Periodic Reporting for period 1 - mirADAM (Blood microRNAs ADAM10-target: novel diagnostic and prognostic biomarkers for Alzheimer's disease by electrochemical sensors)
Reporting period: 2025-03-01 to 2026-02-28
Summary of the context and overall objectives of the project
This project addresses the growing global challenge of Alzheimer’s Disease (AD), a neurodegenerative disorder that significantly impacts the elderly population. AD is closely linked to Type 2 Diabetes (T2D), with evidence suggesting that T2D increases the risk of developing AD due to shared mechanisms such as insulin resistance and amyloid-beta (Aβ) accumulation. ADAM10, a protein involved in the non-amyloidogenic cleavage of amyloid precursor protein (APP), has emerged as a promising biomarker for AD. However, its regulation, particularly by microRNAs (miRNAs), remains poorly understood. Additionally, current diagnostic methods for AD are invasive, expensive, and inaccessible to many, highlighting the need for innovative, cost-effective, and less invasive solutions.
Objectives:
1. Identify and validate circulating miRNAs targeting ADAM10 that are deregulated in AD, T2D, and AD+T2D patients, and assess their diagnostic and prognostic value.
2. Develop a novel diagnostic device using validated miRNA panels for ADAM10 detection, leveraging electrochemical sensors for a low-cost, sensitive, and non-invasive approach.
3. Advance understanding of ADAM10 regulation and its role in the pathophysiology of AD and T2D, contributing to the development of therapeutic strategies.
Project Pathway to Impact:
1. Scientific Impact: The project will enhance knowledge of miRNA regulation of ADAM10, providing insights into the molecular mechanisms underlying AD and T2D. It will also establish miRNA panels as potential biomarkers for early diagnosis and prognosis of AD.
2. Technological Innovation: The development of a patented diagnostic device based on miRNA panels will offer a low-cost, sensitive, and accessible alternative to current diagnostic methods, reducing patient discomfort and healthcare costs.
3. Societal Impact: Early detection of AD through the diagnostic device could enable timely interventions, potentially delaying disease progression and improving quality of life for patients. The device’s affordability and ease of use will make it accessible to underserved populations, addressing health inequities.
4. Economic Impact: By reducing the costs associated with AD diagnosis and treatment, the project could alleviate financial pressures on healthcare systems, particularly in low-income settings.
Pathway:
• Work Package 1 (WP1): Screening and identification of miRNAs targeting ADAM10 in healthy controls, AD, T2D, and AD+T2D groups.
• Work Package 2 (WP2): Validation of identified miRNAs and their application in a diagnostic device using electrochemical sensors.
• Work Package 3 (WP3): Regular meetings and collaboration among international research teams to ensure effective project management.
• Work Package 4 (WP4): Training and skill development for the researcher, including patent writing and innovation courses.
• Work Package 5 (WP5): Dissemination and exploitation of results through scientific publications, conferences, outreach activities, and patent applications.
By integrating molecular biology, biotechnology, pharmacology, and gerontology, this interdisciplinary project aims to deliver impactful scientific, societal, and economic outcomes, fostering innovation in AD diagnosis and treatment.
Objectives:
1. Identify and validate circulating miRNAs targeting ADAM10 that are deregulated in AD, T2D, and AD+T2D patients, and assess their diagnostic and prognostic value.
2. Develop a novel diagnostic device using validated miRNA panels for ADAM10 detection, leveraging electrochemical sensors for a low-cost, sensitive, and non-invasive approach.
3. Advance understanding of ADAM10 regulation and its role in the pathophysiology of AD and T2D, contributing to the development of therapeutic strategies.
Project Pathway to Impact:
1. Scientific Impact: The project will enhance knowledge of miRNA regulation of ADAM10, providing insights into the molecular mechanisms underlying AD and T2D. It will also establish miRNA panels as potential biomarkers for early diagnosis and prognosis of AD.
2. Technological Innovation: The development of a patented diagnostic device based on miRNA panels will offer a low-cost, sensitive, and accessible alternative to current diagnostic methods, reducing patient discomfort and healthcare costs.
3. Societal Impact: Early detection of AD through the diagnostic device could enable timely interventions, potentially delaying disease progression and improving quality of life for patients. The device’s affordability and ease of use will make it accessible to underserved populations, addressing health inequities.
