C-terminal binding protein-2 promotes cell proliferation and migration in breast cancer via suppression of p16INK4A

C-terminal binding protein-2 (CtBP2) enhances cancer proliferation and metastasis. The role and mechanism of CtBP2 in breast cancer remains to be elucidated. Western blot and immunochemistry were employed to evaluate the level of CtBP2 and p16INK4A in breast cancer. Genetic manipulation was used to study the expression of p16INK4A and its downstream genes regulated by CtBP2. Functional assays, including colony formation, wound healing, transwell invasion, anchorage-independent growth assay and a xenograft tumor model were used to determine the oncogenic role of CtBP2 in breast cancer progression. The expression of CtBP2 was increased in breast cancer tissues and cell lines. The expression of p16INK4A were inversely correlated CtBP2 (r2 = 0.43, P < 0.01). The expression of both CtBP2 and p16INK4A were significantly related to histological differentiation (P < 0.01 and P = 0.004, respectively) and metastasis (P = 0.046 and 0.047, respectively). The overall survival rate was lower in patients with increased CtBP2 expression and lower p16INK4A expression. Knockdown of CtBP2 resulted in the activation of p16INK4A and down–regulation of cell cycle regulators cyclin D, cyclin E and cyclin-dependent kinase 2 and 4. This down-regulation also led to a decreased transition of the G1-S phase in breast cancer cells. Moreover, gain-of-function experiments showed that CtBP2 suppressed p16INK4A and matrix metalloproteinase-2, subsequently enhancing the migration in breast cancer. However, the silence of CtBP2 abrogated this effect. Collectively, these findings provide insight into the role CtBP2 plays in promoting proliferation and migration in breast cancer by the inhibition of p16INK4A.


INTRODUCTION
Breast cancer is currently one of the most common malignancies, and the second leading cause of cancer death among women worldwide [1]. Despite the progress in combined modality therapies, the long-term outcome of patients with breast cancer is far from satisfactory [2]. This outcome is mainly attributed to the induction and progression of breast cancer. Several studies show that aberrant transcriptional activities of major oncogenes and tumor suppressor genes are involved in the tumorigenesis of breast cancer [3]. Although the identification and characterization of transcriptional co-activators and corepressors further the understanding of regulation on oncogene transcription, the underlying mechanisms by which specific transcription factors play a role in breast cancer are still unclear.

C-terminal binding protein-2 promotes cell proliferation and migration in breast cancer via suppression of p16 INK4A
and the repression of migration [18,19]. The protein expression of p16 INK4A is reduced in human primary tumors, including those of ESCC (esophageal squamous cell carcinoma) [14], urothelial cancer [20], ovarian cancer [21], non-small cell lung carcinoma, glioma and breast carcinoma [22]. p16 INK4A blocks the cell cycle progression by binding to either CDK4 or CDK6, and inhibiting the action of cyclin D [23][24][25][26]. In the presence of an imbalance between p16 INK4A and cyclin D, the p16 INK4A accelerates the G1-S phase checkpoint, resulting in abnormal cell growth and tumor development [27]. Down-regulation of p16 INK4A promotes migration in breast cancer because of an increased secretion level of matrix metalloproteinase-2 (MMP-2) [19]. In the present study, we investigated the detailed mechanisms by which CtBP2 contributes to the development of breast cancer and the predictive value of CtBP2 and associated pathways in the prognosis of breast cancer.

The expression of CtBP2 and p16 INK4A in breast cancer tissues and cells
Immunohistochemical staining was used to determine the physiological and pathological interaction between CtBP2, p16 INK4A and the proliferation index Ki-67 in tissue samples from patients with both benign breast disease and breast cancer. The results are shown in Figure 1A and Table 1. Representative examples of reactivity for CtBP2, p16 INK4A and Ki-67 are shown in Figure 1A. CtBP2 and p16 INK4A were both expressed mainly in the nuclei. Our results showed that the immunoreactivity of CtBP2 and Ki-67 was weak in the nucleus of (Figure 1Aa, 1Ac) mammary epithelial and myoepithelial cells in tissue samples from benign breast disease. The expression of p16 INK4A was strong ( Figure 1Ab); however, CtBP2 was strongly expressed in tissues from breast cancer samples and four breast cancer cell lines. Interestingly, the expression of CtBP2 was higher in MDA-MB-231 and MCF-7 cells when compared with other cell lines ( Figure 1C, 1D). Moreover, we found that the level of p16 INK4A was inversely related to the level of CtBP2 in both breast carcinoma specimens and cell lines of breast cancer ( Figure 1B).
The expression of CtBP2 was positively related to Ki-67 in breast cancer specimens ( Figure 2). In addition, the proportion of p16 INK4A -positive tumor cells was negatively correlated with the proportion of CtBP2positive and Ki-67-positive tumor cells ( Figure 2).

