Brain Imaging of Nicotinic Receptors in Alzheimer's Disease

Neuronal nicotinic acetylcholine receptors (nAChRs) are a family of ligand-gated ion channels which are widely distributed in the human brain. Several lines of evidence suggest that two major subtypes (α4β2 and α7) of nAChRs play an important role in the pathophysiology of Alzheimer's disease (AD). Postmortem studies demonstrated alterations in the density of these subtypes of nAChRs in the brain of patients with AD. Currently, nAChRs are one of the most attractive therapeutic targets for AD. Therefore, several researchers have made an effort to develop novel radioligands that can be used to study quantitatively the distribution of these two subtypes in the human brain with positron emission tomography (PET) and single-photon emission computed tomography (SPECT). In this paper, we discuss the current topics on in vivo imaging of two subtypes of nAChRs in the brain of patients with AD.


Introduction
Alzheimer's disease (AD) is the most common neurodegenerative disorder in the elderly and has become a major worldwide health problem. Several reports indicated that it is affecting almost 1 in 10 individuals over the age of 65 [1], and as life expectancy increases, over 37 million people suffer with AD, and it is projected to quadruple by 2050 [2]. AD accounts for over 50% of senile dementia and the majority of presenile dementia cases and is characterized by progressive deterioration of higher cognitive functions including the loss of memory [3,4].
van Duijn and Hofman [5] reported the inverse relationship between smoking history and early onset AD, suggesting that smoking may protect against AD [6]. Furthermore, Rusted and Trawley [7] reported acute improvements in prospective memory following nicotine administration. Although Swan and Lessov-Schlaggar [8] discuss the effects of tobacco smoke and nicotine on cognition in their review, smoking is associated with increased risk for negative preclinical and cognitive outcomes in younger people as well as in older adults. More recently, a meta-analysis including longitudinal studies published between 1995 and 2007 reported that current smokers relative to never-smokers were at increased risk of AD, vascular dementia, any dementia, and cognitive decline, in over the age of 65 [9]. Several lines of evidence demonstrated that smoking almost doubled the risk of AD and that smoking cessation might contribute to a reduction of risk factors for AD and cardiovascular disease [10,11]. Noteworthy, the later is also known as a risk factor for AD. These results suggest that smoking cessation may play an important role in not only primary but also secondary prevention of AD. In contrast, although the discussion about neuroprotection by smoking has been continued, it is possible that nicotinic acetylcholine receptors (nAChRs) in the brain might play a role in the pathophysiology of AD.
The nAChRs are one of the main classes of AChRs, which have a pentameric structure composed of five membrane spanning subunits, of which nine different types have thus far been identified and cloned. To date, twelve neuronal nAChR subunits have been described [12]; nine (α2-α10) code for subunits [12] based on the presence of adjacent cysteine residues in the predicted protein sequences, in a region homologous to the putative agonist-binding site of the muscle, a subunit (α1) and three referred to as non-α or βsubunits (β2-β4). Among the several nAChR subtypes in the human central nervous system (CNS), the heteromeric α4β2 and homomeric α7 subtypes ( Figure 1) are predominant in the brain [13,14]. It has been reported that other subtypes (e.g., α3, α6) exist in the brain [15,16] and that α6 subtype might be mainly involved in the pathophysiology of Parkinson's disease [16]. Furthermore, studies using postmortem human brain samples have demonstrated alterations in the levels of α4 and α7 nAChR in the brains of patients with AD [15,[17][18][19]. Despite its lower number, loss of α3 subtype consistent with α4 and α7 nAChR subtypes was also observed in the brains of patients with AD [15]. Taken together, it is likely that these two subtypes (α4β2 and α7) of nAChR might play a role in the pathogenesis of AD. Therefore, it is of great interest to examine whether these two subtypes of nAChR are altered in the living brain of patients with AD using brain imaging techniques.
In this paper, we discuss the recent findings on imaging of these two nAChRs (α4β2 and α7) in the brain with AD using positron emission tomography (PET) and singlephoton emission computed tomography (SPECT).

