Biological information was obtained to investigate the mRNA levels of KIF22 in tumor and normal tissues and investigate the association between KIF2C and patient prognosis. Data on survival rates were obtained from The Cancer Genome Atlas (TCGA) database. Gene Expression Profiling Interactive Analysis (http://gepia.cancer-pku.cn/detail.php?gene=KIF22/) was used to collate and analyze TCGA (https://www.cancer.gov/about-nci/organization/ccg/research/structural-genomics/tcga) data with a threshold of P<0.05 and LogFC>1 or <−1 for differential genes; the median was used as the basis for dividing patients into two groups: i) The high expression, or ii) low expression groups for Kaplan-Meier survival analysis. The log rank test was used to determine any statistically significant differences in patient survival.

The following antibodies were used in the present study: Anti-KIF22 [1:200 dilution for immunohistochemistry (IHC), 1:2,000 dilution for western blot analysis and 1:50 dilution for chromatin immunoprecipitation (ChIP) assay; MA5-15912; Invitrogen; Thermo Fisher Scientfic, Inc.], anti-CDCA3 (1:200 dilution; ab166902; Abcam), anti-β-actin (1:2,000 dilution; 60008-1-Ig; ProteinTech Group, Inc.), anti-Ki67 (1:1,000 dilution; 27309-1-AP; ProteinTech Group, Inc.), anti-proliferating cell nuclear antigen (PCNA) (1:500 dilution; SAB2108448; Sigma-Aldrich; Merck KGaA), anti-cyclin D1 (1:1,000 dilution; ab16663, Abcam) and anti-cyclin A2 (1:1,000 dilution; ab181591; Abcam).

The primer sequences used for reverse transcription-quantitative PCR (RT-qPCR) were as follows: KIF22 forward, 5′-GAT CTC AGG AGC TGG TCG C-3′ and reverse, 5′-GTT CCA TCC ACA AAT GGC CG-3′; CDCA3 forward, 5′-TGG TAT TGC ACG GAC ACC TA-3′ and reverse, 5′-TGT TTC ACC AGT GGG CTT G-3′; and GAPDH forward, 5′-CGA CCA CTT TGT CAA GCT CA-3′ and reverse, 5′-GGT TGA GCA CAG GGT ACT TTA TT-3′.

The shRNA clone of KIF22 was purchased from Addgene, Inc. The pcDNA3.1-KIF22, pcDNA-CDCA3 and pGL-CDCA3 plasmids were constructed in the laboratory of Department of Urology of Tianjin Third Central Hospital Affiliated to Nankai University. The pcDNA3.1-vector served as the control of pcDNA3.1-KIF22 and pcDNA3.1-CDCA3, and the pGL-vector served as the control of pGL-CDCA3. The scrambled control plasmid (5′-ATG GTA CTG ACC TCC AGA G-3′) was used as the negative control (NC). The method used to harvest viral supernatant was through ultracentrifugation (600 × g, 5 min, 4°C). A total of 1×10⁵ T24 or 5637 cells were seeded into 6-well plates and 0.5 µg plasmids were used. The cells were transfected using 10 µl Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) in each well. Following incubation for 20 min at 20°C, the transfection was completed. The efficiency was measured using both reverse transcription-quantitative PCR (RT-qPCR) and western blot analysis after 48 h. Two shRNAs were initially used to avoid off-target effects, and the one with a higher silencing efficiency was selected for use in subsequent experiments in vitro and in vivo. The high silencing efficiency shRNA sequence of KIF22 was 5′-AAG CAA GAT TGG AGC TAC TCG TC-3′. The shRNA sequence of the negative control was 5′-CTT GGA GAA TGA GGC AGG GCA GA-3′.

A total of 131 human bladder cancer tissues were obtained from Tianjin Third Central Hospital (Tianjin, China) from patients who underwent transurethral resection of the bladder tumor. All procedures in this experiment were approved and conducted in accordance with the standards upheld by the Ethics Committee of Tianjin Third Central Hospital Affiliated to Nankai University. All patients signed informed consent.

