Circular RNA circ-Foxo3 induced cell apoptosis in urothelial carcinoma via interaction with miR-191-5p.

BACKGROUND: Circular RNAs (circRNAs) play a critical role in cancer. Emerging evidence has shown circ-Foxo3, a circRNA, was dysregulated in a variety of tumor types. However, the exact role of circ-Foxo3 in bladder cancer has never been studied.
METHODS: We measured the expression level of circ-Foxo3 in human and murine bladder cancer tissues and in various human bladder cancer cell lines. We induced bladder cancer in mice by a carcinogen N-butyl-N-(4-hydroxybutyl)nitrosamine (BBN). circ-Foxo3 was overexpressed in mice by lentiviral gene transfer and in cultured cells via overexpression plasmid. The effect of circ-Foxo3 on apoptosis was examined via apoptotic marker staining, Western blot, and flow cytometry. We further characterized the interaction between circ-Foxo3 and miR-191 and its functional impact on bladder cancer cells.
RESULTS: circ-Foxo3 was downregulated in bladder cancer in vivo and in vitro, and was upregulated in response to apoptotic stress. Overexpression of circ-Foxo3 promoted bladder cancer cell apoptosis in BBN mice and in human bladder cancer cell lines. miR-191-5p suppressed circ-Foxo3 expression and the pro-apoptotic effect of circ-Foxo3 in bladder cancer cells via directly targeting the 3'-untranslated region (3'-UTR) of circ-Foxo3.
CONCLUSION: circ-Foxo3 was downregulated in bladder cancer in vivo and in vitro, and promoted bladder cancer apoptosis via direct interaction with miR-191. circ-Foxo3 could be a potential therapeutic target for bladder cancer.

Background

Bladder cancer is the 9th most commonly diagnosed cancer worldwide and the 4th most common cancer in men.1,2 Globally, about 430,000 new bladder cancer cases are diagnosed annually.1 In the United States, over 80,000 new cases of bladder cancer are diagnosed each year, and over 17,000 patients die from bladder cancer annually.2 Urothelial carcinoma, also known as transitional cell carcinoma, is the most common type of bladder cancer. Urothelial carcinoma originates from the urothelial cells that line the inner layer of urinary tract. The development of urothelial carcinoma is predisposed by certain risk factors, including smoking and exposure to certain industrial carcinogens such as aromatic amines. To date, there is no definitive cure to urothelial carcinoma, and its mortality rate has maintained stable. Therefore, identifying early diagnostic markers and novel therapeutic targets for bladder cancer, particularly urothelial carcinoma, remains a major public health need.

Circular RNA (circRNA) is a type of single-stranded RNA that forms a closed loop by joining the 5ʹ and 3ʹ ends of a linear RNA. CircRNAs are ubiquitously expressed from archaea to eukaryotes and are evolutionary conserved, strongly suggesting their functional importance.3 CircRNA can be protein-coding or non-coding.4,5 Although the detailed function of most circRNAs remains unclear, one major role of those non-coding circRNAs is shown to be gene regulation,6 potentially by interaction with microRNAs or RNA-binding proteins.7,8

circ-Foxo3 is a circRNA derived from the FOXO3 gene that also encodes the linear FOXO3 mRNA.9 The expression of circ-Foxo3 appeared independent of the expression of FOXO3 mRNA, and may have regulatory function on targets beyond the linear FOXO3 mRNA.10 Recently, emerging evidence has indicated that circ-Foxo3 was detectable in multiple cancers, and that circ-Foxo3 was associated with cell cycle retardation or apoptosis.10–14 However, the role of circ-Foxo3 in bladder cancer has yet to be fully understood.

Here, we explored the role of circ-Foxo3 in bladder cancer. We found circ-Foxo3 was dysregulated in bladder cancer tissue in vivo and in vitro. circ-Foxo3 was a direct target of the microRNA miR-191-5p. Together, a miR-191-circ-Foxo3 axis appeared critical to bladder cancer apoptosis.

Methods

Human bladder cancer samples

The study involving human samples was approved by the Institutional Review Board of The First Affiliated Hospital of Harbin Medical University. Written informed consent was obtained from all patients upon recruiting. Thirty pairs of fresh bladder cancer tissues and adjacent normal bladder tissue were excised during partial or radical cystectomy from patients with confirmed diagnosis of urothelial carcinoma at the Department of Urology of The First Affiliated Hospital of Harbin Medical University between 2016 and 2017. None of the patients had medication or radiation therapy prior to surgery. The specimens were snap-frozen in liquid nitrogen immediately after excision. Pathological and histological diagnoses were performed by two pathologists independently, and the diagnosis of urothelial carcinoma was confirmed in all samples. The grades and stages of specimens were classified using 2004 World Health Organization Consensus Classification and Staging System.

