Abstract
The EGADi project aims to identify innovative early biomarkers and therapeutic targets for Alzheimer's disease (AD) by investigating molecular mechanisms underlying its pathogenesis. Using advanced molecular biology techniques, EGADi will analyze biomarkers in plasma, cerebrospinal fluid, and extracellular vesicles (EVs) from healthy controls, prodromal subjects, and AD patients. The project will develop in vitro models derived from patient fibroblasts to study intercellular communication mediated by EVs. Expected outcomes include a comprehensive biomarker profile for early AD diagnosis and identification of new pharmacological targets, enabling improved diagnostic approaches and innovative treatments, ultimately promoting healthy aging and reducing the social burden of AD.
Results achieved
The primary goal of the EGADi project was to investigate the complex network of pathways involved in AD pathogenesis, with the aim of identifying innovative early biomarkers. This would facilitate the development of advanced diagnostic methods and therapeutic strategies, marking a new era in AD treatment. The specific objectives of EGADi were the identification of sensitive biomarkers for the early diagnosis of AD, particularly by creating a panel of circulating biomarkers combining Abeta and tau levels with synaptic failure assessments, as well as free and encapsulated miRNAs in extracellular vesicles (Milestone 1); and the exploration of potential disease-modifying strategies using innovative AD cellular models to identify the key cellular pathways influenced by EVs signaling (Milestone 2). Besides the significant outcome that the project would have in the research and innovation field, the EGADi project was expected to have a significant long-term impact on several levels: citizen health enhancing the diagnosis and management of AD patients; in social economy reducing the economic burden of AD; and on the environment diminishing the pharmaceutical waste due to the use of ineffective medications. EGADi project team was articulated in three independent research units (UNIBO, UNIMI, and UNIBS) equally represented in terms of gender and directed by high-profile researchers. At the beginning of the project, UNIBS initiated the recruitment of healthy individuals and patients across the AD spectrum, from prodromal stages to dementia. Clinical and neuropsychological evaluations were performed according to the NIA-AA Research Framework, together with structural MRI and biomarker analyses in cerebrospinal fluid (CSF), plasma, and EVs. Over the course of the project, recruitment continued progressively, allowing the establishment of a well-characterized cohort. The overall clinical cohort eventually included 205 subjects, comprising 150 AD patients and 55 healthy controls. In parallel, UNIBS recruited newly diagnosed AD and non-AD patients with CSF-confirmed diagnoses (n=80) for plasma biomarker studies and initiated the collection of skin biopsies from selected participants to generate patient-derived cellular models. The biological samples collected by UNIBS were distributed to the partner institutions for molecular analyses. UNIBO performed the extraction of total RNA from plasma and CSF samples and carried out miRNA profiling to identify potential AD-related biomarkers involved in synaptic transmission, neuroinflammation, cell signaling, and neuronal structural maintenance. The analysis identified several significantly deregulated miRNAs in AD patients. In particular, miR-125b-5p, miR-150-5p, miR-223-3p, and miR-101-3p were significantly downregulated. Extracellular vesicles were also isolated from plasma samples, and sequencing libraries for EV-associated miRNAs were prepared. Furthermore, the expression of inflammatory mediators released in culture media was quantified by ELISA assays. Finally, UNIBO generated induced pluripotent stem cell (iPSC)-derived neural cell types, including astrocytes, neurons, and microglia, from healthy controls, subjects with mild cognitive impairment (MCI), and AD patients. Astrocyte identity was confirmed through marker staining, and these cells were further characterized based on the expression levels of miRNAs implicated in AD progression. Indeed, deregulated miRNAs identified in plasma were validated in astrocytes and vesicles derived from human samples. In addition, UNIBO initiated the generation of brain organoids, providing a three-dimensional model to study disease mechanisms in a more physiologically relevant environment. In parallel, UNIMI focused on the analysis of synaptic proteins, particularly sAPPα and CAP2, and their relationship with classical AD biomarkers. The analysis of CAP2 levels in CSF revealed