The samples were fixed with 10% formalin for 24 h at 98°C, embedded with resin (Epoxy resin; Sigma-Aldrich; Merck KGaA), and divided into 5-µm-thick sections. The sections were dewaxed with xylene at 65°C, then rehydrated in a gradient ethanol series. The samples were immersed in citrate buffer (pH 6.0) at 98°C for 30 min and placed in a microwave for incubation for 10 min for antigen retrieval at 20°C. Hydrogen peroxide was then added to block endogenous peroxidase activity and the samples were incubated at 20°C for 10 min, followed by blocking with 2% BSA (Sigma-Aldrich; Merck KGaA) for 20 min at room temperature. Subsequently, the samples were incubated with the primary antibody of KIF22 and CDCA3 at room temperature for 2 h. Finally, the samples were washed with PBS four times and incubated with the secondary antibody (anti-Rabbit HRP; 1:200 dilution; ab205718; Abcam). Diaminobenzidine was used as a chromogen substrate. Images were captured using an Olympus inverted fluorescence microscope (IX71; Carl Zeiss AG).

The KIF22 protein is located in both the cytoplasm and nucleus of bladder cancer tissues (26). The expression level of KIF22 was classified into four groups based on the staining intensity (0, negative; 1, low; 2, medium; and 3, high). Additionally, the proportion of stained cells was as follows: 0, 0% stained cells; 1,1-25% stained cells; 2, 26-50% stained cells; and 3, 51-100% stained cells. A staining intensity score x the score of the percentage of stained cells <2 was considered as weak staining, 2-3 as moderate staining and >4 as strong staining.

A staining index (score, 0-12) was determined by multiplying the score for the positive area and the staining intensity. The expression of CDCA3 was scored as follows: 0, negative; 1, weak; 2, moderate; 3, strong). The property of positive cells was defined as follows: 0, <5% positively stained cells; 1, 5-25; 2, 26-50; 3, 51-75; and 4, >75% positively stained cells). A score of 0 was considered negative, scores of 1-6 were considered low expression and scores of 7-12 were considered as a high CDCA3 expression.

The sections of each patient were observed within five visual fields, and an experienced pathologist examined the sections.

The T24 and 5637 human bladder cancer cell lines were purchased from ATCC. Both cells were maintained in RPMI-1640 culture medium, supplemented with 10% of fetal bovine serum and incubated at 37°C in a 5% CO₂ incubator.

A total of 0.5 µg control or KIF22 shRNA plasmids were transfected into the bladder cancer cells using Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.). The cells in the shControl (shNC) group were transfected with a negative control plasmid, and those in the shKIF22 group were transfected with a KIF22 shRNA plasmid. After 48 h, the subsequent assays were performed. The stable KIF22 knockdown cell line was screened using lentivirus infection and used in the in vivo assays.

Total RNA was extracted from the T24 and 5637 cells using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.). Subsequently, the total RNA was reverse transcribed using M-MLV reverse transcriptase at 42°C for 60 min (Promega Corporation). qPCR was conducted using the SYBR PrimeScript RT-PCR kit II (cat. no. DRR083; Takara Biotechnology Co., Ltd.) and the relative expression levels of KIF22 was normalized to the mRNA expression levels of β-actin. The following thermocycling conditions were used: Initial denaturation at 95°C for 3 min; followed by 30 cycles of denaturation at 95°C for 30 sec, annealing at 58°C for 30 sec and extension at 72°C for 30 sec. The 2^−ΔΔCq method was used to quantify the results (27). The relative expression level of KIF22 was normalized to GAPDH. The primer sequences used for RT-qPCR are described above.

Bladder cancer cells or tissue samples were lysed with lysis buffer (60 mM Tris-HCl, pH 6.8, 2% SDS, 20% glycerol, 0.25% bromophenol blue, 1.25% 2-mercaptoethanol and protease inhibitor cocktail, Beyotime Institute of Biotechnology). Total protein was separated by 10% SDS-PAGE and sequentially transferred onto PVDF membranes (IPSN07852; EMD Millipore). The PVDF membranes were then blocked with 5% dry milk at room temperature for 2 h in TBST buffer and subsequently incubated with the primary antibodies, including KIF22, Ki67, PCNA, cyclin D1, cyclin A2, or β-actin antibody for 2 h at room temperature. After washing with TBST 3 times, the membranes were incubated with secondary antibody (rabbit; 1:5,000 dilution; cat. no. ab205718; Abcam) for 45 min at room temperature. Each blot was subsequently visualized with the use of an ECL kit (RPN 2109; Cytiva). The blot intensity was analyzed using ImageJ 9.0 software (National Institutes of Health).

The T24 and 5637 cells were re-suspended and plated into 6-well plates at a density of 2,000 cells/well and grown for 2 weeks. The colonies were then fixed with methanol at −20°C for 5 min and stained with 0.1% crystal violet for 20 min. Colonies were then photographed using an Olympus inverted fluorescence microscope and images captured (IX71; Carl Zeiss AG) and the differences in colony numbers between the control and KIF22-silenced bladder cancer cells were calculated. A colony was counted when it included >100 cells. The number of colonies per visual field area visible under an Olympus inverted fluorescence microscope (IX71; Carl Zeiss AG) was counted.

