MicroRNA-381 inhibits bladder cancer pathogenesis through directly controlling BMI1 and triggering the Rho/ROCK signaling pathway defect CURRENT STATUS:

Background Emerging evidence has noted the important participation of microRNAs (miRNAs) in several human diseases including cancer control. This research was launched to probe the function of miR-381 in bladder cancer (BCs) progression. Methods Aberrantly expressed miRNAs in BCs tissues were analyzed using miRNA microarrays. miR-381 expression in the bladder and paired tumor tissues, and in BCs and normal cell lines was determined. The target relationship between miR-381 and BMI1 was predicted online and validated through a luciferase assay. Gain-of-functions of miR-381 and BMI1 were performed to identify their functions on BCs cell behaviors as well as tumor growth in vivo . Results miR-381 was poor regulated in BCs tissues and cells. A higher miR-381 level indicates a better prognosis of patients with BCs. Artificial up-regulation of miR-381 inhibited proliferation, invasion, migration, resistance to apoptosis, and tumor formation ability of BCs cells. miR-381 directly binds to BMI1 expression. Overexpression of BMI1 partially blocked the tumor suppressing roles of miR-381 in cell malignancy and tumor growth. In addition, miR-381 led to decreased RhoA phosphorylation and ROCK2 activation, which were also reversed by BMI1. The study evidenced that miR-381 may act as a beneficiary biomarker in BCs patients. Up-regulation of miR-381 could suppress BCs development both in vivo and in vitro through BMI1 down-regulation and the Rho/ROCK inactivation. and and increased cell apoptosis. In addition, similar results were found in animal experiments, where cells with stable miR-381 overexpression contributed to a significant inhibition in tumor growth in volume and weight

non-muscle invasive BCs (NMIBD, < T2 stage) and the more risky and metastatic type, muscle invasive BCs (MIBD, T2-T4 stage) with the current or future distant invasion potential [3]. However, tumor recurrence is a major challenge of the BCs treatment with a rate of 61% in the first year, and the approximately 3%-15% recurrent NMIBD may progress to the invasive type MIBD when obtaining additional genetic mutations [4]. The current treatment for BCs is far below expectations, with the survival rate of patients saw little improvement during the last decades [2]. Identifying more molecular mechanisms may help develop novel effective treatments for this disease.
MicroRNAs (miRNAs), a large family of non-coding RNAs with 22 nucleotides in length, regulate multiple biological processes and are aberrantly expressed in human diseases including cancers [5].
miRNAs are well-known to exert key functions by inducing complementary target mRNA repression post-transcriptionally through the 3'untranslated region (3'UTR) [6]. Here in the paper, the miRNA microarray analysis was performed with miR-381 found as a major down-regulated miRNA in the BCs tissues. miR-381 has been noted as a tumor inhibitor in many human cancers [7,8]. In addition, it is noteworthy that miR-381-3p may inhibit CDK6 and MET-mediated epithelial-mesenchymal transition (EMT) and cell cycle progression of BCs cells [9]. But the roles of miR-381 in BCs progression are far away from being fully researched. Here, our study identified BMI1 as a putative mRNA target of miR-381. BMI1 is a key protein of polycomb repressive complex 1 that functions in integrity maintenance, and it has been increasingly known as a factor in the development of multiple human cancers [10].
Thereby, we proposed that miR-381 inhibits malignant behaviors of BCs through the down-regulation of BMI1. Altered expression of miR-381 and BMI1 was induced to explore their roles in cell malignant behaviors as well as the tumor growth in vivo, and the potential downstream signaling pathway was identified.

