All animal experiments in this study were performed following protocols approved by the Institutional Animal Care and Use Committee at the South China Agricultural University, China (approval ID: SCAU#0011). All the experiments were carried out in compliance with the regulations and guidelines established by this committee. The authors declare that all animal experiments under the consent of the animal owners. All efforts were made to minimize suffering to animal. The authors declare no additional considerations of the study in cases where vulnerable populations were involved.

Two chicken groups, ALV-J-infected broilers and non-infected broilers at 140 days old, were used. The animals were the same as those used in our previous RNA sequencing study (Li et al., 2015).

SCAU-HN06, a strain of ALV-J, was kindly provided by Assoc. Prof. Weisheng Cao (South China Agricultural University). The chicken embryonic fibroblast cell line (DF-1 cells) was obtained from the Cell Bank of the Committee on Type Culture Collection of the Chinese Academy of Sciences and it was cultivated adherently in cell culture vessels containing Dulbecco's Modified Eagle's Medium (DMEM; Gibco, Life Technologies, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, Life Technologies, Grand Island, NY, USA) at 37°C in an atmosphere of 5% CO₂. When the cells grew to a density of 80% confluence, they were infected with the ALV-J strain SCAU-HN06. The inoculum was discarded 2 h later and it was replaced with DMEM containing 1% FBS, 100 U/mL penicillin, and 100 ug/mL streptomycin. After incubation for 6 days in culture, the cell supernatants (which contained the viral particles) were harvested as described previously (Dai et al., 2016). The virus was diluted 10-fold using serial dilution techniques in DMEM and the suspension was then applied to DF-1 cells for 6 days in a 96-well plate. The virus was titrated to determine the 50% tissue culture infectious doses (TCID₅₀) in accordance with the method described by Reed and Muench (1938).

Total RNA was extracted from the tissue samples and DF-1 cells using TRIzol® as recommended by the supplier (Invitrogen Life Technologies, Carlsbad, CA, USA). The integrity and quantity of RNA were assessed using 1.5% agarose gel electrophoresis and spectrophotometry (ND-2000, NanoDrop Inc., Wilmington, DE, USA), respectively. A PrimeScript RT Reagent Kit (Takara, Otsu, Japan) was used to carry out cDNA synthesis of mRNA, following the manufacture's protocol. The synthesized cDNA was stored at −20°C until subsequent analysis using real-time quantitative polymerase chain reaction (RT-qPCR).

RT-qPCR primers that were specific for MDA5, ALV-J Env, Gag, Pol, Mx1, and β-actin were designed using Premier Primer 5.0 software. RT-qPCR primers that were specific for genes in the MDA5 signaling pathway, including interferon-β promoter stimulator 1 (IPS-1), interferon regulatory factor-3 (IRF-3), interferon-β (IFNβ), double-stranded RNA-dependent protein kinase (PKR), 2′, 5′-oligoadenylate synthetase (OAS), myxovirus (influenza virus) resistance 1, interferon-inducible protein p78 (mouse) (Mx1), and major histocompatibility complex class I (MHC class I), are referenced in previous publications (Supplementary Table 1; Lee C. C. et al., 2014). PCR primers for the full-length MDA5 coding sequence clone and the 3′ UTR of the MDA5 clone were also designed using the Premier Primer 5.0 software (Supplementary Table 2). All the above primers were synthesized by Sangon Biotech Co., Ltd. (Guangzhou, China). A bulge-loop™ Reverse Transcription primer and RT-qPCR primers that were specific for gga-miR-34b-5p were designed and synthesized by RiboBio (Guangzhou, China).

Gga-miR-34b-5p mimics, mimic control duplexes, small interfering RNA (siRNA) targeted against the MDA5 gene (si-MDA5), and a non-specific siRNA control duplex were designed and synthesized by RiboBio (Guangzhou, China). Oligonucleotide sequences of si-MDA5 were showed in Supplementary Table 3. To construct a gga-miR-34b-5p target luciferase reporter vector (pmirGLO-WT MDA5-3′UTR), the segment sequence of the MDA5 3′ UTR (666 bp) that contained the putative gga-miR-34b-5p binding sequence was amplified by PCR using a cDNA template synthesized from total RNA. Subsequently, the PCR product was sub-cloned into NheI/SalI restriction sites in the pmirGLO dual-luciferase reporter vector (Promega, Madison, WI, USA) to generate the pmirGLO- WT-MDA5-3′UTR reporter vector.

