Imaging Cholinergic Receptors in the Brain by Positron Emission Tomography

Cholinergic receptors represent a promising class of diagnostic and therapeutic targets due to their significant involvement in cognitive decline associated with neurological disorders and neurodegenerative diseases as well as cardiovascular impairment. Positron emission tomography (PET) is a noninvasive molecular imaging tool that has helped to shed light on the roles these receptors play in disease development and their diverse functions throughout the central nervous system (CNS). In recent years, there has been a notable advancement in the development of PET probes targeting cholinergic receptors. The purpose of this review is to provide a comprehensive overview of the recent progress in the development of these PET probes for cholinergic receptors with a specific focus on ligand structure, radiochemistry, and pharmacology as well as in vivo performance and applications in neuroimaging. The review covers the structural design, pharmacological properties, radiosynthesis approaches, and preclinical and clinical evaluations of current state-of-the-art PET probes for cholinergic receptors.


INTRODUCTION
Acetylcholine (ACh) is an essential neurotransmitter that is involved in various biological processes.Its role in the central nervous system (CNS) includes cognitive processes such as learning, memory, and attention as well as chemical transmission at the neuromuscular junction and autonomic function in the peripheral nervous system (PNS). 1,2The action of ACh is mediated through the activation of cholinergic receptors, which can be divided into two classes based on their signaling mechanism: G-protein-coupled muscarinic acetylcholine receptors (mAChRs) and ligand-gated nicotinic acetylcholine receptors (nAChRs; Figure 1). 3Among the mAChRs, there are .Typical structures of cholinergic receptors.Schematic illustration of (A) mAChRs and (B) nAChRs.Crystal structures of (C) M4 mAChR (PDB 6KP6) 17 and (D) nAChR (PDB 2BG9). 18ive subtypes (M1, M2, M3, M4, and M5), which are encoded by distinct genes (CHRM1−CHRM5) and have distinct regional distributions in the CNS and peripheral tissues. 4,5M1, M4, and M5 mAChRs are predominantly expressed in the CNS and are critical in normal neuronal function.On the other hand, M2 and M3 mAChRs exhibit much wider expression in the periphery and are involved in cardiovascular and secretory processes and gut motility.
On the other hand, nAChRs are composed of pentameric proteins formed by a combination of homomeric or heteromeric subunits among 12 different subunits (α2−10 and β2−4). 6In the mammalian brain, the most predominant nAChRs are the homomeric α7 nAChRs and the heteromeric α4β2 nAChRs, both of which are involved in a variety of neuronal processes. 1The α7 nAChRs are highly expressed in the hippocampus and cerebral cortex and are associated with cognitive functions such as working memory, long-term memory, and sensory gating. 7−11 Additionally, in

Journal of Medicinal Chemistry
−14 On the other hand, α4β2 nAChRs have been linked to addiction and hedonic processes and are highly expressed in the hippocampus, cerebral cortex, ventral tegmental area, and substantia nigra. 15Varenicline, a partial agonist of α4β2 nAChRs, has been approved for the treatment of smoking cessation. 16−21 Compared to other imaging technologies, PET offers several advantages, such as superior tissue penetration, increased quantification capabilities, and feasibility for clinical translation.−26 In this review, we aim to provide an extensive and current summary of recently discovered PET tracers that selectively target subunits of cholinergic receptors.Additionally, we will discuss their structural diversity, pharmacological properties, and in vivo imaging applications.We hope that this review will facilitate the development of more potent and selective PET ligands targeting cholinergic receptors for clinical use.

