CircZNF609 promotes bladder cancer progression and inhibits cisplatin sensitivity via miR-1200/CDC25B pathway

Circular RNAs (circRNAs) have been extensively studied in tumor development and treatment. CircZNF609 (hsa_circ_0000615) has been shown to serve as an oncogene in all kinds of solid tumors and may act as the novel biomarker in tumor diagnosis and therapy in tumor early diagnosis and therapy. However, the underlying character and mechanism of circZNF609 in cisplatin chemosensitivity and bladder cancer (BCa) development were unknown. The expression level of cell division cycle 25B (CDC25B), microRNA 1200 (miR-1200), and circZNF609 in BCa cells and tissues depended on quantitative real-time PCR (qRT-PCR). CDC25B protein level was assayed with Western blot. Functional assays in vitro and in vivo had been conducted to inspect the important role of circZNF609 on BCa progression and cisplatin chemosensitivity in BCa. RNA sequencing and online databases were used to predict the interactions among circZNF609, miR-1200, and CDC25B. Mechanistic exploration was confirmed by RNA pull-down assay, RNA fluorescence in situ hybridization (FISH) and Dual luciferase reporter assay. CircZNF609 expression was increased significantly in BCa cell lines and tissues. For BCa patients, increased expression of circZNF609 was correlated with a worse survival. In vitro and in vivo, enforced expression of circZNF609 enhanced BCa cells proliferation, migration, and cisplatin chemoresistance. Mechanistically, circZNF609 alleviated the inhibition effect on target CDC25B expression by sponging miR-1200. CircZNF609 promoted tumor growth through novel circZNF609/miR-1200/CDC25B axis, implying that circZNF609 has significant potential to act as a new diagnostic biomarker and therapeutic target in BCa. Enhancing cisplatin sensitivity is an important direction for bladder cancer management. 1. This research reveals that circZNF609 improves bladder cancer progression and inhibits cisplatin sensitivity by inducing G1/S cell cycle arrest via a novel miR-1200/CDC25B cascades. 2. CircZNF609 was confirmed associated with worse survival of bladder cancer patients. 3. CircZNF609 act as a prognostic biomarker for bladder cancer treatment.


Introduction
Bladder cancer (BCa), a common urological carcinoma type, has a high incidence and mortality on a global scale (Teoh et al. 2020).Non-muscle invasive bladder cancer (NMIBC) and muscle invasive bladder cancer (MIBC) are two types of bladder cancer (Tran et al. 2021), which account for approximately 80% and 20% of the incidence, respectively.In comparison with MIBC, which has a high mortality, approximately 90% of patients with NMIBC have a favorable 5-year survival rate (Knowles and Hurst 2015).However, patients with NMIBC are prone to recurrence (Knowles and Hurst 2015).For patients with MIBC, cisplatin-based neoadjuvant and adjuvant chemotherapy are first-line therapeutic strategies (Plimack et al. 2014).However, resistance to cisplatin-based chemotherapy results in a poor prognosis (Sagalowsky 2013).Hence, further research is needed to elucidate the internal mechanism of resistance to chemotherapy, which may offer new insights into BCa treatment.
Circular RNAs (circRNAs) are a new class of RNAs that are identified as back-splicing of exons or introns of parental pre-mRNAs (Memczak et al. 2013).To date, most circRNAs are exon-containing circRNAs and are located in the cytoplasm (Dong et al. 2017).CircRNAs have high stability and are evolutionarily conserved, cell-specific, and tissuespecific, as determined using bioinformatics analysis and high-throughput sequencing (Li et al. 2015).Since their discovery in 1970s, circRNAs have gained increasing attention for their potential clinical utility.CircRNAs could modulate the development, progression, relapse, and chemosensitivity in various malignant tumors (Xu et al. 2020).In the presence of microRNA (miRNA) binding sites, some circRNAs develop their roles by suppressing miRNA functions via sponge adsorption (Kristensen et al. 2018;Tang et al. 2021).Our studies have shown that circ-ITCH inhibits BCa development by sponging miR-224/miR-17 (Yang et al. 2018) and circ-Cdr1as increases BCa susceptibility to cisplatin via miR-1270/ APAF1 axis (Yuan et al. 2019).CircZNF609 has been extensively explored in various tumor types but not in BCa (Liu et al. 2021b;He et al. 2020), making it important for BCa treatment.
Cisplatin is a highly effective chemotherapy regimen for individuals with BCa.However, cisplatin-based chemotherapy would be effective in less than 50% of BCa patients (Sagalowsky 2013).Cell-cycle regulation is one of the major determinants of cisplatin chemosensitivity (Yang et al. 2018;Borst et al. 2014).When more cells are enriched in the cell division phase and fewer cells are enriched in the G1 phase, BCa cells become less susceptible to cisplatin chemotherapy (Donaldson et al. 1994;Shah and Schwartz 2001).Cell division cycle 25B (CDC25B) is a well-characterized cell cycle regulator that is required for the accelerated cell cycle transition (Boutros et al. 2007;Aressy and Ducommun 2008).The increased expression of CDC25B may promote tumor progression and predict a worse prognosis in all kinds of human neoplasia, including BCa (Z Zhang et al. 2014;Sur and Agrawal 2016), whereas the role of CDC25B in modulating cisplatin chemosensitivity remains unknown.
The main findings of the present research are as follows: (1) CircZNF609 increased in BCa cell lines and tissues and is associated with poor prognosis for patients with BCa; (2) enforced expression of circ-ZNF609 could accelerate BCa cells' proliferation and cisplatin resistance in vitro and in vivo via an unrevealed circZNF609/miR-1200/CDC25B axis.

