NF-κB-modulated miR-130a targets TNF-α in cervical cancer cells

The human cervical cancer cell lines HeLa and C₃₃A were grown in RPMI1640 medium (GIBCO BRL, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS), 100 IU/ml of penicillin and 100 mg/ml of streptomycin. The cell lines were incubated at 37°C in a humidified chamber supplemented with 5% CO₂. Transfections were performed using the Lipofectamine 2000 Reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. All transfections were performed in three independent experiments.

miRNA targets were predicted using the following algorithms: Target-Scan, PicTar, and MiRBase Targets.

To construct the pcDNA3/NF-κB vector expressing p50 subunit, we amplified a DNA fragment carrying p50 by PCR using sense and antisense NF-κB primers, and then the fragment was cloned into the pcDNA3 vector at the EcoRI and Xhol restriction sites.

To construct the pSilencer/shR-NF-κB vector interfere p50 subunit expression, a 70-bp double-stranded fragment was obtained via an annealing reaction using two single-strands. The fragment was then cloned into a pSilencer 2.1 vector plasmid (Ambion) at the BamHI and Hind III sites.

To construct a pcDNA3/pri-miR-130a vector expressing miR-130a, we amplified a DNA fragment carrying pri-miR-130a from genomic DNA using sense and antisense Pri-130a primers. The 3’UTR, including predicted target sites, was amplified by PCR using TNF-α-3’UTR-sense and -antisense primers, and the amplified sequence was cloned into an expression plasmid (pcDNA3/EGFP) downstream of the EGFP gene between the EcoRI and BamHI restriction sites. Similarly, the 3’UTR containing mutated miR-130a binding sites was amplified using PCR site-directed mutagenesis and cloned into the pcDNA3/EGFP plasmid between the same restriction sites with TNF-α-3’UTR-mut-sense and -antisense primers. The resulting vectors were designated as pcDNA3/EGFP-TNF-α-3’UTR and pcDNA3/EGFP-TNF-α-3’UTRmut.

The human TNF-α mRNA lacking the 3’UTR was amplified from human HeLa cell cDNA by PCR using sense and antisense TNF-α primers, which are listed in Table 1, and then cloned into the pcDNA3 vector at the EcoRI and Xhol restriction sites, which created the pcDNA3/TNF-α plasmid.

All DNA oligonucleotides used are listed in Table 1, and all constructs were confirmed by DNA sequencing.

HeLa and C₃₃A cells were co-transfected with pcDNA3/pri-miR-130a or ASO-miR-130a in a 48-well plate followed by the pcDNA3/EGFP-TNF-α-3’UTR or pcDNA3/EGFP-TNF-α-3’UTR-mut reporter plasmids. The RFP expression vector, pDsRed2-N1 (Clontech, Mountain View, CA), was used for normalization. Cells were lysed with RIPA 48 h after transfection, and proteins were collected. EGFP and RFP fluorescence intensities were determined using an F-4500 fluorescence spectrophotometer (HITACHI, Tokyo, Japan).

HeLa and C₃₃A cells were seeded in 24-well plates overnight and then transfected with plasmid (1 μg each) or oligonucleotides (final concentration 200 nM). Afterwards, cells were trypsinized and counted, seeded in 96-well plates (in triplicate) for an MTT assay at a density of 8,000 cells/well (HeLa) or 10,000 cells/well (C₃₃A), and incubated at 37°C for 24 h. Then, at 24, 48 and 72 h after cell seeding, 10 μl MTT (final concentration, 0.5 mg/ml) was added, and the cells were maintained at 37°C for another 4 h. The medium was removed, and the precipitate was dissolved in 100 μl DMSO. After shaking for 15 min, the absorbance at 570 nm (A570) was measured using an uQuant Universal Microplate Spectrophotometer (Bio-Tek Instruments, Winooski, USA).

Following transfection in 24-well plates as described above, HeLa and C₃₃A cells were counted and seeded in 12-well plates (in triplicate) at a density of 200 cells/well. The plates were incubated at 37°C in a 5% CO₂ humidified incubator. The culture medium was replaced every 3 days. After 14 days in culture, cells were stained with crystal violet and counted. Colonies with at least 50 cells were considered for quantification.

