MicroRNA-424/503 cluster members regulate bovine granulosa cell proliferation and cell cycle progression by targeting SMAD7 gene through activin signalling pathway

The mature sequences of miR-424-5p (miR-424) and miR-503-5p (miR-503) were obtained from the miRBase ( http://www.mirbase.org/ ) database. We performed an in silico target prediction for potential putative targets using the miRWalk database (http://​zmf.​umm.​uni-heidelberg.​de/​apps/​zmf/​mirwalk/​). The miRNA-mRNA binding site prediction in bovine sequences was performed using TargetScan 6.2 [34]. Target gene predictions were considered according to rank based on the predicted efficacy or targeting as calculated using ‘cumulative weighted context++ scores’ of the sites [34] and probability of conserved targeting (P_CT) [35]. Accordingly, among several genes, the SMAD7 [36] and ACVR2A genes, which are ubiquitously expressed in the ovarian follicle and important in reproductive performance [37], were selected for functional analysis. The secondary structure of miR-424 and miR-503 was predicted by RNAhybrid ( http://bibiserv.techfak.uni-bielefeld.de/rnahybrid ).

Bovine ovaries as sources of bGCs were collected from a local slaughterhouse. Ovaries were processed to obtain follicular fluid and isolation of granulosa cells as described previously [12]. Further, a total of 2.0–2.5 × 10⁵ bGCs per well were seeded into CytoOne® 24-well plate (Starlab International GmbH, Germany) in the F12+ culture media. The bGCs were cultured in 37 °C with 5% CO₂ in humidified environment. The bGCs were incubated for 48 h to attach and pre-confluent (60–70%) for treatment or transfection purpose. In the culture medium FSH, IGF1 or other factors were not added to avoid its effect on bovine granulosa cell proliferation. In some experiments cells were cultured in the presence of Recombinant Human/Mouse/Rat Activin A (R&D Systems, MN, USA).

The chemically synthesized miRNA-424-5p mimic and inhibitor, miR-503-5p mimic and inhibitor, and the corresponding negative controls (NC) were used to transfect (Qiagen GmbH, Germany) bGCs. The miRNAs and/or plasmids were diluted in Opti-MEM I reduced-serum medium (Invitrogen). Sub-confluent cultured bGCs (70–80% confluent) were co-transfected with 500 ng of the wild-type or mutant-construct plasmid and 50 nM individual microRNA mimic or mimic control. For miR-424/503 gain- and loss-of-function analysis, 50 nM individual microRNA mimic, inhibitor or corresponding negative controls were co-transfected to sub-confluent cultured bGCs. The transfection was performed using Lipofectamine 2000 transfection reagent (Life Technologies, Germany).

To validate whether the SMAD7 and ACVR2A gene are real targets of the miR-424/503 cluster, fragments of the 3´-UTR of SMAD7 or 3´-UTR of ACVR2A containing the binding sites for miR-424-5p (miR-424) and miR-503-5p (miR-503) (wild type) or with mutations in the seed sequences of miR-424/503 (mutant type) (Fig. 1) were cloned and inserted between the SacI and XhoI restriction sites of the pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega Corporation, USA). The cDNA from ovarian bGCs was used to amplify the predicted miRNA-mRNA binding site in the 3´-UTR region of the SMAD7 or ACVR2A mRNA. Specific primers and 50-mer mutated oligonucleotides were designed based on bovine SMAD7 (XM_005224232.3) or ACVR2A (NM_174227) mRNA sequences in GenBank (Additional file 1: Table S1). The luciferase activity was measured 48 h after transfection using the pmirGLO Dual Luciferase® Reporter Assay System (Promega Corporation, USA) according to the manufacturer’s protocol. Firefly and Renilla luciferase activity was detected by measuring the absorbance on a Centro LB 960 Microplate Luminometer (Berthold Technologies GmbH, Germany).

