High-Throughput Fluorescence-Based Screen Identifies the Neuronal MicroRNA miR-124 as a Positive Regulator of Alphavirus Infection

Arthropod-borne (arbo) viruses are part of a class of pathogens that are transmitted to their final hosts by insects. Because of climate change, the habitat of some of these insects, such as mosquitoes, is shifting, thereby facilitating the emergence of viral epidemics. Among the pathologies associated with arbovirus infection, neurological diseases such as meningitis and encephalitis represent a significant health burden. Using a genome-wide miRNA screen, we identified neuronal miR-124 as a positive regulator of the Sindbis and chikungunya alphaviruses. We also showed that this effect was in part direct, thereby opening novel avenues to treat alphavirus infections.

ABSTRACT

MicroRNAs (miRNAs) are small regulatory RNAs which act by modulating the expression of target genes. In addition to their role in maintaining essential physiological functions in the cell, miRNAs can also regulate viral infections. They can do so directly by targeting RNAs of viral origin or indirectly by targeting host mRNAs, and this can result in a positive or negative outcome for the virus. Here, we performed a fluorescence-based miRNA genome-wide screen in order to identify cellular miRNAs involved in the regulation of arbovirus infection in human cells. We identified 16 miRNAs showing a positive effect on Sindbis virus (SINV) expressing green fluorescent protein (GFP), among which were a number of neuron-specific ones such as miR-124. We confirmed that overexpression of miR-124 increases both SINV structural protein translation and viral production and that this effect is mediated by its seed sequence. We further demonstrated that the SINV genome possesses a binding site for miR-124. Both inhibition of miR-124 and silent mutations to disrupt this binding site in the viral RNA abolished positive regulation. We also proved that miR-124 inhibition reduces SINV infection in human differentiated neuronal cells. Finally, we showed that the proviral effect of miR-124 is conserved in other alphaviruses, as its inhibition reduces chikungunya virus (CHIKV) production in human cells. Altogether, our work expands the panel of positive regulation of the viral cycle by direct binding of host miRNAs to the viral RNA and provides new insights into the role of cellular miRNAs as regulators of alphavirus infection.

IMPORTANCE Arthropod-borne (arbo) viruses are part of a class of pathogens that are transmitted to their final hosts by insects. Because of climate change, the habitat of some of these insects, such as mosquitoes, is shifting, thereby facilitating the emergence of viral epidemics. Among the pathologies associated with arbovirus infection, neurological diseases such as meningitis and encephalitis represent a significant health burden. Using a genome-wide miRNA screen, we identified neuronal miR-124 as a positive regulator of the Sindbis and chikungunya alphaviruses. We also showed that this effect was in part direct, thereby opening novel avenues to treat alphavirus infections.

INTRODUCTION

Infectious diseases, and among them viral diseases, remain a leading cause of morbidity and mortality worldwide. In addition to the direct consequences of viral infections, many health disorders are indirectly linked to viruses. Although vaccines are available for some viruses, this is not the case for a large number of them, and it is essential to find novel antiviral compounds to fight them. Alphaviruses, from the Togaviridae family, are arthropod-borne viruses (arboviruses) transmitted to vertebrates by a mosquito vector and form a group of widely distributed human and animal pathogens. They are small, enveloped, positive single-stranded RNA viruses. Their RNA genome of ∼11 kb is capped and polyadenylated. It has two open reading frames (ORFs) encoding nonstructural and structural proteins. ORF2 is expressed through the production of a subgenomic RNA from an internal promoter in the minus-strand RNA replication intermediate. In addition to protein-coding sequences, alphavirus RNAs contain important regulatory structures, such as the 5′ and 3′ untranscribed regions (UTRs) (1).

Alphaviruses represent an emerging public health threat, as they can induce febrile and arthritogenic diseases, as well as other highly debilitating diseases such as encephalitis (2). Sindbis virus (SINV) is considered the prototypical alphavirus and is widely used as a laboratory model. Although the infection has been mainly associated with a rash, arthritis, and myalgia in humans (3), SINV displays a neuronal tropism in developing rodent brain cells and is associated with encephalomyelitis (4). Another virus from the same genus is chikungunya virus (CHIKV), which causes outbreaks of severe acute and chronic rheumatic diseases in humans (5). CHIKV has also been reported to affect the human nervous system, causing encephalopathy in newborns, infants, and adults (6). Due to its ability to quickly spread into new regions, CHIKV is classified as an emergent virus (7, 8) for which preventive or curative antiviral strategies are needed.

MicroRNAs (miRNAs) are small, 22-nucleotide-long noncoding RNAs which act as guides for effector proteins to posttranscriptionally regulate the expression of target cellular mRNAs (9) but also viral RNAs (10–17). These small RNAs have been identified in almost all eukaryotic species, and a number of them are conserved throughout evolution (18). They derive from longer precursors, which are transcribed by RNA polymerase II, and are sequentially processed by the RNase III enzymes Drosha and Dicer. The mature miRNA is then assembled in a protein of the Argonaute family to guide it to target RNAs. Once bound to its target, the Argonaute protein regulates its expression by recruiting proteins to inhibit translation initiation and induce its destabilization by deadenylation (19). The main determinant of miRNA sequence specificity is its seed sequence, which corresponds to a short region at the 5′ end of miRNAs (nucleotides 2 to 7) (20). Perfect pairing of the miRNA seed with the target RNA represents the minimal requirement for efficient Argonaute binding and function. In some cases, additional base pairing toward the 3′ end of the mature miRNA (so-called 3′ compensatory sites) may compensate for suboptimal pairing in the seed region (21, 22).

Identifying miRNAs that alter virus replication has illuminated roles for these molecules in virus replication and highlighted therapeutic opportunities. Target predictions based on the concept of “seed” initially identified binding sites for liver-specific miR-122 in the 5′ UTR of hepatitis C virus (HCV), which turned out to be positively regulated by this miRNA (23). Further work from different teams later showed that miR-122 can positively regulate the virus by increasing the stability and translation of the viral RNA (24, 25). Interestingly, the use of inhibitors of miR-122 in HCV infection is currently in a clinical trial as an miRNA-based antiviral therapy (26).

Here, we performed a genome-wide miRNA overexpression and inhibition screen to identify cellular miRNAs involved in the regulation of SINV infection. We found 16 miRNAs with a positive effect on virus accumulation. Among them, miR-124-3p was shown to positively regulate the virus by increasing viral structural protein translation and viral production, and we identified a binding site for this small RNA in the viral genome. miR-124 is the most expressed, conserved, and specific microRNA in the central nervous system (CNS), and it has been described as a master regulator of neuronal differentiation (27). It is upregulated as neuronal progenitors exit mitosis and begin to differentiate (28), and its expression has been shown to be sufficient to drive cells toward the neuronal pathway (29). Mutations in the miR-124 seed region or in the binding site on the viral RNA abolished regulation. We also proved that the miR-124 inhibition strategy via antisense oligonucleotides reduces both SINV and CHIKV production in human cells expressing this miRNA. These studies highlight a novel role for miR-124 in arbovirus infection and suggest that targeting alphavirus infection via miRNA modulation could be used therapeutically.

