Circadian Clock Gene Polymorphisms and Sleep/Wake Disturbance in Alzheimer’s Disease

doi:10.1097/JGP.0b013e31820d92b2

Am J Geriatr Psychiatry. Author manuscript; available in PMC 2012 July 1.

Published in final edited form as:

Am J Geriatr Psychiatry. 2011 July; 19(7): 635–643.

doi: 10.1097/JGP.0b013e31820d92b2

PMCID: PMC3128424

NIHMSID: NIHMS268837

Circadian Clock Gene Polymorphisms and Sleep/Wake Disturbance in Alzheimer’s Disease

Jerome A. Yesavage, M.D.,^1,² Art Noda, M.S.,² Beatriz Hernandez, B.A.,² Leah Friedman, Ph.D.,² Jauhtai J. Cheng, M.D., Ph.D.,¹ Jared R. Tinklenberg, M.D.,^1,² Joachim Hallmayer, M.D.,² Ruth O’Hara, Ph.D.,^1,² Renaud David, M.D.,³ Philippe Robert, M.D.,³ Elizabeth Landsverk, M.D.,⁴ the Alzheimer’s Disease Neuroimaging Initiative,^* and Jamie M. Zeitzer, Ph.D.^1,²

¹ Department of Veterans Affairs Health Care System, Palo Alto, California

² Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Palo Alto, California

³ Memory Center (CMRR), CIU-S, Université de Nice, France

⁴ ElderConsult Medical Associates, Burlingame, California

Address correspondence to: Jerome Yesavage, M.D., Palo Alto VA Health Care System (151Y), 3801 Miranda Avenue, Palo Alto, CA 94304 ; Email: yesavage/at/stanford.edu Phone: 650-852-3287 Fax: 650-852-3297

^*Data used in the preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (www.loni.ucla.edu/ADNI). As such, the investigators within the ADNI provided data but did not participate in analysis or writing of this report. ADNI is the result of efforts of many co-investigators from a broad range of academic institutions and private corporations (see www.adni-info.org.)

The publisher's final edited version of this article is available at Am J Geriatr Psychiatry

See other articles in PMC that cite the published article.

Abstract

Objectives

One of the hypothesized causes of the breakdown in sleep/wake consolidation often occurring in individuals with Alzheimer’s disease (AD) is dysfunction of the circadian clock. The goal of this study is to report indices of sleep/wake function collected from individuals with AD in relation to relevant polymorphisms in circadian clock-related genes.

Design

One week of ad libitum ambulatory sleep data collection.

Setting

At-home collection of sleep data and in-laboratory questionnaire.

Participants

Two cohorts of AD participants. Cohort 1 (n=124): individuals with probable AD recruited from the Stanford/Veterans Affairs NIA Alzheimer’s Disease Core Center (n=81) and the Memory Disorders Clinic at the University of Nice School of Medicine (n=43). Cohort 2 (n=176): individuals with probable AD derived from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) data set.

Measurements

Determination of sleep/wake state was obtained by wrist actigraphy data for seven days in Cohort 1 and by the Neuropsychiatric Inventory (NPI-Q) for Cohort 2. Both cohorts were genotyped using an Illumina Beadstation and 122 circadian-related SNPs were examined. In Cohort 1, an additional polymorphism (variable number tandem repeat in per3) was also determined.

Results

Adjusting for multiple tests, none of the candidate gene SNPs were significantly associated with the amount of wake after sleep onset (WASO), a marker of sleep consolidation. Although the study was powered sufficiently to identify moderate-sized correlations, we found no relationships likely to be of clinical relevance.

Conclusions

It is unlikely that a relationship with a clinically meaningful correlation exists between the circadian rhythm-associated SNPs and WASO in individuals with AD.

Keywords: Alzheimer’s Disease, Sleep/wake Disturbances, Circadian Rhythm

OBJECTIVE

Nocturnal wakefulness and daytime napping often characterize the sleep/wake disturbance frequently associated with Alzheimer’s disease (AD). Previously, we longitudinally followed sleep/wake disturbances in AD participants and found significant nocturnal wake after initial sleep onset (WASO) (1). We also found that sleep/wake deterioration in individuals with AD behaved as a “trait”, i.e., decline was consistently manifested in some individuals while others never manifested such decline over their illness course (2). This led us to search for genetic variations to explain such “traits”. One hypothesized cause of this breakdown in sleep/wake consolidation is degeneration of the circadian clock (3). A limited number of genes likely underlie the core circadian oscillation found in neurons of the suprachiasmatic nucleus (SCN), the location of the circadian clock in mammals (4). Variation within the small number of circadian genes could result in several relevant phenotypes becoming more prominent as compensatory mechanisms are lost during the course of AD. The potential impact of human circadian genetic variation is suggested by recent studies showing that variation in the human CLOCK and per3 genes may be associated with differences in sleep/wake function (5,6). CLOCK and per3 proteins interact with other gene products in the basic circadian system (per1, per2, bmal-1, csnk1e, cry1 and cry2).

