RAC1/miR-3613/RAC1 feedback loop contributes to ovarian cancer progression

Changzhong Li (  15168888909@163.com ) Shandong University A liated Hospital: Shandong Provincial Hospital Ruobing Leng Shandong University A liated Hospital: Shandong Provincial Hospital Yunfang Meng Shandong University A liated Hospital: Shandong Provincial Hospital Na Li Shandong University A liated Hospital: Shandong Provincial Hospital Feifei Li Shandong University A liated Hospital: Shandong Provincial Hospital Yingzi Zhao Shandong University A liated Hospital: Shandong Provincial Hospital


Introduction
Ovarian cancer is one of the malignant tumors with the highest incidence in women. Its most common type is epithelial ovarian cancer (EOC), which has the leading mortality rate among gynecological tumors and thus seriously threatens women's lives [1,2]. The mainstream treatment of ovarian cancer is surgery to minimize the tumor burden, followed by chemotherapy to kill as many residual cancer cells as possible. However, many patients cannot tolerate the pain of chemotherapy [3]. In the era of molecular targeted therapy, people aim to extend the time from disease recurrence to death or the time interval to achieve a progress-free status; although the use of latest drugs bevacizumab and olaparib can delay this cancer progression, the opportunity to extend the lives of patients is still limited [4,5].
The abnormal expression of cancer-related genes renders current anticancer therapies less effective or even ineffective. Rho-GTPases RhoA, RAC1, and Cdc42 are important regulators of cytoskeletal dynamics. Although many in vitro and in vivo experiments indicated the pro-tumor effects of activated Rho-GTPase, some of its tumor suppressor functions have also been described [6,7]. Small GTPase is a molecular switch that alternates between the active GTP binding state and the inactive GDP binding state and regulates almost all aspects of cell biology through a series of complex and dynamic biochemical signal networks. RAC1 was rst described as promoting phagocytosis by activating NADPH and producing superoxide and growth factors to form membrane folds through actin polymerization [8,9].
Since then, RAC1 has been found to participate in various cellular processes, such as cell-matrix adhesion, migration, invasion, and proliferation [10][11][12]. Some studies con rmed the existence of nuclear RAC1 and revealed the mechanism that promotes RAC1 nuclear localization and its functional implications. For example, RAC1 directly interacts with BCA3 to promote nuclear NF-kB signaling [13].
Nuclear RAC1 directly interacts with nuclear phosphorus 1 (NPM1), which limits the GTP load of RAC1 and reduces cell proliferation [14].
On the one hand, miR-3613 binds to RAC1 mRNA to inhibit its translation. On the other hand, RAC1 inhibits the transcription of miR-3613 to relieve the inhibiting effect of the latter on the expression of the former and promote tumor progression to the greatest extent. This research enhances the understanding on the biological mechanism of RAC1 in tumors and may provide certain theoretical support for the development of anti-tumor drugs targeting RAC1.

Materials And Methods
MicroRNA array SKOV3 cells were transfected with RAC1 siRNA or siRNA NC. Total RNA was then isolated by TRIzol regent (Takara, Japan) in accordance with the manufacturer's instructions. RNA quality was determined by a spectrophotometer at 260 and 280 nm, and RNA integrity was evaluated by 1.5% formaldehyde denaturing gel. Flash Tag

qRT-PCR
The total RNA of the cells was extracted with Trizol (Sigma, USA) in accordance with the instructions.

MTT
The cells were resuspended in a culture medium containing 10% FBS and inoculated in a 96-well plate with 1,000 cells per well. After 24 hours, lipofectamin3000 was used to transfect miRNA mimics and DNA plasmids into the cells, which were then cultured for 4 days. After transfection, cell viability was detected by MTT method every day. Each well was added with 20 ul of 5 mg/ml MTT solution (Beyotime, China).
After incubation for 4 hours, the medium was discarded, 150 ul of DMSO was added, and the mixture was shaken for 10 minutes. The absorption value of each hole was measured at a wavelength of 450 nm.
Each experiment was performed in triplicate.

Colony formation assay
The cells in the logarithmic growth phase were digested with 0.25% trypsin and pipetted into single cells.
The cells suspended in culture medium containing 2% FBS were seeded into a six-well plate at a density of 1000 cells per well. All cells were maintained in an incubator containing 5% CO2 at 37 °C. The medium was changed every 3 days until visible clones were detected. Afterward, the medium was discarded, and the cells were carefully washed twice with PBS, xed with 4% paraformaldehyde for 15 minutes, stained with GIMSA for 10-30 minutes, and slowly washed with running water. Images were capture to record the number of clones. Each experiment was performed in triplicate.

