LncRNA MBNL1-AS1 Represses Proliferation and Cancer Stem-Like Properties of Breast Cancer through MBNL1-AS1/ZFP36/CENPA Axis

Background Emerging studies have revealed long noncoding RNAs (lncRNAs) were key regulators of cancer progression. In this research, the expression and roles of MBNL1-AS1 were explored in breast cancer (BC). Methods In this study, the MBNL1-AS1 expression in breast cancer tissue, as well as in cell line, was studied by qRT-PCR assays. The effects of MBNL1-AS1 on proliferation and stemness were evaluated by MTT assays, colony formation assays, orthotopic breast tumor mice models, extreme limiting dilution analysis (ELDA), fluorescence in situ hybridization (FISH), flow cytometry assays, and sphere formation assays. Flexmap 3D assays were performed to show that MBNL1-AS1 downregulated the centromere protein A (CENPA) secretion in BC cells. Western blot, RNA pull-down assays, RNA immunoprecipitation (RIP) assays, and FISH were conducted to detect the mechanism. Results The results showed that the expression levels of MBNL1-AS1 were downregulated in breast cancer tissues and cell lines. In vitro and in vivo studies demonstrated that overexpression of MBNL1-AS1 markedly inhibited BC cells proliferation and stemness. RNA pull-down assay, RIP assay, western blot assay, and qRT-PCR assay showed that MBNL1-AS1 downregulated CENPA mRNA via directly interacting with Zinc Finger Protein 36 (ZFP36) and subsequently decreased the stability of CENPA mRNA. Restoration assays also confirmed that MBNL1-AS1 suppressed the CENPA-mediated proliferation and stemness in breast cancer cells. Conclusions The new mechanism of how MBNL1-AS1 regulates BC phenotype is elucidated, and the MBNL1-AS1/ZFP36/CENPA axis may be served as a therapeutic target for BC patients.


Introduction
Female breast cancer has surpassed lung cancer as the most common cancer, with about 2.3 million new cases (11.7%) [1]. It is the main cause of cancer-associated death for women around the world [2]. Approximately 15.4% of women died of breast cancer, although the diagnosis and treatment strategies improved greatly over the past several decades [1]. The abnormal proliferation of tumor cells is a striking feature of malignant tumors. Cancer stem cells (CSCs) have vital roles in intra-and intertumoral heterogeneity, which are related to tumor progression, treatment resistance, and disease relapse [3,4]. Therefore, exploring the potential molecular mechanisms underlying breast cancer proliferation and stemness is of paramount importance in identifying effective and novel therapeutic strategies.
Accumulating researches had showed that the expression pattern of lncRNAs was correlated with cancer progression, proliferation, and stemness [10][11][12][13]. Emerging evidence demonstrated that lncRNAs regulated the expression level of a target gene by binding to RNA-binding proteins (RBPs). For instance, the lncRNAs interacted with the RBPs and subsequently regulated cancer progression [14,15]. These researches showed that lncRNAs were pivotal players in cancer pathogenesis and vital new biomarkers in cancer early detection and therapy. Recently, two cohorts were analyzed to find lncRNAs of differential expression by research in BC. MBNL1-AS1 was regarded as one of the downregulated lncRNAs in BC [13]. However, the phenotype and mechanism of MBNL1-AS1 in BC have not been detected.
CENPA (a 17 kDa variant of histone H3) was located in the active centromeres [16,17]. A previous study has found that CENPA was correlated with human pluripotent stem cell self-renewal [18]. Low expression of CENPA induced cell cycle arrest and promoted apoptosis [19]. Subsequent evidence showed that upregulation of CENPA promoted initiation and progression in several cancers [17,20]. Higher expression of CENPA was correlated with increased invasiveness and higher-grade cancers [19,21]. In contrast, downregulation of CENPA showed to inhibit the HCC cells proliferation [19]. Moreover, CENPA had predictive value in breast cancer and could contribute to disease progression as a marker of proliferation [21,22]. In this research, MBNL1-AS1 was founded markedly decreased the expression of CENPA mRNA.
Here, the study showed that MBNL1-AS1 in BC was downregulated and the decreased expression of MBNL1-AS1 was correlated with survival. MBNL1-AS1 inhibited BC proliferation, and stem cell properties were confirmed by functional studies in vitro and in vivo. Mechanistically, MBNL1-AS1 directly interacted with ZFP36, an RNAbinding protein, subsequently reduced the stabilization of CENPA mRNA. This study might bring new insights into therapeutic targets for BC. . Indicated cells were cultured in DMEM containing 10% FBS (Thermo Fisher Scientific) and were supplied at 37°C in a humidified atmosphere of 5% CO 2 . 60 BC samples were obtained from the Department of Breast Surgery, The Second Affiliated Hospital of Harbin Medical University (246 Xuefu Street, Nangang District, Harbin, China). The 60 patients were Asian women, aged from 33 to 77 years. Fresh BC tissues and paired adjacent normal tissues were frozen in liquid nitrogen immediately after surgical excision and saved at -80°C. The tissues were validated by H&E staining. This research was approved by the Ethic Committee of Harbin Medical University.

