CDCA2 promotes the proliferation of colorectal cancer cells by activating the AKT/CCND1 pathway in vitro and in vivo

A total of 120 CRC and 115 adjacent non-tumor colorectal specimens were obtained from Jiangsu Province Hospital between June 2014 and June 2016. The Research Ethics Committee of Nanjing Medical University has approved the research, and we obtained written informed consent from all patients. All the samples were obtained surgically and conserved at − 80 °C. Patients were not included in this study if they received any preoperative treatment. For the in vitro experiments, cell lines, including five types of CRC cells (SW480, LoVo, DLD-1, HCT116, HT29) and a intestinal mucosal epithelial cell (NCM460), all were conserved in the laboratory. The cell culture medium consisted mostly of Dulbecco’s modified Eagle’s medium (DMEM), including 100 U/mL penicillin, 100 μg/mL streptomycin and 10% fetal bovine serum (Wisent, Canada). All cells were cultured in a 5% CO2 atmosphere at about 37 °C. We purchased the PI3K inhibitor LY294002 from Cell Signaling Technology (Danvers, MA, USA) and the inhibitor was used to treat CRC cells at 10 μM.

IHC was conducted to detect protein expression of CDCA2 in 30 CRC tissues and 25 non-tumor tissues. IHC staining, which was performed according to standard immunoperoxidase staining procedure, was independently reviewed by two experienced pathologists. The staining intensity score was calculated as follows: 0, negative; 1, weak; 2, moderate; and 3, strong. The percentage of positive cells was calculated as follows: 0, negative; 1, < 33%;2, 34–66%; 3, > 67%. The final scores were based on the sum of these two scores, scored as follows: -(total score 0); + (total score 1 and 2); ++ (total score 3 and 4) and +++ (total score 5 and 6).

Total RNA was extracted using a TRIzol extraction kit (Invitrogen, Carlsbad, CA, USA) followed by the manufacturer’s protocol. The qRT-PCR assay was carried out by means of a PCR kit (Roche Diagnostics, Indianapolis, IN, USA). Subsequently, the final step was conducted with the StepOnePlus Real-time System (Applied Biosystems, Foster City, CA, USA). The gene-specific primer sequences were as follows: CDCA2: forward 5′-TGCCGAATTACCTCCTAATCCT-3′ and reverse 5′- TGCTCTACGGTTACTGTGGAAA-3′, p21: forward 5′- TGTCCGTCAGAACCCATGC-3′ and reverse 5′- AAAGTCGAAGTTCCATCGCTC-3′, p27: forward 5′- AGGAGGAGATAGAAGCGCAGA-3′ and reverse 5′- GTGCGGACTTGGTACAGGT-3′, CCND1: forward 5′- GCTGCGAAGTGGAAACCATC-3′ and reverse 5′- CCTCCTTCTGCACACATTTGAA-3′, CCNB1: forward 5′- AATAAGGCGAAGATCAACATGGC-3′ and reverse 5′- TTTGTTACCAATGTCCCCAAGAG-3′, CCNE1: forward 5′- AAGGAGCGGGACACCATGA-3′ and reverse 5′- ACGGTCACGTTTGCCTTCC-3′, CDK2: forward 5′- CCAGGAGTTACTTCTATGCCTGA-3′ and reverse 5′- TTCATCCAGGGGAGGTACAAC-3′, GAPDH: forward 5′- GGAGCGAGATCCCTCCAAAAT-3′ and reverse 5′- GGCTGTTGTCATACTTCTCATGG-3′. The 2^–ΔΔCt method was used to analyzed the data. All qRT-PCR processes were carried out in triplicate.

We downloaded normalized RNA expression data (level 3) from the RNASeq v2 system, which were provided by the Cancer Genome Atlas project (TCGA) (https://​portal.​gdc.​cancer.​gov/​). We defined the exclusion criteria: i) histological results excluded a CRC tissue; ii) patients had other malignant diseases at the same time; iii) samples lack of necessary data; and iv) patients receiving neoadjuvant therapy. At last, 351 tumor tissues and 32 normal tissues were used in this study.