Wound healing
Approximately 5x10 5 cells were seeded in a 24-well plate and culture for 24 hours until the cell density was more than 90%. A10 ul pipette tip was then used to make a straight scratch. The cells were washed three times with PBS to remove the suspended cells. A serum-free medium was added, and the cells were placed in a 37 °C 5% CO2 incubator for culture. After 24 hours, images were captured under a microscope, and the cell migration rate was calculated. Each experiment was performed in triplicate.

Transwell assay
In brief, 5 μg/μl Matrigel (BD, USA) was added to the upper chamber (Corning, USA) and placed at 37 °C for 30 min. The digested cells were washed with PBS, resuspended in a serum-free medium containing BSA, and seeded at a concentration of 10 5 per well. The lower chamber was added with 500 ul of medium containing 10% fetal bovine serum and cultured for 24 hours. The medium was then discarded, and the cells in the chamber were removed and xed with 90% alcohol at room temperature for 30 minutes. The cells were then stained with 0.1% crystal violet at room temperature for 10 minutes. After being washed with water, the invasive cells were recorded under a microscope (Olympus, DP50, Japan). Each experiment was performed in triplicate.

Western blot
The total protein of the cells was harvested with RIPA lysate (Beyotime, China), quanti ed by BCA method, and then separated by 10% SDS-PAGE. The separated protein was transferred to the PVDF membrane. After being washed three times with TBST, the membrane was blocked in 5% BSA overnight at 4 °C and then incubated with the following diluted primary antibody for 2 hours at room temperature: RAC1 polyclonal antibody (Abcam, ab155938, UK), N-cadherin polyclonal antibody (A nity, AF4039, China), vimentin polyclonal antibody (A nity, AF7013, China), and GAPDH polyclonal antibody (A nity, AF0911, China). HRP-labeled secondary antibody was then added to react at room temperature for 2 hours, and the specimens were nally visualized by ECL luminescence reagent (Beyotime, China) until the protein is clear. Each experiment was performed in triplicate.

Plasmid, oligo, and transfection
Pre-miR-3613 genomic sequence was synthesized and inserted into pcDNA3.1 vector to construct the miR-3613 expression plasmid. RAC1 expression plasmid was purchased from SinoBiological Inc.

Luciferase activity assay
The wild type and mutant 3 -UTR of RAC1 were synthesized and cloned into the pmirGLO vector. The cells were seeded into a 48-well plate. After 24 hours, the luciferase reporter plasmids were transfected when the cell density reached approximately 70%. After 48 hours, the culture medium was discarded, and the cells were washed with 100 μl of PBS. Each well was then added with 50 μl of diluted lysis buffer and then shaken for 20 minutes on a shaker to ensure that the cells were completely lysed. Fire y luciferase and renilla luciferase activities were detected following the instructions of Dual-Lumi™ dual luciferase reporter gene detection kit (Beyotime, China). Each experiment was performed in triplicate.

In vivo experiments
Eighteen 5-week-old female BALB/c nude mice were purchased from Charles River Laboratories, Inc. and randomly classi ed into three groups, namely, vector control, miR-3613 overexpression group, and miR-3613 and RAC1 overexpression group. The prepared cells were injected subcutaneously into the nude mice at a concentration of 5x10 6 cells per mouse. Tumor volumes were monitored every 3 days. After 1 month, all animals were euthanized through the intravenous injection of sodium pentobarbital at a nal concentration of 100 mg/kg. The surgically harvested solid tumor tissues were xed in 4% formalin, and their volume was calculated as follows: volume= (length×width2)/2. All the animals were well taken care of, and all the experiments were performed in accordance with the ethical standards of the Institutional Animal Use and Care Committee of the Shandong Provincial Hospital A liated to Shandong First Medical University. Ethical approval was obtained prior to the study. Each experiment was performed in triplicate.

Immunohistochemistry
Tumor tissues was embedded in para n and cut into 4 μm-thick sections. After heating in a microwave oven and blocking the antigen with 3% H2O2 solution, non-speci c binding was blocked with 5% bovine serum albumin at 37°C for 30 minutes. Then the sections were incubated overnight at 4°C with the following primary antibody, RAC1 (Abcam, USA, 1:50). The signal was detected by the DAB method (Beyotime, China) and examined under a microscope (Nikon, Japan). The immunohistochemical staining score was evaluated by two pathologists. The scoring system used is as follows: no positive cells are scored as 0; no positive cells are scored as 0. The yellow, light brown and dark brown staining of positive cells are 1-3 respectively. Each experiment was repeated three times.