RESULTS

Identification of miRNAs involved in the regulation of SINV infection by fluorescence-based screening.

To identify miRNAs that are involved in the regulation of SINV infection in a systematic and comprehensive way, we performed a fluorescence microscopy-based high-throughput screening using two libraries of ∼2,000 human miRNA mimics and ∼2,000 antimiRNA oligonucleotides (Fig. 1A). Control miRNA mimics or antimiRNAs with no sequence homology to any known human miRNA were used as negative controls. A small interfering RNA (siRNA) against the 3′ UTR of SINV, able to decrease green fluorescent protein (GFP) expression without inducing cell death, was used as a functional positive control (Fig. 1B). Huh7.5.1 cells were transfected with either the library of miRNA mimics or antimiRNAs, and 72 h later, the cells were infected with a SINV strain encoding the green fluorescent protein (GFP) (Fig. 1A). The number of GFP-positive cells and the GFP intensity were measured 24 h postinfection (hpi) by automated image analysis, and the robust strictly standardized mean difference (SSMD* or SSMDr) value was calculated for each miRNA to identify significant hits. While the antimiRNA screen did not reveal any significant hit and the mimic screen did not identify any miRNAs with a negative effect on GFP accumulation, we identified 16 miRNA mimics that increased the GFP signal significantly (SSMDr > 1.28) without affecting cell viability (Fig. 2A). All of the top hit miRNAs were undetectable or expressed at a very low level in Huh7.5.1 cells as assessed by small RNA sequencing (Fig. 2B and C), which explains why transfection of the corresponding antimiRNAs showed no effect on GFP levels. Among the top hits, miR-124-3p and miR-129-5p, which are known to be expressed in neuronal cells (30, 31), had the most striking effect (SSMDr > 2.20), suggesting a link between neuron-enriched miRNAs and positive regulation of the virus. To determine whether the positive effect of these miRNAs on SINV-GFP expression (Fig. 2D) implied an increase in viral production, we measured the effect of the overexpression of miR-124-3p and miR-129-5p on SINV titers compared to that for a negative miRNA mimic control (Fig. 2E). As observed from the measurement of GFP fluorescence, miR-124-3p and miR-129-5p also increased SINV-GFP viral titers in Huh7.5.1 cells, providing additional evidence for their involvement in the positive regulation of viral infection.

FIG 1.

Design of the phenotypic fluorescence-based screen and validation with siRNA directed against SINV-GFP. (A) Schematic representation of the screening protocol. Huh7.5.1 cells were reverse transfected with a complete library of mimics or antimiRNAs corresponding to all identified human miRNAs, and 72 h posttransfection, they were infected with SINV-GFP at a low multiplicity of infection (MOI of 10⁻³) for 24 h. The infection level was analyzed by automated microscopy analysis based on GFP fluorescence. (B, top) schematic representation of SINV-GFP genomic structure with the repeated region within the 3′ UTR of SINV (starting at positions 12345 and 12437) that is targeted by the siRNA SINV (siSINV) used as a functional control for the genome-wide screen. (Middle) Representative fluorescence microscopy images of SINV-GFP infected Huh7.5.1 cells after transfection with a control siRNA or with the siRNA directed against the virus (siSINV). (Bottom) Histograms represent the percentages of GFP-positive cells (left) and cell counts (right) to evaluate toxicity after negative-control siRNA (CTR1, CTR2, and CTR4) or siSINV transfection. The means ± standard errors of the means (SEMs) from three independent experiments are presented. ****, P < 0.0001, unpaired Student's t test.

FIG 2.

Identification of cellular miRNAs positively regulating SINV infection in Huh7.5.1 cells. (A) Dot plot representation of the strictly standardized mean difference (SSMDr) in fluorescence for each mimic or antimiRNA. Candidates with an absolute SSMDr equal to or higher than the threshold 1.28 were considered significant. In red, significant mimic candidates. (B) List of significant candidates identified in the mimic screen with their respective SSMDr and associated toxicity values. miRNA expression levels in Huh7.5.1 cells determined by small RNA sequencing (small RNA-seq) are indicated as reads per million miRNA reads (rpm). (C) Top 30 microRNAs expressed in Huh7.5.1 cells. Ranking was based on the reads normalized per million microRNA reads (rpm) in the small RNA-seq experiment. (D) Representative fluorescence images from the mimic screen of some of the top candidates compared to the controls (Neg2 and miR-137). (Left) GFP signals from infected cells; (right) merges of GFP signals and Hoechst staining of cell nuclei. Pictures taken at ×10 magnification. (E) Representative plaque assay image after crystal violet staining of SINV-infected Huh7.5.1 cells transfected with control or candidate mimics.

miR-124 positively regulates SINV by increasing viral structural protein synthesis and viral particle production in a seed-dependent manner.

Because miR-124 was the top hit of our screen, the most expressed miRNA in neuronal tissues, and one of the most conserved microRNAs in metazoans, we decided to focus our efforts on the characterization of the mechanism underlying its effect on the SINV cycle. We first assessed the impact of miR-124-3p expression during SINV binding and internalization, the early steps of the viral cycle. We thus performed a SINV attachment and entry assay in cells transfected with a control or miR-124-3p mimic. At 72 h posttransfection, we detected a decrease rather than an increase of SINV attachment (Fig. 3A, attachment). Moreover, miR-124-3p overexpression did not affect the efficiency of SINV entry into Huh7.5.1 cells compared to that of cells transfected with control mimic (Fig. 3A, entry). Collectively, these data suggest that the positive effect of miR-124-3p on SINV is not due to increased viral binding or internalization.

FIG 3.

