Long non‐coding RNA SOX21‐AS1 enhances the stemness of breast cancer cells via the Hippo pathway

Previous reports have confirmed that long non‐coding RNA SOX21‐AS1 could regulate the properties of breast cancer stem cells (BCSCs), although the underlying molecular mechanism is unclear. We verified that SOX21‐AS1 was highly expressed in breast cancer tissues and cell lines, especially in BCSCs. SOX21‐AS1 could promote the stemness viability, proliferation, migration and invasion of BCSCs by inhibiting the Hippo pathway.


Introduction
Breast cancer is a type of malignancy and the global leading cause of cancer-related death in women [1]. Studies have confirmed that stem cell-like cell populations exist in tumor tissues [2][3][4]. Breast cancer stem cells (BCSCs) with self-renewal ability and multidirectional differentiation potential in breast cancer tissues are closely associated with the occurrence, development, post-treatment metastasis and the relapse of breast cancer [5,6].
Long non-coding RNA (lncRNA) is a class of RNA transcript that consists of more than 200 nucleotides and does not encode proteins [7]. Increasing evidence shows that lncRNA plays an important role in cell differentiation, proliferation and apoptosis [8]. At the same time, the role of lncRNA dysregulation in tumorigenesis and development has attracted more and more attention. Recent studies have shown that lncRNA SOX21-AS1 is located at chromosome 13q32.1 and transcribed into a 2986 nucleotide transcript, which exerts important functions in tumorigenesis and maintenance of tumor stem cell characteristics [9]. For example, SOX21-AS1 modulates breast cancer stem cells properties and carcinogenesis via targeting SOX2 [10]. Although there is evidence that SOX21-AS1 is involved in the proliferation, invasion and maintenance of the stemness of BCSCs [10][11][12], the underlying molecular mechanisms remain unclear.
Research has shown that the Hippo signaling pathway has important functions in organ size regulation, tissue regeneration, cancer development and stem cell function as a result of the regulation of cell proliferation, apoptosis and stem cell self-renewal ability [13,14]. In detail, the Hippo signaling pathway is a kinase cascade. Upon the induction of the Hippo signaling pathway by the activators WWC1 and Nf2 [15], the MST1/2 kinase forms a complex with SAV1 and phosphorylates LAST1/2, and the phosphorylated LAST1/2 kinase then phosphorylates the key downstream protein effectors of the Hippo signaling pathway, YAP and TAZ. pYAP can bind to proteins in the cytoplasm and stay in the cytoplasm, causing ubiquitination and degradation, thereby inhibiting the functions of YAP with respect to promoting growth, resisting apoptosis and maintaining stemness [16]. By contrast, when the Hippo signaling pathway is inhibited, YAP can be transported to the nucleus to bind to transcription factors such as TEAD1-4, thereby inducing gene expression, which promotes cell proliferation and maintains cell stemness [17]. The Hippo pathway maintains the characteristics of cancer stem cells and SOX21-AS1 can promote the development of cancer stem cells, although whether SOX21-AS1 acts on BCSC through this pathway is unknown. Accordingly, the present study aimed to investigate the potential mechanism of SOX21-AS1 on BCSCs and to provide potential therapeutic targets for the treatment of breast cancer.

Tissue sample collection
Breast cancer tissues and the tumor-adjacent non-tumor tissues were collected from Linyi People's Hospital in Shandong Province. Tumor-adjacent, non-tumorous tissues were collected ≥ 3 cm away from the cancer margin. The patients did not receive any chemoradiotherapy before surgical resection. This study was approved by the Ethics Committee of Linyi People's Hospital. The clinical features of the patients are listed in Table 1. Fresh tissues were sent for western blot and quantitative reverse transcriptase PCR (qRT-PCR) analysis immediately after resection or immediately frozen using liquid nitrogen. The study was granted permission by the Ethics Committee of Linyi People's Hospital and all patients or their guardians provided their written informed consent for inclusion in the study. The study methodologies conformed to the standards set by the Declaration of Helsinki.

Cell culture
Normal human breast epithelial cell line (MCF-10A) and human breast cancer cell lines (MCF-7, MDA-MB-231) were purchased from the American Type Culture Collection (Manassas, VA, USA). All three cell lines were successfully cultivated in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum at 37°C with 5% CO 2 . The BCSCs CSC-MCF-7 and CSC-MDA-MB-231 were cultured and selected as described previously [18,19].

