable therapeutic strategy for the treatment of AD, schizophrenia, and other neuropsychiatric disorders. While developing α7 nAChR modulators, selectivity and toxicity profiles should be further improved. And before clinical trials, scientific and well-rounded clinical plans should be designed.