Comprehensive analysis of the lncRNA-associated ceRNA network identifies neuroinflammation biomarkers for Alzheimer's disease.

ABSTRACT

Accumulating evidence has highlighted the important roles of long non-coding RNAs (lncRNAs) acting as competing endogenous RNAs (ceRNAs) in Alzheimer's disease (AD). In this study, we constructed an AD-derived lncRNA-associated ceRNA network (LncACeNET) based on the ceRNA hypothesis and co-expressed correlation analysis of RNAs (miRNAs, mRNAs and lncRNAs) from AD patients. Based on this network, we preliminarily identified new potential AD biomarkers including hsa-miR-155-5p, CERS6-AS1, and CTB-89H12.4. The functional enrichment analysis demonstrated that these inferred biomarkers were significantly correlated with AD-related biological processes such as neuron projection development and neuron projection morphogenesis. Notably, lncRNA CTB-89H12.4 is significantly associated with "calcium ion-regulated exocytosis of neurotransmitter", "chemical synaptic transmission", "presynaptic membrane assembly", "receptor localization to synapse", and "learning". This indicates the important role of CTB-89H12.4 as a promising target for AD therapy. Subsequently, we used the computational pipeline DTINet and discovered 19 lines of probable therapeutic relationships between FDA-approved drugs and CTB-89H12.4, which offered a new avenue to repurpose existing FDA-approved drugs for AD indication. Our study provides a new landscape for LncACeNET in AD, and will benefit mechanism study and new drug development for AD.