Stem cell isolation
The cells were cultured in serum-free DMEM medium containing 2% B27, 20 ngÁmL À1 epidermal growth factor, 5 mgÁL À1 insulin and 4 gÁL À1 BSA in an ultra-low adherent plate and cultured at 37°C in a 5% CO 2 humidified incubator for 15-20 days.

Side population (SP) assay
For the SP assay, cells were suspended at a density of 1 9 10 6 cellsÁmL À1 and then incubated with 5 lgÁmL À1 Hoechst 33342 (Sigma-Aldrich, St Louis, MO, USA) at 37°C for 60 min. Then, 10 lM Verapamil (Sigma-Aldrich) was used as the blocker. Then, the cells were analyzed using a flow cytometer (Thermo Fisher).

Sphere formation assay
For sphere-forming assay, cells were seeded into the 48well ultra-low attachment plate (Corning Inc., Corning, NY, USA) in serum-free DMEM or RPMI 1640. After 7 days of culture in the medium supplemented with B27 (dilution 1:50; Invitrogen), 20 ngÁmL À1 epidermal growth factor (Invitrogen) and 20 ngÁmL À1 basic fibroblast growth factor (Invitrogen), the number of spherical cells were observed under a stereomicroscope (Olympus, Tokyo, Japan).

Clone formation assay
Briefly, CSC-MCF-7 cells were trypsinized into single-cell suspensions and seeded into six-well plates (500 cells per well) and then cultured in DMEM containing 10% fetal bovine serum for 14 days. The medium was replaced every 3 days. On day 14, the cells were fixed with 4% paraformaldehyde for 15 min and then stained using 1% crystal violet. The specific number of colonies was calculated manually.

Wound healing assay
After the transfected cells formed a monolayer, a 200-lL pipette tip was used to scratch a wound and the detached cells were washed away. The cells were cultured at 37°C for 48 h. Scratch wound healing was observed and monitored by microscopy (XSP-11CC; Caikon, Shanghai, China). All wound area measurements were evaluated using

Transwell assay
The migration and invasion of cells were examined using the transwell assay. For migration assay, CSC-MCF-7 cells (5 9 10 4 per well) were suspended in 200 mL of DMEM without serum and seeded into the chamber. For invasion assays, the Transwell membrane was coated with Matrigel (BD Biosciences, Frankline Lakes, NJ, USA). Twenty-four hours after incubation, the non-invaded cells were removed from the upper surface of the membrane and the cells that penetrated the pore were fixed and then stained with crystal violet.

Statistical analysis
All data are reported as the mean AE SD). SPSS, version 20.0 software (IBM Corp., Armonk, NY, USA) was used for statistical analysis. Differences between groups were measured by Student 0 s t-test or one-way analysis of variance. P < 0.05 was considered statistically significant. All experiments were performed at least three times.

