In Vivo and In Vitro Characteristics of Radiolabeled Vesamicol Analogs as the Vesicular Acetylcholine Transporter Imaging Agents

The vesicular acetylcholine transporter (VAChT), a presynaptic cholinergic neuron marker, is a potential internal molecular target for the development of an imaging agent for early diagnosis of neurodegenerative disorders with cognitive decline such as Alzheimer's disease (AD). Since vesamicol has been reported to bind to VAChT with high affinity, many vesamicol analogs have been studied as VAChT imaging agents for the diagnosis of cholinergic neurodeficit disorder. However, because many vesamicol analogs, as well as vesamicol, bound to sigma receptors (σ1 and σ2) besides VAChT, almost all the vesamicol analogs have been shown to be unsuitable for clinical trials. In this report, the relationships between the chemical structure and the biological characteristics of these developed vesamicol analogs were investigated, especially the in vitro binding profile and the in vivo regional brain accumulation.


Introduction
Many clinical trials for early diagnosis of Alzheimer's disease by amyloid PET imaging have been reported [1][2][3][4][5][6][7][8][9][10]. Many researchers have reported that amyloid imaging is useful for early diagnosis of AD, but there have been many reports showing no significant association between the brain accumulation of amyloid imaging agents and the severity of dementia in AD [11][12][13][14][15]. Recently, there were many reports that tau imaging was useful for the diagnosis of the severity of dementia and early diagnosis of AD [16][17][18][19][20]. Evaluation of the diagnostic efficacy of tau imaging regarding AD will continue for several years. e onset of AD, which is a progressive neurological disease characterized by reduction in cognitive function and memory, is thought to be caused by a hypothesized amyloid cascade ( Figure 1) [21]. Namely, (1) amyloid β (Aβ40 and Aβ42), which is produced by an abnormal cleavage of the amyloid precursor protein (APP) by β-and c-secretase, is aggregated and accumulated extracellularly in cranial nerve cells. (2) Neurofibrillary tangles (NFTs) are formed by the accumulation of a tau protein phosphorylated excessively in the cytoplasm. (3) Nerve degeneration, neurologic function deficiency, and metabolism deficiency occur in the neuronal cell. (4) Neuronal cell death occurs, which causes the onset of AD. Abnormal accumulation of amyloid β based on the amyloid cascade supports the usefulness of amyloid imaging for early diagnosis of AD. On the contrary, studies of amyloid β immunotherapy showed that reduction of amyloid β plaques in patients with Alzheimer's disease did not prevent progressive neurodegeneration [22]. e amyloid β plaque is an antecedent marker of Alzheimer's disease [22], and amyloid imaging will not be useful to evaluate the therapeutic efficacy of AD treatment. Neuronal degeneration, neurologic function deficiency, and metabolic deficiency in the third step of the amyloid cascade are thought to be important internal molecular targets for the development of an imaging agent for the early diagnosis of neurodegenerative disorders with cognitive decline such as AD. Acetylcholine esterase inhibitors such as donepezil are commonly used for treatment of cognitive dysfunction in AD [23,24]. e dysfunction of cholinergic neurons is associated with AD symptoms such as cognitive dysfunction, memory impairment, and learning disorders [25]. Presynaptic cholinergic function, such as loss of choline acetyl transferase (ChAT), the enzyme for synthesis of acetylcholine (ACh) from choline and acetyl-coenzyme A, and the vesicular acetylcholine transporter (VAChT), the transporter for the accumulation of acetylcholine (ACh) inside the synaptic vesicles, is changed in AD [26,27].
us, the internal molecules in the cholinergic nerve system will be suitable as the cranial molecular target of an imaging agent for early diagnosis of AD.
ere are five main molecular targets: choline acetyl transferase (ChAT), vesicular acetylcholine transporter (VAChT), choline transporter (ChT), acetylcholine esterase (AChE), and postsynaptic receptors in the cholinergic synaptic terminal (Figure 2). A small molecule compound binding to ChAT with high affinity has not yet been found, and only hemicholinium-3 (HC-3) with a positive electric charge as a small molecule compound binding to ChT, the transporter for reuptake of choline (Ch) released by ACh hydrolysis in the synaptic cleft, has been found, which makes the development of ChAT and ChT imaging agents difficult. A reduction in AChE activity in AD patients was shown by PET imaging using [ 11 C]MP4A and [ 11 C]PMP [28][29][30][31]. However, these AChE imaging agents, a selective substrate for AChE, show low stability in the blood, and quantitative measurement is thought to be difficult. Changes in presynaptic cholinergic functions, such as ChAT and VAChT activity in AD, are thought to be more significant than changes in postsynaptic cholinergic functions, such as the cholinergic muscarinic receptors (mAChR) [26,27,32]. erefore, VAChT is an excellent in vivo target substrate for the early diagnosis of AD. Many vesamicol analogs have been developed as potential VAChT imaging agents for PET or SPECT, since vesamicol (2-(4-phenylpiperidino) cyclohexanol) was reported to bind to VAChT [33,34]. However, many of the reported vesamicol analogs were shown to be insufficient for use as VAChT imaging agents due to binding to sigma receptors (σ 1 and σ 2 ) or low accumulation in brain in vivo. Many vesamicol analogs were developed with the aim of improving VAChT affinity and decreasing the affinity for sigma receptors (σ 1 and σ 2 ).
In this report, the biological characteristics of these developed vesamicol analogs were investigated, especially the in vitro binding profile and the in vivo regional brain accumulation. PET ligands for VAChT had been reviewed by Giboureau previously [35]. We tried to compare in vitro and in vivo characteristics of many VAChT ligands including these PET ligands under the same conditions as much as possible.