To confirm the expression of target and marker genes in treated bGCs at each stage of the experiment, harvested bGCs were resuspended in lysis buffer, and the subsequent RNA isolation was performed using an miRNeasy® mini kit (Qiagen GmbH, Germany) following the manufacturer’s protocol. Total RNA quantity and purity (260/280 ratios) was measured with a NanoDrop 8000 spectrophotometer (NanoDrop products, USA). After validating the quality and concentration of the RNA samples, the cDNA synthesis was performed using a RevertAid first stand cDNA synthesis kit (Thermo Fisher Scientific, Germany). Briefly, each RNA sample (1 μg) was co-incubated with 1 μl of Oligo (dT)₁₈ primer and dH₂O to a total of 11 μl at 65 °C for 5 min and then chilled on ice for 5 min. The reverse transcription was performed in a total mixture volume of 20 μl consisting of 4 μl of 5× Reaction Buffer, 1 μl of RevertAid Reverse Transcriptase, 2 μl of dNTP mix and 1 μl of RiboLock RNase Inhibitor. The reactions were carried out in a thermocycler programmed at 37 °C for 60 min followed by 70 °C for 5 min.

Primers for specific genes were designed using the Primer-BLAST program (http://​www.​ncbi.​nlm.​nih.​gov/​tools/​primer-blast/​). The details of the primers are described in Additional file 1: Table S2. The specificity of each primer amplicon was confirmed by sequencing the PCR products using a GenomeLab GeXP Genetic Analysis System (Beckman Coulter GmbH, Germany). The qRT-PCR analysis of mRNA was performed in an Applied Biosystem® StepOnePlus™ System (Thermo Fisher Scientific Inc., USA), using iTaq™ Universal SYBR® Green Supermix (Bio-Rad Laboratories GmbH, Germany), with the program described previously [23, 29] The mRNA expression data were analysed using the comparative Ct (2^-∆∆Ct) method [38], and β-ACTIN was used as an internal control.

Candidate miRNAs were quantified as described previously [23, 29]. Briefly, cDNA was synthesized using 80 ng of miRNA-enriched total RNA using a miRCURY LNA™ Universal cDNA synthesis kit (Exiqon, Denmark) according to the manufacturer’s instructions. The synthesized cDNA was diluted 40× and used for the qRT-PCR analysis of candidate miRNAs using ExiLENT SYBR Green Master mix (Exiqon, Denmark). The thermal cycling program was used as described previously [23, 29]. The specificity of the miRNA amplification was evaluated by melting curve analysis. The 5 s ribosomal RNA (5 s rRNA) (miRCURY LNA™ Universal RT microRNA PCR) was used as the reference gene primer. The qRT-PCR data were analysed using the comparative Ct (2^-∆∆Ct) method [38].

Total protein from cultured bGCs was isolated using 1× PLB (Promega Corporation, USA). The total protein concentration was determined using the Bradford method [39]. Western blotting was performed as described previously [12, 13]. The antibodies used were (Santa Cruz Biotechnology Inc., Germany): anti-ACTR-II2A goat polyclonal antibody (product no. sc-5667), anti-SMAD7 rabbit polyclonal antibody (product no. sc-11,392), anti-PCNA rabbit polyclonal antibody (product no. sc-7907), anti-STAR rabbit polyclonal antibody (product no. sc-25,806), anti-SMAD2/3 rabbit poly clonal antibody (product no.sct-5678), anti-psmad2/3 rabbit monoclonal antibody (sct-8828) or anti-β-ACTIN mouse monoclonal antibody (product no. sc-47,778). At the end of the incubation period, the membrane was washed six times with 1× TBST and incubated with the corresponding donkey anti-goat, goat anti-rabbit, or goat anti-mouse secondary antibody conjugated to horseradish peroxidase (Santa Cruz Biotechnology). The detection of the protein signal was then performed using Clarity Western ECL Substrate (Bio-Rad Laboratories). The relative band intensity was determined by ImageJ program (https://​imagej.​nih.​gov/​ij/​).

