Up-regulation of miR-30b suppresses glioblastoma by negatively regulating MEF2D through Wnt/β-catenin signaling pathway

Purpose: To study miR-30b’ significance on glioblastoma, and its underlying mechanism of action. Methods: Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) while 3-(4,5)dimethylthiahiazo (-z-y1)-3,5-diphenytetrazoliumromide (MTT), Transwell, and xenograft tumor formation assays were carried out to study miR-30b’s effect on glioblastoma while luciferase reporter assay was employed to study the interaction between MEF2D and miR-30b. Glioblastoma cells treatment with miR-30 mimic or inhibitor were subjected to Western blot assay to study the effect of Wnt/β-catenin signaling on miR-30b/MEF2D axis-mediated cell progression. Results: MiR-30b was lowly expressed in glioblastoma tissues (p = 0.007), and this was associated with poor prognosis of patients (p = 0.022). The direct target of miR-30b was identified as MEF2D (p = 0.036). Increasing miR-30b blocked MEF2D expression in glioblastoma cells (p = 0.029). Moreover, MEF2D overturned miR-30b’ inhibitory effect on glioblastoma cell progression (p = 0.041; p = 0.006; p = 0.037). In vivo, restoration of miR-30b inhibited tumor growth (p = 0.01) and MEF2D. Interestingly, restoration of miR-30b inhibited epithelial-to-mesenchymal transition (EMT) and Wnt/β-catenin signaling pathways. Conclusion: These results indicate the critical role of miR-30b/MEF2D axis in glioblastoma via EMT and Wnt/β-catenin pathways. Thus, the miR-30b/MEF2D axis might be a beneficial therapeutic target for the management of glioblastoma patients.


INTRODUCTION
Glioblastoma is a common type of brain tumor with a high incidence [1]. Although significant progress has been made in the treatment of glioblastoma, the disease is still a big problem in clinical practice due to its poor prognosis. Therefore, it is necessary to seek innovative and specific therapeutic methods for the disease.
Accumulating evidence indicate that miRNAs are abnormally expressed in many malignancies, including glioblastoma. For example, miR-29a was reported to be lowly expressed in glioblastoma, resulting in suppression of the disease via TRAF4/AKT signaling pathway [2]. Increasing miR-785 repressed glioblastoma cell proliferation and metastasis by targeting ZBTB20 [3]. In glioblastoma, miR-30b was reported to be decreased, while restoration of miR-30b suppressed cell proliferation and apoptosis [4,5]. However, miR-30b' mechanism in glioblastoma is still unclear.
Myocyte enhancer factor 2D (MEF2D), a transcription factor, belongs to the MEF2 family [6] which has been showed to be involved not only in muscle development but also in tumor suppression, initiation and progression [7]. Increasing evidence show that MEF2D is upregulated in multiple cancers, and is linked to the etiology and prognosis of tumors [8,9]. However, its regulatory effect on glioblastoma has not been reported. This study was carried out to study miR-30b' functional role on glioblastoma, and the molecular mechanism involved.

EXPERIMENTAL Tissue samples
Forty-four glioblastoma tissues were obtained from glioblastoma patients undergoing a surgical procedure at Weihai Central Hospital, Weihai, Shandong, China. All participants included in this study received neither radiotherapy nor chemotherapy before surgery. The tissues were taken from the pathology department and stored at -80 °C in a refrigerator until RNA isolation. Before specimen collection, the written informed consent forms were signed by all participates participated in the study. The procedures used in this research were approved by the ethic committee of Weihai Central Hospital (approval no. 2017-014). All procedures are carried out according to the "Declaration of Helsinki" [10].

Cell lines and transfection
The glioblastoma cell lines (U251, T98G, A172 and LN18) and primary normal human astrocytes NHA cells were cultured following the instructions of manufacturer. Cells were incubated at 37 °C under 5 % CO2 atmosphere.
All transfection procedures were conducted with the aid of Lipofectamine 2000™ reagent (Invitrogen). Glioblastoma cells were transfected with miR-30b mimic/inhibitor to force or silence miR-30b expression, respectively.

qRT-PCR
Total RNA was isolated using TRIzol reagent (Invitrogen). The RNA was reverse-transcribed to cDNA with All-in-OneTM miRNA First-Strand cDNA Synthesis Kit. The quantification of all gene transcripts was conducted using TaqMan PCR kit with the method of 2 ™ ΔΔCt.

Western blot
The western blot was performed as described previously [11]. Briefly, proteins were separated and then transferred. Subsequently, they were blocked with 5 % skim milk, followed by primary antibodies and secondary antibodies. The loading control used was GADPH.