However, to generate a gga-miR-34b-5p target-mutated reporter vector (pmirGLO-MT-MDA5-3′UTR), mutations were achieved by changing the gga-miR-34b-5p binding seed sequences from ACTGCCT to GACTATC using the megaprimer PCR method (Ke and Madison, 1997). An MDA5 overexpression construct was generated by amplifying the MDA5 coding sequence, and it was subsequently cloned into the overexpression plasmid vector, pSDS-20218, which was purchased from Shanghai SiDanSai Biotechnology Co., Ltd., China (http://www.sidansai.com/).

When DF-1 cells grew to a density of 50% confluence, they were transfected with (a) the MDA5 overexpression plasmid, (b) si-MDA5, or (c) the gga-miR-34b-5p mimic using Lipofectamine 3000 Reagent (Life Technologies, USA), in accordance with the manufacture's recommended protocol., After 12 h, the cells were inoculated with TCID₅₀ of ALV-J. After 2 h of incubation, the supernatant was discarded and the infected cells were replenished with DMEM medium containing 1% FBS, 100 U/mL penicillin, and 100 ug/mL streptomycin.

Luciferase activity was measured using Dual-GLO® Luciferase Assay System Kits (Promega, Madison, WI, USA) following the manufacturer's instructions. DF-1 cells were seeded at a density of 1 × 10³ cells per well in 96-well plates. After 24 h, the cells were co-transfected with 100 ng pmir-GLO- WT-MDA5-3′ UTR- (wild-type) or pmir-GLO-MT-MDA5-3′ UTR(mutant-type) plasmids, or 100 nM gga-miR-34b-5p mimic and miR-NC using the Lipofectamine 3000 Reagent. Fourty-eight hours after transfection, luciferase assays were performed using a Fluorescence/Multi-Detection Microplate Reader (Synergy 2, Biotek, Winooski, VT, USA). The values obtained were normalized to the levels of a Renilla luciferase plasmid (pRL-TK Vector) levels.

The RT-qPCR reactions were performed using a Bio-Rad CFX96 Real-Time PCR Detection System using iTaq™ Universal SYBR® Green Supermix Kits (Bio-Rad Laboratories Inc., Hercules, CA, USA) to detect the mRNA expression level associated with each gene. The chicken β-actin gene was used as an internal control. The analysis was carried out using the 2^−ΔΔCt method, as described previously (Livak and Schmittgen, 2001).

We also carried out a wound healing migration assay. First, DF-1 cells were seeded in 12-well plates. When the cells grew to a density of 50% confluence, they were transfected with the MDA5 overexpression plasmid, si-MDA5, or a gga-miR-34b-5p mimic, and then they were infected with ALV-J, as described above. When the cells reached a density of 100% confluence, a linear wound was generated by scratching the monolayer of cells with a 1 mL pipette tip. The cells that migrated into the wound area were captured at 0 (control), 24, 48, and 72 h using a microscope (Nikon, Shinagawa, Tokyo, Japan). Width of the scratches was measured using the Image Pro Plus 6.0 software (Media Cybernetics). The data of each scratch width were presented in Supplementary Table 4. The formula to calculate the cell migration rate was as follows: (W_0h − W_xh) × 100%/W_0h, where W_0h, represents the mean wound width at 0 h and W_xh represents the mean wound width at 24, 48, and 72 h.

For the flow cytometry analysis of the cell cycle, DF-1 cells were seeded in 12-well plates. When the cells grew to a density of 50% confluence, they were transfected and then infected with ALV-J, as described above. After being infected, the cells were incubated for 72 h, collected and fixed overnight with 75% ethanol at 4°C. The fixed cells were stained with 50 μg/mL propidium iodide solution (Sigma Life Science, St. Louis, MO, USA) containing 10 μg/mL RNase A (Takare) and 0.2% (v/v) Triton X-100 (Sigma), and they were incubated in the dark for 30 min at 37°C. The cell cycle phase analysis of the incubated cells was performed using a BD Accuri C6 flow cytometer (BD, San Jose, CA, USA).