PET TRACERS TARGETING MUSCARINIC ACETYLCHOLINE RECEPTORS (MACHRS)
mAChRs can be activated by both the endogenous neurotransmitter acetylcholine (ACh) and the exogenous compound muscarine. 1,3,27As members of the ACh receptor family and G-protein-coupled receptors (GPCRs), mAChRs play a vital role in regulating cholinergic neurotransmission and various Journal of Medicinal Chemistry physiological functions.They signal via the activation of intracellular GTP (guanosine triphosphate)-binding regulatory proteins, also known as G-proteins.Each subtype of mAChRs (M1−5) couples to distinct G-proteins.This allows for the modulation of diverse ion channels and other signaling proteins.The M1, M3, and M5 subtypes preferentially couple to G q/11 proteins, leading to the activation of phospholipase C and an increase in intracellular calcium levels, thereby promoting an excitatory response.On the other hand, the M2 and M4 subtypes primarily couple to G i/o proteins, resulting in the inhibition of adenylate cyclase and a reduction in intracellular cyclic adenosine monophosphate (cAMP) levels. 28The abnormal activity of mAChRs has been linked to the development of several pathological conditions, including cancer, as well as psychiatric and neurological disorders such as AD, schizophrenia, Parkinson's disease (PD), and drug abuse.These findings highlight the importance of mAChRs in physiological processes and the potential therapeutic implications of targeting these receptors for the treatment of various CNS diseases. 29−32 2.1.pan mAChR PET Tracers.The roles of mAChRs in a variety of brain dysfunctions have motivated numerous studies on the development of mAChR PET ligands.Initially, efforts were primarily focused on the development of pan mAChR agents that lack subtype selectivity.The representative pan mAChR PET ligands and corresponding radiolabeling methods are summarized in Figures 2 and 3, respectively.
Quinuclidinyl benzilate (QNB) is one of the most extensively studied pan mAChR antagonists with a reported dissociation constant (K d ) ranging from 30 to 80 pM for M1− 5 mAChRs. 33,34Radiolabeling of QNB with carbon-11 led to the discovery of the pan mAChR PET tracer [ 11 C]QNB (1), which was synthesized by one-pot synthesis of [ 11 C]benzylic acid from [ 11 C]CO 2 , followed by esterification with 3quinuclidinol (Figures 2 and 3A). 35n vivo PET studies in baboons utilizing [ 11 C]QNB (1) have shown high accumulation of radioactivity in mAChR-rich regions such as the striatum (1.84%ID/100 mL; %ID = percentage injected dose) and cerebral cortex (2.35%ID/100 mL), while radioactivity rapidly washed out from the cerebellum (approximately 0.5%ID/100 mL), a region known to be deficient in mAChRs. 36Pretreatment with dexetimide, a well-established mAChR antagonist, resulted in a significant reduction of radioactivity in the striatum and cerebral cortex (>83%) with no significant changes observed in the cerebellum compared to the control.These findings demonstrate the ability of [ 11 C]QNB (1) to specifically bind to cerebral mAChRs in primates.However, clinical validation of [ 11 C]QNB ( 1) has yet to be reported, and considering the substantial difference in binding affinities between (R)-QNB and (S)-QNB (a factor of 100), further evaluation of the more active enantiomer [ 11 C](R)-QNB is warranted.
Despite the focus of this review on PET ligands for mAChRs, an exception was made to discuss radioiodinated QNB due to its significance in the development of mAChR ligands.−39 The enantiomer of (R)-QNB with higher binding affinity was obtained through chiral resolution and was initially assigned as (R,R)-4-[ 123 I]IQNB. 40However, it was later correctly designated as (R,S)-4-[ 123 I]IQNB (2). 41To date, (R,S)-4-[ 123 I]IQNB (2) remains the most extensively investigated mAChR radioligand in clinical studies. 22Initial evaluation of (R,S)-4-[ 123 I]IQNB (2) with single-photon emission computed tomography (SPECT) imaging indicated significant differences in cerebral radioactivity between AD patients and healthy controls. 42,43In healthy individuals, high radioactivity levels were observed in the regions such as the occipital cortex, insular cortex, putamen, and basal ganglia, while the cerebellum revealed an extremely low radioactivity uptake.The regional pattern of (R,S)-4-[ 123 I]IQNB (2) correlated well with the known distribution of mAChR in the human brain.However, in AD patients, a significant reduction of radioactivity was observed in the frontal or posterior temporal cortex compared to healthy controls, indicating defects in mAChR in these areas.Autoradiography studies of (R,S)-4-[ 123 I]IQNB (2) demonstrated alterations of mAChRs in postmortem brain tissues from individuals with dementia with Lewy bodies (DLB), AD, or PD and the corresponding controls. 44In vivo SPECT imaging studies of (R,S)-4-[ 123 I]IQNB (2) in schizophrenia patients showed a remarkably decreased mAChR availability (20−33%) in all brain regions except the pons compared with the age-and gender-matched normal controls. 45Apart from the applications in investigating mAChR availability in healthy and diseased subjects, 45−47 (R,S)-4-[ 123 I]IQNB (2) has been used to evaluate the effect of drug treatment.Studies have shown that disease treatment with clozapine, an atypical antipsychotic, led to significantly lower radioactive signals in all cortical regions of interest, indicating lower mAChR availability in these regions. 48This result suggested that clozapine had high rates of anticholinergic side effects, which was validated by clinical studies.Despite these promising clinical results, it is noted that (R,S)-4-[ 123 I]IQNB (2) suffered from the dependence on blood flow and transport across the blood−brain barrier (BBB), which necessitated two clinical scans and pharmacokinetic modeling to differentiate the parameters of blood flow and transport from receptor density. 49,50This complexity along with other considerations has led to the discontinuation of investigations with (R,S)-4-[ 123 I]IQNB (2) in recent years.
[ 11 C]Scopolamine (3) is another pan mAChR PET tracer that has demonstrated high binding affinity (IC 50 = 0.7 nM).It was synthesized in 1988 through phosphite-mediated reductive methylation reactions with [ 11 C]formaldehyde (Figure 3B). 51valuation of [ 11 C]scopolamine in human subjects indicated favorable whole brain uptake with an average 3.2%ID between 70 and 90 min p.i. 52 The radioactive signals were preferentially cleared from brain regions with low mAChR density, such as the cerebellum and thalamus, following initial perfusiondirected radioactivity delivery.In contrast, radioactivity continued to increase during the 2 h scanning period in mAChR-rich areas, such as the cerebral cortex and basal ganglia.However, the cerebral distribution of [ 11 C]scopolamine (3) at later time points did not align with the known mAChR expression patterns, which has limited its clinical utility.
Benztropine mesylate, also known as Cogentin, is a clinically prescribed anticholinergic drug that is commonly used to alleviate extrapyramidal adverse effects associated with antipsychotic therapy with neuroleptics. 53The evaluation of [ 11 C]benztropine (4) as a PET tracer has demonstrated good brain uptake in mice (3.07%ID/g at 30 min p.i.) and a heterogeneous distribution pattern in baboon and human brains. 54The ratios of radioactive levels between the corpus striatum and the cerebellum in baboon and human brains were further determined to be 1.46 and 1.53, respectively.
Radioactivity continued to accumulate during the 80 min scanning period in almost all brain regions, except the cerebellum, which exhibited extremely low mAChR density.Pretreatment with scopolamine or benztropine resulted in a remarkable reduction of radioactive levels in all baboon brain regions (35.8% in the corpus striatum, 32.9% in the cortex, and 5.3% in the cerebellum), indicating a favorable binding specificity of [ 11 C]benztropine (4) toward mAChRs.In addition, metabolic analysis of [ 11 C]benztropine (4) in human plasma has demonstrated reasonable in vivo stability with 83% parent fraction at 30 min post tracer injection.Despite these promising results, the clinical application of [ 11 C]benztropine (4) as a mAChR PET tracer has yet to be fully explored.
(+)2α-Tropanyl benzilate (TRB) is a potent mAChR antagonist with high binding affinity (IC 50 = 0.7 nM), 55 and its radiolabeling with carbon-11 gave rise to the mAChR radioligand [ 11 C]TRB (5). 56,57Unfortunately, [ 11 C]TRB (5)  was not found to be a promising mAChR radioligand due to its high binding rate and low dissociation rate, making it a less ideal choice for imaging studies. 57,58-Methyl-4-piperidylbenzilate (NMPB) is a potent antagonist of mAChR with an IC 50 value of 1.8 nM and K i value of 0.3 nM.The compound is amenable for labeling with carbon-11 to produce [ 11 C]NMPB (6), also known as [ 11 C]4-MPB (6).59 Similar to [ 11 C]benztropine (4), radioactivity continued to increase in all human brain regions except the cerebellum during 60 min PET imaging scanning, and the uptake in the frontal cortex reached a peak of ∼17%ID/L at 60 min post tracer injection.Notably, [ 11 C]4-MPB ( 6) can be used to measure age-related changes in mAChR activity.By using a compartment model with the radioactivity in the cerebellum as an input function, studies have shown a significant age-related decrease in the specific binding of [ 11 C]4-MPB (6) in several brain regions, such as the frontal cortex, temporal cortex, and striatum, in both monkeys and humans.The decrease was found to be even more pronounced in humans, reaching approximately 45%.59,60 Following the initial studies, two congeners with N-ethyl ([ 11 C]4-EPB (7), K i = 0.5 nM) and Npropyl ([ 11 C]4-PPB (8), K i = 36 nM) 61 replacing the Nmethyl residue in [ 11 C]4-MPB (6) were synthesized using [ 11 C]ethyl iodide and [ 11 C]propyl iodide, respectively (Figure 3C).62 PET imaging studies of [ 11 C]4-EPB (7) in nonhuman primates (NHPs) revealed brain kinetics similar to [ 11 C]4-MPB (6), in which the uptake in all brain regions except the cerebellum gradually increased over a 90 min scanning period.However, PET imaging studies of [ 11 C]4-PPB (8) revealed different kinetics with the uptake in all brain regions reaching a peak at around 10 min post tracer injection and then gradually decreased over time.
Regioisomers of [ 11 C]4-MPB (6) with regard to the piperidyl position were also investigated, resulting in a pair of enantiomers: the more active [ 11 C](+)3-MPB (9) (K i = 1.7 nM) and its inactive isomer [ 11 C](−)3-MPB. 63Similar to [ 11 C]4-MPB (6), kinetic analysis of [ 11 C](+)3-MPB (9) in conscious monkeys showed an age-related decrease of specific binding in the frontal cortex, temporal cortex, and striatum, reflecting reduced mAChR densities. 64By utilizing PET monkey studies with [ 11 C](+)3-MPB (9), Tsukada and coworkers demonstrated a remarkable positive correlation between mAChR occupancy and oxybutynin or scopolamineinduced cognitive impairment in the striatum, cortices, brainstem, and thalamus. 65,66Moreover, the thresholds of mAChR occupancy that could induce cognitive impairment have been proposed to be around 20−30% in the brainstem and around 30−40% in the cortices.It is suggested that a favorable anticholinergic agent should not exceed these thresholds.Furthermore, the ability of [ 11 C](+)3-MPB (9)  to detect mAChR binding alterations in patients with chronic fatigue syndrome (CFS) has also been demonstrated.Studies have revealed that brain binding of [ 11 C](+)3-MPB (9) in CFS patients with positive tests of CFS(+) antibodies was significantly lower than that in CFS patients with negative results of CFS(−) antibodies and normal controls (Figure 4). 67ructural modification of  (11).This is likely attributed to a higher dissociation rate constant.Additionally, it was observed that the sensitivity of these radioligands toward ACh had a negative correlation with their binding affinities, and the least potent congener, [ 11 C](+)3-PPB (11), proved to be the most sensitive to the concentration of endogenous ACh.However, the complex synthesis of [ 11 C](+)3-PPB (11), which involved the technically challenging preparation of [ 11 C]propyl iodide, has hampered its widespread application.Recently, radioligand [ 11 C](S,R)-12, an analogue of [ 11 C](+)3-PPB (11), was developed, which exhibited comparable binding affinity to mAChRs with a K i value of 3.5 nM in comparison to [ 11 C](+)3-PPB (11) with a K i value of 7.9 nM. 69PET studies of [ 11 C](S,R)-12 in NHPs have yielded promising results, with high and heterogeneous brain uptake (maximum standardized uptake value (SUV max ) = ∼8.7 in the caudate and putamen), good specific binding, and a comparable kinetic profile to that of [ 11 C](+)3-PPB (11).These findings suggest that [ 11 C]-(S,R)-12 may have potential as a PET probe for monitoring changes of endogenous ACh concentrations.Further validation through preclinical animal models and clinical studies is required to fully assess the utility of [ 11 C](S,R)-12 as a PET probe.
The properties and molecular imaging results of pan mAChR PET ligands discussed in this section are displayed in Table 1.