Clinical sample collection
BCs tissues and the paired adjacent tissues were obtained from 28 BCs patients (22 males and 6 females) who admitted into the First Affiliated Hospital of Jiamusi University from January 2013 to January 2015 with a mean age of 65 ± 5.67 years. A 4-year follow-up study was performed to record the prognosis of patients at a 3-month interval. The patients were divided into T1 to T4 stages according to the Tumor-Node-Metastasis (TNM) staging, with 7 in T1, 13 in T2, 5 in T3, and 3 in T4. All included patients had neither pre-surgery treatment history nor chemoradiotherapy history, and they were pathologically diagnosed as primary BCs free of other diseases. The tumor tissues and the adjacent tissues (at least 3 cm away from the tumor tissues) were resected during surgery and preserved in liquid nitrogen. The demographic characteristics of the respondents are exhibited in Table 1. miRNA microarray analysis Four patients in T1, T2, T3 or T4 stage (one in each stage) were enrolled for miRNA microarray analysis. The tumor and the paired adjacent tissues from patients were collected for total RNA extraction using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). Total RNA was concentrated in isopropanol, and then the RNA purity was detected using NanoDrop 2000C (Thermo Fisher Scientific Inc., Waltham, MA, USA) and the RNA quality was determined by formaldehyde denatured agarose gel electrophoresis. Then, 50 µg total RNA was purified using a Taqman miRNA ABC purification kit (Thermo Fisher) and hybridized with the beads. The beads were then washed with magnet, and the miRNAs were eluted 3 times to collect the miRNAs. Thereafter, the collected miRNAs were labeled using a miRNA Complete Labeling and Hyb Kit (Agilent, USA). In brief, 100 µg total RNA was incubated in 10 µg Labeling Spike-In at 37℃ for 30 minutes, treated with DMSO at 16℃ for 2 hours, dried in a vacuum concentrator at 45℃ for 3 hours, and then treated with Hybridization Mixture and Hyb labeling at 55℃ for 20 hours for hybridization. Then, the miRNAs were further hybridized with Human miRNA Microarray Release 14.0,8 (Agilent, USA) and scanned using a SureScan Dx Microarray Scanner (Agilent, USA). The obtained data were subjected to Quality Center analysis and normalization to produce the heatmap for differentially expressed miRNAs.
Reverse transcription quantitative polymerase chain reaction (RT-qPCR) Total RNA for cells or tissues was extracted using TRIzol Reagent and reversely transcribed into cDNA using a PromeScript TM RT Master Mix Kit (Takara Bio, Jan) and amplified. Relative gene expression was quantitated using a SYBR®Premix Ex Taq™II Kit (Takara Bio, Jan). The qPCR was performed on a Mx3005P System (Stratagene, USA). Quantification of each sample was repeated by three times.
Relative expression was evaluated using the 2 −ΔΔCt method with U6 and GAPDH as internal references. Table 2 exhibits the primer sequences.

Cell migration by scratch test
Guide lines were produced on the back side of 6-well plates at a 1-cm interval, and 5 × 10 5 cells were added into the plates. On the second day, a pipette was used to produce a scratch perpendicular to the back side guide lines, and the scratched cells were washed away by phosphate buffer saline (PBS). Then the plates were loaded with serum-free medium and cultured at 37℃ with 5% CO 2 .
Twenty-four hours later, the cells were observed and photographed under a microscope (CX22, Olympus). The scratch area was analyzed using the Image J software (version 1.48, NIH, USA).

Cell invasion by transwell assay
Matrigel was diluted in serum-free DMEM, and then 100 µL Matrigel was loaded onto each apical chamber at 37℃ for 4 hours. Then, cells were resuspended in 10% FBS to 5 × 10 5 cells/mL, and each apical chamber was loaded with 200 µL cell suspension, while each basolateral chamber was loaded with 600 µL cell medium (containing 5 µg/mL fibronectin). Then transwells were incubated at 37℃ for 20 hours, and the non-invaded cells in the apical chambers were removed, while the invaded cells in each well were stained with 500 µL 0.1% crystal violet at 37℃ for 30 minutes. Then the number of invaded cells were observed and counted under the microscope with 5 random fields included.