Extreme Limiting Dilution
Assay. Extreme limiting dilution assay was performed as previously described [24]. Briefly, MCF-7 or MDA-MB-231 cells were plated in 96well ultralow attachment surface plates at a density of 100, 250, 500, and 1000 cells/well, with 12 replicates for each cell density at 37°C in a humidified atmosphere of 5% CO 2 . The number of wells presenting visible tumor spheres was counted on after 14 days, and then the data was put into the ELDA online software (http://bioinf.wehi.edu.au/ software/elda/). The stem cell frequency was calculated.
2.8. Flow Cytometry Assay. Flow cytometry assay was performed as previously described [25,26]. Briefly, using flow cytometry with the use of propidium iodide (Sigma-Aldrich, MO China) staining (5 μg/ml), cell cycle analysis was performed after transfection. The cells were trypsinized and dispersed into cell suspension. Next, 1 mL of cell suspension was centrifuged at 1,500 rpm for 10 min. Then, the indicated cells were added with chilled ethanol and saved at 4°C overnight. Then, the BC cells were centrifuged and resuspended in PBS and appended with 100 μL RNase at room temperature. After     Journal of Oncology blot. The assays were experimented in RNase-free conditions.
2.11. RNA Immunoprecipitation (RIP) Assay. RIP was performed with the use of the Magna RIP kit (Millipore, USA). The spectrophotometer (Thermo Scientific, USA) was used to examine the RNA concentration, and the bioanalyzer (Agilent, USA) was utilized to detect the RNA quality. The input control was the total RNA. Following, qRT-PCR analysis was used to test the results to demonstrate the bound targets.

Immunofluorescence Staining (IF) and Fluorescence In
Situ Hybridization (FISH) Assay. IF and FISH assays were performed as previously described [14,26]. For FISH, the MBNL1-AS1 subcellular localization was assessed by the lncRNA FISH Probe Mix and FISH kit which was obtained from Guangzhou RiboBio Co., Ltd. The 4% paraformaldehyde-fixed cells were fixed in PBS. After washed by phosphate buffered solution with tween for three times, the anti-fluorescence quencher sealed the indicated cells, and the images were taken by the FV1000 laser microscope (Olympus, Japan). For IF, cells were fixed with 4% formaldehyde and subjected to standard immunofluorescence staining. The anti-CENPA and secondary antibodies were obtained from Thermo Fisher Scientific. The data was analyzed by Image J software. The nuclei were dyed with DAPI. The 3D-cultured spheroids were conducted as previously described [26].
2.14. mRNA Decay Assay. Stable cells were added with 5 μg/ mL actinomycin D. Following, qRT-PCR was performed to determine the CENPA mRNA. The data was performed independently in triplicate.