Correlation between CtBP2, p16 INK4A expression and clinicopathological variables in breast cancer
As shown in Table 1, the level of CtBP2 was positively correlated with the histologic grade (P < 0.001), metastasis (P = 0.046) and tumor size (P = 0.011). However, CtBP2 expression was not related to the age, histology, estrogen receptors (ER), progesterone receptors (PR) or HER2 status in patients with breast cancer. In contrast, the level of p16 INK4A expression was inversely correlated with histologic grade (P = 0.004), metastasis (P = 0.047) and tumor size (P = 0.043), and no significant correlation was found between p16 INK4A expression and other variables.

The expression of CtBP2 and p16 INK4A in relation to prognosis in patients with breast cancer
At the end of clinical follow-up, survival information was available for a total of 80 patients. The survival rate of patients with a high level of CtBP2 was significantly lower than that of patients with a low level of CtBP2 (31.2%, (18/57) and 78.3% (18/23), respectively), as shown in Table 2. Univariable analysis was performed to study the expression of CtBP2 and p16 INK4A in relation to survival status (Table 2). Kaplan-Meier analysis showed that increased expression of CtBP2 was significantly associated with shorter overall survival (P = 0.042, Figure 3A), whereas a high level of p16 INK4A was associated with longer overall survival (P < 0.001, Figure 3B). Patients with a high expression of CtBP2 and low expression of p16 INK4A had a poorer overall survival rate when compared to the other patients (P < 0.001, Figure 3C). The Cox's proportional hazards regression model demonstrated that expression level of CtBP2 and p16 INK4A , histological grade, tumor size and metastases were independently predictive factors for an adverse prognosis in patients with breast cancer ( Table 3).

The expression of CtBP2 and p16 INK4A was correlated to cell cycles in the MDA-MB-231 breast cancer cell line
The correlation between the cell cycle stages and the expression levels of CtBP2 and p16 INK4A was further examined in MDA-MB-231 cells. After the synchronization of cell cycles at the G0/G1 phase by serum deprivation for 48 h, the cells were released and allowed to progress to the S phase by serum stimulus. The progress of the entire cell cycle was monitored by flow cytometry as the time indicated ( Figure 4A). We found a time-dependent increase of CtBP2 expression ( Figure 4B). In contrast, the p16 INK4A expression showed a trend of time-dependent decrease ( Figure 4C). Collectively, our findings suggest that CtBP2 and p16 INK4A expression is related to the cell cycle. Our results are consistent with a previous study by Guan et al. [14].

CtBP2 promotes proliferation in breast cancer cell lines
We further investigated the mechanisms by which CtBP2 stimulates proliferation in cell lines of MDA-MB-231 and MCF-7. The transfection of pcDNA3.1-EGFP and pcDNA3.1-EGFP-CtBP2 vectors in MDA-MB-231 cells was tested by light microscopy. In addition, CtBP2 protein was analyzed using Western blot ( Figure 4D). The expression of p16 INK4A was significantly decreased in cells transfected with pcDNA3.1-EGFP-CtBP2 vectors when compared to cells transfected with pcDNA3.1-EGFP vectors ( Figure 4E). Concomitantly, a decrease of p16 INK4A expression was detected ( Figure 4F). After transfection, cell proliferation was evaluated by CCK-8 assay at indicated times (0, 12, 24, 48, and 72 h). The results demonstrate a remarkable increase of cell proliferation ( Figure 4G). In addition, fluorescence activated cell sorter analysis of cell cycle distribution revealed that the number of cells in S phase was significantly increased in cells transfected with pcDNA3.1-EGFP-CtBP2 vectors ( Figure 4H). Our findings suggest that CtBP2 promotes cell proliferation by inhibiting p16 INK4A , resulting in a shorter transition of the cell cycle in breast cancer cells.
To further investigate the CtBP2-p16 INK4A pathway described above, we established loss-of-function models ( Figure 5). MDA-MB-231 and MCF-7 cells were transfected with either CtBP2-shRNA or control vectors for 48 h. The efficiency of transfection was assessed by Western blot ( Figure 5A). We found that the expression of p16 INK4A was significantly increased after the transfection of CtBP2-shRNA vectors ( Figure 5A, 5B, 5C). Moreover, CCK8 and colony formation assays revealed that the proliferation rate of CtBP2-shRNA positive cells was significantly slower ( Figure 5D, 5E). Additionally, the analysis of cell cycle distribution revealed an accumulation of cells at the G0/G1 phase and a concomitant reduction of cells at S phase after the transfection of CtBP2-shRNA vectors ( Figure 5F). The expression of several key cell cycle regulators, including CDK2, CDK4, CyclinD and CyclinE was decreased in CtBP2-shRNA positive cells   Figure 5G). We also measured the expression of p21 and Bax which are additional CtBP2-targeted pathways [4]. We found that the expression of p21 and Bax was increased in CtBP2-shRNA positive cells ( Figure 5G).