Flow cytometry
In short, 27 mL ddH 2 O was used to dilute 3 mL Binding Buffer (10 ×), and cells in each group were resuspended in 1 mL 1 × Binding Buffer, centrifuged at 300 g for 10 min to discard the supernatant, and resuspended in 1 mL 1 × Binding Buffer again till a density of 1 × 10 6 cells/mL. Next, each tube Dual-luciferase reporter gene assay The binding relationship between miRNA and mRNA was predicted on StarBase (http://starbase.sysu.edu.cn/). The 3'UTR sequence of BMI2 and miR-381 was inserted into pMIR-REPORT™ luciferase reporter vector (Thermo Fisher) using Lipofectamine 3000 (Invitrogen) and then transfected into HEK293T cells. Cells were lysed 24 hours later, and the luciferase activity was determined using a Dual-Luciferase Reporter Assay System (Promega Corporation, WI, USA) [13].
Three independent experiments were performed.

Western blot analysis
Radio-Immunoprecipitation assay cell lysis buffer (Amresco Inc., Texus, USA) was used to collect total protein, which was then and centrifuged at 800 g at 4℃ for 5 min to collect the cells. Then the cells were ice-bated in a 5-fold volume of lysis buffer for 10 min, and then centrifuged at 12,000 g at 4℃ for 10 min to collect the supernatant. The protein was run on SDS-PAGE and transferred to polyvinylidene fluoride membranes (EMD Millipore). Subsequently, the membranes were blocked with 5% skimmed milk and then incubated with the primary antibodies against RhoA (1:5000, ab187027), Differences in multiple groups were analyzed using one-way or two-way analysis of variance (ANOVA).
Tukey's multiple comparisons test was used for the post-hoc test after ANOVA analysis. Survival curve was drawn using the Kalpan-Meier method and analyzed using the log rank test. p was obtained by two-tailed tests and p < 0.05 was regarded statistically significant.

Results
Poor miR-381 expression in BCs patients indicates unfavorable prognosis The tumor tissues and paired adjacent tissues from 4 BCs patients (T1 to T4) were used for miRNA microarray analysis. It was found that miR-381 was the greatest down-regulated one in tumor tissues as compared to the normal tissues (Fig. 1A). Then, the RT-qPCR results validated that miR-381 was significantly poorly expressed in the tumor tissues in all the included BCs patients (Fig. 1B). According to the average miR-381 expression (0.46), the patients were allocated into high-expression group (n = 16) and low-expression group (n = 12). The follow-up study data suggested that patients with higher miR-381 had relatively higher survival rates and longer lifetime (Fig. 1C). In addition, we also determined miR-381 expression in cell levels, and it was noteworthy that miR-381 expression was much lower in BCs cell lines T24 and RT4 than that in normal bladder epithelial cells (Fig. 1D).

miR-381 inhibits malignant behaviors of BCs cells
To further figure out the roles of miR-381 in BCs development, up-regulation of miR-381 was introduced in both T24 and RT4 cell lines by administrating miR-381 mimic or mimic control ( Fig. 2A).
Then, we found that miR-381 mimic treatment led to a significant decline in proliferation in both cells (Fig. 2B). Accordingly, cells with increased miR-381 expression presented decreased migration distance (Fig. 2C) and reduced number of invaded cells (Fig. 2D). In addition, it was found that miR-381 mimic led to increased cell apoptosis (Fig. 2E). As for in vivo experiments, the implantation of cells with miR-381 mimic led to a significant decline in tumor growth in mice, and the tumor weight was decreased on the 28th day (n = 8) (Fig. 2F).
miR-381 targets BMI1 to inactivate the Rho/Rock pathway We next explored the potential downstream molecules involved. We first predicted BMI1 as a target of miR-381 on a computer-based bioinformation system, and then had the binding relationship between miR-381 and BMI1 validated through a luciferase assay (Fig. 3A). Then, it was found that BMI1 expression was increased in tumor tissues as compared to the adjacent tissues (Fig. 3B), and similarly, the BMI1 expression was also higher in BCs cell lines than that in bladder epithelial cells (Fig. 3C). The BMI1 expression in the tumor tissues of BCs patients was negatively correlated with miR-381 expression (Fig. 3D). BMI1-OE vectors were administrated into T24 cells and RT4 cells, and BMI1-OE#2 presented a best overexpressing efficiency (Fig. 3E). After miR-381 control, miR-381 mimic, BMI1-NC or BMI1-OE vector transfection, the protein levels of RhoA and ROCK2 were detected.
It was found that miR-381 mimic inhibited the phosphorylation of RhoA and ROCK2 while BMI1-OE led to increased activation of the Rho/ROCK signaling pathway (Fig. 3F). The above results indicated that miR-381 could target BMI1 to inactivate the Rho/Rock pathway.

