Reduction of Bladder Cancer Chemosensitivity Induced by the Effect of HOXA-AS3 as a ceRNA for miR-455-5p That Upregulates Notch1

Chemoresistance is one of the main causes of recurrence in bladder cancer patients and leads to poor prognosis. Recently, long non-coding RNAs, like HOXA-AS3, have been reported to regulate chemoresistance in several types of cancer. In this study, we aimed to determine whether HOXA-AS3 can mediate cisplatin resistance in bladder cancer, and its potential mechanism of action. We determined the viability, proliferation, and apoptosis of bladder cancer cells using a CCK-8 assay, EdU staining, and flow cytometry, respectively. We used western blot analysis to assess the expression of markers of epithelial-mesenchymal transition (EMT) and Notch1. We then confirmed expression of these EMT-related markers by immunofluorescence analysis. We found that hypoxia promoted resistance to cisplatin and upregulated the level of HOXA-AS3 in BC cells. Inhibition of HOXA-AS3 enhanced hypoxia-induced cisplatin sensitivity by regulating EMT and Notch1 in BC cells. A dual-luciferase reporter assay confirmed that HOXA-AS3 directly targets miR-455-5p and that Notch1 was a potential target of miRNA-455-5p. We also found that the positive effect of HOXA-AS3 inhibition on cisplatin resistance and tumorigenesis was alleviated when BC cells were transfected with miR-455-5p. Finally, we showed combining HOXA-AS3 small interfering RNA (siRNA) with cisplatin treatment inhibited tumorigenesis in a BALB/c nu/nu mouse model. Our findings indicate that HOXA-AS3 may function as a competing endogenous RNA (ceRNA) of miR-455-5p to regulate Notch1 and play an important role in regulating chemotherapeutic drug sensitivity in BC cells. Therefore, HOXA-AS3 may be a novel therapeutic target for treating bladder cancer.


