Circulating micro-RNAs Differentially Expressed in Korean Alzheimer’s Patients with Brain Aβ Accumulation Activate Amyloidogenesis

doi:10.21203/rs.3.rs-944590/v1

Download PDF

Research

Circulating micro-RNAs Differentially Expressed in Korean Alzheimer’s Patients with Brain Aβ Accumulation Activate Amyloidogenesis

https://doi.org/10.21203/rs.3.rs-944590/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

BACKGROUND

A role for extracellular vesicles (EVs) enriched with micro-RNAs (miRNAs) has been proposed in Alzheimer’s disease (AD) pathogenesis, leading to the discovery of blood miRNAs as biomarkers of AD. However, the diagnostic utility of specific miRNAs is not consistent. This study aimed to discover blood miRNAs that are differentially expressed in Korean AD patients, evaluate their clinical performance in plasma or plasma EVs, and investigate their role in amyloidogenesis.

METHODS

Blood from 15 (7 cognitively normal [CN] and 8 AD) out of 262 subjects (59 CN, 105 mild cognitive impairment [MCI], 98 AD) and 8 Parkinson’s disease (PD) patients was used to discover miRNAs differentially expressed in AD. We evaluated the clinical performance of these miRNAs in plasma of a subgroup of 100 subjects (51 CN, 22 MCI, 27 AD) and in plasma EVs isolated from the total cohort in a cross-sectional design. The effects of miRNAs on amyloid b (Ab) production and expression of their target genes were investigated in neuronal culture systems.

RESULTS

Among 17 upregulated, and one downregulated miRNAs in AD (>2-fold), three upregulated miRNAs (miR-122-5p, miR-210-3p, and miR-590-5p) that were differentially expressed in AD compared with CN or PD were selected. Diagnostic utility for discrimination of AD or MCI from CN of the selected miRNAs in plasma or plasma EV was not high. Nevertheless, the levels of three miRNAs in plasma or plasma EVs of subjects who were Ab positive on positron emission tomography (PET) were significantly higher than those from subjects who were Ab-PET negative. Furthermore, the selected miRNAs induced Ab production through activation of b-cleavage of amyloid precursor protein and downregulated their target genes. Pathway enrichment and protein interaction network analysis of target genes of the miRNAs further supported the roles of the selected miRNAs in amyloidogenesis.

CONCLUSION

The diagnostic utility of circulating miR-122-5p, miR-210-3p, and miR-590-5p to discriminate AD from CN was modest. However, these miRNAs were highly expressed in patients with amyloid accumulation, which was supported by in vitro analysis of amyloidogenesis. Our results suggest that blood-based miRNA biomarkers may play a role in amyloidogenesis during AD onset and progression.

Neurology

Cellular & Molecular Neuroscience

Cognitive Neuroscience

Micro-RNA

Alzheimer’s disease

Blood Biomarkers

Beta-amyloid

Amyloidogenesis

Alzheimer’s disease (AD) is the most prevalent type of dementia globally and is clinically diagnosed by progressive memory loss and cognitive impairment. Recently, new guidelines have been developed that aim to improve diagnosis, strengthen autopsy reporting, and establish a research agenda for future progress in early AD detection and greater diagnostic accuracy. These guidelines incorporate imaging and cerebrospinal fluid (CSF) biomarkers in the definition of preclinical AD, mild cognitive impairment (MCI) due to AD, and dementia due to AD [1-3]. Given that core AD CSF biomarkers, amyloid β (Aβ) 42, total tau and tau phosphorylated at Thr181, are useful to increase diagnostic accuracy, their use in clinical practice and research has been recommended in the guidelines [4]. However, CSF analysis using immunoassay platforms requires an invasive lumbar puncture CSF collection procedure and raises issues of preanalytical and analytical variability in immunoassay techniques, both within and among laboratories. Consequently, the development of accessible blood-based analytes is a highly attractive alternative for reliable AD biomarkers.

Micro-RNAs (miRNAs) are short noncoding single-stranded RNAs of approximately 22–24 nucleotides in length that play diverse roles in biologic and pathologic processes by regulating the expression of their target genes. Various miRNAs detected in human biologic fluid or tissues of animal models have been proposed to participate in AD pathogenesis [5, 6]. Although the species or type and concentration of miRNA in the peripheral circulation closely reflect cognitive status, imaging biomarkers, CSF biomarkers, or actual clinical diagnosis are not consistent. Nevertheless, systematic reviews suggest that circulating miRNAs have a certain diagnostic value for the detection of cognitive impairment, or diagnosis of AD with high sensitivity and specificity [7-9]. Circulating miRNAs are enriched in small-sized extracellular vesicles (EV) such as exosomes and therefore, EV may be a reliable source of miRNA for biomarkers of AD [10]. Specific miRNAs involved in a disease are actively encapsulated into exosomes and can be efficiently taken up by the recipient target cells. Therefore, circulating exosomal miRNAs may be used as diagnostic and prognostic biomarkers of AD. In addition, the profile of exosomal miRNAs as compared with that of plasma miRNAs may give us an insight into the role of exosomes in the pathogenesis or progression of AD. However, there are a couple of issues that need to be overcome before obtaining solid and reliable expression profiles of plasma miRNAs in AD patients. First, a major issue of miRNA library construction occurs if the amount of input RNA is low. For example, if the total RNA is less than 200 ng, the 5′ and 3′ adapters can form dimers that compete in the RT-PCR reaction with the miRNA inserts and do not synthesize the expected product [11]. Second, RNA ligase enzymes have been known to cause a high level of sequence-dependent bias in the miRNA-adaptor-ligation process [12]. Third, the amount of total RNA from plasma is not sufficient to generate a miRNA library efficiently. In general, the construction of an unbiased miRNA library from plasma is a challenging task. Fourth, because of the complexity of miRNA library construction, it is also essential to mitigate batch effects. Batch effects arise from variability in individual sample processing, such as reaction conditions, pipetting accuracy, and even differences in technical proficiency [13]. Because of these issues, miRNA expression profiling in AD patients is highly variable from one study to another, save for a few common miRNAs. Furthermore, it is necessary to increase the diagnostic accuracy of miRNAs in plasma or in circulating exosomes in order to differentiate AD from non-AD dementia as well as from healthy older persons. Currently, such studies are limited.

Against this background, our aim in this study was to identify circulating miRNAs that are differentially expressed in Korean AD patients as compared with cognitively normal older people and to evaluate the clinical potential of miRNAs in plasma or circulating exosomes to differentiate AD or MCI patients from CN. In addition, we evaluated the function of discovered miRNAs in amyloidogenesis in vitro. Finally, we confirmed the effect of miRNAs on target gene regulation in vitro and analyzed the enriched pathway of the miRNA target genes.