SDS-PAGE and Western blotting were performed using methods that have been described in detail previously (Feng et al., 2016). The primary antibodies used were as follows: ALV-J-specific monoclonal antibody JE-9 (which was kindly provided by Prof. Aijian Qin, Yangzhou University, China) and mouse polyclonal anti-glyceraldehyde 3-phosphate dehydrogenase (anti-GAPDH; Bioworld Technology, Inc., USA). Goat anti-rabbit IgG (heavy and light chains (H+L))-horseradish peroxidase (HRP) and goat anti-mouse IgG (H+L)-HRP (Bioworld Technology, Inc., USA) served as secondary antibodies. The DF-1 cells were seeded in 6-well plates. When the cells grew to a density of 50% confluence, they were transfected, and then infected with ALV-J, as described above. After 3 days in culture, the DF-1 cells were subjected to the SDS-PAGE and Western blot analyses. Detection of HRP-conjugated secondary antibodies was performed using Western Lightning Chemiluminescence Reagent Plus (Perkin Elmer Life Sciences, Inc, Boston, MA, USA), and images were developed on X-ray films.

For the ELISA, the DF-1 cells were seeded in 96-well plates. When the cells grew to a density of 50% confluence, they were transfected, and then infected with ALV-J, as described above. At the indicated times (24, 48, 72, 96, and 120 h), the supernatants were collected. After three repetitions of freezing and thawing the supernatants, the ALV-J p27 protein expression level was measured using an ELISA ALV-J diagnosis kit, as described in the manufacturer's protocol (IDEXX, USA).

For the cell growth assays, DF-1 cells were seeded in 96-well plates. When the cells grew to a density of 50% confluence, they were transfected, and then infected with ALV-J, as described above. The viable cells were detected every 24 h using a CKK-8 (Dojindo, Kumamoto, Japan), in accordance with the manufacturer's protocol. Absorbance was measured using a Model 680 Microplate Reader (Bio-Rad, Hercules, California, USA) by optical density at a wavelength of 450 nm.

Each experiment was performed in triplicate. The data are presented as the mean ± the standard error of the mean (SEM) of each set of three independent experiments. Where applicable, the statistical significance of the data was tested using one-sample or paired t-tests. The types of tests and the P-values, when applicable, are indicated in the figure legends.

ALV-J, an oncogenic retrovirus, induces several leukemia-like proliferative diseases in the hemopoietic systems of chickens. Based on our previous miRNA sequencing data (GEO database accession numbers, GSE63676), miR-34b-5p was found to be significantly (P = 1.62E-71) upregulated in the spleens of ALV-J-infected chickens compared to the spleens of non-infected chickens (Figure 1A). In addition, RT-qPCR involving immune system-related tissues showed that the expression of miR-34b-5p was significantly increased (P = 0.0036 and P = 0.0197) in ALV-J-infected spleens and thymuses compared to non-infected spleens and thymuses, respectively (Figure 1B). In the liver, the expression of miR-34b-5p was also increased in ALV-J-infected chickens, although the difference was not significant (P = 0.09). These findings demonstrate that the expression of miR-34b-5p is significantly dysregulated in ALV-J-infected chickens and it might be involved in virus invasion.

Previous studies demonstrated that miRNAs can affect various aspects of cancer biology, including cell proliferation, migration, and apoptosis (Zhao et al., 2014; Shi W. et al., 2016). To explore the role of miR-34b-5p on ALV-J infection, we investigated its influence on cellular processes in ALV-J-infected cells. Firstly, we transiently transfected an miR-34b-5p mimic or a negative control mimic (empty plasmid; miR-NC) into DF-1 cells. After 12 h in culture media, the transfected DF-1 cells were infected with ALV-J, and then cell proliferation and migration analyses were carried out. The result of the CCK-8 assay showed that overexpression of miR-34b-5p could significantly (at P < 0.05 or P < 0.01) promote the proliferation of ALV-J-infected cells compared to miR-34b-5p negative control (miR-NC) at 24~96 h after the ALV-J infection (Figure 2A). Moreover, using flow cytometry, we found that overexpression of miR-34b-5p significantly increased the number of cells in S phase (P < 0.001) and reduced the number of cells in G2 phase (P < 0.05) compared to the negative control group (Figure 2B).

