P.5.b.007 Cerebrospinal fluid microRNA profile in Alzheimer's disease

Abstract

Within the last decade accumulating evidence suggests that epigenetic regulation is associated with the pathogenesis of Alzheimer’s disease (AD). Beside DNA methylation and histone modifications, microRNA (miRNA) describe a novel class of non-coding RNA molecules which exist in a functionally active form able to epigenetically regulate gene expression at the post-transcriptional level. First investigations assign miRNAs important roles in the pathophysiology of AD [1] offering the possibility to act as biomarker in an altered biological process and to link therapeutic options capable of targeting the pathological mechanism. In a miRNome profiling study we made use of open array technology to screen 1178 (miRBase v14) endogenous mature miRNAs in ante-mortem cerebrospinal fluid (CSF) for informative signals in a set of AD patients (n = 22) and heterogeneous disease controls (n = 28). The test group comprised clinically secure AD cases whereas cognitive healthy and disease controls with other neurodegenerative disorders were included in the control group. Applying Cq 34 as cut-off, 451 positive and 727 negative signals were identified in our CSF cohort (n = 50). The screening demonstrated that not less than 38% signals are constitutively expressed in brain which coincides with numbers of known protein-coding genes in cerebral tissues. Among our positive signals we identified 66 miRNAs as downand 55 miRNAs as upregulated between test groups above a fold change in expression of at least 1.5. Statistical analysis using the explorative method ‘Measure of Relevance’ resulted in 13 significant signals including a subset of 4 abundant and 9 less abundant miRNAs. Confirmatory MANCOVA revealed abundant miR-100−5p, miR-146a-5p and miR-1274a as differentially expressed in AD reaching Bonferroni corrected significance. T-tests confirmed also differential expression of less abundant miR-103a-3p, miR-375, miR-505−5p, miR-708−5p and miR-4467 as marginal significant between test groups. Discrimination analysis using a combination of miR-100−5p, miR-103a-3p and miR375 was able to detect AD in cerebrospinal fluid by positively classifying controls with 96.4% and AD cases with 95.5%. Using the computational microRNA target filter tool implemented in the web-based Ingenuity Pathway Analysis application, a list of AD associated mRNAs as potential targets of our informative miRNA subsets was identified. This list included genes such as MAPT, BACE1 and ADAM19 as highly predicted suggesting our miRNAs to be of functional relevance in AD and as potentially involved in regulating pathogenic mechanisms respectively. We provide a redundant CSF miRNA profile in AD and conclude that significantly expressed miRNAs in ante-mortem CSF could possibly act as novel biomarker for the detection of AD and might allow to design alternative strategies to treat AD in future. Continuing work is required to elucidate the applicability of our miRNA subset as a potential diagnostic test for AD. Differential expression and in silico predicted targets need to be further validated with RT-qPCR and in vitro cell culture. References