T. Mano1, K. Sato1, T. Ikeuchi2, T. Toda1, T. Iwatsubo3, A. Iwata4, Japanese Alzheimer’s Disease Neuroimaging Initiative5
1. Department of Neurology, Graduate School of Medicine, The University of Tokyo, Bunkyo-ku, Tokyo, Japan; 2. Department of Molecular Genetics, Brain Research Institute, Niigata University, Chuo-ku, Niigata, Japan; 3. Department of Neuropathology, Graduate School of Medicine, The University of Tokyo, Bunkyo-ku, Tokyo, Japan; 4. Tokyo Metropolitan Geriatric Hospital and Institute of Gerontology, Itabashi-ku, Tokyo, Japan; 5. The full list of members of the Japanese Alzheimer’s Disease Neuroimaging Initiative is provided in the supplementary file, «J-ADNI co-investigators».
Corresponding Author: Tatsuo Mano, Department of Neurology, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo Bunkyo-ku, Tokyo 113-8655, Japan, Email: firstname.lastname@example.org, Phone +81-3-5800-8672, Fax +81-3-5800-6548
J Prev Alz Dis 2021;
Published online June 9, 2021, http://dx.doi.org/10.14283/jpad.2021.31
BACKGROUND: Recent biomarker studies demonstrated that the central nervous system (CNS) environment can be observed from peripherally-derived samples. In a previous study, we demonstrated significant hypomethylation of the BRCA1 promoter region in neuronal cells from post-mortem brains of Alzheimer’s disease patients through neuron-specific methylome analysis. Thus, we investigate the methylation changes in the BRCA1 promoter region in the blood samples.
OBJECTIVES: To analyze the methylation level of the BRCA1 promoter in peripheral blood from AD patients and normal controls.
DESIGN, SETTING, PARTICIPANTS: Genomic DNA samples from peripheral blood were obtained from the J-ADNI repository, and their biomarker data were obtained J-ADNI from the National Bioscience Database Center. Genomic DNA samples from an independent cohort for validation was obtained from Niigata University Hospital (Niigata, Japan). Amyloid positivity was defied by visual inspection of amyloid PET or a CSF Aβ42 value ≤ 333 pg/mL at the baseline.
MEASUREMENTS: Methylation level of the BRCA1 promoter was analyzed by pyrosequencing.
RESULTS: Compared to normal controls, methylation of the BRCA1 promoter in AD patients was not significantly changed; however, in AD patients, it showed a positive correlation with AD risk factors.
CONCLUSIONS: Our data confirmed the importance of cell-type specific methylome analysis and also suggested that environmental changes in the CNS can be detected by observing the peripheral blood, implying that the peripheral BRCA1 methylation level can be a surrogate for AD.
Key words: Alzheimer’s disease, methylome, BRCA1, peripheral blood.
Abbreviations: AD: Alzheimer’s disease; Aβ: amyloid β; CNS: central nervous system; NC: normal control; NFT: neurofibrillary tangle; PET: positron emission tomography; CSF: cerebrospinal fluid.
BRCA1 is a nuclear DNA repair protein, and its loss-of-function causes breast and ovarian cancers. In familial cases, insufficient DNA repair caused by loss-of-function mutations leads to genomic instability and contributes to cancer development (20). In sporadic breast cancers, hypermethylation of BRCA1 leads to its downregulation and results in insufficient DNA repair, causing cancer (1, 19). Thus, proper BRCA1 function is thought to be crucial in maintaining genomic DNA homeostasis.
In our previous study, aberrant hypomethylation of the BRCA1 promoter was observed in brains from Alzheimer’s disease (AD) patients (13). In contrast to cancer, BRCA1 was upregulated in association with promoter demethylation. A series of experiments showed that BRCA1 upregulation was a cellular protective response to amyloid β (Aβ)-induced DNA double strand breaks. Despite its upregulation, BRCA1 was sequestered to neurofibrillary tangles (NFTs) and mis-localized to the cytoplasm. This suppressed its function and led to the accumulation of significant DNA damage in AD neurons. Aberrant hypomethylation was observed in neurons as well as glial cells, suggesting that Aβ toxicity affected all types of cells in the central nervous system (CNS).
