Integrative Analysis Extracts a Core ceRNA Network of the Fetal Hippocampus With Down Syndrome

Abstract

Accumulating evidence suggests that circular RNAs (circRNAs)—miRNA–mRNA ceRNA regulatory network—may play an important role in neurological disorders, such as Alzheimer’s disease (AD). Interestingly, neuropathological changes that closely resemble AD have been found in nearly all Down syndrome (DS) cases > 35 years. However, few studies have reported circRNA transcriptional profiling in DS cases, which is caused by a chromosomal aberration of trisomy 21. Here, we characterized the expression profiles of circRNAs in the fetal hippocampus of DS patients (n = 8) and controls (n = 6) by using microarray. MiRNA, mRNA expression profiling of DS from our previous study and scRNA-seq data describing normal fetal hippocampus development (GEO) were also integrated into the analysis. The similarity between circRNAs/genes with traits/cell-types was calculated by weighted correlation network analysis (WGCNA). miRanda and miRWalk2 were used to predict ceRNA network interactions. We identified a total of 7,078 significantly differentially expressed (DE) circRNAs, including 2,637 upregulated and 4,441 downregulated genes, respectively. WGCNA obtained 15 hub circRNAs and 6 modules with cell type–specific expression patterns among scRNA-seq data. Finally, a core ceRNA network was constructed by 14 hub circRNAs, 17 DE miRNA targets and 245 DE mRNA targets with a cell type–specific expression pattern annotation. Known functional molecules in DS or neurodegeneration (e.g., miR-138, OLIG1, and TPM2) were also included in this network. Our findings are the first to delineate the landscape of circRNAs in DS and the first to effectively integrate ceRNA regulation with scRNA-seq data. These data may provide a valuable resource for further research on the molecular mechanisms or therapeutic targets underlying DS neuropathy.