Down-regulation of BTG1 by miR-454-3p enhances cellular radiosensitivity in renal carcinoma cells

Human renal carcinoma 786-O cells were cultured in RPMI-1640 media (GIBCO, NY, USA) supplemented with 10% fetal bovine serum (Hyclone, MA, USA) at 37°C and 5% CO₂. Irradiation was carried out by laboratory X-ray source (RX-650, Faxitron Bioptics, USA) [27]. The dose rate was 0.8 Gy/min (100 keV, 5 mA). Cells were seeded into 12-well plates and grown to <70% confluence at the time of irradiation.

Cells were harvested and fixed as described previously [28]. Prior to analysis, fixed cells were washed twice with PBS, treated with 100 μg/mL RNase A and 50 μg/mL propidium iodide (BD Biosciences, California, USA) for 20 min, and analyzed using FACS Calibur flow cytometry (Becton Dickinson, NJ, USA). Cell cycle distribution was analyzed with the FlowJo software package. DNA content of samples was measured with CellQuest (Becton Dickinson).

Western blotting was performed as described previously [29]. Briefly, cells were collected and lysed with RIPA buffer solution (Beyotime, Haimen, China). Supernatants were collected at 10,000 g for 10 min at 4°C. The concentrations of samples were determined using a BCA protein assay kit (Pierce, IL, USA). Equal amounts of protein were loaded onto 10% SDS-polyacrylamide gels for electrophoresis, and proteins were transferred onto PVDF membrane by western blotting. GAPDH was used as loading control. Membranes were then blocked with nonfat milk and incubated overnight at 4°C with the following primary antibodies: anti-GAPDH (1:5,000, Santa Cruz), anti-BTG1 (1:1,000, Abnova, Taipei, Taiwan). Immunological complexes were detected by enhanced electrochemiluminescence (Millipore, Darmstadt, Germany) with either an anti-rabbit peroxidase (1:5000, Santa Cruz, Texas, USA), or an anti-mouse peroxidase (1:4000, Santa Cruz) antibody. The fold changes of protein levels were analyzed by the Image J software.

MicroRNA mimic and negative control mimic were synthesized and purified by RiboBio Co. (Guangzhou, China) [29]. Transfections of the miRNA duplexes were performed with 40–60% confluent cells using Lipofectamine 2000 (Invitrogen, California, USA). The medium was replaced with new culture medium for 5 h after transfection.

Cell viability is calculated as the number of viable cells divided by the total number of cells within the grids on the hemacytometer. Cells which take in trypan blue are considered non-viable. Cells were harvested with trypsin, suspended in PBS and mixed with 0.4% solution of trypan blue stain (Invitrogen) after various treatments. Count at least 500 cells for calculation. The percentage of viable cells = [1.00 - (Number of blue cells/Number of total cells)]*100%.

To evaluate the activity of caspase-3, the caspase-3 activity kit (Beyotime) was used. Cells were collected and lysed with reaction buffer and the total protein concentration is 1-3 mg/mL. In the samples, activated caspase-3 cleaves substrate (Ac-DEVD-pNA) (2 mM) between DEVD and pNA, quantitatively generating pNA that can be detected using an ELISA reader at an absorbance of 405 nm. In the caspase-3 colorimetric calibration, the value of R² should be greater than 0.999.

The clonogenic assay was conducted as described previously [30]. Briefly, cells were harvested and an appropriated number of cells were seeded onto each of the 60 mm dishes to produce about 50–120 colonies. After 8–10 days incubation, the colonies were washed with 1 × PBS softly, fixed with 70% ethanol for 5 min and stained with 0.5% crystal violet for 3 min at room temperature. Colonies with more than 50 cells were counted.

