MicroRNA-485-5p inhibits glioblastoma progression by suppressing E2F transcription factor 1 under cisplatin treatment

ABSTRACT Cisplatin (CDDP) has been widely used for glioblastoma treatment. miR-485-5p and E2F transcription factor 1 (E2F1) dysfunction has been reported in glioblastoma. Nonetheless, whether CDDP affects glioblastoma progression via the miR-485-5p-E2F1 axis requires investigation. The expression of miR-485-5p and E2F1 was investigated by quantitative real-time polymerase chain reaction or western blotting in glioblastoma tissues and cell lines. The interaction between miR-485-5p and E2F1 was confirmed using a luciferase assay. The malignancy of glioblastoma was detected using Cell Counting Kit-8, bromodeoxyuridine (BrdU), cell adhesion, flow cytometry, and transwell assays. We identified miR-485-5p downregulation and E2F1 upregulation in glioblastoma, and miR-485-5p inhibited cell growth and elevated cell apoptosis in glioblastoma cells after CDDP treatment. Moreover, miR-485-5p targeting E2F1 repressed cell growth and improved cell apoptosis in glioblastoma cells after CDDP treatment. Our study revealed that CDDP retarded glioblastoma cell development via the miR-485-5p-E2F1 axis, which may be a new direction for glioblastoma therapy.


Introduction
Glioblastoma is the most common central nervous system malignancy with a poor prognosis [1,2]. Despite advances in understanding the molecular mechanisms of glioblastoma, patients still present a high recurrence rate and a low 5-year survival rate [3,4]. Thus, additional potential mechanisms underlying glioblastoma initiation and progression are crucial for glioblastoma diagnosis and treatment.
Cisplatin (CDDP) has been widely used as a cytotoxic drug in multiple tumors, including lung, bladder, and esophageal cancers [5][6][7]. CDDP treatment has been frequently applied in the diagnosis and treatment of glioblastoma [8]. For instance, Wang et al. [9] identified miR-152-3p as a CDDP sensitizer for glioblastoma in 2019. In 2020, Kim et al. [10] studied the effects of CDDP on cytotoxicity, apoptosis, autophagy, and the AKT/ mTOR pathway in U87MG and U373MG cells. Therefore, we aimed to explore the mechanism of CDDP in glioblastoma, thereby improving the efficacy of CDDP in glioblastoma therapy.
The E2F transcription factor 1 (E2F1) gene is located on chromosome 20q11. 22 and consists of seven exons. It encodes a member of the E2F family of transcription factors, which plays a crucial role in controlling the cell cycle and regulating tumor suppressor proteins [17]. Evidence displays that E2F1 aggravates cancer progression as an oncogene [18]. E2F1 has been reported to facilitate glioblastoma development, where it accelerates cell growth under CDDP treatment [19]. One study revealed that miR-136 remarkably repressed E2F1 expression to promote CDDP chemosensitivity in glioma cells [20]. Notably, the function of the miR-485-5p-E2F1 axis in glioblastoma with CDDP treatment is unclear.
Therefore, we aimed to investigate the effect of the miR-485-5p-E2F1 axis in glioblastoma cells under CDDP treatment. It was hypothesized that CDDP-assisted miR-485-5p might reduce E2F1 expression following CDDP treatment in glioblastoma. This study provides novel insights into the molecular mechanisms underlying glioblastoma.

Clinical Tissues, Cell Culture, and Cell Transfection
Human glioblastoma tissues and adjacent normal tissues were collected from 34 patients in our hospital with informed consent and approved by the Ethics Committee of our hospital. The patients' baseline data are presented in Table 1. Human glioblastoma U251 and U87 cells were obtained from ATCC (VA, USA) and were maintained in Dulbecco's Modified Eagle's Medium (Gibco, TX, USA) supplemented with 10% fetal bovine serum (FBS; Gibco) at 37°C and 5% CO 2 . For CDDP or temozolomide (TMZ) treatment, CDDP (Cat#: 232,120, Sigma-Aldrich, MO, USA) or TMZ (Cat#: 34,219, Sigma-Aldrich) was dissolved in phosphate-buffered saline (PBS), and final concentrations of 10, 20, and 30 μM were added to the medium. The miR-485-5p inhibitor, mimic, E2F1 overexpression plasmid (OE), and their negative controls (NC) were obtained from GenePharm (Shanghai, China). When cells reached 50% confluence, U251 and U87 cells were transfected using Lipofectamine 3000 (Invitrogen, MA, USA), and the following experiments were conducted after 48 h of transfection.

