Competing endogenous RNA network profiling reveals novel host dependency factors required for MERS-CoV propagation

Circular RNAs (circRNAs) are an integral component of the host competitive endogenous RNA (ceRNA) network. These noncoding RNAs are characterized by their unique splicing reactions to form covalently closed loop structures and play important RNA regulatory roles in cells. Recent studies showed that circRNA expressions were perturbed in viral infections and circRNAs might serve as potential antiviral targets. We investigated the host ceRNA network changes and biological relevance of circRNAs in human lung adenocarcinoma epithelial (Calu-3) cells infected with the highly pathogenic Middle East respiratory syndrome coronavirus (MERS-CoV). A total of >= 49337 putative circRNAs were predicted. Among the 7845 genes which generated putative circRNAs, 147 (1.9%) of them each generated >= 30 putative circRNAs and were involved in various biological, cellular, and metabolic processes, including viral infections. Differential expression (DE) analysis showed that the proportion of DE circRNAs significantly (P < 0.001) increased at 24 h-post infection. These DE circRNAs were clustered into 4 groups according to their time-course expression patterns and demonstrated inter-cluster and intra-cluster variations in the predicted functions of their host genes. Our comprehensive circRNA-miRNA-mRNA network identified 7 key DE circRNAs involved in various biological processes upon MERS-CoV infection. Specific siRNA knockdown of two selected DE circRNAs (circFNDC3B and circCNOT1) significantly reduced MERS-CoV load and their target mRNA expression which modulates various biological pathways, including the mitogen-activated protein kinase (MAPK) and ubiquitination pathways. These results provided novel insights into the ceRNA network perturbations, biological relevance of circRNAs, and potential host-targeting antiviral strategies for MERS-CoV infection.