Both bladder cancer cells were plated in 96-well plates at a density of 5×10⁴ cells/well and cultured for 48 h (MTT) or 5 days (CCK-8) at 37°C. For the MTT assays, the cells were then treated with MTT (Beyotime Institute of Biotechnology) for 3 h and washed with PBS. Cells were then extracted using 150 µl DMSO and the absorbance value at a wavelength of 570 nm was measured and analyzed (28). For CCK-8 assays, the cells were then treated with CCK-8 (Beyotime, China) for 2 h and the absorbance value at a wave-length of 490 nm was measured using a Multiscan Spectrum (K3; Thermo Fisher Scientific, Inc.).

Following transfection for 48 h, the cells were collected and washed with PBS twice. The cells were then fixed with precooled in 70% ethanol at −20°C for 1 h. Subsequently, the fixed cells were washed with PBS twice and subjected to RNase I (BD Biosciences) treatment at 37°C for 30 min. Finally, the cells were stained with propidium iodide (PI, 200 µg/ml; BD Biosciences) at 4°C for a further 30 min and analyzed using a BD FACSCalibur™ flow cytometer (BD Biosciences).

The experiments involving animals were approved by the Institutional Animal Care and Use Committee (IACUC) of the Tianjin Third Central Hospital Affiliated to Nankai University (approval no. SYXK 2019-0318). Nude BALB/c mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. A total of 16 male nude BalB/c mice (8 in each group, 8-weeks-old; 18-22 g) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd., and fed with food and water ad libitum and at specific pathogen-free conditions (20°C; 60% humidity and alternating 12-h light/dark cycles).

For the tumor growth assay, T24 cells stably transfected with control or KIF22 shRNA lentivirus were subcutaneously injected into the right flanks of female nude mice. Almost 2 weeks later, tumors (150 mm³) were established, and the tumor volume was measured each week and calculated [length x (width)²/2]. The mice were euthanized by an intraperitoneal injection of 120 mg/kg sodium pentobarbital before the tumors were removed at the 49-day time point. The hearts of the mice were then monitored, and death was confirmed by cardiac arrest. Tumor growth curves were plotted according to the tumor volume in the different groups.

ChIP assay was performed using a ChIP assay kit (ab500; Abcam). T24 cells (~10⁸) were cross-linked with 1% formaldehyde (Sigma-Aldrich; Merck KGaA), resuspended and lysed by RIPA buffer (Beyotime Institute of Biotechnology), then sonicated to shear the DNA into a range of 500-1,000 bp. The DNA and protein complex were then immunoprecipitated with anti-KIF22 antibody for 2 h at room temperature, and the complex was enriched using protein A agarose (Beyotime Institute of Biotechnology). Magnetic beads (Beyotime Institute of Biotechnology) were isolated and washed. Isolated DNA was further purified using a QIAquick PCR Purification kit (cat. no. 28104, Qiagen, Inc.) and amplified using AmpliTaq Gold^® 360 Master Mix (Life Technologies; Thermo Fisher Scientific, Inc.).

Luciferase assay was performed using the luciferase assay system as per the manufacturer's instructions (E1500, Promega Corporation) to detect the activities of the promoter of the CDCA3 gene from which the 3′ untranslated regions (UTRs) were obtained. Briefly, T24 cells were cultured and transfected with pGL-CDCA3, pGL-Basic and pcDNA3.1 plasmids (0.5 µg) overnight using Lipofectamine^® 3000 (10 µl; Invitrogen; Thermo Fisher Scientific, Inc.). Following transfection for 48 h, the cells were washed and the luciferase activities were measured following the addition of prepared solutions. The relative luciferase activities were calculated by normalizing the Firefly luciferase activity to Renilla luciferase activity.

GraphPad 5.0 software (GraphPad Software, Inc.), was used to perform the statistical analysis in the present study. Data are represented as the mean ± SEM. The statistical significance of the differences between two groups was analyzed using a unpaired Student's t-test. The statistical significance of the difference among more than two groups was analyzed using one-way ANOVA and a Turkey's post hoc test. A value of P<0.05 was considered to indicate a statistically significant difference. Kaplan-Meier survival analysis with the log-rank test was performed to assess patient prognosis, and the Chi-squared test (χ² test) was performed to assess the association between protein expression levels and the clinical features of patients. Pearson's correlation coefficient (Pearson's R) was used to analyze the correlation between the expression of KIF22 and CDCA3 in bladder cancer tissues.