CtBP2 facilitates breast cancer cell migration and invasion
To investigate the molecular mechanisms by which CtBP2 enhances breast cancer cell migration, EGFP-CtBP2 and/or CtBP2-shRNA MDA-MB-231 and MCF-7 clones were established and subjected to wound healing assays. We observed that the track length of EGFP-CtBP2 breast cancer cells migrating into the cell free areas was significantly longer than those of CtBP2-shRNA cells ( Figure 6A, 6B). A transwell assay was employed to assess the effect of CtBP2 on cell invasion. We found that the invasion was substantially augmented in breast cancer cells transfected with EGFP-CtBP2 vectors ( Figure 6C, 6D). After the transfection of the EGFP-CtBP2 vectors, the expression of the epithelial marker E-cadherin was decreased, while the expression of vimentin was increased. In contrast, CtBP2-shRNA positive cells showed an increased expression of E-cadherin and a decreased expression of vimentin ( Figure 6E). We subsequently detected the expression of p16 INK4A -targeted MMP-2. This expression of MMP-2 was decreased in the cells transfected with CtBP2-shRNA, while MMP-2 was increased in EGFP-CtBP2 positive cells ( Figure 6E). These results showed that CtBP2 promoted cell migration and invasion, which correlates with the expression of E-cadherin and p16 INK4A .

Effect of CtBP2 on tumorigenicity
As shown in Figures 7A and 7B, EGFP-CtBP2 cells demonstrated a significant increase in the anchorage-   independent growth ability in soft agar; however CtBP2-shRNA cells displayed a decrease of the anchorageindependent growth ability in soft agar. Our in vitro studies indicated that functional overexpression of CtBP2 makes breast cancer cells phenotypically more malignant, and underexpression of CtBP2 makes the same cells less malignant. Furthermore, we evaluated the effect of CtBP2 on tumorigenicity in nude mice. The tumor growth was measured every 4 days. Similarly, we observed that EGFP-CtBP2 positive tumors grew significantly faster, whereas the tumors formed by CtBP2-shRNA cells grew at a much slower rate ( Figure 7C, 7D). The size of tumors formed by CtBP2-shRNA cells was significantly smaller compared with those formed by EGFP-CtBP2 positive cells (P < 0.05, Figure 7D).

