Alterations in Phase-Related Prefrontal Activation During Cognitive Tasks and Nicotinic α 4β2 Receptor Availability in Alzheimer’s Disease

Abstract

Background: Evidence shows that the cholinergic system plays an important role in regulating working memory and that working memory-related prefrontal activation decreases with age and neuronal degeneration, such as Alzheimer’s disease (AD). However, the relation between attention-related α4β2 nicotinic cholinergic function and task-induced prefrontal activation especially time course-related activation remains to be explored.

Objective: We aimed to elucidate the relationship between changes in task-induced oxy-hemoglobin concentration (cerebral blood flow, CBF) in the prefrontal cortex and the availability of α4β2 nicotinic receptors in the brain of AD patients in light of their task performance.

Methods: Eleven mild-to-moderate AD patients and eleven normal elderly subjects underwent the near-infrared spectroscopy during easy and difficult working memory tasks for estimating prefrontal CBF changes and positron emission tomography with the α4β2 tracer [¹⁸F]2FA-85380 ([¹⁸F]2FA) for measuring the α4β2 nicotinic receptor binding.

Results: Significant correlations between mean oxy-hemoglobin concentration in the channels with significant [group] main effects and prefrontal [¹⁸F]2FA binding were observed during the early easy task period in the normal group and during the late difficult task in the AD group. In addition, those prefrontal CBF responses were significantly correlated with not correct performance but the execution time to spend.

Conclusion: The α4β2 nicotinic acetylcholine receptors in the prefrontal cortex play an important role in increasing prefrontal activation when attending to novel stimuli, irrespective of the accuracy of the outcome. A delay in the cholinergic-induced increase in prefrontal activation in AD patients might explain their delayed responses in the cognitive task.

Keywords

Alzheimer’s disease cerebral blood flow nicotinic acetylcholine receptor prefrontal cortex working memory

INTRODUCTION

The prefrontal cortex (PFC) controls higher brain functions such as working memory, attention, and decision-making [1]. Particularly, the anterior part of the PFC (aPFC), which is also known as the rostral prefrontal cortex or Brodmann’s area 10, plays important roles in prospective memory, reallocation of attention, episodic goal monitoring, and shifting from internal to external attention [2, 3], and encoding the outcome of a decision before the answer is recognized [4]. Previous reports show that executive functions, including attention, task-switching, multi-tasking, and working memory, are impaired in Alzheimer’s disease (AD) [5 –7]. Whether prefrontal activation occurs during cognitive tasks in AD, however, remains controversial. In AD patients, the resting cerebral blood flow (CBF) was decreased by 10% to 30% in the cerebral cortex, and especially in the frontal cortex, compared with age-matched elderly subjects, and CBF was correlated with dementia severity [8]. Task-related reductions in CBF were reported in the dorsolateral PFC (DLPFC) of AD patients using near-infrared spectroscopy (NIRS), which demonstrated that DLPFC activation is decreased during a verbal fluency test in mild-to-moderate AD patients [9]. In contrast, task-related increases in prefrontal CBF have been reported as a potential compensatory mechanism in AD. In a [¹⁵O]H₂O positron emission tomography (PET) study during semantic and recognition task, mild AD patients showed increased prefrontal activity during a cognitive task, and the degree of this recruitment was associated with task performance and compensation for reduced activity in the hippocampus and parietal cortex [10]. In addition, the compensatory activation observed in early-stage AD patients during a prefrontal task was localized to the premotor cortex [11] and left inferior PFC [12].

It has been reported that the nicotinic acetylcholine receptor (nAChR) system is involved in modulating attention and enhancing cognitive performance [13, 14]. For the purpose of visualizing nAChRs in the living humans, (S)(–)¹¹C-nicotine ([¹¹C]nicotine) was the first tracer applied in the clinical setting [15]. With this tracer, the degrees of attention in Digit symbol test and Trail Making Test were shown to relate to the [¹¹C]nicotine binding in the frontal and parietal areas in mild AD patients [16]. Intervention studies with treatment of cholinesterase inhibitors showed an increase of [¹¹C]nicotine binding [17, 18] and a reinforcing the relation between the fronto-parietal [¹¹C]nicotine binding and attentional performance in mild AD [19]. As suggested in a series of [¹¹C]nicotine PET studies with treatment, the cortical [¹¹C]nicotine binding is considered to be associated with attention rather than other cognition such as memory [20]. However, a major defect of [¹¹C]nicotine may be that the binding specificity of the tracer to nAChRs is low. Among the nAChRs subtypes, the α4β2 heteromer is the most abundant in the brain and has a high affinity for agonists, including nicotine [21]. The α4β2 subtype of nAChRs is involved in many cognitive and behavioral actions, especially in attentional control both in animal and human studies [22, 23]. In contrast with [¹¹C]nicotine that is nonselective for nAChR subtypes, [¹⁸F]2FA-85380 ([¹⁸F]2FA) is considered to be sensitive to the α4β2 subtype. In AD, a significant reduction of α4β2 nAChRs was reported [24, 25] and the level of α4β2 nAChR availability was likely associated with cognitive impairment [25, 26]. From the pharmacological view point, although the galantamine-induced improvement of cognition was not related to α4β2 nAChR [27], some α4β2 nAChR-mediated agonists were reported to affect performance in working memory and attention-requiring tasks [28, 29]. Taken together, α4β2 nAChRs are considered to play an important role in regulation of cognitive functions especially attentional valence.

NIRS is a suitable method for measuring temporal hemodynamic changes. We recently reported different brain activation patterns during the course of working memory tasks in elderly and young individuals [30]. In this study, elderly subjects showed working memory-related CBF responses in the aPFC that were reduced during the pre-task period but gradually increased later in the task period, suggesting that the phase-dependent brain activation may vary with the aging process and perhaps in dementia, such as in AD. NIRS studies that have examined the hemodynamics of AD patients reported decreased activation in the PFC during driving [31], the loss of hemispheric asymmetry [32] and reduced cerebral oxygenation [33, 34]. So far, pathophysiological alterations have been reported mainly at the group level, hence the NIRS approach would provide a valuable contribution [35] by focusing on the task periods in which CBF increases occur and on an individual basis.