Total RNA extracted from HeLa and C₃₃A cells was using the TRIZOL reagent (Invitrogen, Carlsbad, CA), according to the manufacturer’s instructions. RNA extraction from tissue samples was performed by using the mirVana miRNA Isolation Kit (Ambion) according to the manufacturer’s instructions. To assess RNA integrity, a portion each RNA sample was used for concentration and purity measurements (A260 and A280 by spectrophotometry), and a separate portion was subjected to denaturing electrophoresis in a 1% agarose gel stained with ethidium bromide.

qRT-PCR used to detect the miRNA and mRNA levels and was performed using an iQ5 real-time PCR detection system (Bio-Rad). Moloney murine leukemia virus (M-MLV) reverse transcriptase (Promega, Madison, WI) was used to reverse transcribe cDNA from RNA samples. The SYBR Premix Ex Taq™ kit (TaKaRa, Otsu, Shiga, Japan) was used to measure amplified DNA. Related primers were purchased from AuGCT, Inc. (Beijing, China), and sequences are shown in Table 1.

To detect miR-130a, 2 μg of total RNA (in triplicate) was reverse transcribed to cDNA using M-MLV. U6 was used as a housekeeping gene to normalize gene expression. The following PCR cycles were used: 94°C for 3 min followed by 40 cycles of 94°C for 30 s, 56°C for 30 s, and 72°C for 30 s. To detect target genes, 5 μg of total RNA (in triplicate) was reverse transcribed into cDNA. Endogenous β-actin gene levels were used as a control to normalize gene expression. The following PCR cycles were used: 94°C for 3 min followed by 40 cycles of 94°C for 30 s, 58°C for 30 s, and 72°C for 30 s.

Western blotting was performed to determine protein expression. Briefly, HeLa and C₃₃A cells were transfected with plasmid (4 μg each) or control oligonucleotides (final concentration of 200 nM). After seeding in 25-ml culture flasks, cells were washed with PBS and lysed after 30 min in RIPA lysis buffer at 4°C to harvest total protein. To collect nucleoproteins, cells were harvested with PBS, centrifuged to obtain a cell precipitate, 0.4% NP-40 added, and the cell precipitate resuspended. After centrifugation, the supernatant contained the nucleoproteins. Next, 0.1% NP-40 was added to the remaining cell precipitate, the precipitate was resuspended and centrifuged, the supernatant was discarded, and the remaining material resuspended in RIPA buffer to obtain cytoplasmic proteins. Equal amounts of protein were subjected to electrophoresis using a 10% SDS-polyacrylamide gel under denaturing conditions and were then transferred onto a nitrocellulose membrane. To assess protein levels, membranes were incubated in blocking buffer for 4 h at room temperature and then with antibodies raised against p50 or TNF-α overnight at 4°C. Membranes were then washed and subsequently incubated with secondary antibody. Protein expression was assessed by enhanced chemiluminescence and exposure to chemiluminescent film. The LabWorks™ Image Acquisition and Analysis Software (UVP, Upland, CA) were used to quantify the protein band intensities. All the antibodies were purchased from Tianjin Saier Biotech (Tianjin, China).

Cells were seeded on 14-well slides at a concentration of 2000 cells/well and incubated overnight. After the cells adhered to the slides, the culture medium was replaced with medium containing TNF-α (Sigma-Aldrich, St. Louis, MO) at a final concentration of 20 ng/mL, and the cells were cultured for 24 h. Cells were fixed with 4% paraformaldehyde for 30 min, permeabilized with 0.5% Triton X-100 for 5 min, and blocked with 10% donkey serum at room temperature for 2 h. Following blocking, primary antibodies were diluted in 1% normal donkey serum, placed on the slide surface, and incubated at 4°C overnight. Cells were washed, and a FITC-labeled donkey anti-rabbit secondary antibody was diluted in 1% normal donkey serum and added to the slides. Finally, DAPI was used to stain the nuclei. Fluorescent images were acquired using a Digital Sight DS-U1 scanning microscope (Nikon, Tokyo, Japan) at an excitation wavelength of 488 nm. Images were superimposed using the NIS Elements F 2.20 imaging software (Nikon).

For in vivo tumor study, 3 × 10⁶ Hela cells transfected with pri-miR-130a and it is control vector were suspended in 150 μl of serum-free 1640 for each mouse. Each mouse (6 in each group, female, BALB/c-nu/nu at 6 weeks of age) were implanted subcutaneously in the right armpit of nude mice. All mice were killed 14 days after implantation. The samples were frozen in liquid nitrogen or fixed with phosphate-buffered neutral formalin. All studies were performed under the American Association for the Accreditation of Laboratory Animal Care guidelines for humane treatment of animals and adhered to national and international standards. Ethical approval was given by the medical ethics committee of the Ethics Committees of Tianjin Medical University with the following reference number: TMUhMEC 2011018.