Study population

Among 262 participants from three memory clinics (Inha University Hospital and Yonsei University Severance Hospital) across South Korea who agreed to provide blood samples, 59 subjects were cognitively normal (CN), 105 subjects were diagnosed as MCI, and 98 subjects were diagnosed as probable AD, according to the criteria for clinical diagnosis as previously described [14]. Plasma samples from 15 subjects (7 CN and 8 AD) out of 262 participants and an additional 8 patients with Parkinson’s disease (PD) were used to discover differentially expressed miRNAs able to discriminate AD patients from healthy controls or PD. To test the diagnostic utility of discovered plasma miRNAs, samples from 100 subjects (51 CN, 22 MCI, and 27 AD) recruited from Inha University Hospital were used for comparing the levels of differentially expressed plasma miRNAs. For comparing levels of the differentially expressed miRNAs in plasma EVs, we added 162 plasma samples from other independent cohorts from Yonsei University Hospital and Inha University Hospital.

Participants underwent amyloid imaging via ¹⁸F-flutemetamol positron emission tomography (PET). Amyloid PET positivity was determined independently by clinical neurologists. The Institutional Review Board and Ethical Committee of Inha University Hospital approved the study (approval No.: INHAUH2016-06-010-002). Written informed consent to participate was obtained from all volunteers. The demographic characteristics and clinical information of the subjects are summarized in Table 1.

Table 1

Demographic and clinical information of 262 participants.
Diagnosis	n (M:F)	Aβ-PET (+/-, n)^a	Age, yr (mean±S.D.)	Education, yr (mean±S.D.)	MMSE (mean±S.D.)
CN	59 (18:41)	4/49	68±6.5	9.6±4.3	27.5±2.1
MCI	105 (37:68)	43/62	73±6.8^***	8.3±4.9	24.3±3.3^***
AD	98 (25:73)	71/22	75±8.5^***	8.1±4.7	16.7±5.9^***,###
P value	n.s. ^b	<0.001 ^b	<0.001	n.s.	<0.001
Total	262 (80:182)	118/133	73±7.9	8.5±4.8	22.1±6.2
^a6 of CN and 5 of AD did not undergo Aβ-PET analysis. ^bChi-square test, ^***p<0.001 vs. CN, ^###p<0.001 vs. MCI by Dunn’s multiple comparison. n.s., not significant.

Blood collection and plasma preparation

To exclude the contribution of anticoagulants to the pre-analytical variability in the EV yield [15] or the subsequent quantitative polymerase chain reaction (qPCR) [16], we selected the most commonly used anticoagulant, ethylenediaminetetraacetic acid (EDTA) to prepare the plasma. All participants provided their fasting blood, which was drawn from the antecubital vein into EDTA tubes (BD Vacutainer K2E-EDTA; Beckton and Dickinson, Franklin Lake, NJ, USA). Plasma was prepared by immediate centrifugation at 3,000 ⋅ g for 10 min at room temperature (RT). To remove platelets, cell debris, and large particles, the plasma was centrifuged again at 10,000 ⋅ g for 20 min at RT. The supernatant was stored at –70 °C until EV preparation.

Extraction of plasma small RNAs

We collected 1.5 mL blood in PAXgene Blood RNA tubes (PreAnalytiX) from patients with AD (n = 8), patients with PD (n = 8), and from CN (n = 7). After centrifugation at 15,000 rpm, we collected 0.5 mL plasma and added 0.5 mL of extraction buffer. We added 0.2 mL of chloroform to the mixture of plasma and extraction buffer and incubated for 5 min. After centrifugation for 10 min at 15,000 rpm, we transferred the supernatant to a new tube and added 1 ⋅ volume of isopropanol. By filtering the mixed solution through a MinElute column (Qiagen, Venlo, Netherlands), we collected RNAs that were washed with 0.5 mL of Qiagen washing buffer (RPE buffer), and dissolved in DEPC-treated water.

Plasma miRNAome analysis

For small RNA sequencing, the quality of the constructed library was checked using an Agilent Bioanalyzer (Agilent, Santa Clara, CA, USA) and sequenced on the Illumina Nextseq 500 (Illumina, San Diego, CA, USA) platform. The quality and band size of the libraries were assessed using an Agilent 2100 Bioanalyzer (Agilent). Libraries were quantified by qPCR using the CFX96 Real-Time System (Bio-Rad, Hercules, CA, USA). After normalization, library sequencing was conducted on the Nextseq 500 system (Illumina) with 75 bp single end reads. We obtained 41, 36, and 35 million clean reads on average for patients with AD, patients with PD, and from CN, respectively. Potentially existing sequencing adapters and raw quality bases in the low reads were trimmed using Skewer and FastaQC. The cleaned high-quality reads, after trimming the low-quality bases and sequencing adapters were mapped to the reference genome using QuickMIRseq. To quantify the mapped reads on the reference genome into the miRNA expression values, QuickMIRseq with default options was used. The miRNA and hairpin annotation of the reference genome from miRBase was used and the expression values were calculated in Reads Per Ten Million (RPTM) units. The differentially expressed miRNAs between the two selected biological conditions were analyzed by NOISeq in Bioconductor. The expression levels of miRNAs were calculated by normalizing the miRNA counts with the total number of clean reads in the small RNA libraries.

EV isolation, extraction of RNAs and amplification of specific EV miRNAs

For isolation of EVs in plasma, we modified the manufacturer’s provided protocol, as previously described [17]. Briefly, we used a commercially available PP-based kit (miRCURY Exosome Plasma Kit, #76603, Qiagen, Venlo, Netherlands) with pre-incubation of plasma with thrombin (#36402.01, SERVA, Heidelberg, Germany) and proteinase K (0.5 mg/mL; Ambion, Austin, TX, USA). We used the serum-like supernatant from plasma samples (0.6 mL) pre-incubated with thrombin and centrifuged for defibrinating. Then, the supernatant was mixed with proteinase K and incubated at 37 °C for 10 min, followed by incubation with precipitation buffer at 4 °C for 1 h. RNA pellets were obtained following centrifugation at 1,000 ⋅ g for 5 min. Subsequently, we isolated total RNA using a commercial RNA extraction kit (miRNeasy Plasma Advanced Kit, #217204, Qiagen) following the manufacturer’s instructions. EV pellets were lysed with 0.5 mL RNase-free water and the lysed EVs were mixed with lysis RPL buffer (150 µL) including 0.625 ng synthetic RNA (#339390, Qiagen), followed by precipitation in 50 µL of RPP buffer and centrifugation at 12,000 ⋅ g for 3 min at RT. The supernatants were transferred to new tubes and mixed with an equal volume of isopropanol. The mixtures were eluted through a miRNA spin column (RNeasy UCP MinElute column, Qiagen) after washing with 20 µL of RNase-free water. Using 2 µL of the extracted RNA, cDNA was synthesized with a miRCURY LNA RT Kit (#339340, Qiagen) following the manufacturer’s instructions. cDNA was processed through a reverse transcription step at 42 °C for 60 min and an inactivation reaction step at 95 °C for 5 min on a PCR Thermal Cycler (TP600, TAKARA BIO INC., Japan). After reverse transcription, specific miRNAs were amplified using cDNA with 10-fold dilution. We used primer sets for miRNAs from Qiagen, and carried out qPCR under conditions of the miRCURY LNA SYBR Green PCR kit (#339345, Qiagen) on a CFX Connect Real-Time PCR Detection System (#1855201, Bio-Rad, Hercules, CA, USA). The PCR cycling conditions consisted of PCR initial heat activation at 95 °C for 2 min and 45 cycles of 2-step cycling with denaturation at 95 °C for 10 s and combined annealing/extension at 56 °C for 60 s, followed by melting curve analysis from 60 °C to 95 °C. The quantitative real-time PCR (qRT-PCR) data were normalized to UniSp2 as an exogenous reference gene. The primer sequences are summarized in

