LncRNA ZNF503-AS1 acts as a tumor suppressor in bladder cancer by up-regulating Ca²⁺ concentration via transcription factor GATA6

Abstract

Purpose

Ca²⁺ homeostasis plays a pivotal role in regulating proliferation and apoptosis during cancer development. This study intended to examine the potential tumor-suppressing role of ZNF503 antisense RNA 1 (ZNF503-AS1) in bladder cancer, which may be implicated in the regulation of Ca²⁺ homeostasis.

Methods

Differentially expressed long non-coding RNAs (lncRNAs) related to bladder cancer were identified using microarray analysis, followed by the verification of transcription factors to which they bind. The relationship between ZNF503-AS1, GATA6 and SLC8A1 was assessed using dual luciferase reporter, RIP and ChIP assays. The expression levels of ZNF503-AS1, GATA6 and SLC8A1 were modulated to examine their effects on the tumorigenic potential, intracellular Ca²⁺ concentration and Ca²⁺-ATPase activity in bladder cancer cells. The in vivo tumorigenic ability was validated in nude mice.

Results

Microarray-based expression profile analysis of the GEO GSE61615 dataset revealed that the expression of ZNF503-AS1 was decreased in bladder cancer. Subsequently, we found that ZNF503-AS1 can bind to the transcription factor GATA6 to up-regulate the expression of SLC8A1. ZNF503-AS1 and SLC8A1 were found to be down-regulated in both primary bladder cancer tissues and cells. Exogenous overexpression of ZNF503-AS1 or SLC8A1 attenuated bladder cancer cell proliferation, invasion and migration, but promoted their apoptosis, accompanied by decreased Ca²⁺-ATPase activities and increased intracellular Ca²⁺ concentrations. Additional in vivo experiments validated the inhibitory effect of ZNF503-AS1 overexpression on the tumorigenic capacity of bladder cancer cells in nude mice.

Conclusion

ZNF503-AS1 can recruit transcription factor GATA6 to up-regulate SLC8A1 expression, thereby increasing the intracellular Ca²⁺ concentration and repressing the proliferation, invasion and migration, and enhancing the apoptosis of bladder cancer cells.

Electronic supplementary material

The online version of this article (10.1007/s13402-020-00563-z) contains supplementary material, which is available to authorized users.

Introduction

Bladder cancer is regarded as a heterogeneous disease, with up to 30% of the cases exhibiting muscle-invasive characteristics accompanied by a high risk of distant metastasis, while 70% of the cases present with superficial tumors vulnerable to recurrence [1]. Statistic data indicate bladder cancer as the 9th most frequently occurring cancer in all individuals, the 7th in men across the world, and the 13th cause of death [2]. Evidence indicates that calcium ion (Ca²⁺) signaling may be remodeled in cancer cells to sustain proliferation and decrease cell death, suggesting a potential for therapeutic intervention [3]. Bee venom treatment has, for example, been proven to trigger apoptosis via the death pathway mediated by intracellular Ca²⁺ concentration alterations in bladder cancer cells [4]. Tobacco- and environmental-gene interactions have also been identified as risk factors contributing to the occurrence and development of bladder cancer, which has led to on-going studies to examine the functional relevance of differentially expressed genes in bladder cancer [5]. Recent progresses in genomic and genetic understanding, combined with urologic cancer-specific features, have led to improvements in treatment outcome in patients with urologic cancer, including bladder cancer [6, 7]. Accordingly, the current study aims to identify promising biomarkers or molecular pathways that may facilitate the development of tailored therapies for bladder cancer.

Recent studies have implicated long non-coding RNAs (lncRNAs) and their functional mechanisms in modulating the pathogenesis and progression of bladder cancer [8–10]. LncRNAs represent a heterogeneous group of functional molecules >200 nucleotides long, presenting with little or no coding potential [11]. Of note, it has been found that aberrantly expressed lncRNAs may affect cellular behavior and, ultimately, prevent or stimulate carcinogenesis [12, 13]. In the current study, we found that the expression of ZNF503 antisense RNA 1 (ZNF503-AS1) was decreased in bladder cancer, and predicted to bind to the transcription factor GATA binding protein 6 (GATA6) to mediate the expression of solute carrier family 8 member A1 (SLC8A1). ZNF503-AS1 has previously been reported to be negatively correlated with the expression of zinc finger protein 503 (ZNF503), as ZNF503-AS1 is transcribed from the antisense strand of the ZNF503 gene locus [14]. In addition, it has been found that ZNF503/Zpo2 stimulates the growth of aggressive breast cancer cells by down-regulating the expression of transcription factor GATA3 [15]. A prior genomic profiling study also revealed that GATA6 over-expression may enhance the oncogenic potential of pancreatic cancer cells, suggesting that GATA6 may act as an oncogene and, thus, may be employed for the design of novel therapeutic strategies [16]. Furthermore, the SLC8A1 gene, encoding a Na⁺/Ca²⁺ exchanger, has been found to play a pivotal role in calcium homeostasis, and its dysregulation to be linked to apoptosis and proliferation in penile cancer through Ca²⁺ concentration alteration [17]. Therefore, we set out to explore the functional relevance of the ZNF503-AS1/GATA6/SLC8A1 axis and its involvement in Ca²⁺ concentration alterations in bladder cancer.

Materials and methods

Ethical statement

The current study was conducted under approval of the Ethics Committee of The Second Xiangya Hospital, Central South University. Signed informed consents were obtained from all participants prior to enrollment. Nude mice were employed for in vivo studies and cared for in accordance with the principles of the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health, with all efforts made to ensure minimal suffering of the animals used in the study.

Microarray-based analysis

Firstly, the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo) was used to download the bladder cancer-related microarray data (GSE61615) and annotation probe files, which were obtained from the Agilent-028004 SurePrint G3 Human GE 8x60K Microarray (Probe Name Version). Background correction and normalization of the microarray data were performed using the Affy installation package of the R software [18]. Next, the linear model-experience Bayesian statistical method of the Limma installation package combined with the traditional t-test was employed for non-specific filtering of the expression profile data. Multiple-testing correction was performed using the Benjamini-Hochberg false discovery rate (FDR). LncRNAs and mRNAs presenting with a FDR < 0.05 and an absolute fold-change ≥1.5 were considered to be statistically significant [19].

