Transcription‐based drug repurposing in Alzheimer disease psychosis

Abstract

AbstractBackgroundThere are at least 50 million people with Alzheimer’s Disease (AD) or other dementias worldwide. About 50% of people with AD may develop psychotic symptoms (delusions and hallucinations). However, there is no effective pharmacological therapy for psychosis in AD and all current treatments are based on antipsychotics originally designed for other disorders like schizophrenia. In this study, we identify the molecular signature of psychosis in AD in post‐mortem brain samples and then use the connectivity map (CMAP) and library of network‐based cellular signatures (LINC) to predict candidate drugs based on the disease transcriptional signature.MethodWe generated RNA‐Seq data (Illumina Stranded mRNA Prep) and compared the gene expression levels across 4 groups: Control, AD with no psychosis (AD), AD with Delusions (ADD), and AD with both Delusions and Hallucinations (ADDH). The final identified differentially expressed up and down‐regulated genes (DEGs) were used as query signatures in the CMAP and LINC databases to identify drugs that could potentially reverse the gene expression signatures in disease samples. To investigate the disease mechanisms and the effect of repurposed drugs on it, a drug‐target interaction network was reconstructed and combined with the human interactome. Enrichment analysis of this network identifies AD and psychosis‐related pathways and the potential effect the nominated drugs have on them.ResultAfter filtering out the common DEGs, 357, 171, and 66 unique genes remained in the AD, ADD, and ADDH signature sets, respectively. The signatureSearch package was used to search transcriptome signatures against CMAP and LINC. The top ranked drugs, based on negative connectivity scores, were selected as final results. Gene set enrichment analysis of the drug target genes presented some key pathways and activities related to psychiatric disorders. For example, serotonin receptor signalling pathway is one of the most over‐represented pathway of target genes in both ADD and ADDH groups.ConclusionThis study can improve our understanding of the molecular mechanisms underlying AD psychosis. The candidate drugs we have identified could be repurposed for clinical trials of psychosis in AD and provide valuable knowledge for future pharmacological research.