SOX21-AS1 is highly expressed in breast cancer tissues and cell lines
The expression level of SOX21-AS1 in clinical breast cancer samples and cultured cell lines was detected by qRT-PCR. The results are shown in Fig. 1. Compared to the tumor-adjacent normal samples and human breast epithelial cell line, SOX21-AS1 was highly expressed in clinical breast cancer samples and cultivated cell lines. Moreover, the expression level of SOX21-AS1 in CSC-MCF-7 and CSC-MDA-MB-231 cells was higher than that of MCF-7 and MDA-MB-231 cells, among which the expression level of SOX21-AS1 in CSC-MCF-7 was the highest. Therefore, CSC-MCF-7 cells were selected for the subsequent experiments.
Next, the effect of SOX21-AS1 on the stemness of CSC-MCF-7 cells was determined by flow cytometry analysis. In Fig. 2B, the ratio of CD44 + /CD24in the Blank, pcDNA3.1-SOX21-AS1, pcDNA3.1, si-SOX21-AS1#1, si-SOX21-AS1#2 and si-NC groups was 8.96 AE 0.40%, 16.97 AE 0.51%, 8.72 AE 0.35%, 4.53 AE 0.26%, 3.79 AE 0.43%, 9.06 AE 0.21%, respectively. The results of flow cytometry showed that the ratio of CD44 + /CD24in the pcDNA3.1-SOX21-AS1 group was the highest, followed by the Blank, pcDNA3.1 and si-NC groups, and there was no significant difference with respect to the ratio of CD44 + / CD24in these three groups. The ratio of CD44 + / CD24was lower in the si-SOX21-AS1#1 group and the si-SOX21-AS1#2 group had the lowest ratio. From the results of flow cytometry, we inferred that SOX21-AS1 overexpression could promote the stemness of BCSC. Western blot analysis was used to detect the expression levels of stemness relative proteins, and the results are shown in Fig. 2C. The expression levels of SOX2, OCT4, ABCG2, NANOG and ALDHA1 proteins in the pcDNA3.1-SOX21-AS1 group were significantly upregulated, followed by the Blank, pcDNA3.1 and si-NC groups, whereas, in the si-SOX21-AS1#1 and si-SOX21-AS1#2 groups, the expression of SOX2, OCT4, ABCG2, NANOG and ALDHA1 was significantly inhibited. More importantly, the results of SP and sphere formation assays are shown in Fig. 2D,E. We found that, compared to the Blank, pcDNA3.1 and si-NC groups, overexpression of SOX21-AS1 in the pcDNA3.1-SOX21-AS1 group increased the percentage of SP and the sphere formation ability of the cells. However, after silencing the expression of SOX21-AS1, the percentage of SP and the cell sphere formation ability were decreased in the si-SOX21-AS1#1 and si-SOX21-AS1#2 groups.
Cell stemness is closely associated with the proliferation, migration and invasion of tumor cells. Therefore, we conducted experiments to determine the influence of SOX21-AS1 on CSC-MCF-7 cells. First, we explored the effect of SOX21-AS1 on the cell viability of CSC-MCF-7 cells using the CCK-8 assay. As shown in Fig. 3A, overexpression of SOX21-AS1 in the pcDNA3.1-SOX21-AS1 group significantly enhanced the cell viability of CSC-MCF-7 cells compared to the Blank, pcDNA3.1 and si-NC groups, whereas cell viability in the si-SOX21-AS1#1 and si-SOX21-AS1#2 groups was significantly decreased. The results of the CCK-8 assay showed that SOX21-AS1 could increase the viability of CSC-MCF-7 cells. Subsequently, the effect of SOX21-AS1 on the proliferation of CSC-MCF-7 cells was detected by a clone formation assay. The results shown in Fig. 3B indicate that the number of colonies in the pcDNA3.1-SOX21-AS1 group was significantly increased compared to the Blank, pcDNA3.1 and si-NC groups, whereas those in the si-SOX21-AS1#1 and si-SOX21-AS1#2 groups were significantly decreased. Among them, the number of colonies in si-SOX21-AS1#2 group was the lowest. From these results, we found that SOX21-AS1 promoted the proliferation of CSC-MCF-7 cells.
To investigate the effect of SOX21-AS1 on the migration ability of CSC-MCF-7 cells, a wound healing assay was performed. The outcomes are shown in Fig. 3C. The overexpression of SOX21-AS1 in pcDNA3.1-SOX21-AS1 group significantly accelerated the migration rate of CSC-MCF-7 cells, whereas the migration rate of the si-SOX21-AS1#1 and si-SOX21-AS1#2 groups was significantly reduced compared to the Blank, pcDNA3.1 and si-NC groups. These outcomes indicated that SOX21-AS1 could increase the migration ability of CSC-MCF-7.
Next, the effect of SOX21-AS1 on the invasion and migration ability of CSC-MCF-7 cells was further verified through a transwell assay, and the outcomes are shown in Fig. 3D. Compared to Blank, pcDNA3.1 and si-NC groups, the cell migration and invasion rate of the pcDNA3.1-SOX21-AS1 group increased significantly,   The results are shown in Fig. 4A. Compared to the Blank, pcDNA3.1 and si-NC groups, the expression levels of WWC1, Nf2, pMST1 and pLATS2 in the pcDNA3.1-SOX21-AS1 group were all decreased, and the expression level of pYAP was also significantly diminished. However, in the si-SOX21-AS1#1 and si-SOX21-AS1#2 groups, the expression levels of WWC1, Nf2, pMST1, pLATS2 and pYAP were significantly increased. Therefore, we considered that SOX21-AS1 could affect the stemness, proliferation, migration and invasion of CSC-MCF-7 cells by action on the Hippo signaling pathway.