VAChT Imaging Agent Based on
Vesamicol Analogs e molecular structures of VAChT imaging agents based on vesamicol analogs are shown in Figure 3. (1). Rogers et al. investigated the binding affinity of 84 vesamicol analogs to VAChT in an in vitro binding assay [36] and reported benzovesamicol as one of the vesamicol analogs binding to VAChT with a high affinity. Benzovesamicol (BV) is the vesamicol analog with a benzene ring in ring A of the vesamicol skeleton ( Figure 3). e affinity (IC 50 � 50 nM) of the racemate of BV to VAChT is similar to that (IC 50 � 40 nM) of the racemate of vesamicol in the in vitro binding assay [36].    (1)- (7). Trozamicol analogs: (8) and (9). Vesamicol analogs with a radionuclide into the C ring: (10)- (12). Vesamicol analogs with alicyclic groups into the A ring: (13)- (17). Vesamicol analogs with a carbonyl group between the B ring and C ring: (18)- (22).

Summary regarding Benzovesamicol Analogs.
Although benzovesamicol (BV) has higher affinity and selectivity for VAChT than vesamicol, several benzovesamicol (BV) analogs were synthesized to improve VAChT affinity and to decrease the affinity for sigma receptors (σ 1 and σ 2 ). A BV analog with phenylpiperazine instead of phenylpiperidine in BV (5) [51,52], a BV analog with benzylpiperidine instead of phenylpyridine (6), and a BV analog with pyridinyl piperidine instead of phenyl piperidine (7) showed a lower VAChT affinity than BV [53]. e affinity of 7 to the σ1 receptor increased.