Animal Studies.
Animal studies were ratified by the ethics committee of Harbin Medical University (Harbin, China). The BC cells were injected in the flank of fourweek-old female nude mice (each group has 3 mice). The volume of tumor was recorded every week. After intraperitoneal injection of D-luciferin, all mice were imaged by the Xenogen IVIS Spectrum Imaging System (Caliper Life Sciences, USA) before being sacrificed. Primary tumor volume was measured. The mice were euthanized by i.p. injection of sodium pentobarbital (200 mg/kg).
2.16. Statistical Analysis and Database. All analyses were experimented three times at least. The measurement data in this study were exhibited as means ± SD. Student's t-test was applied to compare the numeric variables between two groups. The differences between more than two groups of variables were evaluated by χ 2 or Fisher's exact χ 2 -test. R software package version 3.0.0 and GraphPad Prism 5 were used. The TCGA data set from the GEPIA2 Platform was used to analyze the expression of MBNL1-AS1 in tumor and normal tissues. And Kaplan-Meier Plotter was used to analyze the RFS in high and low expression of MBNL1-AS1. lncLocator database (http://www.csbio.sjtu.edu.cn/ bioinf/lncLocator/) was used to predict the subcellar localization of MBNL1-AS1. A p-value<0.05 was regarded as significant; p-value <0.01 was very significant.

MBNL1-AS1 Expression Was Downregulated in BC Cells
Which Was Correlated with Poor Prognosis. MBNL1-AS1 was regarded as one of the LncRNA suppressor genes associated with breast cancer, according to the bioinformatics analysis [13]. However, expression and function in breast tissue have not been confirmed yet. Therefore, the MBNL1-AS1 expression was detected in the cancer genome atlas (TCGA) database. As indicated in Figure 1(a), the MBNL1-AS1 expression pattern was marked lower than normal tissues in BC tissues. The MBNL1-AS1 levels in different BC subtypes were explored for the following study. The MBNL1-AS1 levels in tumor tissues were found markedly decreased in HER2+, Luminal A, and Luminal B subtypes compared with normal tissues. However, no statistical significance of MBNL1-AS1 expression was found in basal-like subtype between tumor tissues and normal tissues (Figure 1(b)). Next, the MBNL1-AS1 expression level was examined in 60 BC tissues and normal breast tissues. The H&E staining of the BC patients and normal sample is shown in Supplementary Figure 1A. Through the qRT-PCR, MBNL1-AS1 expression significantly decreased in the BC tissues (Figure 1(c)). Then, we linked the expression of MBNL1-AS1 with the clinicopathological features of the BC patients. We observed that the expression of MBNL1-AS1 had negative relationship with TNM stage and lymph node metastasis (Figure 1(d) and Table 3). Then, we detected the MBNL1-AS1 expression in several BC cell lines. Compared with the immortal MCF-10A cells and low-metastatic BC cell line MCF-7, the MBNL1-AS1 expression levels were observed to be much lower in BC cell lines, particularly in MDA-MB-468 and MDA-MB-231 cell lines with the higher metastatic features (Figure 1(e)).   (Figure 2(a)). As shown in Figure 2

12
Journal of Oncology significantly inhibited the CENPA secretion; however, other proteins were not of statistical significance (Figures 3(a)  and 3(b)). We, therefore, set out to detect the regulatory effect of MBNL1-AS1 on CENPA in BC cells. Interestingly, MBNL1-AS1 significantly reduced the levels of CENPA protein and mRNA in BC cell lines (Figures 3(c) and 3(d)).
As shown in Figure 3   Besides, researches showed that ZFP36 can specifically bind to AU-rich elements (ARE) in the mRNA 3'UTR and subsequently induce mRNA decay [27,28]. RIP assays were used to examine whether MBNL1-AS1 was physically associated with ZFP36. As indicated in Figure 4(a), these results demonstrated that MBNL1-AS1 could directly bind with ZFP36 in indicated cells. RNA pull-down analysis further demonstrated the interaction of MBNL1-AS1 and ZFP36. ZFP36 proteins were pulled down by MBNL1-AS1, but the antisense RNA was not pulled down (Figure 4(b)). To examine whether MBNL1-AS1 bound to ZFP36 and modulated the CENPA mRNA stability, we tested the CENPA mRNA half-life in indicated cells after ZFP36 was silenced. Before that, western blot assay was used to examine the ZFP36 knockdown efficiency in the BC cells (Figure 4(c)). As shown in Figures 4(c) and 4(d), MBNL1-AS1 could not affect the CENPA mRNA stability after ZFP36 silenced, nor decrease the protein pattern of CENPA in indicated cells. To conclude, MBNL1-AS1 directly interacted with ZFP36 and reduced the CENPA mRNA stability.