miRNA target prediction and pathway enrichment analysis

Using the results from differential miRNA expression data, gene–miRNA interaction was predicted using miRDB, which is an online database for miRNA target prediction and functional annotations (version 6.0) [18]. Candidate transcripts with scores ≥ 80 are presented as predicted targets for further analysis. Pathway enrichment analysis was employed to investigate the function of significantly differentially expressed miRNAs. Metascape (http://metascape.org) was used to carry out gene ontology (GO) term analysis for enriched pathways, providing gene list annotation and analysis source [19]. Pairwise similarities between any two enriched terms are computed based on a Kappa-test score (threshold of a 0.3 similarity). In addition, to explore further the relationship between differentially expressed genes (DEGs) at the protein level, we carried out protein–protein interaction (PPI) network analysis using the following databases: STRING, BioGrid, OmniPath, and InWeb IM. If the network contains between 3 and 1000 proteins with physical interaction, the plug-ins Molecular Complex Detection (MCODE) algorithm was applied to identify densely connected network components.

Cell culture and transfection

H4 cells overexpressing the Swedish FAD mutant APP695 (H4-APPswe) were grown in Dulbecco’s modified Eagle’s medium (Invitrogen) with 10% fetal bovine serum (FBS), 25 mM glucose, 400 µg/mL G418, 100 U/mL penicillin, and 100 µg/mL streptomycin at 37 °C in a humidified atmosphere of 95% air and 5% CO₂. For transfection of miRNA, 400,000 cells per well were plated in 6-well plates. The next day, cells were transfected with 100 nM of pre-miRNAs (BIONEER, Daejeon, Korea) or scrambled sequence using Lipofectamine RNAiMAX (Invitrogen, MA, USA) following the manufacturer’s instructions. The transfected cells were grown in complete medium for 24 h and refed with serum-free medium for an additional 24 h. Harvested cells and culture media (CM) were used for immunoblots, qRT-PCR, or ELISA.

Measurement of Aβ in culture media

We collected CM of H4-APPswe cells at 24 h after miRNA transfection. Media were centrifuged to remove particulates and stored at –80 °C until use. The quantity of Aβ40 and Aβ42 was measured using commercially available ELISA kits (Invitrogen, MA, USA) according to the manufacturer’s instructions. The CM was diluted appropriately and transferred into the supplied microplate pre-coated with the target-specific capture antibody. Next, the sandwich was formed by incubating with a detection antibody for 3 h at RT. Enzyme labeled antibody was added and reacted with substrate solution. Absorbance was read at 450 nm.

Amplification of targeted mRNAs

For RNA extraction, whole cells were collected using the Trizol reagent (Ambion, Carlsbad, CA, USA), and subjected to RNA isolation as previously described [20]. Total RNA quality and quantity were measured using a NanoDrop 1000 (Thermo Fisher Scientific, MA, USA), before 20 µL reactions containing RNA template (1 µg) were converted to cDNA using the PrimeScript™ 1st strand cDNA Synthesis Kit (Takara, Japan). The mRNA expression levels of three miRNA target genes (ADAM10, BDNF, and JAG1) were estimated via qRT-PCR. Amplification reactions contained 10 µL of 2X SYBR supermix (#170-8882, Bio-Rad), each forward and reverse primer (1 pM), 4 µL of nuclease-free water, and 2 µL of cDNA. The qRT-PCR was carried out under the following conditions: Initial step at 95 °C for 10 min; 40 cycles of 95 °C 15 s; 58 °C 25 s, and 72 °C 35 s, followed by melting curve analysis from 60 °C to 95 °C. Primers were designed using the NCBI primer blast tool and synthesized by BIONIC INC. (Seoul, ROK). The primer sequences are summarized in Supplementary Table 2. The relative expression level was calculated using the 2^−ΔΔCt method (ΔCt = Ct_{Target gene} – Ct_GAPDH; ΔΔCt = ΔCt_{Target miRNA} – ΔCt_{scrambled miRNA}), after normalization with the glyceraldehyde-3-phosphate-dehydrogenase (GADPH) gene as a housekeeping gene.

SDS-PAGE and immunoblot analysis

Cells were treated with miRNA for 24 h and collected to evaluate the expression of amyloid precursor protein (APP) and its C-terminal fragments (CTFs). To prepare whole-cell lysates, cells were washed with ice-cold PBS, scraped into radioimmunoprecipitation assay (RIPA) buffer (GE Healthcare Bio-Sciences, Pittsburgh, PA, USA) containing a cocktail of both protease inhibitors and phosphatase inhibitors (Sigma-Aldrich, St. Louis, MO, USA), sonicated to homogenize the cell suspension, and centrifuged to remove cell debris. Total protein extract (10–100 µg) was subjected to western blot analysis with an appropriate antibody. The isolated protein was electrophoresed in a 10% or 15% polyacrylamide-SDS gel and transferred to a nitrocellulose membrane. After blocking for 1 h with 5% skim milk, the membrane was incubated with a primary antibody. After washing, the membrane was then incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG or goat anti-mouse IgG antibody. Immunoreactive signals were detected with an enhanced chemiluminescence detection kit (Merck Millipore, Temecula, CA, USA) and analyzed using a Chemi-Doc System (Bio-Rad).

Statistical analysis

Groups were compared using chi-square tests for categorical variables or the Kruskal–Wallis test followed by post hoc Dunn’s multiple comparisons among three groups or the Mann–Whitney U test to compare two groups for continuous variables. To evaluate the effects of miRNA on the levels of target gene expression or APP protein expression and fragmentation, we used the Mann–Whitney U test or one-sample t tests to compare data from 4 or 6 independent experiments, respectively. Receiver-operating characteristic (ROC) curve analyses were performed to assess the diagnostic performance of various miRNAs to distinguish patients from normal subjects using Prism (v. 8.0; GraphPad Software, San Diego, CA, USA). We considered P < 0.05 as significant.

