Blood RNA transcripts show changes in inflammation and lipid metabolism in Alzheimer's disease and mitochondrial function in mild cognitive impairment

Abstract

Background

Abnormal immunity in the periphery has been reported in the pathogenesis of Alzheimer's disease (AD).

Objective

In this study, blood transcriptome analyses of patients with AD, those with mild cognitive impairment (MCI) due to AD, and heathy controls were performed to elucidate immune-related pathophysiology.

Methods

The sample included 63 participants from a complete enumeration study of elderly people in Nakayama town (the Nakayama Study), who were over 65 years of age, diagnosed as (1) healthy controls (N = 21, mean age: 83.8 years), (2) having MCI due to AD (N = 20, mean age: 82.6 years), or (3) having AD (N = 21, mean age: 84.2 years). Every participant underwent blood tests, magnetic resonance imaging, and questionnaires about lifestyle and cognitive function. With transcriptome analysis, differential gene expressions in the blood of the three groups were evaluated by gene ontology, pathway enrichment, and ingenuity pathway analyses, and quantitative real-time PCR was performed.

Results

Neutrophil extracellular trap signaling was increased, and lipid metabolism (FXR/RXR activation, triacylglycerol degradation) was decreased in AD, whereas MCI showed protective responses via decreased neutrophil extracellular trap signaling and mitochondrial functions such as upregulation of the sirtuin pathway and downregulation of oxidative stress.

Conclusions

Based on these findings and consistent with other published studies, immune cells appear to have important roles in the pathogenesis of AD, and the transcriptome in blood may be useful as a biomarker for diagnosis via monitoring immunity in MCI and AD.

Keywords

Alzheimer's disease blood mild cognitive impairment immune lipid mitochondria transcriptome

Introduction

In recent years, disease-modifying drugs for Alzheimer's disease (AD) have emerged, and the importance of biomarkers for early detection and treatment assessment has been reported.¹ Positron emission tomography and spinal fluid testing for amyloid-β and phosphorylated tau have been used for its biomarkers and are highly useful, but they have not yet reached the stage of routine use due to a lack of available facilities, high cost, and high invasiveness.² In contrast, blood biomarkers have the advantage of being less invasive, and their validity has been actively examined in recent years with the aim of generalizing their use.³ Use of whole blood transcriptomic analysis has increased in recent years due to its advantages in terms of minimal invasiveness and repeatable measurement of a large number of gene expressions. For example, pathway and gene-set enrichment analyses of the blood of AD showed differences in immune response-related pathways enriched throughout the analysis, despite different underlying gene sets for different races.^4,5 The use of small sets of blood transcripts from inflammation, stress, and epigenetics has also shown the potential to distinguish between patients with neurodegenerative diseases, such as AD, and healthy controls.⁶ Inflammation and redox genes selected from whole blood transcriptomic analysis have also been suggested as potentially useful biomarkers for monitoring anti-inflammatory therapy in mild AD.⁷ Although still small, the number of studies of whole blood transcriptomic analysis of AD is increasing and mainly detects immune abnormalities, but it is hoped that AD of varying stages in different races will be studied cross-sectionally and longitudinally.

Recently, it has been suggested that immune cells are not only biomarkers for AD, but are also involved in the pathogenesis of the disease.^8,9 For example, the migration of peripherally derived immune cells, such as T cells and neutrophils, to the brain has been shown to be involved in brain aging and neurodegeneration.^10–12 Serum C-reactive protein (CRP) and inflammatory cytokines, indicators of systemic inflammation, have been shown to be elevated in AD.^13,14 We have recently reported that elevated serum high-sensitivity CRP levels were associated with AD prevalence and temporal lobe atrophy in a large cohort study.¹⁵ It has also been suggested that anti-inflammatory agents may slow the progression of AD.^16,17 Recent data suggest that peripheral blood may be an appropriate proxy for gene expression profiles identified in AD, such as aberrant immune and inflammatory system function and mitochondrial dysfunction.^9,18–23 Transcriptomic studies using postmortem AD brains also showed alterations in pathways including cell proliferation, apoptosis, immune response, synaptic transmission, and energy metabolism.^24,25 The genetic etiology of AD using genome-wide association study (GWAS) data also confirmed the involvement of immunity in AD.²⁶ These results suggest that abnormal immunity in the periphery is involved in the pathogenesis of AD, and blood gene expression analysis is useful both for identifying biomarkers and for studying the pathogenesis of AD.

In this study, blood transcriptome analyses of patients with AD, those with mild cognitive impairment (MCI) due to AD, and healthy controls were performed to elucidate immune-related pathophysiology.

Methods

Participants

The sample included 62 participants from a complete enumeration study of elderly people over 65 years of age in Nakayama town, Ehime prefecture, Japan (the 5th Nakayama Study, in which 927 community-dwelling older persons living in the area participated)^27–29 who were clinically diagnosed as (1) healthy controls (N = 21), (2) having MCI due to AD (N = 20), or (3) having AD (N = 21) (Table 1). There were 20 samples of patients diagnosed with MCI due to AD. Samples of 21 patients with age- and sex-matched AD and 21 age- and sex-matched participants with normal cognitive function were selected. All samples were selected from participants with few physical complications that could affect gene expression in the blood. All subjects or their families gave written, informed consent using a format approved by the ethics committee of Ehime University Graduate School of Medicine (Approval Number: 1610004). This study was conducted in accordance with the Declaration of Helsinki.

Table 1.

Participants’ characteristics.