Patient enrollment and cell lines

A total of 86 patients with bladder cancer referred to The Second Xiangya Hospital, Central South University from September 2016 to November 2017 were enrolled in the current study. All patients were pathologically diagnosed with bladder cancer and, subsequently, bladder cancer tissues and adjacent normal tissues were surgically obtained and immediately frozen in liquid nitrogen and stored at −80 °C for later use. None of the patients underwent preoperative chemotherapy or radiotherapy. Additionally, human bladder cancer cell lines 5673, J82, H/RB-CL2 and H/RB-M, and human embryonic bladder tissue-derived cells CCC-HB-2 were purchased from the National Infrastructure of Cell Line Resource (http://www.cellresource.cn/index.aspx). All cell lines were cultured in a humidified incubator with 5% CO₂ in air at 37 °C with Roswell Park Memorial Institute (RPMI)-1640 medium (72,400,120, Thermo Fisher, USA) containing 10% fetal bovine serum (FBS), 100 μg/ml streptomycin, and 100 IU/ml penicillin. Upon reaching 90% confluence, the cells were passaged using 0.25% trypsin. After the digestion was terminated, the cells were pipetted into a uniform single cell suspension. The expression levels of ZNF503-AS1 in the four human bladder cancer cell lines were determined by qRT-PCR, after which the cell line exhibiting the lowest ZNF503-AS1 expression was selected for subsequent experiments.

Cell treatment assays

The sequence of ZNF503-AS1 was obtained from the National Center for Biotechnology Information (NCBI) database. Subsequently, an over-expression (oe)-ZNF503-AS1 sequence (Sigma, USA) was ligated into a PLKO-Puro vector (Sigma, USA). After sequence verification, the plasmid was co-transfected with psPAX2 and pMD2.G (Addgen, USA) into human embryonic kidney 293 (HEK293) cells, after which infected cells containing lentiviral particles were collected and used for the establishment of stable cell lines. Next, cells were transfected with various plasmids including a short hairpin RNA negative control (sh-NC), sh-ZNF503-AS1, oe-NC and oe-ZNF503-AS1, or plasmids as oe-NC, oe-SLC8A1, sh-ZNF503-AS1 + oe-GATA6 + oe -SLC8A1, and sh-ZNF503-AS1 + oe-GATA6 + oe-NC. All the aforementioned plasmids were purchased from Dharmacon (Lafayette, CO, USA). Bladder cancer cells were seeded in a 6-well plate at a density of 3 × 10⁵ cells/well, and transfected according to the protocols supplied with the Lipofectamin 2000 kit (Invitrogen, USA) when the cells reached 90% confluence. Next, 4 μg of the target plasmid and 10 μl Lipofectamin 2000 were diluted in 250 μl Opti-minimum essential medium (MEM) (Gibco, USA). The mixture was kept at room temperature for 20 min, and then added dropwise to the cell culture wells, followed by incubation in 5% CO₂ in air at 37 °C. The medium was renewed after 6 h. Finally, the cells were harvested 36–48 h after transfection.

RNA extraction and qRT-PCR

Total RNA was extracted from tissues or cells using TRIzol (15,596,026, Invitrogen, USA), and the obtained RNA was reverse transcribed into complementary DNA (cDNA) using a reverse transcription kit (RR047A, Takara, Japan). Next, the samples were loaded using a SYBR Premix EX Taq kit (RR420A, Takara), and subjected to qRT-PCR on an ABI 7500 instrument (Applied Biosystems, Foster City, CA, USA). Three replicate wells were set for each sample. The primers used were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China) (Table 1). mRNA expression levels were determined by qRT-PCR and normalized to U6 levels [20].

Table 1.

Primer sequences for qRT-PCR

Target	Sequence
ZNF503-AS1	Forward: 5’-CATTCTCCACCCTGCCACAT-3’
	Reverse: 5’-GCCCTCACTTGGAATCCTCC-3’
SLC8A1	Forward: 5’-ACAACATGCGGCGATTAAGTC-3’
	Reverse: 5’-GCTCTAGCAATTTTGTCCCCA-3’
MMP-2	Forward: 5’-TACAGGATCATTGGCTACACACC-3’
	Reverse: 5’-GGTCACATCGCTCCAGACT-3’
MMP-9	Forward: 5’-TGTACCGCTATGGTTACACTCG-3’
	Reverse: 5’-GGCAGGGACAGTTGCTTCT-3’
Ki67	Forward: 5’-ACTTGCCTCCTAATACGCC-3’
	Reverse: 5’-TTACTACATCTGCCCATGA-3’
Bax	Forward: 5’-CCCGAGAGGTCTTTTTCCGAG-3’
	Reverse: 5’-CCAGCCCATGATGGTTCTGAT-3’
β-actin	Forward: 5’-AGGTCATCACTATTGGCAAC-3’
	Reverse: 5’-ACTCATCGTACTCCTGCTTG-3’

Western blot analysis

Total protein was extracted from tissues or cells using radioimmunoprecipitation assay (RIPA) lysis buffer containing phenylmethylsulphonyl fluoride (PMSF), followed by incubation on ice for 30 min. Subsequently, the lysates were centrifuged at 8000 g for 10 min at 4 °C and the supernatants collected. Next, each sample was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes. The membranes were blocked in 5% skimmed milk powder at room temperature for 1 h, followed by overnight incubation at 4 °C with primary antibodies (Abcam, UK) directed against SLC8A1 (ab177952, dilution ratio of 1:1000), matrix metalloproteinase (MMP)-2 (ab37150, dilution ratio of 1:1000), MMP-9 (ab73734, dilution ratio of 1:10000), Bcl-2-associated X protein (Bax) (ab32503, dilution ratio of 1:1000), Ki67 (ab32124, dilution ratio of 1:1000) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (ab9485, dilution ratio of 1:2500). Next, the membranes were incubated with a horseradish peroxidase (HRP)-labeled secondary antibody for 1 h, and rinsed with Tris-buffered saline Tween-20 (TBST). Enhanced chemiluminescence (ECL) (Cat. No. BB-3501, Amersham, UK) was employed to visualize the proteins, which were photographed using a Bio-Rad Image Analysis System (BIO-RAD, USA), and quantified using Quantity One v4.6.2 software. Relative protein levels were expressed as the ratio of gray value of the corresponding protein band to that of the GAPDH protein band. The experiments were repeated three times, with the average values calculated.

