Identification of biomarkers and pathways to identify novel therapeutic targets in Alzheimer’s disease: Insights from a systems biomedicine perspective

Abstract Alzheimer’s disease (AD) is a progressive neurodegenerative disease characterized by the accumulation of amyloid plaques and neurofibrillary tangles in the brain. However, there are no peripheral biomarkers available that can detect AD onset. This study aimed to identify such AD biomarkers through an integrative analysis of blood cell gene expression data. We used two microarray datasets (GSE4226 and GSE4229) comparing blood cell transcriptomes of AD patients and matched controls to identify differentially expressed genes (DEGs). Geneset and protein overrepresentation analysis, protein-protein interaction (PPI), DEGs-Transcription Factor interactions, DEGs-MicroRNAs interactions, protein-drug interactions, and protein subcellular localizations analyses were performed on DEGs common to the datasets. We identified 25 common DEGs between the two datasets. Integration of genome scale transcriptome datasets with biomolecular networks revealed hub genes (TUBB, ATF3, NOL6, UQCRC1, SND1, CASP2, BTF3, INPP5K, VCAM1, and CSTF1), common transcription factors (FOXC1, ZNF3, GEMIN7, and SMG9) and miRNAs (mir-20a-5p, mir-93-5p, mir-16-5p, let-7b-5p, mir-708-5p, mir-24-3p, mir-26b-5p, mir-17-5p, mir-4270, and mir-4441). These included 8 previously identified AD markers (BTF3, VCAM, FOXC1, mir-26b-5p, mir-20a-5p, mir-93-5p, let-7b-5p, mir-24-3p) and 13 novel AD biomarkers. Evaluation of histone modification revealed that hub genes and transcription factors possess several histone modification sites associated with AD. Protein-drug interactions revealed 10 compounds that affect the identified AD candidate markers, including anti-neoplastic agents (Vinorelbine, Vincristine, Vinblastine, Epothilone D, Epothilone B, CYT997, and ZEN-012), a dermatological (Podofilox) and an immunosuppressive agent (Colchicine). The subcellular localization of markers varied, including nuclear, plasma membrane and cytosolic proteins. This study identified blood-cell derived molecular biomarker signatures that might be useful as peripheral biomarkers in AD. This study also identified drug and epigenetic data associated with these molecules that may be useful in designing therapeutic approaches to ameliorate AD.