Inhibited Long Non-Coding RNA OIP5-AS1 Elevated microRNA-92a to Suppress Proliferation and Metastasis of Ovarian Cancer Cells by Silencing ITGA6

Objective: Over the years, long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) have been identied as essential biomarkers during the development of malignancies. This study was performed to verify the roles of lncRNA opa-interacting protein 5 antisense transcript 1 (OIP5-AS1) and miR-92a in ovarian cancer (OC). Methods: OIP5-AS1, miR-92a and ITGA6 expression in tissues and cells was assessed. The screened OC cells were respectively with integrin alpha 6 (ITGA6)/OIP5-AS1 silenced vector, miR-92a mimic/inhibitor or their negative controls. The viability, migration, invasion and apoptosis of the cells were determined and the levels of epithelial-mesenchymal transition (EMT)-related proteins were also measured. The interactions between OIP5-AS1 and miR-92a, and between miR-92a and ITGA6 were conrmed by dual luciferase report gene assay and/or RNA pull-down assay. Results: OIP5-AS1 and ITGA6 were upregulated while miR-92a was downregulated in OC tissues versus the adjacent normal tissues. Inhibited OIP5-AS1 or elevated miR-92a repressed EMT, viability, migration and invasion of OC cells, and promoted OC cell apoptosis. These effects that induced by silenced OIP5-AS1 could be reversed by miR-92a inhibitor. The reduction of ITGA6 restricted EMT in OC cells. MiR-92a was a target of OIP5-AS1 and ITGA6 was targeted by miR-92a. Conclusion: OIP5-AS1 silencing promoted miR-92a to repress proliferation and metastasis of OC cells through inhibiting ITGA6. This research may provide potential biomarkers for OC. alpha 6; GAPDH, glyceraldehyde phosphate dehydrogenase.


Introduction
Ovarian cancer (OC) is the most fatal among all reproductive cancers, the 11st commonest type and the 5th major reason of cancer-related death in women [1]. It was estimated by the most recent global statistic that there were 295,414 newly diagnosed cases of OC each year and 184,799 OC-related annual death [2]. This high mortality rate is induced by the absent or nonspeci c symptoms in early stages of OC, resulting in delayed diagnoses until late stages. More than 75% of OC cases were not diagnosed before advanced stages and the 5-year survival rate of late stage OC is about 30% [3]. In general, the initial treatments for OC are surgery and chemotherapy, which are effective for most patients. However, the recurrence appeared in around 80% of OC patients within several years and treatment for recurrence is seldom curative [4]. Thus, novel targets for OC treatment are urgently needed.
Long noncoding RNAs (lncRNA) are functional RNA molecules containing over 200 nucleotides. LncRNAs modulate the expression of key genes via epigenetic modi cation and transcriptional and posttranscriptional regulation [5]. It has been identi ed that lncRNA HOTTIP acted as a predictive role in OC prognosis [6] and lncRNA LINC00152 has been demonstrated to promote OC cell proliferation through regulating mitochondrial apoptosis pathways [7]. OPA-interacting protein 5 antisense transcript 1 (OIP5-AS1) situated at chromosome 15q15.1 and is evolutionarily conserved in vertebrates [8]. OIP5-AS1 has been revealed to participate in the progression of cervical cancer [9] and breast cancer [10]. Nevertheless, the role of OIP5-AS1 in OC remains rarely studied. It is known that lncRNAs are able to repress miRNA functions via serving as competing endogenous RNAs (ceRNAs) in human diseases [11]. MicroRNAs (miRNAs) are non-coding RNAs comprised of about 22 nucleotides, which post-transcriptional function on gene expression [12]. Some particular miRNAs have been reported to be implicated in OC. For instance, miR-203a-3p regulated the biological behaviors of OC cells [13], and miR-1307 has been identi ed to affect the chemosensitivity of OC cells [14]. MiR-92a is one of the miRNAs that was considered to be related to the progression of OC [15] and aggressive breast cancer [16]. Nevertheless, the combined effect of OIP5-AS1 and miR-92a is still unexplored. Integrin alpha 6 (ITGA6) is a 150-kDa transmembrane protein [17] that has been revealed to be related to drug resistance and prognosis in OC [18]. However, the target relation between miR-92a and ITGA6 remains unexplored.
We designed this research to investigate the role of OIP5-AS1/miR-92a/ITGA6 axis in OC, and we speculate that OIP5-AS1 may serve as a ceRNA to sponge miR-92a, thus regulating the biological processes of OC cells with the involvement of ITGA6.