	Control	MCI	AD	p
N	21	20	21
Age (SD) y	83.8 (4.6)	82.6 (4.9)	84.2 (4.6)	0.50
Male (%)	10 (47.6)	8 (40.0)	11 (52.4)	0.73
Height (SD) cm	150.0 (7.6)	150.7 (8.5)	146.7 (9.9)	0.28
Weight (SD) kg	53.2 (9.0)	52.7 (10.6)	51.3 (8.0)	0.79
BMI (SD) kg/m2	23.7 (3.8)	23.1 (3.3)	23.9 (3.0)	0.75
Education (N)
Finished ≤ primary school	1	1	5	0.07
Finished ≤ junior high school	10	14	13
Finished ≤ high school	8	5	3
Finished ≤ college, university	2	0	0
MMSE (SD)	29.1 (0.8)	25.5 (1.8)	19.0 (3.5)	<0.01
GDS15 (SD)	2.0 (2.1)	3.2 (2.6)	3.3 (3.2)	0.20
Medical history N (%)
Hypertension	15 (71.4)	13 (65.0)	12 (57.1)	0.63
Diabetes mellitus	3 (14.3)	2 (10.0)	4 (19.0)	0.71
Hyperlipidemia	10 (47.6)	5 (25.0)	4 (19.0)	0.11
Drinking	3 (14.3)	5 (25.0)	4 (19.0)	0.69
Smoking N (%)
Current	1 (4.8)	1 (5.0)	2 (9.5)	0.89
Never	15(71.4)	16 (80.0)	14 (66.7)
Past	5(23.8)	3 (15.0)	5(23.8)

BMI: body mass index; MMSE: Mini-Mental State Examination; GDS15: Geriatric Depression Scale 15.

Blood samples were generally taken 2 h after eating. A self-administered questionnaire was used to obtain data on education, tobacco and alcohol history, daily physical activity level, medical history, and use of medication. After data acquisition, the data were reviewed by trained interviewers. Height and weight were determined with the patients wearing light clothing, and the body mass index (BMI) was calculated from the measurements. The Mini-Mental State Examination (MMSE) and brain magnetic resonance imaging (MRI) were also conducted. All AD patients fulfilled the diagnostic criteria of the National Institute on Aging/Alzheimer's Association,³⁰ and a diagnosis of MCI due to AD was established for patients who satisfied the following criteria: (1) MMSE score ≥ 23 and normal general cognitive function; (2) Wechsler Memory Scale-Revised delayed recall test below the cut-off point for any cognitive impairment according to education status (≤8 points for 16 years of education, ≤4 points for 8–15 years, and ≤2 points for 0–7 years); (3) neuropsychiatric evaluation showing lack of dementia or depression as assessed by geriatric neuropsychiatrists according to the Diagnostic and Statistical Manual of Mental Disorders, 3rd edition, revised (DSM-III R) (American Psychiatric Association, 1997); (4) no disability in activities of daily living; and (5) atrophy of the hippocampus and parietal lobes without apparent cerebrovascular disease on brain MRI.³¹ A joint meeting of the JPSC-AD²⁷ research group was then held, and a final diagnosis was made.

RNA isolation and complementary DNA synthesis

Blood was taken into PaxGene Blood RNA Systems tubes (BD, Tokyo, Japan), and RNA was isolated according to the manufacturer's protocol. RNA concentration and quality were assessed with the NanoDrop 1000 system (Thermo Fisher Scientific, Yokohama, Japan). The RNA integrity number was also determined with the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) and Agilent RNA6000 Nano kit (Agilent Technologies). All samples fulfilled the following conditions: A260/A280 ≥ 1.8, A260/A230 ≥ 2.0, and RNA integrity number ≥ 7.0. With the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Santa Clara, CA, USA), 1.0 μg of RNA was used to synthesize 40-μL reaction mixtures of complementary DNA (cDNA).

RNA-Sequencing (transcriptome analysis)

RNA-Seq library preparation was performed using 145 ng of total RNA from each sample. An RNA-seq library of each sample was prepared using the QuantiSeq 3′mRNA-Seq Library Prep Kit for Illumina (LEXOGEN, Greenland, NH, USA) in combination with the Globin Block Modules for QuantSeq (LEXOGEN) according to the manufacturer's protocol. Then, 75-bp single-end sequencing was performed using the NextSeq 500 (Illumina, San Diego, CA, USA). Approximately 2.2 to 3.6 million reads were obtained for each sample. All data analysis was conducted using QIAGEN CLC Main Workbench 23.0 (QIAGEN, Germantown, MD, USA) with default parameters. Briefly, trimming of the adapter sequence, low-quality base of the 3′ end, polyA sequence, and the 5′ end first 12 bases derived from the random primer of fastq, was performed according to the library construction manual. Trimmed reads were aligned to the human reference data GRCh38, and significantly differentially expressed genes (DEGs) were detected based on Trimmed Mean of M-values (TMM) normalization and the negative binomial generalized linear model (GLM). A p-value of 0.05 was set as the threshold for DEGs, as no genes fulfilled p < 0.05 and FDR q < 0.05. All RNA-Seq data were deposited in the GEO database (accession number GSE 249477).

Gene ontology analysis and pathway enrichment analysis of the DEGs

Gene ontology (GO) analysis with default settings was performed to analyze the main functions of DEGs using DAVID tools (https://david.ncifcrf.gov/home.jsp). DAVID tools help researchers understand the biological implications behind many genes by delivering a comprehensive set of functional annotation tools.^32,33 Significant pathways and GO items, including biological processes (BPs), cellular components (CCs), molecular functions (MFs), and KEGG pathway (PW) were defined as pathways with p < 0.05.