In the present study, we used NIRS and PET with the α4β2 nicotinic receptor tracer [¹⁸F]2FA to compare the temporal patterns of task-related prefrontal CBF changes in healthy elderly individuals and AD patients. In addition, we examined the relationship between the degree of CBF changes during the task periods and [¹⁸F]2FA binding to determine whether phase-related CBF responses might be associated with cholinergic functioning, and specifically α4β2 nAChRs, in the PFC of AD patients.

MATERIALS AND METHODS

Participants

Seven drug-naïve and four drug-free non-smoking patients with early-to-moderately severe AD (six women and five men; mean age, 63.4 ± 7.9 y) and 11 drug-naïve non-smoking healthy elderly control (HC) subjects (seven women and four men; mean age, 72.2 ± 5.9 y) took part in this study (Table 1). Four drug-free patients had been taking irregularly a different kind of anticholinesterase drugs prescribed at clinics within a year before the study. To exclude the effect of the drugs on the PET findings, we asked all patients to stop taking all medicines but anti-hypertensive or anti-hyperlipidemia for 4 days as a washout period. Under the exclusion criteria that any candidates with nootropic drugs taken routinely and/or smoking history were excluded, only 22 people in total were able to enter this study. This recruitment caused to yield a significant age difference between groups. As shown in Table 1, several neuropsychological tests were conducted at entry to evaluate cognitive ability and included the following: the Mini-Mental State Examination (MMSE) for general cognition, the Frontal Assessment Battery (FAB) for frontal lobe functioning or executive functions, the Rivermead Behavioral Memory Test (RBMT) for everyday memory performance, the logical memory I (LM1) and logical memory II (LM2) of the Wechsler Memory Scale-Revised (WMS-R) for verbal and episodic memory, and the Zung self-rating depression scale (SDS) for mood. Our inclusion criteria for healthy control were as follows: 1) Clinical Dementia Rating (CDR) score was 0, 2) MMSE score was over 26 (more than 10 years of education), and was over 24 (less than 10 years of education), 3) WMS-R logical memory >3 in order to exclude probable mild cognitive impairment (MCI) subjects. All participants did not take anything that could interfere with their cognition or wakefulness, such as caffeine or alcohol. They were given adequate rest prior to any cognitive testing in order to eliminate physical and mental stress before measurements. A questionnaire of daily and nocturnal activities revealed that no participants had trouble with sleep, so none took sleeping pills regularly. The present study was approved by the Ethics Committee of Hamamatsu Medical Center, and all procedures were performed in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to their enrollment.

Magnetic resonance imaging data acquisition and analysis

All participants underwent a 3D magnetic resonance imaging (MRI) scan. Magnetic resonance images were obtained using a static magnet (0.3 T MRP7000AD, Hitachi) with the following acquisition parameters: 3D mode sampling, repetition time/echo time (200/23), 75° flip angle, 2-mm slice thickness with no gap, and 256×56 matrices (x = 0.938 mm, y = 0.938 mm, z = 1.2 mm) [36].

Because the PFC of AD patients is reported to have greater volumetric atrophy than that of healthy elderly individuals [37], we measured the distance between the scalp and the cortical surface along the AC-PC line on the MRI image using ImageJ (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, http://imagej.nih.gov/ij/). As the distance between the two center probes of the NIRS device placed across the midline are 30 mm, we manually measured the distance at a point that was 15 mm from the midline. We then compared this distance between the HCs and AD patients using the unpaired t-test and found no significant difference between the two groups (Supplementary Table 1).

NIRS data acquisition

To measure changes in the concentrations of oxygenated hemoglobin (oxy-Hb), deoxygenated hemoglobin (deoxy-Hb), and total hemoglobin (total-Hb), we used a continuous wave NIRS device (OEG-16, Spectratech Inc.) with two wavelengths of near-infrared light (770 nm and 840 nm) and based on the Modified Lambert-Beer law [38]. The emitter-detector spacing was 3 cm, which was assumed to measure approximately 2-3 cm beneath the scalp [39, 40]. The optode probe consisted of two rows of photodiodes (three emitters and three detectors) in each row, resulting in a total of 16 channels that were placed on the forehead over the prefrontal area. The midpoint between a pair of photodiodes (an emitter and a detector) located in the center of the bottom row was located at Fpz according to the international 10/20 system for electroencephalography. We examined the positions of the emitter and detector probes for the current NIRS system using MRI (0.3 T MRP7000AD, Hitachi) and confirmed that the photodiodes in the center of the bottom row were located on the rostrum of the PFC (Fig. 1B). Because the continuous wave system cannot obtain path length information, the absolute value of the change in hemoglobin concentration was not determined. Hence, the unit of measurement is expressed as mmol×mm [41]. The sampling rate for the recording was 1.526 Hz.

Activation task

The activation task was the same visual working memory task used in our previous study [30]. In the previous PET study, we used a hand-shape image memory task [42] and modified for NIRS measurement. As shown in Fig. 1A, one experimental session consisted of six task blocks interleaved with a fixed-duration resting period lasting 30 s. One task block (28.8 s duration) consisted of two types of working memory tasks, which are shown in Fig. 1A. The easy visual working memory task (eT) was presented during the first half of the task block (14.4 s duration), whereas the difficult visual working memory task (dT) was presented during the second half of the task block (14.4 s duration). In the recognition phase of both the eT and dT, subjects were asked to select a picture of a hand that was presented in the acquisition phase. Non-flipped images of the hand were presented in the eT and flipped images (i.e., the rear side hand was shown) were presented in the dT. The eT and dT periods each included two trials, with each trial lasting 7.2 s. As shown in Fig. 1A, each trial included a start cue (“!”) that was presented for 1000 ms, an acquisition phase (500 ms) in which the hand shape was memorized, a maintenance phase (4000 ms) with a white cross to keep the image in mind, and a recognition phase (1700 ms), in which the picture of the correct hand was selected from two pictures. The subjects were asked to quickly press the corresponding button on a controller device that they held with both hands.

Before the NIRS measurements began, researchers provided the subjects with instructions on how to perform the task. Each subject was familiarized and fully informed of the task sequences and the lengths of the resting periods. Subjects were instructed not to move their head and, additionally, to stare attentively at a blue cross on the monitor without thinking during the resting periods.

