The NF-κB-modulated microRNAs miR-195 and miR-497 inhibit myoblast proliferation by targeting Igf1r, Insr and cyclin genes

MicroRNAs (miRNAs; miR) were first discovered in 1993 (Lee et al., 1993; Wightman et al., 1993) and have since attracted global attention to their various functions in a broad range of biological processes. Skeletal-muscle-specific miRNAs, including miR-1, miR-133, miR-206 and miR-208b/499 (Chen et al., 2006; Kim et al., 2006; McCarthy et al., 2009; Yang et al., 2015), muscle-enriched miRNAs including miR-486 (Xu et al., 2012), and ubiquitous miRNAs including miR-21 (Bai et al., 2015) have all been shown to play important roles in myogenesis.

miR-195 and miR-497 are closely linked in the genome and have been shown to share the same seed sequences. Both belong to the miR-15 family, which contains two more members, namely miR-15 and miR-16. In this family, miR-15 and miR-16 have been reported to target the well-known anti-apoptotic gene Bcl2 (Cimmino et al., 2005). Therefore, members of the miR-15 family are thought to play some roles in tumorigenesis. Consistent with this notion, miR-195 and miR-497 have been shown to be associated with various types of cancers. For example, miR-195 has been shown to act as a tumor suppressor in gastric cancer (Deng et al., 2013), hepatocellular carcinoma (Ding et al., 2013; Xu et al., 2009), bladder cancer (Lin et al., 2012), breast cancer (Luo et al., 2014; Singh and Saini, 2012), leukemia (Diaz-Beya et al., 2012) and other cancer types. miR-497 and miR-195 play similar roles in most physiological processes and miR-497 has also been found to be involved in tumorigenesis (Waters et al., 2012). The CCND1, CCND3, CCNE1, CDC25A and CDK4 genes are usually shown as being the targets of miR-195 and miR-497 in some cancer cells (Furuta et al., 2013; Li et al., 2011). Therefore, miR-195 and miR-497 are potential therapeutic targets in many types of cancers (Furuta et al., 2013; Guo et al., 2012; Itesako et al., 2014; Li et al., 2011; Ozata et al., 2011). Additionally, miR-195 has been implicated in neurodevelopment (Ai et al., 2013; Liu et al., 2013; Zhu et al., 2012), placental pathogenesis (Bai et al., 2012; Hu et al., 2009) and glucose metabolism (Joglekar et al., 2007; Ortega et al., 2014), suggesting its importance in a range of biological processes.

Previous studies have suggested that miR-195 and miR-497 participate in muscle development. In a previous work from our laboratory, we found that both miR-195 and miR-497 were significantly upregulated in longissimus muscle tissue during muscle development in pigs (Huang et al., 2008). Moreover, miR-195 exhibited higher expression in the gluteus medius in horses suffering from polysaccharide storage myopathy (PSSM) or recurrent exertional rhabdomyolysis (RER) (Barrey et al., 2011). Furthermore, miR-195 might regulate the proliferation and apoptosis of cardiomyocytes, and it is also associated with heart development and heart disease; for example, hypertrophy, heart failure and lipotoxic cardiomyopathy (Porrello et al., 2011; van Rooij et al., 2006; Zhu et al., 2011). Finally, miR-195 and miR-497 have been reported to induce quiescence of satellite cells at the postnatal stage in mice (Sato et al., 2014). However, the direct roles of miR-195 and miR-497 in skeletal muscle development remain largely undefined.

In this study, we examined the role of the miR-195 and miR-497 cluster in skeletal muscle development. Our data provide insight into the targets of these miRNAs and demonstrate the involvement of nuclear factor κB (NF-κB) in the regulation of this miRNA cluster. Furthermore, our results support the idea that miRNAs play important roles in skeletal muscle growth and development.

First, we used quantitative real-time polymerase chain reaction (qPCR) to analyze miR-195 and miR-497 expression patterns during myogenesis in C2C12 cells that were stimulated to differentiate into myotubes (Fig. 1A), and observed an upregulation of miR-195 (Fig. 1B). Consistent with this finding, miR-195 was also gradually upregulated during muscle development from 2 days to 6 weeks of age in the thigh muscles of mice (Fig. 1C). miR-497 exhibited a similar expression profile, but the fold change of miR-497 was lower than that of miR-195 in muscle tissue (Fig. 1B,C). Thus, expression of miR-195 and miR-497 was upregulated during skeletal muscle growth and myoblast differentiation.