Ingenuity pathway analysis

Ingenuity pathway analysis (IPA) with default settings was conducted to further explore relevant pathways associated with DEGs (QIAGEN) (https://www.qiagenbioinformatics.com/ products/ingenuity-pathway-analysis).³⁴ IPA is a web-based software application for the analysis, integration, and interpretation of data from RNA-Seq, microRNA, and SNP microarrays, metabolomics, proteomics, and gene and chemical listings, including small-scale experiments that generate downstream effects analyses to predict cellular functions, disease processes, and other phenotypes affected by patterns in the analyzed dataset. In addition, upstream regulator analyses were conducted to identify regulators (e.g., transcription factors, cytokines, kinases, etc.) that are directly related to targets in the analyzed data and whose activation or inhibition may explain the observed changes, with a positive Z-score meaning that the pathway was promoted and a negative Z-score meaning that the pathway was inhibited.

Quantitative polymerase chain reaction analysis

To validate microarray data, the expression levels of various mRNAs were determined by quantitative reverse transcription-polymerase chain reaction (qPCR) using the StepOnePlus Real-Time PCR System (Applied Biosystems). The mRNA expression levels were measured in duplicate using a StepOnePlus Real-Time PCR System with the PrimeTime Gene Expression Master Mix (Integrated DNA Technologies, Inc., Coralville, IA, USA) and mRNA-specific probes. The following were used as mRNA-specific probes: Hs.PT.58.4982621 for ITGB2, Hs.PT.58.20344216 for NFKB1, Hs.PT.58.25125759 for EGF, and Hs.PT.58.20622816 for CHRFAM7A. Hs.PT.39a.22214836 for GAPDH was used as an internal standard. RT-qPCR was conducted in 10-μL mixtures with 0.5 μL of cDNA. Following the manufacturer's protocol, the thermal cycling conditions were as follows: initial denaturation step (50°C for 2 min, and 95°C for 10 min), and 40 cycles of denaturation steps (95°C for 15 s, 60°C for 60 s annealing and elongation). The expression levels were calculated using the ΔΔCt method. To adjust for differences between plates, the same sample was placed in all plates as a calibration.

Statistical analyses

Continuous variables with a normal distribution are represented as mean ± SD. Normality was assessed by the Kolmogorov-Smirnov test. Equal variance was assessed using the F test. Mean differences among three groups were examined by one-way analysis of variance. In qPCR analysis, comparisons between two groups of continuous variables were made using the Student's t-tests. Categorical variables are presented as proportions (%). Comparisons between categorical variables were performed using the chi-squared test. Differences were considered significant if the two-tailed test yielded p < 0.05. All data were statistically analyzed using GraphPad Prism software (version 10.0.2; GraphPad Software, Boston, MA, USA).

Results

Characteristics of the participants

Characteristics of the participants including age, sex, height, weight, BMI, education, MMSE scores, Geriatric Depression Scale 15 (GDS15), medical history, and smoking status are presented in Table 1. The MMSE scores were significantly lower in AD (Control versus MCI versus AD: 29.1 versus 25.5 versus 19.0, p < 0.01). For the other characteristics, including medical history, there were no significant differences. There were no genes with p < 0.05 and FDR q < 0.05, indicating that gene expression in blood was not markedly different among AD, MCI, and Control participants. To ensure that the number of genes could withstand downstream analysis to detect slight changes, genes fulfilling only p < 0.05 were considered DEGs.

Transcriptome and DEG analysis

A total of 1091 genes were found to differentially distinguish patients with AD from Control, 938 genes were found to distinguish MCI from Control, and 788 genes were found to distinguish AD from MCI using p < 0.05 as the threshold for DEGs (Supplemental Table 1).

Gene ontology and pathway enrichment analyses (AD versus control)

DEGs were further analyzed using DAVID tools, which were also used to analyze BPs, CCs and MFs. As shown in Figure 1, GO analysis using upregulated DEGs identified several important BPs and PWs, including regulation of cholesterol transport and cholesterol metabolism. Significant GO downregulation in AD included steroid biosynthesis, fungicide, and Fc gamma R-mediated phagocytosis. Of these, GOs related to immunity and cholesterol metabolism were significantly accumulated in both upregulated and downregulated genes, indicating that the dysregulation of cholesterol metabolism and the immune system plays an important part in the process of AD.

Figure 1.

Gene ontology (GO) analysis (AD versus Control). The x-axis shows the –log10 p-value. GO and pathway analyses of upregulated DEGs (1A: Orange bars). GOs of regulation of cholesterol transport and cholesterol metabolism are significantly enriched. GO and pathway analysis of downregulated DEGs (1B: Blue bars). GOs of steroid biosynthesis, fungicide, and Fc gamma R-mediated phagocytosis are significantly enriched.

IPA analysis (AD versus control)

Bar graphs of the enrichment of canonical pathways of DEGs were plotted using the IPA tool. The y-axis represents the -log10 p-value of the significance of the enrichment of the IPA pathways on Fisher's exact test. Most pathways were positively regulated (Figure 2). FXR/RXR activation had the highest Z-score. Tumor microenvironment and neutrophil extracellular trap signaling had a positive Z-score. Top upstream regulators included ESR2, beta-estradiol, aflatoxin B1, CCN5, and trans-hydroxytamoxifen, and the causal network included BECN1, nastorazepide, PTPN13, CYP3A4, and PRMT5.

Figure 2.

IPA analysis of DEGs (AD versus control). The y-axis shows the –log10 p-value. The top 15 pathways are presented. FXR/RXR activation has the highest Z-score. Tumor microenvironment and neutrophil extracellular trap signaling have a highly positive Z-score.

Gene ontology and pathway enrichment analyses (MCI versus control)

GO analysis using upregulated DEGs showed several important BPs and MFs, as shown in Figure 3, including negative regulation of the execution phase of apoptosis, positive regulation of the JNK cascade, and RAGE receptor binding. Significant GO downregulation in MCI included receptor-mediated endocytosis, mitochondrial ATP synthesis-coupled proton transport, pathways of neurodegeneration - multiple diseases, AD, and oxidative phosphorylation. Of these, GOs associated with mitochondrial function and neurodegeneration, including AD, were significantly enriched in both up- and downregulated genes, suggesting that mitochondrial function, immune response, and neurodegeneration have important roles in the process of MCI.