α4β2 nAChRs Subtype
2.1. Relationship between Amyloid-β and α4β2 nAChR. Amyloid β protein (Aβ) is a major constituent of senile plaques and one of the candidates for the cause of the neurodegeneration found in AD. It has been shown that the accumulation of Aβ precedes other pathological changes and causes neurodegeneration or neuronal death in vitro and in vivo [20,21]. The loss of memory seen in AD is thought to be associated with Aβ-induced impairment of synaptic plasticity such as long-term potentiation (LTP) in the hippocampus. There are lines of evidence suggesting that nAChR activation provides protection against Aβ-induced neurotoxicity in cultured cortical neurons [22,23]. These results indicated that nicotine protects against Aβ-induced neuronal death, and similar effect has been also observed in those selective α4β2 nAChR agonists such as cytosine and epibatidine, but this neuroprotection is blocked by the selective α4β2 nAChR antagonist dihydro-β-erythroidine (DHβE). Moreover, recently, Wu et al. [24]  These findings suggest that α4β2 nAChR, which can be directly activated by Aβ, is required for Aβ suppression of LTP in vivo. The mechanisms by which nicotine enhanced the inhibition of LTP by Aβ were not clear. A possible explanation is that nicotine could activate nAChRs present in inhibitory interneurons, thereby potentiating inhibitory inputs to hippocampal neurons.

Cognition and α4β2 nAChR
Agonists. It is likely that reduced density of nAChR is related to dementia severity, assessed using a global rating. Nicotine has been postulated to be a possible treatment for AD, improving cognition in humans [25]. Recently, Loughead et al. [26] reported novel evidence that the α4β2 partial agonist varenicline increased working memory-related brain activity after 3 days of nicotine abstinence, particularly at high levels of task difficulty, with associated improvements in cognitive performance among highly dependent smokers.

Postmortem Studies of α4β2 nAChR in the Brain of
Patients with AD. Not only transmitter release but also receptor-binding sites may be altered in the brain of AD patients [27][28][29]. Postmortem studies showed the reduction (up to 50%) of α4β2 subtype of nAChRs in brain of patients with AD [30], and it may occur very early in the course of AD [31]. Both α4 and α7 subunits are known to be important constituents in α4β2 and α7 receptor subtypes, respectively. Investigation using the autopsy samples of human cerebral cortex has clearly shown that these two subtypes (α4 and α7 isoforms) are significantly decreased in their protein amount in the cortices of AD patients [15,19,32].

Imaging of α4β2 nAChR Subtype.
Considering the role of α4β2 nAChR in the pathophysiology of AD, it is of great interest to study α4β2 nAChR in the living human brain using PET/SPECT. Much effort has been devoted to visualize α4β2 nAChR in the brain by PET/SPECT. Currently, two PET ligands, including

[ 11 C]
Nicotine. The development of radiolabelled nicotine [36,37] has allowed for evaluating the uptake and distribution of nAChR in the living human brain [38][39][40]. The data obtained by [ 11 C]nicotine is generally consistent with the known pattern of nAChR measured by in vitro binding in autopsy brain tissue [39]. [ 11 C]nicotine-PET has been used to study α4β2 nAChR in human brain, and a severe loss of the nAChR has been detected in the brain of patients with AD [13]. Cortical nAChRs in mild AD patients are robustly associated with the cognitive function of attention [35] and have revealed a significant negative correlation between severity of cognitive impairment and density of brain nAChR [40]. It will be, therefore, of interest to study an alteration in α4β2 nAChR at a presymptomatic stage of AD. Furthermore, the in vivo cortical AChE inhibition and  [ 11 C]nicotine binding were associated with changes in the attention domain of cognition rather than episodic memory when administering galantamine [41]. Thus, [ 11 C]nicotine-PET may be also used for monitoring treatment efficacy in AD patients [41,42].
Unfortunately, [ 11 C]nicotine displays high levels of nonspecific binding, rapid metabolism, and rapid washout of the brain [43]. The heterogeneity of [ 11 C]nicotine binding in the brain also precludes the identification of a reference region which may be used to accurately determine nonspecific binding. Taken together, it is unlikely that [ 11 C]nicotine might be a suitable PET ligand for in vivo imaging of α4β2 nAChR in human brain.