Subtype-Selective mAChR PET Tracers
. mAChR ligands can be divided into three categories based on their binding sites: orthosteric, allosteric, and bitopic. 78Orthosteric ligands bind to the same site as the endogenous ligand, acetylcholine, while allosteric ligands bind to a distinct site and modulate the activity of the receptor.Bitopic ligands, on the other hand, possess the ability to bind to both the orthosteric and the allosteric sites.As depicted in Figure 5, the orthosteric site of mAChRs has been extensively studied for decades, leading to a thorough understanding of its structure−activity relationships (SARs) and high-affinity binding.However, achieving subtype selectivity at the orthosteric site has proven  to be challenging.On the other hand, the allosteric site provides a promising avenue for achieving subtype selectivity.However, modifying certain scaffolds, as done with many orthosteric drugs, may pose challenges in improving the binding affinity. 79Notably, bitopic ligands have the potential to target both the orthosteric and the allosteric sites, combining high binding affinity with subtype selectivity.This offers an exciting opportunity for future development of mAChR ligands.
Therefore, compared to pan mAChR ligands, which lack adequate receptor subtype selectivity, the development of allosteric or bitopic mAChR PET tracers with high subtype selectivity has been a major focus in recent years.These subtype-selective tracers enable the specific imaging of certain subtypes of mAChRs, thereby providing valuable information on their expression and distribution in the brain.This, in turn, can aid in the development of more targeted and effective therapeutic approaches for several pathological conditions, such as cancer as well as psychiatric and neurological disorders.To date, subtype-selective PET tracers have been primarily developed for M1, M2, and M4 mAChRs, while rare examples have been reported for M3 and M5 subtypes.This limitation can be attributed to the underdevelopment of nonradiolabeled ligands/drug candidates specifically targeting these two subtypes.Selective PET tracers that target specific mAChRs are herein discussed.
Subtype-Selective PET Tracers for M1 Receptor.The M1 receptor accounts for up to 60% of the total expression of mAChRs in the CNS and is predominantly found in postsynaptic glutamatergic and striatonigral pyramidal neurons.It is widely expressed in major forebrain areas, including the cerebral cortex, hippocampus, and neostriatum, and plays a crucial role in regulating synaptic plasticity, neurotransmission, and learning and memory.M1 receptor has been implicated in the cognitive decline associated with various CNS pathologies, including AD, and activation of M1 receptor has been shown to improve learning and memory deficits in affected patients. 80everal PET ligands that selectively target M1 receptors have been developed, and the representative M1-selective PET ligands and radiolabeling methods are given in Figures 6 and 7, respectively.
The development of M1 receptor PET ligands commenced with the synthesis of the agonist tracers [ 11 C]xanomeline (18)  and [ 11 C]butylthio-TZTP (19). 81These tracers exhibited high affinity (K i < 10 nM) to M1 receptors and high selectivity toward M2 receptors. 82In preclinical studies, [ 11 C]xanomeline (18) and [ 11 C]butylthio-TZTP ( 19) could efficiently penetrate the BBB of cynomolgus monkeys with ∼10% of injected dose (%ID) and accumulated in the cortex and striatum.However, in healthy volunteers, despite high brain uptake (>5%ID), [ 11 C]xanomeline (18) had a homogeneous distribution in the brain, while [ 11 C]butylthio-TZTP (19) only exhibited slightly higher signals in the striatum and neocortex than that of the cerebellum.The lack of desirable PET performance was likely attributed to their poor selectivity toward sigma binding sites.
The stereoselective synthesis of fluorine-containing quinuclidinyl benzilate (QNB) derivatives led to the identification of compounds (R,R)-FMeQNB and (R,S)-FMeQNB, which exhibited high affinity (K i < 1 nM) and moderate selectivity toward M1 and M2 receptors. 83Among these compounds, (R,R)-FMeQNB demonstrated 8-fold higher affinity for M1 over M2 receptors, with K i values of 0.11 and 0.84 nM, respectively.Upon 18 F labeling (Figure 7A), (R,R)-[ 18 F]-FMeQNB (20) showed higher signals in the regions of the rat brain known to have high M1 receptor expression, including the cortex, hippocampus, and caudate, albeit at low levels (<0.45%ID/g). 84However, the data obtained from coinjection and displacement experiments with (R,R)-[ 18 F]FMeQNB (20)  were complex and may be influenced by its binding to M1/M2 receptors.AF150(S) is an agonist of M1 receptors with moderate affinity (K i = 390 nM) and high selectivity toward M2 receptors (22 000-fold). 85[ 11 C]AF150(S) (21) was synthesized and subsequently evaluated by in vitro rat brain autoradiography (ARG), ex vivo biodistribution, PET imaging, and radiometabolite analysis. 86,87The results of these studies indicated that while [ 11 C]AF150(S) (21) could rapidly penetrate the BBB, it also had a relatively fast clearance rate from all brain regions.Despite this, apparent specific binding of [ 11 C]AF150(S) (21) was observed in brain regions that have a high density of M1 receptors, suggesting its potential for imaging the active pool of M1 receptors in vivo.However, the rapid metabolism, hydrophobicity, and moderate binding affinity present significant challenges for its utility in PET studies.
As a positive allosteric modulator (PAM) and agonist, GSK1034702 has moderate affinity (EC 50 = 7.9 nM) for M1 receptors and high selectivity over other muscarinic receptors (>790 nM).In order to evaluate its utility as a ligand for PET imaging, [ 11 C]GSK1034702 (22) was prepared through Pdcatalyzed methylation of aryl stannane precursor with a reasonable yield (10% decay corrected, Figure 7B). 88The ligand was evaluated in both NHP and human PET imaging. 89owever, the results of these imaging studies showed that [ 11 C]GSK1034702 (22) displayed homogeneous and marginal regional heterogeneity in the brain, with regional distribution volume (V T ) values of 3.9 and 4.2 in NHP and the human brain, respectively.These results indicated that [ 11 C]-GSK1034702 (22) was not suitable for imaging M1 receptors in the brain.Nevertheless, the study provided important information to support the clinical development of GSK1034702.More recently, a 18 F-labeled regioisomer of GSK1034702, [ 18 F]23, has been developed, but no preclinical evaluation has been reported yet. 90The preparation of [ 18 F]23 was achieved via alcohol-enhanced Cu-mediated radiofluorination with a moderate yield of 17%.
LSN3172176 is a bitopic M1-preferring agonist and partial allosteric modulator, which has several similarities with GSK1034702 in chemical structure and has high affinity to M1 receptors (K d = 1.5 nM using [ 3 H]LSN3172176). 91lthough the K d values for [ 3 H]LSN3172176 at M1 (K d = 1.5 nM), M2 (K d = 0.6 nM), and M4 (K d = 0.8 nM) were overlapping in human recombinant cells, the dominant expression of M1 mAChRs in the brain regions may confer its selectivity in vivo.Further experiments using M2 or M4 knockout mice are necessary to validate this postulation in addition to M1 KO target occupancy studies.In PET imaging of rhesus monkeys, [ 11 C]LSN3172176 (24) exhibited high brain uptake (V T 10−18 mL/cm 3 ) in M1 receptor-rich regions of the basal ganglia, frontal cortex, and hippocampus but with low V T in the cerebellum (poor receptor expression). 92The high specific binding of [ 11 C]LSN3172176 (24) to M1 receptors was strongly demonstrated by the pretreatment experiments with scopolamine (a muscarinic antagonist) and AZD6088 (a M1 receptor-selective partial agonist).When advanced to PET imaging in healthy volunteers, [ 11 C]-LSN3172176 (24) showed selective accumulation in the striatum, neocortical regions, and white matter, and the lowest radioactivity was found in the cerebellum. 93Thus, [ 11 C]-LSN3172176 (24) has excellent test−retest reproducibility in estimation of regional acetylcholine concentration variation in living human brain. 94lthough several M1 radioligands have progressed to clinical trials in healthy volunteers, further efforts are needed to facilitate their translation into the diagnosis and treatment of pathophysiological conditions.It is worth noting that most in vivo evaluated M1 radioligands to date have been limited to 11 C-labeled compounds.Therefore, future attention should focus on the development and application of more 18 F-labeled ligands for M1 receptors.This would not only enable more sensitive and accurate imaging of M1 receptors in vivo but also expand the scope of applications of M1 PET imaging in various CNS disorders.
The properties and molecular imaging results of M1selective PET ligands discussed in this chapter are summarized in Table 2.
Subtype-Selective PET Tracers for M2 Receptors.The M2 receptor is also widely expressed in the CNS and is among the earliest targets for the development of subtype-selective mAChR radiotracers. 95It is abundantly expressed in the basal forebrain, thalamus, neocortex, and hippocampus, especially in noncholinergic terminals in the cortex and hippocampus.M2 receptor also plays a crucial role in learning, recognition memory, and hippocampal plasticity.In rat models, M2 antagonism has been shown to rescue cognitive deficits associated with neurodegeneration. 96In light of its significance in the brain, PET ligands that selectively target M2 have been developed and evaluated (Figures 8 and 9).
AF-DX384, an M2 receptor antagonist with high affinities of 0.28 and 28 nM (K i values for high-and low-affinity binding sites, respectively), represented a promising starting point for PET tracer development despite its marginal selectivity (K i (M1)/K i (M2) = 5). 97As a result, [ 11 C]AF-DX384 (25)  was synthesized via urea formation of aniline precursor with [ 11 C]phosgene in a high decay-corrected yield of 25−40% (Figure 9A). 98However, the low BBB penetration and the presence of brain-permeable radiometabolites have impeded its further application in PET imaging.A structurally similar compound BIBN99 was reported to have moderate affinity for M2 receptors (pK i = 7.6 nM) and improved selectivity toward M1 receptors (K i (M1)/K i (M2) = 33).[ 11 C]BIBN99 (26) was obtained with a decay-corrected yield of 26% from [ 11 C]ethyl iodide, but to date, no studies have been conducted to investigate its potential use as a PET imaging agent. 99Despite sharing similar chemical structures with several reported M1selective PET ligands, such as [ 11 C]xanomeline (18) and [ 11 C]butylthio-TZTP (19), [ 18 F]FP-TZTP (27) proved to exhibit higher binding affinity for M2 receptors (K i = 2.2 nM) in comparison to M1 receptors (K i = 7.4 nM). 100 The selectivity was confirmed through ex vivo ARG studies in knockout mice 101 as well as in vitro studies in muscarinic receptor-expressing cells and rat brain sections, which suggest that the slower dissociation kinetics of [ 18 F]FP-TZTP (27)  from M2 receptors may be responsible for this selectivity. 102iodistribution results in rats have demonstrated that [ 18 F]FP-TZTP ( 27) had rapid uptake and clearance from the brain with relatively uniform accumulation across different brain regions, ranging from 1.1%ID/g in the cortex to 0.6%ID/g in the thalamus at 15 min postinjection. 100PET imaging in humans showed that the distribution of [ 18 F]FP-TZTP ( 27) using V T values in the cortical, subcortical, and cerebellar areas is consistent with M2 receptor expression and correlates with age. 103However, additional clinical studies are needed to validate the practical application of [ 18 F]FP-TZTP (27) in human M2 receptor PET imaging. 104,105R,S)-FMeQNB has been found to exhibit higher affinity (K i = 0.13 nM) for M2 receptors compared to (R,R)-FMeQNB (20) (Figure 6) and demonstrated a 7-fold selectivity over M1 receptors (K i = 0.89 nM).83 In a study using rat brain, (R,S)-[ 18 F]FMeQNB (28) displayed a uniform distribution among different brain regions.84 In a coinjection experiment with unlabeled (R,S)-FMeQNB, the most significant inhibition (97%) of radioactivity uptake was observed in the heart, an organ known to have a high concentration of M2 receptors.Radioactivity was also blocked by 36−54% in all brain regions.Furthermore, in a displacement experiment with unlabeled (R,S)-FMeQNB at 60 min postinjection of the radioligand, a higher reduction of 30−50% radioactivity was observed in the pons, medulla, and cerebellum, the regions known to contain a high proportion of M2 receptors.These results indicated a reasonable subtype selectivity of (R,S)-[ 18 F]FMeQNB (28) in the rat brain.In rhesus monkey, prolonged uptake and retention of (R,S)-[ 18 F]FMeQNB (28) was observed in the brain.In a coinjection experiment with unlabeled (R,S)-FMeQNB, significant radioactivity inhibition was observed in the heart, thalamus, and pons but not in the cerebellum, which also has a high concentration of M2 receptor.(R,S)-[ 18 F]FMeQNB (28) was found to degrade rapidly in monkey plasma with less than 5% of the parent compound remaining at 30 min postinjection, which may be likely induced by radiodefluorination in vivo as confirmed by significant bone uptake.
Compared to other mAChR subtypes, the development of M2-selective radioligands has lagged far behind.Most M2 tracers were reported before 1997, and no significant advances have been made in the last two decades.This is partly due to the lack of ligands that exhibit high selectivity between M2 and M1.Given this scenario, future research efforts should focus on the development of highly selective M2 ligands based on recent development of an allosteric binding mechanism (cf. Figure 5) on mAChR ligands.
The properties and molecular imaging results of M2selective PET ligands discussed in this section are given in Table 3.
Subtype-Selective PET Tracers for M4 Receptors.M4 receptor is predominantly expressed in brain regions such as the striatum, hippocampus, and cortex and has been implicated in several CNS disorders, including AD and schizophrenia. 106ence, M4 receptor presents a promising target for the treatment of these diseases.Additionally, M4 is coexpressed with dopamine receptors on striatal projection neurons, 107 regulating dopamine release and inhibiting dopamine D1 receptor function.This renders M4 as a potential target for the treatment of PD as significant loss of dopamine neuron projecting to the striatum has been implicated in the disease. 108Given the critical role of M4 in various brain diseases, the development of subtype-selective M4 PET ligands has been an active area of research, resulting in the design and synthesis of PET tracers with various structural cores for M4 receptors, as shown in Figures 10 and 11.
In 2019, [  10) were developed from a series of potent PAMs. 111,112These radioligands were developed with the intention of being used as tools for the quantification of therapeutic drug occupancy of the M4 receptor via PET imaging.The lead radiotracer [ 11 C]34 was prepared through palladium-catalyzed cyanation reaction of aryl bromide precursor with [ 11 C]HCN (Figure 11A).non-displaceable binding potential (BP ND , termed as the ratio of available receptor sites to the equilibrium distribution constant) values of 0.15−0.37 in rhesus monkey brain with PET imaging, which was considered insufficient for accurate   quantification of the therapeutic drug occupancy of the M4 receptor via PET imaging. 111ortunately, compound MK-6884 was identified as a highly potent and selective PAM of the M4 receptor with high affinity and selectivity (K i = 0.19 nM and greater than 3600-fold selectivity over other receptors).In addition, in vitro binding potential values were favorable with B max /K d = 14.4 and 7.8 for monkey and human brain tissues, respectively.In ARG studies using [ 3 H]MK-6884, regional binding density was found to be highest in the striatum, followed a decreasing order in the cortex ≈ hippocampus > thalamus ≫ cerebellum in both monkey and human brain sections.Furthermore, the binding signal was significantly enhanced by the orthosteric mAChR agonist carbachol. 113In a PET study with rhesus monkeys, [ 11 C]MK-6884 (36) was found to rapidly penetrate the BBB and accumulate in the striatum with a robust BP ND value of 0.83.In addition, the specific binding of [ 11 C]MK-6884 (36)  to M4 receptors in the brains of rhesus monkeys was confirmed through blocking experiments with another M4 PAM. 111 Encouraged by these findings, [ 11 C]MK-6884 (36) was subsequently employed in imaging studies of M4 receptors in healthy volunteers and patients suffering from AD (Figure 12). 113,114These studies revealed a significant difference of BP ND values observed between the temporal cortex of healthy elderly participants (0.83 ± 0.15) and AD patients (0.46 ± 0.18).Overall, the development and application of [ 11 C]MK-6884 (36) as a radioligand highlights its utility in the discovery and evaluation of M4 PAMs as well as the study of M4 receptor changes in neuropathological conditions such as AD.
The development of M4-selective radioligands has faced several challenges, and most of the currently available radioligands have limitations.However, [ 11 C]MK-6884 (36)  has shown promise and has been successfully translated into human use, indicating that further investigation is warranted.Additionally, the use of longer lived isotopes such as 18 F may help overcome some of the limitations associated with shortlived isotopes and may be a priority for future research in the development of M4 radioligands.
The properties and molecular imaging results of M4selective PET ligands discussed in this section are shown in Table 4.