Clinical samples and BCa cell lines
The BCa tissues used the present study were obtained from patients who received radical surgery at the first affiliated hospital of Nanjing Medical University.
BCa samples and their corresponding normal tissues (n = 48) were included between 2014 and 2019.All human-related tissues in this research were granted by The First Affiliated Hospital of Nanjing Medical University Ethics Committee.Meanwhile, all participants signed informed consent before any clinical specimens were collected.Liquid nitrogen was used to preserve samples.Each sample was scored by the same team of pathologists and subjected to an independent review by at least two pathologists.
MiR-1200 mimics and miRNA controls were obtained from GenePharma Co. (Shanghai, China).Based on the manufacturer's guidelines, the Lipofectamine 3000 kit (Invitrogen, USA) was used for transfection.

RNA isolation and quantitative real time-PCR (qRT-PCR)
In accordance with the manufacturer's protocol, total RNAs were separated from cells and tissues by using TRIzol reagent (Invitrogen, USA).HiScript II (Vazyme, China) was used to synthesize cDNA.The qRT-PCR assays of circRNA, miRNA, and mRNA were conducted on LightCycler 480 (Roche, USA) or StepOne Plus Real-Time PCR system (Applied Biosystems, USA).U6 or β-actin was used as a control.By calculating CT values, data were analyzed, and each experiment was repeated thrice.All PCR primers needed were obtained by TsingKe (TsingKe, Nanjing, China) (Table S1).
Protein extraction and Western blot RIPA buffer (Sigma, USA) was used to lyse the tissues or cells.Protein extractions were absorbed, and then protein concentration was tested by using bicinchoninic acid (BCA) assessment (Beyotime, China).Protein was isolated and transferred to polyvinylidene fluoride membranes (Millipore, USA) by using 10% SDS-PAGE.Membranes were first blocked in 5% skim milk, and then incubated with primary (Protech, USA) and secondary antibodies (Protech, USA).Chemiluminescence (Bio-Rad, USA) was used to identify signals, which were then evaluated using Image Lab Software.