To determine the roles of miR-195 and miR-497 in myoblasts, miR-195 and miR-497 were overexpressed by transfecting synthetic miRNA mimics in C2C12 cells, and the cell cycle distribution was determined by flow cytometry. The flow cytometry results showed that the proportion of cells arrested in the G₁ or G₀ phase was significantly higher in cells transfected with miR-195 or miR-497 mimics than in cells transfected with the negative control (NC) mimics, and the proliferation index was also decreased following the transfection of miR-195 or miR-497 mimics (P<0.01; Fig. 2A). These results indicate that both miR-195 and miR-497 inhibit the proliferation of myoblasts by blocking the G₁-S transition.

Additionally, we confirmed the effects of miR-195 and miR-497 on myoblast proliferation with 5-ethynyl-2′-deoxyuridine (EdU) assays. First, a miR-195 and miR-497 overexpression vector, named miR-195/497-C1, was constructed by inserting a 738-bp DNA fragment, which contained the miR-195 and miR-497 primary transcript cluster, into the pEGFP-C1 vector. qPCR results indicated a significant overexpression of both miR-195 and miR-497 following the transfection of C2C12 cells with the miR-195/497-C1 construct (P<0.01; Fig. 2B). Moreover, EdU staining showed that the proportion of EdU-positive cells was significantly lower following transfection with miR-195/497-C1 than with pEGFP-C1 (P<0.01; Fig. 2C,D). These results indicate that DNA synthesis was inhibited in myoblasts when miR-195 and miR-497 were overexpressed. Taken together, the above results confirmed that miR-195 and miR-497 could inhibit the proliferation of myoblasts.

To elucidate the mechanisms through which miR-195 and miR-497 affected cell proliferation, the potential targets were predicted by TargetScan (http://www.targetscan.org/). Genes encoding cyclin E1 and cyclin D2 (Ccne1 and Ccnd2, respectively), which were found to contain multiple binding sites for miR-195 and miR-497 at their 3′ untranslated regions (UTRs), were primarily selected as potential candidates (Fig. 3A).

The dual luciferase reporter system was first used to analyze the interaction between miR-195 and/or miR-497 and these two candidate genes. Fragments of 3′UTR containing the binding sites were inserted into the psiCHECK-2 vector and then the constructs were transfected into C2C12 cells together with miRNA mimics. Overexpression of miR-195 and miR-497 resulted in decreased relative luciferase activity when the proximal binding site of Ccne1 3′UTR was inserted. However, no significant difference was detected when the distal binding site was inserted (Fig. 3B). Moreover, overexpression of miR-195 and miR-497 decreased the relative luciferase activity when the proximal or distal binding sites of Ccnd2 3′UTR were inserted (Fig. 3C). These results indicate that miR-195 and miR-497 can interact with the Ccne1 and Ccnd2 3′UTR. Furthermore, western blot analysis demonstrated that cyclin E1 and cyclin D2 proteins were decreased when miR-195 or miR-497 was overexpressed (Fig. 3D). Therefore, miR-195 and miR-497 could directly target the Ccne1 and Ccnd2 cell cycle genes in myoblasts.

In addition to cell-cycle-related genes, we also predicted that other genes related to cell growth might be targeted by these miRNAs. TargetScan prediction results showed that the 3′UTR of insulin-like growth factor 1 receptor (Igf1r or IGF-1R) contained one binding site for miR-195 and miR-497, whereas that for insulin receptor (Insr) contained two binding sites for miR-195 and miR-497 (Fig. 4A).