Cell culture and treatment

Three cell lines of human transitional cell carcinoma, T24, UM-UC-3 and J82, and a normal human uroepithelial cell line SV-HUC-1, were obtained from the American Type Culture Collection (ATCC, USA). All cells were maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum (Gibco, USA). Cells were kept in an incubator at 37°C with humidified atmosphere containing 5% CO2. Medium was replaced every 2–3 days.

Cells receiving transfection were treated in RPMI-1640 basal medium (serum-free) containing 0.4 mM hydrogen peroxide (H2O2), or 1 μg/mL doxorubicin (DOX), or 2 μg/mL cisplatin (CP) for 24 h.

Bladder tumor model in mice

All animal protocols were approved by the Ethical Committee of The First Affiliated Hospital of Harbin Medical University and adhered to the Guide for the Care and Use of Laboratory Animals (NRC 2011, 8th Edition). Male, 8-week-old C57BL/6 mice were obtained from The First Affiliated Hospital of Harbin Medical University. Mice were housed in a temperature-controlled environment with 12-h light/dark cycles and unrestricted access to food and drinking water. To induce carcinogenesis in mouse bladder, 0.05% N-butyl-N-(4-hydroxybutyl)nitrosamine (BBN; Tokyo Chemical Industry Co Ltd, Japan) was added to the drinking water for 17 weeks until tissue harvesting. BBN-free water (vehicle) was added in the drinking water in equal volume for 17 weeks as control treatment. The induction of bladder cancer was histologically confirmed at tissue harvesting.

Histology

Mice were euthanized by CO2 inhalation. The bladders were exposed. A small fraction of bladder tumor tissue and the adjacent normal bladder tissue was removed for RNA extraction. The remaining bladder was inflated and soaked in 4% formaldehyde in phosphate-buffered saline (PBS) overnight. Fixed bladders were cut in half, positioned in cassettes, soaked in 70% alcohol overnight, and embedded in paraffin. The paraffin-embedded blocks were cut into 6-μm-thick serial sections, stained with hematoxylin and eosin (H&E), and gold enhanced for visualization by light microscopy.

Circrna vectors

The circ-Foxo3 over-expression plasmid was constructed by sub-cloning the human circ-Foxo3 cDNA (synthesized by TSINGKE, China) onto a pCD-ciR circRNA expression vector (Geneseed Biotech, China), which has a front circular frame and a back circular frame to ensure functional circRNA transcription. The sequence-verified plasmid was transfected into cells using Lipofectamine 2000 (Thermo Fisher Scientific, USA). The empty pCD-ciR vector was used as control.

The control (circControl-GFP) and circ-Foxo3 over-expression (circFoxo3-GFP) lentiviruses were generated by Hanbio (China). For somatic gene transfer, circControl-GFP or circFoxo3-GFP lentivirus (2×108 viral genome copies/μL) was injected into the male 8-week-old C57BL/6 mice (5 μL each). Mice were divided into the following 4 groups for treatments at 2 weeks after lentiviral injection: circControl + vehicle, circFoxo3+ vehicle, circControl + BBN, and circFoxo3+ BBN.

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

Total RNA was isolated from homogenized tissue or cultured cells using an RNeasy Mini Kit (Qiagen, USA) following the manufacturer’s instruction. cDNA was synthesized using PrimeScript RT Master Mix (Takara Bio, China) with 2 μg total RNA input in 20 μL reaction. A portion of the reverse transcription product (1 μL, equal to 0.1 μg cDNA) was saved for regular PCR with two appropriate primers. qPCR was performed using the miScript SYBR Green PCR Kit (Qiagen, USA) with 1 μL cDNA template. The relative expression of RNA was quantified using the ΔΔCq method.15

Apoptosis assay

Cells were grown on 6-well plates after transfection. Apoptosis was quantified using the Annexin V-FITC apoptosis kit (Thermo Fisher Scientific, USA). Briefly, cells were detached 48 h post-transfection, washed with PBS, and resuspended in 500 μL binding buffer containing 5 μL propidium iodide (PI) and 5 μL Annexin V-FITC. After incubation at room temperature for 10 min, cells were washed and resuspend in binding buffer. Apoptosis was analyzed by a FACScalibur flow cytometer (BD Biosciences, USA).

