Blood MAPT expression and methylation status in Alzheimer's disease

Abstract Aim This study aimed to investigate the expression levels and methylation status of microtubule‐associated protein tau (MAPT) in the blood of Alzheimer's disease (AD) patients and age‐ and sex‐matched healthy controls. Methods Fifty AD outpatients and 50 healthy contorls were enrolled. Blood samples were collected for processing of complementary DNA and genomic DNA. MAPT messenger ribonucleic acid (mRNA) expression was analyzed by real‐time quantitative polymerase chain reaction. The methylation rates of four cytosine‐phosphate‐guanine (CpG) sites in the upstream region of MAPT exon1 were evaluated by the pyrosequencing method. Results No significant differences in MAPT mRNA expression levels were found between AD and control subjects (AD 0.97 ± 0.49 vs. control 1.0 ± 0.64, p = 0.62). MAPT mRNA expression levels were not correlated with any other clinical characteristics or results of psychological tests. MAPT mRNA expression levels were significantly higher in AD subjects treated with acetylcholinesterase inhibitors (AchEIs) (n = 25) than in subjects not treated with AChEIs (n = 25) (unmedicated 0.83 ± 0.33 vs. medicated 1.12 ± 0.59, p = 0.049). The AD subjects did not differ from the control subjects in methylation rates at selected CpG sites. MAPT methylation status were not correlated with clinical characteristics, the results of psychological tests, or MAPT mRNA expression. Conclusion MAPT mRNA expression levels and methylation status in blood do not appear useful as biomarkers for AD or the examined CpG sites were not genetically significant for MAPT gene expression or AD pathology. However, AChEIs may alter MAPT mRNA expression. Further studies are needed to explore blood biomarkers that can discriminate AD patients from controls.


INTRODUCTION
Alzheimer's disease (AD) is a chronic neurodegenerative disease that causes apraxia, aphasia, impairment of memory, and loss of motivation.Pathologically, it is characterized by senile plaques and neurofibrillary tangles. 1 Senile plaques are extracellular accumulations of amyloid beta encoded by amyloid precursor protein on chromosome 21q21. 2Neurofibrillary tangles are the abnormal intracellular accumulation of the hyperphosphorylated protein tau encoded by microtubule-associated protein tau (MAPT) on chromosome 17q21. 3PT has been associated with various neurodegenerative diseases.One of the most typical MAPT-related disorders is frontotemporal dementia with parkinsonism linked to chromosome 17, which is caused by MAPT mutations. 4Progressive supranuclear palsy is also associated with the more common allele (H1) and genotype (H1H1) in MAPT. 5 With regard to AD, increased MAPT messenger ribonucleic acid (mRNA) expression was found in the temporal lobe of sporadic AD patients, 6 but no difference was found between AD and control subjects with regard to MAPT mRNA expression in peripheral blood mononuclear cells. 7The H2-haplotype in MAPT has also been reported to reduce the risk of late-onset AD, 8 and rs393152, a single-nucleotide polymorphism (SNP) at the MAPT locus, was reported to significantly increase the risk of AD. 9 Changes in gene expression without changing DNA sequences are termed epigenetic.DNA methylation is one of several epigenetic processes that regulate gene expression 10 and has been associated with the pathogenesis of various brain diseases. 11,12In the promoter region of MAPT, the DNA methylation status of a cytosine-phosphate-guanine (CpG) island clearly affects mRNA expression. 6,13In brains with sporadic AD, both neuronal and non-neuronal cells had reduced methylation of this CpG island, suggesting that hypomethylation in AD patients increases tau expression, leading to tau aggregation and pathological spread throughout the brain, as in prion disease. 6In Parkinson's disease (PD) patients, MAPT hypermethylation has been reported to occur in the cerebellum, with hypomethylation in the putamen. 14On the other hand, it has been reported that methylation is not altered in AD, PD, and frontotemporal dementia. 15rrently, β-amyloid (1-42), total tau, and phospho-tau-181 in cerebrospinal fluid are useful biomarkers for AD diagnosis. 16wever, lumbar puncture is an invasive and expensive test, and minimally invasive biomarkers are needed that can differentiate AD patients from healthy controls.Our group has previously reported that the methylation status of synuclein alpha (SNCA) in blood could be a useful biomarker for AD. 17 The involvement of immune function in the pathogenesis of AD has also been actively reported in recent years, [18][19][20] attracting attention to blood biomarkers. 21We therefore hypothesized that MAPT mRNA expression levels and the methylation status in blood are biomarkers of AD.In particular, the methylation levels of MAPT in the blood of AD patients have not been investigated, therefore the expression levels and the methylation status of MAPT in blood of AD patients and age-and sexmatched healthy controls were investigated.

AD subjects and healthy control subjects
Demographic data for each group of participants are shown in Table 1.Fifty AD patients (17 males and 33 females, mean age ± standard deviation (SD) = 78.3± 6.2 years), who were outpatients at Ehime University Hospital and Zaidan Niihama Hospital, Ehime, Japan, were enrolled.AD subjects were diagnosed according to the Aging/Alzheimer's Association criteria. 22ey were diagnosed as having probable AD and lived with at least one caregiver.Patients with AD without cerebrovascular lesions on head computed tomography or magnetic resonance imaging were included.AD subjects were assessed with the Mini-

Mental State Examination (MMSE) and Alzheimer's Disease
Assessment Scale (ADAS), which assesses cognitive functions, 23,24 the Montgomery-Åsberg Depression Rating Scale (MADRS) to assess depression symptoms, 25 Clinical Dementia Rating (CDR) by family caregivers, 26 and the Neuropsychiatric Inventory (NPI) 27   (Applied Biosystems).9][30] The final volume of each reaction containing TaqMan Universal Master Mix (Applied Biosystems) was 10 µl.Expression levels were examined by duplicate measurements.The ΔΔC t method and StepOne software (Applied Biosystems) were used to measure relative expression levels.