Cell proliferation experiment
To determine the cell viability of T24 or BIU87 cells, we seeded the cells in 96-well plates with starting number of 2000 or 5000 cells, respectively.Cell counting kit-8 (CCK-8) assay (Dojindo, Japan) was applied to record cell viability every 24 h (24, 48, 72, and 96 h).The absorbance value (at 450 nm peak) was measured on a microplate reader (Tecan, Switzerland).
Cloning formation T24 or BIU87 BCa cells were planted in six-well plates (1000 or 1500 cells/well).After 2 weeks, the cells were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet.Image J (NIH, USA) was applied to visualize and calculate cell colonies.

Scratch wound healing assay
When T24 or BIU87 cells reached 90-100% confluence in six-well plates, the monolayer cells were scraped off with the tip of a 200 μl pipette and cultured in serum-free medium.Images of cells were captured every 24 h (24 and 48 h) by using a microscope (Olympus, Japan).As a control, images of BCa cells at 0 h were also obtained.The migration rate was determined using a formula based on the distance traveled by BCa cells divided by the initial width.

Transwell assays
To examine the migration ability, we seeded 200 µl of serum-free medium with 4 × 10 4 T24 or 6 × 10 4 BIU87 cells in the upper chamber, while 600 µl of culture medium with 20% FBS was added into the lower chamber (Corning, USA).After 24 h of cultivation, top chambers were fixed with 4% paraformaldehyde and then stained for 15 min with 0.5% crystal violet.Finally, the transwell chambers were photographed using an Olympus microscope.

Evaluation of cell cycle
To detect the distinct phases of cell cycle, we fixed the BCa cells in 75% ethanol and dyed them via cycle test and using the DNA reagent kit (BD Biosciences, USA).Then, by using flow cytometry (Beckman Coulter, USA), the percentage of various phases was enumerated and analyzed using ModFit software (Version 5.0).

IC50 determination
After an overnight incubation of 5000 BCa cells per well in a 96-well plate, various concentration gradients of cisplatin (128,64,50,40,32,16,8,4, 2, and 1 µM, TCA, Japan) were added.After 24 h of cultivation, the cell growth ability was detected using the CCK-8 method, and the IC50 value was obtained using the probit regression model.

RNA-fluorescence in situ hybridization (RNA-FISH) assay
Cy3-labeled circZNF609 and FAM-labeled miR-1200 probes were obtained from Genepharma (Shanghai, China).T24 cells were stained by relative probes by using the RNA-FISH kit (Genepharma, Shanghai).These images were captured from the Zeiss LSM880 NLO confocal microscope system (Leica Microsystems, Germany).
Biotin-coupled probe pull-down assay Biotinylated circZNF609 and miR-1200 (GenePharma, China) pull down assays were conducted.The biotinylated probe was synthesized to bind the junction area of circZNF609, with oligonucleotide probes as negative controls.A total of 10 7 cells were collected and then lysed using the lysis buffer.All lysed products were cultured at room temperature for 2 h with 3 μg biotinylated probes.
Afterward, streptavidin magnetic beads (50 μl) were incubated with the products generated in the previous step for 4 h.The RNA complexes attached to the beads were evaluated by qRT-PCR assay after being washed five times with lysis buffer and isolated using TRIzol reagent.

Luciferase reporter assay
In 96-well plates, 293 T cells were planted and cotransfected by using Lipofectamine3000 (Invitrogen, Foster city, CA, USA) with miR-1200 mimics and plasmids expressing wild-type or mutant fragments.After incubating the cells for 24 h, renilla and firefly luciferase activities were determined using dual luciferase reporter assay (Promega, USA).

Xenograft in nude mice
Approximately 1 × 10 7 T24 cells transfected with sh circZNF609, circZNF609 over expression, or their negative control were subcutaneously injected into the axilla of the athymic BALB/C nude mice (five mice/group, 18-22 g, 4-6 weeks old).Mice in experimental were received intraperitoneally injected by intraperitoneally injected cisplatin (2.5 mg/kg) every 3 days for 1 week following inoculation, while the mice in the control groups received saline injections.The volume (V) was determined each week by monitoring the width (W) and length (L) by using a caliper rule, and the calculation formula V = (W 2 × L)/2 was used.After 4 weeks, we euthanized all animals and quantified the tumor bulk.Ethical approval for all animal experiments was granted by the institutional animal welfare committee of Nanjing Medical University.