The 3′-untranslated region (3′-UTR) of human BTG1 transcript was cloned downstream of the luciferase gene between the Xho I and Sal I sites of the pmirGLO dual-luciferase vector (Promega, WI, USA). A pmirGLO dual-luciferase vector containing one mutated seed sequences of miR-454-3p was constructed. The sequencing of constructed plasmids was verified by Shanghai Sangon Biotechnology Co. (Shanghai, China). 1.5 × 10⁵ 786-O cells in 12-well plate were co-transfected with 300 ng DNA (pmirGLO-3′ UTR constructs or derived mutants) and 30 nM of either miR-454-3p mimics using transfection reagent Lipofectamine 2000 (Invitrogen). Luciferase activity was measured 48 h later using the Dual Luciferase Reporter Assay System (Promega) [31] with a Tecan Infinite M200 Pro microplate reader (Tecan, Mannedorf, Switzerland).

For qRT-PCR, total RNAs were extracted from cultured cells using TRIzol Reagent (Invitrogen) according to the manufacturer’s protocol. Reverse transcription and quantitative RT-PCR were performed according to the protocol of the qRT-PCR Detection Kit (Promega). All of the stem-loop RT primers were purchased from RiboBio Co. (Guangzhou, China) to detect miR-454-3p or U6. U6 was used as an endogenous control for miRNAs and GAPDH for coding genes. Other gene-specific primers were as follows: BTG1, 5′-TCCATAATCCATCCCCAAGA-3′ and 5′-GGATGCAATCCTGGACATTT-3′, SKA2, 5′-CCGCTTTAAACCAGTTGCTG-3′ and 5′-CTCTGCCGCAGTTTTCTCTT-3′, GAPDH, 5′-GTGGACCTGACCTGCCGTCT-3′ and 5′-GGAGGAGTGGGTGTCGCTGT-3′. Gene-specific primers were synthesized from Shanghai Sangon Biotechnology Co. (Shanghai, China).

All experiments were repeated at least three times, and data were presented as means ± SE. The statistical significance of the results was determined by Student’s t-tests using Microsoft Excel (Microsoft Campus, Redmond, WA, USA).

Many tumor suppressor genes are recognized as responders to ionizing radiation (IR) [32]. To investigate whether BTG1 functions as a responder to IR, we examined the protein levels of BTG1 in response to 5 Gy of X-rays in renal carcinoma 786-O cells by Western blot analysis. The protein levels of BTG1 were significantly increased 8 h after irradiation (Figure 1A). We also treated 786-O cells with DMSO and serum starvation to investigate whether BTG1 responds to other types of extracellular stressors and a similar increase in BTG1 levels was observed. Consistent with the results of Western blotting, qRT-PCR showed that the mRNA levels of BTG1 were increased by more than 2-fold upon treatment with irradiation (Figure 1B), DMSO (Figure 1C) or serum starvation (Figure 1D). These results indicate that, like other tumor suppressor genes, BTG1 is up-regulated in response to X-ray. Meanwhile, the BTG1 can also be induced by other sources of extracelluar stressors, including the treatments by DMSO and serum starvation.

Because BTG1 is involved in the regulation of cell cycle progression and responds to IR, we examined the effects of BTG1 expression on the radiosensitivity of renal carcinoma 786-O cells. We transfected 786-O cells with the BTG1 expression vector pcDNA3.0-BTG1 in order to up-regulate the protein levels of BTG1. The efficiency of transfection was confirmed by Western blotting (Figure 2A). The protein levels of BTG1 increased notably at 24 h after transfection and lasted for at least three days.