Identification of differentially expressed miRNAs in plasma

Small RNA was extracted from the plasma samples of 23 individuals (8 AD patients, 8 PD patients, and 7 age-matched healthy control subjects). Following RNA quality control, we performed miRNAome analysis using the Illumina sequencing platform. Although we identified 2060 miRNAs from all the tested samples, we further analyzed only the 613 miRNAs that had a total expression of at least 10 RPTM. To rule out individual variation in the miRNA profiles, we made average readouts of each testing group. Then, we used the average readouts of each testing group as a standard for calculating the individual change ratio (log₂ΔRPTM value). We found that the median expression levels of miRNAs were slightly variable among individuals in each group. For instance, the log₂ΔRPTM values of CON 1 to CON 7 were –0.01, 0.08, 0.00, –0.02, 0.29, –0.33, and –0.31, showing the range of fluctuation was around ±40% (±0.2 change ratio) (Figure 1A). Similarly, the log₂ΔRPTM values of AD patients and PD patients fluctuated within the ±40% range (±0.21 and ±0.24 change ratio, respectively) (Figure 1B and 1C). Next, we compared the average read counts of healthy controls with those of AD and PD patients. The change ratios of the control average versus the AD average and PD average were 0.16 and 0.12, respectively (Figure 1D). These results indicated that the expression changes of total miRNAs were not dramatic in both AD and PD patients compared with healthy controls. Moreover, the up-and-down expression of miRNAs less than ±40% (±0.2 change ratio) should be considered an individual variation rather than a reflection of illness. Therefore, we selected 17 miRNAs that were upregulated in all the tested AD patients at least 2-fold (log₂ΔRPTM > 1.0) (Figure 1E).

In addition to these >2-fold upregulated miRNAs, we selected miR-873-5p, miR-222-3p, and miR-451a as negative or positive control markers from previous studies [19, 21, 22]. miR-873-5p relates to liver fibrosis, whereas miR-222-3p and miR-451a link to AD and PD, respectively. Through ROC curve analysis and one-way analysis of variance (ANOVA) of sequencing results, we assumed that 4 miRNAs (miR-210-3p, miR-4454, miR-590-5p, and miR-122-5p) could be candidates for AD biomarkers that discriminate AD from healthy controls (Figure 2, Supplementary Figure 1). For example, although the number of subjects was limited, the expression levels of miR-210-3p, miR-4454, miR-590-5p, and miR-122-5p were higher in AD than in the healthy controls that provided the area under the curve (AUC) values of 0.93, 1.0, 1.0, and 0.98, respectively in ROC analysis. Moreover, the comparison of miR-210-3p, miR-590-5p, and miR-122-5p expression between AD and PD provided AUC values of 0.80, 1.0, and 0.91, respectively, showing the AD-specificity of these miRNAs. One-way ANOVA with Tukey’s post hoc test confirmed that the expressions of miR-210-3p, miR-590-5p, and miR-122-3p in AD patients were significantly different from those of PD patients and healthy controls (Supplementary Figure 1). Furthermore, we investigated significantly downregulated miRNAs in AD as compared with control or PD. We found that only miR-1299 was downregulated by >2-fold (log₂ΔRPTM > –1.0) and 17 miRNAs were downregulated by >1-fold (log₂ΔRPTM > –0.5) in all the tested AD patients (Supplementary Figure 2). Even though miR-1299 was notably downregulated in AD patients, it was also reduced in PD patients, implying that miR-1299 could not be used as an AD-specific marker. Hence, we ruled out the reduced miRNAs in the following analyses.

Comparison of miRNAs levels in plasma according to disease or amyloid PET status

To evaluate the levels of differentially expressed miRNAs in plasma of subjects along an AD continuum (i.e., CN, MCI, and AD), we quantified the miRNAs in plasma by qRT-PCR in a subgroup of 100 subjects (Figure 3). The level of miR-122 showed higher levels in MCI (P = 0.0348, Kruskal–Wallis test and P < 0.05 for CN versus MCI by multiple comparisons) but not in AD, while miR-210 in the MCI or AD group was significantly higher than CN (P = 0.0184, Kruskal–Wallis test and P < 0.05 for CN versus MCI or AD by multiple comparisons). The level of miR-590 showed a marginally significant difference among groups (P = 0.0551 by the Kruskal–Wallis test). However, when we compared the levels of these miRNAs between amyloid PET-positive and amyloid PET-negative subjects, all miRNAs in amyloid PET-positive subjects showed higher levels than those in amyloid PET-negative subjects. We consistently found that amyloid PET-positive MCI or AD subjects showed higher levels of miRNAs than those in amyloid PET-negative CN subjects. In particular, MCI with amyloid positivity showed higher levels of all selected miRNAs than those in amyloid-negative MCI patients, which was not observed in AD patients. Taken together, these results suggest that the selected miRNAs differentially expressed in AD plasma may be associated with amyloid accumulation during disease progression.

Comparison of miRNAs levels in plasma EV

Selected miRNAs are actively packed into EVs according to the disease condition. EV-encapsulated miRNAs are more stable and protected from dilution in the extracellular environment and therefore are deemed promising biomarkers. We quantified miRNAs in EVs isolated from plasma using a qRT-PCR method, as previously described [17]. In accordance with the results of total plasma, when we measured the levels of selected miRNAs in plasma EV of 262 subjects, the levels of miR-122, miR-210, and miR-590 in amyloid PET-positive subjects were higher than those in PET-negative subjects (Figure 4). In particular, the levels of miR-122 and miR-590 in MCI with amyloid positivity, but not in amyloid-positive AD, were significantly higher than those in amyloid-negative CN. The level of miR-210 in amyloid-positive MCI or AD was higher than in amyloid-negative patients. When we evaluated the diagnostic utility of those miRNAs in total plasma or plasma EVs to differentiate MCI or AD from CN using ROC analysis, the diagnostic performance was not significant (AUC < 0.6). This suggests that although the selected miRNAs in plasma EV have little clinical diagnostic value, these miRNAs may be associated with amyloid accumulation in the early stages of AD progression.

Effects of miRNAs on amyloidogenesis

To evaluate the effects of selected miRNAs on Aβ production, we measured the expression of APP and its CTFs in H4-APPswe cells. All selected miRNAs significantly upregulated the expression of the mature form of APP (112~125% of control, Figure 5A and B), and all miRNAs significantly increased the levels of CTF-β (130–156% of control). The levels of CTF-α were significantly upregulated by miR-122 (123% of control), but not by other miRNAs. The ratio CTF-β/CTF-α was significantly increased by miR-590 (≈125% of control) and miR-122 (≈127% of control), and increased by miR-210 but with marginal significance (114% of control) (Figure 5C-E). When we measured the levels of Aβ40 and Aβ42 in culture media of miRNA-treated H4-APPswe cells, all selected miRNAs significantly increased the levels of Aβ40 (112–116% of control) and Aβ42 (111–118% of control) (Figure 5F and G).