In AD brains, neurons are far more vulnerable than glial cells to Aβ toxicity. This difference could be explained by the absence of NFTs in glial cells. Thus, functional BRCA1 provided resistance against DNA damage. Hypomethylation was also observed in the cerebellum, suggesting that Aβ was affecting the entire brain. However, its effect on other peripheral tissues and cells is not yet known.
The blood-brain-barrier was once thought to separate the CNS from peripheral blood, with the exception of several small molecules. However, there are a number of studies showing that the CNS environment can be observed from peripherally-derived samples (8, 12). Among patients with bipolar disorders and schizophrenia, there is evidence that genes encoding molecules reported to be involved in these diseases show an altered epigenome in peripheral blood (2, 9, 18). A series of recent studies on AD strongly indicates that fibrillar Aβ accumulation in the CNS can be detected by measuring Aβ levels in peripheral blood samples (5, 15, 16, 21). Thus, we believed that it was worthwhile to analyze the methylation level of the BRCA1 promoter in peripheral blood from AD patients.
Here, we report the levels of BRCA1 promoter methylation using peripheral blood DNA derived from clinically diagnosed AD patients. We also analyzed the methylation levels in J-ADNI participants who underwent either Pittsburgh compound B amyloid positron emission tomography (PET) or cerebrospinal fluid (CSF) Aβ42 analysis that identified them as amyloid pathology-positive, and examined the relationship between the methylation level and clinical features.
All participants provided written informed consent. This study was approved by the ethics committee of the University of Tokyo (approvals G2183-18 and 11628-(1)). All experiments were performed in accordance with the principles of the Declaration of Helsinki.
Genomic DNA samples
Peripheral blood samples were obtained from the J-ADNI repository upon approval from the sample sharing committee. We obtained J-ADNI biomarker data from the National Bioscience Database Center with approval from its data access committee (https://humandbs.biosciencedbc.jp/en/hum0043-v1). We also validated the results of the J-ADNI samples in an independent cohort of patients admitted at Niigata University Hospital (Niigata, Japan). Patients were diagnosed clinically as AD or NC.
Amyloid positivity was defined clinically by visual inspection of PET images using Pittsburgh Compound B as a ligand by trained radiologists who were blinded to any clinical information (22) and/or a CSF Aβ42 value ≤ 333 pg/mL at the baseline, using the cut-off value determined in a previous report (11). When both data were available, we defined a sample as «positive» when at least one test met the «positive» criterion.
Bisulfite-conversion of DNA was performed by applying 100 ng of genomic DNA to an EpiTect Bisulfite Kit (Qiagen, Hilden, Germany), following the manufacturer’s instructions; the final product was eluted with 50 μL buffer. The primers used in this study were published previously (13) (Supplementary Figure 1). Samples were applied to PyroMark Q24 using PyroMark Gold Q24 Reagents (Qiagen), following amplification of bisulfite-converted DNA by the polymerase chain reaction using a PyroMark PCR Kit (Qiagen). The results were analyzed using PyroMark Q24 software (Qiagen). Primer set designs were verified using universal methylated and de-methylated human DNA standards.
Prism 6 (GraphPad, San Diego, CA, USA) and R software were used for statistical analyses. Unless otherwise noted, the significance of differences between groups was determined using the t-test. P < 0.05 was considered significant.
We analyzed the methylation level of the BRCA1 promoter in peripheral blood from normal control (NC) and clinically diagnosed AD patients. The demographics of the subjects are shown in Table 1. We performed pyrosequencing analysis of genomic DNA samples derived from 76 subjects. In total, we analyzed seven CpGs of the BRCA1 promoter region that were differentially methylated in AD brains (13). The linearity of all primer sets was validated using universal methylated and unmethylated human genomic DNA (Figure S1). There were no statistically significant differences in the methylation levels between NC and AD patients in all the probes analyzed (Figure 1). Because these subjects were diagnosed based only on clinical features without knowing brain Aβ accumulation, there remains a possibility that insufficient diagnostic accuracy could have affected the results. Therefore, amyloid positivity should be confirmed for an accurate diagnosis of AD.
* Fisher’s exact test; †The APOE ε4 genotype was not available in five NC samples.
The methylation levels of CpG probes in BRCA1 were analyzed using the validation cohort. Black dots and gray boxes represent normal control (NC) and Alzheimer’s disease (AD) patients, respectively. Error bars represent means + SD. Statistical significance was determined using a two-tailed t-test. n = 40 (NC) and 36 (AD).