PET TRACERS TARGETING NICOTINIC ACETYLCHOLINE RECEPTORS (NACHRS)
nAChRs are a type of transmembrane receptor that plays a vital role in fast synaptic transmission in the CNS.These receptors can be activated by both endogenous ligand, ACh, and exogenous ligands, such as nicotine. 6,117,118They are also involved in the regulation of processes like cell excitability, transmitter release, and neuronal integration by acetylcholinemediated innervation, which contributes to their critical role in modulating a wide range of physiological processes, including pain, fatigue, cognition function, sleep, and arousal. 119−122 Neuronal nAChRs are pentamers of heteromeric or homomeric combinations of α (α2-α10) and β (β2-β4) subunits, which gives rise to multiple nAChR subtypes. 119In this section, we will focus specifically on the development of PET tracers for the two most expressed nAChR subtypes found in mammalian brains: α7 and α4β2.
3.1.PET Tracers for Imaging α7 nAChRs.The α7 nAChR subtype is unique as it is a homomeric complex that contains five equivalent agonist-binding sites in the extrac-  ellular domain.These receptors are expressed on both pre-and postsynaptic membranes and act as modulators of circuit activity in the brain. 123Studies have shown that α7 nAChRs are involved in the pathogenesis of various neurological disorders. 124−127 Varieties of PET tracers for imaging α7 nAChRs have been designed and synthesized (Figures 13 and 14).
GTS-21, also known as DMXB-A, is a partial α7 nAChR agonist that has been demonstrated to improve memory in multiple animal models. 128,129This compound has also been translated into clinical trials for the enhancement of cognition in patients with schizophrenia. 10,130In 2007, Kim and colleagues reported the radiolabeling of GTS-21 with carbon-11 at two different positions as well as the radiolabeling of its two in vivo metabolites, thereby creating four radioligands:  13). 131The pharmacokinetic properties of [2-11 C]GTS-21 (40) and [4-11 C]GTS-21 (41) were evaluated in baboons utilizing PET.These studies revealed that both radioligands had good initial brain uptake, reaching a peak of 0.036 and 0.037%ID/cc at 1.4 and 1.7 min post tracer injection, respectively.Additionally, the tracers exhibited a rapid clearance with a half-life of less than 15 min.Further evaluation of another ligand [4-11 C]2-OH-GTS-21 showed significant uptake in the baboon brain, while [2-11 C]4-OH-GTS-21 (42) had negligible accumulation.These findings suggest that the metabolite 2-OH-GTS-21 may partially contribute to the therapeutic effects of GTS-21.
As the first full α7 nAChR agonist, AR-R17779 (44) features a 1-aza-bicyclo[2.2.2]octane scaffold and shows reasonable binding affinity (K i = 92 nM) and selectivity toward α7 nAChRs over α4β2 nAChRs (K i = 16 μM). 132Some derivatives of AR-R17779 (44) were patented as α7 nAChR agonists; however, there is no available pharmacological data. 133The radiolabeling of one compound in this series with [ 11 C]phosgene resulted in radioligand [ 11 C]2 (45). 134Ex vivo biodistribution studies of [ 11 C]2 (45) in rats demonstrated reasonable brain uptake with 0.8−1.2%ID/gamong various brain regions.However, no preferential radioactivity accumulation was observed in regions known to be rich in α7 nAChR, such as the hippocampus, colliculi, and pons.Pretreatment studies utilizing nicotine and nonradioactive compound 2 indicated that there was no specific binding of this ligand toward α7 nAChRs.As a result, [ 11 C]2 ( 45) is not a suitable PET ligand for in vivo visualization of α7 nAChRs.In 2005, Pomper and co-workers described a method for the radiolabeling of two potent α7 agonists containing an 1-azabicyclo[2.2.2]octane scaffold with carbon-11.These radioligands exhibited high binding affinity and selectivity toward α7 nAChRs over α4β2 nAChRs (K i = 0.54 nM and α4β2/α7 > 22 000 for [ 11 C]3 (46); K i = 5.8 nM and α4β2/α7 = 14 000 for [ 11 C]4 ( 47)). 135Notably, the radiosynthesis of [ 11 C]3 (46) was achieved in a reasonable radiochemical yield (6.5%) via a reductive amination reaction of the corresponding desmethylated precursor with [ 11 C]formaldehyde, which was in situ generated from [ 11 C]CO 2 .However, biodistribution studies of these two radioligands in mice revealed limited brain uptake, homogeneous distribution among various brain regions, and minimal receptor blockade, thus impeding their further in vivo applications.Subsequent studies led to the discovery of a potent α7 nAChR PET ligand [ 11 C](R)-MeQAA (48) (K i = 41 nM). 136Evaluation of the ligand [ 11 C](R)-MeQAA (48) in mice demonstrated high initial brain uptake, reaching a peak of 7.68%ID/g at 5 min post tracer injection.A heterogeneous radioactivity distribution was observed with higher levels in the hippocampus and lower levels in the cerebellum, which is in agreement with the known α7 nAChR density in the mouse brain.Notably, pretreatment with MLA, an α7 nAChR antagonist, resulted in a significant 32% blockade of radioactive signals in the hippocampus, providing evidence for specific binding of the ligand to α7 nAChR. 137Follow-up PET imaging studies of [ 11 C](R)-MeQAA (48) in rhesus monkeys also demonstrated regional brain uptake with high levels in the thalamus (approximately 0.033%ID/mL), moderate levels in the cortex, and low levels in the cerebellum, which is consistent with the α7 nAChR distribution in the monkey brain.
In addition to the 1-aza-bicyclo[2.2.2]octane scaffold, 1,4diaza-bicyclo[3.2.2]nonane is also prevalent in several α7 nAChR agonists.One such example is SSR180711, which has been shown to have a favorable binding affinity for rat (K i = 22 nM) and human (K i = 14 nM) α7 nAChRs, respectively. 141,142adiolabeling of SSR180711 was achieved through two approaches, namely, radiobromination of the corresponding phenyl−stannane precursor with bromine-76 using chloramine-T as oxidant (Figure 14B) and carbamate formation of 1,4-diazabicyclo[3.2.2]nonane with [ 11 C]phosgene.These methods provided two radioligands [ 76 Br]SSR180711 (53) 143 and [ 11 C]SSR180711 (54), 144 respectively.While no biological data was disclosed for [ 11 C]SSR180711 ( 54), [ 76 Br]-SSR180711 (53) was evaluated in conscious monkeys, revealing a heterogeneous distribution pattern and favorable uptake in the brain.The highest radioactivity level was detected in the temporal cortex, reaching a peak of ∼0.0135% ID/mL at around 30 min post tracer injection.In contrast, the cerebellum, a α7 nAChR-deficient region, exhibited the lowest radioactive signals.Pretreatment with nonradioactive SSR180711 rendered remarkably decreased uptake of [ 76 Br]-SSR180711 (53) in various brain regions with the exception of the cerebellum, whereas no obvious alteration of radioactivity levels was observed upon the administration of the selective α4β2 nAChR agonist A85380.
Further studies led to the discovery of a SSR180711 analogue, [ 11 C]CHIBA-1001 (55), as a potent α7 nAChR PET probe. 143 55) exhibited comparable uptake in the cerebellum (SUV max = ∼4.7 at around 13 min p.i.) to that in the cortex, which was in contrast to the low radioactivity accumulation in the cerebellum of monkeys.This difference in uptake was likely attributed to its relatively low binding affinity (K d = 120−180 nM). 146n 2009, a potent 18 F-labeled 1,4-diaza-bicyclo[3.2.2]nonane-derived PET ligand [ 18 F]NS10743 (56) was developed with improved binding affinity to α7 nAChRs (K i = 11.6 nM, K d = 8.99 nM). 147,148Ex vivo biodistribution studies of [ 18 F]NS10743 (56) in mice indicated good BBB permeability with a peak brain uptake of 4.83%ID/g at 5 min p.i.The target specificity was also validated through preinjection with the selective α7 nAChR agonist SSR180711, which resulted in 28% reduction of radioactive signals in the mouse brain.In vivo PET imaging studies of [ 18 F]NS10743 (56) in juvenile pigs also demonstrated good brain uptake with high radioactive signals in α7 nAChR-rich regions, such as the hippocampus, thalamus, colliculi, and temporal cortex. 149The highest uptake was observed in the colliculi, with a SUV max of 2.48, while the olfactory bulb exhibited the lowest radioactivity level (SUV max = 1.53 at 7 min p.i.).To further evaluate the specificity of [ 18 F]NS10743 (56), preadministration of the α7 nAChRselective ligand NS6740 was conducted, which decreased the binding of [ 18 F]NS10743 (56) by 16−35% based on BP ND (the ratio of specifically bound radiotracer to that of nondisplaceable radiotracer) throughout the brain, except for the olfactory bulb.These results suggest that while the binding of [ 18 F]NS10743 (56) to α7 nAChR is reasonable, it may not be sufficiently specific in vivo.
Another high-affinity and selective α7 nAChR partial agonist NS14492 was radiolabeled with carbon-11 to provide the radioligand [ 11 C]NS14492 (57) (K i = 2.2 nM for α7 nAChRs; K i = 2.8 μM for α4β2 nAChRs). 150In vivo evaluation in Danish Landrace pigs indicated that [ 11 C]NS14492 (57) could readily penetrate the BBB with high radioactive signals in the thalamus and cerebral cortex, moderate signals in the striatum, and relatively low signals in the cerebellum, which was in accordance with the distribution of α7 nAChRs in the pig brain.Notably, the radioactive signals were found to be dosedependently blocked by preadministration of NS14492 and SSR180711, resulting in up to 81% reduction, supporting that [ 11 C]NS14492 (57) selectively binds to α7 nAChRs.These results demonstrate that [ 11 C]NS14492 (57) is a promising radioligand for imaging of α7 nAChRs in the brain.
Recently, [ 18 F]NS14490 (58) was developed as a 18 Flabeled α7 nAChR PET ligand with high binding affinity and selectivity (K i = 2.5 nM for α7 nAChRs; K i > 1 μM for α4β2 nAChRs). 151However, subsequent in vivo evaluation revealed limited brain uptake of the radioligand.Ex vivo biodistribution studies in mice demonstrated a low brain uptake level of only 0.16%ID/g at 5 min p.i. 152 Similarly, in vivo PET imaging studies in pigs revealed a SUV max of 0.54. 153These results indicate that [ 18 F]NS14490 (58) may not be an ideal radioligand for in vivo mapping of cerebral α7 nAChRs.