RNA Extraction and quantitative real-time polymerase chain reaction (qRT-PCR)
TRIzol reagent (Invitrogen) was used for RNA isolation, and the miRNeasy Mini Kit (Cat#: 217,004, QIAGEN, Hilden, Germany) was used for miRNA isolation from U251 and U87 cells. Next, cDNA transcribed from RNA was applied using the PrimeScript First Strand cDNA Synthesis Kit (Takara Bio, Shiga, Japan), and cDNA transcribed from miRNA was obtained using a miScript II RT Kit (Cat#: 218,161, QIAGEN). qRT-PCR was performed using SYBR Premix Ex Taq (Takara Bio) for E2F1 expression and the miScript SYBR Green PCR Kit (Cat#: #218,075, QIAGEN) for miR-485-5p expression. Relative expression was normalized to Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or small RNA U6 (U6) using the 2-ΔΔCt method [21]. All primer sequences are listed in Table 2.

Cell counting kit-8 (CCK8) assay
U251 and U87 cells were seeded at a density of 1 × 10 4 cells/well in 96-well plates. After adherence, the cells were treated with different concentrations of CDDP or TMZ for 48 h, followed by the incubation with 10 μL CCK8 solution (Dojindo, Kumamoto, Japan) for 2 h. Furthermore, a different set of cells was treated with 10 μM CDDP and incubated with 10 μL CCK8 solution for 2 h at four time points (0, 24, 48, and 72 h). The OD value at 450 nm was detected using a multimode plate reader (Thermo Fisher Scientific, MA, USA) [20,22].

Bromodeoxyuridine (BrdU) assay
The BrdU assay was performed according to a previous study [23]. U251 and U87 cells were seeded at a density of 1 × 10 4 cells/well in 96-well plates. After cells reached 80% confluence, the cells were labeled with BrdU (Cat#: ab126556, Abcam, Cambridge, UK) for 12 h. After washing twice with PBS, the cells were permeabilized and incubated with BrdU antibody for 2 h at room temperature. Then, the cells were incubated with antimouse antibody (Abcam, Cambridge, UK) for 1 h and subjected to a multimode-plate-reader at OD 450 nm (Thermo Fisher Scientific).

Cell adhesion assay
Collagen I solution (Sigma-Aldrich) was added to a 96-well plate for cell adhesion detection, and 2 × 10 4 U251 and U87 cells were cultured into the plate at 37°C for 4 h. Post culture, the medium was discarded, and the cells were treated with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide reagent (C0009S, Beyotime, Jiangsu, China) for 2 h at 37°C. Then, 100 µL dimethyl sulfoxide was added to each well, and the OD value at 570 nm was detected using a multimodeplate-reader (Thermo Fisher Scientific) [24].

Cell apoptosis assay
As previously described [25], the Annexin V-FITC Apoptosis Detection Kit (Cat#: 556,547; BD, NJ, USA) was used to detect apoptosis in U251 and U87 cells. Cells were harvested and treated with 5 µL fluorescein isothiocyanate and propidium iodide. After incubation for 20 min in the dark, the cells were washed twice, suspended, and analyzed by flow cytometry. The sum of the two right quadrants represented cell apoptosis, which was calculated using the FlowJo software (Tree Star, OR, USA).

Transwell assay
U251 and U87 cell migration was measured using a transwell chamber (Cat#: 3244, Corning, NY, USA) in a 24-well plate. The bottom compartment was prepared with 10% FBS medium, and cell suspensions (2 × 10 5 cells) without FBS were added to the top compartment. After 24 h of incubation, cells in the upper layer were removed, and cells in the lower layer were treated with 4% paraformaldehyde and 0.5% crystal violet for 15 min. Finally, the migrated cells were randomly captured using a microscope (Olympus, Tokyo, Japan) [26].

Luciferase reporter assay
This assay was performed according to a previous study [27]. Wild-type (WT) sequence of E2F1 3 -UTR based on the binding site, and a random mutated (MUT) sequence was constructed into the pmiRGLO vectors. U251 and U87 cells were co-transfected with pmiRGLO-E2F1-WT or pmiRGLO-E2F1-MUT and either miR-NC or miR-485-5p using Lipofectamine 3000. After 72 h, Firefly and Renilla luciferase activities were applied to the Luciferase Assay Kit (Abcam). Renilla luciferase activity was used as an internal control.

Western blot
E2F1 protein expression was measured by western blotting, as previously described [28]. The protein from the cells was extracted using a cell lysis buffer. Then, 20 µg protein was separated using 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred onto a polyvinylidene fluoride membrane, and blocked with 5% nonfat milk.