Using actigraphy it is feasible to collect sleep/wake phenotype data in large numbers of AD participants and to analyze their genotype data on candidate gene single nucleotide polymorphisms (SNPs) using gene chip technology. This study reports actigraphy sleep/wake data collected from AD participants in relation to their circadian candidate gene polymorphisms.

METHODS

Participants

The analyses in this report were performed on two cohorts of AD participants.

Cohort 1

Inclusion/exclusion criteria required a diagnosis of probable AD by NINCDS-ADRDA criteria (7) based on relevant neurological, medical, neuroimaging, and neuropsychological assessments. Participants were excluded if they had active major medical conditions that would have precluded collection of actigraphy data. The 124 participants in this cohort are from two sources: 1) 81 participants in an ongoing longitudinal study of AD at the Stanford/Veterans Affairs NIA Alzheimer’s Disease Core Center (ADCC), and 2) 43 French participants from the Memory Disorders Clinic at the University of Nice School of Medicine that is collaborating with our group on AD sleep studies. For this analysis, only Caucasians were selected from both sources and participants in Cohort 1, as a whole, were 46% male. At the time of actigraphy, the average age of the 124 participants in Cohort 1 was 75.2 years (SD = 8.1; Range = 45 to 88) and their mean MMSE (8) was 19.8 (SD = 4.8; Range = 4 to 29).

Cohort 2

Data were also obtained from a second cohort of 176 Caucasian participants in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (www.loni.ucla.edu/ADNI) with a diagnosis of probable AD by NINCDS-ADRDA criteria.

The ADNI was launched in 2003 by the National Institute on Aging (NIA), the National Institute of Biomedical Imaging and Bioengineering (NIBIB), the Food and Drug Administration (FDA), private pharmaceutical companies and non-profit organizations, as a $60 million, 5-year public-private partnership. The primary goal of ADNI has been to test whether serial magnetic resonance imaging (MRI), positron emission tomography (PET), other biological markers, and clinical and neuropsychological assessment can be combined to measure the progression of mild cognitive impairment (MCI) and early Alzheimer’s disease (AD). Determination of sensitive and specific markers of very early AD progression is intended to aid researchers and clinicians to develop new treatments and monitor their effectiveness, as well as lessen the time and cost of clinical trials.

The Principal Investigator of the ADNI is Michael W. Weiner, M.D., VA Medical Center and University of California – San Francisco. ADNI is the result of efforts of many co-investigators from a broad range of academic institutions and private corporations, and participants have been recruited from over 50 sites across the U.S. and Canada. The initial goal of ADNI was to recruit 800 adults, ages 55 to 90, to participate in the research.

The ADNI participants were studied using the same Illumina gene chip technology as that used for Cohort 1, described above. They did not receive actigraphy but received the Neuropsychiatric Inventory (NPI-Q) (9), which includes sleep and other behavioral ratings. The major reason for including the ADNI/NPI cohort was to provide additional data regarding the possibility that facets of behavior being described by the NPI-Q might be in any way affected by SNPs in the same manner as the sleep data from Cohort 1. Thus since both cohorts were measured on the same SNPs, we felt it worthwhile to provide all available data on behavioral correlates of these SNPs in one analysis.

Only Caucasians were selected from the ADNI database, the cohort was 55% male and the average age was 75.6 years (SD = 7.6; Range = 55 to 91). The mean MMSE (8) for the 176 Cohort 2 participants at the time the NPI data were collected was 23.3 (SD = 2.0; Range = 18 to 27).

The Institutional Review Board for Human Subjects Research at each site in the current analysis approved their respective research protocols. Written informed consent was obtained from each participant or legally authorized representative.