NIRS data analysis

The NIRS data were inspected for motion artifacts before the data were analyzed. For the analysis, we used a Brain Vision Analyzer (Brain Products GmbH, Gilching, Germany) and Matlab software (MathWorks, Natick MA). We used the concentration changes of the oxy-hemoglobin concentration value as an indicator of hemodynamic changes because it is considered to reflect mostly regional changes in CBF [43]. We evaluated the concentration changes of oxy-Hb (Δ[oxy-Hb]) quantitatively throughout the entire experimental session on all 16 channels. The values of each sampling point were normalized by dividing each sampling point by its standard deviation calculated from the whole experimental period for each channel. To minimize physiological noise and correct for drift artifacts, we used low-cut 0.005 Hz and high-cut 0.1 Hz filters, as well as global direct current trend correction [44].

We used the Δ[oxy-Hb] level measured during the resting period from 25 to 15 s before the task began as a baseline. We selected this baseline period because one of our main interests was to examine differences in the pre-task activation between HC subjects and AD patients. We averaged six blocks of normalized Δ[oxy-Hb] for each group and divided the averaged data according to the three task periods. The mean Δ[oxy-Hb] was then calculated for each segment.

PET data acquisition and analysis

The PET procedure was based on a previous report [25], but the later scan was adopted in the current study. In brief, data were collected using a high-resolution brain PET scanner (SHR12000, Hamamatsu Photonics K.K., Hamamatsu, Japan) [45]. After head fixation was achieved using a thermoplastic facemask, a 10-min transmission scan was performed. Then, a total duration of 30 min (3×10 min) scan was performed 210 min after the bolus injection of [¹⁸F]2FA (4 MBq/kg) without arterial blood sampling. [¹⁸F]2FA binding was estimated as a ratio (ratio index of binding potential; BP_RI) of the [¹⁸F]2FA PET count at a target region to the PET count at a pseudo-reference region (the corpus callosum), because this structure is a region with low nAChR density and minimal age-related changes [46] or a region with negligible amount of nAChR [47, 48]. A caveat is that the selected reference region was not completely devoid of nAChR. These parametric BP_RI images were proceeded using image processing software (Dr View, Asahi Kasei Co., Tokyo, Japan) on a SUN workstation (Ultrasparc, SUN Microsystems, San Diego, CA, USA), as described elsewhere [45, 49]. In this procedure, no partial volume correction was performed because a preliminary volumetric analysis disclosed no significant reduction in volume in AD (data not shown). This finding might be partly due to the age difference, i.e., nearly 10 years younger in AD than healthy control. The parametric BP_RI images were normalized to the standardized Montreal Neurological Institute (MNI) template using voxelwise statistical parametric mapping (SPM) software (SPM8, Wellcome Department of Cognitive Neurology, London, UK) as previously reported [25].

Statistical analysis

The mean reaction time and the correct answer ratio (scores for each subject/full scores) in the working memory task were analyzed using unpaired t-tests between the HC and AD groups. Among the neuropsychological tests, the scores for the RBMT, SDS, and WMS-R (LM1, LM2) were analyzed for 10 HC subjects and 10 AD patients because one subject in each group missed the tests. Statistical significance was set to p < 0.05. All statistical analyses were performed with SPSS ver. 19 for Windows.

We performed the chi-square test to examine the effect of the numbers of men and women between groups and to confirm that any differences did not affect the results (χ² = 0.188, df = 1, p = 0.5). Because the unpaired t-test revealed a significant age difference between groups (t = 2.957, df = 20, p = 0.008), and the mean age of the HC group was higher than that of AD group, we performed a multiple regression analysis to investigate whether age contributed to changes in CBF. We verified that age did not contribute to changes in CBF (see Supplementary Table 2). Rather, older patients might exhibit greater atrophy in the frontal cortex, which could have been a greater nuisance in the NIRS study in general.

To analyze differences in the temporal pattern of changes in Δ[oxy-Hb], we performed a two-way ANOVA [group (HC, AD) and period (preT, eT, dT)] for each channel. Furthermore, to examine the differences in Δ[oxy-Hb] values among the task periods and between the normal and AD subjects, we performed an additional analysis using an unpaired t-test in each channel, which revealed a significant main effect [group].

For the correlation analysis between hemodynamic changes and behavioral data, we analyzed the correlation between the Δ[oxy-Hb] values in each task period (preT, eT and dT) and the behavioral data (correct answer ratio, reaction time) in each group using Pearson’s correlation coefficient.

For PET data analysis, individual MRI images were first co-registered to [¹⁸F]2FA PET data, and the co-registered MRI images were spatially normalized to the T1-weighted MRI template using the algorithm provided with SPM8. Then, using the generated normalization parameters, [¹⁸F]2FA BP_RI parametric images of each individual were normalized to the MNI space. The normalized images were smoothed with an isotropic Gaussian kernel of 6 mm full width half maximum. Because there was an age difference between the HC and AD groups, all the between-group and correlation analyses were conducted using age as a covariate. First we analyzed differences in [¹⁸F]2FA binding between the HC and AD groups. Then, we performed voxel-wise correlation analyses using task-induced CBF changes measured by NIRS as covariates. We used the mean Δ[oxy-Hb] values of the channels in which the significant main effect [group] (i.e., ch4, 7, 9, 10, 11, 12, 13, and 14) was shown. The level of significance was set at p < 0.001 uncorrected for peak height with a cluster size larger than 50 because the regions of focus were known a priori as areas with cholinergic innervation[50].

RESULTS

Cognitive performance

The unpaired t-test revealed that the reaction time in the HC group was significantly shorter than that in the AD group (t = –4.121, df = 18, p < 0.001), and the correct answer ratio was significantly higher in the HC group than in the AD group (t = 5.175, df = 20, p < 0.001) (Table 1).