Overexpression of miR-195 or miR-497 significantly decreased the luciferase activity of the construct containing the binding site from the Igf1r gene as compared to that of the NC control (P<0.01; Fig. 4B). Mutant miR-195 or miR-497 generated through mutation of the seed sequence did not inhibit the luciferase activity (Fig. 4B). This result indicates that miR-195 and miR-497 can specifically interact with the Igf1r gene. For the Insr gene, miR-195 significantly inhibited both binding sites (P<0.01, P<0.05), whereas miR-497 significantly inhibited the first binding site (P<0.05), but not the second binding site (Fig. 4B). This result indicates that miR-195 and miR-497 target the Insr gene. Furthermore, the western blotting results confirmed that miR-195 and miR-497 inhibited the expression of Igf1r and Insr protein in C2C12 cells (Fig. 4C). Taken together, these results demonstrate that miR-195 and miR-497 directly target the Igf1r and Insr 3′UTRs in C2C12 cells.

Several studies indicate that alterations in Igf1r and/or Insr genes in modified L6 myoblasts or muscle satellite cells affect the sensitivity of these cells to growth-related hormone, thereby altering cell proliferation (Arabkhari et al., 2010; Bonnesen et al., 2010; Dong et al., 2013; Kim et al., 2005; Quinn and Roh, 1993). Here, we also studied the effects of Igf1r and Insr on the proliferation of C2C12 myoblasts. First, we used flow cytometry to analyze the effects of Igf1r or Insr knockdown on the cell cycle distribution in C2C12 cells. More cells were arrested in the G₁-G₀ phase in cells transfected with small interfering RNA (siRNA) against Igf1r1 (si-Igfr-1) and Insr1 (si-Insr-1) than in control-transfected cells (Fig. 5A,B; P<0.05). Moreover, the proliferation index was also decreased following knockdown of Igf1r or Insr (Fig. 5C, P<0.05). These results indicate that downregulation of Igf1r and Insr blocks the G₁-to-S phase transition in C2C12 cells. Additionally, the interaction between miR-195/497 and the IGF-1/insulin signaling pathway was also studied. An EdU incorporation assay indicated that the proportion of EdU-positive cells was significantly decreased in cells transfected with miR-195/497-C1 as compared with control cells. Moreover, following stimulation with IGF-1, the proportion of EdU-positive cells was slightly increased in cells transfected with miR-195/497-C1, but the level was still lower than that in control cells (Fig. 5D). Taken together, these results imply that Igf1r and Insr promotes C2C12 myoblast proliferation, and that miR-195 and miR-497 inhibit myoblast proliferation through targeting the Igf1r and Insr genes.

Next, we evaluated the roles of miR-195 and miR-497 during myoblast differentiation. Following transfection of miR-195, miR-497, or negative control mimics, differentiation was induced in C2C12 cells for 72 h, and the fused myotubes were stained for the myogenic marker myosin (Fig. 6A). The number of fused myotubes in each treatment was analyzed (Fig. 6B). There were no obvious differences in differentiation between cells transfected with miR-195 or miR-497 mimics and cells transfected with negative control mimics. Moreover, qPCR showed that the expression of the myogenic marker genes myogenin, MyHC (also known as myosin heavy chain 1, MYH1) and MCK (also known as CKM) was not significantly different among groups (Fig. 6C). These results indicate that myogenic differentiation was not affected by miR-195 and miR-497 in C2C12 cells.

Because miR-195 and miR-497 play roles in regulating myoblast proliferation, we next investigated their transcriptional mechanism. Based on their genomic structure and expression pattern, we hypothesized that miR-195 and miR-497 were transcribed from the same primary transcript and share the same regulatory mechanism during myogenesis. To test this hypothesis, we first predicted the transcription start site (TSS) of the Mus musculus miR-195 and miR-497 cluster according to the gene annotation database of humans. Then, the potential binding sites for transcription factors in a 3-kb promoter region of the miR-195 and miR-497 cluster were predicted using TFSEARCH software. According to these results, three potential binding sites for p65 (also known as RelA, the active subunit of NF-κB), located at −604, −337 and +106 bp, were found in the promoter region (Fig. 7A). These binding sites share highly conserved nucleotides with the consensus sequence (GGGRNNYYC) for p65 (Fig. 7B). Furthermore, a 3090-bp fragment harboring these three p65-binding sites was cloned into the pGL3 luciferase reporter vector. Deletion mutations showed that the relative luciferase activity of the miR-195 and miR-497 promoter was significantly elevated when the first p65-binding site (−604 bp) was deleted. However, no significant variation was observed when the second (−337 bp) or third (+106 bp) binding sites were deleted (Fig. 7B). As expected, a strong gel-shift signal was detected when the first p65-binding site probe was added in the C2C12 nuclear extracts in electrophoretic mobility shift assays (EMSAs). However, there was almost no gel-shift signal for the probes of the second or third p65-binding sites (Fig. 7C). The shift band signal generated using the probe for the first p65-binding site was decreased when excess unlabeled competitor probe was added, but was not affected by the same amount of mutant competitor probe (Fig. 7D). Moreover, this gel-shift band was intensified in a concentration-dependent manner when nuclear extract was added, and was dramatically reduced when the 10-fold to 100-fold excess of unlabeled competitor probe was added (Fig. 7E). Chromatin immunoprecipitation (ChIP)-PCR was then performed to detect the binding of p65 with these three sites in C2C12 cells. Similar to the results of the luciferase and gel-shift assays, ChIP-PCR showed that the DNA fragments containing the first binding site were pulled down by anti-p65 antibodies (Fig. 7F). Therefore, these results confirm that NF-κB can directly bind to the specific site of the promoter of the miR-195 and miR-497 cluster in C2C12 cells.