RNA fluorescence in situ hybridization (FISH) assay

The subcellular localization of ZNF503-AS1 in bladder cancer cells was determined by FISH according to the instructions of Ribo™ lncRNA FISH Probe Mix (Red) (RiboBio Co., Ltd., Guangzhou, Guangdong, China) [21]. Next, the cells were evaluated and photographed in 5 randomly selected fields under a fluorescent microscope (Olympus, Japan).

RNA binding protein immunoprecipitation (RIP) assay

The binding of ZNF503-AS1 to the GATA6 protein was evaluated using a RIP kit (Millipore Corp., Bedford, MA, USA). Briefly, the cells were lysed with RIPA and centrifuged, with the supernatant discarded. Part of the cell extract was taken as input, and the remaining part was co-precipitated through incubation with an anti-GATA6 antibody (ab22600, dilution ratio of 1:1000, Abcam, UK), which was allowed to stand at room temperature for 30 min. Immunoglobulin G (IgG) (ab109489, dilution ratio of 1:100, Abcam, UK) was employed as negative control (NC). In short, in each co-precipitation reaction, 50 μl magnetic beads were re-vortexed with 100 μl RIP Wash Buffer. Next, 5 μg antibody was added according to the experimental grouping for binding. The magnetic bead-antibody complexes were then washed and resuspended in 900 μl RIP Wash Buffer, followed by incubation with 100 μl cell extract overnight at 4 °C. Next, the samples were placed on a magnetic stand to collect the magnetic bead-protein complexes. The samples and input were separately digested with proteinase K to extract the RNA for subsequent PCR detection.

Dual luciferase reporter assay

The UCSC (http://genome.ucsc.edu/) and JASPAR (http://jaspar.genereg.net/) websites were employed to detect putative binding of the GATA6 protein to the promoter region of SLC8A1. Next, SLC8A1 recombinant luciferase reporter vectors with truncated or mutated binding sites were constructed and co-transfected with the GATA6 expression vector into H/RB-CL2 cells to assess specific binding of the GATA6 protein to SLC8A1 DNA. In addition, the target site sequence (WT) of the 3′-untranslated region (3’-UTR) of SLC8A1 mRNA and site-directed mutants of the WT target site (MUT1, MUT2, MUT1/2) were synthesized. The WT and MUT1, MUT2 and MUT1/2 target gene fragments were inserted into a pGL3 vector and, after sequence verification, used for subsequent transfections. Vectors containing MUT and WT fragments were co-transfected with oe-NC or oe-GATA6 into H/RB-CL2 cells, respectively. The resulting cells were harvested and lysed after a 48 h transfection period. Subsequent experiments were performed using a luciferase assay kit (K801–200, Biovision, Bay Area, San Francisco, CA, USA) in conjunction with an analysis system (Promega Corporation, Madison, WI, USA). Using Renilla luciferase as internal reference, the luciferase activity of the target reporter gene was expressed as the ratio of the relative luciferase activity of Firefly luciferase to that of Renilla luciferase.

Chromatin immunoprecipitation (ChIP) assay

Cells were fixed with formaldehyde for 10 min to yield DNA-protein cross-linking, and then sonicated to break the chromatin into fragments, followed by centrifugation to collect the supernatant. Next, negative control rabbit IgG (ab109489, dilution ratio of 1:300, Abcam, Shanghai, China) and an anti-GATA6 antibody (sc-137065, dilution ratio of 1:1000, Santa Cruz Biotechnology, Shanghai, China) were added to the supernatant for incubation overnight at 4 °C. Next, DNA-protein complexes were precipitated with Protein Agarose/Sepharose, followed by centrifugation at 12000 g for 5 min, with the supernatant discarded. Non-specific complexes were washed, followed by overnight cross-linking at 65 °C. Next, the DNA fragments were extracted, purified and recovered using phenol/chloroform. Primers were designed that could amplify GATA6 binding site 1 within the SLC8A1 promoter (F: 5’-TGACAGGATCATGTGGTGGGAAATGGGC-3′, R: 5’-CCAGAAAGAAAGGACCTGGGATGCTTGT-3′). The amplified target sequence was 247 bp long, and located 66 bp upstream of the transcription start site (TSS). In addition, a distal primer (a primer that amplifies a sequence away from the SLC8A1 DNA promoter region) was designed as a negative control (NC) for the site 1 primer (F: 5’-GGTAAACACAAACAGACTCACAGACAC-3′, R: 5’-ATCCATCAGGAGAAGAAAGGAAAAATA-3′). The amplified product was 333 bp long and located 7587 bp from the TSS. With the recovered DNA fragments used as templates, site 1 primers and distal primers were added, respectively, to perform PCR to verify whether site 1 of SLC8A1 DNA was the binding site of GATA6.

RNA pull-down assay

H/RB-CL2 cells were transfected with 50 nM bio-ZNF503-AS1. After 48 h, the cells were collected, rinsed with phosphate-buffered saline (PBS) and suspended. Next, the cells were incubated with lysis buffer (Ambion, Company, Austin, TX, USA) for 10 min, after which 60 ml of the cell lysate was divided into aliquots. The residual lysate was then incubated with M-280 streptavidin magnetic beads (Sigma, St. Louis, MO, USA) pre-coated with RNase-free yeast tRNA (purchased from Sigma, St. Louis, MO, USA) for 3 h at 4 °C. Next, the lysate was washed twice with cold lysis buffer, thrice with low-salt buffer and once with high-salt buffer. Total protein was finally extracted using RIPA lysis buffer, and GATA6 expression patterns were detected using Western blotting.