4. Economic Impact: By reducing the costs associated with AD diagnosis and treatment, the project could alleviate financial pressures on healthcare systems, particularly in low-income settings.
Pathway:
• Work Package 1 (WP1): Screening and identification of miRNAs targeting ADAM10 in healthy controls, AD, T2D, and AD+T2D groups.
• Work Package 2 (WP2): Validation of identified miRNAs and their application in a diagnostic device using electrochemical sensors.
• Work Package 3 (WP3): Regular meetings and collaboration among international research teams to ensure effective project management.
• Work Package 4 (WP4): Training and skill development for the researcher, including patent writing and innovation courses.
• Work Package 5 (WP5): Dissemination and exploitation of results through scientific publications, conferences, outreach activities, and patent applications.
By integrating molecular biology, biotechnology, pharmacology, and gerontology, this interdisciplinary project aims to deliver impactful scientific, societal, and economic outcomes, fostering innovation in AD diagnosis and treatment.
Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far
1.2 Explanation of the work carried out per WP
1.2.1 Work Package 1
Ethical aspects
The project was submitted and approved by the bioethics committee of the University of Barcelona according to the Institutional Review Board IRB00003099 CER052440.
MicroRNAs-ADAM10 targeted for AD's disease and diabetes mellitus type 2 (D2T)
The miRTargetLink 2.0 is the updated version of miRTargetLink Human developed by the staff of the Chair for Clinical Bioinformatics at Saarland University. The current version of the tool includes miRNA, target and pathway annotations for human, mouse and rat. In essence, miRTargetLink provides users with a visualization interface to explore and analyze interaction networks between miRNAs and target genes. Users can choose to include validated targets (strong or weak support) from miRTarBase and miRATBase or predicted targets compiled in mirDIP and miRDB. From the miRTargetLink 2.0 website, the ADAM10 gene and the edit network for miRNA targets (weak or strong or predicted validated), 433 entries were observed. Initially, the search tools for miRNAs targets and pathways (strong validated) were added in a unidirectional search using the interaction landscape for ADAM10. In this initial search, n = 7 entries, with n = 6 miRNAs were found: hsa-miR-103-3p, hsa-miR-122-5p, hsa-miR-448, hsa-miR-449a, hsa-miR-451a, hsa-miR-655-3p.
From these miRNAs (n = 6), two tables were displayed, (1) interaction table - contains all nodes and interactions included in the network, as well as experimental information, support, source database and references, and (2) Node annotation table – different annotations available for each node. In this study, n = 7 and n = 436 entries were detected in these two tables’ results, respectively. In Suppl. table 2, the filter “AD’s disease” was added, and n = 4 entries were found, with only n = 1 miR (hsa-miR-103a-3p), belonging to the Published Diseases category. For “D2T mellitus” filter no entries were found, so further analysis was performed.
MiRTargetLink is integrated with other tools from the Chair for Clinical Bioinformatics, which Enrichment analyses (EA) is performed by miRNA enrichment analysis or gene enrichment analysis with miEAA 2.0 or GeneTrail 3.0 respectively. These databases utilize a fundamental enrichment analysis technique known as over-representation analysis (ORA). This approach aims to determine whether a predefined miR set or functional annotation is significantly overrepresented within a given gene or protein list. It employs statistical tests, such as hypergeometric distribution or Fisher's exact test, to assess the significance of the enrichment. These tests evaluate the probability of the observed level of overlap between the query miR set and the predefined gene set occurring by chance. The resulting p-values or false discovery rates help establish the statistical significance of the enrichment.
In miEAA 2.0 173 entries were found for these n = 6 miRNAs-ADAM10 targeted selected in figure 2A, for which the filters “AD” AND “D2T Mellitus” were subsequently inserted in the miEAA results table. For this analysis, n = 1 entry were found for AD (n = 6 miRs) and n = 1 specific for DM (n = 5 miRs), and n = 5 miRs previous observed (has-miR-103a-3p, has-miR-451a, has-miR-448, has-miR-449a and has-miR-655-3p) were found concomitantly for both diseases. Given that no entries were noted specifically for T2DM, more comprehensive analyses were performed.