After confirming the binding of NF-κB to the promoter of the miR-195 and miR-497 cluster, we studied the relationship between NF-κB and the expression of miR-195 and miR-497. Immunofluorescence staining showed that p65 was highly expressed in the nucleus of myoblasts, but weakly expressed in the nucleus and predominantly expressed in the cytoplasm of myotubes (Fig. 8A). These results indicate that NF-κB is deactivated during myoblast differentiation. By contrast, the expression of miR-195 and miR-497 was opposite to the activity of NF-κB during myogenesis. Taken together with the above results showing that NF-κB can directly bind to their promoter region, we hypothesized that NF-κB could directly regulate the expression of miR-195 and miR-497. To test this hypothesis, we studied the relationship between the activity of NF-κB and the promoter activity of the miR-195 and miR-497 cluster in vitro and in vivo. Overexpression of p65 significantly increased NF-κB activity but significantly reduced the promoter activity of miR-195 and miR-497 cluster in C2C12 cells (Fig. 8B,C). In contrast, knockdown of p65 decreased NF-κB activity but enhanced the promoter activity of miR-195 and miR-497 cluster (Fig. 8B,D). Furthermore, qPCR showed that miR-195 and miR-497 expression was decreased when p65 was overexpressed and increased when p65 was knocked down (Fig. 8E).

For the in vivo study, lipopolysaccharides (LPS) was used to activate NF-κB. Expression of miR-195 and miR-497 in the skeletal muscle tissue was significantly lower than that in the phosphate-buffered saline (PBS) control group 24 h after LPS treatment, whereas a decreasing tendency that did not reach significance was observed at 48 h after LPS treatment (Fig. 8F). Western blot results showed that, along with upregulation of p65, the protein levels of Igf1r and Insr, targets of miR-195 and miR-497, were also increased at 24 h after LPS treatment (Fig. 8G). These results indicate that transcription of miR-195 and miR-497 is also negatively regulated by NF-κB in the physiological context.

Taken together, these results demonstrate that NF-κB can directly downregulate miR-195 and miR-497 in both myoblasts and skeletal muscle tissue.

Previous studies have demonstrated that miR-195 and miR-497 play important roles in tumorigenesis, and miR-195 and miR-497 have also been shown to be involved in muscle development. In this study, we observed that miR-195 and miR-497 were upregulated during myogenesis in vivo and in vitro, consistent with our previous study in pigs (Huang et al., 2008). Moreover, we found that overexpression of miR-195 and miR-497 could prohibit myoblast proliferation. These results indicate that miR-195 and miR-497 play important roles in myogenesis.

Our present study identified Ccnd2 as a new target of miR-195 and miR-497, besides Ccne1, in C2C12 cells. These target genes are important for cell cycle progression, which might at least partially provide a basis for the proliferation inhibition mediated by miR-195 and miR-497 in myoblasts. Consistent with our results, recent studies have shown that miR-195 and miR-497 act as tumor suppressors by inhibiting multiple cell cycle regulators. miR-195 has been shown to directly target CCND1 and CCNE1 in human glioma and breast cancer (Hui et al., 2013; Luo et al., 2014), and both miR-195 and miR-497 have been shown to target CCND1, CCND3, CCNE1, CDC25A and CDK4 in human breast cancer and hepatocellular carcinoma (Furuta et al., 2013; Li et al., 2011). These results indicate that miR-195 and miR-497 can target many cell cycle genes and inhibit cell proliferation in both pathological and physiological conditions.