Cerebral blood flow responses

The grand average waveform of Δ[oxy-Hb] tended to be lower in the AD group than that in the HC group, showing that the initial activation was delayed and remained elevated during the task, especially during dT in the AD group (Fig. 2A and B). Two-way ANOVA for each channel showed a significant main effect of [period] in the following channels: 1, 2, 3, 4, 5, 11, 13, 14, 15, 16 (F (2, 40) = 8.716 to 16.791, p < 0.001), 6, 8, 12 (F (2, 40) = 6.24 to 7.993, p < 0.01) and 9 (F (2, 40) = 3.871, p < 0.05). There was a significant main effect of [group] in the following channels: 4 (F (1, 20) = 5.707, p < 0.05), 7 (F (1, 20) = 8.377, p < 0.01), 9 (F (1, 20) = 11.238, p < 0.01), 10 (F (1, 20) = 8.87, p < 0.01), 11 (F (1, 20) = 6.322, p < 0.05), 12 (F (1, 20) = 7.399, p < 0.05), 13 (F (1, 20) = 5.546, p < 0.01) and 14 (F (1, 20) = 9.375, p < 0.01) (Fig. 3). Repeated measures ANOVA did not show a significant interaction between the period×group factors in any channel. A post-hoc unpaired t-test revealed that Δ[oxy-Hb] in the AD group was significantly lower than that of the HC group in channels 9 (t = 2.282, p < 0.05), 10 (t = 2.901, p < 0.01), 11 (t = 2.849, p < 0.01), 12 (t = 2.622, p < 0.05) and 13 (t = 2.6, p < 0.05) during the pre-task period, in channels 7 (t = 2.9, p < 0.01), 9 (t = 2.828, p < 0.05), 10 (t = 2.645, p < 0.05), 11 (t = 2.246, p < 0.05), 12 (t = 2.251, p < 0.05), 13 (t = 2.259, p < 0.05) and 14 (t = 2.614, p < 0.05) during the eT period, and in channels 4 (t = 2.893, p < 0.01), 7 (t = 2.126, p < 0.05), 9 (t = 2.975, p < 0.01) and 14 (t = 2.713, p < 0.05) during the dT period (Fig. 3).

Correlation between cerebral blood flow response and cognitive function

The correlation between task performance (reaction time and correct answer ratio) and Δ[oxy-Hb], there was a significant positive correlation between reaction time and Δ[oxy-Hb] during the eT period, in which there was a remarkable elevation of CBF in the medial PFC in HC group (Table 2, Fig. 2B). Within the AD group, we found no correlation between the task performance and hemodynamic responses.

Regional [¹⁸F]2FA binding and task-induced oxy-hemoglobin concentration in the PFC

Between-group SPM analysis showed a significant reduction in [¹⁸F]2FA BP_RI level in AD patients compared to HC subjects in the cholinergic projection region (Fig. 4A), as was previously reported [25]. No significant correlations were found between [¹⁸F]2FA BP_RI and Δ[oxy-Hb] in any channel during the pre-task period in either the HC or AD group. However, there were significant correlations between [¹⁸F]2FA BP_RI and Δ[oxy-Hb] level during the eT period in the HC group (Fig. 4B, Table 3), and during the dT period in the AD group (Fig. 4C, Table 3). These observations indicated that the periods showing increased CBF were interestingly linked with α4β2 nAChR availability in the PFC, irrespective of the disease entity.

DISCUSSION

The present study showed different patterns of CBF responses between normal elderly subjects and AD patients during working memory tasks. These differences in CBF responses were dependent on the task phase, and AD patients were slow to respond during the task, which was consistent with the delayed increase in CBF that occurred later in the task period. Interestingly, the degree of the CBF increase correlated with the [¹⁸F]2FA BP_RI level in the cholinergic projection region. The present results suggested that nicotinic α4β2 acetylcholine receptors are important for augmenting neuronal activation in the PFC in response to cognitive demands under both aging and pathological conditions (AD). This α4β2 effect may also be present in early-stage AD patients.

Behavioral data

The present data regarding cognitive performance showed that AD patients compared to HC exhibited a significantly slower performance (delayed reaction time) and showed greater difficulty in responding correctly in the working memory task. This can also be observed empirically in their daily lives. Consistent with our behavioral findings, previous reports indicated that deficits in executive functioning developed in not only early-stage AD patients [51] but also pre-clinical AD subjects [52 –56]. In addition, a longitudinal cohort study that included a 16-year observation period indicated a working memory decline during the prodromal phase of AD that developed earlier than deficits in other cognitivefunctions [57].

The difference in temporal hemodynamics between the HC and AD groups

The present results showed significant group differences in CBF responses in the aPFC betweenHC subjects and AD patients (channels 4/7/9/10/11/12/13/14). Although a significant interaction between group×period was not found in any channel, a delayed CBF increase and a prolonged return-to-baseline response after task cessation were found in AD patients, as demonstrated in the waveform shown in Fig. 2B. The CBF increase in the pre-task period, which was more remarkable in healthy young adults than elderly adults [30], was not observed in AD patients, though the present HC subjects showed larger responses to some degree in some channels. Rather, the activation peak in the HC group was shown in the eT period in many channels, which is consistent with our previous data recorded in 60 subjects [30].

An increase in CBF Δ[oxy-Hb] during eT and the following decrease during dT in HC subjects may be partly ascribed to habituation effects because attention to novel stimuli reduces with habituation to the task [58]. This eT-phase activation in the HC group was likely to occur in the rostral part of aPFC (i.e., ch10 in Figs. 2 and 3), and this observation is consistent with the data from our previous study [30]. In contrast, the activation peak in the DLPFC region was present during the dT period in elderly subjects, which was also observed in the previous study. It is speculated that the difference in CBF responses between the rostral aPFC and DLPFC may result from a preparatory mechanism in which aPFC activity enhances DLPFC activity via the posterior-to-anterior hierarchy in the PFC [59]. In a previous report, the aPFC was reported to become active earlier than the DLPFC during the practice phase, possibly because a higher-level task set representation was loaded from long-term memory into the aPFC during preparation for experienced task [60, 61]. One caveat on the regional difference is that the aPFC and DLPFC regions could not be anatomically distinguished, and its functional area overlapped in the present NIRS results.

