Host microRNAs exhibit differential propensity to interact with SARS-CoV-2 and variants of concern

Abstract

A significant number of SARS-CoV-2-infected individuals naturally overcome viral infection, suggesting the existence of a potent endogenous antiviral mechanism. As an innate defense mechanism, microRNA (miRNA) pathways in mammals have evolved to restrict viruses, besides regulating endogenous mRNAs. In this study, we systematically examined the complete repertoire of human miRNAs for potential binding sites on SARS-CoV-2 Wuhan-Hu-1, Beta, Delta, and Omicron. Human miRNA and viral genome interaction were analyzed using RNAhybrid 2.2 with stringent parameters to identify highly bonafide miRNA targets. Using publicly available data, we filtered for miRNAs expressed in lung epithelial cells/tissue and oral keratinocytes, concentrating on the miRNAs that target SARS-CoV-2 S protein mRNAs. Our results show a significant loss of human miRNA and SARS-CoV-2 interactions in Omicron (130 miRNAs) compared to Wuhan-Hu-1 (271 miRNAs), Beta (279 miRNAs), and Delta (275 miRNAs). In particular, hsa-miR-3150b-3p and hsa-miR-4784 show binding affinity for S protein of Wuhan strain but not Beta, Delta, and Omicron. Loss of miRNA binding sites on N protein was also observed for Omicron. Through Ingenuity Pathway Analysis (IPA), we examined the experimentally validated and highly predicted functional role of these miRNAs. We found that hsa-miR-3150b-3p and hsa-miR-4784 have several experimentally validated or highly predicted target genes in the Toll-like receptor, IL-17, Th1, Th2, interferon, and coronavirus pathogenesis pathways. Focusing on the coronavirus pathogenesis pathway, we found that hsa-miR-3150b-3p and hsa-miR-4784 are highly predicted to target MAPK13. Exploring miRNAs to manipulate viral genome/gene expression can provide a promising strategy with successful outcomes by targeting specific VOCs.

Keywords: Antiviral immunity; MicroRNAs; Post-transcriptional regulation; SARS-CoV-2; Variants of concern.

Fig. 1

In silico prediction of host miRNA and SARS-CoV-2 interactions. (a) Computational pipeline used to predict human miRNAs that uniquely target SARS-CoV-2 S gene of Wuhan-Hu-1, Beta (B.1.351), Delta (B.1.617.2), and Omicron (BA.1). All 2636 mature human miRNAs (miRbase v. 22.1) were aligned against the whole genomes of each SARS-CoV-2 strain (Wuhan-Hu-1, GenBank: NC_045512.2; Beta, GenBank: MW981442.1); Delta, GenBank: MZ359841.1; Omicron, GISAID: EPI_ISL_6640916) via RNAhybrid 2.2. Target sequences identified by RNAhybrid 2.2 were filtered to select miRNAs (1) specifically target SARS-CoV-2 S gene. p-Value = nothing (helix constraint and p-values cannot be used simultaneously), (2) expressed in human oral keratinocytes and (3) expressed in Calu3 human lung epithelial cell lines, alveolar epithelial primary cells, and normal lung tissue. (b) Schematic of representative Spike protein mutations in Beta (B.1.351), Delta (B.1.617.2), and Omicron (BA.1) SARS-CoV-2 variants. Key mutations are noted in red. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Host miRNAs targeting SARS-CoV-2 Wuhan-Hu-1, Beta (B.1.351), Delta (B.1.617.2), and Omicron (BA.1) spike and expressed in oral epithelial cells, lung epithelial cells, and lung tissue. (a) 28 representative host miRNAs with lowest minimum free energy (mfe) values common to at least two SARS-CoV-2 strains (274 in total; * not shown here; refer to Supplementary Fig. 2). (b) Venn diagram showing commonly and differentially targeting host miRNAs. Host miRNAs targeting SARS-CoV-2 Wuhan-Hu-1, Beta (B.1.351), Delta (B.1.617.2), and Omicron (BA.1) spike and expressed in oral epithelial cells, lung epithelial cells, and lung tissue. (a) 28 representative host miRNAs with lowest minimum free energy (mfe) values common to at least two SARS-CoV-2 strains (274 in total; * not shown here; refer to Supplementary Fig. 2). (b) Venn diagram showing commonly and differentially targeting host miRNAs.

Fig. 3

Representative mutations resulting in differential targeting of SARS-CoV-2 spike by host miRNAs. (a) Regions of mutations resulting in the binding of uniquely specific host miRNAs (noted in red) to Beta, Delta, and Omicron. Host miRNAs lost in the variants (but bind to Wuhan-Hu-1) due to these mutations are noted in black. x = no miRNA interaction in the region of mutation. (b) Representative multiple sequence alignment analysis of Wuhan-Hu-1 (GenBank: NC_045512.2), Beta (GenBank: MW981442.1), Delta (GenBank: MZ359841.1), and Omicron (GISAID: EPI_ISL_6640916; hCov-19/Botswana) spike. Nucleobase substitutions and deletions are in red. Relative to Wuhan-Hu-1, specific mutations resulting in the addition of novel target binding sites or alter an existing target binding sites are numbered from I to VIII. (c) Omicron N-protein mutations resulting in the loss of miRNA binding sites previously present in Wuhan-Hu-1. x = no miRNA interaction in the region of mutation. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Coronavirus pathogenesis pathway target genes of hsa-miR-326, hsa-miR-6796-5p, hsa-miR-380-5p, hsa-miR-3150b-3p, and hsa-miR-4784. These host miRNAs with predicted binding affinity to Wuhan-Hu-1 are substituted by miRNAs uniquely specific to Beta, Delta, and Omicron. We used IPA's miRNA Target Filter Analysis to analyze whether these substitutions result in the loss of key miRNA-mediated regulation of coronavirus pathogenesis pathways. To narrow down the potential mRNA targets, we focused on experimentally validated and highly predicted target genes of hsa-miR-326, hsa-miR-6796-5p, hsa-miR-380-5p, hsa-miR-3150b-3p, and hsa-miR-4784 in the coronavirus pathogenesis pathway. Experimentally validated and highly predicted target genes are highlighted in orange. Moderately predicted target genes are highlighted in blue. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Supplementary Fig. 1

Search pipeline used during miRna target filter analysis (IPA, QIAGEN) of miRnas with binding to Wuhan-Hu-1 but displaced due to VOC mutations.

Supplementary Fig. 2

Whole genome phylogenetic analysis of SARS-CoV-2 VOCs. Clade 1 (20C): Beta. Clade 2 (20B): Alpha, Omicron, and Gamma. Clade 3 (21A) Delta.