Importance and mechanism of miR-124-3p and its seed in the positive regulation of SINV infection. (A) Huh7.5.1 cells transfected with mimic CTL or 124 for 72 h were incubated with SINV-GFP (MOI of 5) at 4°C for 30 min and washed with PBS three times to remove unbound SINV particles. Total RNA was extracted and used for quantification of SINV genomic RNA by RT-qPCR. Data are presented as means ± SEMs from three independent experiments normalized to GAPDH. In parallel, transfected cells were incubated with SINV-GFP (MOI of 5) at 4°C for 30 min, washed with PBS, and then transferred to 37°C for 30 min to allow virus entry. Total RNA was extracted and used for quantification of SINV genomic RNA by RT-qPCR. Data are presented as means ± SEMs from three independent experiments. (B) miR-124 accumulation after mimic transfection quantified by RT-qPCR and relative to control mimic and normalized to snRNA U6. (C) RT-qPCR analysis of CDK4 expression and SINV genomic and subgenomic RNA accumulation in mimic-124-transfected cells following SINV infection at an MOI of 10⁻³ for 24 h relative to mimic CTL transfection and normalized to GAPDH. (D) Strand-specific RT-PCR on SINV antigenome produced in Huh7.5.1 cells transfected with mimic-124 or CTL and infected with SINV-GFP at an MOI of 10⁻³ for 24 h. PCR products from 15, 20, 25, and 30 cycles are shown. (E) Western blot analysis of viral nsP2 and capsid protein synthesis in Huh7.5.1 transfected cells with mimic-124 or CTL and infected with SINV at an MOI of 10⁻³ for 24 h. Tubulin is used as a loading control. (F) Diagram of miRNA mimic-124 and mimic-124-mut (seed mutant). Seed sequence is indicated in blue and mutations in red. Western blot analysis of viral capsid protein accumulation (G) and plaque assay quantification of SINV-GFP viral titers (H) after transfection of control mimic, mimic-124, or mimic-124-mut in Huh7.5.1 cells and SINV infection at an MOI of 10⁻³ for 24 h. The means ± SEMs from three independent experiments are presented. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, nonsignificant, unpaired Student's t test.

We then analyzed the effect of miR-124-3p expression on SINV RNA replication by measuring viral genomic, subgenomic, and antigenomic RNA accumulation. Transfection of miR-124-3p mimic, which could readily be measured by reverse transcriptase quantitative PCR (RT-qPCR) (Fig. 3B), induced repression of a known miR-124 target, the cyclin-dependent kinase 4 (CDK4) (32). However, neither SINV genomic nor subgenomic RNA levels were significantly affected (Fig. 3C).

To better assess the miR-124 effect on SINV RNA replication, we performed semiquantitative strand-specific RT on the negative-strand antigenomic RNA. We observed that a similar amount of antigenome accumulated in Huh7.5.1 cells transfected with control or miR-124 mimic (Fig. 3D), suggesting that the positive effect of miR-124-3p on SINV is not due to an increased antigenome synthesis.

Finally, we analyzed the accumulation of either nonstructural (i.e., nsP2) or structural viral proteins (i.e., capsid) upon mimic transfection and SINV-GFP infection. The results showed an increase in the structural capsid protein but not in the nonstructural protein nsP2 upon mimic-124 transfection (Fig. 3E), strongly suggesting that positive regulation by miR-124 is due to an enhancement of viral structural proteins synthesis from the subgenomic RNA.

Among the top candidates showing a proviral effect in the screen (Fig. 2A and B), we found miR-506-3p, which belongs to the same family as miR-124 and shares similar physiological functions (33). This strongly suggests a possible implication of their seed region in the mechanism of viral regulation. In order to assess whether the proviral effect observed depended on the miRNA seed sequence, we transfected control mimics or mimics containing either the miR-124-3p wild-type seed sequence (mimic-124) or a miR-124-3p seed with three mutations at positions 4 to 6 (mimic-124mut) (Fig. 3F) in Huh7.5.1 cells before infecting them with SINV-GFP. Viral capsid protein production demonstrated that the positive effect of miR-124-3p expression was lost when cells were transfected with the seed mutant mimic (Fig. 3G). In addition, the same effect was observed on the viral titer quantification (Fig. 3H), strongly suggesting that miR-124-3p positively regulates SINV via its seed region, by enhancing viral structural protein translation and viral production.

Mutation of a miR-124-3p binding site within the SINV E1 coding region prevents miRNA-mediated regulation.

As miRNA-mediated viral regulation can happen by direct binding of the miRNA to the viral RNA (34), we bioinformatically searched for predicted miR-124-3p binding sites on the SINV genome. Using the prediction algorithm ViTa (35), we found three putative binding sites at positions 2454 to 2475, 3828 to 3853, and 10904 to 10926 in the viral genome. We further studied the latter, localized within the virus ORF2, and more specifically, in the glycoprotein E1 coding region (Fig. 4A), which possesses a canonical 6-nucleotide (nt) seed match (nucleotides 2 to 7) for miR-124-3p.

FIG 4.

Acquisition and conservation of miR-124-3p binding site in natural SINV isolates. (A) Schematic representation of SINV-GFP genomic structure with a diagram of miR-124-3p binding site within glycoprotein E1 coding region. (B) Alignment of 16 SINV complete genomic sequences from natural isolates corresponding to nucleotide positions 10905 to 10925 in the reference genome (GenBank accession no. NC_001547). Blue rectangle delimits predicted miR-124-3p binding site. Nonconserved nucleotides are highlighted in red. (C) Phylogenetic tree generated from NCBI virus sequence alignment of the entire genomic sequence of SINV strains in panel B. In blue, the SINV reference sequence; in red, the isolates without miR-124-3p binding site conservation. For each strain, the GenBank accession number, host, geographical region, and collection date are shown. Scale bar represents genetic distance.

Using available SINV sequences in the NCBI virus database (36), we sought to determine the conservation of this region among SINV strains. Sequence alignment of the entire genomic sequences of 16 SINV strains isolated worldwide from various insects and vertebrates between 1969 and 2016 compared to the reference SINV genome (NC_001547) revealed that most of the strains (14 of 16) share the same nucleotide composition at the binding site, independently of the host, the year, or the geographical areas in which they were isolated (Fig. 4B and C). Moreover, two of the most ancient and phylogenetically distant strains (MG182396 and KF981618) display silent point mutations in the miRNA binding site (Fig. 4B).

With the aim of disrupting the putative binding site without interfering with the viral coding sequence, we introduced two silent mutations at positions 10919 and 10922 (Fig. 5A) in the viral genome. We first characterized the mutant virus (SINV-GFP-U10919C-C10922A, referred to as SINV-GFP-mut) by an infection kinetic study compared to the wild-type SINV-GFP in Huh7.5.1 cells. SINV-GFP-mut growth kinetics showed no significant differences in viral titers compared to those of the wild-type SINV-GFP (Fig. 5B). Furthermore, we assessed the release and spread of the mutant virus by verifying the plaque size phenotype, and we observed no difference in plaque size between the two viruses (Fig. 5C). We then verified that miRNA-mediated regulation depends on the binding of this particular site. We transfected Huh7.5.1 cells with a control mimic or with different doses of a miR-124 mutant mimic (124comp) containing compensatory mutations within its seed (Fig. 5D) followed by infection with either SINV-GFP or SINV-GFP-mut. Viral titer quantification by plaque assay shows that mimic 124comp significantly enhanced SINV-GFP-mut but did not affect SINV-GFP (Fig. 5D).