Figure 3.

Go analysis (MCI versus control). The y-axis shows the –log10 p-value. GO and pathway analyses of upregulated DEGs (3A: Orange bars). GOs of negative regulation of the execution phase of apoptosis, positive regulation of the JNK cascade, and RAGE receptor binding are significantly enriched. GO and pathway analyses of downregulated DEGs (3B: Blue bars). GOs of the receptor-mediated endocytosis, mitochondrial ATP synthesis-coupled proton transport, pathways of neurodegeneration - multiple diseases, AD, and oxidative phosphorylation are significantly enriched.

IPA analysis (MCI versus control)

Mitochondrial dysfunction had the highest Z-score in MCI versus CT (Figure 4). The sirtuin pathway had a positive Z-score, and oxidative phosphorylation had a positive Z-score. Neutrophil extracellular trap signaling had a highly negative Z-score. Top upstream regulators included HNF4A, CD3, LARP1, RICTOR, and PRMT8, and the causal network included CMA1, miR-381-3p, ARG2, and mir-210.

Figure 4.

IPA analysis of DEGs (MCI versus control). The y-axis shows the –log10 p-value. The top 15 pathways are presented. The sirtuin pathway has a positive Z-score, and oxidative phosphorylation has a positive Z-score. Neutrophil extracellular trap signaling has a highly negative Z-score.

Gene ontology and pathway enrichment analyses (AD versus MCI)

As shown in Figure 5, GO analysis using upregulated DEGs showed several important BPs and PWs, including membrane disruption in other organisms, innate immune response in mucosa, viral myocarditis, and allograft rejection. Significant GO downregulation in AD included telencephalon development, malaria, and human papillomavirus infection. Of these, GOs associated with the immune response were significantly clustered in both upregulated and downregulated genes. This suggests that the immune response is important in the conversion from MCI to AD.

Figure 5.

Go analysis (AD versus MCI). The y-axis shows the –log10 p-value. GO and pathway analyses of upregulated DEGs (5A: Orange bars). GOs of membrane disruption in other organisms, innate immune response in mucosa, viral myocarditis, and allograft rejection are significantly enriched. GO and pathway analysis of downregulated DEGs (5B: Blue bars). GOs of telencephalon development, malaria, and human papillomavirus infection are significantly enriched.

IPA analysis (AD versus MCI)

Most pathways were negatively regulated (Figure 6). Triacylglycerol degradation had a negative Z-score, whereas PPARα/RXRα activation and neutrophil extracellular trap signaling had positive Z-scores. Top upstream regulators included tretinoin, MRTFB, TGFBR2, IL6, and filgrastim, and causal network included ZNF703, ERBB2, WNT3A, zidovudine, and GATA3.

Figure 6.

IPA analyses of DEGs (AD versus MCI). The y-axis shows the –log10 p-value. The top 15 pathways are presented. Triacylglycerol degradation has a negative Z-score, and PPARα/RXRα activation and neutrophil extracellular trap signaling have positive Z-scores.

Real-time PCR for the validation of DEGs in transcriptome analysis

To check the reliability of the transcriptome analysis, ITGB2 (neutrophil extracellular trap signaling pathway), NFKB1 (sirtuin signaling pathway and neutrophil extracellular trap signaling pathway), EGF (tumor microenvironment pathway), and CHRFAM7A (ID1 signaling pathway) were selected (Figure 7). ITGB2 expression was lower in MCI than in Control, but there were no significant differences (Control versus MCI = 0.94 ± 0.19 versus 0.82 ± 0.20, Student's t-test p = 0.06). NFKB1 expression was significantly lower in MCI than in Control (Control versus MCI = 0.83 ± 0.23 versus 0.68 ± 0.14, Student's t-test p = 0.02). EGF expression was significantly higher in AD than in Control (Control versus AD = 0.97 ± 0.48 versus 1.36 ± 0.67, Student's t-test p = 0.04). CHRFAM7A expression was lower in AD than in MCI, but there were no significant differences (MCI versus AD = 1.33 ± 0.77 versus 0.94 ± 0.48, Student's t-test p = 0.06).

Figure 7.

ITGB2, NFKB1, EGF, and CHRFAM7A expression levels. ITGB2 expression is lower in MCI (Control versus MCI = 0.94 ± 0.19 versus 0.82 ± 0.20, Student's t-test p = 0.06). NFKB1 expression is significantly lower in MCI (Control versus MCI = 0.83 ± 0.23 versus 0.68 ± 0.14, Student's t-test p = 0.02). EGF expression is significantly lower in AD (control versus AD = 1.36 ± 0.67 versus 0.97 ± 0.48, Student's t-test p = 0.04). CHRFAM7A expression is higher in AD (Control versus AD = 0.94 ± 0.48 versus 1.31 ± 0.76, Student's t-test p = 0.07).

Discussion

There are three findings in this study.