Measures

Actigraphy

Rest/activity data were collected for Cohort 1 by means of a wrist-worn, watch-size ambulatory motion-detecting device, the actigraph (Ambulatory Monitoring Systems, Inc., Ardsley, NY 10502). Participants were asked to wear an actigraph 24 hours a day on their non-dominant wrists for 7 consecutive days and were instructed to remove the device only for bathing or swimming. The actigraph was set to record motion in 30-second epochs.

Measures of nighttime sleep/wake behavior, in particular wake after sleep onset (WASO), were obtained using the computer scoring program supplied by the manufacturer (ACTION software version 1.3). The ACTION software scores the actigraph recordings following entry of participants’ evening bed times and final morning out-of-bed times derived from daily sleep logs completed by the caregiver. The scores used in the data analyses were averages of the consecutive days of actigraphy; night-to-night variability was not examined in this paper. Not all data were usable due to occasional technical failures of the device, and the amount of actigraph data collected varied across participants depending on their compliance, but 84% of the participants had at least 4 or more days of data per recording session. The mean WASO for the 124 participants in Cohort 1 was 85 minutes (SD = 64; range = 5 to 414). The staff in Nice, France were trained in the use of actigraphy at Stanford and used the same device. Further, all data collected in Nice were double-checked by U.S. staff and both groups used the same diagnostic criteria for AD. An advantage of the actigraph as an outcome measure is that it is not language dependent.

Neuropsychiatric Inventory (NPI-Q)

The NPI-Q is a brief, informant-completed questionnaire assessing the participant’s neuropsychiatric symptoms and the caregiver’s level of associated distress. The NPI-Q is designed to assess dementia patients’ behaviors (delusions, hallucinations, agitation/aggression, depression/dysphoria, anxiety, elation/euphoria, apathy/indifference, disinhibition, irritability/lability, motor disturbance, nighttime behaviors and appetite/eating) regarding their severity and distress to the caregiver during the past month. The item relating to sleep/wake disturbance in nighttime behaviors on the NPI-Q was analyzed for the 176 AD participants in Cohort 2: “Does the patient awaken you during the night, rise too early in the morning, or take excessive naps during the day? (Item K)” On this item, 74% were scored as ‘not a problem’, 17% were ‘mild’, 7% ‘moderate’ and 2% ‘severe’ (see Table 1, which shows the association tests between gene SNPs and NPI severity items). None of the measures used required translation (the NPI-Q reported here was only administered to the U.S.-based ADNI cohort).

Table 1

Table of Association Tests Between 122 Candidate Gene SNPs and NPI Severity Items^*

Genotyping Procedures

Genomic DNA extraction from frozen EDTA-containing whole blood or saliva samples was conducted as previously described (10). Genotyping was performed by an Illumina Beadstation (Illumina, San Diego CA) using the manufacturer-provided procedures. We utilized the Human 610 Quad gene chip that genotypes approximately 658,000 SNPs and has the same coverage as that used in ADNI. SNPs on all relevant circadian candidate genes were genotyped for both cohorts.

Additional Genotyping Procedures

The variable number tandem repeat (VNTR) polymorphism in per3, where a 54-nucleotide coding region motif is repeated in 4 or 5 units, has been linked with multiple sleep phenotypic parameters (6). This VNTR polymorphism could not be determined using Illumina technology so a separate procedure was performed in which the target hPer3 gene fragment was amplified using sense primer (5′-CAAAATTTTATGACACTACCAGAATGGCTGAC-3′) and reverse (5′-AACCTTGTACTTCCACATCAGTGCCTGG-3′) primers. The PCR reaction was carried out in a final volume of 15 μL consisting of 30 ng of genomic DNA, 50 ng each of sense and antisense primers, 7.5 μL of Taq PCR Master mix (Qiagen #201445) and 10% DMSO. The PCR conditions included an initial denaturation step at 95°C for 3 minutes, followed by 35 cycles of denaturation at 95°C for 30 seconds, annealing at 63°C for 1 minute and extension at 72°C for 1 minute and 45 seconds, with a final extension of 8 minutes at 72°C. The 4-repeat polymorphism with a size of 581 bp and the 5-repeat with a size of 635 bp were detected by Western blot.