Journal of Medicinal Chemistry pubs.acs.org/jmc
Perspective compared to [ 18 F]DBT10 (60), reaching a peak of 7.5%ID/g at 5 min post tracer injection. 154,155,157Furthermore, a heterogeneous radioactivity distribution was observed for [ 18 F]ASEM (59) with high levels in the hippocampus, frontal cortex, and colliculus, intermediate levels in the striatum, and low levels in the cerebellum.This distribution pattern aligns with the known α7 nAChR density in the mouse brain.The uptake of [ 18 F]ASEM (59) could be dose-dependently blocked by pretreatment with the α7 nAChR agonist DMXB-A or SSR180711, indicating good in vivo specific binding.Notably, in a rodent model of schizophrenia, mutant DISC1 mice, the uptake of [ 18 F]ASEM (59) in the brain was remarkably lower in comparison to control mice, which is in line with postmortem human results. 157Further PET imaging studies of [ 18 F]ASEM (59) in baboons also demonstrated excellent BBB penetration with a peak SUV of 5 at 5 min p.i. 157 Regional radioactivity uptake was observed with V T values ranging from 14 in the cerebellum to 23 in the thalamus.The in vivo specific binding in α7 nAChR-rich NHP brain regions was determined to be approximately 80−90%, as demonstrated by dosedependent blockade of radioactivity uptake with α7 nAChR partial agonist SSR180711.Similar results were also obtained in a study evaluating [ 18 F]ASEM (59) in pigs. 158ncouraged by the promising preclinical results, the first in human PET studies of [ 18 F]ASEM (59) were conducted, which revealed high and regional brain uptake (SUV max = 4). 159,160Specifically, high levels of radioactivity accumulation were found in the parietal cortex (V T = 22), putamen (V T = 21.8),thalamus (V T = 20.9),temporal lobes (V T = 19.7),cingulate (V T = 19.6),frontal lobes (V T = 19.3),and hippocampus (V T = 17.9), and a significantly lower radioactivity level was observed in the corpus callosum (V T = 9.9).The regional distribution pattern of [ 18 F]ASEM (59) in humans matches the postmortem data from humans and NHPs.Furthermore, the potential relationship between α7 nAChR availability and healthy aging was also assessed using [ 18 F]ASEM (59) (Figure 15). 160,161By utilizing Spearman's rank correlation analysis, the V T values of [ 18 F]ASEM (59) in nine regions of interest in the brain, including the cortical and subcortical regions, were found to be positively correlated with age.Additionally, the tissue volume ratios of six brain regions, including the striatum and five cortical regions (temporal, occipital, cingulate, frontal, and parietal cortices), were negatively correlated with age, which was in accordance with the previous reports on the regional decreases of brain volume over normal aging.Recently, [ 18 F]DBT10 (60) was also evaluated in NHPs and exhibited relatively lower but still notable brain uptake (SUV = 2.9−3.7)when compared to that of [ 18 F]ASEM (59). 162Given the structural similarity of [ 18 F]DBT10 (60) and [ 18 F]ASEM (59), no further validation of [ 18 F]DBT10 (60) was pursued.
In 2018, Zhang and co-workers developed a new 18 F-labeled α7 nAChR ligand [ 18 F]YLF-DW (61) (K i = 2.98 nM). 163reliminary biodistribution and PET imaging studies in mice showed favorable brain uptake (8.98%ID/g at 5 min p.i.); however, further validation of the specific binding and evaluation in higher species is necessary.Most recently, [ 18 F]YLF-DW (61) was applied in identifying vulnerable atherosclerotic plaques in carotid arteries. 164Further studies conducted in ApoE −/− mouse models have demonstrated that [ 18 F]YLF-DW (61) has the potential to be used as a diagnostic tool for the detection of atherosclerotic plaques.In these studies, a significant radioactive signal was observed in the carotid artery of ApoE −/− mice, while no radioactivity was accumulated in the carotid artery of normal control mice.This was consistent with the findings from oil red staining and immunohistochemistry.
The properties and molecular imaging results of α7 nAChRselective PET ligands discussed in this section are demonstrated in Table 5.