First, activation of FXR/RXR, tumor microenvironment, and neutrophil extracellular trap signaling was detected in AD. FXR/RXR are receptors involved in cholesterol regulation, and IPA analysis identified FXR/RXR activation in 5XFAD and Trem2 KO AD model mice, highlighting the significance of energy dysregulation and inflammatory processes in the progression of AD-related pathology.³⁵ Retinoids are thought to be involved in AD through their impact on oxidative stress (which may be related to mitochondrial dysfunction), amyloid-β deposits, inflammation, and neurotransmission.³⁶ Previous studies have shown that cholesterol and its derivatives affect T cell aging.³⁷ The tumor microenvironment consists of the cancer cells, the cytokine milieu, the extracellular matrix, subsets of immune cells, and other components. Not only do tumors manage to escape the host immune system, they also efficiently take advantage of invading cells by altering their function to create a microenvironment favorable to tumor progression.³⁸ The tumor microenvironment may be similar, with immune responses to pathogens such as amyloid plaques in AD.^39,40

Second, upregulation of the sirtuin pathway and downregulation of oxidative stress were detected in MCI. Sirtuins can alleviate neuroinflammation, whatever its cause, by inhibiting an intrinsic inflammatory pathway dependent on NF-κB,⁴¹ and they also regulate many AD-related events, including APP and tau processing, mitochondrial activity, oxidative damage, and neuroinflammation. Mitochondrial DNA (mtDNA) defects in AD can lead to energy dysfunction (1) caused by defective complexes I, III, IV, and V (2), promotion of amyloid-β deposition (3), and elevated oxidative stress (4), which in turn worsens mtDNA damage and increases the production of reactive oxygen species, creating a perpetual cycle of malfunctioning and injured mitochondria.⁴² Blood gene expression of people with MCI may reflect protective or resistant responses to the pathogenesis of AD via sirtuins and mitochondria.

Third, triacylglycerol degradation had a negative Z-score, whereas PPARα/RXRα activation and neutrophil extracellular trap signaling had a positive Z-score in the comparison of AD and MCI on IPA. GO analysis also showed that the immune response is altered in comparisons of AD and MCI. A higher triglyceride level was related to higher dementia risk in patients aged <60 years, but the inverse was observed for patients aged ≥60 years.⁴³ Current evidence suggests that the presence of increased blood total cholesterol (TC) or triglycerides (TG) is linearly related to an increase in the relative risk of AD, supporting the idea that high levels of TC and TG appear to have a causal effect in the pathogenesis of AD.⁴⁴ Triglycerides and fatty acids are important ingredients of the plaque environment to which T cells from the circulation are exposed, and there is growing evidence that fatty acids affect T cell activity.⁴⁵ In AD, neutrophils adhere to and spread within brain vasculature and are present in the parenchyma, as well as neutrophil extracellular traps.⁴⁶ In AD, neutrophil-vascular interactions cause myeloperoxidase accumulation in the brain.¹¹

This study may help to link several independent fields of research, such as diabetes, hypertension, obesity, lipid droplet formation, and age-related mitochondrial dysfunction, not only in the brain, but also in the blood.^9,19,47 Insulin resistance is involved in the progression of AD and is an important turning point in this disease.^28,48 For example, the SGLT2 inhibitor empagliflozin improves endothelial function and thereby reduces related pathologies.^49,50 In diabetes, glucose metabolism abnormalities and insulin resistance negatively affect cognitive function.⁵¹ Empagliflozin reduces oxidative stress and improves frailty in hypertensive and diabetic patients.⁵² Abnormal mitochondrial dynamics lead to neurodegeneration, but normalizing these dynamics may prevent its progression.⁵³ Metformin improves cognitive function in hyperglycemic patients and reduces oxidative stress in brain microvascular endothelial cells.⁵⁴ Mitochondrial quality control is crucial in neurodegenerative diseases, and mitophagy plays a key role in treatment.⁵⁵ Mitochondrial oxidative stress provides important insights into frail diabetic patients.^56,57 Muscle weakness and cognitive decline are associated with factors related to aging, such as mitochondrial dysfunction.⁵⁸ Mitophagy plays an important role in mitochondrial quality control, providing important insights into the mechanisms and potential treatments for neurodegenerative diseases.^59,60 Obesity is a risk factor for AD, and it destroys the microenvironment of adipose tissue, causing chronic inflammation and mitochondrial dysfunction, and inducing oxidative stress.⁶¹ Obesity contributes to cognitive decline through brain mitochondrial dysfunction and insulin resistance.⁶² There is a paradox regarding the role of adiponectin in AD, suggesting a relationship with amyloid formation.^63,64 A clinical trial has demonstrated that L-arginine supplementation, which may improve hypertension and endothelial function is effective in improving cognitive impairment in frail older adults with hypertension.⁶⁵

The strength of our research is that the clinical characteristics that are important in the pathology of dementia, such as the age and sex of the samples and physical complications such as diabetes, are neatly aligned between the three groups. In addition, since we are analyzing gene expression in the blood using PaxGene Blood RNA Systems tubes, it is easy to use as a diagnostic biomarker in clinical settings, and it can also detect abnormalities in the immune system. On the other hand, the present study had limitations. First, the sample size was relatively small, although age, sex, and rates of physical comorbidity were adequately controlled. Second, the blood samples were obtained from a cross-sectional survey, so no causal relationships can be inferred. The diagnosis of AD or MCI due to AD is clinical and does not involve confirmation of AD pathology through cerebrospinal fluid tests or amyloid PET scans. Not all MCI due to AD will develop into AD. Third, the limitations of generalizability should be pointed out, because this study intentionally selected an age- and sex-matched sample with relatively few physical complications from a large number of cohort samples. Fourth, it should be noted that no correction for multiple comparisons was made because no genes met the criteria for p < 0.05 and FDR q < 0.05, and that this study may be interpreted as a negative study and there may be many false positives. Gene expression in blood showed that there were no marked differences among AD, MCI, and Control, but even small changes may be important in the pathology. In the future, it will be necessary to observe longitudinal changes to clarify the relationships between the present results and the pathophysiology of AD and MCI.

