Respiratory syncytial virus infection‐induced mucus secretion by down‐regulation of miR‐34b/c‐5p expression in airway epithelial cells

Abstract Severe RSV infection is the main cause of hospitalization to children under the age of five. The regulation of miRNAs on the severity of RSV infection is unclear. The aim of the study was to identify the critical differential expression miRNAs (DE miRNAs) that can regulate the pathological response in RSV‐infected airway epithelial cells. In this study, miRNA and mRNA chips of RSV‐infected airway epithelia from Gene Expression Omnibus (GEO) were screened and analysed, separately. DE miRNAs‐targeted genes were performed for further pathway and process enrichment analysis. DE miRNA‐targeted gene functional network was constructed on the basis of miRNA‐mRNA interaction. The screened critical miRNA was also investigated by bioinformatics analysis. Then, RSV‐infected human bronchial epithelial cells (HBECs) were constructed to verify the expression of the DE miRNAs. Finally, specific synthetic DE miRNAs mimics were used to confirm the effect of DE miRNAs on the RSV‐infected HBECs. 45 DE miRNAs were identified from GEO62306 dataset. Our results showed that hsa‐mir‐34b‐5p and hsa‐mir‐34c‐5p decreased significantly in HBECs after RSV infection. Consistent with the biometric analysis, hsa‐mir‐34b/c‐5p is involved in the regulation of mucin expression gene MUC5AC. In RSV‐infected HBECs, the inducement of MUC5AC production by decreased hsa‐mir‐34b/c‐5p was partly mediated through activation of c‐Jun. These findings provide new insights into the mechanism of mucus obstruction after RSV infection and represent valuable targets for RSV infection and airway obstruction treatment.


| INTRODUC TI ON
Respiratory syncytial virus (RSV) infection in early childhood has been associated with the development of asthma. It is worth noting that severe early RSV infection trigger an increased susceptibility of asthma and the exacerbations in most of children with asthma, which caused substantial public health cost and economical cost. 1,2 Although epidemiologic studies have revealed that over 95% children have been infected with RSV before 2 years old, 3 severe RSV infection only occurs in a small number of RSV-infected children. 4 Given the significant connections between RSV infection response and the development and exacerbation of asthma, accumulating studies have demonstrated that the differential antiviral responses in asthmatics may explain these connections. 5 The immune response and clinical manifestations after RSV infection vary greatly among children after RSV infection. 6,7 Moreover, airway obstruction is one of the main clinical manifestations of hospitalization for severe RSV infection. 8 Compared with healthy children, asthmatic children have lung function impairment during childhood which persists into adulthood. Moreover, airway remodelling is already present in the early stage of childhood asthma. 9 The lungs of infants with severe RSV infection also showed that desquamated epithelial cells and inflammatory cell could cause airway obstruction. 10 It is also of particular concern that increased mucus secretion has shown to be related to the severity of RSV infection. Moreover, RSV infection could increase the composition of airway epithelial cells that secrete MUC5AC. 11 Then, the excessive mucus secretion and reduced mucus clearance by ciliated cells lead to mucus deposition, airway obstruction, further infection and even increased bacterial colonization. 12 However, the inner mechanism of airway mucus secretion after RSV infection is far from clear.
Interestingly, a series of recent evidences demonstrated that RSV infection regulates epithelial genes expression through epigenetic mechanisms. 13 Epigenetic mechanisms such as DNA methylation, histone modifications and microRNAs expression could regulate transcription activities of target genes without alterations of nucleotide sequence. It has been demonstrated that aberrant miRNA expression plays a critical role in promoting the development and progression of RSV infection. 14 miRNAs were also screened to serve as a therapeutic and prognostic factor for RSV infections. 15,16 miRNA is a type of noncoding RNA between 22 and 24 nucleotide (nt). 17 Acting as a negative regulator of gene expression, miR-NAs are predicted to target 60% of all human protein-coding genes and involved in many biological processes, including proliferative responses and inflammatory/immune responses. 18 Several studies have investigated the altered miRNA in RSV-infected airway epithelial cells and the role of the altered miRNAs has recently been highlighted. 14,16,19 However, the specific role of the differential miRNAs in the pathogenesis between mild RSV infection and severe RSV infection has not been fully understood.
Here, we analysed differential expression miRNAs (DE miRNAs) between healthy individuals, mildly RSV-infected patients and severely RSV-infected patients by screening miRNA profiling from public datasets. Then, pathway enrichment analysis, and miRNA-mRNA function regulated network were constructed to further uncover the function of the DE miRNAs. Bioinformatics analysis screened out that hsa-mir-34b-5p and hsa-mir-34c-5p were negatively correlated with the severity of RSV infection. RSV-infected airway epithelial cells further confirmed that RSV infection decreased the expression of hsa-mir-34b/c-5p, which further induced mucus secretion through the activation of c-Jun/AP-1. Together, these results explored a new possible mechanism for mucus secretion after RSV infection in airway epithelial cells which further provided a potential target for the prevention and treatment of asthma.