Because the previous study on the neuron-specific methylome in AD brains demonstrated that the presence of Aβ was essential for aberrant methylation of the BRCA1 promoter region (13), we examined methylation in peripheral blood from NC and AD subjects with known amyloid pathology from the J-ADNI cohort. The demographics of this cohort are shown in Table 2. Among the 537 subjects enrolled in that study, there were 51 NC and 49 AD subjects for whom amyloid data from amyloid PET or CSF Aβ42 analysis were available. No significant change in the methylation level of the BRCA1 promoter region was observed (Figure 2).
CSF: cerebrospinal fluid
The CpG probes in BRCA1 that were differentially methylated in neuronal cells (13])were analyzed by pyrosequencing. Black dots and the gray boxes represent normal control (NC) and Alzheimer’s disease (AD) patients, respectively. Error bars represent means + SD. Statistical significance was determined using a two-tailed t-test. n = 51 (NC) and 49 (AD).
Because the methylation levels of some CpGs correlate with age, and the age of the participants examined in this study differed between NC and AD subjects, we considered the possibility that age-related effects might have diminished the difference in methylation levels of BRCA1 between NC and AD patients. However, the methylation level of the CpGs in the BRCA1 promoter region were not affected by age (Figure 3). Thus, we concluded that the methylation level of the BRCA1 promoter region derived from peripheral blood did not reflect the accumulation of fibrillar Aβ in the brain.
Correlation plots of the pyrosequenced methylation levels of each BRCA1 probe and age at death. For normal control (NC) patients, blue dots and lines represent individual data and regression lines, respectively. For Alzheimer’s disease (AD) patients, red dots and lines represent individual data and regression lines, respectively. Gray bands represent the 95% confidence interval of each linear regression. Pearson’s product-moment correlation coefficient (r) values between the methylation level and age of NC and AD patients, and subsequent significance, are shown in each graph.
In the brain, methylation levels of the BRCA1 promoter region in neuronal cells were associated with the number of APOE ε4 alleles only in the AD cohort (13). Thus, we analyzed the relationship between sex differences or APOE ε4 and the methylation level of this region in the NC and AD cohorts. As shown in Figure 4A, female sex had a significant positive effect on the methylation level of the BRCA1 promoter region only in the AD group. The number of APOE ε4 alleles had opposite effects on the methylation levels in NC and AD groups; in NCs, the methylation level was negatively correlated with the APOE ε4 allele number, while in AD the level had a positive correlation.
(A, B) Comparison of the methylation level of each differential methylation position based on sex difference (A) and the number of APOE ε4 alleles (B). (A) Orange, green, blue, and purple represent normal control (NC) males, NC females, Alzheimer’s disease (AD) males, and AD females, respectively. Significance was determined using a two-way analysis of variance (ANOVA) followed by the post hoc Sidak method. (B) Orange, yellow, light green, green, blue, and purple represent NC without the APOE ε4 allele, NC with one APOE ε4 allele, NC with two APOE ε4 alleles, AD without APOE ε4 allele, AD with one APOE ε4 allele, and AD with two APOE ε4 alleles, respectively. Whiskers were defined by Tukey’s boxplot method. Significance was determined for NC and AD subjects using a two-way ANOVA followed by the post hoc Tukey’s (for NC) or Sidak (for AD) method. Boxes extend from the 25th–75th percentile. Lines in the boxes are medians.
In this study, we clearly demonstrated that the methylation level of the BRCA1 promoter in peripheral blood was unchanged between NC and AD patients. This was in stark contrast to the global methylation change in the CNS. In the brain, toxic Aβ induces DNA damage regardless of the region or the cell type, resulting in BRCA1 up-regulation through promoter hypomethylation. This allows sufficient DNA repair only in the absence of cytoplasmic aggregated tau because BRCA1 co-aggregates with NFTs and loses its proper function exclusively in neurons. In peripheral blood cells that are apparently free from NFTs, even if peripheral toxic Aβ had any effect on DNA damage, proper repair should occur. The absence of different methylation levels in NC and AD subjects could be attributed to a low Aβ concentration in the peripheral blood leading to a less prominent response of peripheral cells towards Aβ toxicity. Another explanation is that, in peripheral blood, Aβ toxicity could differ from that of the CNS, even if blood biomarker studies show that CNS Aβ accumulation can be detected by an analysis of the peripheral blood. This suggests that the CNS and peripheral blood share common properties of Aβ species.