In summary, neutrophil extracellular trap signaling was increased, and lipid metabolism (FXR/RXR activation, triacylglycerol degradation) was decreased in AD, whereas MCI showed protective responses via decreased neutrophil extracellular trap signaling and mitochondrial functions such as upregulation of the sirtuin pathway and downregulation of oxidative stress (Figure 8). The present study's results are consistent with previous reports that whole blood transcriptome analysis shows immune and inflammatory abnormalities and mitochondrial abnormalities, but the changes were small and, when corrected for multiple comparisons, were not significant (Supplemental Figure 1). These findings show that immune cells may play an important role in the development of AD, and that it is possible to monitor the immune system in MCI and AD through the blood transcriptome.

Figure 8.

A pictorial representation of the main results. In AD, neutrophil extracellular trap signaling is increased, and lipid metabolism is decreased, whereas in MCI, neutrophil extracellular trap signaling is decreased, and mitochondrial function is enhanced. Created with BioRender.com.

Supplemental Material

sj-docx-1-alr-10.1177_25424823241307878 - Supplemental material for Blood RNA transcripts show changes in inflammation and lipid metabolism in Alzheimer's disease and mitochondrial function in mild cognitive impairment

Supplemental material, sj-docx-1-alr-10.1177_25424823241307878 for Blood RNA transcripts show changes in inflammation and lipid metabolism in Alzheimer's disease and mitochondrial function in mild cognitive impairment by Jun-ichi Iga, Yuta Yoshino, Tomoki Ozaki, Ayumi Tachibana, Hiroshi Kumon, Yu Funahashi, Hiroaki Mori, Mariko Ueno, Yuki Ozaki, Kiyohiro Yamazaki, Shinichiro Ochi, Masakatsu Yamashita and Shu-ichi Ueno in Journal of Alzheimer's Disease Reports

Supplemental Material

sj-xlsx-2-alr-10.1177_25424823241307878 - Supplemental material for Blood RNA transcripts show changes in inflammation and lipid metabolism in Alzheimer's disease and mitochondrial function in mild cognitive impairment

Supplemental material, sj-xlsx-2-alr-10.1177_25424823241307878 for Blood RNA transcripts show changes in inflammation and lipid metabolism in Alzheimer's disease and mitochondrial function in mild cognitive impairment by Jun-ichi Iga, Yuta Yoshino, Tomoki Ozaki, Ayumi Tachibana, Hiroshi Kumon, Yu Funahashi, Hiroaki Mori, Mariko Ueno, Yuki Ozaki, Kiyohiro Yamazaki, Shinichiro Ochi, Masakatsu Yamashita and Shu-ichi Ueno in Journal of Alzheimer's Disease Reports

Footnotes

Acknowledgments

The authors would like to thank Ms. Chiemi Onishi for her technical assistance. This study was assisted by the Division of Laboratory Animal Research and the Division of Analytical Bio-Medicine at the Advanced Research Support Center (ADRES), Ehime University. The authors also thank Dr Naohito Tokunaga for technical advice on RNA-sequencing.

ORCID iDs

Jun-ichi Iga

Yuta Yoshino

Ayumi Tachibana

Kiyohiro Yamazaki

Shinichiro Ochi

Masakatsu Yamashita

Shu-ichi Ueno

Author contributions

Jun-ichi Iga (Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Software); Yuta Yoshino (Conceptualization; Resources; Supervision; Writing – review & editing); Tomoki Ozaki (Investigation; Resources); Ayumi Tachibana (Data curation; Investigation; Resources); Hiroshi Kumon (Investigation; Resources); Yu Funahashi (Data curation; Investigation); Hiroaki Mori (Data curation; Investigation); Mariko Ueno (Data curation; Investigation); Yuki Ozaki (Resources; Writing – review & editing); Kiyohiro Yamazaki (Resources; Writing – review & editing); Shinichiro Ochi (Resources; Writing – review & editing); Masakatsu Yamashita (Supervision; Writing – review & editing); Shu-ichi Ueno (Resources; Supervision; Writing – review & editing).

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Japan Agency for Medical Research and Development (JP22dk0207053) and JSPS KAKENHI Grant Numbers 20K07971, 22K07597, 22K07562, 22K15752, 22K15765 and 23K07016.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

All of the data that were generated or analyzed during this study are included in this published article and in the supplementary information files. All RNAseq data were deposited in the GEO database (accession number GSE249477).

Supplemental material

Supplemental material for this article is available online.

References

Frisoni

Altomare

Ribaldi

, et al. Dementia prevention in memory clinics: recommendations from the European task force for brain health services. Lancet Reg Health Eur 2023; 26: 100576.

Hameed

Fuh

Senanarong

, et al. Role of fluid biomarkers and PET imaging in early diagnosis and its clinical implication in the management of Alzheimer's disease. J Alzheimers Dis Rep 2020; 4: 21–37.

Hampel

Cummings

, et al. Blood-based biomarkers for Alzheimer's disease: current state and future use in a transformed global healthcare landscape. Neuron 2023; 111: 2781–2799.

Mondal

Noreen

Loffredo

, et al. Transcriptomic analysis of Alzheimer's disease pathways in a Pakistani population. J Alzheimers Dis Rep 2024; 8: 479–493.

Griswold

Sivasankaran

Van Booven

, et al. Immune and inflammatory pathways implicated by whole blood transcriptomic analysis in a diverse ancestry Alzheimer's disease cohort. J Alzheimers Dis 2020; 76: 1047–1060.

Huseby

Delvaux

Brokaw

, et al. Blood transcript biomarkers selected by machine learning algorithm classify neurodegenerative diseases including Alzheimer's disease. Biomolecules 2022; 12: 1592.

Milanesi

Dobre

Cucos

, et al. Whole blood expression pattern of inflammation and redox genes in mild Alzheimer's disease. J Inflamm Res 2021; 14: 6085–6102.