Negative control (NC) and small interfering RNAs (siRNAs) of CDCA2 were constructed by GenePharma Corporation (Shanghai, China). The siRNA sequences of CDCA2 were as follows: siRNA1, 5′-CACCUGCCUUUCUAAAUAUTT-3′; siRNA2, 5′-GGGCAAAGGAUCAAGUGAUTT-3′; siRNA3, 5′-CUGCCUUGGAAAGGAUUGATT-3′. Transfection was performed with Lipofectamine 3000 (Invitrogen) following the manufacturer’s instructions. The assays were performed 48 h after transfection to assess the knockdown efficiency. A CDCA2 inhibitor lentivirus (shCDCA2) was then constructed according to siRNA1. To upregulate the expression of CDCA2 in DLD-1 cells, mammalian expression plasmids (pReceiver-M02-CDCA2) designed to specifically express CDCA2 were obtained from GeneCopoeia (Rockville, MD, USA).

Five hundred cells were cultured in the wells of six-well plates 48 h after transfection. Two weeks later, each well of the plates was bathed with cold phosphate-buffered saline (PBS) 2–3 times, soaked in 95% alcohol for about 30 s, then dyed with crystal violet for 10 min. Afterwards, a Nikon light microscope (Nikon Corporation, Tokyo, Japan) were employed to count the spots (≥50 cells/spot). And a Canon digital camera (Canon DS126211, Inc., Tokyo, Japan) were used to capture images.

The Cell Counting Kit-8 (CCK-8; Dojindo, Tokyo, Japan) assay was used to detect cell viability. Firstly, 96-well plates were seeded with 2 × 10³ cells. Then, 100 μL reagent, 10% of which was CCK-8 reagent, was supplemented to each well 24, 48, 72, and 96 h later. Two hours later, We measured the absorbance at a test wavelength (450 nm) and a reference wavelength (630 nm) by using a microplate reader.

5-Ethynyl-2′-deoxyuridine (EdU) assay kit (RiboBio, China, C10310–3) was used to measure cell proliferation. Briefly, before the addition of EdU (50 μM), cells were seeded into 24-well plates (2 × 10⁴ cells/well) and cultured with DMEM for 24 h. The cells were then immersed in formaldehyde (4%) for 30 min and soaked in 0.5% Triton X-100 for 10 min. 400 μL 1× ApolloR reaction cocktail was then supplemented. After a 30 min reaction, to bring out the nuclei, Hoechest 33,342 (400 μL) was added. 30 min later, images of the cells were shot by a Nikon microscope (Nikon, Japan). To assess cell proliferation, we then randomly selected three fields and calculated the mean number of cells.

The cells were diposed by trypsin and centrifuged for 5 min at 1200 rpm. The cells were firstly washed with PBS 2–3 times and then soaked in 75% ethanol before saved at − 20 °C overnight. After being washed twice with PBS and incubated with RNAse, the cells were stained with PI staining solution (500 μL) for about 15 min at room temperature. PI (10 μg/mL; Sigma-Aldrich) and Annexin V-FITC (50 μg/mL, BD Biosciences) were used to incubate with the apoptotic cells in dark place for about 15 min. The data were acquired using a FACScan flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA).

Followed by the manufacturer’s guidelines, protein lysates from the different cells were treated with a RIPA kit (Beyotime, Shanghai, China). Before transferring to polyvinylidene difluoride membranes (Millipore, Bedford, MA, USA), the different-weight proteins would be separated on 10% sodium dodecyl sulfate-polyacrylamide gels in running buffer. The membranes were soaked in 5% skim milk for 2 h at room temperature and then soaked in primary antibodies at 4 °C overnight. The membranes were then incubated with anti-mouse or anti-rabbit IgG for 2 h at room temperature and then washed with TBST buffer three times. The bands were exposed by ECL Plus (Millipore, Billerica, MA, USA) in a Bio-Imaging System. The antibodies included were as follows: CDCA2 (1:1000, no. ab45129), CCND1 (1:10000, no.ab134175), CCNE1 (1:1000, no.ab33911), CCNB1 (1:25000, no. ab32053), p-AKT (1:500, no.ab38449), AKT (1:500, no.ab8805), anti-rabbit secondary antibodies (1:5000, no. GAB007), and anti-mouse secondary antibodies (1:5000, no. GAM007). We used GAPDH (1:5000, no.ab8245) as a control.

