Respiratory syncytial virus modifies microRNAs regulating host genes that affect virus replication

Abstract

Respiratory syncytial virus (RSV) causes substantial morbidity and life-threatening lower respiratory tract disease in infants, young children and the elderly. Understanding the host response to RSV infection is critical for developing disease-intervention approaches. The role of microRNAs (miRNAs) in post-transcriptional regulation of host genes responding to RSV infection is not well understood. In this study, it was shown that RSV infection of a human alveolar epithelial cell line (A549) induced five miRNAs (let-7f, miR-24, miR-337-3p, miR-26b and miR-520a-5p) and repressed two miRNAs (miR-198 and miR-595), and showed that RSV G protein triggered let-7f expression. Luciferase-untranslated region reporters and miRNA mimics and inhibitors validated the predicted targets, which included cell-cycle genes (CCND1, DYRK2 and ELF4), a chemokine gene (CCL7) and the suppressor of cytokine signalling 3 gene (SOCS3). Modulating let-7 family miRNA levels with miRNA mimics and inhibitors affected RSV replication, indicating that RSV modulates host miRNA expression to affect the outcome of the antiviral host response, and this was mediated in part through RSV G protein expression.

Fig. 1.

6340WT infection deregulates host miRNA expression. A549 cells were infected or mock infected with RSV 6340WT virus at an m.o.i. of 1 for 24 h. The data represent the mean qPCR fold change±sem of let-7f (let-7f), miR-337-3p (miR-337), miR-520a-5p (miR-520a), miR-24, miR-26b, miR-198 and miR-595 from three independent experiments relative to mock-infected cells, with values >1.0 considered to be upregulation and values below 1.0 considered to be downregulation.

Fig. 2.

RSV G protein regulates miRNA expression during infection. (a) Expression of let-7f, miR-24, miR-520a, miR-26b and miR-337 was measured in A549 cells infected with 6340WT or 6340ΔG virus at 24 h p.i. (b) Expression of let-7f in cells treated with purified RSV G protein (1.0 µg ml⁻¹) or RSV F protein (1.0 µg ml⁻¹) and at 24 h after infection with 6340WT or 6340ΔG virus. Data represent mean fold changes in copy number±sem from three independent experiments relative to mock-infected cells. Statistical significance is indicated: ***P<0.001, **P<0.01, *P<0.05.

Fig. 3.

Luciferase (Luc)–UTR assays used to validate predicted let-7f gene targets. RSV G protein induced let-7f and other miRNAs regulate multiple genes during RSV infection. (a) Venn diagram depicting the overlap between predicted let-7f gene targets and genes deregulated during RSV infection. Genes that were examined further are shown in bold. SOCS3, Suppressor of cytokine signalling 3; CCND1, cyclin D1; SMOX, spermine oxidase; HOXA1, homeobox A1 transcription factor; TNFAIP3, tumour necrosis factor α-induced protein 3; ELF4, E74-like factor 4; DYRK2, dual-specificity tyrosine phosphorylation regulated kinase 2; CCL7, chemokine (C-C motif) ligand 7; PLAUR, plasminogen activator, urokinase receptor; VLDLR, very low density lipoprotein receptor; GLRX3, glutaredoxin 3; SERPING, serpin peptidase inhibitor, clade G. The dashed line indicates the baseline value. (b) Seed sequence (nt 2–8) conservation (boxed area) among miRNAs of the let-7 family. (c) Sequence of let-7f and miR-24 inhibitor and mimic sequences. The nature of chemical modifications (N_0–16) on inhibitors and mimics are proprietary and not known. (d) Sequence alignment of various gene 3′UTRs and let-7f and miR-24. Numbers correspond to nucleotides in the 3′UTR. (e) let-7f regulates multiple genes during RSV infection. Luc–3′UTRs of putative let-7f targets were co-transfected into A549 cells with pSEAP2-Control (transfection control) and inhibitors or mimics for let-7f and/or miR-24. Data represent mean fold change±sem in Luc values [measured in relative light units (RLU)] from three independent experiments between inhibitor- and mimic-transfected cells relative to a non-target control (NTC) inhibitor or mimic. Statistical significance is indicated for all transfections represented in (e) and (f): ***P<0.001; **P<0.01; *P<0.05. (f) Cooperative activity of let-7f and miR-24 on DYRK2–Luc expression. A549 cells were transfected with DYRK2-pMLC plasmid and let-7f /miR-24 inhibitor/mimic alone or with DYRK2-pMLC plasmid and equimolar concentrations of let-7f+miR-24 inhibitor/mimic together with pSEAP2-Control plasmid as a transfection control. Data represent the fold change in Luc expression±sem from two independent experiments with the dashed line indicating the baseline value.

Fig. 4.

RISC complexes from RSV 6340WT-infected cells are enriched for CCND1 and let-7f transcripts. (a) RISC-associated RNA from mock-, 6340WT- and 6340ΔG-infected cells were assayed for CCND1 by PCR. CCND1 UTR amplicons (0.6 kb) were amplified as described in Methods in two independent experiments. 18S rRNA was used as a loading control. RAgo, Anti-Ago2-precipitated RNA from RSV-infected cells; RBU, anti-BrdU-precipitated RNA from RSV-infected cells; VAgo, mock-infected Vero cell RNA precipitated with anti-Ago2; VBU, mock-infected Vero cell RNA precipitated anti-BrdU. (b) Enrichment of let-7f in RISC immunoprecipitated RNA from 6340WT- and RSVΔG-infected cells was assayed by qPCR and normalized to that of mock-infected cells from two independent experiments. Results are shown as means±sem, and Student’s t-test was used to measure the statistical significance of the data: ***P<0.001.

Fig. 5.

Modulation of miRNA levels deregulates virus replication. A549 cells were mock transfected or transfected in two independent experiments with inhibitors of let-7f and miR-24 separately and together (let-7f+miR-24), followed by infection with rgRSV at an m.o.i. of 0.5. The number of RSV p.f.u. was measured at day 3 p.i. using an anti-RSV F-based plaque assay relative to mock-infected cells. siRNA against the RSV N gene (siRSV) was used as a silencing control. The dashed line indicates basal p.f.u. levels in the mock-transfected control.