FIG 5.

Mutation of one miR-124-3p binding site within SINV genomic RNA prevents regulation. (A) Schematic representation of SINV-GFP genomic structure with a diagram of miR-124-3p binding site within glycoprotein E1 coding region. The miRNA seed is indicated in blue; point mutations introduced in the mutant virus (SINV-GFP-mut) and in the compensatory mutant miR-124 mimic are indicated in red. (B) Plaque assay quantification of viral titers produced by SINV-GFP and SINV-GFP-mut on Huh7.5.1 cells at 4, 8, 12, or 24 hpi. Error bars represent means ± standard deviations (SDs). ns, nonsignificant, two-way analysis of variance (ANOVA) with Bonferroni’s correction. (C) Plaques for SINV-GFP and SINV-GFP-mut visualized by crystal violet staining on infected Vero cells to assess plaque size phenotype. (D) Differential fold change in viral titers produced by SINV-GFP and SINV-GFP-mut in Huh7.5.1 cells after transfection of control mimic or mimic-124comp (compensating the viral binding site mutation). (E) miR-124-3p expression in lentiviral transduced Huh7.5.1 cells stably expressing miR-124-3p (LV124), wild-type Huh7.5.1 cells, and SK-N-BE(2) neuroblastoma cells. (F) Northern blot analysis of miR-124-3p expression in LV124 after antimiRNA transfection. snRNA U6 is used as a loading control. (G) Fold differences in viral titers produced by SINV-GFP or SINV-GFP-mut in LV124 after antimiRNA transfection. (H) Western blot analysis of viral capsid protein accumulation in LV124 after antimiR transfection. Tubulin is used as a loading control. Error bars represent means ± SEMs from three independent experiments. **, P < 0.01; ns, nonsignificant, unpaired Student's t test.

In order to measure the effect of miR-124-3p inhibition on the wild-type (WT) and mutant SINV, we established a Huh7.5.1 cell line stably expressing the human mir-124-1 gene (LV124 cells). miR-124-3p levels were verified in LV124 cells and compared to the expression in differentiated human neuroblastoma SK-N-BE(2) cells, used as a positive control. Huh7.5.1 cells were used as a negative control. Northern blot analysis showed that miR-124 expression in the LV124 cells was similar to the expression level in differentiated neuroblastoma cells, known to express miR-124 (37) (Fig. 5E). Antisense oligonucleotides have been previously used as a strategy to sequester and block miRNAs (38). To test whether miR-124-3p inhibition reduced SINV infection, we transfected an antimiR-124 or a control antimiRNA into LV124 cells, which resulted in a complete depletion of the mature miRNA as assessed by Northern blotting (Fig. 5F). We then evaluated the effect of miR-124 inhibition in this setup on SINV-GFP and SINV-GFP-mut infection. After antimiR-124 transfection, SINV-GFP viral titers as well as the capsid viral protein synthesis were significantly reduced (Fig. 5G and H). In contrast, SINV-GFP-mut viral titers and protein levels remained unchanged upon miR-124-3p inhibition (Fig. 5G and H). These results show that inhibition of miR-124-3p negatively regulates SINV infection and that mutations introduced in the binding site on the viral RNA abolish the miRNA-mediated regulation. In addition, regulation of the mutant virus can be rescued by overexpression of a compensatory mutant mimic in a dose-dependent manner.

miR-124-3p inhibition in differentiated neuroblastoma cells restricts SINV-GFP infection.

Since miR-124-3p is a neuron-specific microRNA, we turned to a system where the miRNA was expressed endogenously to verify both the miRNA effect on SINV infection and the effect of virus infection on miRNA expression. The human neuroblastoma SK-N-BE(2) cells can be differentiated into neuron-like cells by retinoic acid (RA) treatment (39). We first verified neurite outgrowth, which is a morphological hallmark of neuroblastoma cell differentiation in vitro (Fig. 6A). To confirm the correct differentiation, we verified by RT-qPCR analysis that the expression of the neuron proliferation marker MYCN decreased after RA treatment (40) (Fig. 6B). miR-124-3p endogenous expression increased at 6 days postdifferentiation compared to that in proliferating cells but was not affected by SINV infection (Fig. 6C). We assessed the effect of miR-124-3p inhibition on the production of SINV-GFP wild-type and mutant in these cells by transfecting miR-124-3p or control antimiR prior to differentiation and infection. Northern blot analysis of miR-124-3p expression confirmed the depletion of miR-124-3p under both mock and SINV-infected conditions (Fig. 6D). While SINV-GFP wild-type viral titers as well as viral capsid protein levels were reduced about 50% in antimiR-124-transfected cells compared to that in the control (Fig. 6E and F), the mutation in the viral RNA (SINV-GFP-mut) abolished the effect. These results show that inhibition of miR-124-3p negatively affects SINV-GFP viral infection in human neuronal differentiated cells and that this regulation involves the miR-124 binding site at position 10904 to 10926 in the viral genome.

FIG 6.

Sequestration of miR-124-3p in differentiated neuronal SK-N-BE(2) cells attenuates SINV-GFP infection. (A) Microscopy images of proliferating (top) or 6-day-differentiated (bottom) SK-N-BE(2) cells by 10 μM RA treatment. Bars, 25 μm. (B) RT-qPCR analysis of relative MYCN expression in proliferating or differentiated cells. (C) Northern blot analysis of miR-124-3p expression levels in proliferating or 6-day-differentiated cells, mock or SINV-GFP infected. (D) Northern blot analysis of miR-124-3p expression levels upon antimiRNA transfection in differentiated cells at 6 days post-RA treatment. Plaque assay quantification of viral titer production of SINV-GFP or SINV-GFP-mut, shown as relative fold change compared to that under the control condition (E), and Western blot analysis of viral capsid protein accumulation (F) after antimiRNA transfection and 6-day post-RA differentiation of SK-N-BE(2) cells. Tubulin is used as a loading control. Data are presented as means ± SEMs from three independent experiments. **, P < 0.01; ***, P < 0.005, unpaired Student's t test.

Inhibition of miR-124-3p restricts CHIKV infection.

