GC tissues and paired normal adjacent mucosa tissues were derived from 150 patients with GC who had undergone surgical resection at the Department of General Surgery, The Affiliated Hospital of Nantong University (Nantong, China). Histological classifications and tumor-node-metastasis (TNM) stages were independently determined by two senior pathologists according to the classification criteria of the American Joint Committee on Cancer. Resected samples were snap-frozen in liquid nitrogen until RNA and protein extraction could be performed. Tissue specimens were flash frozen immediately following surgery and stored in liquid nitrogen until RNA extraction. All patients provided written informed consent prior to surgery and the Ethics Committee of Affiliated Hospital of Nantong University, approved the present study. Signed informed consents were obtained from all subjects and no scientific research was conducted without the informed contents. Nantong University approved the present study. The study was approved by the Ethics Committee of the Affiliated Hospital of Nantong University. All procedures performed in this study were in accordance with the 1964 Helsinki Declaration and its later amendments.

Five human GC cell lines (BGC823, AGS, MKN45, MKN28 and SGC7901) and a human normal gastric epithelial cell line (GES-1) were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA), and the National Infrastructure of Cell Line Resource (Beijing, China). The cell lines were maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS), ampicillin and streptomycin, in a humidified chamber at 37°C with 5% CO₂.

Total RNA was isolated from paired GC and adjacent non-tumor tissues, 5 human GC cell lines and GES-1 cells by employing TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer's protocol. PrimeScript RT reagent (Takara Bio, Inc., Otsu, Japan) was used to transcribe isolated total RNA into cDNA and then SYBR Premix Ex Taq (Takara Bio, Inc.) was used to evaluate mRNA expression levels by quantitative real-time PCR assays. The primer sequences were as follows: CDCA3 sense, 5′-TGGTATTGCACGGACACCTA-3′ and antisense, 5′-TGTTTCACCAGTGGGCTTG-3′; β-actin sense, 5′-AGAGCCTCGCCTTTGCCGATCC-3′ and antisense, 5′-CTGGGCCTCGTCGCCCACATA-3′. All experiments were run in triplicate.

Paraffin-embedded sections (4-µm thick) were incubated with polyclonal rabbit anti-CDCA3 antibody (dilution 1:200; cat. no. ab167037; Abcam, Cambridge, UK) at 4°C overnight using SP-9000 Histostain™-Plus kits (ZSGB-BIO; OriGene Technologies, Inc., Beijing, China) according to the manufacturer's protocol. The immunoreactive scores (IRS) for the proportion of positive cells and he staining grade were calculated for each specimen. The quantity score evaluating the proportion of positive cells of CDCA3 was scored as 0 (negative), 1 (1–10% labeled cells), 2 (11–50% labeled cells), 3 (>50% labeled cells). The intensity score evaluating the intensity of staining was scored as 0 (negative staining), 1 (weakly positive), 2 (moderately positive) and 3 (strongly positive). Multiplication of the extent scores and the staining intensity was performed to calculate the IRS and 3 was the optimal cutoff value: Samples with an IRS ≥3 were defined as high CDCA3 expression and samples with an IRS <3 were regarded as low CDCA3 expression.

Based on the CDCA3 nucleotide sequence from the human cDNA library (GenBank: NM_001297602.2), the full-length open reading frame (ORF) of the human CDCA3 gene was amplified using polymerase chain reaction according to the manufacturer's protocol (PrimeStar PCR; Takara Bio, Co., Ltd., Dalian, China). For the construction of CDCA3, the expression vector pcDNA3.1B containing the full-length open reading frame (ORF) of the human CDCA3 gene was used to generate pcDNA3.1B-CDCA3. The sequence of the forward primer was as follows: 5-EcoRI-AGAGAATTCATGGGCTCAGCCAAGAGCGT-3, and sequence of the reverse primer was 5-BamHI-AGAGGATCCCTAGCTCTCCACCAAGGGA-3. The construct was verified by sequencing.