We purchased totally 12 male mice (3–5 weeks, 12–16 g) from the Laboratory Animal Center. We then randomly divided the mice into two groups (shNC and shCDCA2) and injected 2 × 10⁶ CRC cells (shCDCA2 or shNC SW480 cells) subcutaneously per rat. We measured bi-dimensional tumor extent every 4 days. Four weeks later, all the mice were executed by dislocation of the cervical vertebra and all implanted tumors were surgically collected. The formula to figure up the tumor volume was as follows: volume = (width² × length)/2. All the experiments above were carried out followed by the protocols of the NJMU Institutional Animal Care and Use Committee.

We used the Statistical Program for Social Sciences 20.0 software (SPSS, CA, USA) and GraphPad Prism 5.0 (GraphPad Software, CA, USA) to analyze the data. Chi-square test (Table 1) was used to analyze the clinical features. The Wilcoxon rank-sum test was used in Table 2. Pearson’s correlation test was carried out to examine the relationship between CDCA2 and Ki-67. One-way analysis of variance or student’s t-test were used to analyze the treated and control groups. A P-value < 0.05 indicated statistical significance.

To detect the functional role of CDCA2, we analyzed CDCA2 expression in published profiles from TCGA (https://​cancergenome.​nih.​gov/​) and found that it was overexpressed in tumor sample (337 cases) compared with normal tissues (32 cases) (P < 0.001; Fig. 1a). Next, we analyzed CDCA2 expression in a total of 32 paired CRC tissues in this dataset to further confirm the above result and found that it was significantly upregulated in tumor samples (P < 0.001; Fig. 1b). To further detect the potential role of CDCA2, we compared the level of CDCA2 mRNA expression between 90 pairs of CRC and adjacent normal tissues using qRT-PCR assay, normalized it to GAPDH (P < 0.001; Fig. 1c), and investigate the connection between the level of CDCA2 expression and the clinical factors of CRC. According to the median CDCA2 expression, we classified the 90 CRC cases into two groups: the high CDCA2 expression group (n = 45) and the low CDCA2 expression (n = 45) (Table 1). Obviously, high CDCA2 expression was significantly related to larger tumor size (P = 0.002) and lymph node invasion (P = 0.011). However, the expression of CDCA2 was not associated with other clinical features, like age (P = 0.649) and primary tumor site (P = 0.294). We then detected its expression using IHC staining in 55 paraffin-embedded tissues, including 25 normal tissues and 30 CRC tissues. Table 2 and Fig. 1d showed that protein expression of CDCA2 was obviously overexpressed in cancerous samples compared to non-tumor tissues. Furthermore, CDCA2 expression were raised in all five cancer cell lines compared to NCM460 cells in both protein and mRNA levels (Fig. 1e and f). Moreover, through analyzing the data from TCGA, we found that expression of CDCA2 was positively associated with expression of Ki-67 (r = 0.5635, P < 0.001) (Fig. 1g). Taken together, all of the above results strongly indicated that CDCA2 expression is upregulated in CRC cells and may be related to advanced clinicopathological features of CRC.

To further confirm the function of CDCA2 in CRC, SW480 cells were transiently incubated with small interfering RNAs targeting CDCA2, while CDCA2 expression was upregulated in DLD-1 cells (Fig. 1e and f). The qRT-PCR results revealed that siRNA1-CDCA2 presented a higher efficiency of interference than siRNA2-CDCA2 and siRNA3-CDCA2 (Fig. 2a). DLD-1 cells were transiently transfected with CDCA2-carrying plasmids, as well as empty vectors. The effect was determined by western blot (Fig. 2b).

Considering that CDCA2 encodes a targeting subunit of the cell-cycle associated protein [8], we speculated whether CDCA2 played a role in promoting cell proliferation. The colony formation assay indicated that groups transfected with siRNA1 presented fewer spots than siRNA-NC-transfected groups (Fig. 2c). Similarly, the results of the CCK-8 assay showed that siRNA-CDCA2-transfected CRC cells showed evidently damaged proliferative capacity (Fig. 2d). EdU (red)/Hoechst 33342 (blue) immunostaining showed the same trend that downregulation of CDCA2 significantly inhibited the proliferative capacity of CRC cells (Fig. 2e). Finally, exogenous CDCA2 expression in DLD-1 cells enhanced their proliferation (Fig. 2c, d, and e). According to the results of the three assays, we concluded that CDCA2 can promote proliferation of CRC cell lines.