Analytic Procedures

The analytic approach proceeded along the following steps. We first identified the SNPs available for analysis from the eight circadian candidate genes (CLOCK, per1, per2, per3, bmal-1, csnk1e, cry1 and cry2) using a database linking SNPs on the Illumina Human 610 Quad gene chip with parent genes. There were 136 SNPs from the candidate circadian genes on the Illumina chip. We then performed quality control using Ilumina’s BeadStudio Software (Version 3.1, San Diego CA) and visually examined each SNP to validate the called genotype clusters. We discovered one SNP with problematic genotyping because one signal channel was not performing adequately (rs1441351) and excluded this SNP from further analysis, leaving 135 SNPs. Next, using Golden Helix SNP and Variation Suite (SVS Version 7.2.2, Bozeman MT) software, SNPs were filtered to exclude failed SNPs according to one or more of the following criteria: call rate < 0.95; minor allele frequency < 5%; Fisher’s Exact Test for HWE P < 0.001. This filtering reduced the number of candidate gene SNPs to 122. Finally, we used the SVS software genotype association test with the following options: 1) Additive model testing, 2) Correlation/Trend test, 3) Drop missing values, 4) Full scan permutation with 10,000 permutations (a correction for multiple testing), and 5) Correcting batch effects/stratification with Principal Components Analysis (PCA).

RESULTS

Table 2 shows the association testing between the 122 candidate gene SNPs and the relevant sleep/wake phenotype variable, WASO. None of the candidate gene SNPs or the VNTR polymorphism was significantly associated with WASO score after full scan permutation. The full scan permutation is a permutation test over the distribution of minima for the p-values calculated in the full genome-wide scan for association provided by the Golden Helix Software. “Full scan”, as defined by Golden Helix, means that the software determines the minimal p-value over the entire set of p-values for the study, randomizing the phenotype with each re-sampling iteration (11).

Table 2

Table of Association Tests Between 122 Candidate Gene SNPs and WASO

The VNTR polymorphism appeared to be associated with one of the PER3 SNPs in our sample (rs228729, see Table 3). This is a different SNP from that previously found (rs2640909) to be associated with the PER3 VNTR in a Japanese sample (12). These two SNPs are not well associated in Caucasian (r²=0.117) or Japanese individuals (r²=0.014) (SNP Annotation and Proxy Search, www.broadinstitute.org/mpg/snap/ldsearch.php, based on HapMap release 22).

Table 3

The VNTR Polymorphism Association Tabulated with one PER3 SNP

A similar analysis was also negative using the same SNPs in the 176 cases from the ADNI dataset and their NPI measure of nighttime behavioral disturbance. Parenthetically, we found that there were no full-scan permutation-corrected significant correlations between the selected SNPs and any of the NPI severity measures (see Table 1, which shows the association tests between gene SNPs and NPI severity items).

CONCLUSIONS

The analysis conducted with 124 participants and 122 SNPs has over 80% power to detect a two-tailed correlation of 0.38 for any one of the 122 SNPs with WASO scores. Thus we conclude that it is unlikely that a clinically relevant relationship (one with a “medium” effect size) (13) exists between these circadian rhythm-associated SNPs and WASO, our primary measure of nocturnal sleep disturbance, in individuals with AD, although this is based on a possible 20% false negative rate. It is possible that there may be an association between these SNPs and sleep pathologies commonly found in this population (e.g., sleep disordered breathing and restless legs syndrome), though this remains for future study. These results do not rule out the possibility that smaller correlations exist; however, correlations smaller than 0.38 would explain less than 14% of the variance of the clinical phenomena, hence may not be clinically relevant. The data in Table 2 suggest the possibility that relationships of a smaller magnitude might exist with some SNPs on cry1 or per3; however, none of these correlations at the < 0.30 level were significant after the full scan permutation correction. Such smaller relationships may still be of theoretical interest to those examining basic physiological relations of genetic measures and sleep/wake phenomena, even if they may not be clinically relevant. Exploratory analyses were also performed using the ADNI dataset and scores on other NPI items. In these analyses no SNPs were significantly associated with any of the other NPI measures. We feel it would be premature to draw any conclusions from these results other than to suggest the need for further examination of these relationships in other datasets to determine if they are replicable.

In summary, we examined genetic sources of variability in two independent samples of AD patients with different but complementary measures of sleep/wake disturbances. We did not find relationships that are likely to be of clinical relevance, even though the study was powered sufficiently to identify correlations of a moderate size. We conclude that sources of variation of neuropsychiatric symptoms in AD do not lie in simple relationships to specific SNPs associated with circadian rhythms but possibly depend upon other physiological mechanisms or interactions among a number of genetic markers.