Overexpression of BMI1 promotes BCs progression
To further identify the exact roles of BMI1 in BCs pathogenesis, overexpression of BMI1 was artificially up-regulated in T24 and RT4 cells. BMI1-OE vector led to an increase in cell proliferation (Fig. 4A).
Likewise, the migration rate (Fig. 4B) and invasion rate (Fig. 4C) of cells was notably increased following BMI1 overexpression. In addition, overexpression of BMI1 reduced the apoptosis rate in both cell lines (Fig. 4D). As for in vivo, we found cells transfected with BMI1-OE led to significantly increased tumor growth rate and weight in nude mice (n = 8) (Fig. 4E). These results identified BMI1 presented a reverse regulating role in BCs cells as relative to miR-381.

BMI1 partially blocks the role of miR-381
To further confirm the involvement of BMI1 in the miR-381-mediated events. Overexpression of BMI1 was introduced in cells pre-transfected with miR-381 mimic (Fig. 5A). It was found that further overexpression of BMI1 partially recovered cell proliferation (Fig. 5B), cell migration (Fig. 5C) and cell invasion (Fig. 5D) that was suppressed by miR-381 mimic. In addition, the miR-381-triggerred cell apoptosis was inhibited when BMI1 was further overexpressed (Fig. 5E). Accordingly, cells transfected miR-381 mimic and BMI-OE vector led to an increased tumor growth rate in nude mice compared to cells with miR-381 mimic alone (Fig. 5F). These findings suggested that BMI1 partially antagonized the effects of miR-381 mimic.