Immunohistochemistry

Bladder cancer tissues and adjacent normal tissues were fixed in paraformaldehyde for 48–72 h, paraffin-embedded and sliced into 5-μm-thick sections. The tissue sections were then dewaxed with xylene and dehydrated using an alcohol gradient. After treatment in 0.01 M citrate solution and PBS equilibration, the sections were treated with 3% hydrogen peroxide to block endogenous peroxidase activity. Next, the sections were blocked with goat serum at 37 °C for 20 min and, subsequently, probed with rabbit anti-SLC8A1 (ab135735, dilution ratio of 1:100, Abcam, UK) overnight at 4 °C, and with PBS as negative control for the primary antibody. After rewarming with primary antibody for 1 h, the sections were re-probed with a secondary goat anti-rabbit IgG (ab150117, dilution ratio of 1:1000, Abcam, UK) at 37 °C for 30 min, followed by incubation with a streptavidin biotin peroxidase complex (SABC) at 37 °C for 30 min. The proteins were then visualized using 3, 3′-diaminobenzidine (DAB) under a microscope. Next, the sections were subjected to hematoxylin staining, washing in water, decolorization in 1% hydrochloric acid alcohol, washing under running water, dehydration, saturated aluminum carbonate staining and permeabilization in xylene.

5-ethynyl-2′-deoxyuridine (EdU) assay

Poly-L-lysine-coated coverslips were allowed to stand overnight in a 24-well plate for later use. Bladder cancer cells at passage 3 were seeded at a density of l × 10⁵ cells/ml/well. EdU solution (C00052, Ribio, China) was diluted in cell culture medium at a ratio of 1000:1 to prepare 50 μM EdU medium. Next, 100 μl of 50 μM EdU medium was added to each well for 2 h, after which the medium was discarded. After further incubation with 2 mg/ml glycine (Sigma, USA) for 10 min at room temperature, the cells were permeabilized with 0.5% Triton X-100 (100 μl per well, Sigma, USA). Apollo staining solution (Beyotime, Jiangsu, China) was used to incubate the cells at room temperature for 30 min, followed by nucleus staining with 4′,6-diamidino-2-phenylindole (DAPI) (Sigma, USA) for 10 min. The cells were finally evaluated and photographed under an inverted fluorescence microscope to calculate the rate of EdU positive cells/DAPI positive cells.

Transwell migration and invasion assays

Matrigel (40111ES08, Yeason, Shanghai, China) diluted in previously pre-cooled serum-free Dulbecco’s modified Eagle’s medium (DMEM) (Matrigel: DMEM = 1:8) was placed in the upper Transwell culture chambers (3413, Unique Biotechnology Co. Ltd., Beijing, China); this step was omitted in the migration assay. The chambers were then incubated at 37 °C for 4–5 h for solidification. Next, the transfected cells were diluted in 100 μl serum-free medium to prepare a cell suspension (density 1 × 10⁶ cells/ml). After cell seeding, the lower chambers were supplemented with 500 μl DMEM containing 20% FBS. Three replicate wells were set for each group. After incubation for 24 h with 5% CO₂ in air at 37 °C, the Transwell chambers were removed and fixed with 5% glutaraldehyde. The cells were then stained with 0.1% crystal violet for 5 min at 4 °C. Cells on the upper surface were wiped off with cotton balls. Under an inverted fluorescence microscope (TE2000, Nikon, China) 5 fields were randomly selected for evaluation and photography. The average numbers of cells passing through the membranes were calculated from three independent experiments.

Flow cytometry

On the 2nd day after transfection, the cells were digested with 0.25% trypsin. The digestion was terminated with RPMI-1640 medium containing 10% FBS, followed by centrifugation for 5 min at 1000 rpm, with the supernatant discarded. Next, the cells were fixed with pre-cooled 70% ethanol at 4 °C, and the cell density was adjusted to 1 × 10⁶ cells/ml. Annexin V-FITC/PI (556,547, Shanghai Shuojia Biotechnology Co., Ltd., China) was applied in a refrigerator at 4 °C for 15–30 min, after which cell apoptosis was determined by flow cytometry (Beckman Coulter, Fullerton, CA, USA) at an excitation wavelength of 480 nm. Apoptosis rates were expressed as percentage of the number of apoptotic cells in the total number of cells.

Intracellular Ca²⁺ concentration determination

Cells were centrifuged and the supernatants were removed such that about 50 μl was left. Next, the cells were resuspended by adding 1 ml 5 μmol/L Fluo-3 AM Ca²⁺ fluorescent probe (AAT-21010, AAT Bioquest, USA), followed by incubation at 37 °C for 30 min. Next, supernatants were discarded after centrifugation, and the cells were resuspended in 1 ml serum-free RPMI-1640 medium. The intracellular Ca²⁺ concentrations were measured by flow cytometry over time. The experiments were conducted in triplicate to obtain average values.

Ca²⁺-ATPase activity determination

Cell culture supernatants (0.1 ml) were extracted from each group and incubated with a reaction solution (containing 30 nM imidazole, 80 nM NaCl, 15 mM KCl, 3 mM MgCl₂, 0.1 mM ouabain, 4 mM ATP, 0.1 mM CaCl₂, Ph = 7.1, total volume of 1 ml) for 1 h at 37 °C. The reaction was terminated with 1 ml 10% trichloroacetic acid. Next, the mixture was centrifuged to collect 0.5 ml of the supernatant to determine Ca²⁺-ATPase activity using a Ca²⁺-ATPase activity kit (MAK113-1KT, Sigma, USA). Enzyme activity was expressed as the amount of inorganic phosphorus produced per gram of tissue protein per hour, i.e., μmolpi h⁻¹/mg pro.

In vivo tumor formation assay

A total of 36 specific pathogen-free (SPF) BALB/C male nude mice (aged 3–5 weeks, weighing 16–22 g) were purchased from Hunan SLAC Laboratory Animals Co., Ltd. (Changsha, Hunan, China). The mice were randomly divided into 3 groups (n = 12/group) for tumor growth assessment. Next, H/RB-CL2 cells were infected with lentiviral particles (Shanghai GenePharma Co., Ltd., Shanghai, China) of LV-oe-NC, LV-sh-NC, LV-oe-ZNF503-AS1 and LV-sh-SLC8A1, and stably-infected cells were selected. Next, the stably-infected cells from the three groups were respectively resuspended in 50% Matrigel (BD Biosciences, Bedford, MA), with the concentration adjusted to 5 × 10⁶ cells/ml. Next, 0.2 ml single cell suspension (containing 1 × 10⁶ cells) was injected subcutaneously into the left axilla of each nude mouse. After 8 days of injection, the tumor sizes were examined every 3 days using a Vernier caliper. The tumor volumes (mm³) were calculated using the following formula: tumor volume = length × width² × 0.5. Thirty days after tumor cell injection, all mice were euthanized and the tumors were removed and weighed.