PET Tracers for
Imaging α4β2 nAChRs.The α4β2 subtype of nAChRs is a heteromeric receptor composed of α4 and β2 subunits, and it is the most abundant subtype of nAChRs in the brain. 167Due to their critical roles in a variety of neurological disorders such as nicotine addiction, 168,169 AD, 170,171 PD, 172,173 and depression, 174,175 α4β2 nAChRs have become one of the most extensively studied subtypes of nAChRs.PET imaging has emerged as a promising tool for measuring α4β2-nAChR levels in vivo, which could aid in disease diagnosis and the identification of patient subpopulations that would benefit from α4β2-nAChR-targeted therapy.Figures 16 and 17 provide an overview of representative α4β2 nAChR-selective PET ligands and selected radiosynthetic methods, respectively.
The first attempts to utilize PET studies for measuring α4β2-nAChRs involved the use of the radiolabeled nicotine, [ 11 C]nicotine (66), on nonhuman primates, healthy human volunteers, and patients with AD. 176,177 However, due to the nonspecific nature of [ 11 C]nicotine (66) and its rapid metabolism, this approach was deemed unfavorable. 178,179To overcome these limitations, researchers developed two 3pyridylether derivatives, namely, 2-[ 18 F]F-A85380 (67, also known as 2-[ 18 F]FA) and 6-[ 18 F]F-A85380 (68) (Figure 16). 180While the derivatives 2-[ 18 F]FA (67) and 6-[ 18 F]F-A85380 (68) have shown unfavorable brain kinetics for quantitative kinetic modeling, 181,182 the high affinity and specificity of 2-[ 18 F]FA (67) (K i = 0.05 nM) toward α4β2-nAChRs has allowed noninvasive visualization of these receptors in the human thalamus. 183In addition to the 18   (69), K i = 0.047 nM) 184 has also been investigated as a tracer for α4β2-nAChR in PET imaging studies with rhesus monkeys. 185he majority of subsequent attempts to develop improved α4β2-nAChR PET ligands have focused on addressing the limitation of slow brain washout, as observed in the case of 2-[ 18 F]FA (67), which precluded conventional kinetic tracer modeling and required up to 5-fold prolonged scanning times.Of note, Zhang et al. hypothesized that the slow brain kinetics of 2-[ 18 F]FA (67) may be attributed to suboptimal lipophilicity and subsequently designed a library of derivatives with a broad range of lipophilicity.Two of these derivatives showed 2.5-fold improved binding potential compared to 2-[ 18 F]FA (67) in rhesus monkeys, and suitable precursors were radiofluorinated to furnish radioligands [ 18 F]NIDA-52158 (70) (K i = 9 pM) and [ 18 F]NIDA-52159 (71, Figure 17) (K i = 4.9 pM), respectively. 186However, further studies are needed to evaluate the brain kinetics of these probes in order to understand their potential for use in PET imaging.
Additional studies led to the discovery of two epibatidine derivatives [ 18 F]FPhEP (72) and [ 18 F]F 2 PhEP (73) as potential tracers for α4β2-nAChRs (Figure 16).These compounds exhibited high to excellent in vitro affinity (K i = 733 and 78 pM, respectively) toward the α4β2 subtype of nAChRs.However, although [ 18 F]FPhEP (72) demonstrated superior brain kinetics compared to 2-[ 18 F]FA (67), both ligands were not suitable for PET imaging applications, primarily due to limited in vivo specificity. 187,188Several other homoepibatidine derivatives have been developed as potential tracers for α4β2 in PET imaging.For example, the two enantiomers of [ 18 F]flubatine ([ 18 F]NCFHEB) (74)  showed high binding affinity toward α4β2-nAChRs (K i = 0.064−0.112pM) but exhibited slow washout from the thalamus in monkey studies. 189,190The epibatidine derivatives, such as [ 18 F]AZAN (75) and [ 18 F]XTRA (76), have demonstrated slightly improved performance characteristics. 191n vivo baboon PET studies with [ 18 F]AZAN (75) have shown high specific binding and a reduced study time to obtain quantitative information on α4β2-nAChR abundance. 192imilarly, [ 18 F]XTRA (76) ([ 18 F](−)-JHU86428) exhibited promising in vitro binding characteristics, including subnanomolar binding affinity and improved lipophilicity over that of 2-[ 18 F]FA (67). 191ukherjee and colleagues have made significant contributions to the field of α4β2-nAChRs PET imaging by developing several radioligands.Initially, they reported on the successful development of 5-(3-[ 1 8 78)), which both possessed faster binding kinetics than previously reported agents and showed regional uptake patterns that corresponded with the known expression patterns of α4β2-nAChRs in the rat and monkey brains. 193,194ubsequently, 3-(2-(S)-azetidinylmethoxy)-5-(3′-[ 18 F]fluoropropyl)pyridine ([ 18 F]nifzetidine (79)) was designed to reduce off-target binding that was observed with [ 18 F]nifrolidine (77) by substituting the pyrrolidine ring in the [ 18 F]nifrolidine (77) structure with an azetidine ring.In vivo rhesus monkey PET imaging experiments using [ 18 F]nifzetidine (79) revealed high levels of tracer uptake in the thalamus and extrathalamic regions, which are known to be α4β2-rich brain regions.Nonetheless, the slow kinetics of [ 18 F]nifzetidine (79) necessitated a prolonged imaging time of greater than 3 h for quantitative PET studies. 195Further optimization led to the development of 3-(2-(S)-3,4dehydropyrrolinylmethoxy)-5-(3′-[ 18 F]fluoropropyl)pyridine ([ 18 F]nifrolene (80)), which has a binding affinity for α4β2-nAChRs similar to that of 2-[ 18 F]FA (67). 196In vivo rhesus monkey PET studies of [ 18 F]nifrolene revealed the highest binding in the thalamus followed by regions of the lateral cingulated and temporal cortex and least binding in the cerebellum.Of note, the thalamus to cerebellum ratio in the monkey brain was >3 at 120 min, which was higher than that for [ 18 F]nifrolidine and [ 18 F]nifzetidine, suggesting promise of [ 18 F]nifrolene as a new PET imaging agent for α4β2 nAChR.A few promising α4β2-targeted PET ligands have advanced to clinical studies and have demonstrated encouraging results in their initial applications in humans.−199 Furthermore, it has been utilized to evaluate the effects of nicotine-induced upregulation of nAChRs in active smokers and during smoking cessation. 200These findings have led to further clinical investigations using 2-[ 18 F]FA (67) to assess changes in cerebral nAChRs in healthy nonsmokers, ex smokers, and heavy and light situational smokers (NCT01038245 and NCT01046513).Similar to 2-[ 18 F]FA (67), first in human studies have indicated that [ 18 F]nifene (78) was a reliable and reproducible radiotracer to visualize α4β2-nAChRs in the human brain.Its favorable brain kinetics enabled kinetic tracer modeling with substantially reduced scan times, which is particularly suitable for use in vulnerable populations. 201Furthermore, the enantiomerically pure α4β2-nAChR PET radioligand (−)-[ 18 F]flubatine (81) exhibited high in vivo stability, fast brain kinetics, and rapid uptake and equilibration between free and receptor-bound tracer in the human brain, making it a valuable tool for imaging α4β2-nAChRs in neuropsychiatric disorders. 202Another probe [ 18 F]AZAN (75) was also evaluated in humans, and a 90 min PET scan with [ 18 F]AZAN (75) was sufficient for performing quantitative analysis of α4β2-nAChR in the living human brain. 203[ 18 F]AZAN (75) has been used to study the relationship between the extent of α4β2-nAChRs occupancy induced by varenicline and the magnitude of dopamine release following nicotine use. 204Lastly, [ 18 F]XTRA (76), a radioligand that exhibits relatively high binding potentials in extrathalamic regions of the baboon brain, has been successfully utilized for quantitative human neuroimaging of the extrathalamic α4β2-nAChR.This finding provides further support for the concept that α4β2-nAChR may play an important role in age-dependent remodeling of the human brain. 205,206verall, a large number of nAChR PET radioligands have been developed, but many of them are still in preclinical investigation or in clinical evaluation for visualizing cerebral nAChRs in neurodegenerative diseases.Recent interest has grown in the potential of nAChRs for diagnosing and treating  other neurological diseases beyond neurodegenerative diseases. 207This renewed interest may stimulate further development and application of nAChR PET radioligands for these purposes.
The properties and molecular imaging results of α4β2 nAChR-selective PET ligands discussed in this section are presented in Table 6.
Apart from vast application in identifying the regional nAChR and mAChR availability under normal conditions and their alteration under disease conditions and/or following drug exposure, PET imaging can also assess endogenous ACh concentrations, which are quantified by regional differences in receptor occupancy.For instance, α4β2-nAChR PET radioligand (−)-[ 18 F]flubatine (81) has demonstrated sensitivity to changes in ACh levels caused by the acetylcholinesterase inhibitor physostigmine. 208Subsequently, Hillmer and coworkers used (−)-[ 18 F]flubatine (81) and M1-preferring radioligand [ 11 C]LSN3172176 (24) to measure baseline variation in ACh concentration across regions in the living human brain, offering a novel approach to estimate potential ACh imbalances in clinical use. 209