Subsequently, to assess the effect of miR-34b-5p overexpression on the cell migration, a wound healing assay was performed in ALV-J-infected DF-1 cells. We found that the migration rate of DF-1 cells transfected with miR-34b-5p mimics was significantly increased (P < 0.05 and P < 0.01 at 48 and 72 h after the ALV-J infection, respectively) compared to the negative control group (Figures 2C,D). Taken together, our data suggested that miR-34b-5p promotes the growth of ALV-J-infected DF-1 cells by accelerating their proliferation and migration.

Our previous RNA sequencing data (GSE63226) demonstrated that MDA5 was significantly downregulated (P = 1.34E-06) in the spleens of ALV-J-infected chickens compared to non-infected chickens (Figure 3A). The RT-qPCR results revealed that the expression of MDA5 was significantly downregulated in immune system-related tissues including the livers (P = 0.046), spleens (P = 0.0022), and thymuses (P = 0.022) of ALV-J-infected chickens compared to those from non-infected chickens (Figure 3B). In addition, miR-34b-5p had a contrasting expression pattern in ALV-J-infected chickens compared to non-infected chickens (Figures 1, 3A,B). Using miRNA target prediction software, miRDB (http://www.mirdb.org/miRDB) (Wong and Wang, 2015), MDA5 was predicted to be a potential target gene for miR-34b-5p. The miR-34b-5p mature sequence and its binding seed sequence in MDA5 are shown in Figure 3C. To confirm whether miR-34b-5p directly targets the 3′ UTR of MDA5, a dual-luciferase reporter gene assay was carried out in DF-1 cells. The transfection efficiency analysis showed that 50 and 100 nM of an miR-34b-5p mimic could significantly (P = 0.019) and highly significantly (P = 0.0027) decrease the MDA5 expression levels in DF-1 cells at 48 h after transfection, respectively (Figure 3D). Therefore, we co-transfected 100 nM miR-34b-5p mimic or mi-NC into DF-1 cells with reporter vectors. The dual luciferase assay showed that the wild-type group (WT-MDA5-3′UTR) induced a significant decrease (P < 0.05) in luciferase activity compared to the control group, whereas no significant difference was observed between the mutant-type group (MT-MDA5-3′UTR) and control group (Figure 3E). Taken together, these results suggested that MDA5 is a direct target of miR-34b-5p.

To explore whether the effect of miR-34b-5p on cell processes in ALV-J-infected cells was mediated by MDA5, we performed CCK-8, cell cycle and wound healing assays in MDA5 overexpression and knockdown DF-1 cells. Overexpression of MDA5 significantly inhibited (P ≤ 0.01) cell proliferation compared to the control group at 24–120 h after the ALV-J infection (Figure 4A). However, MDA5 knockdown significantly promoted (P ≤ 0.05) cell proliferation at 48–120 h after the ALV-J infection (Figure 4B). On the other hand, MDA5 overexpression significantly (P < 0.001) decreased the proportion of ALV-J-infected cells at S phase but notably increased the proportion of cells at G1 phase (P < 0.0001) and G2 phase (P < 0.05) compared to the control group (Figure 4C). In contrast, knockdown of MDA5 using siRNA led to a significant (P < 0.01) elevation in the proportion of cells at S phase, and a clear reduction in the proportion of cells at G1 phase (P < 0.05) or G2 phase (P < 0.01; Figure 4D). Moreover, the migration rate of DF-1 cells that overexpressed MDA5 was significantly decreased (P ≤ 0.01) compared to that in the control group (Figures 4E,F). In contrast, the migration rate of MDA5 knockdown cells was significantly increased (P ≤ 0.01) compared to those transfected with the small interfering RNA negative control (si-NC; Figures 4G,H). These results indicated that miR-34b-5p might modulate the proliferation, migration, and cell cycle of ALV-J-infected DF-1 cells by targeting MDA5.