The reduced PFC activation in AD

In the present study, AD patients compared with HC subjects showed a lesser and slower degree of CBF responses entirely in the PFC. The reduced response in the PFC is in line with the previous finding that executive function impairment was correlated with a reduction in brain glucose metabolism in the left lateral prefrontal cortex of AD patients [62]. In contrast to this reduction in the PFC response, some reports describe an increase in CBF responses in terms of compensatory activation in subjects with amnestic MCI [63] and in cognitively intact people carrying APOE ɛ4 [64]. However, such hyperactivation would gradually disappear with a subsiding episodic memory capacity and a reduction in the hippocampal volume [64]. It was also reported that compensatory hyperactivation that developed in the right PFC of MCI subjects with mild cognitive deficits subsequently disappeared in those with the severe cognitive deficits [65]. Likewise, hypoperfusion was reportedly present in the left medial frontal lobe in AD patients in comparison with amnestic MCI subjects who exhibited prefrontal hyperactivation compared with cognitively healthy subjects [63]. These lines of evidence suggest that the brains of AD patients, who showed mean MMSE scores of 15.6 ± 4.6 in this study, might have little capacity to respond to working memory stimuli in a compensatory manner. Another explanation for the lack of a prefrontal compensatory response in AD is that the given task might be too difficult for the patient to perform because the relationship between the task load and activation is considered to form an inverted-U pattern [66, 67]. In reality, the hypoactivation was often observed in AD patients with poorer performances in various tasks [68].

The relationship between nicotinic α4β2 receptor and CBF change

In the present study, SPM analyses showed significant positive correlations between [¹⁸F]2FA BP_RI and Δ[oxy-Hb] during the eT period in the HC group and with Δ[oxy-Hb] during the dT period in the AD group. This finding suggests that the increase in CBF responses in each period was linked with α4β2 nAChR availability in the PFC. In other words, a higher availability of α4β2 nAChRs in the PFC might result in increased brain activity during the period when the brain activity is augmented enough to execute the given task, irrespective of task performance. The functions of α4β2 nAChR are associated with attention and memory regulation, and a significant correlation was found between [¹⁸F]2FA BP_RI in the prefrontal areas and FAB scores in healthy subjects [25], suggesting that impairment in α4β2 nAChR function would affect executive functions, and most of all, attention [69]. Indeed, an α4β2 partial agonist was shown to increase working memory-related brain activity in humans [28]. Thus, the present result showing that its receptor binding was related to activation during the period in which the task-induced activation was most pronounced suggests that the working memory-task-induced activation could have been boosted by the functions of the nAChR, irrespective of a disease state. As illustrated in Fig. 4, the magnitude of CBF in the eT-phase was positively correlated with [¹⁸F]2FA binding in the lateral and dorsal PFC in the HC group, whereas the dT-phase CBF magnitude was positively associated with [¹⁸F]2FA binding in the medial PFC in the AD group. Because there was a positive correlation between the eT-phase CBF response and the execution time to spend but not with the number of correct answers (achievement) in the HC group, it is possible to speculate that α4β2 nAChR function might be an important driving force for executing cognitive tasks in the elderly. This contention may be true because the delayed activation during the dT period was not relevant to task outcome in AD patients. Indeed, it was reported that cholinergic stimulation by intravenous injection of the acetylcholine-esterase inhibitor physostigmine enhanced encoding but interfered with retrieval in humans [70]. Therefore, the α4β2 nAChR function in the PFC in AD patients may be physiologically important for driving them to perform a cognitive task, irrespective of the outcome (right or wrong).

Limitations

There were some limitations in this study. First, the number of subjects was small due to the strict rule for the present inclusion criteria. From the statistical viewpoint under this situation, we evaluated the number of sample using power analysis because of a risk of type II error. We performed a study of a continuous response variable from independent control and experimental subjects with 1 control per experimental subject. The response within each subject group was expected to be normally distributed with standard deviation 0.15. If the true difference between the experimental and control means is 0.2, we would need to study 10 experimental subjects and 10 control subjects to be able to reject the null hypothesis that the population means of the patients and control groups are equal with probability (power) 0.8 under p value <0.05. Even though the statistic passed the number, a larger size would allow comparisons of different brain loci that are implicated in α4β2 nAChR regulation in cognitive tasks. Second, since previous studies of substance-induced effect on the α4β2 nAChR showing that specific binding of the [¹⁸F]2FA might exist in the corpus callosum would hamper appropriate quantification of [¹⁸F]2FA binding, our [¹⁸F]2FA data might be based on the “pseudo-reference region” analysis. However, our study was conducted without any influence of drugs or chemicals that could perturb the tracer kinetics of [¹⁸F]2FA in the tissue (specific or non-specific). Nonetheless, the BP_RI method without arterial input correction may confound the exact binding of nAChR in the frontal cortex due to varying blood flow. Third, the pathlength of the light used with NIRS was unknown and therefore the Δ[oxy-Hb] used in this study was not the absolute value. Accordingly, we used a baseline measure of CBF to allow quantitative comparisons of CBF responses. Finally, the NIRS data included physiological noise (i.e., Mayer wave) and skin blood flow, all of which could not be excluded [71]. Specifically, a stress-induced perfusion change under the skin during NIRS measurements would contaminate the brain-evoked NIRS data. In order to reduce this effect, we paid careful attention to have participants feel comfortable and relaxed in the experimental setting. No special stress was confirmed by the answer for each inquiry after NIRS measurements.

CONCLUSION

The temporal changes in CBF responses in the PFC during cognitive tasks were different between the HC and AD groups in that the response appeared later in AD patients. The correlation of the PFC α4β2 nAChR availability with the activation in both healthy elderly individuals and AD patients suggests a boostingeffect of the α4β2 nAChR system in the PFC. Reduced α4β2 nAChR in AD patients may cause a weaker enhancing effect in executing cognitive tasks.

Footnotes

ACKNOWLEDGMENTS

We would like to thank Mr. Hiroyuki Okada, Mr. Masami Futatsubashi, Ms. Tomomi Shinke (Hamamatsu Photonics K.K.), and Mr. Tohihiko Kanno (Hamamatsu Medical Photonics Foundation), for their dedicated support. Additionally, we would thank the participants in this study. This work was supported by grants from the New Energy and Industrial Technology Development Organization (NEDO), from the Japanese Ministry of Education, Culture, Sports, Science and Technology and from the Takeda Science Foundation.

Authors’ disclosures available online ().

Appendix

References

Rajah

, D’Esposito

(2005) Region-specific changes in prefrontal function with age: A review of PET and fMRI studies on working and episodic memory. Brain 128, 1964–1983.