PERSPECTIVES AND CONCLUSION
The development of PET tracers that selectively target cholinergic receptors has sparked a vibrant and rapidly advancing research field, holding immense promise for deepening our understanding of these receptors and their associated disorders.Despite a number of cholinergic receptortargeted PET tracers having been disclosed, many of them failed in preclinical investigation.The most common challenge resides in flow-dependent tracer accumulation, which confounds accurate and reliable quantification of cholinergic receptors in tissues of interest.Other challenges include metabolic instability, limited binding specificity, and inability to cross the BBB.Therefore, it is not surprising that future cholinergic receptor-targeted PET tracer development should combine conventional medicinal chemistry with rational pharmacological design in the mechanism of action including orthosteric, allosteric, and bitopic binding to optimize ligand characteristics.Further, although several cholinergic receptortargeted PET tracers have been advanced into clinical evaluation, the vast majority of them focus on imaging of cerebral cholinergic receptors.Given the advent of whole-body PET, a deeper understanding of cholinergic receptors in the periphery might be of interest as a complementary finding to CNS studies.Ultimately, the development and application of cholinergic receptor-targeted PET tracers will have a significant impact on the diagnosis, treatment, and management of cholinergic receptor-related disorders.

Table 1 . 77 a
Properties and Molecular Imaging Results of pan mAChR PET Ligands35,36,41−52,54,56−60,62−71,73−The definition of antagonist or agonist is based on literature reports.PV, preclinical validation; PA, preclinical application; CV, clinical validation; CA, clinical application; Refs, references.The light green shading indicates the investigation stage of translation into human use.

Figure 5 .
Figure 5. Modes of binding to mAChRs by different classes of ligands.
[ 11  C]32 displayed moderate affinity for the human striatum (K d = 17.6 nM) but did not show specific binding in the brains of rhesus monkeys.Meanwhile, [ 11 C]33, [ 11 C]34, and [ 11 C]35 showed inadequate

Figure 9 .
Figure 9. Representative radiolabeling methods for the preparation of M2-selective PET ligands.

Figure 12 .
Figure 12.Representative PET images of [ 11 C]MK-6884 (36) in a healthy adult volunteer taken under baseline conditions and AD patients receiving treatment with donepezil, an acetylcholinesterase inhibitor.Reprinted with permission from 113.Copyright 2022 American Association for the Advancement of Science.

Table 4 . 116 a
Properties and Molecular Imaging Results of M4-Selective PET Ligands 109−The definition of antagonist or agonist is based on literature reports.PV, preclinical validation; PA, preclinical application; CV, clinical validation; CA, clinical application.The light green shading indicates the investigation stage of translation into human use.