MBNL1-AS1 Suppressed Proliferation and Stemness of
Breast Cancer Cells by Interacting with ZFP36. To further validate our findings, MTT assays, colony formation assays, and sphere formation assays were performed to examine the function of ZFP36 in breast cancer cell lines and xenograft tumor models. ZFP36 knockdown significantly enhanced the growth of MCF-7 cells; however, knockdown of MBNL1-AS1 did not revert the proliferation abilities ( Figures 5(a) and 5(b)). The same results also showed that MBNL1-AS1 overexpression did not affect the growth of MDA-MB-231 cells after ZFP36 knockdown. Next, we examined the stemness abilities of indicated cells. As indicated in Figure 5(c), ZFP36 knockdown enhanced the stemness abilities of MCF-7 cells. However, knockdown of MBNL1-AS1 did not revert the stemness abilities of indi-cated cells. Similarly, sphere formation assays showed that MBNL1-AS1 knockdown or overexpression failed to revert the increase of sphere numbers in si-ZFP36 transfected BC cells. Meanwhile, the results of in vivo assays also confirmed that transfection of MBNL1-AS1 or si-MBNL1-AS1 did not revert the tumorigenic abilities after ZFP36 knockdown in orthotopic breast cancer mice models ( Figure 5(d)). Our data revealed that the interaction of MBNL1-AS1 and ZFP36 inhibited the proliferation and stemness of BC cells.