The tissue samples were fixed in phosphate-buffered neutral formalin, embedded in paraffin, and cut into 5 μm thick sections. Tissue sections were deparaffinized, rehydrated, and microwave-heated in sodium citrate buffer for antigen retrieval. The sections were then incubated with 0.3% hydrogen peroxide/phosphate-buffered saline for 30 min. Sections were incubated with a primary antibody against TNF-α at a 1:50 dilution and incubated overnight at 4°C. Detection of the primary antibody was performed using goat anti-rabbit-HRP for 1 hour at room temperature and visualized with DAB substrate.

Data are expressed as the mean ± standard deviation (SD). Statistical analyses were performed with a paired t-test. P < 0.05 was considered statistically significant. Experiments were performed in triplicate, and one representative experiment is shown in the Figures.

NF-κB has been shown to promote tumorigenesis, tumor cell proliferation, invasion and metastasis[39]. To investigate the role of NF-κB in cervical cancer cells, we constructed an NF-κB overexpression vector (pcDNA3/NF-κB) and an NF-κB interference vector (pSilencer/shR-NF-κB) which over express or knock down p50 protein that were validated by western blot in transfected HeLa and C₃₃A cells (Figure 1A and B). Next, an MTT assay showed that NF-κB overexpression increased HeLa and C₃₃A cell viability compared with the control group at 24, 48, and 72 h after cell seeding (Figure 1C). Conversely, the knock down of NF-κB suppressed cell viability (Figure 1D). The colony formation assay also demonstrated that NF-κB overexpression increased HeLa and C₃₃A colony numbers by approximate 25% and 40%, respectively, compared with the control vector (Figure 1E), but decreased NF-κB levels reduced the colony numbers (Figure 1F). These results indicate that NF-κB promotes cervical cancer cell viability and growth.

In our previous experimental work, gene chip results showed some miRNAs which may regulated by NF-κB, miR-130a was in them and relatively few existing reports about it, so we picked it as our object of study for a further explore. When HeLa and C₃₃A cells were transfected with pcDNA3/NF-κB or pSilencer/shR-NF-κB, qRT-PCR indicated that NF-κB overexpression increased the miR-130a levels by approximately 50% and 40% compared to that of control vector, but a knock down of NF-κB decreased the miR-130a levels by 25% in both HeLa and C₃₃A cells, respectively (Figure 2A). These results indicate that NF-κB may induce miR-130a expression.

Some reports have shown that miR-130a can mediate endothelial cancer cell growth[40, 41], and our data also showed that NF-κB enhanced the growth of cervical cancer cells and induced miR-130a expression. Thus, we speculated that miR-130a may mediate NF-κB’s cell growth effects. First, we validated the efficiencies of pcDNA3/pri-miR-130a or miR-130a antisense oligomers (ASO-miR-130a). qRT-PCR revealed that cells transfected with pcDNA3/pri-miR-130a produced 7.5 and 6-fold increases in miR-130a levels. However, ASO-miR-130a reduced miR-130a levels by almost 50% in transfected HeLa and C₃₃A cells (Figure 2B). Then, we performed MTT and colony formation assays in transfected HeLa and C₃₃A cells and found that pcDNA3/pri-miR-130a increased cell viability and the colony formation rate by approximately 20% compared with the control group (Figure 2C and E). Moreover, ASO-miR-130a led to a suppression of cell viability and colony formation (Figure 2D and F) rate. These results indicate that miR-130a facilitates HeLa and C₃₃A cell viability and growth, which is consistent with NF-κB’s effects on HeLa and C₃₃A cells.As miR-130a and NF-κB plays the same role on the growth of cancer cells, which prompt us miR-130a maybe one of the regulatory mechanisms of NF-κB, in order to verify this, we performed an experiment using by NF-κB in cells with miR-130a knockdown. Results showed that as miR-130a knockdown, cell growth ability decreased compare with the group only overexpressed NF-κB both in the MTT and colony formation assays (Figure 2G and H).