Analysis was extended by adding weak and strong validated miRNAs targets and pathways. In this search, n = 12 entries, with n = 11 miRNAs were found: hsa-miR-103-3p, hsa-miR-122-5p, hsa-miR-155-5p, hsa-miR-32-5p, hsa-miR-448, hsa-miR-449a, hsa-miR-451a, hsa-miR-484, hsa-miR-655-3p, hsa-miR-92a-3p and hsa-miR-92b-3p.
In this study, n = 12 and n = 584 entries were detected in the interaction table and node annotation table results, respectively. The “AD’s disease” filter was added, and n = 5 entries with n = 3 miRs (hsa-miR-32-5p, hsa-miR-103a-3p, hsa-miR-484), belonging to the Published Diseases category were found. For the “D2T mellitus” filter no entries were found, so EA-ORA was also performed in the miEAA 2.0.
In miEAA 2.0 8,285 entries were found for these n = 11 miRNAs-ADAM10 targeted selected in figure 2B, for which the filters “AD” AND “D2T Mellitus” were subsequently inserted in the miEAA results table. For this analysis, n = 5 entries were found for AD and n = 7 for DM, and all n = 11 miRs previous observed (has-miR-155-5p, has-miR-92a-3p, has-miR-92b-3p, has-miR-122-5p, has-miR-103a-3p, has-miR-484, has-miR-451a, has-miR-32-5p, has-miR-448, has-miR-449a and has-miR-655-3p) were found concomitantly for both diseases (Table 2). With an adjustment to target only for miRs involved in T2D, n = 9 were selected for analysis.
To further refine the search, the “Node annotation table” was also used to verify possible miRs candidates already analyzed in biological samples of blood and/or plasma. Thus, it was verified that has-miR-155-5p (blood), and has-miR-92a-3p, has-miR-122-5p, has-miR-32-5p (plasma) are the most promising candidates for this study, considering the proposed analysis material. However, due to the reduced number of miRs, it was included for analysis all miRs-ADAM10 proposed by EA that were weak or strong validated and linked to AD + T2DM (n = 9).
Among the n=9 miRs (hsa-miR-155-5p, hsa-miR-92b-3p, hsa-miR-122-5p, hsa-miR-103a-3p, hsa-miR-92a-3p, hsa-miR-484, hsa-miR-451a, hsa-miR-32-5p, hsa-miR-448), that present intersection between the diseases (AD+T2D), we also performed an analysis in miRNet (https://www.mirnet.ca/miRNet/home.xhtml(opens in new window)) between the two diseases. MiRNet is a miRNA-centric network visual analytics platform that integrates data from 14 different miRNA databases (TarBase, miRTarBase, miRecords, miRanda, miR2Disease, HMDD, PhenomiR, SM2miR, PharmacomiR, EpimiR, starBase, TransmiR, ADmiRE, and TAM 2.0). The microRNA and gene targets databases - Tarbase 9.0 and miRTarBase 9.0 were updated at 06/27/2024 and database and functions to distinguish mature vs pre-miR at 01/15/2025.
In this analysis, n = 409 miRs and n = 532 edges were observed, among which, n = 123 are common to both diseases. When inserting the n = 9 miRs selected by miRTarget 2.0 into these results, it was observed that only the miRs, hsa-miR-92a-3p and hsa-miR-103a-3p were common for both analysis platforms. For this reason, it was decided to include the search and analysis of the miRs-AD-T2DM-ADAM10 interaction also in MirNet. For this analysis, interactions were found among n = 3,943 genes and n = 676 miRs, with a total of n = 6,718 edges. The inclusion of the ADAM10, n = 7 miRs were identified (hsa-miR-let-7a-5p, hsa-let-7f-5p, hsa-miR-29b-3p, hsa-miR-26a-5p, hsa-miR-101-3p, hsa-miR-103-3p, hsa-miR-92a-3p), with the last two highlighted also being indicated by miRTarget 2.0.