Chang

Chen

, et al. Advances in the study of APOE and innate immunity in Alzheimer's disease. J Alzheimers Dis 2023; 93: 1195–1210.

Ochi

Iga

Funahashi

, et al. Identifying blood transcriptome biomarkers of Alzheimer's disease using transgenic mice. Mol Neurobiol 2020; 57: 4941–4951.

10.

Chou

Babamale

Walker

, et al. Blood-brain crosstalk: the roles of neutrophils, platelets, and neutrophil extracellular traps in neuropathologies. Trends Neurosci 2023; 46: 764–779.

11.

Smyth

LCD

Murray

Hill

, et al. Neutrophil-vascular interactions drive myeloperoxidase accumulation in the brain in Alzheimer's disease. Acta Neuropathol Commun 2022; 10: 38.

12.

Liston

Yshii

. T cells drive aging of the brain. Nat Immunol 2023; 24: 12–13.

13.

Tachibana

Iga

Ozaki

, et al. Serum high-sensitivity C-reactive protein and dementia in a community-dwelling Japanese older population (JPSC-AD). Sci Rep 2024; 14: 7374.

14.

Krance

Zou

, et al. The complement cascade in Alzheimer's disease: a systematic review and meta-analysis. Mol Psychiatry 2021; 26: 5532–5541.

15.

Tachibana

Iga

Tatewaki

, et al. Late-life high blood pressure and enlarged perivascular spaces in the putaminal regions of community-dwelling Japanese older persons. J Geriatr Psychiatry Neurol 2024; 37: 61–72.

16.

Gonzales

Garbarino

Kautz

, et al. Senolytic therapy in mild Alzheimer's disease: a phase 1 feasibility trial. Nat Med 2023; 29: 2481–2488.

17.

Tseng

Zeng

Suen

, et al. Efficacy and acceptability of anti-inflammatory eicosapentaenoic acid for cognitive function in Alzheimer's dementia: a network meta-analysis of randomized, placebo-controlled trials with omega-3 fatty acids and FDA-approved pharmacotherapy. Brain Behav Immun 2023; 111: 352–364.

18.

Ueno

Yoshino

Mori

, et al. Association study and meta-analysis of polymorphisms and blood mRNA expression of the ALDH2 gene in patients with Alzheimer's disease. J Alzheimers Dis 2022; 87: 863–871.

19.

Mori

Funahashi

Yoshino

, et al. Blood CDKN2A gene expression in aging and neurodegenerative diseases. J Alzheimers Dis 2021; 82: 1737–1744.

20.

Kumon

Yoshino

Funahashi

, et al. PICALM mRNA expression in the blood of patients with neurodegenerative diseases and geriatric depression. J Alzheimers Dis 2021; 79: 1055–1062.

21.

Sao

Yoshino

Yamazaki

, et al. TREM1 mRNA expression in leukocytes and cognitive function in Japanese patients with Alzheimer's disease. J Alzheimers Dis 2018; 64: 1275–1284.

22.

Sao

Yoshino

Yamazaki

, et al. MEF2C mRNA expression and cognitive function in Japanese patients with Alzheimer's disease. Psychiatry Clin Neurosci 2018; 72: 160–167.

23.

Yamazaki

Yoshino

Mori

, et al. Gene expression and methylation analysis of ABCA7 in patients with Alzheimer's disease. J Alzheimers Dis 2017; 57: 171–181.

24.

Ciryam

Kundra

Freer

, et al. A transcriptional signature of Alzheimer's disease is associated with a metastable subproteome at risk for aggregation. Proc Natl Acad Sci U S A 2016; 113: 4753–4758.

25.

Long

, et al. Integrated genomic approaches identify major pathways and upstream regulators in late onset Alzheimer's disease. Sci Rep 2015; 5: 12393.

26.

Bellenguez

Kucukali

Jansen

, et al. New insights into the genetic etiology of Alzheimer's disease and related dementias. Nat Genet 2022; 54: 412–436.

27.

Ninomiya

Nakaji

Maeda

, et al. Study design and baseline characteristics of a population-based prospective cohort study of dementia in Japan: the Japan Prospective Studies Collaboration for Aging and Dementia (JPSC-AD). Environ Health Prev Med 2020; 25: 64.

28.

Yoshida

Mori

Shimizu

, et al. Analysis of factors related to cognitive impairment in a community-based, complete enumeration survey in Japan: the Nakayama study. Psychogeriatrics 2023; 23: 876–884.

29.

Shimizu

Mori

Yoshida

, et al. Secular trends in the prevalence of dementia based on a community-based complete enumeration in Japan: the Nakayama Study. Psychogeriatrics 2022; 22: 631–641.

30.

McKhann

Knopman

Chertkow

, et al. The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011; 7: 263–269.

31.

Albert

DeKosky

Dickson

, et al. The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011; 7: 270–279.

32.

Sherman

Hao

Qiu

, et al. DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res 2022; 50: W216–W221.

33.

Huang da

Sherman

Lempicki

. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 2009; 4: 44–57.

34.

Kramer

Green

Pollard

Jr. , et al. Causal analysis approaches in ingenuity pathway analysis. Bioinformatics 2014; 30: 523–530.

35.

Weller

Ferraro

Doyle

, et al. Single nucleus transcriptome data from Alzheimer's disease mouse models yield new insight into pathophysiology. J Alzheimers Dis 2022; 90: 1233–1247.

36.

Sodhi

Singh

. Retinoids as potential targets for Alzheimer's disease. Pharmacol Biochem Behav 2014; 120: 117–123.

37.

Zhu

, et al. Cholesterol metabolism in T-cell aging: accomplices or victims. FASEB J 2023; 37: e23136.

38.