Badre

, D’Esposito

(2009) Is the rostro-caudal axis of the frontal lobe hierarchical?. Nat Rev Neurosci 10, 659–669.

Ramnani

, Owen

(2004) Anterior prefrontal cortex: Insights into function from anatomy and neuroimaging. Nat Rev Neurosci 5, 184–194.

Soon

, Brass

, Heinze

, Haynes

(2008) Unconscious determinants of free decisions in the human brain. Nat Neurosci 11, 543–545.

Perry

, Hodges

(1999) Attention and executive deficits in Alzheimer’s disease. A critical review. Brain 122(Pt 3), 383–404.

Godefroy

, Azouvi

, Robert

, Roussel

, LeGall

, Meulemans

(2010) Dysexecutive syndrome: Diagnostic criteria and validation study. Ann Neurol 68, 855–864.

Amieva

, Phillips

, Della Sala

, Henry

(2004) Inhibitory functioning in Alzheimer’s disease. Brain 127, 949–964.

Nicolakakis

, Hamel

(2011) Neurovascular function in Alzheimer’s disease patients and experimental models. J Cereb Blood Flow Metab 31, 1354–1370.

Herrmann

, Langer

, Jacob

, Ehlis

, Fallgatter

(2008) Reduced prefrontal oxygenation in Alzheimer disease during verbal fluency tasks. Am J Geriatr Psychiatry 16, 125–135.

10.

Grady

, McIntosh

, Beig

, Keightley

, Burian

, Black

(2003) Evidence from functional neuroimaging of a compensatory prefrontal network in Alzheimer’s disease. J Neurosci 23, 986–993.

11.

Ouchi

, Yoshikawa

, Futatsubashi

, Okada

, Torizuka

, Kaneko

(2004) Activation in the premotor cortex during mental calculation in patients with Alzheimer’s disease: Relevance of reduction in posterior cingulate metabolism. Neuroimage 22, 155–163.

12.

Backman

, Andersson

, Nyberg

, Winblad

, Nordberg

, Almkvist

(1999) Brain regions associated with episodic retrieval in normal aging and Alzheimer’s disease. Neurology 52, 1861–1870.

13.

Picciotto

, Caldarone

, King

, Zachariou

(2000) Nicotinic receptors in the brain. Links between molecular biology and behavior. Neuropsychopharmacology 22, 451–465.

14.

Sarter

, Parikh

, Howe

(2009) nAChR agonist-induced cognition enhancement: Integration of cognitive and neuronal mechanisms. Biochem Pharmacol 78, 658–667.

15.

Kadir

, Nordberg

(2010) Target-specific PET probes for neurodegenerative disorders related to dementia. J Nucl Med 51, 1418–1430.

16.

Kadir

, Almkvist

, Wall

, Langstrom

, Nordberg

(2006) PET imaging of cortical 11C-nicotine binding correlates with the cognitive function of attention in Alzheimer’s disease. Psychopharmacology (Berl) 188, 509–520.

17.

Nordberg

, Lilja

, Lundqvist

, Hartvig

, Amberla

, Viitanen

, Warpman

, Johansson

, Hellstrom-Lindahl

, Bjurling

, et al (1992) Tacrine restores cholinergic nicotinic receptors and glucose metabolism in Alzheimer patients as visualized by positron emission tomography. Neurobiol Aging 13, 747–758.

18.

Nordberg

, Amberla

, Shigeta

, Lundqvist

, Viitanen

, Hellstrom-Lindahl

, Johansson

, Andersson

, Hartvig

, Lilja

, Langstrom

, Winblad

(1998) Long-term tacrine treatment in three mild Alzheimer patients: Effects on nicotinic receptors, cerebral blood flow, glucose metabolism, EEG, and cognitive abilities. Alzheimer Dis Assoc Disord 12, 228–237.

19.

Kadir

, Darreh-Shori

, Almkvist

, Wall

, Langstrom

, Nordberg

(2007) Changes in brain 11C-nicotine binding sites in patients with mild Alzheimer’s disease following rivastigmine treatment as assessed by PET. Psychopharmacology (Berl) 191, 1005–1014.

20.

Kadir

, Darreh-Shori

, Almkvist

, Wall

, Grut

, Strandberg

, Ringheim

, Eriksson

, Blomquist

, Langstrom

, Nordberg

(2008) PET imaging of the in vivo brain acetylcholinesterase activity and nicotine binding in galantamine-treated patients with AD. Neurobiol Aging 29, 1204–1217.

21.

Nordberg

(1992) Neuroreceptor changes in Alzheimer disease. Cerebrovasc Brain Metab Rev 4, 303–328.

22.

Guillem

, Bloem

, Poorthuis

, Loos

, Smit

, Maskos

, Spijker

, Mansvelder

(2011) Nicotinic acetylcholine receptor beta2 subunits in the medial prefrontal cortex control attention. Science 333, 888–891.

23.

Demeter

, Sarter

(2013) Leveraging the cortical cholinergic system to enhance attention. Neuropharmacology 64, 294–304.

24.

Kendziorra

, Wolf

, Meyer

, Barthel

, Hesse

, Becker

, Luthardt

, Schildan

, Patt

, Sorger

, Seese

, Gertz

, Sabri

(2011) Decreased cerebral alpha4beta2* nicotinic acetylcholine receptor availability in patients with mild cognitive impairment and Alzheimer’s disease assessed with positron emission tomography. Eur J Nucl Med Mol Imaging 38, 515–525.

25.

Okada

, Ouchi

, Ogawa

, Futatsubashi

, Saito

, Yoshikawa

, Terada

, Oboshi

, Tsukada

, Ueki

, Watanabe

, Yamashita

, Magata

(2013) Alterations in alpha4beta2 nicotinic receptors in cognitive decline in Alzheimer’s aetiopathology. Brain 136, 3004–3017.

26.

Sabri

, Kendziorra

, Wolf

, Gertz

, Brust

(2008) Acetylcholine receptors in dementia and mild cognitive impairment. Eur J Nucl Med Mol Imaging 35(Suppl 1), S30–S45.

27.