DISCUSSION
An increasing body of evidence indicates that CtBP2 is involved in tumorigenesis and tumor progression by the regulation of several essential cellular processes, such as transcriptional repression [9], and is correlated with poor prognosis in a number of tumor types [11,14,[28][29][30][31]. CtBP2 works against important tumor suppressors such as E-cadherin [32], p16 INK4A [5], p15 Ink4b , PTEN, HIPK2 [4], Ink4a/Arf [8] and APC [33], and enhances    cell proliferation, migration and invasion. Meanwhile, down-regulation of p16 INK4A shows much aggressiveness in breast cancer cells by increasing the expression level of MMP-2 [19]. In ESCC tissues and cells, CtBP2 protein level is elevated by a proteomic study and Guan et al (2013). revealed that CtBP2 contributes to malignant development of ESCC by regulation of p16 INK4A [14]. However, the molecular mechanism of CtBP2 in human breast cancer is unclear. Therefore, the present study is aimed to investigate the role and mechanism of CtBP2 in human breast cancer.
Our study shows that the expression of CtBP2 in breast cancer tissues is increased, which was consistent with a previous report by Liu et al. [11]. Meanwhile, p16 INK4A expression was decreased in breast cancer tissues. IHC analysis revealed that the expression of CtBP2 in breast cancer samples was positively correlated with breast cancer malignancy. We also found that the increased expression of CtBP2 was a predictor of poor survival. The expression of p16 INK4A was significantly lower in breast cancer and was correlated with poor survival in patients with breast cancer. Patients with the combined phenotype of CtBP2 (high) and p16 INK4A (low) showed a poorer overall prognosis. This study further showed that CtBP2 negatively regulated the expression of p16 INK4A , and was positively correlated with cellular proliferation. Thus, these results suggest that CtBP2 might enhance the progression of breast cancer by suppressing the p16 INK4A pathway.
This study also showed that CtBP2 is related to histologic grade, metastasis and tumor size, which implies that CtBP2 may be involved in the proliferation and migration of breast cancer. The expression of CtBP2 was detected during the progression of the cell cycle in breast cancer cells. Upon serum stimulation and release from G1, the expression of CtBP2 was clearly increased, concurrent with a decrease of p16 INK4A expression. The promotion of growth in the breast cells by wild-type CtBP2 could be explained by the accelerated cell cycle. Down-expression of CtBP2 resulted in suppressed cell growth and an arrest in the cell cycle transition. Similar results were also demonstrated by the colony formation assay and anchorage-independent growth assays. These findings suggest that CtBP2 stimulates cell proliferation and growth in breast cancer. In agreement with previous studies, the present study also found that CtBP2 enhances migration in breast cancer [11]. Genetic manipulation substantiated that CtBP2 promotes migration and invasion by the inhibition of E-cadherin and stimulation of MMP-2.
The in vivo animal model confirmed that CtBP2 promotes the tumor growth. However, our study has some limitations. While we have confirmed that there are inverse correlations between the expression of CtBP2 and p16 INK4A , the mechanism by which CtBP2 regulates p16 INK4A expression is still unclear, and follow-up experiments are suggested.
In summary, our results demonstrate that CtBP2 might contribute to the progression of breast cancer by promoting cell proliferation, enhancing cell migration or invasion and inhibiting the expression of p16 INK4A . A better understanding of the molecular mechanism of CtBP2 in breast cancer development and progression provides novel therapeutic strategies for breast cancer patients.

Specimens
All samples were obtained preoperatively by biopsy. Eighty breast specimens were obtained from January 2000 to December 2003 in the department of pathology, the Affiliated Sixth People's Hospital, Shanghai Jiaotong University. This study was approved by the ethical committee of local institute. Written informed consent was obtained from all patients. The mean post-operative follow-up period for these patients was 78 months (range: 17-129 months). Breast tumors were reviewed regarding histopathologic type based on the WHO classification.

Immunohistochemistry (IHC) and immunohistochemical analyses
The specimens were fixed in 10% formalin and embedded in paraffin. The procedures were carried out as described as previous methods [11]. Sections were incubated overnight at 4°C with purified CtBP2, p16 INK4A and Ki-67 antibodies, which were diluted all at 1:100 with 10 % normal serum in phosphate-buffered saline (PBS). Two pathologists independently scored the results of the staining, and similar results were obtained. CtBP2, p16 INK4A and Ki-67 indices were determined as the percentage of all immunostained cells. The mean percentage of CtBP2-positive cells was 61.36%. The samples were considered CtBP2-positive when the percentage of positive cells was > 61.36% and negative when the percentage was ≤ 61.36%. Meanwhile the mean percentage of p16 INK4A -positive cells was 37.14%, so the samples were considered p16 INK4A -positive when the percentage of positive cells was > 37.14% and negative when the percentage was ≤ 37.14%.