Immunohistochemistry (IHC)
Mouse tumor was fixed in paraffin and mounted on 4 μm slide each.Tumor slides were rehydrated by different concentrations of ethanol.Antigen was separated by microwave heating.The slides were soaked in 3% H 2 O 2 for 10 min and then treated with CDC25B antibody (1:500, Proteintech, USA) in a humidified incubator at 4 °C overnight.Then, the slides were incubated with HRP-conjugated antibody at room temperature for 30 min.The data were collected by using a microscope.

Data processing
Data was processed using SPSS (IBM, version 22.0) and were reported as mean ± standard deviation (means ± SD).Kaplan-Meier analysis was performed to plot the survival curves, and differences were analyzed using log-rank test.To analyze differences among each group, we used one-way ANOVA and Student's t-test.P values were two-sided, and statistical significance was considered at P value of 0.05.

Results
CircZNF609 expression was highly expressed in BCa tissues and cell lines, and this condition is related to histological grades and prognosis in BCa patients CircZNF609 was validated as a circular product by Sanger sequencing after amplified by PCR with divergent primers (Fig. 1A).We performed actinomycin D and RNase R treatment experiments to establish that circZNF609 was more stable than linear ZNF609 (Figs. 1B-C).CircZNF609 expression was remarkably over-expressed in BCa cell lines or tissues compared with adjacent normal tissues or SV-HUC, an immortalized, but non-tumorigenic urothelial cell line (Figs.1D-E).Furthermore, we observed a positive association between circZNF609 expression and histological grade (P = 0.040).No significant correlation was observed between circZNF609 and other clinicopathological parameters, such as age, gender, tumor size, histological grade, and tumor node metastasis (TNM) stage (Table 1).Interestingly, a near-significant "gender"-dependent association was observed, in which males are associated with high levels of circ-ZNF609, whereas females were associated with low levels.We then further explored the "Age + Gender" group and "Tumor size + Gender" groups.However, considering the limited sample size, no significant differences were observed (Table S2, Table S3).Additionally, prognostic indicates uncovered that high circZNF609 expression level results in poor overall survival for patients with BCa (Fig. 1F).
Fig. 4 MiR-1200 inhibits the progression and cisplatin resistance of BCa cells.A CCK8 assay showed miR-1200 could inhibit T24 and BIU87 cells proliferation (**P < 0.01, ***P < 0.001, Student's t-test).B Colony formation assay indicated miR-1200 could inhibit the colony numbers of T24 and BIU87 cells (**P < 0.01, Student's t-test).C Flow cytometry assays were applied to investigate the function of miR-1200 in T24 cells.D More T24 cells were distributed in G1 phase and less T24 cells were distributed in S phase when miR-1200 was overexpression (*P < 0.05, **P < 0.01, Student's t-test).E Flow cytometry assays were applied to investigate the function of miR-1200 in BIU87 cells.F More BIU87 cells were distributed in G1 phase and less BIU87 cells were distributed in S phase when miR-1200 was overexpression (*P < 0.05, Student's t-test).G IC50 assays showed that miR-1200 improved sensitivity of T24 cells to cisplatin (**P < 0.01, Student's t-test).H IC50 assays showed that miR-1200 improved sensitivity of BIU87 cells to cisplatin (***P < 0.001, Student's t-test).Data are mean ± SD, n = 3