More importantly, the present study identified Igf1r and Insr as new targets of miR-195 and miR-497 in C2C12 cells. We confirmed that miR-195 and miR-497 could inhibit proliferation of C2C12 by targeting IGF1 signaling. Many studies have confirmed that the IGF-1 and insulin signaling pathways play important roles in the proliferation of myoblasts and the progression of myogenesis. IGF-1 and insulin have been shown to stimulate the proliferation and differentiation of muscle cells in a time- and concentration-dependent manner (Allen et al., 1985; Cassar-Malek et al., 1999; Cen et al., 2008; Engert et al., 1996; Grabiec et al., 2014); these important mediators also modulate protein metabolism (Castillo et al., 2004; Crown et al., 2000; Fryburg, 1994). Moreover, the myogenesis program can be influenced when the expression of IGF-1 and insulin receptors are altered (Arabkhari et al., 2010; Bonnesen et al., 2010). In rat and bovine myoblasts, overexpression of Igfr1 has been shown to enhance cell sensitivity to IGF-1 and to increase the proliferation and differentiation of the cells (Quinn and Haugk, 1996; Quinn and Roh, 1993). In contrast, satellite cells isolated from Igf1r^+/− mouse muscles showed reduced proliferation and differentiation (Dong et al., 2013). It has also been shown that MKR mice, which have impaired Igf1r and Insr function, did not respond to growth hormone treatment and fail to exhibit increased muscle mass, proliferation and differentiation of satellite cells (Kim et al., 2005). Therefore, this study provides a new mechanism underlying the inhibition of myoblast proliferation by miR-195 and miR-497 through targeting Igf1r and Insr genes, which might explain the recent finding that miR-195 and miR-497 induce quiescence of satellite cells (Sato et al., 2014).

In this study, we also found that transcription of the miR-195 and miR-497 cluster was negatively regulated by NF-κB in both myoblasts and skeletal muscle tissue. Previous studies have indicated that NF-κB is constitutively activated in myoblasts, and its expression is decreased following the differentiation of C2C12 cells (Bakkar et al., 2008; Guttridge et al., 1999). In this study, our immunofluorescence results showed that p65 translocated from the nucleus to the cytoplasm during myoblast differentiation; thus, the reduced activation of NF-κB was mainly caused by translocation alteration during myogenic differentiation. Furthermore, NF-κB has been shown to play an important role in myogenesis by regulating cell cycle genes and myogenic genes. Many studies have indicated that NF-κB can increase the proliferation of myoblasts and inhibit myogenic differentiation (Bakkar et al., 2008; Guttridge et al., 1999; Langen et al., 2001). In a previous study, NF-κB was shown to enhance Ccnd1 transcription and promote retinoblastoma protein hyperphosphorylation, both of which are crucial for myoblast proliferation (Guttridge et al., 1999). Moreover, NF-κB inhibits MyoD expression at the post-transcription level and can promote the degradation of myoD mRNA (Guttridge et al., 2000).

NF-κB could also affect myogenesis through regulating crucial miRNAs. In this study, we demonstrated that NF-κB could negatively regulate the miR-195 and miR-497 cluster, which could inhibit the proliferation of myoblasts by targeting cell cycle and growth factor receptor genes. Thus, our results identified a new pathway mediating the proliferation of myoblasts, as shown in Fig. 8H. In this pathway, NF-κB regulated the transcription of the miR-195 and miR-497 cluster, which subsequently mediated the expression of Igf1r, Insr, Ccne1 and Ccnd2 to affect myoblast proliferation. Similarly, another myoblast proliferation inhibitor, miR-29, is also negatively regulated by NF-κB (Mott et al., 2010; Wang et al., 2008). These results confirm that NF-κB regulates the myogenesis program through modulating not only functional genes but also miRNAs.