To investigate the role of KIF22 in the development of bladder cancer, the present study used tumor and adjacent tissue samples from 131 patients with surgically treated bladder cancer to detect its expression. IHC assays were conducted to detect the expression of KIF22. It was found KIF22 was localized and distributed in the cytoplasm and nucleus of the bladder cancer tissues (Fig. 1A). However, the expression of KIF22 was evidently low in the normal adjacent tissues (Fig. 1B), compared with that in the tumor tissues. These findings demonstrated the high expression of KIF22 in human bladder cancer tissues, suggesting a potential link between KIF22 and bladder cancer.

Subsequently, according to the staining intensity of KIF22 expression in the bladder tumor tissues, the samples were classified into two groups as follows: The KIF22-low (n=44) and KIF22-high (n=87; Fig. 1A) expression groups. The differences in the clinicopathological characteristics of the patients between KIF22-high and KIF22-low groups were then analyzed. The data demonstrated that KIF22 expression in bladder cancer was significantly associated with tumor stage (P=0.003) and recurrence (P=0.016), whereas no obvious association was revealed between KIF22 and other clinical features, such as patient age (P=0.422), gender (P=0.856), tumor grade (P=0.198) and lymph node metastasis (P=0.354; Table I).

Through bioinformatics analysis, it was found that KIF22 was highly expressed in bladder cancer clinical samples (Fig. 1B), consistent with the authors' expectations. In addition, by performing Kaplan-Meier analysis, it was revealed that patients with a low expression of KIF22 had a higher overall survival and disease-free survival rate, compared with those with a high KIF22 expression (Fig. 1C). Collectively, these results confirmed that KIF22 was associated with the clinical characteristics and the prognosis of patients with bladder cancer.

The present study then investigated whether KIF22 affects bladder cancer progression via the regulation of cell proliferation. KIF22 expression was silenced using KIF22 shRNA plasmids in the T24 and 5637 human bladder cancer cell lines. The silencing efficiency of KIF22 shRNA was determined using RT-qPCR and western blot analysis. The results revealed that KIF22 expression in the KIF22-shRNA-transfected T24 and 5637 bladder cancer cells was significantly decreased (Fig. 2).

On this basis, the role of KIF22 in the proliferation of bladder cancer was explored using colony formation and MTT assays. The results revealed significantly decreased colony numbers in the KIF22-silenced T24 and 5637 cells (Fig. 3A and B). Furthermore, MTT assays revealed KIF22 silencing led to a significant decrease in the proliferation of the two bladder cancer cell lines (Fig. 3C). In addition, using CCK-8 assays, it was found that the OD value decreased from day 3 following KIF22 silencing in the T24 and 5637 cells (Fig. 3D).

Subsequently, the expression of the cell proliferation-related markers, PCNA and Ki67, was examined. A markedly decreased expression of PCNA and Ki67 was observed in the group in which KIF22 was silenced, indicating a decline in the cell proliferative ability induced by KIF22 silencing (Fig. 3E-G).

The disruption of the cell cycle leads to an abnormal proliferation and further triggers tumorigenesis. Thus, the present study detected the differences in the cell cycle between the cells in which KIF22 was silenced and the controls. It was noted that KIF22 silencing significantly increased the percentage of cells in the S phase and decreased that of cells in the G2/M phase (Fig. 4A). The expression of cyclin D1 and cyclin A2 was then detected and a significant decrease in expression was observed in the KIF22-silenced T24 and 5637 cells (Fig. 4B). Therefore, the knockdown of KIF22 led to a significant arrest of the cell cycle, thus suppressing the proliferation of bladder cancer cells.

The aforementioned results revealed a critical role of KIF22 in the regulation of the proliferation of bladder cancer cells. To further determine whether KIF22 silencing suppresses bladder cancer progression, an in vivo assay was performed. T24 cells transfected with control or KIF22 shRNA lentivirus were injected into nude mice, and after 2 weeks, the tumor volume was measured each week. Representative tumor photographs from each group were obtained and are presented in Fig. 5A. Tumor growth curves were drawn and the tumor volume in the mice injected with KIF22-silenced cells was significantly smaller than that of the controls (Fig. 5B). KIF22 expression was then detected in the tumor tissues, and the results of both western blot analysis and IHC confirmed that the expression of KIF22 in the tumor tissues from mice injected with KIF22-silenced cells was markedly decreased compared with the controls (Fig. 5C and D). Thus, these results confirmed that KIF22 promoted bladder tumorigenesis in vivo.