CircZNF609 promoted the cisplatin resistance and process of BCa cells in vitro
To investigate the function of circZNF609 in BCa cells, we constructed circZNF609 overexpression and knockdown in low-potential-malignancy BCa cell line BIU87 and high-potential-malignancy BCa cell line T24 (Figure S1A-D).Further functional experiments were performed to explore the effect of circZNF609 in BCa.The proliferation (cell number) of T24 and BIU87 BCa cells was assessed using CCK-8 assay (Figs.2A-B).The results of cloning formation assays Vol.: (0123456789) revealed a positive association between circZNF609 expression and colony numbers (Figs.2C-D).When the expression level of circZNF609 decreased, more cells were distributed in the G1 phase and less in the S phase (Figs.2E-F), and vice versa (Figs.2G-H).In cisplatin treatment experiments, cell viability curves were applied to calculate the IC50 value of cisplatin (Figures S2A-B).The IC50 value of cisplatin was decreased in T24 and BIU87 cells when circZNF609 was reduced by knockdown (Figs.2I-J), but a high IC50 value was obtained when circZNF609 was overexpressed (Figs.2K-L).Additionally, a positive correlation was observed between circZNF609 and T24 and BIU87 cell migration during scratch wound healing and transwell assays (Figures S3A-C).
In 48 pairs of BCa tissues, adjacent normal tissues expressed higher levels of miR-1200 compared with tumor tissues by qRT-PCR (Fig. 3G).A negative relationship between circZNF609 and miR-1200 was discovered by Pearson correlation analysis.(Fig. 3H).Patients with high level of miR-1200 expression had a favorable prognosis (Fig. 3I).

MiR-1200 suppressed cisplatin resistance and the progression of BCa cells in in vitro assays
Functional experiments were conducted to confirm the role of miR-1200 in BCa cells.CCK8 (Fig. 4A) and cloning assays (Fig. 4B) demonstrated that miR-1200 suppressed the growth ability of T24 and BIU87 cells.When miR-1200 was overexpressed, flow cytometry results showed that more T24 or BIU87 cells were distributed in G1 phase and less in S phase (Figs.4C-F).MiR-1200 could also increase the cisplatin chemosensitivity of T24 and BIU87 cells (Fig. 4G-H).Moreover, miR-1200 could suppress the migration ability of T24 and BIU87 cell based on scratch wound healing and transwell assays (Figures S3A-C).