In order to test whether miR-124-3p could positively regulate other positive single-stranded RNA viruses, we tested two different alphavirus CHIKV strains (CHIKV La Réunion and Caribbean strains) and two strains of the flavivirus Zika virus (ZIKV African and French Polynesia strains). CHIKV La Réunion viral production was significantly increased following miR-124-3p overexpression. Though to a lesser extent, the viral production from the Caribbean strain also followed the tendency of increased titers following miR-124-3-p overexpression. In contrast, no significant effect was observed on the two ZIKV strains tested (Fig. 7A). To verify whether miR-124-3p inhibition could reduce CHIKV infection, we transfected an antimiR-124 or a control antimiRNA in LV124 cells and we infected them with a CHIKV La Réunion strain expressing GFP (Fig. 7B). After antimiR-124 transfection, CHIKV-GFP viral titers were reduced about 50% (Fig. 7C). Moreover, antimiR-124 treatment significantly reduced CHIKV-GFP viral titers in differentiated neuroblastoma SK-N-BE(2) cells (Fig. 7D). Interestingly, sequence alignment of the CHIKV genomic sequence corresponding to the binding site present in the SINV genome revealed the presence of a suboptimal seed match with miR-124-3p, which could be compensated by 3′ pairing (Fig. 7E). These results extend the proviral role of miR-124-3p to another alphavirus and confirm that miR-124-3p inhibition might be a promising strategy to block these viruses.

FIG 7.

miR-124-3p positively regulates CHIKV infection. (A) Plaque assay quantification of viral titers, shown as relative fold change, produced by CHIKV La Réunion strain, CHIKV Caribbean strain, ZIKV African strain, and ZIKV French Polynesia strain after transfection of mimic-124-3p (blue) in Huh7.5.1 cells compared to that after transfection of control mimic (black). (B) Schematic representation of CHIKV-GFP genomic structure. Plaque assay quantification of viral titer production of CHIKV-GFP in LV124 cells (C) and 6-days post-RA differentiation in SK-N-BE(2) cells (D), shown as relative fold change of antimiR-124 compared to that under the control condition. Data are presented as means ± SEMs from three independent experiments. *, P < 0.05; **, P < 0.01, for two-way ANOVA with Bonferroni’s correction (A) or unpaired Student's t test (C and D). (E) Sequence alignment of miR-124 binding site on SINV and CHIKV RNAs. The miRNA seed is indicated in blue, and the unpaired nucleotide in the seed match for miR-124 in the CHIKV genome is shown in red.

DISCUSSION

In addition to their role as fine-tuners of cellular functions, microRNAs are emerging as important regulators of host-pathogen interactions (34). They can regulate viral infection either directly, by targeting RNAs of viral origin, or indirectly, by targeting host RNAs, with a positive or negative outcome for the virus. Given the cell specificity of certain miRNAs (41), the resulting interactions with viral RNAs can participate in determining the tissue tropism of viral pathogens. For instance, stable expression of the liver-specific miR-122 increases HCV replication in nonhepatic cells (42). Another example is the inhibition of the Eastern equine encephalitis virus (EEEV) cycle by miR-142-3p, which binds to the 3′ UTR of the virus in hematopoietic/myeloid cells (17).

SINV belongs to the Alphavirus genus, which includes viruses already emerged or with the potential to emerge as important human pathogens (8). Interestingly, as the alphavirus genome mimics the host mRNA and its replication takes place in the cytoplasm, the incoming viral RNA has the potential to directly interact with cellular miRNAs. In agreement, binding sites for cellular microRNAs were identified by Argonaute crosslinking immunoprecipitation (AGO-CLIP) within the alphavirus CHIKV, SINV, and Venezuelan equine encephalitis virus (VEEV) genomes (15).

In this study, we took advantage of a functional genome-wide screen approach (43, 44) to identify pro- and antiviral miRNA activity on SINV. The gain-of-function approach identified 16 miRNAs whose overexpression had a positive effect on the virus. Since we used GFP expressed from the virus as a proxy for measuring viral accumulation, it might be that we have enriched for miRNA regulating viral gene expression rather than just virus replication, but this allowed us to efficiently screen a large number of candidates. Surprisingly, we could not identify any miRNA mimic with an antiviral effect, which might reflect the fact that many viruses are refractory to inhibition by cellular miRNAs, as assessed by the depletion of Dicer (45). In contrast, our loss-of-function screen based on the use of antimicroRNAs was not suitable to identify any phenotypic effect. This could reflect the fact that Huh7.5.1 cells lack miRNAs that could play a role in SINV infection; hence, no miRNAs naturally expressed in those cells can have a pro- or antiviral effect on SINV. Among the hits, we uncovered hsa-miR-124-3p as a novel positive regulator of SINV, and we demonstrated that its overexpression increases SINV structural protein synthesis and viral production while mutations in its seed sequence abolish this effect. We identified and validated a binding site for hsa-miR-124-3p in the SINV genome within positions 10904 and 10926. We focused our efforts on this site for two reasons: first, it is conserved in several strains of SINV found in nature, and second, it lies in a position which corresponds at the same time to the 3′ UTR of the viral genomic RNA and to ORF2, which is only translated when expressed from the subgenomic RNA. Our results indicate that miR-124-3p binding more likely occurs on the viral subgenomic RNA, since it affects viral structural protein translation only.

As SINV may have a neuronal tropism (46), the binding of the cellular miRNA to the viral genome could provide an evolutionary advantage and could be considered a novel example of coevolution between the virus and its host. Indeed, we were able to show that miR-124 inhibitors reduced SINV infection in differentiated human neuroblastoma cells expressing miR-124-3p endogenously. The importance of this miRNA in alphavirus regulation is reinforced by our results obtained with CHIKV, which is also regulated by miR-124. The binding site that we validated in SINV is partially conserved in the CHIKV genome, with one mutation in the seed that could be compensated by extended pairing in the 3′ end of the miRNA.

While we were able to validate that miR-124 appears to regulate SINV by direct binding to the viral RNA, we do not formally exclude that the regulation of cellular RNAs by miR-124 could also participate in the process for both SINV and CHIKV. It has indeed been reported that cellular miRNAs can also regulate viral infections by targeting host mRNAs positively or negatively involved in the host response (34). In particular, this has already been observed for miR-124 regulation of measles virus (47), Japanese encephalitis virus (JEV) (48), and human immunodeficiency virus (HIV) (49) infections. In addition, miR-124 expression is modulated by different viruses, including ZIKV (50), enterovirus 71 (EV71) (51), human cytomegalovirus (HCMV) (52), and influenza H1N1 (53), suggesting a possible implication of the miRNA in the regulation of these infections as well.