| Search strategy
A systematic searching strategy was constructed to identify miRNA and mRNA expression profiles in public platform. Firstly, a comprehensive search was performed in GEO (www.ncbi.nlm.nih.gov/geo/) and Array Express (www.ebi.ac.uk/array express), separately. The search string used such terms as "microRNA" AND "microarray", "RSV" OR "respiratory syncytial virus" OR " respiratory infection". Secondly, datasets-relevant studies and their reference lists were further reviewed to ensure no potential researches have been missed. Datasets were excluded according to the following criteria: (a) only cell lines were used; (b) non-human organism; (c) non-respiratory tissue.

| Identification of DE miRNAs
GSE62306 was finally selected for analysis according to the above standards. Using the GEO2R program (https://www.ncbi.nlm.nih. gov/geo/geo2r/), normalization and identification of DE miRNAs from GSE62306 dataset were performed. Raw data were normalized, and the overall characteristics of value distributions and samples which were not median-centred values were excluded. At mechanism of mucus obstruction after RSV infection and represent valuable targets for RSV infection and airway obstruction treatment.

K E Y W O R D S
airway epithelial cells, miRNA, mucus secretion, pathway and process enrichment analysis, respiratory syncytial virus this stage, the nasal mucosal samples from healthy children were assigned to 'control group'. Severe RSV infection samples were assigned to 'severe group' according to grade information. The DE miRNAs were calculated by limma package built-in R software and adjusted for multiple test by the Benjamini and Hochberg (BH) method, with the threshold criterion of adjusted P-value < 0.05. 20 Results of included miRNA expression were ranked by adjusted P-values. were sorted by P-value, and P-value less than 0.05 was considered to be significantly enriched.

| miRNA-target regulatory functional network construction
To identify a more valuable miRNA-mRNA interactions network, we

| RT-qPCR for c-Jun, c-Fos, MUC5AC and MUC5B
Total RNA was prepared from HBECs and quantified on a SmartSpec™ Plus spectrophotometer (Bio-rad, USA). RT-PCR was conducted according to the PrimeScript™ RT Master Mix Kit (Takara, Japan). Quantitative PCR (qPCR) was performed on a CFX96 Touch™ Deep Well Real-Time PCR Detection System (Bio-rad, USA) by the use of TB Green ® Premix Ex Taq (Takara, Japan) with thermal cycling conditions. 24 Primer sequences were described in Supplementary Table 1.

| miRNA RT-qPCR
Total miRNAs was prepared with TRIzol reagent (Thermo Fisher Scientific, USA) and Phenol-Chloroform extraction from HBECs and quantified on a SmartSpec™ Plus spectrophotometer (Bio-rad, USA). 25 cDNA was reverse transcribed by use of the miRcute miRNA cDNA kit (Tiangen Biotech, China), and RT products were then detected by the miRcute Plus miRNA qPCR Kit (Tiangen Biotech, China). U6 was used as an internal reference. The thermal cycling conditions were the following: 95˚C for 5 minutes, followed by 40 cycles at 95˚C for 15 seconds, 60˚C for 30 seconds and 72˚C for 20 seconds. The relative expression was calculated with the 2-ΔΔCt method. The miRNA primer sequences were described in Table S1.

| Western blot
Protein extraction from HBECs was performed according to previous procedures. 26 In brief, 50 µg protein was isolated and separated from HBECs by 10% SDS-PAGE and transferred to a polyvinylidene fluoride (PVDF) membrane. Then, the PVDF membrane was incubated with primary antibody for 12 hours and next incubated with Horseradish Peroxidase (HRP) conjugated secondary antibody.
Expressions of c-Jun (Santa Cruz, sc-74543) and phosphorylated c-Jun (Santa Cruz, sc-822) were determined with corresponding antibodies.
DAPI was used for nuclear staining. 27 IF samples were visualized using a fluorescence microscope (Carl Zeiss MicroImaging GmbH, Göttingen, Germany). Images were captured with a digital camera (Axio-Cam ICc3, Spectra Service, Ontario, NY, USA) and analysed with AxioVision Rel. 4.7 software (Zeiss).