Western blot analysis
Western blot was performed similarly to previously described methods [34], using anti-CtBP2

Reverse Transcriptase PCR (RT-PCR) and real-time quantitative PCR (RT-qPCR)
Total RNA was prepared from MDA-MB-231 and MCF-7 cells using a Trizol extraction kit according to the manufacturer's procedure. cDNA was synthesized using the Thermo Script RT-PCR system (Invitrogen). Primer pairs for p16 INK4A were: sense, 5′-GGGTAGAGGAGG TGCGG-3′ and antisense, 5′-CGGGGATGTCTGAGGGA-3′. The primer pairs for p21 were: sense, 5′-ATGTCAGAA CCGGCTGGGGATGTC-3′, and antisense, 5′-GGGCTT CCTCTTGGAGAAGATC-3′. Cycling conditions were: 95°C for 45 s, 55°C for 45 s, 72°C for 30 s, and a total of 30 cycles. The last cycle was followed by an additional extension step of 72°C for 10 min. GAPDH was used as internal control and was detected using the primers sense, 5′-TGATGACATCAAGAAGGTGGTGAAG-3′ and antisense, 5′-TCCTTGGAGGCCATGTGGGCCAT-3′. Cycling conditions were: 94°C for 30 s, 55°C for 30 s, 72°C for 30 s, and a total of 28 cycles. Densitometric analysis of PCR products was performed with computer software and standardized to the GAPDH product. Quantitative real-time polymerase chain reaction was analyzed by a Lightcycler 480 Detection System (Roche Molecular Biochemicals). RT-qPCR products were detected using SYBR Green. Transcript levels were quantifed by using the 2 ΔCT method (ΔCt = Ct GAPDH -Ct Target ).

Cell proliferation assay
To evaluate the effect of transfection of EGFP-CtBP2 and CtBP2-shRNA, cells were seeded on a 96-well cell culture cluster (Corning Inc., Corning, NY) at a concentration of 2 × 10 4 /well in 100 μL medium and grew overnight. CCK-8 reagents (Dojindo, Kumamoto, Japan) were added to the different subset wells and then incubated at 37°C. The absorbance was quantified using an automated plate reader at a test wave length of 570 nm at different times.

Colony formation assays
Cells were plated in 60 mm plates (0.5 × 10 3 cells per plate) and cultured for 10 days. The colonies were stained with 1% crystal violet for 30 s after fixation with 10% formaldehyde for 5 min.

Wound healing assays
After transfected 48 h, cells were serum starved for 12 h. Then scratching the monolayer with a 10 ml pipette tip, cells were washed with PBS, cultured in 5% FBS-DMEM at 5% CO 2 and 37°C, and photographed under 20×objective lens every 4 h by inverted Leica phasecontrast microscope (Leica DFC 300 FX).

In vitro invasion assay
A 24-well transwell plate (8 µm pore size, Corning, USA) was used to measure the invasive ability of MDA-MB-231 and MCF-7. Chamber inserts were coated with 200 mg/mL of Matrigel and dried overnight under sterile conditions. Then, 1 × 10 5 cells were plated in the top chamber. The experiment was carried out for each cell line in triplicates.

Anchorage-independent growth ability assay
Five hundred cells were trypsinized and suspended in 2 mL complete medium plus 0.3% agar (Sigma, Saint Louis, MI). The agar-cell mixture was planted on top of a bottom layer with 1% agar completed medium mixture. After 10 days, viable colonies that were larger than 0.1 mm were counted. The experiments were repeated at least three times.

Xenografted tumor model
Female nude mice were housed under standard conditions. The animal protocols were done in agreement with SIBS Guide for the Care and Use of Laboratory Animals and approved by Animal Care and Use Committee, Shanghai Institutes for Biological Sciences. Six-week-old female nude mice were divided into 3 groups (n = 8 per group), and the mice were s.c. injected at one site in the left flank with 1 × 10 5 breast cancer cells. The resulting tumors were measured with calipers every 4 days; length, width, and thickness measurements were obtained with calipers and tumor volumes were calculated. Four weeks after injection, tumors were harvested.

Statistical analysis
Statistical analyses were performed using SPSS 18.0 software package (SPSS, Inc., Chicago, IL, USA). The association between CtBP2 and p16 INK4A expression and clinicopathological features was analyzed using χ 2 test. CtBP2, p16 INK4A and Ki-67 expression was studied using the Spearman rank correlation test. Survival curves were calculated using the Kaplan-Meier method, and the log-rank test was used for analysis. Multivariate analysis was performed using Cox's proportional hazards model. A P value < 0.05 was considered as statistical significance.