Summary regarding Trozamicol Analogs.
e binding affinity of each trozamicol analog was investigated by a different method of the in vitro binding assay. Furthermore, different tissue membranes were used in the binding assay for VAChT or sigma receptors (σ 1 and σ 2 ). It is difficult to compare the value of the VAChT affinity of each trozamicol analog with the value of sigma receptors (σ 1 and σ 2 ) affinity. us, to compare all vesamicol analogs, the value of the affinity of vesamicol to VAChT and sigma receptors (σ 1 and σ 2 ) was used as the standard value in each binding assay in this report. Namely, the comparison between these vesamicol analogs was expressed as the ratio of the affinity of vesamicol analogs to the affinity of vesamicol to VAChT, σ 1 , and σ 2 , which were obtained by the same in vitro binding assay method (Table 1). e affinity and selectivity of MIBT to VAChT was superior to that of FBT [57]. (+)-FBT (9) showed a higher affinity and selectivity for VAChT than (−)-FBT [60]. e accumulation of (+)-[ 18 F]FBT (9) in the rat brain was higher than that of (+)-[ 125 I]MIBT (9) [56,72]. However, the accumulation of (+)-[ 18 F]FBT (9) in the rat brain is lower than the expected brain accumulation based on the chemical structure of (+)-[ 18 F]FBT (9) [72]. Considering the radiation dose and spatial resolution, the  (4)  - e affinity of these vesamicol analogs for VAChT, σ1, and σ2 had not been reported. * * e affinity of vesamicol, as a reference, for VAChT, σ1, and σ2 was not investigated by the same in vitro binding assay method.

Cholinergic Nerve Systems.
e cholinergic system includes the following three main cholinergic pathways [85][86][87][88][89][90]: (1) numerous cholinergic neurons in the basal forebrain, which supply cholinergic projections throughout the cerebral cortex, the forebrain limbic structures, the diagonal band nucleus, the amygdala, and the hippocampus, (2) the mesopontine regions including the laterodorsal tegmental nucleus and the pedunculopontine nucleus, which project to the forebrain, the thalamus, the hypothalamus, the cerebellar nucleus, and the brainstem, and (3) populations of cholinergic interneurons in the striatum. Cholinergic neurons in the basal forebrain and the mesopontine regions are closely associated with cognition, learning, and memory functions. Cholinergic neurons in the striatum do not project beyond the borders of the striatum. It is important to investigate the change of the cholinergic nerve system in the cerebral cortex, the forebrain limbic structures, the diagonal nucleus, the amygdala, the hippocampus, the thalamus, the hypothalamus, and the cerebellar nucleus in comparison with the striatum in order to diagnose AD early by a VAChT imaging agent.

Distribution of VAChT in Cholinergic Neurons.
e nerve terminal consists of presynaptic and postsynaptic neurons. e regional distribution of VAChT situated at presynaptic ) are found in high density in the cerebral cortex, striatum, diagonal band, hippocampus, amygdala, anterior and intralaminar nuclei of the thalamus, granule and purkinje cell layers of the cerebellum, and motor nuclei of the cranial nerves [91][92][93]. e nicotinic receptor is widely distributed in the anteroventral nucleus of the thalamus [94]. erefore, VAChT-rich presynaptic cholinergic nerve terminals were thought to be widely distributed in various brain regions, including the cerebral cortex, striatum, diagonal band, hippocampus, thalamus, amygdaloid nucleus, cerebellum, and nuclei of cranial nerves. e mAChR concentrations in the striatum are approximately 1.67-fold higher than those in the cerebral cortex. Table 2 shows the accumulation of vesamicol analogs  in the striatum and cortex and the ratio of the striatum to cortex (ST/CTX) for radiolabeled vesamicol analogs. CNS radioligands need to accumulate in the brain through the BBB. e CNS radioligands are required to have 2-3 as the value of the log of the octanol-water partition coefficient (Log P) and be less than 500 as molecular weight (MW) as their chemical characteristics to penetrate the BBB [95,96] (11) displayed high brain uptake in rat. However, other vesamicol analogs displayed low brain uptake in rats or mice against the physicochemical properties. e low brain uptake of vesamicol   (10) and (−)-OMV (11) may be caused by excretion of vesamicol analogs from brain by a P-glycoprotein related to the drug excretion mechanism or the high binding affinity for serum protein. e accumulation of (−)-[ 11 C]TZ659 (20) in the brain increased 2.2-fold by pretreatment with cyclosporine A (CycA), which inhibits a P-glycoprotein related to the drug excretion mechanism, 30 min prior to injection of a radiotracer. e vesamicol analogs except for (−)-oIV (10), (−)-OMV (11), and (−)-OIDV (13) showed a ratio of the striatum to cortex (ST/CTX) more than 1.1. As mentioned above, the VAChT-rich region was widely distributed in the various regions of the brain. ese VAChT imaging agents will distribute in the cerebral cortex, the hippocampus, the thalamus, the hypothalamus, and the cerebellum besides the striatum. However, several vesamicol analogs showed the high concentration in the striatum and the low concentration in the cerebral cortex and the cerebellum.