Western blot

Cells were harvested in RIPA lysis buffer and protein concentration was normalized by the Brandford protein assay. Normalized protein was boiled in Laemmli sample buffer, and resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Protein was transferred onto polyvinylidene fluoride membrane and immunoblotted as described.16 The following primary antibodies were used: anti-cleaved-caspase3 (1:1000), anti-Bcl2 (1:1000), anti-Bax (1:1000), and anti-GAPDH (1:1000) (all from Abcam, UK).

Cell viability assay

Cell viability was measured by the Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Kumamoto, Japan) following the manufacturer’s protocol. Briefly, transfected cells were plated on 96-well plates and cultured in 100 μL/well medium for 24 h. CCK-8 solution was added to the plates (10 μL/well) at 0 h, 24 h, 48 h, 72 h, and 96 h, respectively, and further incubated for 4 hrs for colorimetric reaction. Absorbance at 450 nm was measured with a microplate reader to represent cell viability.

Luciferase reporter assay

The 3ʹ-untranslated region (3ʹ-UTR) of circ-Foxo3 RNA containing the putative miR-191 binding sequence was amplified by PCR and cloned into a pmirGLO (Promega, USA) dual-luciferase miRNA target expression vector (named pmirGlo-CREB1-WT; WT). Targeted mutation of the vector was generated within the circ-Foxo3 3ʹ-UTR sequence by a QuikChange Multi Site-Directed Mutagenesis Kit (Agilent Technologies, USA) resulting a mutant luciferase reporter vector (named pmirGlo-CREB1-MUT; MUT). For the luciferase reporter assays, WT or MUT luciferase reporter vectors were transfected into J82 cells on 6-well plates in combination with miRNA-191 mimics or non-targeting control microRNA mimics using Lipofectamine 2000 (Thermo Fisher Scientific, USA). The firefly luciferase activity was measured 48 h post-transfection using a dual-luciferase reporter assay system (Promega, USA) following the manufacturer’s instruction, and normalized to Renilla luciferase activity.

Terminal deoxynucleotidyl transferase dutp nick end labeling (TUNEL) assay

Apoptotic DNA fragmentation in situ was detected by the TUNEL assay as previously described.17

Statistical method

All data were expressed as mean ± standard deviation (SD). Difference of means between two groups was compared by two-tailed Student’s t-test, with significance set as p <0.05. Difference of means among three or more groups was analyzed by one-way analysis of variance (ANOVA), followed by multiple comparisons test, and multiplicity adjusted p <0.05 was considered statistically significant.

Results

circ-Foxo3 was downregulated in bladder cancer in vivo and in vitro

We first examined the expression of circ-Foxo3 in bladder cancer tissues and cell lines. RT-qPCR showed the RNA expression of circ-Foxo3 was significantly downregulated in tumor tissues isolated from bladder cancer patients as compared with the adjacent normal bladder tissues (Figure 1A). circ-Foxo3 expression was also reduced in murine bladder tumors induced by 17-week oral treatment of BBN, a carcinogen that effectively induces bladder cancer (Figure 1B).

Figure 1

circ-Foxo3 was downregulated in bladder cancer in vivo and in vitro. (A) Expression of circ-Foxo3 RNA in tumor tissues and adjacent normal bladder tissues taken from bladder cancer patients (n=30). (B) Expression of circ-Foxo3 RNA in tumor tissues and adjacent normal bladder tissues from mice treated with vehicle or BBN for 17 weeks (n=10). (C) Expression of circ-Foxo3 RNA in different bladder cell lines (n=5). Mean ± SD, *p<0.05.

We also measured circ-Foxo3 expression in bladder cancer in vitro. The expression of circ-Foxo3 RNA was significantly down-regulated in human bladder cancer cell lines T24, UM-UC-3, and J82, as compared with the immortalized normal bladder cell line SV-HUC-1 (Figure 1C).

circ-Foxo3 was upregulated in response to apoptotic stress

Next, we explored the role of circ-Foxo3 in apoptosis. We first tested the level of apoptosis in J82 cells treated with doxorubicin, cisplatin or H2O2. Western blot showed the apoptotic protein markers was markedly increased following these treatments, as indicated by increased cleaved-caspase3 expression and increased ratio of the pro-apoptotic protein Bax over the pro-survival protein Bcl2 (Figure 2A).