Anderson

Simon

. The tumor microenvironment. Curr Biol 2020; 30: R921–R925.

39.

Miao

Zhuo

Han

, et al. From degenerative disease to malignant tumors: insight to the function of ApoE. Biomed Pharmacother 2023; 158: 114127.

40.

Wang

Chu

Cao

, et al. The emerging role of triggering receptor expressed on myeloid cell-2 in malignant tumor. Cent Eur J Immunol 2022; 47: 373–381.

41.

Yin

Song

, et al. Adsorption of Cd(II) from aqueous solution by Pennisetum sp. straw biochars derived from different modification methods. Environ Sci Pollut Res Int 2019; 26: 7024–7032.

42.

Bazzani

Equisoain Redin

McHale

, et al. Mitochondrial DNA repair in neurodegenerative diseases and ageing. Int J Mol Sci 2022; 23: 11391.

43.

Gong

Harris

Peters

SAE

, et al. Serum lipid traits and the risk of dementia: a cohort study of 254,575 women and 214,891 men in the UK Biobank. EClinicalMedicine 2022; 54: 101695.

44.

Zhu

Liu

Zhu

, et al. Lipid levels and the risk of dementia: a dose-response meta-analysis of prospective cohort studies. Ann Clin Transl Neurol 2022; 9: 296–311.

45.

Reilly

Lutgens

Kuiper

, et al. Effects of fatty acids on T cell function: role in atherosclerosis. Nat Rev Cardiol 2021; 18: 824–837.

46.

Zenaro

Pietronigro

Della Bianca

, et al. Neutrophils promote Alzheimer's disease-like pathology and cognitive decline via LFA-1 integrin. Nat Med 2015; 21: 880–886.

47.

Byrns

Perlegos

Miller

, et al. Senescent glia link mitochondrial dysfunction and lipid accumulation. Nature 2024; 630: 475–483.

48.

Noguchi-Shinohara

Yuki-Nozaki

Abe

, et al. Diabetes mellitus, elevated hemoglobin A1c, and glycated albumin are associated with the presence of all-cause dementia and Alzheimer's disease: the JPSC-AD Study. J Alzheimers Dis 2022; 85: 235–247.

49.

Hierro-Bujalance

Garcia-Alloza

. Empagliflozin reduces brain pathology in Alzheimer's disease and type 2 diabetes. Neural Regen Res 2024; 19: 1189–1190.

50.

Hierro-Bujalance

Infante-Garcia

Del Marco

, et al. Empagliflozin reduces vascular damage and cognitive impairment in a mixed murine model of Alzheimer's disease and type 2 diabetes. Alzheimers Res Ther 2020; 12: 40.

51.

Zhang

Wen

, et al. Cognitive dysfunction in diabetes: abnormal glucose metabolic regulation in the brain. Front Endocrinol (Lausanne) 2023; 14: 1192602.

52.

Mone

Varzideh

Jankauskas

, et al. SGLT2 Inhibition via empagliflozin improves endothelial function and reduces mitochondrial oxidative stress: insights from frail hypertensive and diabetic patients. Hypertension 2022; 79: 1633–1643.

53.

Ishikawa

Yamamoto

Hattori

, et al. Neuronal degeneration and cognitive impairment can be prevented via the normalization of mitochondrial dynamics. Pharmacol Res 2021; 163: 105246.

54.

Mone

Gambardella

Pansini

, et al. Cognitive impairment in frail hypertensive elderly patients: role of hyperglycemia. Cells 2021; 10: 2115.

55.

Wang

Feng

Wang

, et al. Role of mitophagy in mitochondrial quality control: mechanisms and potential implications for neurodegenerative diseases. Pharmacol Res 2021; 165: 105433.

56.

Lynn

Park

Ogunwale

, et al. A tale of two diseases: exploring mechanisms linking diabetes mellitus with Alzheimer's disease. J Alzheimers Dis 2022; 85: 485–501.

57.

de la Monte

. Malignant brain aging: the formidable link between dysregulated signaling through mechanistic target of rapamycin pathways and Alzheimer's disease (type 3 diabetes). J Alzheimers Dis 2023; 95: 1301–1337.

58.

Arosio

Calvani

Ferri

, et al. Sarcopenia and cognitive decline in older adults: targeting the muscle-brain axis. Nutrients 2023; 15: 1853.

59.

Pluta

Bogucka-Kocka

Bogucki

, et al. Apoptosis, autophagy, and mitophagy genes in the CA3 area in an ischemic model of Alzheimer's disease with 2-year survival. J Alzheimers Dis 2024; 99: 1375–1383.

60.

Sharma

Banerjee

Davis

. Early mitochondrial defects in the 5xFAD mouse model of Alzheimer's disease. J Alzheimers Dis 2023; 91: 1323–1338.

61.

Ren

, et al. Oxidative stress: the nexus of obesity and cognitive dysfunction in diabetes. Front Endocrinol (Lausanne) 2023; 14: 1134025.

62.

Sripetchwandee

Chattipakorn

. Links between obesity-induced brain insulin resistance, brain mitochondrial dysfunction, and dementia. Front Endocrinol (Lausanne) 2018; 9: 496.

63.

Yan

Yin

, et al. Adiponectin ameliorates cognitive behaviors and in vivo synaptic plasticity impairments in 3xTg-AD mice. J Alzheimers Dis 2022; 85: 343–357.

64.

Hashimoto

Sugama

, et al. Possible role of activin in the adiponectin paradox-induced progress of Alzheimer's disease. J Alzheimers Dis 2021; 81: 451–458.

65.

Mone

Pansini

Jankauskas

, et al. L-arginine improves cognitive impairment in hypertensive frail older adults. Front Cardiovasc Med 2022; 9: 868521.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.37 MB

0.09 MB