Ethics statement
Written informed consents were acquired from all patients before this study. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) Total RNA in tissues and cells was extracted by TRIzol kits (Invitrogen). A NanoDrop ND-3000 spectrophotometer (Thermo Fisher Scienti c Inc., MA, USA) was employed for RNA quanti cation and the RNA was reversely transcribed into cDNA by PrimeScript RT Master Mix (TaKaRa Biotechnology Co., Ltd., Liaoning, China). The expression levels of target genes were a rmed by SYBR Premix Ex Taq II kits (TaKaRa) and the StepOnePlus system (Applied Biosystems Inc., CA, USA). Glyceraldehyde phosphate dehydrogenase (GAPDH) was used as the standardized control of OIP5-AS1 and ITGA6, and U6 was used as the standardized control of miR-92a. Data were analyzed using 2 −ΔΔCt method and the primer sequences were shown in Table 1.
RNA pull-down assay The DNA fragment with OIP5-AS1 or its NC sequence was performed with PCR ampli cation through the T7-containing primer and bound to GV394 (Invitrogen). Plasmid DNA was digested with restriction enzyme XhoI and the biotin-labeled RNAs were reversely transcripted. Expression of target RNAs was determined by RT-qPCR based on the method in a former publication [21].

Statistical analysis
All data analyses were conducted using SPSS 21.0 software (IBM Corp. Armonk, NY, USA). The measurement data conforming to the normal distribution were expressed as mean ± standard deviation. The unpaired t-test was performed for comparisons between two groups, one-way analysis of variance (ANOVA) was used for comparisons among multiple groups and Tukey's post hoc test was used for pairwise comparisons after one-way ANOVA. P value < 0.05 was indicative of statistically signi cant difference.

Results
lncRNA OIP5-AS1 and ITGA6 are upregulated while miR-92a is downregulated in OC tissues OIP5-AS1, miR-92a and ITGA6 expression in tissues were gauged and we found signi cant higher expression of OIP5-AS1 and ITGA6, and lower expression of miR-92a in OC tissues versus adjacent normal tissues (Fig. 1A, B).

Relationship between OIP5-AS1 expression and clinicopathological characteristics of OC patients
To further explore the relation between OIP5-AS1 expression and clinicopathological characteristics of OC patients, we divided the OC tissues in to the high expression group (n = 53) and the low expression group (n = 53) according to the median of OIP5-AS1 relative expression. The results suggested that OIP5-AS1 expression was higher in OC patients had advanced International Federation of Gynecology and Obstetrics (FIGO) stage (Fig. 1C), lymph node metastasis (LNM) (Fig. 1D) and larger tumor size (Fig. 1E).
Reduced OIP5-AS1 represses proliferation, migration, invasion and epithelial-mesenchymal transition (EMT) of OC cells, and accelerates cell apoptosis SKOV3 and A2780 cells were selected for the determination of OC cell biological processes due to the high OIP5-AS1 expression in the two cell lines (Fig. 2A). In SKOV3 and A2780 cells, inhibition of OIP5-AS1 repressed viability, invasion and migration abilities, and also promoted apoptosis of OC cells. EMT-related protein expression were also in uenced. Detailedly, the protein expression of E-cadherin was increased while that of vimentin was decreased by reduced OIP5-AS1 (Fig. 2B-G).
Elevated miR-92a constrains proliferation, migration, invasion and EMT of OC cells, and accelerates cell apoptosis SKOV3 and A2780 cells had a lower expression level of miR-92a (Fig. 3A), so the two cell lines were selected for subsequent experiments. In SKOV3 and A2780 cells, the upregulation of miR-92a restrained viability, invasion and migration abilities, and also facilitated apoptosis of OC cells. Moreover, the protein expression of E-cadherin was heightened while that of vimentin was lowered after the treatment of miR-92a mimic (Fig. 3B-G).