Thus, through bioinformatic analysis using these two platforms, n = 14 miRs were initially selected for the analysis. Given the limited amount of biological material, it was necessary to carefully select the n = 5 most promising miRs (hsa-miR-103-3p, hsa-miR-92a-3p, hsa-miR-155-5p, hsa-miR-122-5p, hsa-miR-32-5p), considering their presence in both databases and their expression sites (mainly in plasma). Additionally, n = 2 miRs were used as endogenous CTs for analysis (hsa-miR-16 and hsa-miR-U6). Based on a search of the most current and relevant literature, n = 2 miRs were also included for validation assays (hsa-miR-221 and hsa-miR-19).
1.2.2 Work Package 2
Analysis of study participants
Descriptive Analysis of the Study Population
In summary, A total of 73 participants were included in the analyses, distributed as follows: 20 controls (CT), 13 controls with type 2 diabetes mellitus (CT+T2D), 20 patients with Alzheimer’s disease without type 2 diabetes mellitus (AD), and 20 patients with Alzheimer’s disease and type 2 diabetes mellitus (AD+T2D). Participants diagnosed with Alzheimer’s disease (AD and AD+T2D) were significantly older than controls (p = 0.001). Mean age ranged from 65.6 years in the CT group to approximately 74 years in both AD groups. No meaningful age differences were observed between AD participants according to diabetes status. Sex distribution was balanced across groups. Women accounted for 50.0% of participants in the CT and AD groups, 53.8% in the CT+T2D group, and 45.0% in the AD+T2D group, indicating no relevant sex-related imbalance between diagnostic or metabolic categories.
Marked differences were observed across groups in classical AD CSF biomarkers. Both AD groups showed substantially lower CSF Aβ42 concentrations and Aβ42/40 ratios, along with significantly higher CSF total tau and phosphorylated tau levels, compared with both control groups (all p < 0.001). These alterations were present regardless of diabetes status, although numerically higher tau concentrations were observed in the AD+T2D group.
Cognitive and functional measures clearly distinguished AD from control participants. Individuals with AD, with or without T2D, demonstrated lower MMSE scores and higher GDS scores compared to both control groups (all p < 0.001). CDR scores also differed across groups (p = 0.018) reflecting increased functional impairment in participants with cognitive decline. Educational attainment differed significantly between groups, with fewer years of formal education observed in AD participants, particularly in the AD+T2D group (p < 0.001). In contrast, BMI did not differ significantly across the four groups. The frequency of APOE ε4 carriers was markedly higher in the AD group (77.8%) compared with controls, while a lower prevalence was observed among AD participants with T2D (42.9%), suggesting a differential distribution of genetic risk according to metabolic status. Metabolic parameters showed a clear association with diabetes status. Participants with T2D, irrespective of cognitive diagnosis, exhibited significantly higher blood and CSF glucose levels compared with their non-diabetic counterparts (all p < 0.001). Elevated CSF glucose concentrations in T2D groups indicate a strong influence of peripheral metabolic dysregulation on central glucose availability.
Validation of miRNAs by RT-qPCR
Plasma samples were processed for microRNA extraction using the miRNeasy Serum/Plasma Kit (QIAGEN), following the manufacturer’s instructions. Prior to extraction, the samples were thawed on ice and gently homogenized. Each sample underwent a two-step centrifugation procedure to remove residual cells and debris (1,600 × g for 10 minutes followed by 16,000 × g for 10 minutes, both at 4 °C). An exogenous synthetic spike-in CT (cel-miR-39) was added to each sample before lysis to monitor extraction efficiency and technical variability. Subsequently, 200 µL of plasma were transferred to RNase-free tubes and mixed with 1 mL of QIAzol Lysis Reagent, followed by the addition of chloroform to promote phase separation. After centrifugation at high speed, the upper aqueous phase was carefully collected and combined with absolute ethanol according to the kit protocol to facilitate RNA binding to the silica membrane. The mixture was then applied to the RNeasy MinElute columns, which were washed with the appropriate buffers, and the total RNA enriched in small RNAs was finally eluted in 12–14 µL of RNase-free water. The eluates were stored at −80 °C until reverse transcription.
MicroRNA reverse transcription was performed using the TaqMan MicroRNA Reverse Transcription Kit (Thermo Fisher Scientific) together with the specific stem-loop primers for each target miRNA. Reaction mixtures were prepared using the recommended reagent volumes, and 2–5 µL of RNA were used as input, depending on the extraction yield. The reverse transcription was carried out under the standard thermal protocol: 16 °C for 30 minutes, 42 °C for 30 minutes, and 85 °C for 5 minutes, after which the cDNA was stored at 4 °C or −20 °C until qPCR analysis.