Small interference RNAs were chemically synthesized (Shanghai GenePharma Co., Ltd., Shanghai, China). The sequence of siRNA-1069 was: 5′-GGGUACCCAGUUAUCUGUUGAGGAAdTdT-3′ (sense) and 5′-UUCCUCAACAGAUAACUGGGUACCCdTdT-3′ (antisense). Negative control (NC) siRNA synthesized by Shanghai GenePharma Co. was used as a control. The sequence of si-NC was as follows: 5′-UUCUCCGAACGUGUCACGUTT-3′ (sense) and 5′-ACGUGACACGUUCGGAGAATT-3′ (antisense). Synthesized DNA nucleotide fragment encoding shRNA against endogenous CDCA3 was designed according to the Invitrogen (Thermo Fisher Scientific, Inc.) and synthesized by GenePharma Co., Ltd. The sequences were incorporated into the Vector p-SUPER (OligoEngine, Seattle, WA, USA) to generate p-SUPER-sh-1069. The sequence of sh-1069 was as follows: 5′-GATCCCCGGGTACCCAGTTATCTGTTGAGGAATTCAAGAGATTCCTCAACAGATAACTGGGTACCCTTTTTGGAAA-3′ (sense) and 5′-AGCTTTTCCAAAAAGGGTACCCAGTTATCTGTTGAGGAATCTCTTGAATTCCTCAACAGATAACTGGGTACCCGGG-3′ (antisense). The sequence of sh-NC was as follows: 5′-GATCCCCTTCTCCGAACGTGTCACGTTTCAAGAGAACGTGACACGTTCGGAGAATTTTTGGAAA-3′ (sense) and 5′-AGCTTTTCCAAAAATTCTCCGAACGTGTCACGTTCTCTTGAAACGTGACACGTTCGGAGAAGGG-3′ (antisense). The constructs were verified by sequencing.

BGC823 and SGC7901 cells were plated in each well of 6-well plates at a density of 5×10⁴ cells/well. BGC823 cells were transfected with p-SUPER-sh-1069 and p-SUPER-sh-NC, SGC7901 cells were transfected with pcDNA3.1B-CDCA3 and pcDNA3.1B, respectively. Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) was used according to the manufacturer's protocol and transfection was conducted when cells reached ~80% confluency. After transfection, we determined transfection efficiency and conducted subsequent functional experiments.

The transfected cells were seeded at a density of 2,000 cells/well in 96-well plates at daily intervals (every 24 h for 5 days) and Cell Counting Kit-8 (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) was used according to the manufacturer's protocol. The absorbance was measured at 450 nm to evaluate the growth curve of transfected cells. All experiments were run in triplicate.

Each type of SGC7901 (SGC7901 transfected with pcDNA3.1B-CDCA3 SGC7901 transfected with pcDNA3.1B) and BGC823 (BGC823 transfected with p-SUPER-sh-1069 BGC823 transfected with p-SUPER-sh-NC) cells were seeded into 6-well plates at a density of 500 cells/well, and then maintained for 3 weeks in RPMI-1640 medium supplemented with 10% FBS, ampicillin and streptomycin containing G418 (1,000 µg/ml). Paraformaldehyde (4%) (Invitrogen; Thermo Fisher Sceintific, Inc.) was employed to fix proliferating colonies for 30 min at room temperature, which were then stained with 1% crystal violet for 120 min at room temperature. The stained colonies were photographed under a light microscope (Leica Microsystems GmbH, Wetzlar, Germany; magnification, ×40), the numbers of colonies (>50 cells/colony) were counted. All experiments were run in triplicate.

Sixty BALB/c nude male mice, 4-weeks old, were purchased from the Department of Laboratory Animal Center, Nantong University (Nantong, China). Animal experiments were performed according to the protocol approved by the Nantong University Ethics Committee. A total of 100 µl phosphate-buffered saline (PBS) mixed with 5×10⁶ control and transfected cells (SGC7901 transfected with pcDNA3.1B-CDCA3; BGC823 transfected with p-SUPER-sh-1069) were separately and subcutaneously inoculated into each flank of nude mice. The tumor size of inoculated nude mice was monitored and measured every third day following injection. The mice were sacrificed and the subcutaneous tumors were dissected and weighed following 3 weeks. During the experiment, the nude mice were exposed to a natural light-dark cycle at room temperature and a specific-pathogen-free (SPF) environment, and were allowed to feed and drink water ad libitum. Carbon dioxide was used to euthanize the mice and the subcutaneous tumors were dissected out and weighed after 3 weeks. The tumor volume was calculated using the following formula: Tumor volume = length × width² × 0.5. Care of experimental animals was in accordance with institutional animal care and use committee guidelines.

Transfected cells were seeded at a density of 1×10⁵ cells/well into 6-well plates, incubated overnight at 37°C, harvested by trypsinization without EDTA and the cells were washed two times with ice-cold PBS. Following treatment with 75% ethanol at 4°C overnight, the cells were incubated on ice for 1 h. Subsequently, the fixed cells were stained at room temperature for 30 min with 0.5 ml PBS solution containing 0.2 mM EDTA, 20 mg/ml of propidium iodide and 0.2% Triton X-100. The proportion of cells in the G₀/G₁, S and G₂/M phases of the cell cycle was evaluated employing the Beckman Gallios Flow Cytometer (Beckman Coulter, Inc., Brea, CA, USA).