Statistical analysis
Independent sample t-tests for two-group comparisons and one-way analysis of variance with Dunnett's post hoc test for multiple group comparisons were used for data analysis, which was performed using GraphPad (CA, USA). The data are presented as the mean ± standard deviation based on three independent experiments. Pearson analysis was used to determine the correlation between E2F1 expression and miR-485-5p expression in glioblastoma tissues. Statistical significance was set at P < 0.05.

Results
In this study, we hypothesized that miR-485-5p inhibits glioblastoma progression by suppressing E2F1 expression under CDDP treatment. First, we identified the key miRNA (miR-485-5p) and target gene (E2F1) through bioinformatics analysis and found that CDDP treatment upregulated miR-485-5p levels in glioma cells and downregulated E2F1 levels. To investigate the role of CDDP, miR-485-5p, or E2F1 in glioma cells, we evaluated the effects of CDDP combined with miR-485-5p or E2F1 expression on the viability, proliferation, apoptosis, and invasion of glioma cells. Additionally, we validated target binding between miR-485-5p and E2F1. It was revealed that the CDDP-miR-485-5p-E2F1 axis may be a new direction for glioblastoma therapy.

Upregulated E2F1 level and downregulated miR-485-5p level were observed in glioblastoma
The high E2F1 expression in GEPIA glioblastoma data is shown in Figure 1a. GSE103229 from GEO DataSets stored the miRNAs expression profile in glioblastoma samples and normal brain samples. According to the ceRNA mechanism, there is a negative correlation between miRNA and its target genes. Therefore, to identify a potential upstream regulator miRNA of E2F1 in glioblastoma, we intersected the significantly downregulated miRNAs from the GSE103229 data series (screening criteria: P < 0.05 and logFC≤-2) and the predicted miRNAs that could regulate E2F1 by TargetScan algorithm. Seven candidates were identified to target E2F1 and be downregulated in glioblastoma samples (Figure 1b). The candidates are ranked according to their downregulation levels in GSE103229. miR-136-5p has been reported to be a significant tumor suppressor in glioblastoma and is the most significantly downregulated miRNA among the seven; however, it has been previously studied in CDDP resistance in glioblastoma [20]. The second most ranked miRNA, miR-485-5p, is a significant tumor suppressor in glioblastoma; however, it has never been studied in glioblastoma chemoresistance. Thus, we chose miR-485-5p as our study focus.

CDDP inhibited glioblastoma progression by upregulating miR-485-5p
To investigate the function of CDDP and miR-485-5p in glioblastoma, miR-485-5p expression in glioblastoma tissues was examined. The results showed that miR-485-5p expression was downregulated in glioblastoma tissues (Figure 2a). When CDDP was used to treat U251 and U87 cells, miR-485-5p expression increased (Figure 2b). Subsequently, we determined cell viability in U251 and U87 glioblastoma cells treated with CDDP at concentrations of 0, 10, 20,30 μM, and found that cell viability was significantly decreased at 10 μM treatment, which was the half maximal inhibitory concentration (IC50) (Figure 2c). Therefore, we selected 10 μM CDDP for subsequent experiments. In addition, to confirm the CDDP efficacy in glioma cells, cells were also targeted with TMZ. The results showed that cell viability decreased with an increase in TMZ concentration. The OD value of TMZ-treated cells was similar to that of CDDP-treated cells (Supplementary Figure 1). After transfection of the miR-485-5p inhibitor or miR-485-5p mimic in U251 and U87 cells with CDDP treatment, we found that the miR-485-5p inhibitor increased cell viability, miR-485-5p mimic decreased cell viability, and treatment with an inhibitor or mimic accentuated or reversed the inhibitory effect of CDDP on cell viability (Figure 2d). miR-485-5p inhibitor elevated cell proliferation, while the CDDP and miR-485-5p mimic groups reduced cell proliferation (Figure 2e). In addition, the enhanced or decreased cell adhesion of miR-485-5p inhibitor or miR-485-5p mimic were all partially eliminated or sharpened by the inhibitory adhesion of CDDP (Figure 2f). Moreover, 50% reduced cell apoptosis in the miR-485-5p inhibitor group was observed, but cell apoptosis observed in the CDDP or miR-485-5p mimic groups was higher (Figure 2g). Simultaneously, apoptosis changes caused by transfection with mimic or inhibitor were observed to be reversed or aggravated by CDDP treatment (Figure 2g). Finally, cell migration was elevated in the miR-485-5p inhibitor, while the CDDP or miR-485-5p mimic groups repressed cell migration (Figure 2h). Altogether, these data suggest that CDDP suppresses cell growth and enhances cell apoptosis by promoting miR-485-5p in glioblastoma.