Interference of miR-1200 rescued the intensive process and depressed cisplatin sensitivity activated by circZNF609 in BCa cells
Rescue assays were conducted by co-transfecting circZNF609 overexpression lentivirus and miR-1200 mimics into T24 and BIU87 BCa cells.The results of colony and CCK-8 assays demonstrated that circ-ZNF609 overexpression could promote the proliferation of BCa cells, whereas ectopic miR-1200 could repress this effect (Figs. 5A-B).Flow cytometry results indicated that when circZNF609 was overexpressed, less cells were observed in the G1 phase and more cells were observed in the S phase.By Fig. 5 Co-transfection of miR-1200 mimic eliminated the oncogenic function of circZNF609.A CCK-8 experiments indicated that co-transfection with miR-1200 mimics could reverse proliferation-promoting function of circZNF609 in T24 and BIU87 cells (***P < 0.001, Student's t-test).B Cloning formation assays revealed that co-transfection with miR-1200 mimics could reverse the enhanced proliferation induced by circZNF609 in T24 and BIU87 cells (**P < 0.01, ***P < 0.001, Student's t-test).C-D Illustrations of rescued flow cytometry assays in T24 and BIU87 cells.E Less cells were dispersed in G1 phase and more cells were distributed in S phase when circZNF609 was overexpression in T24 cells, and miR-1200 mimic transfection could reverse this phenomenon (*P < 0.05, **P < 0.01, Student's t-test).F Less cells were dispersed in G1 phase and more cells were distributed in S phase when circZNF609 was overexpression in BIU87 cells, and miR-1200 mimic transfection could reverse this oncogenic function (**P < 0.01, ***P < 0.001, Student's t-test).G The IC50 assays indicated that circZNF609 overexpression inhibited sensitivity of T24 cells to cisplatin, while co-transfection of miR-1200 mimics could reverse it in T24 cells (*P < 0.05, **P < 0.01, Student's t-test).H The IC50 assays indicated that circZNF609 overexpression inhibited sensitivity of BIU87 cells to cisplatin, while co-transfection of miR-1200 mimics could reverse it in BIU87 cells (*P < 0.05, **P < 0.01, Student's t-test).Data are mean ± SD, n = 3 ◂ Vol:.( 1234567890) co-transfecting miR-1200 mimics in cells overexpressing circZNF609, miR-1200 could rescue both the cell cycle promotion triggered by circZNF609 (Fig. 5C-F) and the cisplatin resistance induced by circZNF609 (Fig. 5G-H).Additionally, ectopic miR-1200 inhibited the circZNF609-induced increase in migration rates during scratch wound healing and transwell assays (Figures S5A-B).
Vol.: (0123456789) CircZNF609 improved the expression of CDC25B by sponging miR-1200 mRNA sequencing was performed on three pairs of circZNF609 overexpression and relative control T24 BCa cells.Subsequently, Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were conducted to identify related pathways and genes affected by circZNF609.GO analysis indicated that circZNF609 expression was related to cell population proliferation and cell cycle process functions (Fig. 6A).KEGG pathway analysis revealed a correlation between circZNF609 and cancer pathways (Fig. 6B).The prediction databases Targetscan (http:// www.targe tscan.org/ vert_ 72/) and miRDB (http:// mirdb.org/) were used to determine the downstream genes bound to miR-1200.CDC25B was screened as a downstream target gene based on the combined database prediction and mRNA sequencing results (Fig. 6C).
To validate the binding of miR-1200 and CDC25B, we conducted dual luciferase reporter assay.The reporter plasmids contained mutant and wide-type 3′-UTR of CDC25B.Co-transfection of wild-type CDC25B reporter plasmids and miR-1200 mimics dramatically suppressed luciferase vitality, whereas co-transfection of mutated CDC25B plasmids and miR-1200 mimics showed no significant effect (Fig. 6D).
Next, qRT-PCR was conducted to assess the expression signature of CDC25B in 48 pairs of BCa tissues.In comparison with normal tissues, CDC25B was highly expressed in BCa tissues (Fig. 6E).
Patients that express higher level of CDC25B expression suffered from a poor survival rate (Fig. 6F).Furthermore, circZNF609 could enhance CDC25B expression at both mRNA and protein levels (Figs.6G-J), while miR-1200 could inhibit CDC25B expression (Figures S6A-B).Additionally, the transfection of miR-1200 mimics in BCa cells that overexpress circZNF609 rescued the circZNF609-induced high expression of CDC25B (Figs. 6K-L).
Functional experiments were carried out to prove the effect of CDC25B during BCa progression.First, we transfected the short CDC25B interfering RNAs to inhibit its expression (Figures S7A-B).CCK8 and cloning formation assays suggested that CDC25B could increase the proliferation of BCa cells .Transwell and scratch wound healing assays demonstrated that CDC25B could increase the migration rate of BCa cells (Figures S7E-F).Furthermore, CDC25B could induce more BCa cells dispersed in S phase and inhibit cisplatin sensitivity (Figures S7G-J).