Acknowledgments

Grant support: This research was supported by NIH grants MH40041 (NIMH) and AG17824 (NIA), the Sierra-Pacific Mental Illness Research, Education, and Clinical Center (MIRECC), the Medical Research Service of the Department of Veterans Affairs, and an Alzheimer’s disease plan grant (Conseil Général des Alpes-Maritimes, Nice, France). ADNI research was supported by The Foundation for the National Institutes of Health (AG024904), which coordinates the private sector participation of the ADNI public-private partnership as well as non-profit partners, the Alzheimer’s Association and the Institute for the Study of Aging.

We express appreciation to our clinical research staff: Terry Miller, M.D., Helen Davies, R.N.M.S. and Aimee Stepp, for clinical testing and assessments. Special appreciation is expressed to Deryl Wicks who organized the longitudinal actigraphy follow-up and quantification and to Chun-Ping (Phoebe) Liao who performed the Illumina genotyping. The Foundation for the National Institutes of Health (www.fnih.org) coordinates the private sector participation of the $60 million ADNI public-private partnership that was begun by the National Institute on Aging (NIA) and supported by the National Institutes of Health. To date, more than $27 million has been provided to the Foundation for NIH by Abbott, AstraZeneca AB, Bayer Schering Pharma AG, Bristol-Myers Squibb, Eisai Global Clinical Development, Elan Corporation, Genentech, GE Healthcare, GlaxoSmithKline, Innogenetics, Johnson & Johnson, Eli Lilly and Co., Merck & Co., Inc., Novartis AG, Pfizer Inc., F. Hoffmann-La Roche, Schering-Plough, Synarc Inc., and Wyeth, as well as non-profit partners the Alzheimer’s Association and the Institute for the Study of Aging. The Department of Veterans Affairs, War Related Illness and Injury Study Center, Palo Alto, CA provided the support for the purchase and maintenance of the Illumina technology used in this research.

References

1. Yesavage JA, Friedman LF, Kraemer HC, et al. A follow-up study of actigraphic measures in home-residing Alzheimer’s disease patients. J Geriatr Psychiatry Neurol. 1998;11:7–10. [PubMed]

2. Yesavage JA, Taylor JL, Kraemer H, et al. Sleep/wake disturbance in Alzheimer’s disease: How much is due to an inherent trait? Int Psychogeriatr. 2002;14:73–81. [PubMed]

3. Swaab DF, Lucassen PJ, Salehi A, et al. Reduced neuronal activity and reactivation in Alzheimer’s disease. Prog Brain Res. 1998;17:343–377. [PubMed]

4. King DP, Zhao Y, Sangoram AM, et al. Positional cloning of the mouse Clock gene. Cell. 1997;89:641–653. [PubMed]

5. Katzenberg D, Young T, Finn L, et al. A CLOCK polymorphism associated with human diurnal preference. Sleep. 1998;21:569–576. [PubMed]

6. Dijk DJ, Archer SN. PERIOD3, circadian phenotypes, and sleep homeostasis. Sleep Med Rev. 2010;14:151–160. [PubMed]

7. McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer’s disease: A report of the NINCDS-ADRDA work-group under the auspices of the Department of Health and Human Services Task Force on Alzheimer’s disease. Neurology. 1984;34:939–944. [PubMed]

8. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–198. [PubMed]

9. Kaufer DI, Cummings JL, Ketchel P, et al. Validation of the NPI-Q, a brief clinical form of the Neuropsychiatric Inventory. J Neuopsychiatry Clin Neurosci. 2000;12:233–239. [PubMed]

10. Murphy GM, Jr, Taylor J, Kraemer HC, et al. No association between apolipoprotein E epsilon 4 allele and rate of decline in Alzheimer’s disease. Am J Psychiatry. 1997;154:603–608. [PubMed]

11. Westfall PH, Young SS. Resampling-based multiple testing: examples and methods for p-value adjustment. Wiley Interscience; New York: 1993.

12. Ebisawa T, Uchiyama M, Kajimura N, et al. Association of structural polymorphisms in the human period3 gene with delayed sleep phase syndrome. EMBO Rep. 2001;2:342–346. [PubMed]

13. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2. Lawrence Erlbaum Associates; 1988. (See also “A Power Primer” under the Wikipedia entry for “Jacob Cohen”)