E2F1 is a target of miR-485-5p in glioblastoma
To determine whether E2F1 interacted with miR-485-5p, the binding site between E2F1 and miR-485-5p was obtained using TargetScan Human 7.2 ( Figure 3a). The luciferase activity in cells treated with pmiRGLO-E2F1 E2F1 3-UTR WT and miR-485-5p mimic was decreased; however, no difference was observed in cells transfected with pmiRGLO-E2F1 3ʹUTR MUT, suggesting that miR-485-5p interacted with E2F1 in U251 and U87 cells (Figure 3b). Next, we found that E2F1 expression  was upregulated in glioblastoma tissues from 34 patients (Figure 3c), and a negative correlation between E2F1 expression and miR-485-5p expression was observed in glioblastoma tissues from 34 patients (Figure 3d). E2F1 expression was downregulated in U251 and U87 cells following CDDP treatment (Figure 3e). In addition, compared with the control group, the E2F1 protein level was upregulated in the E2F1-OE and miR-485-5p inhibitor groups, and downregulated in the mimic group; CDDP treatment facilitated the effect of miR-485-5p mimic and offset the effect of the inhibitor, suggesting that miR-485-5p is upstream of E2F1 under CDDP treatment in glioblastoma (Figure 3f and g).

CDDP hampered cell progression via the miR-485-5p-E2F1 axis in glioblastoma
To clarify whether the miR-485-5p-E2F1 axis participates in glioblastoma with CDDP treatment, the CCK8 assay was performed; it was observed that E2F1-OE upregulated cell viability, while this effect was reversed by CDDP and miR-485-5p mimic and aggravated by inhibitor (Figure 4a). Furthermore, E2F1-OE enhanced cell proliferation, while the effect was inhibited by CDDP and miR-485-5p mimic and promoted by the inhibitor (Figure 4b). Moreover, E2F1-OE promoted cell adhesion, while this effect was inhibited by CDDP and the miR-485-5p mimic and accelerated by the inhibitor (Figure 4c). Furthermore, E2F1-OE reduced cell apoptosis, while this effect was suppressed by CDDP and the miR-485-5p mimic and facilitated by the inhibitor (Figure 4d). Finally, E2F1-OE enhanced cell migration, but, this effect was hampered by CDDP and miR-485-5p mimics and aggravated by the inhibitor (Figure 4e). Collectively, the data indicate that CDDP hampers cell growth and promotes cell apoptosis via the miR-485-5p-E2F1 axis in glioblastoma.

Discussion
This study revealed miR-485-5p downregulation and E2F1 upregulation in glioblastoma tissues. miR-485-5p hampered cell proliferation, adhesion, and migration, and elevated cell apoptosis in glioblastoma cells following CDDP treatment. Moreover, miR-485-5p targeting E2F1 repressed glioblastoma progression following CDDP treatment. miRNAs have been shown to play pivotal roles in cancer with CDDP treatment. Liu et al. found that the knockdown of miR-6727-5 promoted the sensitivity of cervical cancer cells to CDDP [27]. Chen et al. showed that CDDP inhibited miR-132 expression in oral squamous cell carcinoma, thereby regulating the proliferation, invasion, and migration of cancer cells [26]. Moreover. Yang et al. reported that miR-29a significantly repressed prominin-1 (CD133) expression and contributed to CDDP resistance in CD133 + glioblastoma stem cells [29]. Guo et al. reported that Let-7b downregulation elevated cyclin D1 expression under CDDP treatment in glioblastoma cells [30]. Wang et al. found that 10 μM CDDP was an effective concentration for glioblastoma cell treatment, and miR-152-3p downregulation clearly decreased CDDP sensitivity [9]. Similarly, we found that 10 μM CDDP was an effective concentration for glioblastoma cell treatment. We further demonstrated that miR-485-5p retarded cell proliferation, adhesion, and migration, and elevated cell apoptosis in glioblastoma cells after CDDP treatment.
Although our study has revealed that miR-485-5p inhibits glioblastoma cell progression under CDDP treatment by targeting E2F1, the specific signaling pathways require further exploration. Meanwhile, animal models should be applied to confirm the role of the CDDP-miR-485-5p-E2F1 axis in glioblastoma.

Conclusion
Taken together, our study revealed that repressed cell growth and elevated cell apoptosis of glioblastoma cells occurred via the miR-485-5p-E2F1 axis under CDDP treatment. Thus, the CDDP-miR -485-5p-E2F1 axis may be an innovative direction for glioblastoma therapy.