Regulation of candidate genes targeted by miRNAs

We further validated the function of differentially expressed miRNAs on the expression of target genes associated with amyloidogenesis. Using the miRDB micro-RNA target prediction database, we retrieved predicted target genes. Among genes predicted by miRDB, we identified 92 out of 490 targets, 13 out of 83 targets, and 132 out of 427 targets predicted for miR-122, miR-210, and miR-590, respectively with a target prediction score ≥ 80 (Supplementary Information). Among target genes, through a literature search in PUBMED (www.ncbi.nlm.nih.gov/pubmed, June 18, 2021), we selected target genes associated with amyloidogenesis to validate the function of miRNAs. Among targets of miR-122, we selected ADAM metallopeptidase domain 10 (ADAM10). This gene encodes an α-secretase that promotes anti-amyloidogenic processing of APP [23]. We also found that brain-derived neurotrophic factor (BDNF), an important neuroprotective factor against Aβ-induced neurotoxicity [24] and with anti-amyloidogenic processing activity [25, 26], is a predicted target of miR-210. For miR-590, we selected jagged canonical notch ligand 1 (JAG1) which is associated with BACE1 activity [27]. As shown in Figure 6, the treatment of H4-APPswe cells with miRNA significantly downregulated the target genes (76%, 68%, and 47% of control for ADAM10, BDNF, and JAG1, respectively).

Pathway enrichment analysis of DEGs and PPI network

For each given target gene of the selected miRNAs, pathway and process enrichment analysis was carried out with the following ontology sources: GO Biological Processes, GO Molecular Functions, GO Cellular Components, Reactome Gene Sets, and KEGG pathway. The pathways of miR-122 target genes were enriched in cell projection organization, anion transporter activity, and GTPase activity. The multitest-adjusted P values for the enriched pathway of miR-210 target genes were not significant. For miR-590, the GTPase activity, vasculature development, MAPK regulation, response to growth factor, and ubiquitin–proteasome pathways were enriched. Table 2 summarizes the top 5 enriched pathways for each miRNA target gene. Next, we input a total of 234 predicted targets of miR-122, miR-210, and miR-590 (3 target genes overlapped) into the Metascape website to identify overall pathway enrichments and to build a PPI network of entire target genes. As demonstrated in a cluster heat map of the GO analysis (Figure 7A), the clustered DEGs suggested principal enrichment in GTPase regulator activity, transmembrane RTK signaling pathway, response to growth factor, negative regulation of cell differentiation, and ossification. In addition, an enriched network demonstrating the relationship between the terms was built. In this network, each node symbolized a rich term that was colored by cluster ID (Figure 7B) and P value (Figure 7C).

Table 2

Top 5 enriched clusters with the lowest P values among pathways for each miRNA target genes.
miRNA	Total targets (N)	Candidate (N)	Term_ID	Term name	Gene count	Log10(P)	Log10(q)
miR-17-5p	1333	640	GO:0048667	Cell morphogenesis involved in neuron differentiation	48	-13.90	-9.54
			GO:0007264	Small GTPase mediated signal transduction	40	-11.13	-7.76
			GO:0044257	Cellular protein catabolic process	51	-10.45	-7.24
			GO:0030695	GTPase regulator activity	37	-10.18	-7.00
			GO:0051020	GTPase binding	24	-9.40	-6.27
miR-122-5p	490	92	GO:0031344	Regulation of cell projection organization	11	-5.01	-0.65
			GO:0008509	Anion transmembrane transporter activity	9	-4.68	-0.62
			GO:0048471	Perinuclear region of cytoplasm	10	-3.91	-0.20
			GO:0043087	Regulation of GTPase activity	8	-3.70	-0.20
			GO:0045936	Negative regulation of phosphate metabolic process	8	-3.48	-0.20
miR-210-3p	83	13	GO:0022409	Positive regulation of cell-cell adhesion	3	-3.61	0
			GO:0015980	Energy derivation by oxidation of organic compounds	3	-3.60	0
			GO:0048589	Developmental growth	3	-2.62	0
miR-590-5p	427	132	GO:0043087	Regulation of GTPase activity	14	-7.12	-2.76
			GO:0001944	Vasculature development	17	-6.76	-2.73
			GO:0043408	Regulation of MAPK cascade	16	-6.61	-2.73
			GO:0070848	Response to growth factor	16	-6.42	-2.66
			GO:0043161	Proteasome-mediated ubiquitin-dependent protein catabolic process	12	-6.08	-2.62

For each given total predicted gene targets, pathway and process enrichment analysis has been carried out with the following ontology sources: KEGG Pathway, GO Biological Processes, GO Molecular Functions, and GO Cellular Components. All genes in the genome have been used as the enrichment background. Terms with a p-value < 0.01, a minimum count of 3, and an enrichment factor > 1.5 (the enrichment factor is the ratio between the observed counts and the counts expected by chance) are collected and grouped into clusters based on their membership similarities. More specifically, p-values are calculated based on the accumulative hypergeometric distribution, and q-values are calculated using the Banjamini-Hochberg procedure to account for multiple testings³. Kappa scores⁴ are used as the similarity metric when performing hierachical clustering on the enriched terms, and sub-trees with a similarity of > 0.3 are considered a cluster. The most statistically significant term within a cluster is chosen to represent the cluster. "Log10(P)" is the p-value in log base 10. "Log10(q)" is the multi-test adjusted p-value in log base 10.

As shown in Figure 8, 5 modules in the PPI network were detected when the MCODE algorithm was applied. In the top 3 selected terms for each MCODE, the networks primarily were enriched in ubiquitin or modification-dependent protein catabolic process, T cell differentiation, regulation of neuronal death, glycogen or glucan biosynthetic processes, cellular polysaccharide biosynthetic process, chloride channel or anion channel activity, and chloride channel complex.