Statistical analysis.
All statistical analyses were performed using SPSS 19.0 (SPSS, Chicago, IL, USA). Signi cant differences among two groups were compared using Student's t test. Comparisons among three or more groups were analyzed using ANOVA with post hoc Student-Newman-Keuls test. Statistical signi cance was considered at P < 0.05 and labeled with * .

Results
RAC1 negatively regulates miR-3613 in ovarian cancer cell RAC1 was knocked down in SKOV3 cells to analyze whether this substrate regulates ovarian cancer through miRNA. After TRIzol (Invitrogen, USA) was used to extract total RNA and identify its quality, the expression changes of miRNA were analyzed on the Agilent microarray platform. A heat map was drawn by selecting 34 miRNAs with the most evident expression changes (Fig. 1A). qRT-PCR was then used to detect the expression of 13 differential miRNAs. MiR-3613 was found to be up-regulated most signi cantly after RAC1 knockdown (Fig. 1B). In addition, 20 pairs of ovarian cancer and adjacent tissues were chosen to detect miR-3613 expression. The results showed that the miR-3613 expression was higher in EOC than in normal tissues (Fig. 1C). Finally, 49 patients with ovarian cancer (Table 1) were equally grouped into two according to miR-3613 expression and then analyzed for survival. The results showed that patients with high miR-3613 expression had a long survival time (Fig. 1D). This nding indicates that RAC1 may promote the malignant progression of ovarian cancer by inhibiting miR-3613.

miR-3613 inhibits ovarian cancer cell migration and proliferation
MiR-3613 overexpression plasmids were transfected in ovarian cancer cells OVCAR3 and SKOV3 to verify its role in ovarian cancer cells. The cells transfected with empty vector pcDNA3.1 were used as controls (Figure 2A). MTT and colony formation experiments were conducted for analysis. The results showed that miR-3613 overexpression inhibited cell viability ( Figure 2B) and proliferation ( Figure 2C). Transwell and wound healing assays were also performed to detect cell invasion and migration abilities, and the results showed that miR-3613 inhibited the invasion ( Figure 2D) and migration ( Figure 2E) of ovarian cancer cells and the expression of EMT-related markers N-cadherin and vimentin ( Figure 2F).

miR-3613 directly targets RAC1
The roles of miRNA depend on downstream target genes. Prediction results revealed two miR-3613 binding sites in the 3 -UTR region of RAC1 ( Figure 3A). Western blot analysis con rmed that miR-3613 repressed RAC1 expression ( Figure 3B). Dual-luciferase reporter experiments were then conducted to verify the binding of miR-3613 to RAC1 mRNA. The uorescent reporter vector carrying wild-type or mutant RAC1 mRNA was co-transfected with miR-3613 in OVCAR3 cells ( Figure 3C). The ndings showed that the uorescence intensity of the group carrying wild-type RAC1 mRNA was inhibited by miR-3613 and then recovered after the mutation ( Figure 3D). This result indicates that miR-3613 directly binds to RAC1 mRNA and inhibits the protein expression of RAC1. The expression of RAC1 mRNA and its correlation with miR-3613 were then analyzed in actual patients. RAC1 was found to be highly expressed in patients with ovarian cancer ( Figure 3E) and was negatively correlated with miR-3613 ( Figure 3F).

RAC1 reverses the inhibitory effect of miR-3613 in ovarian cancer cells
The above ndings revealed the existence of the RAC1/miR-3613/RAC1 feedback loop. A rescue experiment to further con rm the role of this pathway in ovarian cancer. After miR-3613 was expressed in OVCAR3 and SKOV3 cells, RAC1 was overexpressed to test its expression and determine whether it could reverse the inhibitory effect of miR-3613. The results showed that the exogenous RAC1 transfection restored ( Figure 4A) and even alleviated the inhibition of miR-3613 against cell viability ( Figure 4B), proliferation ( Figure 4C), migration ( Figure 4D), and invasion ( Figure 4E). Western blot results also showed that the expression of N-cadherin and Vimentin returned to normal after RAC1 expression was restored ( Figure 4F).