Figure 14 .
Figure 14.Representative radiolabeling methods for the preparation of α7 nAChR-selective PET ligands.
[ 11 C]CHIBA-1001 (55) was readily synthesized with excellent radiochemical yield (9.5% nondecay-corrected RCY) and high molar activity (343.7 GBq/μmol) via the palladiumcatalyzed coupling of the corresponding tributyl−stannane precursor with [ 11 C]MeI.PET imaging studies of [ 11 C]-CHIBA-1001 (55) in the conscious monkey brain revealed a heterogeneous pattern of radioactivity distribution, which correlated with the density of α7 nAChRs in various brain regions.Similar to [ 76 Br]SSR180711 (53), [ 11 C]CHIBA-1001 (55) demonstrated the highest uptake in the temporal cortex, reaching a peak of ∼0.048%ID/mL at around 20 min post tracer injection.The specific binding of [ 11 C]CHIBA-1001 (55) was supported by the significant blockade in uptake upon administration of the selective α7 nAChR agonists SR180711 and A844606.Furthermore, subchronic administration of the N-methyl-D-aspartate (NMDA) receptor antagonist phencyclidine (PCP), which could induce a schizophrenia NHP model, resulted in a decrease of [ 11 C]CHIBA-1001 (55) binding in various brain regions with a statistically significant difference observed in the frontal cortex.These promising results prompted the translation of [ 11 C]CHIBA-1001 (55) into clinical studies for the visualization of α7 nAChR in the human brain. 145PET imaging studies of [ 11 C]CHIBA-1001 (55) in humans revealed a heterogeneous brain distribution with V T values ranging from 16.6 to 21.6 in various regions.The highest radioactive signal was observed in the thalamus, reaching a peak of 5.5 SUV at approximately 13 min p.i. Notably, in human subjects, [ 11 C]CHIBA-1001 (

Figure 15 .
Figure 15.Representative V T parametric images from the test and retest scans in a healthy volunteer using [ 18 F]ASEM (59).Reprinted with permission from ref 160.Copyright 2017 Springer.

Figure 17 .
Figure 17.Representative radiofluorination method for the preparation of α4β2 nAChR-selective PET ligands.

■ 18 F
AUTHOR INFORMATION of Sciences (CAS) in 2017.After that, he continued his postdoctoral research with Steven H. Liang at the Division of Nuclear Medicine and Molecular Imaging, Department of Radiology, Massachusetts General Hospital (MGH) and Harvard Medical School (HMS).He is currently a research scientist in the Department of Radiology and Imaging Sciences at Emory University.His research interests include -labeling methodologies and development of PET tracers for the central nervous system.Jiahui Chen obtained his B.S. degree in Pharmacy in 2012 from Guangzhou University of Chinese Medicine (China).He earned his Ph.D. degree in Chemical Biology from Peking University with Zhengying Pan in 2018.In the same year, he started as a postdoctoral fellow with Steven H. Liang at the Division of Nuclear Medicine and Molecular Imaging at Harvard Medical School/Massachusetts General Hospital (HMS/MGH).Recently, he joined Steven H. Liang's group in the Department of Radiology and Imaging Sciences in the School of Medicine in Emory University.His research interests include the construction and evaluation of neuroinflammatory and degenerative disease models and the development and clinical translation of new PET tracers for the central nervous system.Yin-Long Li received his B.Sc. degree in Pharmacy at Shandong University of Traditional Chinese Medicine (2012).He completed his Ph.D. degree (2012−2018) under the supervision of Jun Deng at Tianjin University.Then, he worked as an assistant specialist at the Chinese Academy of Inspection and Quarantine (2018−2021).He joined the group of Steven H. Liang at Massachusetts General Hospital & Harvard Medical School (2021−2022).He is currently working as a postdoctoral researcher at Emory University.His research interests are PET radiopharmaceutical development, fluorine chemistry, and asymmetric catalysis.Chongzhao Ran earned his B.S. degree in Chemistry from South Central University for Nationalities, China, his M.S. degree in Medicinal Chemistry from China Pharmaceutical University, and his Ph.D. degree in Medicinal Chemistry at the Shanghai Institute of Pharmaceutical Industry.He did his postdoctoral training at the University of Chicago and Harvard Medical School.Since 2010, he has held a faculty position at Massachusetts General Hospital (MGH) and Harvard Medical School (HMS).Currently, he is an associate professor of radiology.His research focuses on developing molecular imaging probes and imaging technologies.His research group has successfully developed numerous probes of near-infrared fluorescence, chemiluminescence, bioluminescence, and PET imaging for in vivo detection of amyloid beta and other misfolded proteins in mouse models of AD and other neurodegenerative diseases.Thomas L. Collier received his Ph.D. degree (1989) from Carleton University (Canada) under the guidance of John ApSimon.He started his first industrial postdoctoral position in the environmental analysis company Paracel Laboratories (Canada) and the second one in radiochemistry for use in medical imaging at the University of Tennessee Medical Center at Knoxville (USA) as part of a joint program with CTI Inc. (USA).For the past 20 years, his research has included the use of microfluidics for radiochemistry, and this work continued when he moved to Advion Inc. (USA) about 14 years ago.He currently holds a senior scientist and manager position at Advion-Interchim Scientific Inc. as well as visiting scientist positions at Harvard Medical School, Massachusetts General Hospital, and Emory University.Zhen Chen received his B.S. degree in Applied Chemistry in 2013 from Tianjin University (China).He completed his Ph.D. degree in 2019 under the supervision of Jun-An Ma in Tianjin University (China).From 2017 to 2020, he studied radiochemistry as an exchange Ph.D. student and postdoctoral fellow with Steven H. Liang at Harvard Medical School/Massachusetts General Hospital (HMS/ MGH, USA).From 2020 to 2021, he worked as a postdoctoral research fellow in the group of Nicolai Cramer at EPFL (Switzerland).Currently, he is a full professor at Nanjing Forestry University.His research interests focus on organofluorine chemistry, photochemistry, and the development of PET biomarkers.Steven H. Liang earned his B.S. degree from Tianjin University and Ph.D. degree in Chemistry at the University of British Columbia.He then began his professional career as a NSERC fellow under the mentorship of E. J. Corey at Harvard University.In 2012, he started his junior faculty position at Massachusetts General Hospital (MGH) and Harvard Medical School (HMS) and was then promoted to Director of Radiochemistry (2017) and Associate Professor of Radiology (2019).Recently, he assumed the role of the inaugural head at the newly established Translational PET Center at Emory University.His research interests encompass the development of novel radiochemistry, PET biomarkers, and companion radiotherapy, as well as the translation of these technologies into clinical settings.■ ACKNOWLEDGMENTS We thank the Division of Nuclear Medicine and Molecular Imaging, Radiology, Massachusetts General Hospital and Harvard Medical School and the Department of Radiology and Imaging Sciences, Emory University School of Medicine for general support.A.H. was supported by the Swiss National Science Foundation (SNSF).S.H.L. gratefully acknowledges the support provided, in part, by the Emory Radiology Chair Fund and Emory School of Medicine Endowed Directorship.

Table 2 .
Properties and Molecular Imaging Results of M1-Selective PET Ligands 81,84 , 86−90 ,92−94The definition of antagonist or agonist is based on literature reports.PV, preclinical validation; PA, preclinical application; CV, clinical validation; CA, clinical application.The light green shading indicates the investigation stage of translation into human use.
109,110467485 (29, also known as [11C]-AZ13713945) and its two analogues [ 11 C]30 and [ 11 C]31 (Figure10) were reported as highly selective M4 positive allosteric modulators (PAMs) with EC 50 values of 78.8, 41.4, and 43.4 nM, respectively.109,110InvitroARGresults in the rat brain sections showed that [11C]31 had the best binding specificity, followed by the medium level of [ 11 C]30, while [ 11 C]VU0467485 (29) failed to demonstrate any specific binding.However, [ 11 C]31 was unable to penetrate the BBB in further rat PET imaging.Despite this setback, this work provides a foundation for further chemical optimization in the development of M4-selective PET ligands. Reently, five 11 Clabeled allosteric M4 radioligands (32−36, as depicted in Figure

Table 3 .
Properties and Molecular Imaging Results of M2-Selective PET Ligands 84,98−105 a The definition of antagonist or agonist is based on literature reports.PV, preclinical validation; PA, preclinical application; CV, clinical validation; CA, clinical application.The light green shading indicates the investigation stage of translation into human use.

Table 5 .
Properties and Molecular Imaging Results of α7 nAChR-Selective PET Ligands131,134−136,138−140,143−166The definition of antagonist or agonist is based on literature reports.PV, preclinical validation; PA, preclinical application; CV, clinical validation; CA, clinical application.The light green shading indicates the investigation stage of translation into human use. a