Ellis

, Nathan

, Villemagne

, Mulligan

, Saunder

, Young

, Smith

, Welch

, Woodward

, Wesnes

, Savage

, Rowe

(2009) Galantamine-induced improvements in cognitive function are not related to alterations in alpha(4)beta (2) nicotinic receptors in early Alzheimer’s disease as measured in vivo by 2-[18F]fluoro-A-85380 PET. Psychopharmacology (Berl) 202, 79–91.

28.

Loughead

, Ray

, Wileyto

, Ruparel

, Sanborn

, Siegel

, Gur

, Lerman

(2010) Effects of the alpha4beta2 partial agonist varenicline on brain activity and working memory in abstinent smokers. Biol Psychiatry 67, 715–721.

29.

Dunbar

, Inglis

, Kuchibhatla

, Sharma

, Tomlinson

, Wamsley

(2007) Effect of ispronicline, a neuronal nicotinic acetylcholine receptor partial agonist, in subjects with age associated memory impairment (AAMI). J Psychopharmacol 21, 171–178.

30.

Oboshi

, Kikuchi

, Shimizu

, Yoshimura

, Hiraishi

, Okada

, Magata

, Ouchi

(2014) Pre-task prefrontal activation during cognitive processes in aging: A near-infrared spectroscopy study. PLoS One 9, e98779.

31.

Tomioka

, Yamagata

, Takahashi

, Yano

, Isomura

, Kobayashi

, Mimura

(2009) Detection of hypofrontality in drivers with Alzheimer’s disease by near-infrared spectroscopy. Neurosci Lett 451, 252–256.

32.

Fallgatter

, Roesler

, Sitzmann

, Heidrich

, Mueller

, Strik

(1997) Loss of functional hemispheric asymmetry in Alzheimer’s dementia assessed with near-infrared spectroscopy. Brain Res Cogn Brain Res 6, 67–72.

33.

Hock

, Villringer

, Muller-Spahn

, Wenzel

, Heekeren

, Schuh-Hofer

, Hofmann

, Minoshima

, Schwaiger

, Dirnagl

, Villringer

(1997) Decrease in parietal cerebral hemoglobin oxygenation during performance of a verbal fluency task in patients with Alzheimer’s disease monitored by means of near-infrared spectroscopy (NIRS)–correlation with simultaneous rCBF-PET measurements. Brain Res 755, 293–303.

34.

Arai

, Takano

, Miyakawa

, Ota

, Takahashi

, Asaka

, Kawaguchi

(2006) A quantitative near-infrared spectroscopy study: A decrease in cerebral hemoglobin oxygenation in Alzheimer’s disease and mild cognitive impairment. Brain Cogn 61, 189–194.

35.

Obrig

(2014) NIRS in clinical neurology - a ‘promising’ tool?. Neuroimage 85 Pt 1, 535–546.

36.

Ouchi

, Nobezawa

, Okada

, Yoshikawa

, Futatsubashi

, Kaneko

(1998) Altered glucose metabolism in the hippocampal head in memory impairment. Neurology 51, 136–142.

37.

Salat

, Kaye

, Janowsky

(2001) Selective preservation and degeneration within the prefrontal cortex in aging and Alzheimer disease. Arch Neurol 58, 1403–1408.

38.

Cope

, Delpy

, Reynolds

, Wray

, Wyatt

, van der Zee

(1988) Methods of quantitating cerebral near infrared spectroscopy data. Adv Exp Med Biol 222, 183–189.

39.

Villringer

, Chance

(1997) Non-invasive optical spectroscopy and imaging of human brain function. Trends Neurosci 20, 435–442.

40.

Ohmae

, Ouchi

, Oda

, Suzuki

, Nobesawa

, Kanno

, Yoshikawa

, Futatsubashi

, Ueda

, Okada

, Yamashita

(2006) Cerebral hemodynamics evaluation by near-infrared time-resolved spectroscopy: Correlation with simultaneous positron emission tomography measurements. Neuroimage 29, 697–705.

41.

Maki

, Yamashita

, Ito

, Watanabe

, Mayanagi

, Koizumi

(1995) Spatial and temporal analysis of human motor activity using noninvasive NIR topography. Med Phys 22, 1997–2005.

42.

Kikuchi

, Hirosawa

, Yokokura

, Yagi

, Mori

, Yoshikawa

, Yoshihara

, Sugihara

, Takebayashi

, Iwata

, Suzuki

, Nakamura

, Ueki

, Minabe

, Ouchi

(2011) Effects of brain amyloid deposition and reduced glucose metabolism on the default mode of brain function in normal aging. J Neurosci 31, 11193–11199.

43.

Hoshi

, Kobayashi

, Tamura

(2001) Interpretation of near-infrared spectroscopy signals: A study with a newly developed perfused rat brain model. J Appl Physiol (1985) 90, 1657–1662.

44.

Hennighausen

, Heil

, Rosler

(1993) A correction method for DC drift artifacts. Electroencephalogr Clin Neurophysiol 86, 199–204.

45.

Ouchi

, Kanno

, Okada

, Yoshikawa

, Futatsubashi

, Nobezawa

, Torizuka

, Sakamoto

(1999) Presynaptic and postsynaptic dopaminergic binding densities in the nigrostriatal and mesocortical systems in early Parkinson’s disease: A double-tracer positron emission tomography study. Ann Neurol 46, 723–731.

46.

Ellis

, Nathan

, Villemagne

, Mulligan

, Ellis

, Tochon-Danguy

, Chan

, O’Keefe

Gx J

, Bradley

, Savage

, Rowe

(2009) The relationship between nicotinic receptors and cognitive functioning in healthy aging: An in vivo positron emission tomography (PET) study with 2-[(18)F]fluoro-A-85380. Synapse 63, 752–763.

47.

Brody

, Mandelkern

, London

, Olmstead

, Farahi

, Scheibal

, Jou

, Allen

, Tiongson

, Chefer

, Koren

, Mukhin

(2006) Cigarette smoking saturates brain alpha 4 beta 2 nicotinic acetylcholine receptors. Arch Gen Psychiatry 63, 907–915.

48.

Mukhin

, Kimes

, Chefer

, Matochik

, Contoreggi

, Horti

, Vaupel

, Pavlova

, Stein

(2008) Greater nicotinic acetylcholine receptor density in smokers than in nonsmokers: A PET study with 2-18F-FA-85380. J Nucl Med 49, 1628–1635.

