Functional Mapping of AGO-Associated Zika Virus-Derived Small Interfering RNAs in Neural Stem Cells

Abstract

Viral interfering RNA (viRNA) has been identified from several viral genomes via directly deep RNA sequencing of the virus-infected cells, including zika virus (ZIKV). Once produced by endoribonuclease Dicer, viRNAs are loaded onto the Argonaute (AGO) family proteins of the RNA-induced silencing complexes (RISCs) to pair with their RNA targets and initiate the cleavage of target genes. However, the identities of functional ZIKV viRNAs and their viral RNA targets remain largely unknown. Our recent study has shown that ZIKV capsid protein interacted with Dicer and antagonized its endoribonuclease activity, which requires its histidine residue at the 41^st amino acid. Accordingly, the engineered ZIKV-H41R loss-of-function (LOF) mutant virus no longer suppresses Dicer enzymatic activity nor inhibits miRNA biogenesis in NSCs. By combining AGO-associated RNA sequencing, deep sequencing analysis in ZIKV-infected human neural stem cells (NSCs), and miRanda target scanning, we defined 29 ZIKV derived viRNA profiles in NSCs, and established a complex interaction network between the viRNAs and their viral targets. More importantly, we found that viRNA production from the ZIKV mRNA is dependent on Dicer function and is a limiting factor for ZIKV virulence in NSCs. As a result, much higher levels of viRNAs generated from the ZIKV-H41R virus-infected NSCs. Therefore, our mapping of viRNAs to their RNA targets paves a way to further investigate how viRNAs play the role in anti-viral mechanisms, and perhaps other unknown biological functions.

Keywords: Dicer; RNA interference (RNAi); neural stem cells (NSCs); viral interfering RNA (viRNA); zika virus (ZIKV).

Figure 1

viRNAs are identified via AGO-associated RNA sequencing in ZIKV-WT-infected NSCs. (A) Schematic diagram for AGO-associated RNA sequencing. (B) Size distribution of viRNAs in AGO-associated RNA sequencing in ZIKV-WT-infected NSCs (MOI:1, 48 h post-infection). Red, positive-stranded viRNAs; blue, negative-stranded viRNAs. (C) The distribution and relative abundances of viRNA reads (21~23 nt) from AGO-associated RNA sequencing on the ZIKV genome (Red) and complementary sequence (blue). (D) Read sequence along 5’UTR region of ZIKV genome. Read counts (in brackets), read length, and genomic position are indicated.

Figure 2

Mapping ZIKV viRNAs-viral RNA interactome via miRanda target scanning. (A–D) Circos plot of long-range interaction between viRNA from ZIKV positive (+)- or negative (-)-strand and its predicted viral targets from viral positive (+)- or negative (-)-strand by miRanda (omictools.com/miRanda-tool) in ZIKV (SPH2015) genome. In details, (A) (31 pairings): viRNA(+)-target (+); (B) (four pairings): viRNA(-)-target(+); (C) (76 pairings): viRNA(+)-target(-); (D) (three pairings): viRNA(-)-target(-); (E) Paring between representative viRNAs identified and its viral targets predicted by miRanda.

Figure 3

The identified viRNAs are potentially the products of Dicer activity via directly deep small RNA-seq of ZIKV-infected NSCs. (A) Schematic diagram of comparative analysis for directly deep small RNA-seq in ZIKV-WT- or ZIKV-H41R-infected NSCs. (B) Comparison of read counts in 29 ZIKV viRNA peaks in ZIKV-WT- or ZIKV-H41R-infected NSCs at MOI of 1 for 2 days. The fold change of presented peaks was log2 scale. (C) 90% viRNA peaks identified from AGO-associated RNA sequencing are increased in ZIKV-H41R infected NSCs compared to ZIKV-WT infected ones. (D) Peak landscape of representative viRNA-p18 in AGO-associated RNA-seq and deep small RNA-seq. Value in the small RNA-seq (lower panel) in y-axis indicates reads per 10 million sequenced reads.

Figure 4

Dicer-dependent viRNAs are potentially anti-viral in ZIKV-infected NSCs. (A) Taqman Advanced miRNA assays for miRNA let-7a in NSCs infected with ZIKV-WT or ZIKV-H41R. Mean ± SD; **p < 0.01 by one-way ANOVA and Bonferroni’s post hoc test. n.s., not significant. (B) Customed Taqman Advanced assays for viRNA-p18 (viRNA^p18) detection in NSCs infected with ZIKV-WT or ZIKV-H41R. Mean ± SD; ***p < 0.001 by one-way ANOVA and Bonferroni’s post hoc test. (C) NSCs were infection with ZIKV-WT or ZIKV-H41R mutant virus at MOI of 0.01, and culture supernatant were collected for virus titer determined by plaque assay. (D) Schematic diagram for the involvement of RNAi machinery components in miRNA or viRNA biogenesis. (E) Taqman Advanced miRNA assays for mature miRNA let-7a in scramble, Drosha, or Dicer siRNA-treated NSCs. Mean ± SD; ***p < 0.001 by Student’s t-test. (F) viRNA generation is dependent on Dicer function. 5x10⁵ NSCs transfected with si.scramble, si.Drosha, or si.Dicer for 24 h were infected with ZIKV-WT or ZIKV-H41R at MOI of 0.1, and representative viRNA^p18 expression was detected by customed TaqMan Advanced assays (see Materials and Methods). Mean ± SD; ***p < 0.001 by one-way ANOVA and Bonferroni’s post hoc test. n.s., not significant. (G) Both miRNA and viRNA play important role to limit ZIKV replication in NSCs. 5 x10⁵ NSCs were individually transfected with si.scramble, si.Dicer, or si.Drosha for 24 h, followed by infection with ZIKV-WT or ZIKV-H41R mutant virus at MOI of 0.01, and culture supernatant were collected for virus titer determined by plaque assay. Mean ± SD; **p < 0.01 and ***p < 0.001 by Student’s t-test. n.s., not significant.