INTRODUCTION
Bladder cancer (BC) is ranked as the ninth most common cancer worldwide, and men are more than three times more likely to develop the disease than women (1,2). Cisplatin is extensively used in the clinic as a first-line treatment for many cancers, including liver, ovarian, non-small cell lung cancer (NSCLC), and BC (3)(4)(5); however, its efficacy is limited due to severe side effects and the development of drug resistance (6). Indeed, most patients become irresponsive to cisplatin and eventually die of the disease (7). Therefore, understanding the mechanisms behind cisplatin resistance is key for improving BC treatment and prognosis.
Long non-coding RNAs (lncRNA), a group of evolutionarily conserved RNAs longer than 200 nucleotides, modulate gene expression at the epigenetic, transcriptional, and posttranscriptional levels (8). Recent studies have shown abnormal expression of lncRNA in many types of tumors, playing important roles in cancer progression, including cell proliferation, differentiation, invasion, and metastasis (9,10). Importantly, lncRNAs are considered key regulators in drug resistance, and may also act as promising prognostic and therapeutic targets. Several studies have shown lncRNAs may interact with miRNAs and mutually modulate their expression (11,12). They may function as competing endogenous RNAs (ceRNAs) with miRNAs and subsequently regulate gene expression by post-transcriptional silencing of the target RNAs (13,14). Indeed, the interaction of lncRNAs and miRNAs may provide new insight into cancer biology. For instance, knockdown of the lncRNA SBF2-AS1 increased the chemosensitivity of gemcitabine through inhibiting expression of the twinfilin actin binding protein 1 (TWF1) by competitively binding to miR-142-3p in pancreatic cancer (15). In addition, the absence of the lncRNA FOXD2-AS1 enhanced cisplatin sensitivity of NSCLC cells by regulating the miR-185-5p-SIX1 axis (16). Therefore, understanding the functions of lncRNAs in cancer has become an area of extensive research.
HOXA-AS3 belongs to the clusters of HOX genes, a group of highly homologous transcription factors that regulate embryological development, and also regulate hematopoietic lineage and differentiation (17,18). There is mounting evidence that lncRNAs, such as HOX-AS3, regulate cancer cell growth, metastasis, and resistance to chemotherapy in several types of tumors (19), and this makes them promising targets for novel cancer therapies (20). Previous studies have shown that overexpression of the lncRNA HOXA-AS3 promotes tumor progression and is a predictor of poor prognosis in glioma (21). Increased HOXA-AS3 levels were also found to promote proliferation in lung adenocarcinoma (22). However, the mechanisms of these lncRNAs in the various types of cancer can vary. For instance, knockdown of the lncRNA TUG1 promoted cisplatin sensitivity in an esophageal squamous cell carcinoma cell line (TE-1) by regulating nuclear factor-like 2 (Nrf-2) expression (23). Further, inhibition of the lncRNA HOTAIR enhanced doxorubicin sensitivity in breast cancer by downregulating the PI3K/AKT/mTOR pathway (24). Based on these findings, we sought to unravel the role of HOXA-AS3 in BC.
Hypoxia plays a role in resistance to chemotherapeutic treatments in several cancers (25)(26)(27). It is commonly present in the microenvironment of solid tumors and is associated with tumor invasion, distant metastasis, and epithelial-mesenchymal transition (EMT) (28)(29)(30). HOXA-AS3 inhibition has been shown to induce epithelial-mesenchymal transition (EMT) in NSCLC, and increase its resistance to cisplatin (31). EMT is a critical mechanism of cancer metastasis. During the EMT process, cancer cells lose their epithelial features and acquire a mesenchymal phenotype, and thus obtain enhanced metastatic ability (32). It has been reported that inhibition of SNHG7 could promote tumor growth and the EMT phenotype via upregulation of the targets of miR-34a, including Notch1 (33).Therefore, lncRNAs such as HOXA-AS3 may modulate EMT in various types of cancer, including BC, by regulating the expression of Notch1. Indeed, HOXA-AS3 knockdown was previously shown to inhibit the proliferation, metastasis, and EMT in hepatocellular carcinoma (HCC) cells via the MEK/ERK signaling pathway (34). Thus, HOXA-AS3 may regulate EMT in BC in a similar manner, i.e., by interacting with miRNAs or altering the expression of proteins such as Notch1. However, the effect of HOXA-AS3 on drug sensitivity (e.g., to cisplatin) and the regulation of EMT in BC are yet to be explored.
In this study, we examined whether HOXA-AS3 mediates cisplatin resistance in BC cells, and then unraveled its underlying mechanism(s) of action.

Cell Culture and Human Tissues
Human bladder cancer cells (UM-UC-3, J82, and BIU-87) were purchased from ATCC. All cells were cultured in RPMI 1640 medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin, maintained in humidified air containing 5% CO 2 at 37°C. The tumor tissue samples of BC patients were obtained from First Affiliated Hospital of Zhejiang University. All patients received written informed consent and the study protocol was approved by the Clinical Research Ethics Committee of the First Affiliated Hospital of Zhejiang University.

Cell Viability
A cell counting kit-8 (CCK-8; Dojindo, Kumamoto, Japan) assay was used to determine cell viability. Briefly, BC cells were plated into 96-well plates at a density of 3 × 10 3 cells/well and incubated overnight at 37°C. Subsequently, cells were treated with a series of concentrations of cisplatin (0, 0.625, 1.25, 2.5, 5, or 10 mM) for 48 h or transfected with HOAXA-AS3 siRNA, miR-455-5p inhibitor, or HOXA-AS3 siRNA plus miR-455-5p inhibitor for 48 h. Then, 10 ml of CCK-8 solution was added to each well and cultured for 3 h at 37°C before the absorbance at 450 nm was measured using an MRX II microplate reader (Dynex Technologies, Chantilly, USA).
Relative cell viability was calculated as a percentage of the untreated controls.