To estimate radiosensitivity, the cell viability was assessed by trypan blue dye exclusion assay [33]. In this study, transfection of 786-O cells with pcDNA3.0-BTG1 significantly increased cell viability after radiation exposure (Figure 2B). To further confirm the effects of BTG1 on radiosensitivity, we measured the activity of caspase-3 (see Methods) which is an acknowledged effector among the caspase family members involved in apoptosis [34]. In order to more accurately characterize the changes of caspase-3 activity among all treatments, we drew standard curve and tested the caspase-3 activity at 24 h after irradiation (Additional file 1A and B). At 24 h after exposure to 5 Gy of X-rays, compared to transfection of pcDNA3.0, treatment of 786-O cells with transfection of pcDNA3.0-BTG1 caused a significant decrease in caspase-3 activation (Figure 2C). A colony formation assay was performed to determine the cellular radiation sensitivity. The survival fraction of 786-O cells transfected with pcDNA3.0-BTG1 without irradiation was significantly lower than that transfected with pcDNA3.0, due to the fact that BTG1 is an anti-proliferative gene. It is worth noting that there was no statistically significant difference in the survival fraction of cells exposed to 5 Gy of X-rays, when the group transfected pcDNA3.0-BTG1 was compared with the pcDNA3.0 group (Figure 2D and E). We also used flow cytometry to quantify changes in the sub-G1 peak at 48 h. The peak of transfection of BTG1 expression vector was decreased upon exposure to IR compared to the untransfected (Blank) group and the pcDNA3.0 group (Additional file 1C). These results suggest that BTG1 expression protects cells from the cytotoxic effects of IR.

To further examine the role of BTG1 in radiosensitivity of 786-O cells, we used siRNA oligonucleotides targeting BTG1 to specifically knock down BTG1 expression. Transient transfection of BTG1 siRNA efficiently inhibited BTG1 expression in 786-O cells at 24, 48 and 72 h (Figure 3A). Cell viability assay showed that a treatment with BTG1 siRNA before irradiation promoted the decrease of cell viability induced by 5 Gy of X-rays (Figure 3B). Caspase-3 activity in cells treated with siRNA against BTG1 was increased by more than 3-fold as compared with cells treated with negative control (NC) group (Figure 3C). The survival fraction was also tested in 786-O cells after exposure to 5 Gy of X-rays. As expected, the colony formation efficiency of cells transfected with BTG1 siRNA was significantly lower than that transfected with its negative control (Figure 3D and E). Furthermore, transfection with BTG1 siRNA led to a specific increase in the sub-G1 peak in 786-O cells at 48 h after IR treatment (Additional file 1D). These results suggest that down-regulation of BTG1 significantly increases cellular radiation sensitivity in 786-O cells.

To analyze whether BTG1 can be regulated by miRNAs, we used starBase (sRNA target Base), a miRNA-mRNA interaction prediction software consisting of five different algorithms, to predict the miRNAs which may potentially target the 3′-UTR of the BTG1 mRNA [35]. We selected the miRNAs that were predicted to target BTG1 by four or more algorithms for further analysis. These miRNAs included hsa-mir-130, hsa-mir-301a, hsa-mir-302, hsa-mir-454-3p, and hsa-mir-19b. Next, we constructed a series of pcDNA3.0 vectors that over-express these miRNAs and screened for those that regulate endogenous BTG1 levels at 48 h after transfection in 786-O cells. The BTG1 protein levels in samples transfected with pcDNA3.0-miR-454-3p displayed a significant down-regulation (Additional file 2F), indicating that miR-454-3p can target BTG1. A reproducible repression of endogenous BTG1 protein was also observed for vectors expressing miR-19b and miR-130, suggesting the possibility that multiple miRNAs target BTG1.

To determine whether miR-454-3p directly targets BTG1 via its 3′-UTR, the wild type and a mutant of BTG1 3′-UTR fragment with substitution in the seed region were synthesized and inserted into a luciferase report system (Figure 4A). The activity of the wild type reporter was significantly reduced by miR-454-3p mimic, whereas the miR-454-3p mimic had little effect on the mutant of BTG1 3′-UTR construct (Figure 4B). To further validate these findings, endogenous BTG1 levels were assessed in 786-O cells after transfection with miR-454-3p mimic. The miR-454-3p mimic specifically suppressed both the BTG1 protein and mRNA levels at 48 h post-transfection (Figure 4C and D). These results confirm that BTG1 is a target of miR-454-3p.