Statistical analysis

All data were processed using SPSS 21.0 statistical software (IBM Corp, Armonk, NY, USA). Measurement data are presented as mean ± standard deviation. Data comparisons between bladder cancer tissues and adjacent normal tissues were performed using the paired t-test, and those between the other two groups using the unpaired t-test. Correlation between data of two groups was evaluated using Pearson’s correlation coefficient. Comparisons among multiple groups were performed using one-way analysis of variance (ANOVA) with Tukey’s post-hoc test. Time-based comparisons among multiple groups were conducted by repeated measures ANOVA with Tukey’s post-hoc test. In all statistical analyses, a p-value < 0.05 was regarded statistically significant.

Results

ZNF503-AS1 expression is down-regulated in bladder cancer tissues and cells

Analysis of the GSE61615 dataset revealed that the expression of ZNF503-AS1 was low in bladder cancer tissues (Fig. 1a). In addition, the expression of ZNF503-AS1 was found to be significantly down-regulated in bladder cancer tissues compared to adjacent normal tissues (p < 0.05) (Fig. 1b). Furthermore, all 4 bladder cancer cell lines tested (5637, J82, H/RB-CL2 and H/RB-M) exhibited decreased ZNF503-AS1 expression levels compared to human embryonic bladder cells (CCC-HB-2), with the H/RB-CL2 cell line exhibiting the lowest ZNF503-AS1 expression (p < 0.05) (Fig. 1c). Therefore, this cell line was selected for subsequent experiments. The J82 cell line also showed a reduced ZNF503-AS1 expression, and was used to verify the reliability of key data. ZNF503-AS1 expression analysis by RNA-FISH (Fig. 1d) revealed that it was primarily located in the nucleus in bladder cancer cells. These data indicate that the expression of ZNF503-AS1 is down-regulated in bladder cancer.

Fig. 1.

ZNF503-AS1 is lowly expressed in bladder cancer. a: heat map of the top 20 differentially expressed genes in GSE61615. b: qRT-PCR expression analysis of ZNF503-AS1 in bladder cancer tissues and adjacent normal tissues (n = 86). c: qRT-PCR expression analysis of ZNF503-AS1 in 4 bladder cancer cell lines and human embryonic bladder tissue-derived cells (CCC-HB-2). d: RNA-FISH determining the localization of ZNF503-AS1 in primary patient samples (× 400, scale bar = 25 μm). Data are presented as mean ± standard deviation. Comparisons between bladder cancer tissues and adjacent normal tissues were performed by paired t test. Comparisons among multiple groups were performed using one-way ANOVA with Tukey’s post hoc test. Each experiment was independently repeated three times. * p < 0.05 vs. adjacent normal tissues or CCC-HB-2 cells

ZNF503-AS1 over-expression impedes proliferation, migration and invasion, and stimulates apoptosis of bladder cancer cells

Next, we aimed to investigate the effect of ZNF503-AS1 on various bladder cancer cell characteristics. Verification revealed that ZNF503-AS1 expression was significantly increased in ZNF503-AS1 transfected H/RB-CL2 cells, while being low in un-transfected cells (Fig. 2a & Fig. S1A). In addition, we observed, using EdU incorporation and flow cytometry assays, lower proliferation and higher apoptosis rates in ZNF503-AS1 over-expressing H/RB-CL2 and J82 cells, relative to cells transfected with oe-NC (p < 0.05) (Fig. 2b, c and Fig. S1B, C). Subsequent Transwell assays revealed that the migration and invasion abilities of ZNF503-AS1 overexpressing cells were notably lower than those of cells transfected with oe-NC (p < 0.05) (Fig. 2d and Fig. S1D). The expression levels of cell proliferation-related factor Ki67, apoptosis-related factor Bax and invasion-related factors MMP-2 and MMP-9 were subsequently determined by qRT-PCR and Western blotting, respectively. We found that compared to cells transfected with oe-NC, the expression levels of MMP-2, MMP-9 and Ki67 were all decreased in ZNF503-AS1 overexpressing cells, whereas the expression of Bax was increased (p < 0.05) (Fig. 2e and Fig. S1E). These data indicate that ZNF503-AS1 overexpression impedes proliferation, migration and invasion, and stimulates apoptosis of bladder cancer cells.

Fig. 2.

ZNF503-AS1 overexpression impedes proliferation, migration, invasion and resistance to apoptosis in H/RB-CL2 bladder cancer cells. a: qRT-PCR expression analysis of ZNF503-AS1 in H/RB-CL2 cells transfected with sh-ZNF503-AS1 or oe-ZNF503-AS1. b: EdU incorporation assay showing the effect of ZNF503-AS1 overexpression on the proliferation of H/RB-CL2 cells (× 400, scale bar = 25 μm). c: quantitative flow cytometric detection of the effect of ZNF503-AS1 overexpression on apoptosis of H/RB-CL2 cells. d: Transwell assay showing the effect of ZNF503-AS1 overexpression on the migration and invasion abilities of H/RB-CL2 cells (× 400, scale bar = 25 μm). e: qRT-PCR and Western blot analyses showing the effect of ZNF503-AS1 overexpression on the expression of MMP-2, MMP-9, Ki67 and Bax in H/RB-CL2 cells. Data are presented as mean ± standard deviation and were analyzed by unpaired t test between two groups from three independent experiments. * p < 0.05 vs. the oe-NC group