In summary, our study provides direct evidence that miR-195 and miR-497 participate in skeletal muscle development by regulating myoblast proliferation. The potential mechanism underlying the inhibition of myoblast proliferation by these miRNAs involves direct targeting of Ccnd2, Ccne1, Igf1r and Insr genes. Additionally, miR-195 and miR-497 were negatively regulated by NF-κB. Finally, our data indicate a novel signaling pathway that involves NF-κB–miR-195/497–Igf1r/Insr–Ccnd2/Ccne1 in the modulation of the proliferation of myoblasts in this study. These findings provide important insight into the mechanisms of muscle development and might have great significance on animal breeding and health care.

Balb/c mice were purchased from the Hubei Centre for Disease Control and Prevention (Wuhan, China). Thigh muscle samples were collected from male Balb/c mice at different ages (postnatal 2 days, 2 weeks, 4 weeks and 6 weeks). Samples were frozen in liquid nitrogen and stored at −80°C until RNA isolation. All experiments were performed according to the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, Commission on Life Sciences, National Research Council, 1996).

The growth medium for C2C12 cells was high-glucose Dulbecco's modified Eagle medium (DMEM; Hyclone) plus 10% (v/v) fetal bovine serum (Gibco). The differentiation medium was DMEM plus 2% (v/v) horse serum (Gibco).

Balb/c mice at postnatal 4 weeks were purchased from the Hubei Centre for Disease Control and Prevention (Wuhan, China). Mice were randomly divided into four groups (three mice for every group). Two groups were treated with LPS (1 mg/kg body weight) or PBS for 24 h by intraperitoneal injection. The other two groups of mice were treated with LPS (1 mg/kg body weight) or PBS for 48 h by intraperitoneal injection. The injection was repeated every 12 h. After treatment, thigh muscle samples were collected from every mouse, and samples were frozen in liquid nitrogen and stored at −80°C until RNA and protein isolation.

TRIzol reagent (Invitrogen) was used for total RNA isolation. Before reverse transcription, DNase I (Fermentas) was used to remove the residual DNA. A RevertAid First Strand cDNA Synthesis Kit (Fermentas) was used for reverse transcription, and oligo(dT), random primers, or miRNA-specific reverse transcription primers were added as appropriate. THUNDERBIRD SYBR qPCR Mix (TOYOBO) was used for qPCR, and the results were monitored using a LightCycler 480 II (Roche) system or CFX96 Real-Time PCR Detection System (Bio-Rad). The sequences for PCR primers were as follows: myogenin-F, 5′-CAATGCACTGGAGTTCGGT-3′, myogenin-R, 5′-CTGGGAAGGCAACAGACAT-3′; MyHC-2d-F, 5′-CGCAAGAATGTTCTCAGGCT-3′, MyHC-2d-R, 5′-GCCAGGTTGACATTGGATTG-3′; MCK-F, 5′-GCTTATGGTGGAGATGGAGA-3′, MCK-R, 5′-GGCCATCACGGACTTTTATT-3′; and tubulin-F, 5′-GACTATGGACTCCGTTCGCTC-3′, tubulin-R, 5′-TATTCTTCCCGGATCTTGCTG-3′. The miRNA-specific reverse transcription primers and PCR detection primers for miR-195 and miR-497 were provided by RiboBio.

Fixed cells were incubated with monoclonal anti-myosin (skeletal, fast) antibodies (M4276; Sigma), anti-actin-α1 skeletal muscle antibodies (NB120-7799; Novus Biologicals) or anti-p65 antibodies (sc-372; Santa Cruz Biotechnology) as primary antibodies. Cells were then incubated with Alexa-Fluor-488-conjugated goat anti-mouse-IgG, -IgM (H+L) or Alexa-Fluor-488-conjugated goat anti-rabbit-IgG (H+L) (Invitrogen) secondary antibodies. The cell nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Images were captured using a Nikon ECLIPSE TE2000-S system.

