The Predicted Key Molecules, Functions, and Pathways That Bridge Mild Cognitive Impairment (MCI) and Alzheimer's Disease (AD)

Abstract

To elucidate the key molecules, functions, and pathways that bridge mild cognitive impairment (MCI) and Alzheimer's disease (AD), we investigated open gene expression data sets. Differential gene expression profiles were analyzed and combined with potential MCI- and AD-related gene expression profiles in public databases. Then, weighted gene co-expression network analysis was performed to identify the gene co-expression modules. One module was significantly negatively associated with MCI samples, in which gene ontology function and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis showed that these genes were related to cytosolic ribosome, ribosomal structure, oxidative phosphorylation, AD, and metabolic pathway. The other two modules correlated significantly with AD samples, in which functional and pathway enrichment analysis revealed strong relationships of these genes with cytoplasmic ribosome, protein binding, AD, cancer, and apoptosis. In addition, we regarded the core genes in the module network closely related to MCI and AD as bridge genes and submitted them to protein interaction network analysis to screen for major pathogenic genes according to the connectivity information. Among them, small nuclear ribonucleoprotein D2 polypeptide (SNRPD2), ribosomal protein S3a (RPS3A), S100 calcium binding protein A8 (S100A8), small nuclear ribonucleoprotein polypeptide G (SNRPG), U6 snRNA-associated Sm-like protein LSm3 (LSM3), ribosomal protein S27a (RPS27A), and ATP synthase F1 subunit gamma (ATP5C1) were not only major pathogenic genes of MCI, but also bridge genes. In addition, SNRPD2, RPS3A, S100A8, SNRPG, LSM3, thioredoxin (TXN), proteasome 20S subunit alpha 4 (PSMA4), annexin A1 (ANXA1), DnaJ heat shock protein family member A1 (DNAJA1), and prefoldin subunit 5 (PFDN5) were not only major pathogenic genes of AD, but also bridge genes. Next, we screened for differentially expressed microRNAs (miRNAs) to predict the miRNAs and transcription factors related the MCI and AD modules, respectively. The significance score of miRNAs in each module was calculated using a hypergeometric test to obtain the miRNApivot-Module interaction pair. Thirty-four bridge regulators were analyzed, among which hsa-miR-519d-3p was recognized as the bridge regulator between MCI and AD. Our study contributed to a better understanding of the pathogenic mechanisms of MCI and AD, and might lead to the development of a new strategy for clinical diagnosis and treatment.

Keywords: AD; GO; KEGG; MCI; PPI; WGCNA; microRNA; transcription factor.