ZNF503-AS1 up-regulates SLC8A1 expression by recruiting GATA6

Next, we explored the downstream regulatory effectors of ZNF503-AS1, and found that ZNF503-AS1 can bind to the GATA6 transcription factor and, consequently, up-regulate the expression of SLC8A1, as revealed through the LncMAP website (http://bio-bigdata.hrbmu.edu.cn/LncMAP/index.jsp) and the GEPIA database (http://gepia.cancer-pku.cn). In addition, qRT-PCR and Western blotting were performed to determine SLC8A1 expression in H/RB-CL2 and J82 cells and, by doing so, we confirmed that its expression was significantly increased in cells over-expressing SLC8A1, while being reduced in response to SLC8A1 silencing (Fig. 3a and Fig. S2A). Decreased SLC8A1 expression was also noted in cells transfected with sh-ZNF503-AS1 compared to cells transfected with sh-NC (p < 0.05) (Fig. 3b and Fig. S2B). Relative to the cells transfected with oe-NC, the expression of SLC8A1 was significantly increased in the cells overexpressing ZNF503-AS1 (p < 0.05). In addition, we verified the binding of ZNF503-AS1 to GATA6 using a RIP assay (Fig. 3c and Fig. S2C). Compared with IgG, we found that the binding of GATA6 to ZNF503-AS1 was significantly increased (p < 0.05), and that the anti-GATA6 antibody readily precipitated ZNF503-AS1, indicating that ZNF503-AS1 can form a complex with GATA6. In order to identify the binding sites of the GATA6 protein, we predicted the two most probable binding sites in SLC8A1 DNA using the UCSC (http://genome.ucsc.edu/) and JASPAR (http://jaspar.genereg.net/) databases (Fig. 3d). Using a luciferase reporter assay (Fig. 3e, f) we subsequently found a significant decrease in the ability of oe-NC and oe-GATA6 to activate SLC8A1 (p < 0.05) when SLC8A1 site 1 was truncated or mutated, compared to cells transfected with oe-NC (p < 0.05). Besides, truncation to remove or mutate SLC8A1 site 2 did not affect the ability of GATA6 to activate SLC8A1 (p > 0.05). These results indicate that site 1 is indeed the site of GATA6 binding to SLC8A1 DNA. Additional RNA pull-down assays revealed that lncRNA ZNF503-AS1 can specifically bind to GATA6 (p < 0.05; Fig. 3g). The binding ability of GATA6 to SLC8A1 DNA binding site 1 was subsequently validated through a ChIP assay in H/RB-CL2 cells (Fig. 3h). In addition, we found that the amount of amplification products obtained from SLC8A1 site 1 primers in the GATA6 group was larger than that from distal primers in the IgG group (p < 0.05), while there were no significant differences in the amounts of amplification products between the two pairs of primers in the IgG group (p > 0.05). These results indicate that site 1 of SLC8A1 DNA is indeed the site binding of GATA6.

Fig. 3.

ZNF503-AS1 recruits GATA6 to up-regulate SLC8A1 expression in H/RB-CL2 bladder cancer cells. a: qRT-PCR and Western blot analyses of cells with SLC8A1 overexpression or silencing; * p < 0.05 vs. the sh-NC group, # p < 0.05 vs. the oe-NC group. b: qRT-PCR and Western blot analyses of SLC8A1 expression in cells with ZNF503-AS1 overexpression or silencing. * p < 0.05 vs. the sh-NC group, # p < 0.05 vs. the oe-NC group. c: RIP assay showing binding of ZNF503-AS1 to GATA6. * p < 0.05 vs. the IgG group. d: predicted GATA binding sites in SLC8A1 DNA. e: dual luciferase reporter analysis of cells co-transfected with truncated SLC8A1 recombinant luciferase reporter vector and GATA6 expression vector. * p < 0.05 vs. the oe-NC group. f: dual luciferase reporter assay of cells co-transfected with mutated SLC8A1 recombinant luciferase reporter vector and GATA6 expression vector. * p < 0.05 vs. the oe-NC group. g: RNA pull-down assay showing direct interaction of lncRNA ZNF503-AS1 and GATA6. h. ChIP assay showing binding of GATA6 to the SLC8A1 DNA binding site 1, * p < 0.05 vs. the IgG group. i: qRT-PCR and Western blot analyses of SLC8A1 expression, * p < 0.05 vs. the oe-NC + sh-NC group, # p < 0.05 vs. the oe-ZNF503-AS1 + sh-NC group. Data are presented as mean ± standard deviation and were analyzed by unpaired t test between two groups and compared by one-way ANOVA with Tukey’s post hoc test from three independent experiments

In order to verify a regulatory relationship between ZNF503-AS1, SLC8A1 and GATA6, we employed qRT-PCR and Western blot analyses. We found that compared to cells transfected with oe-NC + sh-NC, those transfected with oe-ZNF503-AS1 + sh-NC exhibited increased SLC8A1 expression levels (p < 0.05) (Fig. 3i and Fig. S2D). SLC8A1 expression was decreased in cells transfected with oe-ZNF503-AS1 + sh-GATA6 compared to those transfected with oe-ZNF503-AS1 + sh-NC (p < 0.05). These results indicate that ZNF503-AS1 can up-regulate SLC8A1 expression by recruiting GATA6.

SLC8A1 over-expression inhibits proliferation and migration of bladder cancer cells by promoting Ca²⁺ influx

Next, we transfected H/RB-CL2 bladder cancer cells with oe-NC and oe-SLC8A1 to assess their proliferation, migration and invasion abilities, as well as their intracellular Ca²⁺ concentrations and Ca²⁺-ATPase activities. SLC8A1 expression was found to be down-regulated in H/RB-CL2 cells compared to embryonic bladder tissue-derived CCC-HB-2 cells (p < 0.05) (Fig. 4a). To further validate the decreased expression of SLC8A1 in bladder cancer, SLC8A1 expression was assessed in primary bladder cancer tissues and adjacent normal tissues using immunohistochemistry. We found that the bladder cancer tissues exhibited a lower SLC8A1 expression compared to the adjacent normal tissues (p < 0.05) (Fig. 4b). Moreover, a positive correlation was observed between ZNF503-AS1 and SLC8A1 expression in the bladder cancer tissues (Fig. 4c).

Fig. 4.