Figure 2

circ-Foxo3 was upregulated in response to apoptotic stress. (A) Western blot of cleaved-caspase3 (c-caspase3), Bax, Bcl2 and GAPDH expression in J82 cells treated with vehicle (Ctrl), doxorubicin (Dox; 1μg/mL), cisplatin (CP; 2 μg/mL), and H2O2 (H2O2; 0.4 mM). The band intensity relative to GAPDH, and the ratio of Bax/Bcl2 band intensity were shown on the right. (B–D) Expression of circ-Foxo3 RNA level in bladder cell lines treated with vehicle (B–D), doxorubicin (B), cisplatin (C), and H2O2 (D), respectively (n=5). Mean ± SD, *p<0.05.

We further characterized the apoptotic stress in additional cell lines. The expression of circ-Foxo3 was significantly upregulated in bladder cancer cell lines T24, UM-UC-3 and J82 as well as the normal bladder epithelial cell line SV-HUC-1 upon treatment with doxorubicin (Figure 2B), cisplatin (Figure 2C), and H2O2 (Figure 2D).

circ-Foxo3 induced apoptosis in bladder tumor cells of BBN mice and in bladder cancer cell lines

To test the effect of circ-Foxo3 on apoptosis in vivo, we injected control (circControl-GFP) or circ-Foxo3 (circFoxo3-GFP) lentiviral vectors for somatic gene transfer in mice. We then treated mice with oral BBN for 17 weeks to induce bladder cancer. Somatic gene transfer of circ-Foxo3 significantly increased cell apoptosis in the bladder tissue of BBN mice as measured by TUNEL staining (Figure 3A). Likewise, Western blot showed that the expression the cleaved-caspase3 and the ratio of Bax to Bcl2 was increased in mice receiving circ-Foxo3 lentiviral injection (Figure 3B).

Figure 3

circ-Foxo3 promoted bladder tumor cell apoptosis in BBN mice. Male, 8-week-old C57BL/6 mice received control (circControl-GFP) and circ-Foxo3 (circFoxo3-GFP) somatic gene transfer by lentiviral injection, followed by exposure to oral vehicle or BBN treatment for 17 weeks. (A) The bladder tumor cell apoptosis was evaluated by the TUNEL assay (left) and quantified (right; n=10). (B)Western blot of cleaved-caspase3 (c-caspase3), Bax, Bcl2 and GAPDH expression in murine bladder tissues. The band intensity relative to GAPDH, and the ratio of Bax/Bcl2 band intensity were quantified (n=5). Mean ± SD, *p<0.05, **p<0.01.

We then examined the effect of circ-Foxo3 overexpression on apoptosis in vitro. We transiently overexpressed circ-Foxo3 RNA in human bladder cancer cell lines with a circ-Foxo3 plasmid (Figure 4A). Overexpression of circ-Foxo3 promoted bladder cancer cell apoptosis, as measured by Western blot of apoptotic markers (Figure 4B) and flow cytometry (Figure 4C).

Figure 4

circ-Foxo3 induced cell apoptosis and reduced viability in bladder cancer cells. T24, UM-UC-3, and J82 bladder cancer cell lines received control (Ctrl) or circ-Foxo3 over-expression (circ-Foxo3) plasmids. (A) Expression of circ-Foxo3 RNA level in bladder cell lines. (B) Western blot of cleaved-caspase3 (c-caspase3), Bax, Bcl2 and GAPDH expression in cells receiving treatments. The band intensity relative to GAPDH, and the ratio of Bax/Bcl2 band intensity were quantified (n=5). (C) Analysis of apoptosis in treated bladder cancer cell lines by flow cytometry (n=5). (D) Cell viability measured by the CCK-8 assay in T24, UM-UC-3, and J82 bladder cancer cell lines. Mean ± SD, *p<0.05.

Furthermore, we examined the viability of bladder cancer cell lines. Overexpression of circ-Foxo3 significantly reduced bladder cancer cell viability as measured by the CCK-8 assay (Figure 4D).

miR-191-5p suppressed circ-Foxo3 expression in bladder cancer cells

Our preliminary analysis revealed a strong negative correlation between the expression of circ-Foxo3 and miR-191-5p RNA in 30 tumor biopsies from bladder cancer patients (Figure 5A). miR-191-5p is a novel microRNA that was implicated in a variety of solid tumors.18–24 The strong correlation between circ-Foxo3 and miR-191-5p RNA expression prompted us to further examine their potential functional interaction in bladder cancer cells.