2-[ 18 F]F-A-85380 and 5-[ 123 I]I-A-85380. A-85380 [3-
(2(S)-azetidinylmethoxy) pyridine] is a potent and selective agonist with high affinity for α4β2 nAChR subtype and low affinity for other nAChR subtypes [44]. A-85380 is effective in a wide range of preclinical models of CNS disorders [45,46]. Recently, A-85380 was successfully labeled using 18 F or 125/123 I with a high affinity (K i = 50 pM for F and K i = 15 pM for I) for α4β2 nAChR [44,47,48]. These radioligands have been evaluated in vitro and in vivo as PET/SPECT radioligands to visualize α4β2 nAChR subtype in the brain [49,50]. In healthy nonsmoking human brain, both 2-[ 18 F]F-A85380 and 5-[ 123 I]I-A85380 have revealed a pattern of highest uptake in the thalamus, intermediate in the midbrain, pons, cerebellum, and cortex, and lowest in white matter [50][51][52], which is consistent with the regional distribution of α4β2 nAChR.
Furthermore, a study of age-related decline in nicotinic receptor availability showed that regional β2 nAChR availability were inversely correlated with decline ranging from 32% (thalamus) to 18% (occipital cortex) over the adult lifespan, or up to 5% per decade [53]. These results may corroborate postmortem reports of decline in high-affinity nicotine binding with age and may aid in elucidating the role of β2-nAChR in cognitive aging. In addition, 2-[ 18 [51,[56][57][58][59][60]. In 17 patients with moderate to severe AD and 6 subjects with amnestic mild cognitive impairment (MCI) compared with 10 healthy control subjects, Sabri et al. [56] found significant reductions of α4β2 nAChR in brain regions (hippocampus, caudate, frontal cortex, temporal cortex, posterior cingulate, anterior cingulate, and parietal cortex) in the brain of AD by using 2-[ 18 F]F-A-85380. Most recently, Kendziorra et al. [57] reported that both patients with AD and those with MCI showed a significant reduction in 2-[ 18 F]F-A-85380 binding potential in typical AD-affected brain regions and that the 2-  (Figure 3). These findings suggest that a reduction in α4β2 nAChR occurs during symptomatic stages of AD and that the α4β2 nAChR availability in these regions correlated with the severity of cognitive impairment. In contrast, there were no differences in distribution volume (DV) of nAChR between the healthy controls and early AD patients ( Figure 4) [51,58].
2-[ 18 F]F-A-85380 PET has been used to observe outcome of drug treatment for the improvements of cognition in patients with mild AD [61]. However, no significant correlations were found between cognitive measures and nAChR simplified DV ( Figure 5). These results are similar to the results reported by Kadir et al. [41] in their studies using [ 11 C]nicotine. The relationship between cognition in AD and cholinergic dysfunction may be related to a number of factors, including the degree of cholinergic system (or receptor) loss, the other nAChR subtypes, or other neurochemical systems.