A sub-analysis revealed several interesting results. Upon stratifying by sex, AD females showed higher methylation levels than males. One of the CpGs showed significantly higher methylation levels even after multiple post-hoc comparisons (Figure 4A, probe cg18372208). As the methylation level of the promoter region generally shows an inverse relationship with the expression of downstream genes (1, 19), the peripheral response to Aβ could be insufficient in females, resulting in vulnerability to toxic Aβ. This response could be related to the fact that being female is a risk factor for AD (4, 17) and also for a rapid decline in cognitive function (3, 6, 7, 10, 14).
In our previous study, we did not observe any sex differences in the BRCA1 promoter methylation level in the CNS (13). This discrepancy could be explained by differences of the organs we analyzed or the disease stage. Specifically, J-ADNI AD patients exhibited mild to moderate dementia, while autopsy patients usually suffer from severe dementia.
When comparing the number of APOE ε4 alleles and BRCA1 methylation levels, the presence of APOE ε4 alleles was negatively associated in NC and positively associated in AD subjects. These opposing results could be explained by a potential protective effect of BRCA1 upregulation in response to peripheral Aβ exposure; APOE ε4 carriers are protected from Aβ toxicity when BRCA1 is upregulated (i.e., a lower methylation level), whereas they are not protected when BRCA1 is downregulated (i.e., higher methylation in AD patients).
The cause of these different responses is not fully clear. Recent studies have shown that Aβ accumulation in the CNS could be detected by analyzing the peripheral blood (5, 15, 16). One study even showed that patients with CNS Aβ accumulation had increased Aβ oligomerization activity in peripheral blood (21). These results imply that aberrant Aβ metabolism could be occurring both in central and peripheral tissues, and could drive BRCA1 upregulation in certain conditions.
In summary, no significant changes in methylation of the BRCA1 promoter were observed in NC and AD patients. Thus, we concluded that BRCA1 promoter methylation cannot be a biomarker for diagnosing AD. Nevertheless, we found distinctive hypomethylation of the BRCA1 promoter in the brain through a neuron-specific methylome analysis. These data emphasized the importance of analyzing specific cell types directly involved in the disease process. However, only in the AD group were risk factors for AD (i.e., female sex and APOE allele ε4 number) correlated with BRCA1 promoter methylation in the peripheral blood, implying that the response to toxic Aβ in terms of BRCA1 expression was shared in the peripheral blood. This could be attributed to recent findings that the brain environment can be partially detected by observing peripheral blood, and provides insights into how the CNS and peripheral blood cross-talk in terms of Aβ accumulation (5, 15, 16, 21). Increased hypomethylation of the BRCA1 promoter has a potentially protective effect against Aβ toxicity. Therefore, while its relevance remains unclear, higher levels of BRCA1 promoter methylation might indicate AD risk, suggesting that peripheral methylation could be a potential biomarker for AD progression.
Author Contributions: Conceptualization, TM and AI; Methodology, TM and KS; Resource, TI and TI; Supervision, TT; Writing – Original Draft Preparation, TM; Writing – Review & Editing, AI; Funding, TM and AI.
Funding: This study was supported by AMED under grant number 17dm0107069h0002, 18dk0207028h0003, 18dk0207020h0004, JP21dk0207057h0001, 21dk0207042h0003, 21dk0207046h0001. JSPS KAKENHI grant numbers 16H05316 and 19K17027, the Cell Science Research Foundation (Osaka, Japan), the Ichiro Kanehara Foundation for the Promotion of Medical Sciences and Medical Care (Tokyo, Japan), the Takeda Science Foundation (Osaka, Japan), The Mochida Memorial Foundation for Medical and Pharmaceutical Research (Tokyo, Japan), Janssen Pharmaceutical K.K. (Tokyo, Japan), and Eisai Co. (Tokyo, Japan).
Acknowledgements: We are grateful for the technical support provided by Yuki Inukai-Mizutani.
Conflicts of interest: None.
1. Baldwin RL, Nemeth E, Tran H, et al., BRCA1 promoter region hypermethylation in ovarian carcinoma: a population-based study, Cancer research 2000;60:5329-5333.