Luciferase Assay
HOXA-AS3-3'UTR wild type or mutant reporter plasmids (which included the binding sequence for miR-455-5p or the mutated binding sites, respectively) were constructed by GenePharma. The plasmids were then transiently co-transfected into cells with luciferase reporter vectors with the miR-455-5p-mimic, miR-455-5p inhibitor, or control using Lipofectamine 2000. After transfection for 48 h, the relative luciferase activity of the wild type or mutant HOXA-AS3-3'-UTR was measured with a dualluciferase reporter assay (Promega, USA).

Flow Cytometry Analysis
The number of apoptotic cells was determined by an Annexin V-FITC Apoptosis Detection Kit (Abcam). In brief, cells (treated as above) were harvested by trypsinization, rinsed with ice-cold phosphate PBS, and centrifuged to remove the supernatant. Then, the cells were resuspended in 100 ml 1× binding buffer and stained with 5 ml annexin V and 5 ml propidium iodide (PI) for 15 min at room temperature in the dark. Finally, the percentage of apoptotic cells was determined using a flow cytometer (LSRII, BD Biosciences, Franklin Lakes, NJ, USA)

Cell Proliferation Assay
Cell proliferation was determined using a Click-iT ® EdU Imaging Kit according to the manufacturer's (Invitrogen) protocols.

Immunofluorescence Analysis
NSCLC cells were seeded into 48-well plates at a density of 3 × 10 3 cells/well.

Nude Mouse Xenograft Model
Female BALB/c nu/nu mice (4-5 weeks old) were purchased from Shanghai SLAC Laboratory Animal Co., Ltd (Shanghai, China). UMUC-3 cells were subcutaneously injected into the left hip of three mice. After the tumor was formed, a small section (1 mm 3 ) of tumor tissue was inoculated into the experimental group nude mice. After 10 days, the tumors had a diameter of 0.5 cm and reached a volume of~50-100 mm 3 . The mice were randomly divided into four groups (n = 16 per group): control, HOXA-AS siRNA (2 nmol), cisplatin (2.5 mg/ kg), or HOXA-AS3 siRNA plus cisplatin. HOXA-AS3 siRNA was injected intratumorally four times from day 0 to 14, while cisplatin was injected into the tail vein once every 2 days for 2 weeks. Tumor volumes were recorded every 2 days and body weight was measured daily. The tumor volume (mm 3 ) was calculated using the formula V = (length × width 2 /2). 8

Terminal Deoxynucleotidyl Transferase dUTP Nick End Labeling
TUNEL was used to identify apoptosis in paraffin-embedded mouse tissue sections (5 mm) with an in situ cell death detection kit (Roche, Basel, Switzerland) according to the manufacturer's instructions. The apoptotic cells were observed under a light microscope (Olympus, Tokyo, Japan). The assay was independently repeated three times. The positive rates were measured using Image-Pro Plus v.6.0 software.

Statistical Analysis
The experimental data are presented as means ± standard deviation (SD). Statistical analysis was performed using GraphPad Prism 7 (GraphPad, San Diego, CA, USA). The significance of differences between two groups was analyzed by Student's t-test and multiple group comparisons were analyzed with one-way ANOVA. *P < 0.05, **P < 0.01, and ***P < 0.001 were considered significant.

Hypoxia Induced a Reduction in Chemosensitivity to Cisplatin and Upregulation of HOXA-AS3 in BC Cells
Hypoxia has been found to induce chemotherapy resistance in various tumors and cancer cells (25,26). Accordingly, the present findings from CCK-8 analysis indicated that hypoxia could induce resistance of BC cells to cisplatin, as indicated by the IC50 values in the histogram in Figure 1A. Next, we used qRT-PCR to detect the level of a series of lncRNAs after cisplatin treatment under hypoxic or normoxic conditions. We found that HOXA-AS3 expression was significantly increased under hypoxic conditions in BC cells ( Figure 1B). Using gene expression profiling interactive analysis (GEPIA), we analyzed  HOXA-AS3 expression across all tumor samples and paired normal tissues and found that HOXA-AS3 was significantly upregulated in bladder tumor tissues compared with normal tissues ( Figure 1C). We also analyzed the clinical analysis of the relationship between HOXA-AS3 expression and the clinicopathological parameters in bladder cancer, showing that high HOXA-AS3 expression in 30 bladder cancer patients was closely related to large tumor size (P=0.0123), the advanced TNM stage (P=0.0173), and invasion(muscle) (P=0.0352) ( Table  2). Furthermore, qRT-PCR confirmed that the level of HOXA-AS3 in cancer tissue was higher than that in adjacent tissue ( Figure 1D). We also confirmed that HOXA-AS3 expression was gradually upregulated with increasing cisplatin concentrations using qRT-PCR analysis (Supplemental Figures 1A). Therefore, hypoxia upregulated HOXA-AS3 and appears to reduce chemosensitivity to cisplatin in BC cells.