CHIKV and SINV are classified as arthritogenic alphaviruses (5) due to the natural symptoms of the infection, which include arthritis and bone pathology (54). This suggests that prior to reaching the CNS, productive infection of cell types other than neurons must take place. Interestingly, in addition to its involvement in the CNS, miR-124 has more recently emerged as a critical modulator of immunity and inflammation (55) by preventing microglia activation (56) or by regulating the adaptive immune response through STAT3 regulation (57). Furthermore, it has also been linked to bone pathology by playing a role in the regulation of osteoclast differentiation (58, 59). Thus, we cannot exclude a role of miR-124–SINV interaction in these cell types. It would be of interest to study miR-124 expression during alphavirus infection in synovial tissues and cartilage, the latter being more physiologically relevant cell types according to these viruses’ tropism. Indeed, this could give more insight into a possible advantage of miR-124–SINV interaction in establishing alphavirus-induced arthritis rather than encephalitis.

We showed that miR-124 is involved in positively regulating viral mRNA translation and viral production, possibly by increasing the amount of structural proteins needed for encapsidation of the viral genome. However, further work will be needed to determine the underlying molecular mechanism. Although the usual outcome of miRNA-mediated regulation is negative, other examples of positive regulation upon miRNA binding have been reported, as it is the case for Bovine viral diarrhea virus and miR-17 (15) or HCV and miR-122 (60). In the former case, miR-17 also binds at the 3′ extremity of the viral genome, similar to our observation with miR-124 and SINV RNA. Nonetheless, our results indicate that miRNAs such as miR-124 could be a potential target for the development of therapeutic drugs to treat diseases associated with alphaviruses. Finally, given the extensive usage of alphaviruses as a vehicle for vaccine and gene therapy delivery (61), the identification of a positive regulation by miR-124 may have a strong impact on the development of more powerful biotechnologies based on the SINV genome.

MATERIALS AND METHODS

Viral stocks, cell culture, and virus infection.

Plasmids carrying a green fluorescent protein (GFP)-SINV genomic sequence or a green fluorescent protein (GFP)-CHIKV La Réunion genomic sequence (kindly provided by Carla Saleh, Institut Pasteur, Paris, France) were linearized with XhoI or NotI, respectively, as in reference 62. They were used as a substrate for in vitro transcription using an mMESSAGE mMACHINE capped RNA transcription kit (Ambion, Thermo Fisher Scientific Inc.) according to the manufacturer’s instructions. GFP expression is driven by duplication of the subgenomic promoter. SINV-GFP and CHIKV-GFP viral stocks were prepared in BHK21 baby hamster kidney cells, and titers were measured by plaque assay. The CHIKV strains used were La Réunion, 06-049 AM258994, and Caribbean (63). The ZIKV strains used were the African HD78788 and French Polynesia, PF-13.

The SINV-GFP-mut plasmid was generated by site-directed mutagenesis using forward (5′-CGCATTTATCAGGACATCAGATGCACCACTGGTCTCA-3′) and reverse (5′-ATGTCCTGATAAATGCGGCGTTCGGGATGTCAATAGA-3′) primers and the In-Fusion HD Cloning kit (TaKaRa) according to the manufacturer’s instructions.

Cells were infected with all viruses at a multiplicity of infection (MOI) of 10⁻³, and samples were harvested at 24 h postinfection (hpi) unless specified otherwise.

Huh7.5.1 cells were maintained in Dulbecco’s modified eagle medium (DMEM) with 4.5 g/liter glucose (Gibco, Thermo Fisher Scientific Inc.) supplemented with 10% fetal bovine serum (FBS; TaKaRa), 1% MEM nonessential amino acids solution (NEAA 100×; Gibco, Thermo Fisher Scientific Inc.), and gentamicin (50 μg/ml; Gibco, Thermo Fisher Scientific Inc.) in a humidified atmosphere of 5% CO₂ at 37°C. HEK293A (QBiogene) and Vero R (88020401; Sigma-Aldrich) cells were maintained in DMEM (Gibco, Thermo Fisher Scientific Inc.) supplemented with 10% FBS (Clontech) in a humidified atmosphere of 5% CO₂ at 37°C.

SK-N-BE(2) cells (95011815; Sigma-Aldrich) were maintained in 1:1 medium composed of Ham’s F12 medium (Gibco, Thermo Fisher Scientific Inc.) supplemented with 15% FBS and Eagle’s minimal essential medium (ENEM) supplemented with 1% NEAA (Gibco, Thermo Fisher Scientific Inc.). For miR-124-3p inhibition experiments in SK-N-BE(2) cells, due to the low transfection efficiency of these cells once differentiated, they were first transfected with 75 nM antimiR specific to miR-124-3p or antimiR-CTL, and 6 h later, differentiation was induced by 10 μM RA treatment (R2625; Sigma-Aldrich).

Lentivirus production and generation of stable cell line.

The human pre-miR-124a-1 gene in locus 8p23.1 flanked by about 200 nucleotides of its upstream and downstream genomic sequences was PCR amplified using the following primers: hsa-pri-miR-124 Fw, 5′-GGGGACAAGTTTGTACAAAAAAGCAGGCTTCGAGCTGCGGCGGGGAGGATGC-3′; hsa-pri-miR-124 Rv, 5′-GGGGACCACTTTGTACAAGAAAGCTGGGTCCCCCTGTCTGTCACAGGCTGC-3′.

The obtained insert was first cloned by the Gateway BP reaction (Invitrogen, Thermo Fisher Scientific Inc.) in pDONOR-221 and then by the Gateway LR reaction (Invitrogen, Thermo Fisher Scientific Inc.) in the lentiviral destination vector pLenti6.2-3×FLAG-V5-ccdB (87072; Addgene). Lentiviruses were produced by cotransfection of pLenti6.2-3×FLAG-V5-hsa-pri-miR-124 with packaging vectors encoding lentiviral Gag and Pol proteins, pPAX, and vesicular stomatitis virus (VSV) envelope glycoprotein pVSV-G. Twenty-four hours posttransfection, the supernatant was collected and used as the inoculum for transduction. Huh7.5.1 cells were transduced with lentiviruses in the presence of 4 μg/ml Polybrene (SC-134220; Santa Cruz Biotechnology) for 6 h. Afterwards, the inoculum was removed, and cells were incubated in complete medium. Selection pressure was applied by supplementing the complete medium with 15 μg/ml of blasticidin (Invivogen). Surviving cells were maintained in culture in the presence of blasticidin as a polyclonal miR-124-3p-expressing cell line (LV124).

High-content miRNA-based phenotypic screening.