The following primers were used to amplify the 3′UTR fragments containing predicted miR-195- and miR-497-binding sites: Ccne1 proximal, 5′-CCGCTCGAGGCAGGAGACAGAATGAC-3′ and 5′-ATAAGAATGCGGCCGCGCAACCTACAACACCC-3′, Ccne1 distal, 5′-CCGCTCGAGTGCTCGGGTGTTGTAGGT-3′ and 5′-ATAAGAATGCGGCCGCGGCGGATTTCTGAGTTTG-3′, Ccnd2 proximal, 5′-CCGCTCGAGGCCAGGACCGCATGAGTAGT-3′ and 5′-ATAAGAATGCGGCCGCTGAGACTTAGGAGCCGTTCACC-3′, Ccnd2 distal, 5′-CCGCTCGAGGAACAGGGCTGCTTTGAAGAGT-3′ and 5′-ATAAGAATGCGGCCGCCAGAAGGACAGGGACAGAGGAA-3′, Igf1r 3′UTR-BS, 5′-CCGCTCGAGGGACTTCTTCATGGGTCTCTCA-3′ and 5′-ATAAGAATGCGGCCGCGAAGCATCAGTTGCCGGAGAA-3′, Insr 3′UTR-BS1, 5′-CCGCTCGAGCTGTGGAGGGCTAACTGTGAA-3′ and 5′-ATAAGAATGCGGCCGCGAAGAAGACACCTGAGCAAGAC-3′, and Insr 3′UTR-BS2, 5′-CCGCTCGAGCGGGCAATGGGAATAACAGAA-3′ and 5′-ATAAGAATGCGGCCGCTAAACTATATGATGCTACTTGGCTCT-3′. The restriction enzymes XhoI (Fermentas) and NotI (Fermentas) were used to insert these fragments into the psiCHECK-2 vector (Promega). Additionally, P195-F, 5′-tcaACGCGTGATGCTCCCATGTGAGACTGTC-3′ and P195-R, 5′-tcaGCTAGCGGTCGAGAGGAAGACAGCACAG-3′ were used to amplify the fragment containing predicted p65-binding sites (lowercase letters represent protective bases for the restriction endounclease site), and the restriction enzymes MluI (Fermentas) and NheI (Fermentas) were used to insert this fragment into the PGL3 vector (Promega). pNFκB-TA-luc (Beyotime) was used to reflect the activity variation of NF-κB. A dual luciferase reporter assay system (Promega) was used to analyze the relative luciferase activity.

Cells were transfected using Lipofectamine 2000 (Invitrogen) according to the manufacturer's recommendations. miRNA mimics, mutants, and scrambled negative control were provided by RiboBio. The sequences of miRNAs used in this study were as follows: miR-195, 5′-UAGCAGCACAGAAAUAUUGGC-3′; miR-195 mut, 5′-UUCGUCGUCAGAAAUAUUGGC-3′; miR-497, 5′-CAGCAGCACACUGUGGUUUGUA-3′; and miR-497 mut, 5′-CUCGUCGUCACUGUGGUUUGUA-3′. The sequences of siRNAs used in this study were as follows: si-Igf1r-1, 5′-CAAUGAGUACAACUACCGCTT-3′; si-Igf1r-2, 5′-CGACUAUCAGCAGCUGAAGTT-3′; si-Igf1r-3, 5′-CAACGAAGCUUCUGUGAUGUUTT-3′; si-Insr-1, 5′-AACAAGACAGCUGGUACCAGGTT-3′; si-Insr-2, 5′-ACACAACCUCACCAUCACUTT-3′; si-Insr-3, 5′-ACUGCAUGGUUGCCCAUGATT-3′; si-Insr-4, 5′-AGAUGACAACGAGGAAUGUTT-3′; and scrambled negative control, 5′-UUCUCCGAACGUGUCACGUTT-3′. siRNA targeting p65 (sip65) was purchased from Invitrogen. p65-F, 5′-GTCaagcttCCCTGACCATGGACGATCTG-3′ and p65-R, 5′-CATgcggccgCACCTTAGGAGCTGATCTGAC-3′ were used to amplify the coding sequence (CDS) of p65 (lowercase letters represent restriction endounclease site), and the restriction enzymes HindIII (Fermentas) and NotI (Fermentas) were used to insert p65 CDS into the pCDNA3.1(+) vector.