2. Booij L, Szyf M, Carballedo A, et al., DNA methylation of the serotonin transporter gene in peripheral cells and stress-related changes in hippocampal volume: a study in depressed patients and healthy controls, PLoS One 2015;10:e0119061.
3. Butchart J, Birch B, Bassily R, Wolfe L, Holmes C, Male sex hormones and systemic inflammation in Alzheimer disease, Alzheimer Dis Assoc Disord 2013;27:153-156.
4. Caracciolo B, Palmer K, Monastero R, et al., Occurrence of cognitive impairment and dementia in the community: a 9-year-long prospective study, Neurology 2008;70:1778-1785.
5. Fandos N, Perez-Grijalba V, Pesini P, et al., Plasma amyloid beta 42/40 ratios as biomarkers for amyloid beta cerebral deposition in cognitively normal individuals, Alzheimers Dement (Amst) 2017;8:179-187.
6. Ferretti MT, Iulita MF, Cavedo E, et al., Sex differences in Alzheimer disease – the gateway to precision medicine, Nat Rev Neurol 2018;14:457-469.
7. Fyfe I, Alzheimer disease: Sex-specific inflammatory link to early Alzheimer pathology, Nat Rev Neurol 2017;13:5.
8. Gaiottino J, Norgren N, Dobson R, et al., Increased neurofilament light chain blood levels in neurodegenerative neurological diseases, PLoS One 2013;8:e75091.
9. Ikegame T, Bundo M, Murata Y, et al., DNA methylation of the BDNF gene and its relevance to psychiatric disorders, J Hum Genet 2013;58:434-438.
10. Iwata A, Iwatsubo T, Ihara R, et al., Effects of sex, educational background, and chronic kidney disease grading on longitudinal cognitive and functional decline in patients in the Japanese Alzheimer’s Disease Neuroimaging Initiative study, Alzheimers Dement (N Y) 2018;4:765-774.
11. Iwatsubo T, Iwata A, Suzuki K, et al., Japanese and North American Alzheimer’s Disease Neuroimaging Initiative studies: Harmonization for international trials, Alzheimer’s & Dementia 2018.
12. Lu CH, Macdonald-Wallis C, Gray E, et al., Neurofilament light chain: A prognostic biomarker in amyotrophic lateral sclerosis, Neurology 2015;84:2247-2257.
13. Mano T, Nagata K, Nonaka T, et al., Neuron-specific methylome analysis reveals epigenetic regulation and tau-related dysfunction of BRCA1 in Alzheimer’s disease, Proceedings of the National Academy of Sciences 2017.
14. Mosconi L, Berti V, Quinn C, et al., Sex differences in Alzheimer risk: Brain imaging of endocrine vs chronologic aging, Neurology 2017;89:1382-1390.
15. Nakamura A, Kaneko N, Villemagne VL, et al., High performance plasma amyloid-β biomarkers for Alzheimer’s disease, Nature 2018.
16. Ovod V, Ramsey KN, Mawuenyega KG, et al., Amyloid beta concentrations and stable isotope labeling kinetics of human plasma specific to central nervous system amyloidosis, Alzheimers Dement 2017;13:841-849.
17. Roberts RO, Geda YE, Knopman DS, et al., The incidence of MCI differs by subtype and is higher in men The Mayo Clinic Study of Aging, Neurology 2012;78:342-351.
18. Turecki G, Ota VK, Belangero SI, Jackowski A, Kaufman J, Early life adversity, genomic plasticity, and psychopathology, The Lancet Psychiatry 2014;1:461-466.
19. Turner NC, Reis-Filho JS, Russell AM, et al., BRCA1 dysfunction in sporadic basal-like breast cancer, Oncogene 2007;26:2126-2132.
20. Venkitaraman AR, Cancer susceptibility and the functions of BRCA1 and BRCA2, Cell 2002;108:171-182.
21. Wang MJ, Yi S, Han JY, et al., Oligomeric forms of amyloid-beta protein in plasma as a potential blood-based biomarker for Alzheimer’s disease, Alzheimers Res Ther 2017;9:98.
22. Yamane T, Ishii K, Sakata M, et al., Inter-rater variability of visual interpretation and comparison with quantitative evaluation of (11)C-PiB PET amyloid images of the Japanese Alzheimer’s Disease Neuroimaging Initiative (J-ADNI) multicenter study, Eur J Nucl Med Mol Imaging 2017;44:850-857.