that AD patients exhibited significantly higher levels compared with healthy controls and patients with other neurodegenerative disorders. CAP2 concentrations were independent of disease severity and APOE genotype but showed a positive correlation with total tau and phosphorylated tau at Thr181 (p-tau181). To investigate a potential mechanistic relationship between CAP2 and tau pathology, CAP2 expression was experimentally downregulated in hippocampal neurons, resulting in a significant increase in p-tau181 without changes in total tau, suggesting a link between synaptic dysfunction and tau pathology in AD. To complement biomarker studies, the project also developed patient-derived cellular models. UNIBS recruited donors for skin biopsies to establish a fibroblast cohort from clinically characterized AD patients and healthy controls. UNIMI isolated and expanded fibroblasts, which were subsequently used both for iPSC reprogramming and for direct neuronal conversion. Using a doxycycline-inducible Tet-On system, fibroblasts were converted into induced neurons (iNs) by overexpressing the pro-neuronal transcription factors Ngn2 and Ascl1. Neuronal cultures were enriched by fluorescence-activated cell sorting (FACS) using the neuronal marker PSA-NCAM, allowing the separation of iNs from residual fibroblasts. Morphological and synaptic characterization confirmed the formation of functional synaptic structures, as demonstrated by the co-localization of the presynaptic marker synaptophysin and the postsynaptic marker PSD-95. The protocol was initially validated on a control donor to optimize the workflow before its extension to the entire cohort. Finally, UNIBS expanded the molecular characterization of the collected biological samples by integrating proteomic analyses in both plasma and CSF. The proteomic signature associated with AD included ptau phosphorylation markers (i.e ptau217, ptau181), glial and neuronal markers (NfL and GFAP), synaptic markers (SNAP-25 and NPTX-R), inflammatory markers (TNF alpha particularly). This complementary approach provided a more comprehensive view of protein alterations associated with neurodegeneration and inflammatory pathways. Overall, the EGADI project successfully combined clinical cohort recruitment, biomarker discovery, and advanced patient-derived cellular models. The integration of clinical data, circulating biomarkers, miRNA profiling, proteomics, and cellular modeling has contributed to identifying novel molecular signatures associated with synaptic dysfunction, tau pathology, and neuroinflammation, offering promising perspectives for the development of early diagnostic markers and therapeutic targets in AD. To ensure progress in the scientific objectives, the consortium implemented an integrated statistical analysis of circulating biomarkers within a panel of 26 plasma biomarkers related to neurodegeneration and inflammation. The dataset included AD patients and healthy controls, and data preprocessing involved numeric conversion, median imputation of missing values, and log-transformation of miRNAs and cytokines. Univariate analyses identified significant differences between groups, particularly for GFAP, pTau217, pTau181, NfL, and several inflammatory mediators. Principal Component Analysis indicated that the main contribution driven by neurodegeneration-related markers, especially tau proteins and GFAP. Feature selection using LASSO logistic regression identified a reduced subset of informative biomarkers, including GFAP, pTau181, pTau217, and TRAC. A Random Forest classification model trained on the full biomarker panel showed excellent discriminative performance (AUC ≈ 0.98), with pTau217, pTau181, GFAP, and NfL emerging as the most important predictors. In individuals with milder cognitive impairment, classification performance remained very high (AUC ≈ 0.99). Individual circulating miRNAs showed limited diagnostic power, but their combined signature significantly improved discrimination in early-stage disease (AUC ≈ 0.90). Finally, correlation analyses suggested associations between miRNAs and neurodegeneration markers. Overall, the project has created a solid foundation for future research activities and collaborative publications, and it is expected to contribute significantly to the improvement of diagnostic strategies and patient management in AD.Project details
Unibo Team Leader: Patrizia Hrelia
Unibo involved Department/s:
Dipartimento di Farmacia e Biotecnologie
Coordinator:
ALMA MATER STUDIORUM - Università di Bologna(Italy)
Total Eu Contribution: Euro (EUR) 300.000,00
Total Unibo Contribution: Euro (EUR) 120.000,00
Project Duration in months: 24
Start Date:
30/11/2023
End Date:
28/02/2026