Figure 5

miR-191-5p suppressed circ-Foxo3 expression in bladder cancer cells. (A) Correlation between of circ-Foxo3 and miR-191-5p RNA expression in tumor biopsy tissue from bladder cancer patients (n=30). (B) Bioinformatic analysis using the TargetScan algorithm () predicted a putative miR-191-5p target site in the 3ʹ-UTR of circ-Foxo3. (C) Bladder cell lines were transfected with a non-targeting microRNA (Ctrl), or a miR-191-5p mimic, or a miR-191-5p inhibitor. The RNA expression of miR-191 was measured by RT-qPCR (n=5). (D) circ-Foxo3 expression luciferase reporter assay. A luciferase reporter vector carrying the wild-type (pmirGlo-CREB1-WT; WT) or mutant (pmirGlo-CREB1-MUT; MUT) 3ʹ-UTR sequence of circ-Foxo3 were transfected into J82 cells in combination with a miR-191-5p mimic (miR-191-5p) or a non-targeting microRNA mimic (Control). The relative luciferase activity (arbitrary units, normalized to Renilla luciferase activity) was analyzed 48 hrs post-transfection. (E) The expression of circ-Foxo3 was measured by RT-qPCR in bladder cancer cells receiving non-targeting microRNA (Ctrl), or a miR-191-5p mimic, or a miR-191-5p inhibitor (n=5). Mean ± SD, *p<0.05.

In an attempt to search for the direct interaction between circ-Foxo3 and miR-191-5p RNA, we identified a predicted miR-191-5p target site within the 3ʹ-UTR of circ-Foxo3 RNA using the TargetScan algorithm () (Figure 5B). To experimentally validate the interaction, we designed luciferase reporters coupled with either wild-type (WT) or mutant (MUT) 3ʹ-UTR sequence of circ-Foxo3, and co-transfected the reporters with a miR-191-5p mimic or inhibitor in bladder cancer cells. The expression of miR-191-5p was significantly enhanced by miR-191-5p mimic and reduced by its inhibitor (Figure 5C). In J82 bladder cancer cells, co-transfection of miR-191-5p significantly suppressed wild-type, but not mutant, circ-Foxo3 luciferase reporter activity (Figure 5D). These results suggested that the 3ʹ-UTR of circ-Foxo3 was a direct binding target of miR-191-5p.

Similarly, ectopic overexpression of miR-191 significantly suppressed the expression of circ-Foxo3 in T24, UM-UC-3 and J82 cells, while silencing of miR-191 significantly enhanced expression of circ-Foxo3 (Figure 5E). These data suggest that transcription of circ-Foxo3 RNA is suppressed by miR-191 in bladder cancer cells.

miR-191-5p suppressed apoptosis via inhibiting circ-Foxo3 in bladder cancer cells

To elucidate the functional relevance of the interaction between miR-191 and circ-Foxo3, we transfected bladder cancer cells with the miR-191-5p mimic, and treated cells with cisplatin to induce apoptosis. Western blot showed cisplatin-induced apoptosis was significantly attenuated by overexpression of miR-191 (Figure 6A). Notably, overexpression of circ-Foxo3 alone also increased apoptosis (Figure 6B). Co-transfection with miR-191 mimic, however, effectively suppressed apoptosis induced by circ-Foxo3 as indicated by Western blot (Figure 6B) and flow cytometry (Figure 6C). Consistent with the cell apoptosis measurements, CCK-8 assay showed that ectopic overexpression of circ-Foxo3 significantly reduced cell viability, which was partially rescued by concurrent transfection of miR-191 mimic in T24, UM-UC-3 and J82 cells, suggesting a pro-survival effect of miR-191 on bladder cancer cells that was partially dependent on targeting of circ-Foxo3 (Figure 6D).