CircZNF609 promoted the growth and cisplatin resistance of xenografted tumor in vivo
To confirm whether circZNF609 affected the in vivo growth ability of tumor, we subcutaneously injected (right forelimb axilla) of T24 cells transfected with circZNF609 overexpression and relative control vector into nude mice.The tumors were removed after 4 weeks (Figs.7A-B).CircZNF609 overexpression remarkably affected the tumor volume and weight growth (Figs.7C-D).IHC experiments indicated that CDC25B levels were positively related to circ-ZNF609 (Fig. 7E).
To confirm whether circZNF609 could influence the cisplatin sensitivity in vivo, we subcutaneously injected (right forelimb axilla) si-circZNF609 transfected and negative control transfected T24 cells into nude mice.The mice in the experimental groups were intraperitoneally injected with cisplatin (2.5 mg per kilogram) every 3 days for 1 week after inoculation, while the mice in the control group received saline Fig. 6 MiR-1200 inhibited BCa progression and cisplatin resistance via targeting CDC25B.A GO terms analysis of mRNA sequencing.B KEGG analysis of mRNA sequencing.C Downstream target CDC25B and binding sites to miR-1200 were predicted by overlapping of database and mRNA sequencing.D Dual luciferase reporter assays were applied to confirm the binding of miR-1200 and CDC25B (***P < 0.001, Student's t-test).E The expression level of CDC25B in 48 pairs of BCa tissues was investigated by qRT-PCR.F The overall survival of BCa patients with different CDC25B expression level was investigated by Kaplan-Meier analysis.G CircZNF609 knockdown decreased the mRNA level of CDC25B in T24 and BIU87 cells (**P < 0.01, Student's t-test).H CircZNF609 knockdown decreased the protein level of CDC25B in T24 and BIU87 cells.I Overexpression of circ-ZNF609 improved the mRNA level of CDC25B in T24 and BIU87 cells (***P < 0.001, Student's t-test).J Overexpression of circZNF609 improved the protein level of CDC25B in T24 and BIU87 cells.K Overexpression of circZNF609 improved the mRNA level of CDC25B in T24 and BIU87 cells, while co-transfection of miR-1200 mimic could reverse it (*P < 0.05, **P < 0.01, ***P < 0.001, Student's t-test).L Overexpression of circZNF609 improved the protein level of CDC25B in T24 and BIU87 cells, while co-transfection of miR-1200 mimic could reverse it.Data are mean ± SD, n = 3 ◂ Vol:.( 1234567890) injection.The tumors were removed after 4 weeks (Figs.7F-G).In comparison with the saline-treated mice, tumor weight and volume in cisplatin-treated mice remarkably decreased (Figs.7H-I).Tumor volume and weight were lower in si-circZNF609 group treated with cisplatin than in the control group treated with cisplatin (Figs.7H-I).