Relationship between Aβ and α7 nAChR.
Of the two major subtypes of nAChRs in the CNS, α7 subtype has lower affinity for ACh compared to α4β2 subtype [62]. Accumulating evidence suggests that α7 nAChR plays a role in the pathophysiology of AD. Aβ has picomolar affinity for α7 nAChR [63,64], which results in the formation of Aβ-α7 nAChR complex. This complex is known to move intracellularly and cause neurotoxicity [63][64][65]. Interestingly, this neurotoxicity is not present in transgenic mouse model of AD overexpressing a mutated form of the human amyloid precursor protein (APP) and lacking the α7 nAChR [66]. Recently, Bencherif and Lippiello [67] pointed out that the α7-JAK2-(NF-κB; STAT3)-Bcl2 prosurvival pathway is important for the neuroprotective role of α7 nAChR ( Figure 6). By blocking cytosolic cytochrome C, which is released from the mitochondria via Aβ 1−−42 , Bcl2 fully counteracts the Aβ 1−−42 -induced apoptosis of cells [68]. The fact that this antiapoptotic pathway is further related with ApoE4 [69], GSK-3β-activated tau phosphorylation [70], and Wnt signaling pathways [71] denotes the critical role of α7 nAChR in pathophysiology of AD.
The 3xTg-AD mice [72], which are triple transgenic mice expressing APP, presinilin-1, and Tau, were shown to have an age-dependent reduction of α7 nAChR. This reduction was limited to brain regions where intraneuronal Aβ 42 accumulation occurred [73]. The early cognitive deficits of 3xTG-AD mice also correlate with intracellular Aβ accumulation, and the clearing of this Aβ accumulation by immunotherapy reverses the early cognitive impairment [74].
Tg2576 transgenic mice (APPswe) dramatically reduced Aβ plaque expression with chronic administration of nicotine for 5.5 months [75]. It is further reported that a 10-day administration of nicotine reduced the guanidinium-soluble Aβ levels by 46 to 66%, whereas the intracellular Aβ levels remained unchanged [76]. This treatment with nicotine also resulted in less glial fibrillary acidic protein-(GFAP-) immunoreactive astrocytes around the amyloid plaques and increased numbers of α7 nAChR in the cortex of APPswe mice [76]. Bencherif [68] points out the importance of these , and Alzheimer's disease (AD) groups. No significant regional differences among the subject groups for any of the 8 regions, including the 4 neocortical regions, were identified. (Data is from the paper of Mitsis et al. [58].) data, as reduction of Aβ with anti-Aβ antibody treatment is reported to rapidly recover the associated neuritic dystrophy in living animals [77].
Orr-Urtreger et al. [78] generated α7 nAChR gene knockout (KO) mice, and the resulting α7 nAChR KO mice did not show any morphological central nervous system abnormalities [78,79], but behavioral tests point out some cognitive deficits in KO mice, such as impaired sustained attention [80,81], impairment in working memory [82], and impairment in performance under high attentional demand [83]. The cognitive deficits seen in APP transgenic mice worsen when α7 nAChR is absent at the same time [84]. These α7 nAChR KO APP mice showed significant reduction in hippocampal and basal forebrain choline acetyltransferase activity and loss of hippocampal neurons and markers; stereological analyses indicated more pronounced loss of hippocampal pyramidal neurons and volume loss compared with APP mice [84]. Taken all together, it is likely that α7 nAChR might play an important role in the process of Aβ disposition which was detected in the brain of patients with AD.
It is reported that nicotine inhibits Aβ deposition and aggregation in the cortex and hippocampus of APP transgenic mice [94]. RNA interference experiments indicated  that these nicotine-mediated effects require α7 nAChR. In another study [70], the selective α7 nAChR agonist A-582941 led to increased phosphorylation of the inhibitory regulating amino acid residue Ser-9 on glycogen synthase kinase 3β (GSK3β), a major kinase responsible for tau hyperphosphorylation in AD neuropathology. This was observed in mouse cingulate cortex and hippocampus and was not observed in α7 nAChR KO mice. S9-GSK3β phosphorylation was also seen in the hippocampus of Tg2576 (APP), as well as wild-type mice by steady-state exposure of A-582941. Moreover, continuous infusions of A-582941 decreased phosphorylation of tau in hippocampal CA3 Mossy fibers in a hypothermia-induced tau hyperphosphorylation mouse model and also decreased spinal motoneurons in AD double transgenic APP/tau mouse line. This group points out that α7 nAChR agonists may have therapeutic potential through GSK3β inhibition followed by reduction of tau hyperphosphorylation and further suggest that this pharmacology may have the potential to provide disease modifying benefit in the treatment of AD. It is reported that the α7 nAChR agonist GTS-21 prevented Aβ 25−−35 -induced impairment of acquisition performance and probe trail test in Morris water maze [95]. Their study showed first in vivo evidence that treatment with GTS-21 ameliorates the Aβ-induced deficit in spatial cognition through not only activating α7 nAChR but also preventing the Aβ-impaired α7 nAChR.
Taken all together, α7 nAChR is one of the therapeutic targets for AD [97,98].