| Statistical analysis
All data were analysed with GraphPad Prism Software (version 6; San Diego, CA, USA) and presented as mean ± SEM. Statistical comparisons were made by one-way ANOVA followed by Dunnett's post hoc test. Differences were considered statistically significant for *P < 0.05, **P < 0.01 and ***P < 0.001.   Figure S1). These results indicated that the samples in the mild group may not be heterogeneous enough to distinguish from the other two groups, which may reduce accuracy of the analysis results.

| DE miRNA expression profiles in RSV-infected children with different degree
Besides, the DE miRNAs in all three groups are relatively limited which may miss some of the important information and cause discrepancy in subsequent bioinformatics analysis. Thus, in our study, healthy controls group and severe RSV group were chosen for comparison with a total of 27 samples. Table S2 Figure 1(A,B).
Principal component analysis (PCA) also showed that two groups were well separated by the evaluation of these DE miRNA expression ( Figure 1C).

| Pathway enrichment analysis of DE miRNA
To further understand the function and pathway of these DE miR- In each database, P-value < 0.05 was selected. Pathways enriched in more than two databases were presented in Table 1 and

| DE miRNAs-targeted gene functional network construction
The potential interacted gene with DE miRNAs was predicted by DIANA-MR-microT database. A total of 4716 genes were obtained under predict score > 0.95. To further explore the differentially expressed genes (DEGs) that may be regulated by miRNA following RSV infection, we compared the predicted target genes with the DEGs derived from RNA-seq datasets, GSE32138, GSE32139 and GSE41374. Specific information of these three datasets was shown in Table S2.

| Decreased hsa-mir-34b/c-5p induced MUC5AC expression through activation of c-Jun in RSV-infected HBECs
As HBECs ( Figure 6C,E), whereas no significant change of RSV replication was observed with SP600125 existent (Figure 6D,F). prevalence and exacerbation of asthma. However, the mechanism for the differential clinical manifestations and severity after RSV infection remains unclear. miRNAs have been shown to participate in modulating antiviral responses through not only the airway immune response pathways but also the virus invasion pathways. 16,19 In this study, our results demonstrated that hsa-mir-34b/c-5p decreased dynamically in RSV-infected HBECs, which was negatively related to the severity of RSV infection. Decreased hsa-mir-34b/c-5p downregulated c-Jun activation which further promoted the production of MUC5AC from HBECs.

| D ISCUSS I ON
By analysing miRNA microarray (GSE62306), we identified that DE miRNAs between the healthy controls and severe RSV-infected groups. This dataset was analysed and revealed differentially expressed miRNAs in mild or severe RSV disease. 30 Then, the results were verified by qPCR with samples of the same individuals.
However, the study did not further explore the mechanism of the DE production. The hypersecretion of MUC5AC increases the concentration of solids in the mucus, which makes the mucus difficult to remove. 48 Although it has been found in previous studies that the c-Jun NH2 terminal kinase (JNK) signalling pathway is activated after RSV infection, no modulation of JNK by mir-34b/c was found in our RSV-infected HBECs. 49 These results indicate that the activation of c-Jun is a combination of JNK pathway activation and miR-34b/c down-regulation.
Although this study demonstrated the critical role of decreased mir-34b/c in RSV-induced mucus secretion from airway epithelial cells, there are still some limitations. Indeed, we did not find a potential binding site in c-Jun 3'UTR or 5'UTR for mir-34b-5p or mir-34c-5p although c-Jun is negatively regulated by mir-34b-5p and mir-34c-5p partly, which indicates that the regulation of mir-34b/c-5p on MUC5AC may be indirect. To probe the possible downstream target genes, we constructed miRNA-target regulatory functional network ( Figure 4C), which indicated that FGFR1 appears as the possible connecting molecules between hsa-mir-34c-5p and c-Jun. Besides, the prediction from TargetScan also showed that there are potential binding sites between FGFR1 and hsa-mir-34b/ c-5p. Moreover, the positive regulation of FGFR1 on c-Jun has been confirmed in the previous literature. 50 However, further work is still needed to confirm the possible target genes. Moreover, it remains unclear about other DE miRNAs or posttranscriptional mechanism in RSV-infected HBECs. In addition, the molecular mechanism of c-Jun activation after RSV infection remains to be further studied.
In summary, this study validated that the decreased expression of hsa-mir-34b/c-5p induced MUC5AC expression in RSV-infected HBECs. Moreover, reduced hsa-mir-34b/c-5p leads to pathological production of mucin through the activation of c-Jun which is mediated through the AP-1 signalling pathway. These results provide some useful insights into the molecular mechanisms of mucus secretion after RSV infection and may also provide some valuable targets for RSV infection and airway obstruction treatment.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data used and analysed in this study are included in this article are available in the GEO database (https://www.ncbi.nlm.nih.gov/geo).