Inhibition of miR-92a reverses the effects of degraded OIP5-AS1 on biological processes of OC cells
In SKOV3 and A2780 cells, reduced miR-92a was able to reverse the impacts of OIP5-AS1 silencing on viability, migration, invasion and apoptosis of OC cells; in relation to the sh-OIP5-AS1 group, the protein expression of E-cadherin was suppressed while that of vimentin was promoted in the sh-OIP5-AS1 + miR-92a inhibitor group (Fig. 4A-F).

MiR-92a is a target of OIP5-AS1
Target relation between miR-92a and OIP5-AS1 was evaluated by a bioinformatic website. It was further con rmed that miR-92a broadly inhibited the luciferase activity of wt-OIP5-AS1 and miR-929a inhibitor obviously strengthened the luciferase activity of wt-OIP5-AS1, while the effect was not found in mut-OIP5-AS1 (Fig. 5A), indicating a direct interaction between miR-92a and OIP5-AS1. Subsequently, RNA pulldown assay was used to con rm whether OIP5-AS1 could pull miR-92a down. We discovered that miR-92a was enriched in OC cells that had been pulled down by biotinylated OIP5-AS1 (Fig. 5B). Moreover, we found through RT-qPCR that OIP5-AS1 expression was lower while miR-92a expression was higher in the sh-OIP5-AS1 group versus the sh-NC group; OIP5-AS1 expression showed no difference while miR-92a was upregulated in the miR-92a mimic group when compared with the mimic-NC group; in contrast to the sh-OIP5-AS1 group, OIP5-AS1 expression didn't alter while miR-92a was downregulated in the sh-OIP5-AS1 + miR-92a inhibitor (Fig. 5C). These data showed that miR-92a directly bound to OIP5-AS1 at miRNA recognition site.
ITGA6 is targeted by miR-92a and the reduction of ITGA6 inhibits EMT of OC cells A bioinformatic website was applied to predict the target relation between miR-92a and ITGA6. It was found in dual luciferase report gene assay that the overexpression of miR-92a apparently repressed activity of ITGA6 3'-untranslated region (3'UTR), while didn't affect the activity of mutant ITGA6 3'UTR; miR-92a inhibitor evidently promoted the activity of ITGA6 3'UTR, while didn't in uence the activity of mutant ITGA6 3'UTR (Fig. 6A). Furthermore, we found that the sh-OIP5-AS1 and miR-92a mimic groups had lower expression of ITGA6 than the sh-NC group and the mimic-NC groups, respectively; in comparison to the sh-OIP5-AS1 group, ITGA6 was overexpressed in the sh-OIP5-AS1 + miR-92a inhibitor group (Fig. 6B, C).
It was observed in SKOV3 and A2780 cells that ITGA6 was knocked down after the treatment of si-ITGA6 (Fig. 6D). The expression levels of EMT-related proteins were measured and the outcomes suggested that the protein expression of E-cadherin was increased while that of vimentin was decreased in the si-ITGA6 group versus the si-NC group (Fig. 6E).