Discussion
Treatment for BCs, especially for the invasive type MIDB remains great challenge owing to high recurrence following surgical resection and drug administration. Aside from the unfavorable prognosis, the huge economic cost in lifetime surveillance with periodic cystoscopy and evaluation of recurrence rate brings considerable burden to BCs patients [14][15][16]. It has been as an emerging consensus that aberrant expression of some miRNAs, either extremely high or poor expression, is closely correlated with BCs pathogenesis [17]. Here, the current study evidenced that miR-381 could serve as a cancer suppressor in BCs with the involvement of BMI1 downregulation and the following Rho/ROCK inactivation.
The initial finding was that miR-381 was poorly expressed in the tumor tissues in BCs patients and was positively correlated with patient prognosis, preliminarily indicating a beneficiary role of miR-381 in BCs prognosis. miRNAs have been emerged as non-invasive biomarkers for BCs detection [18,19] or even potential treating targets in clinical practice [20]. For instance, miR-3622a was noted to promote proliferation and invasion capacities of BCss through binding to LASS2 [11]. On the other hand, some miRNAs have been documented to exert suppressing functions in BCs. miR-502-5p, for example, has been recently found to suppress the malignant behaviors of BCs cells through the different miRNA targets including CCND1, NOP14 and DNMT3B [21]. Similar trends have also been previously in miR-124-3p [22] and miR-203a [23]. As for miR-381, it has been particularly suggested to play critical functions in overcoming cisplatin resistance in breast cancer treatment [24][25][26].
Likewise, its anti-tumor role has been increasingly revealed in other malignancies such as in cervical cancer [27] and in prostate cancer [28] through the different mRNA targets. Here our study found that miR-381 was poorly expressed in BCs cells. Overexpression of miR-381 inhibited the malignant behaviors of BCs T24 and RT4 cell lines, presenting as decreased cell viability, reduced cell migration and invasion, and increased cell apoptosis. In addition, similar results were found in animal experiments, where cells with stable miR-381 overexpression contributed to a significant inhibition in tumor growth in volume and weight regards.
Second, the findings above triggered us to further explore the downstream molecules of miR-381. A computer-based online prediction system suggested BMI1 as a target mRNA of miR-381, and this binding relationship was further validated by a luciferase assay. As a member of the polycomb group family, BMI1, owning stem cell characteristics, is prone to participate in the onset and development of tumors and is linked to tumor metastasis, recurrence and chemo-resistance as well [29]. In addition, BMI1 is well-known to interact with another EMT and stemness promoter, TWIST1 [30,31]. Not surprisingly, silencing of BMI1 has been suggested as a promising therapeutic target in many human cancer therapies [32,33]. Here, our study found that artificial overexpression of BMI1 led to increased cell proliferation, migration, invasion and resistance to death in both T24 and RT4 cells. The involvement of BMI1 in BCs pathogenesis has been relatively largely studied. Its knockdown was noted to inhibit bladder cell growth, self-renewal and progression, and several upstream mediators of BMI1 such as miR-139-5p [34] and miR-200c [35] were identified. In the current paper, in order to further validate that BMI1 silencing holds accountable for the miR-381-mediated events, rescue experiments were performed where cells pre-transfected with miR-381. It was found that the malignant behaviors of cells inhibited by miR-381 mimic was partially recovered following BMI1 overexpression.
The Rho/ROCK signaling is well-known for its implication in cytoskeletal reorganization, which is elemental for cell migration and metastasis [36]. Here, our study identified that the phosphorylation of RhoA and ROCK2 expression was notably inhibited by miR-381, while overexpression of BMI1, notably, led to reversed trends. A Rho-associated kinase inhibitor, Y-27632, has been suggested to inhibit the migration and metastasis of BCss [37]. Therefore, it can be inferred that the Rho/ROCK pathway inactivation is possibly implicated in the miR-381/BMI1-mediated events.

Conclusion
To sum up, the present study evidenced that miR-381 could inhibit BCs progression, presenting as decreased cell proliferation, migration, invasion and resistance to death, as well as tumor growth in vivo. BMI1 is a target mRNA of miR-381 and its down-regulation is, at least partially, implicated in the miR-381-mediated tumor suppressing roles. The current study also suggested that the Rho/ROCK pathway was activated following BMI1 activation, indicating its potential implication in BMI1-induced cell growth and metastasis. However, the exact role of this pathway, and if it abrogates the effect of miR-381 in BCs, were not included in the current study. Anyway, the paper may provide novel insights into gene-based BCs treatment. We hope more studies will be carried to validate our findings and to provide more understanding in BCs control.

Ethics approval and consent to participate
This study was performed with the permission of the Clinical Ethical Committee of the First Affiliated Hospital of Jiamusi University. Signed informed consent was collected from each eligible participant.
All experimental procedures were conducted in line with the ethical guidelines for the study of experimental pain in conscious animals. Great attempts were made to minimize the number and suffering of animals.

Consent for publication
Not applicable.

Availability of data and materials
All the data generated or analyzed during this study are included in this published article. And the identifying/confidential patient data are not be shared.

Competing interests
The authors declare that they have no competing interests.

Funding
Not applicable.

Authors' contributions
DYC, LC proposed the protocol. DYC, LC, HFC, WSL were involved in data collection and management.
HFC, WSL analyzed the data. DYC, LC contributed to manuscript writing and critically revised the manuscript. All authors read and approved the final manuscript.     Diagram for the molecular mechanism. In BCs cells, miRNA-381 directly binds to BMI1 expression to inhibit RhoA phosphorylation and ROCK2 expression, leading to suppressed malignant behaviors of bladder cells.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download. ARRIVEChecklist.docx