e Structure-Activity Relationship of Radioligands
for VAChT Imaging. In in vitro characterization, such as the affinity and selectivity of radioligands for VAChT, (−)-enantiomers of vesamicol analogs based on benzovesamicol ((−)-FEOBV (2) and (−)-FBBV (18)) are superior to other vesamicol analogs (Table 1). However, benzovesamicol analogs showed low brain uptake in the rat and mouse. On the contrary, in in vivo characterization, such as brain uptake of radioligands, vesamicol analogs that incorporated a radionuclide into the C ring of vesamicol ((−)-oIV (10) and (−)-OMV (11)) were superior to other vesamicol analogs (Table 2). However, (−)-oIV (10) and (−)-OMV (11) showed low selectivity for VAChT in vitro and in vivo. Considering abovementioned results, the in vivo characterization of radioligands for VAChT will improve by minimizing the molecular weight of the ligand, and in vitro characterization of radioligands for VAChT will improve by incorporating a carbonyl group between the B ring and C ring of vesamicol analogs. We are interested in vesamicol analogs incorporating three elements such as a carbonyl group between the B ring and C ring, a 4-to 7-membered alicyclic ring as A ring, and a radionuclide in the C ring, together.

Conclusion
Many vesamicol analogs were investigated as an VAChT imaging agent. In this report, 5 types of vesamicol analogs were investigated: (1) vesamicol analogs based on benzovesamicol, (2) vesamicol analogs based on trozamicol, (3) vesamicol analogs that incorporated a radionuclide into the C ring of vesamicol, (4) vesamicol analogs that incorporated alicyclic groups into the A ring of vesamicol, and (5) vesamicol analogs that incorporated a carbonyl group Table 4: Accumulation of enantiomers of vesamicol analogs (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18) in the striatum and cortex and the ratio of the striatum to cortex (ST/CTX) for enantiomers of radiolabeled vesamicol analogs. 10 Contrast Media & Molecular Imaging between the B ring and C ring of benzovesamicol. All vesamicol analogs  are insufficient as an VAChT imaging agent for early diagnosis of Alzheimer's disease. is is because the vesamicol analogs with a high affinity and a high selectivity for VAChT showed low brain uptake, and vesamicol analogs with a high brain uptake showed high affinity for sigma receptors and low selectivity for VAChT. Considering the relationship between the cholinergic nerve system and AD, the development of a VAChT imaging agent is important. It is necessary that the suitable radioligand for VAChT imaging shows a high affinity and high selectivity for VAChT in vitro and in vivo and shows the high accumulation of the regional brain in accordance with the concentration distribution of VAChT in the brain. Furthermore, the ideal radioligand for VAChT imaging will require the fast blood clearance and the resistance to cleavage of the radioligand as described by Giboureau et al. previously. In the future, the suitable cholinergic neuronal degeneration imaging is thought to be found by comparing these three imaging: ChT imaging, VAChT imaging, and ChT imaging, and putting them together. It is necessary to further the development of a radioactive imaging agent for choline transporter (ChT) and choline acetyl transferase (ChAT). Finally, the ideal AD imaging is thought to be obtained by putting amyloid imaging, tau imaging, and cholinergic neuronal imaging together.

Conflicts of Interest
e authors declare that there are no conflicts of interest.