49.

Ouchi

, Yoshikawa

, Futatsubashi

, Yagi

, Ueki

, Nakamura

(2009) Altered brain serotonin transporter and associated glucose metabolism in Alzheimer disease. J Nucl Med 50, 1260–1266.

50.

Gallezot

, Bottlaender

, Gregoire

, Roumenov

, Deverre

, Coulon

, Ottaviani

, Dolle

, Syrota

, Valette

(2005) In vivo imaging of human cerebral nicotinic acetylcholine receptors with 2-18F-fluoro-A-85380 and PET. J Nucl Med 46, 240–247.

51.

Gordon

, Zacks

, Blazey

, Benzinger

, Morris

, Fagan

, Holtzman

, Balota

(2015) Task-evoked fMRI changes in attention networks are associated with preclinical Alzheimer’s disease biomarkers. Neurobiol Aging 36, 1771–1179.

52.

Balota

, Tse

, Hutchison

, Spieler

, Duchek

, Morris

(2010) Predicting conversion to dementia of the Alzheimer’s type in a healthy control sample: The power of errors in Stroop color naming. Psychol Aging 25, 208–218.

53.

Backman

, Jones

, Berger

, Laukka

, Small

(2005) Cognitive impairment in preclinical Alzheimer’s disease: A meta-analysis. Neuropsychology 19, 520–531.

54.

Monsell

, Mock

, Hassenstab

, Roe

, Cairns

, Morris

, Kukull

(2014) Neuropsychological changes in asymptomatic persons with Alzheimer disease neuropathology. Neurology 83, 434–440.

55.

Hutchison

, Balota

, Duchek

(2010) The utility of Stroop task switching as a marker for early-stage Alzheimer’s disease. Psychol Aging 25, 545–559.

56.

Aretouli

, Okonkwo

, Samek

, Brandt

(2011) The fate of the 0.5s: Predictors of 2-year outcome in mild cognitive impairment. J Int Neuropsychol Soc 17, 277–288.

57.

Wilson

, Leurgans

, Boyle

, Bennett

(2011) Cognitive decline in prodromal Alzheimer disease and mild cognitive impairment. Arch Neurol 68, 351–356.

58.

Warach

, Gur

, Skolnick

, Obrist

, Reivich

(1992) Decreases in frontal and parietal lobe regional cerebral blood flow related to habituation. J Cereb Blood Flow Metab 12, 546–553.

59.

Badre

(2008) Cognitive control, hierarchy, and the rostro-caudal organization of the frontal lobes. Trends Cogn Sci 12, 193–200.

60.

Cole

, Bagic

, Kass

, Schneider

(2010) Prefrontal dynamics underlying rapid instructed task learning reverse with practice. J Neurosci 30, 14245–14254.

61.

Sakai

, Passingham

(2006) Prefrontal set activity predicts rule-specific neural processing during subsequent cognitive performance. J Neurosci 26, 1211–1218.

62.

Schroeter

, Vogt

, Frisch

, Becker

, Barthel

, Mueller

, Villringer

, Sabri

(2012) Executive deficits are related to the inferior frontal junction in early dementia. Brain 135, 201–215.

63.

Ding

, Ling

, Zhang

, Huang

, Zhang

, Wang

, Yan

(2014) Pattern of cerebral hyperperfusion in Alzheimer’s disease and amnestic mild cognitive impairment using voxel-based analysis of 3D arterial spin-labeling imaging: Initial experience. Clin Interv Aging 9, 493–500.

64.

Rao

, Bonner-Jackson

, Nielson

, Seidenberg

, Smith

, Woodard

, Durgerian

(2015) Genetic risk for Alzheimer’s disease alters the five-year trajectory of semantic memory activation in cognitively intact elders. Neuroimage 111C, 136–146.

65.

Clement

, Belleville

(2010) Compensation and disease severity on the memory-related activations in mild cognitive impairment. Biol Psychiatry 68, 894–902.

66.

Callicott

, Mattay

, Bertolino

, Finn

, Coppola

, Frank

, Goldberg

, Weinberger

(1999) Physiological characteristics of capacity constraints in working memory as revealed by functional MRI. Cereb Cortex 9, 20–26.

67.

Mattay

, Fera

, Tessitore

, Hariri

, Berman

, Das

, Meyer-Lindenberg

, Goldberg

, Callicott

, Weinberger

(2006) Neurophysiological correlates of age-related changes in working memory capacity. Neurosci Lett 392, 32–37.

68.

Prvulovic

, Van de Ven

, Sack

, Maurer

, Linden

(2005) Functional activation imaging in aging and dementia. Psychiatry Res 140, 97–113.

69.

Granon

, Faure

, Changeux

(2003) Executive and social behaviors under nicotinic receptor regulation. Proc Natl Acad Sci U S A 100, 9596–9601 .

70.

Kukolja

, Thiel

, Fink

(2009) Cholinergic stimulation enhances neural activity associated with encoding but reduces neural activity associated with retrieval in humans. J Neurosci 29, 8119–8128.

71.

Kirilina

, Jelzow

, Heine

, Niessing

, Wabnitz

, Bruhl

, Ittermann

, Jacobs

, Tachtsidis

(2012) The physiological origin of task-evoked systemic artefacts in functional near infrared spectroscopy. Neuroimage 61, 70–81.

Alterations in Phase-Related Prefrontal Activation During Cognitive Tasks and Nicotinic α 4β2 Receptor Availability in Alzheimer’s Disease

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Participants

Magnetic resonance imaging data acquisition and analysis

NIRS data acquisition

Activation task

NIRS data analysis

PET data acquisition and analysis

Statistical analysis

RESULTS

Cognitive performance

Cerebral blood flow responses

Correlation between cerebral blood flow response and cognitive function

Regional [18F]2FA binding and task-induced oxy-hemoglobin concentration in the PFC

DISCUSSION

Behavioral data

The difference in temporal hemodynamics between the HC and AD groups

The reduced PFC activation in AD

The relationship between nicotinic α4β2 receptor and CBF change

Limitations

CONCLUSION

Footnotes

ACKNOWLEDGMENTS

Appendix

References

Regional [¹⁸F]2FA binding and task-induced oxy-hemoglobin concentration in the PFC