Downregulation of HOXA-AS3 Enhances Cisplatin Sensitivity Under Normoxic and Hypoxic Conditions
We showed BIU-87 had the lowest HOXA-AS3 expression among all three human bladder cancer cell lines tested ( Figure  2A). We also showed BIU-87 was more sensitive to cisplatin than J82 and UM-UC-3 cells, indicating the level of HOSA-AS3 was negatively correlated to cisplatin sensitivity in BC cells ( Figure  2A). Next, we examined the effect of HOXA-AS3 on cisplatin sensitivity and its potential mechanism of action in BC cells.
Using a CCK-8 cell viability assay, we found all three BC cell lines that were transfected with HOXA-AS3 siRNA had increased cisplatin sensitivity compared with the negative controls ( Figures 2B-E), while the HOXA-AS3 plasmid had reduced cisplatin sensitivity (Supplemental Figures 2A-D). These results were confirmed by EdU staining analysis of cell proliferation ( Figures 2F, G). qRT-PCR confirmed the expression of HOXA-AS3 following transfection with or without HOXA-AS3 siRNA ( Figure 2E). Finally, as cisplatin-induced refractoriness to apoptosis is an important characteristic of chemoresistance, we examined the effect of HOXA-AS3 on apoptosis using flow cytometry. As shown in Figure 2H, combining HOXA-AS3 siRNA and cisplatin treatment significantly increased the rate of apoptosis of BC cells. Furthermore, knockdown of HOSX-AS3 could reverse hypoxia-induced cisplatin resistance ( Figure  2I). Together, these findings indicate that HOX-AS3 may regulate cisplatin sensitivity in BC cells under normoxic and hypoxic conditions.

HOXA-AS3 Regulates Notch1 Expression and Reverses Hypoxia-Induced EMT in BC Cells
Increasing evidence shows EMT plays an important role in regulating proliferation, migration, and chemoresistance in cancer cells (35,36). Recent studies have reported that HOX-AS3 is closely related to cisplatin resistance and EMT. Therefore, we examined levels of EMT-related proteins in BC cells transfected with either HOXA-AS3 or HOXA-AS3 siRNA. Using western blot, we found that HOXA-AS3 siRNA treatment upregulated the expression of E-cadherin and downregulated Notch1 and vimentin expression in BC cells. Furthermore, transfection with HOXA-AS3 siRNA reversed the effects of hypoxia by upregulating E-cadherin expression and downregulating Notch1 and vimentin expression (Figures 3A, B). Immunofluorescence analysis of E-cadherin and vimentin expression was consistent with these results (Figures 3C, D). These findings indicate that inhibition of HOXA-AS3 could reverse hypoxia-induced EMT in BC cells.