The ALV RNA genome has a typical slowly-transforming replication-competent structure: 5′ R-U5-gag-pol-env-U3-R 3′ that contains three coding genes: Gag, Pol, and Env. The env gene codes for an envelope glycoprotein that determines subgroup specificity, neutralization, and receptor binding (Pan et al., 2012). When miR-34b-5p was normal expression (Figure 5A), the expression level of MDA5 was upregulated (Figure 5B) at 16, 24, and 48 h after infection compared to at 2 h after infection. In DF-1 cells infected with ALV-J, we found that the expression of the Env, Gag and Pol mRNA was significantly upregulated at 16, 24, and 48 h after infection compared to at 2 h after infection (P < 0.05; Figures 5C–E). To investigate whether chicken MDA5 signaling pathway was able to detect ALV-J infections and then to initiate and amplify an innate immune response, RT-qPCR was used to measure the changes in the expression of innate and antiviral genes in the MDA5 signaling pathway during ALV-J infections. The results showed that the levels of all the mRNA, including those associated with IPS-1, IRF-3, IFNβ, PKR, OAS, Mx1, and MHC class I were upregulated at 24 and 48 h after infection compared to at 2 h after infection (Figures 5F–L). In contrast, after overexpression of miR-34b-5p by miRNA mimic (Supplementary Figure 1A) and following inoculation with TCID₅₀ of ALV-J, the expression level of MDA5 was downregulated (Supplementary Figure 1B) in miR-34b-5p overexpression DF-1 cells compared with negative control (NC) group at 2-48 h ALV-J infection time points. In addition, the mRNA expression of ALV-J related genes, including Env, Gag and Pol (Supplementary Figures 1C–E, respectively) were upregulated in miR-34b-5p overexpression DF-1 cells compared with NC group during at 2–48 h ALV-J infection time points. Moreover, the mRNA expression of MDA5 signaling pathway-related innate and antiviral genes (IPS-1, IRF-3, IFNB, PKR, OAS, MX1, MHC class 1) were downregulated compared with control (Supplementary Figures 1F–L). These findings indicate that miR-34b-5p suppresses MDA5 signaling pathway to promote ALV-J replication.

To further elucidate the function of MDA5 on the MDA5 signaling pathway and ALV-J replication, ectopic expression, and knockdown of MDA5 was carried out in DF-1 cells. In DF-1 cells that overexpressed MDA5, the mRNA and protein expression of Env and the ALV-J virion secretion were significantly suppressed (P ≤ 0.05) and the mRNA expression of genes in the MDA5 signaling pathway was significantly upregulated (P ≤ 0.05; Figures 6A–D). In contrast, after knockdown of MDA5 by siRNA, the mRNA and protein expression of env were upregulated (P < 0.05; Figures 6E,F). The expression of genes in the MDA5 signaling pathway was significantly downregulated (P ≤ 0.05), and sample-to-positive ratios of virion secretion in the supernatants were elevated at 4 and 5 days after the infection (P ≤ 0.05; Figures 6G,H). These data suggest that MDA5 can recognize ALV-J infections and initiate and amplify an innate immune response via the MDA5 signaling pathway in chickens.

As our findings suggested that miR-34b-5p might promote the proliferation and migration of ALV-J-infected DF-1 cells by targeting MDA5, we assessed the effects of miR-34b-5p overexpression on the MDA5 signaling pathway to explore the mechanism by which miR-34b-5p contributes to ALV-J infections. In DF-1 cells that overexpressed miR-34b-5p, the mRNA and protein expression level of Env were significantly elevated 3 days after the infection (P < 0.01; Figures 7A,B). Moreover, ectopic expression of miR-34b-5p significantly inhibited the expression of genes in the MDA5 signaling pathway, including MDA5, IPS-1, IFNβ, OAS, Mx1, and MHC class I (P ≤ 0.05), in ALV-J-infected DF-1 cells (Figures 7A,C). The sample-to-positive ratio of virion secreted in the supernatants was also significantly upregulated (P ≤ 0.05) in DF-1 cells that overexpressed miR-34b-5p during ALV-J infections (Figure 7D). These data indicate that miR-34b-5p targets MDA5 to modulate the MDA5 signaling pathway and further promote ALV-J replication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.