A total of 2 × 10⁴ bGCs per well were seeded into a 96-well plate as described previously [12, 13] Individual miR-424/503 mimics, inhibitors or corresponding controls were transfected into sub-confluent cultured bGCs (70–80% confluent). After 48 h of incubation, 10 μl of CCK-8 kit solution (Dojindo EU GmbH, Germany) was added to each well, and the plate was incubated for another 2 h. The optical density (OD) at a wavelength of 450 nm was measured using a Synergy™ H1 Multi-Mode Reader (BioTek Instruments Inc., Germany).

Cultured granulosa cells were transfected with 75 nM miR-424/503 cluster miRNA mimics, inhibitors, or the corresponding negative control (NC) and SMAD7 siRNA, ACVR2A siRNA or NC siRNA. The cells were trypsinized 48 h later and collected in a 15-mL Falcon tube (Thermo Fisher Scientific, Germany), followed by centrifugation at 750×g for 5 min and washing twice with 1 x CMF-PBS. A minimum of ~ 1 × 10⁶ cells were fixed in ice-cold 70% ethanol at 4 °C overnight. The cells were then centrifuged briefly, and the cell pellet was washed twice with 500 μl of 1× CMF-PBS. The cells were then labelled with 50 μg/mL propidium iodide (PI) and treated with 50 μg/mL RNase. The cells were then incubated at 37 °C for 30 min and processed in a BD LSRFortessa™ flow cytometer (BD Biosciences). The cell cycle distribution was analysed using ModFit LT software (http://​www.​vsh.​com/​products/​mflt/​index.​asp).

Bovine specific LNA™ longRNA GapmeRs (Exiqon, USA) were used to inhibit the expression of SMAD7 (5´-TTCGCAGAGTCGGCTA-3′ and 5´-CGATTTTGCTCCGTA-3′) ACVR2A (5´-GTTACTGGATTCGACG-3′ and 5´-GTTGGTCAGTAATCTA-3′). The transfection of 75 nM SMAD7 siRNA, ACVR2A siRNA or control siRNA was performed as described above. The gene expression analysis and cell proliferation assays were performed as described above.

We demonstrated in our previous study that miR-424/503 cluster miRNAs were upregulated in the granulosa cells of preovulatory dominant follicles at day 19 of the oestrous cycle. MicroRNA-424/503 cluster members are transcribed from an intergenic region of chromosome X, which is polycistronic in nature and evolutionary conserved in mammals. The precursors of bta-miR-424-5p (miR-424) and bta-miR-503-5p (miR-503) are 96 bp and 83 bp, respectively. The mature sequences of bta-424-5p and miR-503-5p are CAGCAGCAAUUCAUGUUUUGA and UAGCAGCGGGAACAGUACUG, respectively, which are conserved in other mammalian species. Since miRNAs regulate biological functions by targeting the 3´-UTR of genes post-transcriptionally in a sequence-specific manner, we performed a computational prediction using TargetScan (http://​www.​targetscan.​org) to generate an algorithm to identify putative targets, which revealed that miR-424 and miR-503 target SMAD7 and ACVR2A. On the other side, miRNA prediction for SMAD7 target gene using TargetScan revealed that miR-424-5p was top predict with 8mer site type (an exact match to positions 2–8 of the mature miRNA (the seed + position 8) followed by an ‘A’); while, miR-503-5p showed 7mer-A1 site type (an exact match to positions 2–7 of the mature miRNA (the seed) followed by an ‘A’). The validation of the putative target genes, SMAD7 and ACVR2A, of the miR-424/503 cluster was performed by measuring the luciferase activity of an expression vector carrying the 3´-UTR of the SMAD7 and ACVR2A gene, which contains the miRNA binding sites. The luciferase activity in bovine granulosa cells co-transfected with miR-424 and miR-503 mimics and a plasmid vector harbouring the wild-type SMAD7 and ACVR2A 3’-UTR was significantly reduced compared to bGCs transfected with a control for the individual miRNA mimics and the plasmid vector with the wild-type SMAD7 and ACVR2A 3´-UTR (P < 0.05; Fig. 2). However, there was no significant reduction in luciferase activity in bGCs co-transfected with individual miRNA mimics or a control and a plasmid vector harbouring a mutated SMAD7 and ACVR2A 3´-UTR sequence.