The RT-qPCR was conducted on the StepOne Plus Real-Time PCR System (Applied Biosystems) using the TaqMan Universal PCR Fast Master Mix and Fast 96-well optical plates (Applied Biosystems). Each PCR reaction had a final volume of 10 µL, consistent with Fast chemistry, and consisted of 5 µL of TaqMan Fast Master Mix (2X), 0.5 µL of the 20X TaqMan MicroRNA Assay, 1.0 µL of RT product, and nuclease-free water to volume. The Fast-cycling protocol recommended by the manufacturer was used: 95 °C for 20 seconds, followed by 40 cycles of 95 °C for 1 second and 60 °C for 20 seconds, with fluorescence acquisition during the annealing/extension step. All samples were run in technical duplicate, and no-template CTs were included in each run. Data normalization was performed using the Ct values of the endogenous reference miRNA (has-miR-16). Relative quantification was calculated using the ΔΔCt method, and the spike-in results were used to monitor extraction and RT efficiency across samples.
Analysis of the selected miRs
We observed in all the miRs selected and analyzed an important trend of increasing miRs expression levels in the AD and AD+DMT2 groups. However, these data were not significant for miRs 155 and 122. Several circulating miRNAs showed significant differences across diagnostic and metabolic groups. Among them, miR-92, miR-103, miR-19, and miR-221 demonstrated the greatest ability to discriminate CT from AD, showing consistently higher expression in both AD groups compared to controls. miR-92 exhibited a robust increase in AD regardless of diabetes status, suggesting a strong association with AD pathology per se. miR-103 and miR-19 showed a progressive increase from CT to AD, with the highest levels observed in AD+T2D, indicating sensitivity to combined neurodegenerative and metabolic alterations and supporting its potential as a marker of disease severity. miR-221 also differentiated CT from AD, although with a more moderate effect size. In contrast, miR-32, miR-155, and miR-122 showed greater variability and weaker group separation, limiting their discriminative value. Overall, miR-92, miR-103 and miR-19 emerged as the most promising candidates for distinguishing AD from controls, with miR-103 and miR-19 additionally reflecting increased disease severity associated with the presence of T2D.
1.2.1 Work Package 1
Ethical aspects
The project was submitted and approved by the bioethics committee of the University of Barcelona according to the Institutional Review Board IRB00003099 CER052440.
MicroRNAs-ADAM10 targeted for AD's disease and diabetes mellitus type 2 (D2T)
The miRTargetLink 2.0 is the updated version of miRTargetLink Human developed by the staff of the Chair for Clinical Bioinformatics at Saarland University. The current version of the tool includes miRNA, target and pathway annotations for human, mouse and rat. In essence, miRTargetLink provides users with a visualization interface to explore and analyze interaction networks between miRNAs and target genes. Users can choose to include validated targets (strong or weak support) from miRTarBase and miRATBase or predicted targets compiled in mirDIP and miRDB. From the miRTargetLink 2.0 website, the ADAM10 gene and the edit network for miRNA targets (weak or strong or predicted validated), 433 entries were observed. Initially, the search tools for miRNAs targets and pathways (strong validated) were added in a unidirectional search using the interaction landscape for ADAM10. In this initial search, n = 7 entries, with n = 6 miRNAs were found: hsa-miR-103-3p, hsa-miR-122-5p, hsa-miR-448, hsa-miR-449a, hsa-miR-451a, hsa-miR-655-3p.
From these miRNAs (n = 6), two tables were displayed, (1) interaction table - contains all nodes and interactions included in the network, as well as experimental information, support, source database and references, and (2) Node annotation table – different annotations available for each node. In this study, n = 7 and n = 436 entries were detected in these two tables’ results, respectively. In Suppl. table 2, the filter “AD’s disease” was added, and n = 4 entries were found, with only n = 1 miR (hsa-miR-103a-3p), belonging to the Published Diseases category. For “D2T mellitus” filter no entries were found, so further analysis was performed.