A total of 25 mmol/l Tris-Cl (pH 7.5), 5 mmol/l EDTA, 1% SDS and 1% protein lysate with protease inhibitor were used to extract protein, and we determined the protein concentration using a BCA kit. Total cellular proteins (10 µl protein/lane) were separated by SDS-polyacrylamide gel electrophoresis (10% separation gel and 4% stacking gel), blotted onto polyvinylidene difluoride (PVDF) membranes, the PVDF membranes were blocked by 5% skimmed milk, at room temperature for 1 h and then were incubated with specific primary antibodies at 4°C overnight. Following incubation for 1 h at room temperature with horseradish peroxidase (HRP)-conjugated secondary antibody (dilution 1:2,000; cat. no. Ab205718), each membrane was detected using an enhanced chemiluminescence (ECL) detection system (Pierce Biotechnology, Inc., Rockford, IL, USA) according to the manufacturer's protocol. The inclusion of the primary antibodies used in the present study as follows: Anti-CDCA3 (dilution 1:1,000; cat. no. ab167037), anti-p-RAS (dilution 1:1,000; cat. no. ab214100), anti-t-RAS (dilution 1:5,000; cat. no. ab52939), anti-p-MEK (dilution 1:1,000; cat. no. ab194754), anti-t-MEK (dilution 1:5,000; cat. no. ab178876), anti-p-ERK (dilution 1:1,000; cat. no. ab201015), anti-t-ERK (dilution 1:1,000; cat. no. ab17942), anti-p21Cip1 (dilution 1:1,000; cat. no. ab109520), anti-cyclin D1 (dilution 1:200; cat. no. ab16663), anti-cyclin E (dilution 1:1,000; cat. no. ab33911) and anti-β-actin (dilution 1:1,000; cat. no. ab8226) antibody (all purchased from Abcam).

GC tissues and paired normal adjacent mucosa tissues were derived from 150 patients with GC who had undergone surgeries at the Affiliated Hospital of Nantong University between April 2010 and March 2011. The lump and the regional lymph nodes were radically dissected and the edge of the resection was tumor-free, which was deemed to be curatively resected. Patients with GC with distant metastasis were not included in the present study. Follow-up visits of 150 GC patients ranged from April 2011 to March 2016 and the specified clinicopathological features of the 150 patients with GC are listed Table I.

The study data were analyzed using SPSS 19.0 and GraphPad Prism 5.0 software (GraphPad Software, Inc., La Jolla, CA, USA). A two-tailed Student's t-test was employed to analyze differences between 2 groups. Multiple comparisons between groups was performed using analysis of variance (ANOVA) followed by Student-Newman-Keuls test. The χ² test was employed to determine categorical data. The Kaplan-Meier method was employed to estimate survival rates and the log-rank test was used to compare the significance between survival times. A Univariate Cox regression model was used to investigate factors of prognostic significance and then the factors were further investigated using a multivariate Cox regression model. Quantitative results were expressed as the mean ± SD. P<0.05 was considered to indicate a statistically significant difference.

RT-qPCR was performed to assess the CDCA3 mRNA expression levels in GC and adjacent non-cancerous tissues from 150 patients. The results indicated that CDCA3 mRNA was increased in 106 (70.7%) of the 150 GC specimens compared with the matched normal gastric tissues (Fig. 1A). Furthermore, the expression of CDCA3 in a normal human gastric epithelial cell line (GES-1) and diverse human GC cell lines was also detected by qRT-PCR analysis. Our results indicated that CDCA3 expression was markedly upregulated in all GC-derived cell lines (BGC823, AGS, MKN45, MKN28 and SGC7901) compared with GES-1 (Fig. 1B). IHC analysis was performed to evaluate CDCA3 protein expression in GC specimens and paired normal gastric tissues in the same 150 matched samples. Of these specimens, 86/150 (57.3%) of cancerous specimens exhibited high CDCA3 expression, whereas 43/150 (28.7%) of normal gastric tissues exhibited high CDCA3 expression (Fig. 1C). The collective results indicated an aberrant upregulation of CDCA3 in GC.

To access the clinical significance of ectopic CDCA3 expression in GC progression, the relationship between CDCA3 expression and clinicopathological features were evaluated. The patients with GC were split into 2 groups according to the results obtained by the results of IHC analysis for CDCA3: A high-CDCA3 expression group and a low-CDCA3 expression group, and associations between high CDCA3 expression and tumor size (P=0.008), differentiation (P=0.005), primary tumor (P=0.024), lymph node metastasis (P=0.001) and TNM stage (P=0.001) were identified (Table I). Univariate analysis was applied to investigate all relevant features, and high CDCA3 expression (P<0.001), along with tumor size (P=0.003), tumor differentiation (P=0.005), primary tumor (P<0.001), lymph node metastasis (P<0.001) and tumor TNM stage (P<0.001), were identified to be significantly associated with patient survival. Multivariate regression analysis was subsequently employed to further confirm that CDCA3 expression (P=0.003) and TNM stage (P=0.008) were independent prognostic indicators in GC (Table II).