To determine whether endogenous miR-454-3p levels are regulated physiologically by IR, we assessed miR-454-3p expression in 786-O cells over a timecourse of exposure to 5 Gy of X-rays. While BTG1 mRNA levels increased, miR-454-3p levels decreased upon treatment (Figure 4E). It is worth noting that the BTG1 mRNA level decreased at 16 h after irradiation, whereas the miR-454-3p level increased.

The miR-454-3p gene is located in the first intron of the host gene, SKA2 (spindle and kinetochore-associated protein 2), which plays a critical role in proper mitotic progression, checkpoint silencing, and timely anaphase onset [36]. To test whether the expression of SKA2 correlates with that of miR-454-3p, SKA2 mRNA was quantified by qRT-PCR. Similar to the results for miR-454-3p, the expression of SKA2 was down-regulated in 786-O cells after IR (Figure 4F), providing further verification of our findings. These results indicate that miR-454-3p regulates the expression of BTG1 through a direct interaction with the 3′-UTR of the BTG1, and that a decrease in its expression correlates with BTG1 up-regulation during IR treatment of renal carcinoma cells.

Our results thus far suggest that down-regulation of BTG1 can enhance the radiosensitivity of 786-O cells and that BTG1 is also a target of miR-454-3p. Consequently, we postulated that miR-454-3p may affect cell radiosensitivity. To assess the effects of miR-454-3p on radiosensitivity, we exposed 786-O cells to 5 Gy of X-rays following transfection with miR-454-3p mimic. Trypan blue assay showed that the cell viability of miR-454-3p mimic transfected group relative to the NC group was reduced by 20% (Figure 5A). Similarly, over-expression of miR-454-3p mimic was associated with more than 4-fold increase of caspase-3 activity at 24 h (Figure 5B). Clonogenicity of 786-O cells transfected with miR-454-3p mimic was significantly reduced after exposure to 5 Gy of X-rays (Figure 5C and D). The sub-G1 percentage increased evidently at 48 h as assessed by flow cytometry (Additional file 1E), indicating that miR-454-3p enhances IR-mediated apoptosis.Because cell cycle control is intimately involved in the induction of apoptosis during IR, we also examined whether BTG1 can regulate the cell cycle. 786-O cells were transfected with control pcDNA3.0 vector or pcDNA3.0-BTG1 expression vector, and the effects of IR on the cell cycle distribution were assessed by flow cytometry after 48 h. The majority of the cells were in the G1 phase prior to IR treatment. Exposure to 5 Gy of X-rays caused a strong G2/M arrest for cells transfected with control pcDNA3.0 vector (Figure 5E and F). Whereas, cells transfected with pcDNA3.0-BTG1 were resistant to IR-induced G2/M phase arrest after exposure to 5 Gy of X-rays. These results indicate that high levels of BTG1 can inhibit IR-induced G2/M arrest.To further understand the role of BTG1 in inhibiting IR-induced G2/M arrest and to assess the effects of miR-454-3p on cell cycle arrest, 786-O cells were transfected with either siRNA oligonucleotides against BTG1 or miR-454-3p mimic. Flow cytometry analysis showed that BTG1 siRNA and miR-454-3p had no significant effect on cell cycle distribution in the absence of irradiation (Figure 5G). Surprisingly, however, BTG1 siRNA and miR-454-3p mimic promoted an IR-induced S phase arrest and abolished the G2/M arrest in 786-O cells exposed to 5 Gy of X-rays (Figure 5H). These results indicate that over-expression of BTG1 overcomes the G2/M arrest induced by IR, but knockdown of BTG1 expression causes a switch of the arrest from the G2/M phase to the S phase. One can easily conjecture about that down-regulation of BTG1 would impede the activation of the G1 checkpoint and lead to damage accumulation during S phase and the consequent triggering of the S phase arrest.