MiRTargetLink is integrated with other tools from the Chair for Clinical Bioinformatics, which Enrichment analyses (EA) is performed by miRNA enrichment analysis or gene enrichment analysis with miEAA 2.0 or GeneTrail 3.0 respectively. These databases utilize a fundamental enrichment analysis technique known as over-representation analysis (ORA). This approach aims to determine whether a predefined miR set or functional annotation is significantly overrepresented within a given gene or protein list. It employs statistical tests, such as hypergeometric distribution or Fisher's exact test, to assess the significance of the enrichment. These tests evaluate the probability of the observed level of overlap between the query miR set and the predefined gene set occurring by chance. The resulting p-values or false discovery rates help establish the statistical significance of the enrichment.
In miEAA 2.0 173 entries were found for these n = 6 miRNAs-ADAM10 targeted selected in figure 2A, for which the filters “AD” AND “D2T Mellitus” were subsequently inserted in the miEAA results table. For this analysis, n = 1 entry were found for AD (n = 6 miRs) and n = 1 specific for DM (n = 5 miRs), and n = 5 miRs previous observed (has-miR-103a-3p, has-miR-451a, has-miR-448, has-miR-449a and has-miR-655-3p) were found concomitantly for both diseases. Given that no entries were noted specifically for T2DM, more comprehensive analyses were performed.
Analysis was extended by adding weak and strong validated miRNAs targets and pathways. In this search, n = 12 entries, with n = 11 miRNAs were found: hsa-miR-103-3p, hsa-miR-122-5p, hsa-miR-155-5p, hsa-miR-32-5p, hsa-miR-448, hsa-miR-449a, hsa-miR-451a, hsa-miR-484, hsa-miR-655-3p, hsa-miR-92a-3p and hsa-miR-92b-3p.
In this study, n = 12 and n = 584 entries were detected in the interaction table and node annotation table results, respectively. The “AD’s disease” filter was added, and n = 5 entries with n = 3 miRs (hsa-miR-32-5p, hsa-miR-103a-3p, hsa-miR-484), belonging to the Published Diseases category were found. For the “D2T mellitus” filter no entries were found, so EA-ORA was also performed in the miEAA 2.0.
In miEAA 2.0 8,285 entries were found for these n = 11 miRNAs-ADAM10 targeted selected in figure 2B, for which the filters “AD” AND “D2T Mellitus” were subsequently inserted in the miEAA results table. For this analysis, n = 5 entries were found for AD and n = 7 for DM, and all n = 11 miRs previous observed (has-miR-155-5p, has-miR-92a-3p, has-miR-92b-3p, has-miR-122-5p, has-miR-103a-3p, has-miR-484, has-miR-451a, has-miR-32-5p, has-miR-448, has-miR-449a and has-miR-655-3p) were found concomitantly for both diseases (Table 2). With an adjustment to target only for miRs involved in T2D, n = 9 were selected for analysis.
To further refine the search, the “Node annotation table” was also used to verify possible miRs candidates already analyzed in biological samples of blood and/or plasma. Thus, it was verified that has-miR-155-5p (blood), and has-miR-92a-3p, has-miR-122-5p, has-miR-32-5p (plasma) are the most promising candidates for this study, considering the proposed analysis material. However, due to the reduced number of miRs, it was included for analysis all miRs-ADAM10 proposed by EA that were weak or strong validated and linked to AD + T2DM (n = 9).
Among the n=9 miRs (hsa-miR-155-5p, hsa-miR-92b-3p, hsa-miR-122-5p, hsa-miR-103a-3p, hsa-miR-92a-3p, hsa-miR-484, hsa-miR-451a, hsa-miR-32-5p, hsa-miR-448), that present intersection between the diseases (AD+T2D), we also performed an analysis in miRNet (https://www.mirnet.ca/miRNet/home.xhtml(opens in new window)) between the two diseases. MiRNet is a miRNA-centric network visual analytics platform that integrates data from 14 different miRNA databases (TarBase, miRTarBase, miRecords, miRanda, miR2Disease, HMDD, PhenomiR, SM2miR, PharmacomiR, EpimiR, starBase, TransmiR, ADmiRE, and TAM 2.0). The microRNA and gene targets databases - Tarbase 9.0 and miRTarBase 9.0 were updated at 06/27/2024 and database and functions to distinguish mature vs pre-miR at 01/15/2025.