For the miRNA screen, the miRIDIAN microRNA mimic (catalog number CS-001030) and inhibitor (catalog number IH-001030) libraries (19.0, human microRNAs) were purchased from Dharmacon (Thermo Fisher Scientific Inc.). For the mimic screen, 20 nM each mimic microRNA was transfected into Huh7.5.1 cells (cultured in DMEM with 4.5 g/liter glucose, 10% HyClone fetal calf serum [FCS], 50 μg/ml gentamicin) grown in Greiner μClear 96-well microplates using a high-throughput (HT) reverse chemical transfection with the INTERFERin HTS delivery reagent (Polyplus-transfection SA, Illkirch, France). For the inhibitor screen, 75 nM each inhibitor microRNA was transfected into Huh7.5.1 cells as described above. The HT transfection protocol was optimized for reaching 90% to 95% transfection efficiency with minimal toxicity on a TECAN Freedom EVO liquid handling workstation. The screens were performed in technical triplicates. To limit biological variability, the cell passage (n = 3 after thawing), serum batch, and transfection reagent batch were strictly determined. Internal controls such as positive (siRNA targeting SINV 3′ UTR at nt 12345 and nt 12437: 5′-AACUCGAUGUACUUCCGAGGAUU-3′; Integrated DNA Technologies) and negative siRNA controls (ON-TARGETplus nontargeting siRNA number [no.] 2; Horizon Discovery, Dharmacon), and transfection efficiency control (“PLK1” siRNA that leads to cell death) were added to each microplate to determine parameters for interplate and day-to-day variability. Three days posttransfection, the cells were subjected to SINV-GFP viral infection for 24 h before fixation and staining with Hoechst 3342 (labeling nuclear compartments [nuclei]). High-throughput cell imaging was carried out with the INCELL1000 HCS epifluorescence microscope to collect an average of ∼6,000 cells analyzed per microwell (Hoechst 3342 and GFP channels).

Analysis of the high-content siRNA screening data.

Hoechst 3342 and GFP signals were extracted for all individual cells using the Multi Target Analysis module of the INCELL1000 platform. These parameters describe the GFP⁻ noninfected and GFP⁺ SINV-infected Huh7.5.1 cell populations for each miRNA treatment. The robust strictly standardized mean difference (SSMD* or SSMDr) value for each well was calculated as described below according to reference 64.

$$\text{SSMDr}=\text{median}\left(\text{hit}\right)-\text{median}\left(\text{plate}\right)*1/(\sqrt{2}*\text{MAD}),$$

where MAD is the mean absolute deviation.

Small RNA library preparation, sequencing, and analysis.

RNA was extracted from Huh7.5.1 cells, and a small RNA library was prepared from 25 μg of total RNA as previously described (62, 65, 66). Single-end sequencing was performed at the IGBMC Microarray and Sequencing platform, Illkirch, France, on an Illumina Genome Analyzer IIx machine with a read length of 36 bp. Deep-sequencing data analysis was performed as in reference 62 with slight modifications. Briefly, FASTX-Toolkit (http://hannonlab.cshl.edu/fastx_toolkit) was first applied to remove instances of the 3′ adaptor. Remaining reads between 18 and 32 nt in length were then mapped to the human genome (assembly version hg19; UCSC repository) using Bowtie 1.0.0 (67). Up to 2 mismatches in total with no more than 1 mismatch in the first 15 nucleotides of each read were permitted. In addition, only alignments from the lowest mismatch stratum were recorded, and reads that could map to more than 50 loci were discarded. Finally, expressed human miRNAs (miRBase Release 20 [68]) were identified and quantified using BEDTools 2.16.2 (69) by comparing their genomic coordinates to those of the aligned reads. During the quantification process, multiple mapped reads were weighted by the number of mapping sites in other miRNAs, and the final counts were normalized per million miRNA reads (RPM).

microRNA mimic and siRNA transfection.

For reverse transfection in Huh7.5.1 or HEK293A cells, transfection complexes were prepared using mimics (miRIDIAN microRNA mimics; Dharmacon) or siRNAs (Dharmacon or Integrated DNA Technologies) at a final concentration of 30 nM unless specified differently in INTERFERin-HTS transfection reagent (Polyplus-transfection SA, Illkirch, France) and transfection medium according to the manufacturer’s instructions. Transfection complexes were added to 1.5 × 10⁴ cells in each well of 48-well cell culture plates. Transfected cells were subsequently incubated for 72 h before being infected with SINV-GFP at an MOI of 10⁻³ for 24 h. For the experiment with the mutant mimic of miR-124-3p, mimics for Caenorhabditis elegans elegans miR-67 (5′-UCACAACCUCCUAGAAAGAGdTdT-3′), miR-124-3p (5′-UAAGGCACGCGGUGAAUGCCdTdT-3′), and miR-124-3p mutated in the seed region (5′-UAACCGACGCGGUGAAUGCCdTdT-3′) were purchased from Integrated DNA Technologies.

miRNA inhibition with 2′O-methylated oligonucleotides.

For inhibition of miRNAs with antimiRs, 1.5 × 10⁴ LV124 cells were cultured in 48-well dishes, and SK-N-BE(2) cells were cultured in 6-well plates at a confluence of 2 × 10⁴ cells/cm². Cells were transfected with the 2′O-methylated antimiRs against the endogenously expressed miR-124-3p (5′-GGCAUUCACCGCGUGCCUUA-3′) or with the control sequence of C. elegans cel-miR-67 (5′-UCACAACCUCCUAGAAAGAGUAGA-3′), using INTERFERin-HTS transfection reagent (Polyplus-transfection SA, Illkirch, France) or Lipofectamine 2000 (Invitrogen, Thermo Fisher Scientific Inc.). AntimiRs were used at a final concentration of 75 nM and transfections were performed according to the manufacturer’s instructions. Forty-eight hours posttransfection cells were infected with SINV-GFP at an MOI of 10⁻³ for 24 h.

Standard plaque assay.

Vero R cells seeded either in 96- or 24-well plates were infected with infection supernatants prepared in cascade 10-fold dilutions for 1 h. Afterwards, the inoculum was removed, and cells were cultured in 2.5% carboxymethyl cellulose for 72 h at 37°C in a humidified atmosphere of 5% CO₂. Plaques were counted manually under the microscope. For plaque visualization, the medium was removed, and cells were fixed with 4% formaldehyde for 20 min and stained with 1× crystal violet solution (2% crystal violet [Sigma-Aldrich], 20% ethanol, 4% formaldehyde).

Virus attachment and entry test.

Huh7.5.1 cells were reverse transfected in 48-well plates with control or miR-124 mimic at a final concentration of 30 nM with Lipofectamine 2000 according to the manufacturer’s instructions and incubated for 72 h. Prechilled cells (15 min at 4°C) were incubated with SINV-GFP at an MOI of 5 at 4°C for 30 min (attachment assay). Unbound SINV virions were removed by washing with phosphate-buffered saline (PBS) three times. Total RNA was extracted and used for quantification of SINV RNA by qRT-PCR. SINV entry test was carried out by incubating SINV with cells at 4°C for 30 min to allow viral binding, and then unbound SINV virions were washed rapidly three times with PBS. Finally, cells were shifted to 37°C for 30 min to allow virus internalization, as described previously (70).