MTT assay
Glioblastoma cells were plated in 96-well plate. 20 uL MTT medium was added when cultured for 1, 2, 3, and 4 days. 100 uL dimethyl sulfoxide (DMSO) was then added and incubation for an additional 10 min. The absorbance of each well was detected at 490 nm wavelength.

Transwell assay
Transwell chamber without Matrigel precoating was applied for measuring cell migratory ability, while Transwell chamber precoated with Matrigel was used to determine cell invasive ability. Glioblastoma cells migrated or invaded from the upper to the bottom chamber were stained, and then photographed by a microscope.

Xenograft assays in athymic nude mice
Xenograft tumor formation assay was applied for measuring tumor growth in vivo. The U251 cells treatment with miR-30b mimic were added to culture medium (100 µL) and then 5 × 10 6 cells were injected into the nude mice' right flank (n = 8). A vernier caliper or electronic scale was used to measure the xenograft tumor size or tumor weight, respectively, every 4 days. The mice were sacrificed after 28 days after inoculation.

Luciferase reporter assay
Wild or mutant type of MEF2D-3'UTR fragment was amplified and then cloned into pGL3reporter vectors. The MiR-30b mimic and pGL3reporter vectors were co-transfected into U251 cells for 48 h. Luciferase activity was measured by Dual-Luciferase Reporter Assay System.

Statistical analysis
Data are presented as mean ± standard deviation (SD) in three experiments. Student's ttest or Tukey's post hoc test was employed to compare differences between groups. All statistical analysis were done using SPSS 17.0 statistical software and GraphPad Prism 5. Values of p < 0.05 were regarded as indicated of statistically significant differences.

MiR-30b expression was decreased in glioblastoma
RT-qPCR was applied to measure miR-30b level in glioblastoma. Results displayed that its mRNA level in glioblastoma tissues was obviously decreased, relative to normal tissues (Figure 1 A). Moreover, miR-30b was decreased in all glioblastoma cells versus to normal cells (NHA) ( Figure 1B). Furthermore, results of Kaplan-Meier survival curve revealed that miR-30b high expression was linked to high percentage survival of glioblastoma patients, whereas miR-30b low expression was associated with low survival rate of glioblastoma patients (Figure 1 C). Thus, miR-30b might be an indicator of the prognosis of glioblastoma patients.

MEF2D was a target of miR-30b in glioblastoma
TargetScanHuman was applied to predict the possible target of miR-30b in glioblastoma cells. As shown in Figure 2A, there have binding sites of miR-30b with MEF2D. To verify whether MEF2D was indeed regulated by miR-30b, luciferase reporter assay was applied in U251 cells. The luciferase activity of MEF2D in wild type was remarkably inhibited by miR-30b mimic in U251 cells, whereas there was no significant difference in mutant type ( Figure 2B). Moreover, miR-30b was over-expressed or under-expressed by miR-30b mimic or inhibitor, respectively, in U251 cells ( Figure 2C). Western blotting and RT-PCR analysis demonstrated that over-expression of miR-30b inhibited MEF2D expression, while MEF2D expression was raised by miR-30b inhibitor ( Figure 2D). Next, results discovered that MEF2D was increased in glioblastoma tissues, relative to normal tissues ( Figure 2E). Results in Figure 2F shown that the correlation between miR-30b mRNA and MEF2D was negatively (Figure 2 F). The findings indicate that miR-30b directly targeted MEF2D.

MiR-30b suppressed glioblastoma cell growth by targeting MEF2D
MEF2D expression was measured in U251 and T98G cells treatment with miR-30b mimic or combined with MEF2D plasmid. As shown in Figure 3A, MEF2D expression level was decreased by miR-30b mimic, but the increases were reversed by MEF2D plasmid. Results from MTT displayed that miR-30b over-expression suppressed cell viability, while MEF2D plasmid reversed this inhibitory effect ( Figure 3B). Transwell assay showed that increases in miR-30b concentration blocked glioblastoma cell invasion and migration, whereas MEF2D reversed this inhibitory effect (Figure 3 C and D). These results suggest that miR-30b suppressed glioblastoma cell progression by regulating MEF2D.

MiR-30b inhibited glioblastoma tumor growth in vivo
It has been found that miR-30b suppressed glioblastoma cell development in vitro, further investigations were done to see whether miR-30b had the same effect in vivo. Results revealed that miR-30b significantly reduced tumor volume, when compared to the control (Figure 4 A). Moreover, tumors with miR-30b plasmid grew more slowly than those with miR-NC (Figure 4  B). When the tumors were weighed using an electronic scale, it was found that the tumor was lighter in miR-30b mimic group ( Figure 4C). In addition, western blot analysis of MEF2D displayed that miR-30b over-expression inhibited MEF2D expression in vivo. These findings indicate that miR-30b suppressed glioblastoma tumor growth in vivo.