In this analysis, n = 409 miRs and n = 532 edges were observed, among which, n = 123 are common to both diseases. When inserting the n = 9 miRs selected by miRTarget 2.0 into these results, it was observed that only the miRs, hsa-miR-92a-3p and hsa-miR-103a-3p were common for both analysis platforms. For this reason, it was decided to include the search and analysis of the miRs-AD-T2DM-ADAM10 interaction also in MirNet. For this analysis, interactions were found among n = 3,943 genes and n = 676 miRs, with a total of n = 6,718 edges. The inclusion of the ADAM10, n = 7 miRs were identified (hsa-miR-let-7a-5p, hsa-let-7f-5p, hsa-miR-29b-3p, hsa-miR-26a-5p, hsa-miR-101-3p, hsa-miR-103-3p, hsa-miR-92a-3p), with the last two highlighted also being indicated by miRTarget 2.0.
Thus, through bioinformatic analysis using these two platforms, n = 14 miRs were initially selected for the analysis. Given the limited amount of biological material, it was necessary to carefully select the n = 5 most promising miRs (hsa-miR-103-3p, hsa-miR-92a-3p, hsa-miR-155-5p, hsa-miR-122-5p, hsa-miR-32-5p), considering their presence in both databases and their expression sites (mainly in plasma). Additionally, n = 2 miRs were used as endogenous CTs for analysis (hsa-miR-16 and hsa-miR-U6). Based on a search of the most current and relevant literature, n = 2 miRs were also included for validation assays (hsa-miR-221 and hsa-miR-19).
1.2.2 Work Package 2
Analysis of study participants
Descriptive Analysis of the Study Population
In summary, A total of 73 participants were included in the analyses, distributed as follows: 20 controls (CT), 13 controls with type 2 diabetes mellitus (CT+T2D), 20 patients with Alzheimer’s disease without type 2 diabetes mellitus (AD), and 20 patients with Alzheimer’s disease and type 2 diabetes mellitus (AD+T2D). Participants diagnosed with Alzheimer’s disease (AD and AD+T2D) were significantly older than controls (p = 0.001). Mean age ranged from 65.6 years in the CT group to approximately 74 years in both AD groups. No meaningful age differences were observed between AD participants according to diabetes status. Sex distribution was balanced across groups. Women accounted for 50.0% of participants in the CT and AD groups, 53.8% in the CT+T2D group, and 45.0% in the AD+T2D group, indicating no relevant sex-related imbalance between diagnostic or metabolic categories.
Marked differences were observed across groups in classical AD CSF biomarkers. Both AD groups showed substantially lower CSF Aβ42 concentrations and Aβ42/40 ratios, along with significantly higher CSF total tau and phosphorylated tau levels, compared with both control groups (all p < 0.001). These alterations were present regardless of diabetes status, although numerically higher tau concentrations were observed in the AD+T2D group.
Cognitive and functional measures clearly distinguished AD from control participants. Individuals with AD, with or without T2D, demonstrated lower MMSE scores and higher GDS scores compared to both control groups (all p < 0.001). CDR scores also differed across groups (p = 0.018) reflecting increased functional impairment in participants with cognitive decline. Educational attainment differed significantly between groups, with fewer years of formal education observed in AD participants, particularly in the AD+T2D group (p < 0.001). In contrast, BMI did not differ significantly across the four groups. The frequency of APOE ε4 carriers was markedly higher in the AD group (77.8%) compared with controls, while a lower prevalence was observed among AD participants with T2D (42.9%), suggesting a differential distribution of genetic risk according to metabolic status. Metabolic parameters showed a clear association with diabetes status. Participants with T2D, irrespective of cognitive diagnosis, exhibited significantly higher blood and CSF glucose levels compared with their non-diabetic counterparts (all p < 0.001). Elevated CSF glucose concentrations in T2D groups indicate a strong influence of peripheral metabolic dysregulation on central glucose availability.