To further detect the effect of CDCA2 in tumorigenesis in nude mice, we performed tumorigenesis assay. SW480 cells were incubated with CDCA2 inhibitor lentivirus (shCDCA2) or empty vectors (shNC). Cells (2 × 10⁶) transfected with shCDCA2 or shNC were separately injected subcutaneously into nude rats. Tumor size and volume was respectively measured and calculated every 4 days. All implanted tumors were collected 28 days after injection (Fig. 3a). Figure 3b showed that tumor growth was much slower (Fig. 3b) in shCDCA2 group, which also presented lower mean weight compared with shNC group (Fig. 3c). IHC result revealed that tumors taken from shCDCA2-transfected cells presented lower intensity of Ki-67 staining compared with those derived from cells transfected with empty vector (Fig. 3d). In conclusion, these in vitro and in vivo assays demonstrated that CDCA2 may play an important role in the proliferation of CRC cells.

To explore the possible mechanism by which CDCA2 influences the proliferation of CRC cells, we determined the CDCA2 expression from TCGA via gene set enrichment analysis (GSEA) [13] and found that CDCA2 expression levels were positively correlated with cell proliferation by affecting genes in the cell cycle phase transition, especially in G1-S phase transition (Fig. 4a). To verify the results of this analysis, we measured cell cycle distribution in SW480 and DLD-1 cell lines. As shown in Fig. 4b, siRNA-CDCA2 treatment induced an increase in the percentage of SW480 cells in the G1 phase compared to that by siRNA-NC treatment. Accordingly, upregulation of CDCA2 accelerated the G1/S transition in DLD-1 cells (Fig. 4c). Moreover, qRT-PCR was performed to determine the key checkpoints of the G1-S phase transition, and results revealed that knockdown of CDCA2 led to the downregulation of CCND1, CCNE1, and CCNB1 expression, whereas overexpression of CDCA2 upregulated the expression of these molecules (Fig. 4d). However, altering the expression of CDCA2 had no effect on the expression of p21, p27, and CDK2 (Fig. 4d). We further confirmed this through western blotting. The results showed that CCND1 changed most apparently following up- or down-regulation of CDCA2 (Fig. 4e). Therefore, we believed that the regulation of cell cycle by CDCA2 may be mainly achieved through CCND1.

The PI3K/AKT pathway plays an essential role in cell proliferation in various types of cancer and is closely linked with the adjustment of CCND1 levels [14, 15]. Hence, we performed western blotting to detect whether the PI3K/AKT pathway was involved in the adjustment of proliferation under the influence of CDCA2. As shown in Fig. 5a, the expression of phosphorylated AKT in SW480 cells transfected with siRNA-CDCA2 decreased, with no change observed in the expression of total AKT. The opposite results were obtained in CDCA2-overexpressing DLD-1 cells (Fig. 5a). We therefore questioned whether the PI3K/AKT pathway participates in the regulation of CCND1 expression by CDCA2.

To verify the hypothesis, CDCA2-overexpressing DLD-1 cells were treated with a specific PI3K inhibitor (LY294002). qRT-PCR and western blot analysis were then carried out, and results showed that treatment of CDCA2-overexpressing DLD-1 cells with LY294002 for 24 h obviously inhibited p-AKT activity and promoted the decline of CCND1 expression (Fig. 5b and c), indicating that CCND1 accumulation could be suppressed by the addition of inhibitors in CRC cell lines. We then performed rescue experiments to detect whether the suppression of the PI3K/AKT pathway could reverse the proliferation-promoting effect induced by CDCA2 overexpression. Compared to DMSO treatment, LY294002 treatment significantly weakened the growth ability of CDCA2-overexpressing DLD-1 cells (Fig. 5d and e). These results implicated a role of CDCA2 in cell cycle progression through activation of the PI3K/AKT pathway in CRC.