The present study then investigated the mechanisms underlying the promotion of the proliferation of bladder cancer cells by KIF22. Previous research has confirmed that CDCA3 is involved in cancer progression (24).

In the present study, to confirm the transcriptional regulation of CDCA3 by KIF22 in bladder cancer cells, RT-qPCR was performed to examine the alteration in CDCA3 expression in KIF22-silenced T24 cells. As was expected, the overexpression of KIF22 led to the increased expression of CDCA3, and KIF22 silencing suppressed the expression of CDCA3 in the T24 and 5637 cells (Fig. 6A). The overexpression efficiency of KIF22 and CDCA3 in the T24 cells was further confirmed using western blot analysis (Fig. S1). Additionally, using western blot analysis, it was found that KIF22 overexpression promoted the expression of CDCA3, whereas its silencing led to a decrease in CDCA3 expression (Fig. 6B).

Furthermore, rescue assays were performed to verify whether KIF22 regulates the proliferation of bladder cancer through CDCA3. The pcDNA3.1-CDCA3 plasmid was used to reverse the defects in the proliferation of T24 cells induced by KIF22 silencing. According to the results of MTT assays, the pcDNA3.1-KIF22 plasmid was transfected into the T24 and 5637 cells, and this evidently increased cell proliferation compared with the controls (Fig. 6C). However, cell proliferation was markedly suppressed by KIF22 knockdown. Of note, it was found that CDCA3 overexpression significantly attenuated the defects in proliferation induced by KIF22 silencing in both the T24 and 5637 cells, suggesting that KIF22 promotes the proliferation of bladder cancer cells via CDCA3 (Fig. 6C). Similarly, using colony formation assays, it was found that KIF22 overexpression induced cell proliferation, and its silencing markedly suppressed cell proliferation. Additionally, it was found that CDCA3 overexpression markedly attenuated the decrease in cell proliferation induced by KIF22 silencing in the T24 and 5637 cells (Fig. 6D).

In addition, luciferase reporter assay was used to determine whether KIF22 regulates the transcription of CDCA3. The pGL-CDCA3 plasmid, which contained the promoter region of CDCA3, was co-transfected with pcDNA3.1-vector or pcDNA3.1-KIF22 plasmid (Fig. 6E). Using ChIP assay in the T24 cells, it was found that the promoter fragment of CDCA3 was specifically co-immunoprecipitated by KIF22 antibody, but not by IgG, indicating the binding of KIF22 with the promoter of CDCA3 (Fig. 6F). There results revealed that KIF22 promotes the transcriptional activation of CDCA3 promoter in T24 cells.

Furthermore, western blot analysis was performed to confirm the findings. The expression of Ki67, a marker of proliferating cells, was detected in the T24 cells in the different treatment groups. It was found that KIF22 overexpression induced the expression of Ki67, and its silencing decreased Ki67 expression. Notably, Ki67 expression was increased following the overexpression of CDCA3 in KIF22-silenced cells (Fig. 6G). Collectively, these results revealed that KIF22 transcriptionally activated the expression of CDCA3.

The aforementioned data revealed that KIF22 could bind to the promoter site of CDCA3 and promoted its transcription, and further promoted bladder tumor development through CDCA3. The present study then aimed to confirm these findings in bladder cancer tissues. IHC was performed using tumor tissues from patients with bladder cancer. The expression level of both KIF22 and CDCA3 was detected, and the association between the expression of KIF22 and CDCA3 in bladder cancer tissues was further analyzed (Fig. 7A). Of note, it was found that the expression of CDCA3 was mainly located in the nucleus and was lower or higher in the KIF22 low or high expression groups, suggesting an evident positive association between the expression of KIF22 and CDCA3 (P<0.001, respectively) (Fig. 7A and Table II). Similarly, through Pearson's correlation analysis, it was noted that the expression of KIF22 correlated with the expression of CDCA3 in human bladder cancer tissues (Fig. 7B). Based on these data, we further confirmed that KIF22 promoted bladder cancer through regulating the transcription of CDCA3 in bladder cancer tissues.

Therefore, it was considered that KIF22 was closely associated with the prognosis and pathological features (clinical stage and recurrence) of patients with bladder cancer. KIF22 can regulate bladder cancer cell cycle and proliferation by activating the transcription of CDCA3 and thus promoting the development of bladder cancer (Fig. 8).