Treated cells were fixed and incubated in propidium iodide solution. The cell cycle phase was then analyzed using a FACSCalibur instrument (Becton Dickinson) and ModFit software (Verity Software House). The proliferation index indicates the proportion of mitotic cells (S-phase cells and G₂-phase cells) to total cells.

miR-195/497-F, 5′-cccaagcttaaGGCCATGTTTGCCATTCACAC-3′, and miR-195/497-R, 5′-aaactgcagTGGATCACCTGGGACTCTAAG-3′ (lowercase letters represent the restriction endounclease site and its protective bases), were used to amplify the fragment containing the primary miR-195 and miR-497 cluster transcript, and the restriction enzymes HindIII (Fermentas) and PstI (Fermentas) were used to insert this fragment into the pEGFP-C1 vector. Cells were transfected with the construct miR-195/497-C1 or pEGFP-C1 and incubated in medium containing EdU (Invitrogen). The proportions of EdU- and GFP-positive cells were then determined using flow cytometry.

For experiments using IGF-1 stimulation, cells were transfected with miR-195/497-C1 or pEGFP-C1 for 24 h, stimulated with IGF-1 (100 ng/ml, Sigma) or vehicle (acetic acid) for an additional 12 h, and then subjected to EdU staining and flow cytometry analysis.

Cells protein lysates were obtained using M-PER Mammalian Protein Extraction Reagent (Pierce). SDS-PAGE was used to separate the proteins, and a Mini Trans-Blot Cell (Bio-Rad) was used to transfer protein onto polyvinylidene fluoride (PVDF) membranes (Millipore). Primary antibodies specific for cyclin E1 (Cell Signaling Technology), cyclin D2 (Cell Signaling Technology), IGF1 receptor β (Cell Signaling Technology), insulin receptor β (Cell Signaling Technology), tubulin (Beyotime), and NF-κB p65 (Santa Cruz Biotechnology), as well as horseradish peroxidase (HRP)-labeled anti-rabbit-IgG and anti-mouse-IgG secondary antibodies (Beyotime), were used for immunoblotting. An Image Quant LAS4000 mini (GE Healthcare Bio-Sciences) was used to detect the signal produced by Immobilon Western Chemiluminescent HRP Substrate (Millipore).

Nuclear and cytoplasmic extracts were isolated using an NE-PER Nuclear Protein Extraction Kit (Pierce). The probes were synthesized and double-labeled with FITC (Invitrogen). The sequences of the probes were as follows: probe1, 5′-GCGGGGGAGGGGTTTCCAACGG-3′ and 5′-CCGTTGGAAACCCCTCCCCCGC-3′; probe2, 5′-GGAGGATCGGGCTTTCCTGCTT-3′ and 5′-AAGCAGGAAAGCCCGATCCTCC-3′; probe3, 5′-TCTTGTGGGGGTCCCCCACCCC-3′ and 5′-GGGGTGGGGGACCCCCACAAGA-3′. The reagents and methods used for this experiment were from the EMSA/Gel-Shift kit (Beyotime).

A VCX130 system (Sonics & Materials, Inc.) was used for cell sonication. Anti-NF-κB p65 antibodies (sc-372; Santa Cruz Biotechnology) and normal anti-mouse-IgG agarose-conjugated antibodies (sc-2343; Santa Cruz Biotechnology) were used for the immunoprecipitation reaction. The ChIP Assay kit (P2078) was purchased from Beyotime, and the protocol was performed according to the manufacturer's instructions. The sequences for the following PCR detection primers were as follows: first BS-F, 5′-GTGCTGCTGAAACTGGCAAGGA-3′, first BS-R, 5′-TCTGGCCTGGCTTACCCTCTAG-3′; second BS-F, 5′-GCAGATTCCAGTCCTCCAGGCT-3′, second BS-R, 5′-GCCAAGAGGTCAAGCTCATTGC-3′; third BS-F, 5′-TCTGACTGGGAGTGGAGGAACC-3′, third BS-R: 5′-GGGTGCATCCCTCACATTTGGG-3′.

All results are shown as the mean±s.e.m., and at least three independent individuals or replicates were used per group. Unpaired Student's t-tests were used to determine statistical significance, and P<0.05 was considered significantly different.

The authors would like to thank both Dori Miller and Dr Juming Zhong at Auburn University College of Veterinary Medicine, USA for the language proof of this article.