Circulating miRNAs are emerging biomarkers for AD with diagnostic potential and/or therapeutic targets of AD pathogenesis. Although numerous clinical studies have reported the diagnostic utility of various circulating miRNAs in whole blood, plasma, serum, or exosomes, the diagnostic performance for AD is inconsistent and unreliable. In addition, the differential expression of specific miRNAs in biofluid compartments (e.g., CSF, whole blood, plasma, or plasma exosome) as well as in different types of tissues (i.e., plasma versus brain) has been reported[6, 11]. Using validated miRNA discovery methodologies, we identified several miRNAs in blood that were differentially expressed in AD compared with CN or PD. Although we could not observe a clinically applicable diagnostic performance (i.e., >80% of both diagnostic sensitivity and specificity) of selected miRNAs in both plasma or plasma exosomes, we observed a significant association of selected miRNA levels with amyloid accumulation in the brain. A recent systematic review with meta-analysis reported various blood miRNAs with strong evidence of differential expression[11]. Among these, lower levels of miR-210-3p from 4 studies and miR-122-5p from 3 studies in AD blood-derived samples than those in CN samples were consistently reported. In contrast, we observed higher levels of these miRNAs in plasma and circulating exosomes in AD. In addition, a study evaluating the level of miR-210-3p in AD plasma reported higher levels of miR-210-3p in MCI or AD than in CN subjects [28]. For miR-122-5p, an Alzheimer’s animal model showed upregulation of miR-122, which induced neuronal apoptosis [29]. Thus, there are inconsistencies in the level of miRNAs in plasma that preclude their use as diagnostic AD biomarkers. This disparity may result from pre-analytical variability that influences the levels of miRNAs in blood across the studies. For example, clinical differences such as race, diet, age, and severity of disease and/or pre-analytical variables during sample processing may contribute to variation even before a sample is analyzed. Analytical variables during miRNA quantification including miRNA normalization may also be involved in the disparity. Interestingly, we identified a novel miRNA (miR-590-5p) that is differentially expressed in AD blood, as compared with CN or PD bloods, which should be confirmed in further studies.

We observed higher levels of selected miRNAs (i.e., miR-122-5p, miR-210-3p, and miR-590-5p) in both plasma and plasma exosomes in subjects with amyloid accumulation, indicating that encapsulation of these miRNAs in exosomes may not be affected significantly by amyloid accumulation. Rather, amyloid accumulation may induce the expression of these miRNAs. The level of miRNA in plasma or plasma exosomes represents a level combined from the whole body and is not dependent on production in one specific organ (i.e., brain) suggesting the origin or mechanisms of upregulation of these miRNAs in plasma in AD should be further elucidated.

We further evaluated the possible roles of upregulated miRNAs in amyloidogenesis using an in vitro system and found that selected miRNAs increased Aβ production from APP. Among the targets of each miRNA, miR-122-5p, miR-210-3p, and miR-590-5p downregulated the expression of ADAM10, BDNF, and JAG1, respectively. ADAM10 is an α-secretase that plays a role in the activation of the anti-amyloidogenic pathway[30] p producing CTF-α that is further cleaved to p3 by γ-secretase. Whether BDNF expression is directly involved in Aβ production requires further study, however, in vitro and in vivo evidence has suggested that activation of the BDNF–TrkB signaling pathway reduced Aβ production [25, 31]. Downregulation of JAG1 expression in aged brain has been linked with the Notch signaling pathway downstream [32] and reduced astrogliosis [27]. Levels of Aβ in the brain are regulated by clearance of Aβ by glial cells as well as production of Aβ. Activation of astrocytes also contributes to Aβ clearance. Inhibition of JAG1 expression by miR-590-5p may accelerate amyloid accumulation by inhibition of glial activation. Taken together, this information suggests the upregulation of miR-122-5p, miR-210-3p, and miR-590-5p may play a role in Aβ accumulation mechanisms through silencing their target genes, which may associate with amyloidogenic APP metabolism.

The predicted target genes of selected miRNAs were enriched in several AD-associated pathways. These enriched pathways reportedly participate in amyloidogenesis through Aβ production or clearance [33-35] tau phosphorylation, and neuroinflammation [36], which may be associated with the upregulation of miRNAs in amyloid PET-positive subjects of this study. For example, several GTPases regulate Aβ production through protein trafficking [34] or amyloidogenesis [33]. Receptor tyrosine kinase signaling pathways (e.g., TrkA, Tyro3) regulate BACE1 trafficking and amyloidogenic processing [35, 37]. Other enriched pathways such as “response to growth factor” [38, 39], “negative regulation of cell differentiation” [40], and “ubiquitin proteasome pathway” [41] are associated with AD pathogenesis. In a module analysis of the PPI network, 4 seed genes were selected: HECTD3, the target of miR-122-5p; PPP1R3A, the target of miR-590-5p; TBR1, the target of miR-122-5p, and CLIC2, the target of miR-590-5p. A seed of HECTD3, an E3 ubiquitin ligase, was linked with other proteins of Aβ-clearing ubiquitin-proteasome system, such as ubiquitin-conjugating enzymes and SKP2, which has been reported as a key enzyme for AD via genome-wide expression profiling studies [42, 43], although the role of HECTD3 in AD has not yet been defined. PPP1R3A is a gene encoding a regulatory subunit of protein phosphatase 1 (PP1) which is a key component of insulin signaling and tau phosphorylation [44]. It has not been fully elucidated whether PP1 activity is associated with amyloid deposit and the onset of AD. A genetic association study reported that a polymorphism in PPP1R3 is associated with the risk of late-onset AD [45]. TBR1 is a neocortex-specific transcriptional factor and has been implicated in alterations in the brain that may lead to AD through the disturbance of neuronal migration [46]. The roles of miR-590-5p targeting genes in a cluster of chloride channel genes require further study. Genes in a cluster of growth factors are associated with neuronal survival (target of miR-210-3p), suggesting that upregulation of miR-210-3p in plasma may be associated with mechanisms of AD onset and progression.

To our knowledge, our report is the first to measure miRNAs in both plasma and circulating EVs that are differentially expressed in AD in a Korean population. Nonetheless, our study has some limitations. First, we neither discovered nor evaluated the levels of differentially expressed miRNAs in large clinical cohorts. Our results should be further confirmed in larger cohorts and compared with other ethnic groups. Second, we did not determine the ApoE genotype of participants, which may limit the interpretation of the roles of selected miRNAs in amyloidogenesis. Finally, we did not evaluate the association of miRNA levels with other AD pathologies such as tau phosphorylation or Aβ/Tau/Neurodegeneration classification. Because amyloid pathogenesis is closely associated with tau hyperphosphorylation and subsequent neurodegeneration, the pathogenic mechanisms of selected miRNAs in the Aβ–tau interaction should be further evaluated.

In summary, we identified several miRNAs from AD patients that are differentially upregulated in plasma from CN or PD patients via next-generation sequencing screening and verification by qPCR. The clinical utilities of miR-122-5p, miR-210-3p, and miR-590-5p in plasma and plasma EV to diagnose AD from CN were modest, however, these miRNAs were highly expressed specifically in patients with amyloid accumulation in the brain. We found that target genes of the selected miRNAs are enriched in the pathways associated with amyloid deposition via pathway analysis and PPI network analysis. Some of the targets of selected miRNAs were validated in an in vitro system. Therefore, our results suggest that blood-based miRNA biomarkers may have implications for amyloidogenesis during AD onset and progression.