SLC8A1 inhibits the proliferation and migration of H/RB-CL2 bladder cancer cells by promoting Ca²⁺ influx. a: qRT-PCR and Western blot analyses showing SLC8A1 expression in CCC-HB-2 cells and H/RB-CL2 cells. * p < 0.05 vs. CCC-HB-2 cells. b: immunohistochemical analysis (× 400, scale bar = 25 μm) of SLC8A1 expression in primary bladder cancer tissues and adjacent normal tissues. * p < 0.05 vs. adjacent normal tissues. Data comparisons between bladder cancer tissues and adjacent normal tissues were performed by paired t test (n = 86). c: correlation analysis of ZNF503-AS1 and SLC8A1 expression. d: effect of SLC8A1 overexpression on intracellular Ca²⁺ fluorescence intensity. e: effect of SLC8A1 overexpression on Ca²⁺-ATPase activity. f: qRT-PCR and Western blot analyses showing the expression of MMP-2, MMP-9, Ki67 and Bax in H/RB-CL2 cells after SLC8A1 overexpression. Data are presented as mean ± standard deviation and comparisons between two groups were conducted by unpaired t test. Correlations between data of the two groups were evaluated by Pearson’s correlation coefficient. Time-based comparisons among multiple groups were conducted by repeated measures ANOVA with Tukey’s post hoc test. In panel D-I * p < 0.05 vs. the oe-NC group

Using additional EdU incorporation, flow cytometry and Transwell assays, we found that H/RB-CL2 and J82 cells over-expressing SLC8A1 exhibited reduced proliferation rates, increased apoptosis rates and attenuated invasion and migration rates, compared to cells transfected with oe-NC (p < 0.05) (Fig. S3A-D). Subsequent Ca²⁺ fluorescence intensity and Ca²⁺-ATPase activity assays revealed that, relative to cells transfected with oe-NC, cells over-expressing SLC8A1 exhibited increased intracellular Ca²⁺ concentrations at 50 s, and decreased Ca²⁺-ATPase activities (p < 0.05) (Fig. 4d-e). In addition, compared with cells treated with oe-NC, the expression levels of MMP-2, MMP-9 and Ki67 were found to be decreased, while those of Bax were found to be increased in cells over-expressing SLC8A1 (p < 0.05) (Fig. 4f). These findings suggest that SLC8A1 over-expression may hamper the proliferation and migration of bladder cancer cells by promoting Ca²⁺ influx.

ZNF503-AS1 promotes Ca²⁺ concentration and inhibits bladder cancer cell proliferation and migration by regulating SLC8A1 expression

Next, we set out to assess whether the function of ZNF503-AS1 is achieved through SLC8A1 regulation. First, the expression of SLC8A1 was determined in H/RB-CL2 and J82 cells after different treatments by qRT-PCR and Western blot analyses. Compared to the oe-NC + sh-NC group, the expression of SLC8A1 was significantly increased in the oe-ZNF503-AS1 + sh-NC group, while being reduced in the oe-ZNF503-AS1 + sh-SLC8A1 group relative to the oe-ZNF503-AS1 + sh-NC group (p < 0.05) (Fig. S4A). In addition, EdU incorporation, flow cytometry and Transwell assays revealed that, in contrast to the cells transfected with oe-NC + sh-NC, the cells transfected with oe-ZNF503-AS1 + sh-NC exhibited reduced proliferation rates, increased apoptosis rates, and suppressed migration and invasion rates (p < 0.05). Compared to transfection with oe-ZNF503-AS1 + sh-NC, transfection with oe-ZNF503-AS1 + sh-SLC8A1 increased the proliferation rates, decreased the apoptosis rates, and increased the migration and invasion rates (p < 0.05) (Fig. S4B-E).

Additionally, intracellular Ca²⁺ concentrations and Ca²⁺-ATPase activities were examined (Fig. 5a-b). Relative to cells transfected with oe-NC + sh-NC, cells transfected with oe-ZNF503-AS1 + sh-NC exhibited elevated Ca²⁺ concentrations and reduced Ca²⁺-ATPase activities from 50 s (p < 0.05). In addition, compared to transfection with oe-ZNF503-AS1 + sh-NC, transfection with oe-ZNF503-AS1 + sh-SLC8A1 resulted in decreased Ca²⁺ concentrations and increased Ca²⁺-ATPase activities from 50 s (p < 0.05). Moreover, in response to transfection with oe-NC + sh-NC, the expression levels of MMP-2, MMP-9 and Ki67 were decreased, while those of Bax were increased after transfection with oe-ZNF503-AS1 + sh-NC (p < 0.05). The cells transfected with oe-ZNF503-AS1 + sh-SLC8A1 exhibited higher expression levels of MMP-2, MMP-9 and Ki67, as well as lower expression levels of Bax compared to the cells transfected with oe-ZNF503-AS1 + sh-NC (p < 0.05) (Fig. 5c). These results support the notion that ZNF503-AS1 increases the Ca²⁺ concentration and, consequently, disrupts bladder cancer cell proliferation and migration by regulating SLC8A1.

Fig. 5.

ZNF503-AS1 increases Ca²⁺ levels and inhibits H/RB-CL2 bladder cancer cell proliferation and migration by regulating SLC8A1 expression. a: detection of Ca²⁺ fluorescence intensity in H/RB-CL2 cells over time. b: detection of Ca²⁺-ATPase activity. c: qRT-PCR and Western blot analyses showing expression of MMP-2, MMP-9, Ki67 and Bax in H/RB-CL2 cells. Data are presented as mean ± standard deviation. Comparisons among multiple groups were performed using one-way ANOVA with Tukey’s post hoc test or repeated measures ANOVA for data analysis. The experiments were repeated three times. * p < 0.05 vs. the oe-NC + sh-NC group, # p < 0.05 vs. the oe-ZNF503-AS1 + sh-NC group

ZNF503-AS1 impedes the development of tumor growth in vivo by inhibiting SLC8A1

Next, in vivo assays were performed to confirm the above in vitro results by altering the expression levels of ZNF503-AS1 and SLC8A1 in a mouse xenograft model. We found that, compared to the mice injected with cells transfected with sh-NC + oe-NC, those transfected with oe-ZNF503-AS1 + sh-NC showed a decreased tumorigenic ability from the 9th day onwards after inoculation (p < 0.05). Compared with the mice inoculated with oe-ZNF503-AS1 + sh-NC transfected cells, the tumorigenic ability of oe-ZNF503-AS1 + sh-SLC8A1 transfected cells was significantly enhanced from the 9th day onwards after inoculation (p < 0.05) (Fig. 6a). Additionally, relative to the oe-NC + sh-NC transfected tumor cells, oe-ZNF503-AS1 + sh-NC transfected tumor cells exhibited decreased expression levels of MMP-2, MMP-9 and Ki67, and increased expression levels of Bax (p < 0.05). Besides, oe-ZNF503-AS1 + sh-SLC8A1 transfected tumor cells exhibited increased expression levels of MMP-2, MMP-9 and Ki67, and decreased expression levels of Bax (p < 0.05) (Fig. 6b). Immunohistochemical analysis of SLC8A1 expression indicated that it was increased in oe-ZNF503-AS1 + sh-NC transfected tumor cells compared to sh-NC + oe-NC transfected tumor cells (p < 0.05). Compared to oe-ZNF503-AS1 + sh-NC transfected tumor cells, SLC8A1 expression was found to be decreased in oe-ZNF503-AS1 + sh-SLC8A1 transfected tumor cells (p < 0.05) (Fig. 6c). These findings underscore the notion that ZNF503-AS1 retards bladder cancer growth via down-regulation of SLC8A1 in vivo.