miRNAs regulate a number of cellular functions through the downregulation of target gene expression by binding to 3’UTRs. To predict candidate target genes of miR-130a, we used three algorithm programs, TargetScan, PicTar, and miRanda. Using these programs, we selected TNF-α as a miR-130a target gene for further study because of its ability to suppress cancer cell growth and to activate NF-κB. The complementary sequence between the seed sequence of miR-130a and the TNF-α 3’UTR is shown in Figure 3A. To elucidate miR-130a’s direct regulation of TNF-α, an enhanced green fluorescent protein (EGFP) reporter assay was used to identify the target site in the TNF-α 3’UTR. We first constructed an EGFP reporter plasmid by inserting the miR-130a binding site in the TNF-α 3’UTR downstream of the EGFP stop codon (pcDNA3/EGFP-TNF-α-3’UTR) and a mutant seed sequence in a reporter plasmid (pcDNA3/EGFP-TNF-α-3’UTR-mut). Next, HeLa and C₃₃A cells were cotransfected with the pcDNA3/EGFP-TNF-α-3’UTR reporter plasmid and pcDNA3/pri-miR-130a or ASO-miR-130a plasmids. We found that pcDNA3/pri-miR-130a expression led to a 25% decrease in EGFP fluorescence intensity compared with the control group, and the EGFP fluorescence intensity in cells transfected with ASO-miR-130a increased more than 25% (Figure 3B). However, there were no significant changes in the fluorescence intensities of the pcDNA3/EGFP-TNF-α-3’UTR-mut group regardless of altered miR-130a levels (Figure 3C). Furthermore, we examined endogenous TNF-α that had been downregulated by miR-130a. Overexpression of miR-130a led to an approximately 50% reduction in TNF-α mRNA levels, but conversely, blocking miR-130a expression increased TNF-α mRNA levels by approximately 50% (Figure 3D). Western blot analysis showed that pcDNA3/pri-miR-130a reduced TNF-α protein expression levels by 40% in HeLa cells and 30% in C₃₃A cells (Figure 3E), and blocking endogenous miR-130a expression increased TNF-α protein levels by 50% in both cells types (Figure 3F). We also verified this result in vitro, cells were transfected with pcDNA3/pri-miR-130a and its control vector then injected subcutaneously in the flank and collected the tumor tissue. In the RNA extracted from tissue, miR-130a was increased compared to the control group (Figure 3G), and staining of TNF-α was decline (Figure 3H). Altogether, these results indicate that miR-130a can bind to the TNF-α 3'UTR and negatively regulate its mRNA and protein expression levels.

To confirm that miR-130a promotes cervical carcinoma cell growth through at least a partial downregulation of TNF-α, we generated a TNF-α expression vector (pcDNA3/TNF-α) lacking the TNF-α 3’UTR to minimize miRNA interference. miR-130a was overexpressed with TNF-α in HeLa and C₃₃A cells, and then growth activity was examined by MTT and colony formation assays. Western blot analysis showed that the miR-130a-induced reduction in TNF-α levels was rescued by the pcDNA3/TNF-α plasmid transfected into HeLa and C₃₃A cells (Figure 4A and B). In the colony formation assay, ectopic TNF-α expression abrogated the cell growth enhancement caused by miR-130a compared with the control vector (Figure 4C and D), but in the MTT assay, no obvious changes were observed between the experimental and control groups (Figure 4E and F). Therefore, these results provide further evidence that TNF-α overexpression counteracts the repressive effects of miR-130a on cell growth.

TNF-α is a strong activator of NF-κB[42, 43], and our findings indicated that NF-κB could stimulate miR-130a expression, which in turn downregulated TNF-α expression (Figure 5A). Therefore, we speculated that TNF-α may regulate miR-130a expression through the NF-κB pathway. Twenty-four hours after TNF-α treatment, we extracted cytoplasmic and nuclear proteins from HeLa and C₃₃A cells to detect NF-κB and found that the nuclear p50 content was significantly increased as assessed by Western blot (Figure 5B). Immunofluorescent staining also showed that TNF-α increased nuclear p50 (red) levels with compared with the control (Figure 5C and D). In Figure 2G and H, we seen NF-κB increased the growth capacity of crvical cncer cells through miR-130a, and we have demonstrated that miR-130a can direct targeting TNF-α, so we made a transfection of ASO-miR-130a and NF-κB, we found that the consume of miR-130a rescue the reduction of TNF-α inducted by NF-κB (Figure 5E), these results strongly support our hypothesis. Moreover, TNF-α treatment upregulated miR-130a levels by approximately 5-6.5-fold in HeLa and C₃₃A cells (Figure 5F). Altogether, TNF-α initially stimulated NF-κB activation to induce miR-130a expression, which in turn targeted and down-regulated TNF-α expression and suggests a TNF-α/NF-κB negative feedback loop acting through miR-130a in cervical cancer cells (Figure 5G).