Discussion
The development and progression of BCa is a complex process that has not been fully elucidated.Moreover, the emergence of chemotherapy resistance has complicated BCa treatment.The increasing number of circRNAs is critical to the incidence and treatment of BCa (Yang et al. 2018) (Wei Dong et al. 2019) (Zhang et al. 2021).In the present study, we discovered that patients with BCa who had elevated circ-ZNF609 levels had a higher histological grade and a worse survival.Moreover, circZNF609 facilitated BCa cells' progression and inhibited BCa cells' chemosensitivity to cisplatin via a novel circZNF609/ miR-1200/CDC25B axis in vitro and in vivo.
CircRNAs were characterized as single-stranded and covalently closed, which were created via backsplicing from pre-mRNAs (Capel et al. 1993).Cir-cRNAs are very stable, evolutionary conserved, and tissue-specific (Li et al. 2015).Considering these distinguishing features, circRNAs are increasingly being highlighted as a new candidate for regulating the expression of oncogenes or tumor suppressor genes (Rybak-Wolf et al. 2015;Salzman et al. 2013).
Novel and effective treatment targets should be developed to prevent tumor progression.Numerous circRNAs with critical regulatory roles in carcinogenesis and progression tremendous therapeutic potential (Xu et al. 2020).In BCa, circLIFR could attenuate cisplatin chemoresistance via synergizing with MSH2 (Zhang et al. 2021).CircELP3 which is induced by hypoxia, could additionally inhibit cisplatin chemosensitivity in BCa (Su et al. 2019).Moreover, circ-Cdr1as increases BCa's susceptibility to cisplatin via miR-1270/APAF1 axis (Yuan et al. 2019).Therefore, circRNAs have a regulatory role in cancer therapy.
Cisplatin has performed an essential function in chemotherapy for BCa.However, cisplatin resistance greatly harmed the long-term survival of BCa patient (Yuan et al. 2019).The present study demonstrated for the first time that circZNF609 could induce the cisplatin resistance of BCa.Additional mechanistic investigations demonstrated that circZNF609 enhanced cisplatin resistance by increasing CDC25B expression via sponging miR-1200.CDC25B, as a cell cycle regulator, can alter the distribution of cell cycle phases in tumor cells (Boutros et al. 2007;Aressy and Ducommun 2008).Cisplatin is more toxic to tumor cells in the G1 phase (Donaldson et al. 1994;Shah and Schwartz 2001).Our results showed that increasing the CDC25B expression reduced the amount of BCa cells during G1 cell cycle phase and decreased cisplatin chemosensitivity.CDC25B has an oncogenic potential (Yan et al. 2008;Wang et al. 2020;Li et al. 2019;Jia et al. 2021).Reduced CDC25B expression could inhibit the progression of liver cancer (Yan et al. 2008).CircRNA_102958 accelerated the carcinogenesis of colorectal cancer by increasing CDC25B expression (Li et al. 2019).Moreover, lncRNA FAM83A-AS1 could promote esophageal cell squamous carcinoma progression by increasing CDC25B expression (Jia et al. 2021).
The circZNF609-mediated function network will provide new avenues for BCa treatment (Fig. 8).Fig. 7 CircZNF609 promoted BCa cells proliferation and cisplatin resistance in nude mice.A Picture of nude mice implanted with circZNF609 overexpression or negative control T24 cells (n = 5).B Image of tumor formation in nude mice implanted with circZNF609 overexpression or negative control T24 cells (n = 5).C Volumes of xenograft tumors were measured by electronic scales every week (**P < 0.01, Student's t-test).D Tumor weights of nude mice were measured by electronic scales (***P < 0.001, Student's t-test).E Immunohistochemistry assays were applied to detect the CDC25B in two groups of xenograft tumors.F Picture of the nude mice implanted with circZNF609 knockdown or negative control T24 cells, which injected with cisplatin or saline (n = 5).G Image of xenograft tumor of nude mice implanted with circZNF609 knockdown or negative control T24 cells, which injected with cisplatin or saline (n = 5).H Tumor volumes of nude mice in four groups were measured every week (**P < 0.01, ***P < 0.001, Student's t-test).I Tumor weights of four groups were measured by electronic scales (*P < 0.05, **P < 0.01, ***P < 0.001, Student's t-test).Data are mean ± SD, n = 3 ◂ Additionally, the inhibitors of this network may serve as a potential adjuvant for cisplatin-based chemotherapy in BCa.

Fig. 1
Fig. 1 CircZNF609 was up-regulated in BCa tissues and cell lines and was positively associated with poor survival.A Schematic illustration indicated circZNF609 was formed by circularization of ZNF609 exon 2 and Sanger sequencing confirmed the junction of back splicing.B qRT-PCR confirmed the remaining RNA levels of circZNF609 and ZNF609 mRNA levels in T24 cells following actinomycin D treatment at different time points.C qRT-PCR determined the RNA levels

Fig. 3
Fig. 3 CircZNF609 sponged miR-1200 mostly in BCa cells.A Schematic illustration indicated overlapping of the target miRNAs of circZNF609 predicted by regRNA, miRanda and RNAhybrid.B CircZNF609 pull down assay showed miR-1200 combined with circZNF609 significantly compared with other potential miRNAs (**P < 0.01, Student's t-test).C MiR-1200 pull down assay indicated that biotin coupled miR-1200 could captured more circZNF609 compared with biotin coupled NC (**P < 0.01, Student's t-test).D Dual luciferase reporter assays showed circZNF609 bound with miR-1200 most significantly compared with other potential miRNAs.E Dual luciferase reporter assays confirmed that miR-1200 were

Table 1
Associations between the expression level of circ-ZNF609 and clinicopathological features in BCa patients * P < 0.05