Validation of miRNAs by RT-qPCR
Plasma samples were processed for microRNA extraction using the miRNeasy Serum/Plasma Kit (QIAGEN), following the manufacturer’s instructions. Prior to extraction, the samples were thawed on ice and gently homogenized. Each sample underwent a two-step centrifugation procedure to remove residual cells and debris (1,600 × g for 10 minutes followed by 16,000 × g for 10 minutes, both at 4 °C). An exogenous synthetic spike-in CT (cel-miR-39) was added to each sample before lysis to monitor extraction efficiency and technical variability. Subsequently, 200 µL of plasma were transferred to RNase-free tubes and mixed with 1 mL of QIAzol Lysis Reagent, followed by the addition of chloroform to promote phase separation. After centrifugation at high speed, the upper aqueous phase was carefully collected and combined with absolute ethanol according to the kit protocol to facilitate RNA binding to the silica membrane. The mixture was then applied to the RNeasy MinElute columns, which were washed with the appropriate buffers, and the total RNA enriched in small RNAs was finally eluted in 12–14 µL of RNase-free water. The eluates were stored at −80 °C until reverse transcription.
MicroRNA reverse transcription was performed using the TaqMan MicroRNA Reverse Transcription Kit (Thermo Fisher Scientific) together with the specific stem-loop primers for each target miRNA. Reaction mixtures were prepared using the recommended reagent volumes, and 2–5 µL of RNA were used as input, depending on the extraction yield. The reverse transcription was carried out under the standard thermal protocol: 16 °C for 30 minutes, 42 °C for 30 minutes, and 85 °C for 5 minutes, after which the cDNA was stored at 4 °C or −20 °C until qPCR analysis.
The RT-qPCR was conducted on the StepOne Plus Real-Time PCR System (Applied Biosystems) using the TaqMan Universal PCR Fast Master Mix and Fast 96-well optical plates (Applied Biosystems). Each PCR reaction had a final volume of 10 µL, consistent with Fast chemistry, and consisted of 5 µL of TaqMan Fast Master Mix (2X), 0.5 µL of the 20X TaqMan MicroRNA Assay, 1.0 µL of RT product, and nuclease-free water to volume. The Fast-cycling protocol recommended by the manufacturer was used: 95 °C for 20 seconds, followed by 40 cycles of 95 °C for 1 second and 60 °C for 20 seconds, with fluorescence acquisition during the annealing/extension step. All samples were run in technical duplicate, and no-template CTs were included in each run. Data normalization was performed using the Ct values of the endogenous reference miRNA (has-miR-16). Relative quantification was calculated using the ΔΔCt method, and the spike-in results were used to monitor extraction and RT efficiency across samples.
Analysis of the selected miRs
We observed in all the miRs selected and analyzed an important trend of increasing miRs expression levels in the AD and AD+DMT2 groups. However, these data were not significant for miRs 155 and 122. Several circulating miRNAs showed significant differences across diagnostic and metabolic groups. Among them, miR-92, miR-103, miR-19, and miR-221 demonstrated the greatest ability to discriminate CT from AD, showing consistently higher expression in both AD groups compared to controls. miR-92 exhibited a robust increase in AD regardless of diabetes status, suggesting a strong association with AD pathology per se. miR-103 and miR-19 showed a progressive increase from CT to AD, with the highest levels observed in AD+T2D, indicating sensitivity to combined neurodegenerative and metabolic alterations and supporting its potential as a marker of disease severity. miR-221 also differentiated CT from AD, although with a more moderate effect size. In contrast, miR-32, miR-155, and miR-122 showed greater variability and weaker group separation, limiting their discriminative value. Overall, miR-92, miR-103 and miR-19 emerged as the most promising candidates for distinguishing AD from controls, with miR-103 and miR-19 additionally reflecting increased disease severity associated with the presence of T2D.
Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)
We observed that five miRNAs (n = 5) validated in this study show significant differences among the analyzed groups, particularly between Alzheimer’s disease patients and controls. The results obtained here will be further developed and applied in other scientific projects, with the aim of completing all objectives initiated in this study. We highlight that a project led by the researcher as principal investigator (PI) was recently submitted to the Spanish Ministry in order to advance the findings already observed in this work. In addition, we emphasize that a collaboration has already been established with the Biosensors Group at the Autonomous University of Barcelona, which may also contribute to the continuation and validation of these future findings.