HOXA-AS3 Targets miR-455-5p
Recent studies suggest that lncRNAs could act as a sponge or decoy and compete with other genes for miRNA binding, and thereby, reduce the regulatory effect of miRNAs on their target mRNAs (11). In order to determine whether HOXA-AS3 brings about its effects by interacting with miRNA, we examined the effect of cisplatin on all the miRNAs deposited in the RegRNA2.0 database (http://regrna2.mbc.nctu.edu.tw/). We examined the level of all the miRNAs after treatment with or without cisplatin in BC cells and found that miR-455-5p was significantly downregulated after cisplatin treatment (Supplemental Figure 3A). This indicates that HOXA-AS3 may interact with miR-455-5p via complementary base pairing. Therefore, we constructed two versions of HOXA-AS3, that is, WT-HOXA-AS3 and Mut-HOXA-AS3, and used a luciferase reporter assay to verify the interaction between HOXA-AS3 and miR-455-5p. The 3′UTR residues predicted to interact with miR-455-5-p were mutated in Mut-HOXA-AS3 ( Figure 4A). We found that the addition of an miR-455-5p mimic considerably reduced the luciferase activity of WT-HOXA-AS3, while the addition of an miR-455-5p inhibitor increased its activity ( Figure 4B). However, neither the miR-455-5p mimic nor the miR-455-5p inhibitor had any impact on the luciferase activity of Mut-HOXA-AS3 ( Figure 4B).  Furthermore, miR-455-5p expression was significantly upregulated following transfection with HOXA-AS3 siRNA, while it was downregulated after transfection with the HOA-AS3 plasmid in BC cells compared with the negative control ( Figures 4C, D). In addition, compared with normoxic conditions, hypoxia significantly downregulated the level of miR-455-5p and upregulated the expression of HOXA-AS3 ( Figures 4E, F). We also found HOXA-AS3 levels were significantly downregulated in BC cells transfected with a miR-455-5p mimic but and upregulated by the miR-455-5p inhibitor ( Figure 4G). Moreover, the expression of Notch1 was downregulated by the miR-455-5p mimic, while it was upregulated by the inhibitor (Figure 4H). Luciferase reporter assay also confirmed the interaction between miR-455-5p and Notch1 (Figures 4I, J). The binding of miR-455-5p to HOXA-AS3 was shown to lead to a decrease in HOXA-AS3 transcript levels and a concomitant increase in the translation of the HOXA-AS3-associated mRNA Notch1. The data indicate that HOXA-AS3 may function as a competing endogenous RNA (ceRNA) of miR-455-5p to regulate Notch1 in BC, which in turn, may affect cisplatin sensitivity.

miR-455-5p Mediates the Regulatory Effect of HOXA-AS3 on Cisplatin Sensitivity
Next, we explored whether miR-455-5p is involved in HOXA-AS3-mediated reduction in cisplatin sensitivity in BC cells. Using a CCK-8 (cell viability) assay, we showed miR-455-5p inhibition reduced the sensitivity of BC cells to the effects of cisplatin ( Figures 5A-C). Meanwhile, HOXA-AS3 siRNA enhanced the sensitivity of BC cells to cisplatin, but these positive effects disappeared following the addition of the miR-455-5p inhibitor ( Figures 5A-C). These results were consistent with those of EdU analysis ( Figures 5D, E). We also found that the miR-455-5p inhibitor reduced the apoptosis rate of BC cells after transfection with HOXA-AS3 siRNA along with cisplatin treatment ( Figures  5F, G). Therefore, both miR-455-5p and HOXA-AS3 are involved in modulating the cisplatin sensitivity of BC cells.

Suppression of HOXA-AS3 Enhances Cisplatin Sensitivity of BC Cells In Vivo
To explore the effect of HOXA-AS3 on the sensitivity to cisplatin in vivo, we used UMUC-3 cells to establish a tumor xenograft 12 after tumor cell injection was significantly lower in the cisplatin-treated group than in the other three groups; however, the body weight was higher in mice treated with HOXA-AS3 siRNA plus cisplatin than in those treated with cisplatin alone ( Figure 7B). We also showed that treatment with HOXA-AS3 siRNA plus cisplatin downregulated the cisplatin-induced elevation of Notch1 expression ( Figure 7F). In addition, treatment with HOXA-AS3 siRNA plus cisplatin significantly decreased tumor cell proliferation and increased tumor cell apoptosis compared to the other three groups (Figures 7G, H). Furthermore, treatment with HOXA-AS3 siRNA plus cisplatin significantly downregulated the level of HOXA-AS3, Notch1, and miR-455-5p, as revealed by qRT-PCR (Figures 7I-L). Notch1 protein expression was also reduced following treatment with HOXA-AS3 siRNA plus cisplatin, compared with the cisplatin group, as revealed by western blot analysis ( Figure 7K). These data support our findings from the in vitro studies and, thus, confirm the in vivo function of HOXA-AS3.