Discussion
OC is a malignancy that threatens women's health globally. Although the incidence of OC is lower than that of cervical and endometrial cancers, its mortality is the highest among all gynecological cancers [22].
Accumulating evidence has shown that lncRNAs are capable of acting as ceRNAs to sponge miRNAs, thus modulating cell functions [23]. This research was performed to explore the role of lncRNA OIP5-AS1 in the progression of OC by sponging miR-92a via regulating ITGA6, and the results of our experiments indicated that the reduction of OIP5-AS1 elevated miR-92a to repress proliferation and metastasis of OC cells through downregulating ITGA6.
To begin with, we determined the expression levels of OIP5-AS1, miR-92a and ITGA6 in tissues. It was found that OIP5-AS1 and ITGA6 were upregulated while miR-92a was downregulated in OC tissues when compared with adjacent tissues. Consistently, Song et al. have unveiled that OIP5-AS1 was highly expressed in cervical cancer tissues [9]. A previous study has suggested that miR-92a was downregulated in breast cancer cells [16], and it has been identi ed that the expression of ITGA6 was increased in cisplatin-resistant SKOV3 and cisplatin-resistant A2780 cells, and also in drug-resistant tissues in comparison to the controls [18]. The relationship between OIP5-AS1 and clinicopathological characteristics of OC patients has been analyzed and we discovered that the high expression of OIP5-AS1 was associated with advanced FIGO stage, LNM and larger tumor size of OC patients. Similarly, Yang et al. have a rmed that highly expressed OIP5-AS1 was related to advanced FIGO stage, LNM and poor overall survival of cervical cancer patients [24]. Moreover, we have found that OIP5-AS1 served as a ceRNA to absorb miR-92a, and ITGA6 was targeted by miR-92a. The regulatory relation between OIP5-AS1 and miR-92a remains unexplored while ITGA5 has been unraveled to be a target gene of miR-92a.
However, the target relation of miR-92a and ITGA6 has not been studied yet.
Altered OIP5-AS1, miR-92a and ITGA6 were transfected into the OC cells to observe their roles in the biological behaviors of OC cells. The results of gain-and loss-of-function assays mirrored that silenced OIP5-AS1 or elevated miR-92a restrained proliferation, migration and invasion of OC cells, and also promoted OC cell apoptosis. In accordance with our ndings, Wang et al. have found that OIP5-AS1 promoted proliferation of lung cancer cells [25]. OIP5-AS1 has also been demonstrated to aggravate proliferation and migration of gastric cancer cells, and the cell apoptosis was induced by silenced OIP5-AS1 [26]. A recent publication has uncovered that miR-92a-3p suppressed cell growth in Wilms' tumor [27]. Gu et al. have also illuminated that miR-92a inhibited proliferation and induced apoptosis in acute myeloid leukemia [28]. Moreover, the expression of E-cadherin and vimentin was determined in our study, and we found that the knockdown of OIP5-AS1/ITGA6 or elevation of miR-92a heightened E-cadherin expression while lowered vimentin expression, indicating their repressive roles in EMT progression in OC.
In line with these results, Wang et al. have pointed out that overexpressed OIP5-AS1 facilitated EMT of laryngeal squamous cell carcinoma cells [29], and reduced OIP5-AS1 has been veri ed to inhibit EMT progress in hepatoblastoma cells [30]. Furthermore, it has been recently discovered that miR-92b inhibited EMT in triple negative breast cancer cells [31] and nasopharyngeal cancer cells [32]. In addition, Zhang et al. have clari ed that the oncogenic K-Ras upregulated ITGA6 to promote EMT [33].
To sum up, our study revealed that OIP5-AS1 elevated miR-92a to suppress proliferation and metastasis of OC cells by silencing ITGA6. This research may be helpful for exploring therapeutic strategies for OC.
However, more efforts are required to investigate the detailed mechanisms.

Declarations
Funding None Ethics approval and consent to participate Written informed consents were acquired from all patients before this study. The protocol of this study was con rmed by the Ethic Committee of Sichuan Academy of Medical Sciences Sichuan Provincial People's Hospital and based on the ethical principles for medical research involving human subjects of the Helsinki Declaration.

Con ict of interest
The authors declare that they have no con icts of interest.

Consent for publication
Not applicable Availability of data and material