Fig. 6.

ZNF503-AS1 impedes in vivo tumor growth development by inhibiting SLC8A1 expression. a: quantification of tumorigenic abilities of bladder cancer cells in nude mice. b: qRT-PCR and Western blot analyses showing expression of MMP-2, MMP-9, Ki67 and Bax. c: immunohistochemistry to detect SLC8A1 expression in the tumor tissues (× 400, scale bar = 25 μm). Data are presented as mean ± standard deviation. Comparisons among multiple groups were performed using one-way with Tukey’s post hoc test or repeated measures ANOVA for data analysis from three independent experiments. n = 12. * p < 0.05 vs. the oe-NC + sh-NC group, # p < 0.05 vs. the oe-ZNF503-AS1 + sh-NC group

Discussion

Accumulating evidence indicates that aberrant Ca²⁺ homeostasis can trigger enhanced cancer cell proliferation, angiogenesis and diminished apoptosis, all of which lead to a worse prognosis [22]. Increased understanding of the molecular mechanisms underlying bladder cancer development has led to the identification of various biomarkers that may be used to facilitate its treatment [23]. It has also amply been shown that abnormal expression of lncRNAs may lead to tumor suppression or enhancement through the deregulation of tumor-related mRNAs [24]. In the current study, we found that ZNF503-AS1 can up-regulate SLC8A1 expression by binding to the transcription factor GATA6, thereby increasing intracellular Ca²⁺ concentrations and, ultimately, hampering bladder cancer development (Fig. 7).

Fig. 7.

Schematic representation of the tumor suppressor role of ZNF503-AS1 in bladder cancer. Overexpression of ZNF503-AS1 can result in recruitment of GATA6 to up-regulate SLC8A1 expression and increase the intracellular Ca²⁺ concentration, thereby inhibiting bladder cancer cell proliferation, invasion and migration, as well as its resistance to apoptosis

LncRNAs are known to serve as diagnostic, prognostic and predictive biomarkers in bladder cancer, showing up-regulated or down-regulated expression patterns during tumor progression [25]. Here we found, using a freely available dataset, that the intergenic lncRNA ZNF503-AS1 is lowly expressed in bladder cancer, a result that was subsequently validated in primary bladder cancer tissues and cells. LncRNAs have been implicated in the progression of several malignancies, including bladder cancer, affecting cell proliferation, apoptosis and metastasis [26, 27]. Multiple additional mechanistic studies have associated lncRNAs with biological processes such as cell differentiation, growth and immune responses [28, 29]. ZNF503-AS1 expression negatively correlates with ZNF503 expression, a transcriptional inhibitor reported to facilitate the proliferation and migration of mammary epithelial cells [14]. In addition, a previous study has shown that ZNF503 down-regulation can stimulate breast cancer progression while, on the contrary, its over-expression can stimulate mammary epithelial cell motility and invasion [30]. In the present study, we found that ZNF503-AS1 over-expression is associated with bladder cancer development and affects the overall prognosis in conjunction with ZNF503 expression.

Ca²⁺ signaling is known to play a critical role in cancer cell proliferation and migration, and may result in altered expression of proteins associated with Ca²⁺ fluxes [31]. Additional studies have indicated that Ca²⁺ signaling may also modulate the release of inflammatory cytokines induced by Bacillus Calmette-Guérin in bladder cancer cells [32]. Moreover, augmented Ca²⁺ release induced by bee venom has been shown to be capable of expediting both caspase-dependent and caspase-independent apoptotic death of bladder cancer cells [4]. Accordingly, our current data indicate that when the Ca²⁺ influx is enhanced through ZNF503-AS1 overexpression, the growth of bladder cancer cells may be hampered. In addition, we found that ZNF503-AS1 exerted its effect by regulating the expression of SLC8A1. Previously, low SLC8A1 expression has been associated with Ca²⁺ dysregulation in penile carcinoma, resulting in decreased apoptosis and increased proliferation [17]. More importantly, another study revealed that over-expression of circSLC8A1 hampered bladder cancer cell migration and proliferation, leading to in vivo suppression of bladder cancer growth [33]. Accordingly, our current in vivo assays indicated that ZNF503-AS1 hampered the development of bladder cancer growth by inhibiting SLC8A1 expression.

Additionally, we found that ZNF503-AS1 can up-regulate SLC8A1 expression by recruiting the transcription factor GATA6, thereby elevating intracellular Ca²⁺ concentrations and suppressing bladder cancer development. GATA6 has frequently been correlated with the deregulation of cancer-related genes [16]. In pancreatic cancer cells, for example, GATA6 up-regulation has been found to facilitate proliferation and colony formation by activating Wnt signaling via DKK1 [34]. In addition, GATA6 has previously been highlighted as a predictor of poor prognosis in ovarian cancer [35]. In line with our findings, a recent study revealed that ZNF503 can enhance the aggressiveness of breast cancer cells by modulating GATA3 activity, which suggests that ZNF503 may serve as a promising therapeutic target [15].

Taken conjointly, our current study sheds new light on the ZNF503-AS1/GATA6/SLC8A1 axis and its association with bladder cancer aggressiveness. Over-expression of ZNF503-AS1 can recruit GATA6 to up-regulate SLC8A1 expression, thereby increasing intracellular Ca²⁺ concentrations and decreasing bladder cancer cell growth. The final elucidation of the underlying mechanisms requires further study.

Compliance with ethical standards

Conflict of interest

None declared.

Ethical statement

The study was conducted under approval of the Ethics Committee of The Second Xiangya Hospital, Central South University. All participating patients signed informed consent documentation. Nude mice were used for in vivo studies and were cared for in accordance with the principles of the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health with efforts made to ensure minimal suffering of the animals used in the study.