Figure 6

miR-191-5p suppressed the pro-apoptotic effect of circ-Foxo3 in bladder cancer cells. (A) T24, UM-UC-3, and J82 bladder cancer cell lines were transfected with miR-191 mimic (10 nM) and treated with cisplatin (CP, 2 μg/mL). The expression of cleaved-caspase3 (c-caspase3) was examined by Western blot and quantified relative to GAPDH (n=5). (B–C) T24, UM-UC-3, and J82 bladder cancer cell lines were co-transfected with miR-191 mimic and a circ-Foxo3 over-expression plasmid (n=5). The expression of cleaved-caspase3 (c-caspase3) was examined by Western blot (B) and apoptosis was analyzed by flow cytometry (C). (D) Cell viability measured by the CCK-8 assay in T24, UM-UC-3, and J82 bladder cancer cell lines receiving miR-191 mimic or circ-Foxo3 transfection (n=5). Mean ± SD, *p<0.05, **p<0.01.

Discussion

The primary finding of our study is that circ-Foxo3 regulated bladder cancer growth in vivo and in vitro. The expression of circ-Foxo3 was lower in bladder cancer tissues and bladder cancer cell lines, and was upregulated in response to apoptotic stress. Overexpression of circ-Foxo3 promoted bladder cancer apoptosis in vivo and in vitro, which was partially attributable to the direct targeting of circ-Foxo3 by miR-191.

Our study is the first to identify a critical role of a circRNA in bladder cancer growth. Several recent studies reported that circ-Foxo3 regulated progression and proliferation of breast cancer,11,25 lung cancer,12 and gastric cancer.26 Our current data showed circ-Foxo3 was also dysregulated in bladder cancer and that the dysregulation of circ-Foxo3 could potentially contribute to various solid tumors as a common pathogenic factor.

To date, the detailed mechanism by which circ-Foxo3 regulates tumor progression has not be well understood. Several prior reports suggest the mechanism appears to be multifold. circ-Foxo3 has been shown to increase cellular senescence,14 to arrest cancer cell cycle progression by binding to the cell cycle proteins CDK2 and p21, and to inhibit angiogenesis.25 Our data showed in bladder cancer tissue and bladder cancer cell lines, one major effect of circ-Foxo3 was the induction of apoptosis, indicated by increased cleaved-caspase3 and Bax/Bcl2 ratio. Our results was consistent with the pro-apoptotic effect of circ-Foxo3 in breast carcinoma biopsies and in multiple cancer cell lines.13 Importantly, in breast cancer cells, circ-Foxo3 also increased Foxo3 protein level and promoted p53 ubiquitination and subsequent degradation, suggesting an alternative pathway by which circ-Foxo3 induces cancer cell apoptosis.13

The role of non-coding circRNAs as potential microRNA “sponges” have been recognized.8,27–31 Compared with most linear RNAs, circRNAs have no 3ʹ ends and are therefore more resistant to exonuclease degradation.8 In addition, circRNAs are characterized by high sponging capacity (sequence containing multiple microRNA binding sites) and relatively high expression level. All these characteristics indicate circRNAs may be more effective microRNA-sponges than the linear non-coding RNAs. The sponging effect of circRNAs appears to be microRNA- and tissue-specific: certain circRNAs only “sponge up” microRNAs of a particular family in particular tissues.32 However, the exact microRNA-sponging spectrum of circ-Foxo3 has been controversial. For instance, in non-small cell lung cancer, circ-Foxo3 sequesters miR-155;12 in breast cancer, circ-Foxo3 sequesters eight miRNAs.25 We found for the first time that in bladder cancer cells, there was a strong negative correlation between the expression of circ-Foxo3 and miR-191-5p. Our experiments confirmed the 3ʹ-UTR of circ-Foxo3 was a direct binding site for miR-191, and the pro-apoptotic effect of circ-Foxo3 could be effectively blocked by miR-191 mimic. These results highlighted a critical role of the direct interaction between circ-Foxo3 and miR-191 in regulating bladder cancer apoptosis. Given that the aberrant expression of microRNAs in cancers are highly tissue-specific, this particular interaction between circ-Foxo3 and miR-191 may be indicative of a highly specific therapeutic target for bladder cancer.33,34

Our study has several limitations. First, there may be alternative pathways by which circ-Foxo3 promotes cancer apoptosis,13 which were not examined in our study. In addition, our focus of bladder-cancer-related microRNA is limited to miR-191 and non-exhaustive. There could potentially be many other functionally important microRNAs that interact with circ-Foxo3 in bladder cancer cells.

Conclusion

circ-Foxo3 regulated bladder cancer growth in vivo and in vitro. Overexpression of circ-Foxo3 promoted bladder cancer apoptosis, which was partially attributable to the direct targeting of circ-Foxo3 by miR-191.