DISCUSSION
HOXA-AS3 was previously shown to function as an oncogene and was upregulated in various cancers including glioma tissues, NSCLC, and HCC (21,31,34). Similarly, we found HOXA-AS3 was upregulated in BC tumor tissues and BC cell lines. The high HOXA-AS3 expression in 30 bladder cancer patients was closely related to large tumor size, the advanced TNM stage and invasion(muscle). Moreover, the expression of HOXA-AS3 in BC cells was correlated with the sensitivity to cisplatin, indicating that HOXA-AS3 might be a useful biomarker to predict sensitivity to cisplatin in BC. Our in vitro and in vivo experiments confirmed HOXA-AS3 inhibition could reduce BC cell viability and tumor growth. We also clarified the role of HOXA-AS3 in mediating cisplatin resistance in BC cells. Overall, our findings indicate that targeting HOXA-AS3 could be useful for optimizing cisplatin treatment in BC. Extensive evidence has revealed that lncRNAs could serve as ceRNAs or molecular sponges to directly interact with and negatively regulate miRNAs, thus modulating the expression of specific genes targeted by miRNAs (11,37). For instance, overexpression of the lncRNA UBE2R2-AS1 promoted glioma cell apoptosis via targeting the miR-877-3p/TLR4 axis (38). In addition, suppression of the lncRNA MALAT1 enhanced cisplatin sensitivity via regulating the miR-101-3p/VEGF-C pathway in BC (20). Similarly, we showed miR-455-5p was a direct target of HOXA-AS3: HOXA-AS3 siRNA transfection increased miR-455-5p levels, while HOXA-AS3 transfection decreased them. Indeed, a growing number of studies have demonstrated miRNAs participate in regulating drug sensitivity in several cancers, including BC. Therefore, we explored the effects of miR-455-5p on cisplatin resistance in BC.
Previous studies have shown miR-455-5p can function as either a tumor promoter or suppressor in various cancers. For example, miR-455-5p was downregulated in colorectal carcinoma and gastric cancer, but was upregulated in NSCLC (39)(40)(41). We confirmed that miR-455-5p inhibition could decrease the sensitivity of BC cells to cisplatin, as well as promote BC cell proliferation and reduce apoptosis, indicating miR-455-5p acts as a tumor suppressor in BC. We also found that combining an miR-455-5p inhibitor with HOXA-AS3 siRNA alleviated the effect of HOXA-AS3 siRNA on cisplatin sensitivity. These findings indicate HOXA-AS3 regulates cisplatin sensitivity via regulation of miR-455-5p in BC.
Not only is EMT is strongly correlated with drug resistance, the emergence of drug resistance may also occur as a result of EMT (42,43). Emerging evidence indicates that Notch signaling plays an important role in cell proliferation and apoptosis, which are involved in the development and functioning of many organs (44). Activation of the Notch1 pathway has been commonly observed in many human malignancies including PC (45). Downregulation of the lncRNA XIST in NSCLC cells could suppress cell proliferation and TGF-b1induced EMT through activation of the Notch1 pathway via regulation of miR-137 (46). In addition, recent studies show HOXA-AS3 may regulate cisplatin resistance and metastasis in NSCLC and HCC cells via modulating EMT (31,34). Similarly, we showed that HOXA-AS3 siRNA enhanced cisplatin sensitivity in BC via modulating EMT, and confirmed that Notch1 was mediating these effects.
In conclusion, HOAX-AS3 inhibition may suppress tumor progression and enhance sensitivity to cisplatin in BC via modulation of the miR-455-5p-Notch1 axis. As the HOXA-AS3-miR-455-5p-Notch1 network plays a central role in cisplatin resistance, targeting HOXA-AS3 may help optimize the efficacy of cisplatin treatment for BC in the future.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.