Postmortem
Studies of α7 nAChR in the Brain of Patients with AD. In the postmortem brain of patients with AD, decline of α7 nAChR appears early in the disease and was associated with the progression of cognitive deficits [99][100][101]. Although the protein levels are reduced in the cortex and hippocampus of AD patients [15,19,32,100,102], contradictions arise at the level of gene transcription. For example, levels of α7 nAChR protein were reduced by 36% in the hippocampus of AD patients [15], but α7 nAChR mRNA expression is increased by 65% [18]. Furthermore, no differences in [ 125 I]α-bungarotoxin binding were found in the frontal cortex of AD patients [103] and negative reduction of the α7 nAChR protein levels [104].

3.4.
Imaging of α7 nAChR in the Brain. Given the role of α7 nAChR in the pathogenesis of AD, it is of great interest to study α7 nAChR in the living human brain using PET/SPECT. Much effort has been devoted to visualize α7 nAChR in the brain by PET/SPECT, but the development of a radioligand that depicts α7 nAChR specifically has been problematic due to its relatively low amount in the brain [105][106][107][108]. Generally, α-bungarotoxin and MLA are well known as specific α7 nAChR antagonists. However, due to their large molecular weights, they have difficulty passing through the blood-brain barrier which makes them unfavorable for radioligands [109][110][111]. Consequently, a number of radioligands for α7 nAChR are being developed and evaluated as PET/SPECT radioligand. However, all radioligands except [ 11 C]CHIBA-1001 were unsuccessful [112].  (Figure 7). A PET study using conscious monkeys demonstrated that the distribution of radioactivity in the brain regions after intravenous administration of [ 11 C]CHIBA-1001 was blocked by pretreatment with the selective α7 nAChR agonist SSR180711 (5.0 mg/kg), but not the selective α4β2 nAChR agonist A85380 (1.0 mg/kg) [89]. In addition, we reported that the order of drugs for the inhibition of [ 3 H]CHIBA-1001 binding to rat brain membranes was similar to α7 nAChR pharmacological profiles [113]. We also reported a preliminary PET study of [ 11 C]CHIBA-1001 in a healthy human [114,115]. Very recently, we reported that [ 125 I]CHIBA-1006, an iodine derivative of SSR180711, has a high affinity for α7 nAChR as compared with CHIBA-1001 [116]. Considering the good brain permeability of derivatives (e.g., SSR180711 and CHIBA-1001) of CHIBA-1006, it would be of great interest  to examine whether [ 123 I]CHIBA-1006 and [ 124 I]CHIBA-1006 are suitable radioligands for in vivo labeling of α7 nAChRs in the brain using SPECT and PET, respectively [116].

[
At present, [ 11 C]CHIBA-1001 is the only PET ligand which can be available for in vivo study of α7 nAChRs in intact human brain [114]. PET studies of

Conclusions
Considering the importance of early prevention of onset of AD, it is very important to detect alternations in nAChRs at the presymptomatic stage of AD. In patients with MCI, the early detection and early therapeutic intervention would be beneficial. Therefore, brain imaging of nAChRs using PET and SPECT will be a powerful tool to study the mechanisms underlying pathological brain processes of cognitive disturbances in these patients. Currently, some PET and SPECT ligands for both subtypes (α4β2 nAChR and α7 nAChR) have been used to investigate the changes in receptor densities and functions of patients with AD. Gaining a better understanding of the role of nAChRs in the pathophysiology of AD is expected to provide new perspectives for treating this disorder.

Conflict of Interests
The authors have no conflict of interests.