MiR-3613 inhibits the tumor growth of ovarian cancer in vivo
Finally, the effects of RAC1 and miR-3613 tumor growth were veri ed in nude mice. First, a stable OVCAR3 cell line overexpressing miR-3613 was established. The mice were subcutaneously inoculated with these cells at a concentration of 6x10 5 per mouse. RAC1-overexpressing lentivirus was injected into the tumor with the maximum diameter of 0.5 cm. After the nude mice were euthanized, the solid tumors obtained from surgery were measured and then xed with formaldehyde for subsequent immunohistochemical detection. The results showed that miR-3613 inhibited tumor growth, and RAC1 can reverse this inhibitory effect (Figures 5A and 5B). Immunohistochemical detection also con rmed that miR-3613 suppressed RAC1 expression in solid tumors ( Figure 5C).

Discussion
Malignant tumors are usually characterized by metastasis, which leads to poor prognosis and limited therapeutic possibilities for patients with tumors. Metastasis is one of the coordination processes between tumor cells and their microenvironment during malignant progression, which includes invasion and migration [15,16]. This step involves the active migration of tumor cells guided by integrins after matrix metalloproteinases degrade the extracellular matrix (ECM) [17]. The interaction between metastatic tumor cells and their ECM is essential for mediating the malignant phenotype. The ECM-integrin axis initiates and maintains important cellular functions and is an important regulator of tumor cell phenotype [18]. Rho GTPases contribute to different steps of cancer progression, including invasion and metastasis, and control signal transduction pathways. RAC1 is a small GTPase in the Rho family and mediates stimulus-induced (including integrin) actin cytoskeleton reorganization. Activated RAC1 acts as an intracellular signal transducer and regulates various cellular events, including cytoskeletal dynamics, to maintain cell morphology, polarity, adhesion, and migration. The extensive in uence of RAC1 on tumor cell behavior has led to its potential as a therapeutic target [7,19,20]. In ovarian cancer, Rac1 knockdown can inhibit EMT progression and tumor growth in vivo by increasing E-cadherin and reducing vimentin [21,22]. Here, we mainly focused on why ovarian cancer cells have persistent abnormal RAC1 expression and discovered the RAC1/miR-3613/RAC1 feedback loop. MiR-3613 inhibits the protein translation of RAC1, which in turn can achieve sustained expression by inhibiting miR-3613 expression. This nding served as a reminder that targeted therapy at the transcriptional and post-transcriptional levels of cancer-promoting genes appears to be indispensable.
EMT is a key aspect of tumor invasion and metastasis [23,24]. Ovarian cancer is closely related to EMT, particularly during tumor development and peritoneal metastasis [25,26]. RAC1 promotes EMT for many cancers [27][28][29]. An increased RAC1 activity can drive EMT in ovarian tumor cells. Inhibiting RAC1 activity or expression can restore the epithelial characteristics of ovarian tumor cells and prevent migration and invasion [21,[30][31][32]. MiRNA is a type of small non-coding RNA that controls the expression of target genes by inhibiting mRNA translation and inducing mRNA degradation. Many miRNAs regulate tumor invasion and metastasis through EMT-related mechanisms. In ovarian cancer, the EMT-inducing factor SNAI2 up-regulates PDCD10 by inhibiting miR-222-3p to induce EMT [33]. MiR-186 overexpression in ovarian cancer cells reduces Twist1 expression, thus making the cells highly sensitive to cisplatin in vitro and in vivo [34]. MicroRNA-150-5p promotes the EMT-mediated malignant progression of ovarian cancer by inhibiting c-Myb-mediated Slug suppression [35]. This study found that RAC1 can promote the EMT of ovarian cancer cells. MiR-3613 inhibits the protein translation of RAC1, which in turn suppresses the expression of miR-3613. This relationship allows ovarian cancer cells to continuously express RAC1 and promote the EMT and malignant progression of tumor cells.
In summary, a negative correlation was found between RAC1 and miR-3613 in ovarian cancer. RAC1 can maintain its continuous high expression in ovarian cancer cells by regulating miR-3613, thereby promoting malignant progression. This study is helpful to understand the mechanisms underlying the dysregulation of tumor-related genes and targeted drug development.     RAC1 reverses the effect of miR-3613 on ovarian cancer in vivo. After the OVCAR3 stable cell line overexpressing miR-3613 was injected subcutaneously into nude mice and the tumors had formed, the animals were injected with lentivirus overexpressing RAC1. The solid tumor obtained from surgery was measured and then xed, and IHC was used to detect RAC1 expression. A. Solid tumors after mice were scari ed. B. Tumor volume of each group. C. IHC was performed to detect RAC1 expression in these solid tumor tissues.