Distinct brain extracellular vesicle microRNA profiles differ in frontotemporal dementia and Alzheimer's disease

Dementia is a syndrome caused by various diseases including Alzheimer's disease (AD) and frontotemporal dementia (FTD) with an estimated global prevalence of 60 million individuals. Recently, therapeutic development in the dementia field has accelerated, with the introduction of monoclonal antibody therapeutics such as Lecanemab and Donanemab. However, AD and FTD patients are still either diagnosed too late to benefit from available therapies or are misdiagnosed due to the clinical overlap between dementia subgroups making therapeutic intervention challenging. This highlights a real need to improve early diagnostic tools of neurodegenerative disease (ND) biomarkers. A potential source of such biomarkers come from small extracellular vesicles (sEVs), groups of cell-derived, lipid-bound assemblies with the capability to cross the blood-brain barrier (BBB) and known to carry pathogenic proteins associated with AD and FTD. A known cargo of sEVs is microRNA (miRNA), regulatory molecules that post-transcriptionally silence gene expression including transcripts of autophagic systems, processes which dysfunction in dementia-causing diseases leading to toxic aggregate build-up, causing neurodegeneration. The targeting of functional machineries in macroautophagy (MA) and chaperone-mediated autophagy (CMA) by different miRNA may vary between AD and FTD mutations, leading to potential biomarkers of disease being highlighted. Through isolating sEVs from the frontal cortex of post-mortem brain tissue of AD, FTD-MAPT, FTD-C9orf72, FTD-GRN and no-disease control patients (Manchester Brain Bank), miRNA cargoes were analysed and compared using real-time quantitative PCR (RT-qPCR). Seven autophagy-associated miRNA candidates (MA: miR-124-3p, miR-30a-5p, miR-128-3p; and CMA: miR-224-5p, miR-373-5p, miR-106a-3p and miR-26b-5p) were tested to identify dementia sub-group variations, used alongside small RNA-sequencing to explore broader miRNA variation within sEV populations. Of the miRNA tested miR-224-5p (P = 1.76 × 10^-5) and miR-106a-3p (P = 0.033) showed significant group differences, and further significant pairwise comparison differences [miR-224-5p: AD fold change (FC) = 4.29, MAPT FC = 7.62; miR-106a-5p: AD FC = 5.59] when compared with no disease controls and other dementia subgroups, potentially showing initial diagnostic and differentiating potential. Small RNA-sequencing results revealed 8 AD, 2 FTD-GRN, 52 FTD-MAPT and 12 FTD-C9orf72 differentially expressed sEV-miRNAs when compared with no disease controls. Further direct comparisons between AD versus FTD mutation-derived sEV cargoes, and even FTD mutation versus FTD mutation-derived sEV cargoes, identified additional miRNA with differentiating capabilities. These findings demonstrate sEV-derived miRNA signatures vary across dementia sub-types and suggest potential roles of sEV cargoes in both disease diagnostics and identifying drivers of ND, such as autophagic impairments and signalling pathways.