To obtain further insights into the role of the miR-424/503 cluster in granulosa cell function by regulating the expression of SMAD7 and ACVR2A gene, we transfected cultured bGCs with either the miR-424/503 mimics, inhibitors or the corresponding controls at a concentration of 75 nM. The expression level of the SMAD7 and ACVR2A mRNA was determined by qRT-PCR analysis 48 h after transfection. Transfection of the miR-424/503 cluster members mimic resulted in a significant reduction in the relative abundance of SMAD7 (P < 0.01) and ACVR2A (P < 0.05) mRNA compared to the controls (Fig. 3a). Transfection of miR-424/503 inhibitors did not affect, the expression of SMAD7 mRNA while the expression of ACVR2A mRNA was increased. The western blot analysis showed that bGCs transfected with miR-424/503 cluster mimics had no significant reduction in SMAD7 and ACVR2A protein levels compared to those transfected with negative controls The miR-424/503 cluster inhibitors had no significant change in SMAD7 and ACVR2A protein expression compared to those transfected with controls (Fig. 3b).

MiR-424 mimic transfection in cultured bGCs significantly increased the proliferation activity in cells transfected with a miRNA mimic negative control (Fig. 4a). However, miR-503 had no significant effect on cell proliferation. Conversely, compared to the miRNA inhibitor control, bGCs transfected with miR-424 and miR-503 inhibitors showed no significant differences in proliferation activity. Moreover, the cell proliferation phenotype was accompanied by the higher expression level of the proliferation marker gene PCNA in the bGCs transfected with the miR-424 mimic compared to those transfected with the mimic control (P < 0.01). However, transfection of inhibitors resulted in a slight reduction in the expression of PCNA (Fig. 4b). The PCNA protein expression showed an increasing tendency in bovine granulosa cells transfected with the miR-424/503 cluster mimic (Fig. 4c).

As a marker of differentiation, the expression of the STAR gene was investigated after modulation of the miRNA cluster. Transfection of miR-424/503 cluster mimic resulted in a significant reduction in STAR gene expression compared to the negative control (Fig. 4b, c). These results suggest that miR-424 promotes bovine granulosa cell proliferation and decreases the terminal differentiation of the bGCs, which leads to follicular survival and development.

Furthermore, we investigated the role of miR-424/503 on downstream members of the activin signalling pathway; the western blot results showed that the expression of the SMAD2/3 proteins tended to increase after the overexpression of the miR-424 mimic compared to the negative control (4D). Moreover, the overexpression of the miR-424 mimic did not change the phosphorylated SMAD2/3 protein level (4D). Overexpression of miR-503 mimic could not increase the expression of SMAD2/3 or phosphorylated SMAD2/3 protein level compared to mimic negative control (4D).

To further confirm the proliferation results and to understand the role of the miR-424/503 cluster in modulating the cell cycle of bGCs, we performed flow cytometric analysis after transfection of the miR-424/503 cluster members. The overexpression of the miR-424 mimic resulted in a significant reduction (P < 0.001) in the number of cells in G0/G1-phase, while a significant increase (P < 0.01) in the percentage of cells in S-phase was observed compared to cells transfected with the mimic negative control (Fig. 5g). This result indicated that the overexpression of miR-424 could increase bovine granulosa cell proliferation by promoting G1- to S-phase cell cycle transition. In contrast, the miR-424/503 cluster inhibitor did not cause any measurable changes in the cell cycle profile of bovine granulosa cells (Fig. 5h).