Discussion
LncRNAs are known as diagnostic markers for several kinds of cancers including BC. Previous research demonstrated various lncRNAs as vital players in BC progression [13,26]. In this research, the MBNL1-AS1 expression in breast cancer patients and its functions in BC cells were detected and investigated. The results indicated that expression of MBNL1-AS1 markedly decreased in breast cancer and high-metastatic BC cell lines, which was in line with the studies that revealed that MBNL1-AS1 levels were downregulated in colorectal cancer, NSCLC, retinoblastoma, and bladder cancer. Consistent with our research, another report showed that MBNL1-AS1 inhibited progression of NSCLC by sponging miR-135a-5p, which also identified MBNL1-AS1 as a tumor suppressor factor [29]. Also in NSCLC, a study revealed an inhibitory role of MBNL1-AS1 in CSC drug resistance of NSCLC by upregulating miR-301b-3p-targeted TGFBR2 [30]. Similarly, MBNL1-AS1 inhibited proliferation of bladder cancer through MBNL1-AS1/MiR-362-5p/QKI and MBNL1-AS1/miR-135a-5p/phlpp2/foxo1 signaling pathway [31,32]. The similar mechanism occurs in colon cancer; the study revealed an inhibitory role of MBNL1-AS1 by upregulating miR-412-3p-targeted MYL9 [33]. MBNL1-AS1/miR-338-5p/wnt/β-catenin signaling pathway also showed the anti-tumor effect in retinoblastoma [34]. These strongly demonstrated that MBNL1-AS1 might be a reliable biological marker to diagnose BC. The clinical significance of MBNL1-AS1 still needs to be further verified in more samples for the limitations of the present study. Moreover, MBNL1-AS1 inhibited the proliferation and Mounting evidence showed that lncRNAs regulated the downstream gene expression by binding to RBPs and competing endogenous RNAs (ceRNAs) [14,26,35,36]. ZFP36 was known as RBP, which bound to the target mRNAs untranslated regions subsequently reduced their stability [35,[37][38][39]. It was studied that the ZFP36 expression was downregulated in BC and regulated the stability of CENPA mRNA [38]. However, the upstream regulating factors except ZFP36 are still being researched. Our study demonstrated that the ability of ZFP36 to modulate the stability of CENPA mRNA was regulated by MBNL1-AS1. One research reported that the AP1 transcriptional factor components, including JUN, could transcriptionally regulate the expression of ZFP36 [40]. To clear if MBNL1-AS1 regulates the expression of ZFP36, so, we conducted western blot assays. The results showed that MBNL1-AS1 failed to regulate ZFP36 expression Supplementary Figure 3A). The data could strengthen our hypothesis that MBNL1-AS1 interacts with ZFP36 and promotes the function of ZFP36 as an RNA-binding protein. MBNL1-AS1 directly interacted with ZFP36 and subsequently reduced the stabilization of CENPA mRNA. Similar regulatory mechanisms were also found in previous studies [35,41]. In this research, the stability of CENPA mRNA was modulated by MBNL1-AS1 (Figure 3(a) and Figure 3(f)). Interestingly, in Figure 3(a), PLK1, PAF, YB-1, TWIST, YY1, KLF4, and SALL4 expressions were decreased but with no statistical significance in MBNL1-AS1 transfected cells. We find that MBNL1-AS1 has been reported to inhibit the progression of prostate cancer by sponging miR-181a-5p and regulating PTEN/PI3K/AKT/mTOR signaling [42] and PTEN signaling happens to be regulated by PLK1, PAF, YB-1, TWIST, YY1, KLF4, and SALL4 [43][44][45][46][47][48]. Therefore, PLK1, PAF, YB-1, TWIST, YY1, KLF4, and SALL4 might be the indirect downstream targets of MBNL1-AS1. As for CUG2, E2F8, RAE1, and PTPA, they were reported to be the downstream targets of several other factors, like NPM1, NONO, NKG2D, and UBE3A [49][50][51][52]. So, CUG2, E2F8, RAE1, and PTPA might not be the downstream targets of MBNL1-AS1 in breast cancer cells. Therefore, they may remain unchained. However, the possibility of translation of the MBNL1-AS1 control CENPA mRNA has not yet been detected. Thus, further study is still needed to the development of how MBNL1-AS1 regulated the CENPA expression.
MBNL1-AS1 was regarded as one of the tumor suppressors of lncRNAs in BC [13]. Nevertheless, no studies had confirmed the MBNL1-AS1 expression pattern in BC. Therefore, the expression pattern of MBNL1-AS1 was detected for the first time in breast cancer tissues and cell lines. MBNL1-AS1 was confirmed that the expression in breast cancer and highly metastatic cells was downregulated. That result was consistent with the bioinformatic data from TCGA. In functional assays, an anti-stemness and anti-proliferation function of MBNL1-AS1 was shown by decreasing CENPA expression. In accordance with our study, the researchers also confirmed that MBNL1-AS1 was a tumor-suppressive lncRNA [29,[31][32][33][34]. We demonstrated that MBNL1-AS1 attenuated the abilities of breast cancer stemness and proliferation through reduced stability of CENPA mRNA, which unraveled a novel mechanism of MBNL1-AS1. Undoubtedly, the MBNL1-AS1 and the interplay network and the vital roles of MBNL1-AS1 in BC were enriched and verified by this study.

Conclusions
We demonstrated that MBNL1-AS1 levels were downregulated in BC tissues, which were correlated with prognosis. In vitro and in vivo assays unraveled the anti-stemness and anti-proliferation roles of MBNL1-AS1. In mechanism, MBNL1-AS1 interacted with ZFP36 and subsequently reduced the stabilization of CENPA mRNA ( Figure 5(e)). Therefore, we elucidated a novel mechanism for how MBNL1-AS1 regulated the phenotype of BC and targeting the MBNL1-AS1/ZFP36/CENPA axis might function as therapeutic targets for breast cancer patients.

Data Availability
The data used or analyzed during the current study are available from the corresponding author on reasonable request.

Ethical Approval
This study was approved by the Ethical Committee of Harbin Medical University. The study was performed according to the ethical standards of Declaration of Helsinki. Additionally, the animal experiments were performed in accordance with the Guide for the Administration of Affairs Concerning Experimental Animals. The study was carried out in compliance with the ARRIVE guidelines.

Consent
Patient informed consent for the use of tissues was obtained prior to the initiation of the study.