MiR-30b overexpression blocked Wnt/βcatenin pathway in glioblastoma
Mounting evidence showed that EMT and Wnt/βcatenin pathway participated in glioblastoma development. In this study, the precise molecular mechanism involved in miR-30b' effect in glioblastoma was investigated. U251 and T98G cells were treated with miR-30b mimic or inhibitor. The results revealed that miR-30b overexpression inhibited the expressions of Ncadherin and vimentin, while it increased Ecadherin expression (Figure 5 A). Conversely, miR-30b inhibitor exhibited the opposite effect on these levels. Thus, miR-30b inhibited cell invasion and migration through EMT. Besides, protein levels of p-β-catenin, c-Myc and Cyclin D1 were detected in U251 and T98G cells so as to determine miR-30b' underlying mechanism on cell proliferation. Results displayed that these above proteins were markedly repressed by miR-30b mimic, but they were enhanced by miR-30b inhibition ( Figure 5D). These observations indicate that miR-30b overexpression blocked Wnt/β-catenin pathway in glioblastoma.

DISCUSSION
Glioblastoma is the most common and most malignant astrocytoma of the central nervous system [12]. Even after various treatments including radical surgery, radiation, and chemotherapy, the average survival time of glioblastoma is about one year after diagnosis. Therefore, efforts directed toward better understanding of glioblastoma are essential for the development of efficient therapies. In previous studies, various miRNAs were found to influence glioblastoma tumorigenesis, including miR-422a, miR-1268, and miR-142 [13,14].
These observations have attracted intense investigation into miRNA-based therapies for glioblastoma. In particular, MiR-30b is a wellknown cancer-related miRNA which has become the hotspot of current studies. For instance, miR-30b was associated with the development of breast tumor [15]. Another study discovered a decrease expression of miR-30b in breast cancer, and that it negatively regulated cell invasion [16]. Moreover, a study discovered that miR-30b downregulation in lung cancer inhibited cell growth by targeting EGFR [17]. It is important to note that a single miRNA could regulate many target genes. For example, miR-30b attenuated gallbladder carcinoma via targeting NT5E [18]. Besides, miR-30b hampered cell proliferation and invasion in gastric cancer through regulating IGF-1R [19]. This present study found that miR-30b was lowly expressed in glioblastoma, which consistent with the prior researches reported that miR-30b was under expressed in glioblastoma [5].
Accumulating evidence suggest that MEF2D is involved in the growth of many tumors. A study reported that high expression of MEF2D in pancreatic cancer cells controlled cell growth [20]. Moreover, several studies have identified MEF2D as the target of miR-30a in osteosarcoma and lung cancer [21,22]. The present study verified that MEF2D was the target of miR-30b in regulating glioblastoma. In another study, it was reported that the upregulation of MEF2D in glioblastoma was associated with the prognosis of patients [23]. This is in line with the results that MEF2D was upregulated in glioblastoma. Moreover, it was demonstrated that MEF2D acted as miR-30b' target in inhibiting glioblastoma progression. Furthermore, the inhibitory effect of miR-30b mimic was attenuated by inhibiting the expression of MEF2D.
Wnt/β-catenin pathway shows a carcinogenic role in glioblastoma. For example, miR-505 suppressed glioblastoma through Wnt/β-catenin pathway [24]. Moreover, this signaling which is involved in glioblastoma, is regulated by miR-125b [25]. This study suggests that the Wnt/βcatenin pathway participates in the regulation of glioblastoma by miR-30b.

Limitations of the study
There were limitations in this study. First, the effect of MEF2D over-expression or knockdown on Wnt/β-catenin signaling pathway was not investigated. This would have provided a deeper insight into the mechanism at play. Secondly, the interaction of miR-30b with clinical characteristics was not considered. These areas should be investigated in future studies.

CONCLUSION
The findings of this study demonstrate that miR-33b/MEF2D is a potential alternative and effective approach for controlling abnormal proliferation and metastasis of glioblastoma cells. Thus, these findings underscore the importance of miRNA as a promising molecular tool in glioblastoma research.

Conflict of interest
No conflict of interest is associated with this work.

Contribution of authors
We declare that this work was done by the authors named in this article and all liabilities pertaining to claims relating to the content of this article will be borne by the authors. Jianfeng Bai, as the first author, contributed significantly to data analyses and manuscript preparation. Qingqing Yu performed data analyses and writing of the manuscript. Tongbo Ning, the corresponding author, contributed to the conceptualization of the study. All authors read and approved the final manuscript.

Open Access
This is an Open Access article that uses a funding model which does not charge readers or their institutions for access and distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/ 4.0) and the Budapest Open Access Initiative (http://www.budapestopenaccessinitiative.org/rea d), which permit unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.