AD: Alzheimer’s disease

CSF: Cerebrospinal fluid

MCI: Mild cognitive impairment

miRNA: Micro-RNA

EV: Extracellular vesicle

CN: Cognitively normal controls

PD: Parkinson’s disease

PET: Positron emission tomography

EDTA: Ethylenediaminetetraacetic acid

RT: Room temperature

RPTM: Reads per ten million

qRT-PCR: Quantitative real-time polymerase chain reaction

GO: Gene ontology

DEG: Differentially expressed gene

PPI: Protein-protein interaction

FBS: Fetal bovine serum

CM: Culture medium

APP: Amyloid precursor protein

CTF: C-terminal fragment of APP

RIPA: Radioimmunoprecipitation assay

ROC: Receiver operating characteristic

AUC: Area under the curve

ADAM10: ADAM metallopeptidase domain 10

BDNF: Brain-derived neurotrophic factor

JAG1: Jagged canonical notch ligand 1

PP1: Protein phosphatase 1

Availability of data and materials

All data generated or analyzed during this study are available from the corresponding author on reasonable request, or included in this published article.

Acknowledgement

The authors thank the patients for their participation in this study. We are grateful for the support from Ms. Jin-Ah Lee and Hye-Ran Na (Department of Neurology, Inha University Hospital) for data and sample collection.

Funding

This work was supported by a grant from the Korea Health Technology R&D Project through the KHIDI, funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI16C1118), the Medical Research Center Program through the National Research Foundation of Korea (NRF) funded by the Korean government (MSIP)(2021R1A5A2031612), and the NRF funded by the Ministry of Education, Republic of Korea (2019S1A5C2A03082727).

Author’s contributions

SM, SK contributed equally to the work as coauthors. SHC, PHL, SWY, JHK had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: SHC, PHL, SWY and JHK. Data acquisition, analysis, or interpretation of data: SM, SK, ShM, HBK, DHP, PS, PHL, SWY and JHK. Drafting of the manuscript: SM, SK, PS, SWY and JHK. Critical revision of the manuscript for important intellectual content: All authors. Obtaining funding: PHL, SWY and JHK. Administrative, technical or material support: ShM, PHL, SWY and JHK.

Ethics declaration

All participants provided written informed consent for participation in this study, which was approved by institutional review boards.

Consent for publication

Not applicable.

Competing interests

All authors declare no conflict of interest.

Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, 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(3):270-9.
McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR, Jr., Kawas CH, 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(3):263-9.
Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S, Fagan AM, et al. Toward defining the preclinical stages of 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(3):280-92.
Olsson B, Lautner R, Andreasson U, Ohrfelt A, Portelius E, Bjerke M, et al. CSF and blood biomarkers for the diagnosis of Alzheimer's disease: a systematic review and meta-analysis. Lancet Neurol. 2016;15(7):673-84.
Piscopo P, Lacorte E, Feligioni M, Mayer F, Crestini A, Piccolo L, et al. MicroRNAs and mild cognitive impairment: A systematic review. Ageing Res Rev. 2019;50:131-41.
Samadian M, Gholipour M, Hajiesmaeili M, Taheri M, Ghafouri-Fard S. The Eminent Role of microRNAs in the Pathogenesis of Alzheimer's Disease. Front Aging Neurosci. 2021;13:641080.
Cheng L, Sharples RA, Scicluna BJ, Hill AF. Exosomes provide a protective and enriched source of miRNA for biomarker profiling compared to intracellular and cell-free blood. J Extracell Vesicles. 2014;3:23743
Heinicke F, Zhong X, Zucknick M, Breidenbach J, Sundaram AYM, S TF, et al. Systematic assessment of commercially available low-input miRNA library preparation kits. RNA Biol. 2020;17(1):75-86.
Leek JT, Scharpf RB, Bravo HC, Simcha D, Langmead B, Johnson WE, et al. Tackling the widespread and critical impact of batch effects in high-throughput data. Nat Rev Genet. 2010;11(10):733-9.
Shi D, Han M, Liu W, Tao J, Chen L. Circulating MicroRNAs as Diagnostic Biomarkers of Clinical Cognitive Impairment: A Meta-Analysis. Am J Alzheimers Dis Other Demen. 2020;35:1533317520951686.
Takousis P, Sadlon A, Schulz J, Wohlers I, Dobricic V, Middleton L, et al. Differential expression of microRNAs in Alzheimer's disease brain, blood, and cerebrospinal fluid. Alzheimers Dement. 2019;15(11):1468-77.
Wu HZ, Ong KL, Seeher K, Armstrong NJ, Thalamuthu A, Brodaty H, et al. Circulating microRNAs as Biomarkers of Alzheimer's Disease: A Systematic Review. J Alzheimers Dis. 2016;49(3):755-66.
Zhuang F, Fuchs RT, Sun Z, Zheng Y, Robb GB. Structural bias in T4 RNA ligase-mediated 3'-adapter ligation. Nucleic Acids Res. 2012;40(7):e54.
Hwang J, Jeong JH, Yoon SJ, Park KW, Kim EJ, Yoon B, et al. Clinical and Biomarker Characteristics According to Clinical Spectrum of Alzheimer's Disease (AD) in the Validation Cohort of Korean Brain Aging Study for the Early Diagnosis and Prediction of AD. J Clin Med. 2019;8:341
Gyorgy B, Paloczi K, Kovacs A, Barabas E, Beko G, Varnai K, et al. Improved circulating microparticle analysis in acid-citrate dextrose (ACD) anticoagulant tube. Thromb Res. 2014;133(2):285-92.
Yokota M, Tatsumi N, Nathalang O, Yamada T, Tsuda I. Effects of heparin on polymerase chain reaction for blood white cells. J Clin Lab Anal. 1999;13(3):133-40.
Moon S, Shin DW, Kim S, Lee YS, Mankhong S, Yang SW, et al. Enrichment of Exosome-Like Extracellular Vesicles from Plasma Suitable for Clinical Vesicular miRNA Biomarker Research. J Clin Med. 2019;8:1995
Chen Y, Wang X. miRDB: an online database for prediction of functional microRNA targets. Nucleic Acids Res. 2020;48(D1):D127-D31.
Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523.
Oh YM, Jang EH, Ko JH, Kang JH, Park CS, Han SB, et al. Inhibition of 6-hydroxydopamine-induced endoplasmic reticulum stress by l-carnosine in SH-SY5Y cells. Neurosci Lett. 2009;459(1):7-10.
Gamez-Valero A, Campdelacreu J, Vilas D, Ispierto L, Rene R, Alvarez R, et al. Exploratory study on microRNA profiles from plasma-derived extracellular vesicles in Alzheimer's disease and dementia with Lewy bodies. Transl Neurodegener. 2019;8:31.
Zeng Q, Zou L, Qian L, Zhou F, Nie H, Yu S, et al. Expression of microRNA222 in serum of patients with Alzheimer's disease. Mol Med Rep. 2017;16(4):5575-9.
Kuhn PH, Wang H, Dislich B, Colombo A, Zeitschel U, Ellwart JW, et al. ADAM10 is the physiologically relevant, constitutive alpha-secretase of the amyloid precursor protein in primary neurons. EMBO J. 2010;29(17):3020-32.
Arancibia S, Silhol M, Mouliere F, Meffre J, Hollinger I, Maurice T, et al. Protective effect of BDNF against beta-amyloid induced neurotoxicity in vitro and in vivo in rats. Neurobiol Dis. 2008;31(3):316-26.
Holback S, Adlerz L, Iverfeldt K. Increased processing of APLP2 and APP with concomitant formation of APP intracellular domains in BDNF and retinoic acid-differentiated human neuroblastoma cells. J Neurochem. 2005;95(4):1059-68.
Nascimento CM, Pereira JR, Pires de Andrade L, Garuffi M, Ayan C, Kerr DS, et al. Physical exercise improves peripheral BDNF levels and cognitive functions in mild cognitive impairment elderly with different bdnf Val66Met genotypes. J Alzheimers Dis. 2015;43(1):81-91.
Hu X. Microglia/macrophage polarization: Fantasy or evidence of functional diversity? J Cereb Blood Flow Metab. 2020;40(1_suppl):S134-S6.
Siedlecki-Wullich D, Catala-Solsona J, Fabregas C, Hernandez I, Clarimon J, Lleo A, et al. Altered microRNAs related to synaptic function as potential plasma biomarkers for Alzheimer's disease. Alzheimers Res Ther. 2019;11(1):46.
Gu R, Wang L, Tang M, Li SR, Liu R, Hu X. LncRNA Rpph1 protects amyloid-beta induced neuronal injury in SK-N-SH cells via miR-122/Wnt1 axis. Int J Neurosci. 2020;130(5):443-53.
Lammich S, Kojro E, Postina R, Gilbert S, Pfeiffer R, Jasionowski M, et al. Constitutive and regulated alpha-secretase cleavage of Alzheimer's amyloid precursor protein by a disintegrin metalloprotease. Proc Natl Acad Sci U S A. 1999;96(7):3922-7.
Nigam SM, Xu S, Kritikou JS, Marosi K, Brodin L, Mattson MP. Exercise and BDNF reduce Abeta production by enhancing alpha-secretase processing of APP. J Neurochem. 2017;142(2):286-96.
Sun F, Mao X, Xie L, Ding M, Shao B, Jin K. Notch1 signaling modulates neuronal progenitor activity in the subventricular zone in response to aging and focal ischemia. Aging Cell. 2013;12(6):978-87.
Henderson BW, Gentry EG, Rush T, Troncoso JC, Thambisetty M, Montine TJ, et al. Rho-associated protein kinase 1 (ROCK1) is increased in Alzheimer's disease and ROCK1 depletion reduces amyloid-beta levels in brain. J Neurochem. 2016;138(4):525-31.
Zhang X, Huang TY, Yancey J, Luo H, Zhang YW. Role of Rab GTPases in Alzheimer's Disease. ACS Chem Neurosci. 2019;10(2):828-38.
Zheng Y, Wang Q, Xiao B, Lu Q, Wang Y, Wang X. Involvement of receptor tyrosine kinase Tyro3 in amyloidogenic APP processing and beta-amyloid deposition in Alzheimer's disease models. PLoS One. 2012;7(6):e39035.
Cai R, Wang Y, Huang Z, Zou Q, Pu Y, Yu C, et al. Role of RhoA/ROCK signaling in Alzheimer's disease. Behav Brain Res. 2021;414:113481.
Zhang YW, Xu H. Alteration of beta-secretase traffic by the receptor tyrosine kinase signaling pathway - a new mechanism for regulating Alzheimer's beta-amyloid production. Cell Res. 2007;17(5):385-6.
MacPherson REK. Filling the void: a role for exercise-induced BDNF and brain amyloid precursor protein processing. Am J Physiol Regul Integr Comp Physiol. 2017;313(5):R585-R93.
Yang C, Liu Y, Ni X, Li N, Zhang B, Fang X. Enhancement of the nonamyloidogenic pathway by exogenous NGF in an Alzheimer transgenic mouse model. Neuropeptides. 2014;48(4):233-8.
Liu H, Zhang H, Ma Y. Molecular mechanisms of altered adult hippocampal neurogenesis in Alzheimer's disease. Mech Ageing Dev. 2021;195:111452.
Hegde AN, Smith SG, Duke LM, Pourquoi A, Vaz S. Perturbations of Ubiquitin-Proteasome-Mediated Proteolysis in Aging and Alzheimer's Disease. Front Aging Neurosci. 2019;11:324.
Madadi S, Schwarzenbach H, Saidijam M, Mahjub R, Soleimani M. Potential microRNA-related targets in clearance pathways of amyloid-beta: novel therapeutic approach for the treatment of Alzheimer's disease. Cell Biosci. 2019;9:91.
Wu M, Liu Z, Zhang A, Li N. Identification of key genes and pathways in hepatocellular carcinoma: A preliminary bioinformatics analysis. Medicine (Baltimore). 2019;98(5):e14287.
Liao H, Li Y, Brautigan DL, Gundersen GG. Protein phosphatase 1 is targeted to microtubules by the microtubule-associated protein Tau. J Biol Chem. 1998;273(34):21901-8.
Liolitsa D, Powell J, Lovestone S. Genetic variability in the insulin signalling pathway may contribute to the risk of late onset Alzheimer's disease. J Neurol Neurosurg Psychiatry. 2002;73(3):261-6.
Pombero A, Bueno C, Saglietti L, Rodenas M, Guimera J, Bulfone A, et al. Pallial origin of basal forebrain cholinergic neurons in the nucleus basalis of Meynert and horizontal limb of the diagonal band nucleus. Development. 2011;138(19):4315-26.

SUPPLEMENTS.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Circulating micro-RNAs Differentially Expressed in Korean Alzheimer’s Patients with Brain Aβ Accumulation Activate Amyloidogenesis

Status:

Version 1

Abstract

Figures

Background

Methods

Study population

Blood collection and plasma preparation

Extraction of plasma small RNAs

Plasma miRNAome analysis

EV isolation, extraction of RNAs and amplification of specific EV miRNAs

miRNA target prediction and pathway enrichment analysis

Cell culture and transfection

Measurement of Aβ in culture media

Amplification of targeted mRNAs

SDS-PAGE and immunoblot analysis

Statistical analysis

Results

Identification of differentially expressed miRNAs in plasma

Comparison of miRNAs levels in plasma according to disease or amyloid PET status

Comparison of miRNAs levels in plasma EV

Effects of miRNAs on amyloidogenesis

Regulation of candidate genes targeted by miRNAs

Pathway enrichment analysis of DEGs and PPI network

Discussion

Conclusions

Abbreviations

Declarations

References

Supplementary Files

Status:

Version 1