Bisulfite conversion and pyrosequencing
In pyrosequencing, we analyzed the upstream region of MAPT exon1, which was predicted as a CpG island using the Sequence Manipulation Suite (https://www.bioinformatics.org/sms2/index.html)(Figure 1).The primers of pyrosequencing were designed using PyroMark Assay Design Software (Qiagen).The gDNA (1000 ng/sample) extracted from whole blood was converted with bisulfate using the EpiTect Plus DNA Bisulfite Kit (Qiagen).PCR amplification was then performed using forward primer (5-TGGAAGGTAGTTTAGGATTTTTGTAGG-3) and reverse primer (5-

Participant characteristics
Demographic data and APOE genotypes of the participants are presented in Table 1.There were no differences in sex (P value = 0.18) and age (P value = 0.10) between AD and control subjects.However, there was a significant difference in APOE genotype between the AD and control subjects (P value < 0.001).Clinical characteristics of AD subjects and results of psychological tests are presented in Table 2.

MAPT mRNA expression levels
According to the Shapiro-Wilk test, MAPT expression levels were not normally distributed in both AD group and control subjects.No differences in MAPT mRNA expression levels were found between AD and control subjects (AD 0.97 ± 0.49 vs. control 1.0 ± 0.64, P value = 0.62; Figure 2).
MAPT mRNA expression levels were not significantly different between males and females (AD P value = 0.23, Cnt P value = 0.10) or between ApoE ε4+ and ε4− (AD P value = 0.34, Cnt P value = 0.48).
MAPT mRNA expression levels were significantly higher in AD subjects treated with AChEIs (n = 25) than in subjects not treated with AChEIs (n = 25) (unmedicated 0.83 ± 0.33 vs. medicated 1.12 ± 0.59, P value = 0.049; Figure 3).Between unmedicated AD and control subjects, no significant difference in MAPT mRNA Blood sample collection and processing of cDNA and gDNA Total RNA was collected from whole peripheral blood samples into PaxGene Blood RNA Systems tubes (BD) and extracted according to the manufacturer's protocol.The RNA concentration and purity were measured on a NanoDrop-1000 (Thermo Fisher Scientific) system, with acceptable 260/280 ratios in the range of 1.8-2.0.For each 40µl reaction, complementary DNA (cDNA) was synthesized with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems) with 1.0 µg of RNA as template.The genomic DNA (gDNA) samples were obtained from whole peripheral blood samples collected in potassium ethylenediaminetetraacetic acid tubes.The gDNA was extracted according to the manufacturer's protocol using the QIAamp DNA Blood Mini Kit (Qiagen).PCR procedure Expression of mRNA was analyzed by real-time quantitative polymerase chain reaction (PCR) using a StepOnePlus Real-Time PCR System (Applied Biosystems).The specific Taq-Man probes were Hs00902193 m1 for MAPT and Hs99999905 m1 for GAPDH

F I G U R E 1
Schematic diagram showing the location of MAPT.Correlations between pairs of the four CpGs were analyzed with Spearman's rank correlation coefficient.Statistical significance was defined at P = 0.008 (= 0.05/6) following Bonferroni correction.F, forward primer; R, reverse primer; S, sequence primer; AD, Alzheimer's disease subjects; CpG, cytosine-phosphate-guanine.BLOOD MAPT EXPRESSION AND METHYLATION STATUS IN AD | 3 of 7 performed using Spearman's rank correlation coefficient.Significance was defined at the 95% level (P = 0.05).Statistical significance of correlations between pairs of the 4 CpGs with Spearman's rank correlation coefficient was defined at P = 0.008 (= 0.05/6) following Bonferroni correction.Statistical significance of correlations MAPT mRNA expression with the clinical characteristics was defined at P = 0.004 (= 0.05/14) following Bonferroni correction.Statistical significance of correlations the average of four CpG methylation rates with the clinical characteristics was defined at P = 0.006 (= 0.05/9) following Bonferroni correction.

F I G U R E 2
MAPT mRNA expression levels in AD and control subjects.The mean expression level in AD subjects (average ± SD = 0.97 ± 0.49) is not different from that in control subjects (average ± SD = 1.0 ± 0.64) (Mann-Whitney U test, P = 0.62).The horizontal bars represent the means ± standard error.AD, Alzheimer's disease subjects; Ct, control subjects F I G U R E 3 MAPT mRNA expression levels in AD subjects treated with acetylcholinesterase inhibitors (AChEIs) (n = 25) and those not treated with AChEIs (n = 25).The mean expression level in nonmedicated subjects is significantly different from that in medicated subjects (nonmedicated: 0.83 ± 0.33 vs. medicated: 1.12 ± 0.59, P = 0.049).The horizontal bars represent the means ± standard error.*p < 0.05.
to assess psychological symptoms.Controls were 50 elderly people (11 males and 39 females, mean age ± SD = 76.2 ± 6.2 years) with no cognitive impairment, psychiatric symptoms, or history of previous mental disorders and diagnosed as mentally and cognitively normal by at least two certified psychiatrists based on clinical interviews.All participants were unrelated Japanese and provided written, informed consent approved by the institutional ethics committees of Ehime University Hospital and Zaidan Niihama Hospital (approval numbers: 31-K8, 1901009, and 2109001).