A total of 150 patients were enrolled in the present study. At the last point of follow-up 64 patients had died, of whom 51 exhibited high CDCA3 expression, and 13 exhibited low CDCA3 expression. Among the survivors, the data indicated that 35 patients exhibited high expression levels of CDCA3 and 51 exhibited low CDCA3 expression. Furthermore, the results indicated that the overall survival rate over 5 years for the high CDCA3 group was 40.7%, whereas it was 79.7% for the low CDCA3 group (Fig. 1D).

To explore the role of CDCA3 in the tumor growth of GC, CDCA3 overexpression was established, employing pcDNA3.1B-CDCA3 in the GC cell line SGC7901. Western blotting was used to determine the transfection efficiency of CDCA3 protein expression in the SGC7901 cell line at 48 h following transfection. The results indicated that pcDNA3.1B-CDCA3 effectively upregulated CDCA3 expression in the SGC7901 cell line (Fig. 2A). The ectopic expression of CDCA3 significantly promoted cell growth compared with SGC7901 cells that were transfected with an empty vector (Fig. 2B). In conjunction with this, the number of colonies that formed in CDCA3-transfected SGC7901 cells were significantly increased compared with the empty vector-transfected SGC7901 cells (P<0.05) (Fig. 2C). To further evaluate the mechanism by which CDCA3 promotes cell proliferation, flow cytometry was used to determine the effect of CDCA3 on cell cycle distribution at 24 h following transfection. As illustrated in Fig. 2D the ectopic expression of CDCA3 in SGC7901 cells successfully accomplished cycle transition from the G₀/G₁ to the S phase compared with the empty vector-transfected SGC7901 and the percentage of cells in the S phase increased by 12.52% in the CDCA3-transfected SGC7901 cells (P<0.05) at 24 h. These results demonstrated that CDCA3 is important for the promotion of GC cell growth in vitro.

shRNA-1069 derived from recombinant pSUPER was chemically synthesized to knockdown endogenous CDCA3, concurrently as sh-NC in the BGC823 cell line. Western blotting was used to determine the transfection efficiency of CDCA3 protein expression in the BGC823 cell line at 48 h following transfection. The results indicated that p-SUPER-sh-1069 effectively downregulated CDCA3 expression in the BGC823 cell line (Fig. 3A). The downregulated expression of CDCA3 significantly suppressed cell growth compared with sh-NC transfected cells (P<0.05) (Fig. 3B). In keeping with this, the number of colonies that formed in BGC823 cells transfected with p-SUPER-shRNA-CDCA3 were significantly inhibited in comparison with BGC823 cells transfected with sh-NC (P<0.05) (Fig. 3C). To further evaluate the effect of downregulation of CDCA3 on cell proliferation, flow cytometry was used to determine cell cycle distribution 24 h following transfection. As illustrated in Fig. 3D, knockdown of CDCA3 was associated with cell cycle arrest at the G0/G1 stage, as there was an increase of the percentage of cells in the G₀/G₁ phase in the p-SUPER-sh-CDCA3 transfected group by 15.33% compared to the sh-NC group (P<0.05) at 24 h.

To explore the effect of CDCA3 expression on tumorigenic potential of GC cell lines in vivo, the present study performed a nude mouse xenograft assay. SGC7901 cells overexpressing CDCA3 and BGC823 cells with downregulated CDCA3 expression were injected subcutaneously into the flank of nude mice. CDCA3-overexpressing cells significantly promoted tumor growth in terms of increased capacity for tumorigenesis compared with the control group (Fig. 4A, C and E). Furthermore, the present study also demonstrated that CDCA3 silencing substantially inhibited tumor growth in terms of decreased capacity for tumorigenesis in comparison with the negative controls (Fig. 4B, D and F). These data indicated that the in vivo and in vitro investigations had the same varying trend regarding cell proliferation, and indicated that CDCA3 may function as an accelerator of tumorigenicity.

To further explore the signaling pathway by which CDCA3 enhances the progression of GC, western blotting was performed to determine potential CDCA3-regulated molecules. As illustrated in Fig. 5, the levels of p-Ras, p-MEK, p-ERK, cyclin D1 and cyclin E were increased; however, CDKI (p21Cip1) was decreased in the CDCA3-overexpressing SGC7901 cells when compared with the control group. However, no difference in the t-Ras, t-MEK and t-ERK protein levels were observed between the 2 groups.

The levels of p-Ras, p-MEK, p-ERK, cyclin D1 and cyclin E were decreased, however CDKI (p21Cip1) was increased in CDCA3-downregulated BGC823 cells compared with the control group. Similarly, the expression of t-Ras, t-MEK and t-ERK protein levels were significantly unvaried between the 2 groups (Fig. 5). Collectively, this data demonstrated that the Ras signaling pathway may participate in CDCA3-mediated GC progression.