Next, the possibility that SOX21-AS1 could promote stemness, as well as the proliferation, migration and invasion of CSC-MCF-7 cells, by inhibiting the Hippo signaling pathway was investigated via rescue experiments. YAP, as a co-transcription factor, binds to the transcription factor TEAD, which makes the effect of YAP on tumor activity largely dependent on TEAD-mediated transcription. As an activator of the Hippo pathway, sitagliptin can significantly reduce the nuclear localization of YAP, increase the pYAP level and decrease the expression of TEAD4. Flow cytometry was conducted to verify the effect of SOX21-AS1 on the stemness of CSC-MCF-7 cells through the Hippo pathway. Cells were divided into five groups (Blank, pcDNA3.1-SOX21-AS1 + control, pcDNA3.1 + control, pcDNA3.1-SOX21-AS1 + Hippo activator, pcDNA3.1 + Hippo activator) and the results are shown in Fig. 4B. The CD44 + /CD24ratio in the Blank, pcDNA3.1-SOX21-AS1 + control, pcDNA3.1 + control, pcDNA3.1-SOX21-AS1 + Hippo activator and pcDNA3.1 + Hippo activator groups was 6.11 AE 0.31%, 9.91 AE 0.37%, 6.02 AE 0.35%, 6.11 AE 0.25% and 3.83 AE 0.38%, respectively. In the pcDNA3.1 + Hippo activator group, the CD44 + /CD24ratio of CSC-MCF-7 cells decreased significantly, followed by the Blank, pcDNA3.1 + control and pcDNA3.1-SOX21-AS1 + Hippo activator groups, whereas, in the pcDNA3.1-SOX21-AS1 + control group, the CD44 + /CD24ratio was increased significantly. Compared to the pcDNA3.1-SOX21-AS1 + control group, the effect of SOX21-AS1 on the stemness of CSC-MCF-7 cells in the pcDNA3.1-SOX21-AS1 + Hippo activator group was weakened as a result of the addition of Hippo activator. Subsequently, the stemness relative proteins were detected by western blot analyis, and the results are shown in Fig. 4C. The expression levels of SOX2, OCT4, ABCG2, NANOG and ALDHA1 were significantly upregulated in the pcDNA3.1-SOX21-AS1 + control group and significantly decreased in the pcDNA3.1 + Hippo activator group compared to the Blank, pcDNA3.1 + control and pcDNA3.1-SOX21-AS1 + Hippo activator groups. Meanwhile, the effect of SOX21-AS1 on cell stemness relative proteins was weakened in the pcDNA3.1-SOX21-AS1 + Hippo activator group as a result of the addition of Hippo activator compared to the pcDNA3.1-SOX21-AS1 + control group. Subsequently, SP and sphere formation assays were also used to verify the effect of SOX21-AS on the stemness of CSC-MCF-7 cells through the Hippo pathway. Figure 4D,E shows that, in the pcDNA3.1-SOX21-AS1 + control group, the percentage of SP and sphere formation ability were significantly increased, whereas, in the pcDNA3.1 + Hippo activator group, these two values were significantly decreased compared to the Blank, pcDNA3.1 + control and pcDNA3.1-SOX21-AS1 + Hippo activator groups. It is worth noting that, compared to the pcDNA3.1-SOX21-AS1 + control group, the effect of SOX21-AS1 on cell stemness was significantly reduced in the pcDNA3.1-SOX21-AS1 + Hippo activator group as a result of the addition of Hippo activator. Therefore, the above results confirmed that SOX21-AS1 could promote the stemness of CSC-MCF-7 cells through the Hippo pathway.
The results of the CCK-8 assay shown in Fig. 5A demonstrated that, compared to the Blank, pcDNA3.1 + control and pcDNA3.1-SOX21-AS1 + Hippo activator groups, the cell viability of the pcDNA3.1 + Hippo activator group was significantly decreased, whereas that in the pcDNA3.1-SOX21-AS1 + control group was significantly increased. From a comparison between the pcDNA3.1-SOX21-AS1 + control group and the pcDNA3.1-SOX21-AS1 + Hippo activator group, it can be seen that the addition of Hippo activator weakened the effect of SOX21-AS1, thus decreasing the viability of CSC-MCF-7 cells. These outcomes showed that SOX21-AS1 could significantly increase the cell viability of CSC-MCF-7 cells through the Hippo pathway. Figure 5B shows that the colony number of CSC-MCF-7 cells treated by pcDNA3.1 + Hippo activator was reduced significantly, whereas that in the pcDNA3.1-SOX21-AS1 + control group was significantly increased compared to the Blank, pcDNA3.1 + control and pcDNA3.1-SOX21-AS1 + Hippo activator groups. In the pcDNA3.1-SOX21-AS1 + control group, the ability of SOX21-AS1 with respect to promoting the proliferation of CSC-MCF-7 cells was weakened as a result of the addition of Hippo activator compared to the pcDNA3.1-SOX21-AS1 + Hippo activator groups. The results of the clone formation assay indicated that SOX21-AS1 could increase the cell proliferation of CSC-MCF-7 cells through the Hippo pathway.
To confirm whether SOX21-AS1 can affect the migration of CSC-MCF-7 cells through the Hippo pathway, a wound healing assay was performed. The results are shown in Fig. 5C. We found that cell migration ability was the highest in the pcDNA3.1-SOX21-AS1 + control group, followed by the Blank, pcDNA3.1 + control and pcDNA3.1-SOX21-AS1 + Hippo activator groups, and, finally, the pcDNA3.1 + Hippo activator group. From a comparison between the pcDNA3.1-SOX21-AS1 + control group and the pcDNA3.1-SOX21-AS1 + Hippo activator group, we found that the addition of Hippo activator weakened the effect of SOX21-AS1 on the promotion of CSC-MCF-7 migration. The above results confirmed that SOX21-AS1 promoted the migration of CSC-MCF-7 cells through the Hippo pathway.
A transwell assay was applied to further confirm the effect of SOX21-AS1 on CSC-MCF-7 through the Hippo pathway. Figure 5D shows that the percentage of migrated cells in the Blank, pcDNA3.  whereas the pcDNA3.1 + Hippo activator group had the weakest ability with respect to cell migration and invasion. Compared to the pcDNA3.1-SOX21-AS1 + control group, the ability of SOX21-AS1 with respect to the promotion of CSC-MCF-7 cell migration and invasion was reduced in the pcDNA3.1-SOX21-AS1 + Hippo activator group by the addition of Hippo activator. In summary, we have confirmed that SOX21-AS1 could maintain cell stemness, as well as promote the proliferation, migration and invasion of CSC-MCF-7 cells, by inhibiting the Hippo pathway.