Western blotting.

Proteins were extracted by collecting cell lysates in RIPA buffer (10 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.5 mM EDTA, 0.1% SDS, 1% Triton X-100, 1% sodium deoxycholate, and protease inhibitor). Lysates were cleared by centrifugation at 13,000 rpm for 30 min at 4°C to remove cell debris, and the supernatant was retained for Western blotting. Samples were loaded in a 10% acrylamide–bis-acrylamide gel, and proteins were separated by migration at 100 V in 1× Tris-glycine-SDS buffer. Proteins were transferred to a nitrocellulose membrane by wet transfer in 1× Tris-glycine and 20% ethanol buffer. Viral proteins were detected using primary polyclonal antibodies against SINV capsid protein (CP) and nsP2 (kind gift from Diane Griffin, Johns Hopkins University School of Medicine, Baltimore, MD) and a secondary antibody anti-rabbit-horseradish peroxidase (HRP) (NA9340; GE Healthcare, Thermo Fisher Scientific Inc.). The signal was revealed by incubating the membrane for 10 min with SuperSignal West Femto maximum sensitivity Substrate (Pierce, Thermo Fisher Scientific Inc.). Tubulin was detected with a primary monoclonal antibody (T6557; Sigma-Aldrich) and a secondary antibody anti-mouse-HRP (NXA931; GE Healthcare, Thermo Fisher Scientific Inc.).

Northern blotting.

Total RNA was extracted from cells with TRIzol Reagent (Invitrogen, Thermo Fisher Scientific Inc.) according to the manufacturer’s instructions. Five to ten micrograms of total RNA was loaded on a 17.5% acrylamide-urea 4 M gel and resolved by running in 1× Tris-borate-EDTA for isolation of small RNAs. Nucleic acids were transferred to a nylon membrane by semidry transfer. Small RNAs were cross-linked to the membrane by chemical cross-link using N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC; Sigma-Aldrich). Membrane was prehybridized for 20 min with PerfectHyb Plus hybridization buffer (Sigma-Aldrich). DNA oligonucleotides directed against hsa-miR-124-3p (5′-GGCATTCACCGCGTGCCTTA-3′) and snRNA U6 (5′-GCAGGGGCCATGCTAATCTTCTCTGTATCG-3′) were radiolabeled with 2.5 μCi of γ-ATP by polynucleotide kinase (PNK). After removal of unbound γ-ATP by MicroSpin G-25 column (GE Healthcare, Thermo Fisher Scientific Inc.) purification, the probe was incubated with the membrane in hybridization buffer overnight at 50°C. Membranes were washed twice with SSC 4× solution (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate) at 50°C and exposed on an image plate in a cassette. Imaging of the signal was obtained with Typhoon FLA 7000 laser scanner (GE Healthcare Life Sciences).

RT-qPCR.

For MYCN expression, DNase I (Invitrogen, Thermo Fisher Scientific Inc.) treatment was performed on 1 μg of RNA, which was retrotranscribed using a random nonameric primer and SuperScript IV reverse transcriptase (Invitrogen, Thermo Fisher Scientific Inc.) according to the manufacturer’s instructions. For miR-124-3p, SINV genomic and subgenomic RNA, and CDK4 expression, 250 ng of RNA was retrotranscribed using miScript II (Qiagen) according to the manufacturer’s instructions. Quantitative PCR was performed on a 1/10 dilution of cDNA using SYBR green PCR master mix (Thermo Fisher Scientific Inc.) with the following primers: MYCN forward (5′-GAGCGATTCAGATGATGAAG-3′) and reverse (5′-TCGTTTGAGGATCAGCTC-3′) (40); SINVgenome forward (5′-CCACTACGCAAGCAGAGACG-3′) and reverse (5′-AGTGCCCAGGGCCTGTGTCCG-3′); SINVsubgenome forward (5′-CCACAGATACCGTATAAGGCA-3′) and reverse (5′-TGCAGGTAATGTACTCTTGG-3′); CDK4 forward (5′-CCGTGGTTGTTACACTCTGG-3′) and reverse (5′-ATTTTGCCCAACTGGTCGGC-3′); and GAPDH (glyceraldehyde-3-phosphate dehydrogenase) forward (5′-CTTTGGTATCGTGGAAGGACT-3′) and reverse (5′-CCAGTGAGCTTCCCGTTCAG-3′). For the miScript primer assay, mature hsa-miR-124-3p (MIMAT0000422; 5′-UAAGGCACGCGGUGAAUGCC-3′) or MS00033740 hs_RNU6-2_11 was amplified with a reverse 10× miScript universal primer (Qiagen).

Strand-specific RT-PCR.

Infected cells were lysed with TRIzol, cellular RNA was extracted according to the manufacturer’s instructions, and RNA pellets were resuspended in water. To detect the levels of SINV antigenomic RNA, negative-strand-specific reverse transcription was performed with a plus-sense primer annealing to the 5′ region of the SINV genome (nucleotides 1 to 42) (5′-ATTGACGGCGTAGTACACACTATTGAATCAAACAGCCGACCA-3′). The RT reaction mix was set up with 100 ng of RNA, 1 μl of 2 μM specific primer, 1 μl 10 mM deoxynucleoside triphosphates (dNTPs), and water to reach a final volume of 13 μl. Samples were mixed, incubated at 65°C for 5 min, and placed 2 min on ice. Reaction mix was completed with 4 μl 5× Superscript IV buffer (Invitrogen), 1 μl 0.1 M dithiothreitol (DTT), 1 μl RNase inhibitor, and either 1 μl Superscript IV enzyme (Invitrogen) or water for a negative control. Samples were mixed and incubated at 50°C for 10 min and 80°C for 10 min. One microliter of cDNA products was then amplified by PCR with specific antigenome forward (5′-CATTCTACGAGCCGGTGCGC-3′) and reverse (5′-TAGACGTAGACCCCCAGAGTC-3′) primers using the GoTaq DNA polymerase (Promega). PCR was run as follows: 94°C for 2 min, 15, 20, 25, or 30 cycles of 95°C for 30 s, 55°C for 30 s, and 72°C for 30 s, and a final extension step of 5 min at 72°C. PCR products were loaded on a 1.5% agarose gel for analysis.

Data availability.

The sequencing data discussed in this publication have been deposited in NCBI’s Gene Expression Omnibus (71) and are accessible through GEO Series accession number GSE136740.