In order to investigate the involvement of SMAD7 in bGCs proliferation and differentiation, and to further validate the regulatory role of the miR-424/503 cluster in SMAD7 expression, we performed an independent experiment by knocking down the expression of SMAD7 using small interfering RNA (siRNA). Bovine GCs transfected with the SMAD7 siRNA showed a significant reduction in the expression of both SMAD7 mRNA (P < 0.05) and protein compared to cells transfected with the siRNA negative control (Fig. 6a, e). The transfection of bGCs with the SMAD7 siRNA significantly enhanced proliferation of bGCs compared to siRNA negative control (Fig. 6b), which was accompanied by noteworthy change in the expression of the marker of proliferation PCNA (Fig. 6c). However, flow cytometric analysis showed reduction (P < 0.001) in the number of cells in G0/G1-phase (Fig. 6j), while a significant increase (P < 0.01) in the percentage of cells in S-phase (Fig. 6j) was observed compared to cells transfected following knocking down of SMAD7 with siRNA compared to negative control. Further, the suppression of the SMAD7 gene using siRNA had no effect on the expression of the STAR gene and STAR protein (Fig. 6d, g). Western blot analysis showed that the expression of SMAD2/3 and phosphorylated SMAD2/3 was slightly increased in siRNA-transfected cells compared to cells transfected with the negative control (Fig. 6f).

Similarly, to investigate the possible involvement of ACVR2A in bGCs proliferation and differentiation and to further validate the regulatory role of the miR-424/503 cluster in ACVR2A expression, we suppressed ACVR2A using small interfering RNA (siRNA). Bovine GCs transfected with the ACVR2A siRNA showed a significant reduction in the expression of both ACVR2A mRNA (P < 0.01) and protein compared to cells transfected with the siRNA negative control (Fig. 7a, e). The transfection of bGCs with the ACVR2A siRNA did not result in any measurable effect on the proliferation of bGCs compared to control siRNA-transfected bGCs (Fig. 7b). Similarly, there was no noteworthy change in the expression of the marker of proliferation PCNA following the suppression of the ACVR2A gene (Fig. 7c). However, western blot analysis showed reduced expression of the PCNA protein (Fig. 7f). Flow cytometric analysis showed no change in G1/S cell cycle transition (Fig. 7j). Further, the suppression of the ACVR2A gene using siRNA had no effect on the expression of the STAR gene (Fig. 7d). Western blot analysis showed that the expression of SMAD2/3 and phosphorylated SMAD2/3 was slightly decreased in siRNA-transfected cells compared to cells transfected with the negative control (7G).

Activin has been implicated in various intra-ovarian roles, including granulosa cell proliferation, follicular luteinization retardation and the recruitment of R-SMADs to activate the activin signalling pathway. Here, we investigated the effect of Activin A treatment on the proliferation of granulosa cells and on miR-424/503 cluster expression. We treated bGCs with different doses of Activin A (25 ng/mL, 50 ng/mL and 100 ng/mL). We found that Activin A treatment increased proliferation of bovine granulosa cells in dose-dependent manner (Fig. 8a). Activin A treatment significantly decreased (P < 0.001) the STAR expression compared to control cells (Fig. 8c), which indicates a decrease in bGCs differentiation. Further, we also noticed that Activin A treatment in dose dependent manner significantly decreased the expression of SMAD7 (P < 0.01; Fig. 8e); however, there was no change in the expression of ACVR2A (Fig. 8f). Interestingly, we observed a dose-dependent reduction in miR-424/503 expression upon treatment with Activin A. miRNA-424 was significantly decreased (P < 0.05; Fig. 8g) at the 100 ng/mL dose of Activin A. This result indicates that there might be negative feedback loop between Activin A and the miR-424/503 cluster. These results indicated that there could be possibility that miR-424/503 might be involved in fine tuning of the activin signalling pathway.