Discussion
Breast cancer is one of the common malignant tumors in women [20]. The discovery of tumor stem cells has confirmed that they play important roles in tumor survival, proliferation, metastasis and relapse [21]. In breast cancer, lncRNAs can regulate cell proliferation, migration and invasion, as well as maintain the stemness of BCSC [22][23][24]. It has been documented that the lncRNA level is significantly dysregulated in breast cancer compared to normal mammary cells. Tripathi et al. [25] demonstrated that LINC010187 was involved in oncogenic activity by controlling AKT1 and PAWR genes. Moreover, Zhou et al. [26] showed that lncRNA Hh strengthened cancer stem cell generation in breast cancer via induction of the hedgehog signaling pathway. Meanwhile, the function and mechanism of lncRNA in the development and prognosis of breast cancer have been increasingly emphasized [27]. Therefore, the in-depth exploration of lncRNA has a very important role in the early diagnosis of breast cancer, including the search for potential biomarkers and the prognosis [28]. In the present study, we investigated lncRNA SOX21-AS1 and found that the expression of SOX21-AS1 in breast cancer was disordered. It was confirmed that SOX21-AS1 could promote the stemness of CSC-MCF-7 cells through the detection of stemness-related proteins, as well as side population and sphere formation assays. At the same time, we also found that SOX21-AS1 had a significant promoting effect on the proliferation, migration and invasion of CSC-MCF-7.
The Hippo signaling pathway inhibits cell growth [29]. In mammals, upstream membrane protein receptors of the Hippo signaling pathway receive extracellular growth-inhibitory signals [30]. Once the extracellular growth-inhibitory signals are sensed, they activate a series of kinase cascade phosphorylation reactions and eventually phosphorylate downstream effector YAP and TAZ [31]. Subsequently, the cytoskeleton binds to phosphorylated YAP and TAZ, which causes them to remain in the cytoplasm for degradation, thereby inhibiting their function with respect to promoting proliferation and growth, as well as inhibiting apoptosis [32]. Conversely, unphosphorylated YAP and TAZ can enter the nucleus and bind to TEADs or other transcription factors, thereby inducing the upregulated expression of genes to promote proliferation and inhibit apoptosis [33]. Accordingly, the dysregulation of the Hippo pathway plays an important role in the occurrence and development of tumors. In the present study, we confirmed that the upregulation of SOX21-AS1 suppressed the Hippo pathway. SOX21-AS1 promoted the stemness of CSC-MCF-7 cells and also enhanced the proliferation, migration and invasion of CSC-MCF-7 cells by increasing the nuclear localization of YAP and reducing the level of pYAP.
In summary, we found that SOX21-AS1 could promote breast cancer stem cell stemness, proliferation, migration and invasion through the Hippo signaling pathway. The results of our study provide a reliable research basis for exploring the underlying action mechanism of SOX21-AS1 in breast cancer and indicate a new potential direction for the clinical treatment of breast cancer.