Vasohibin-2 promotes proliferation in human breast cancer cells via upregulation of fibroblast growth factor-2 and growth/differentiation factor-15 expression

Vasohibin-2 (VASH2) is an angiogenic factor, and has been previously reported to be a cancer-related gene, with cytoplasmic and karyotypic forms. In the current study VASH2 expression in human breast cancer tissue and adjacent non-cancerous tissue was investigated with immunohistochemistry. MCF-7 and BT474 human breast cancer cells were transfected with lentiviral constructs to generate in vitro VASH2 overexpression and knockdown models. In addition, BALB/cA nude mice were inoculated subcutaneously with transfected cells to generate in vivo models of VASH2 overexpression and knockdown. The effect of VASH2 on cell proliferation was investigated using a bromodeoxyuridine assay in vitro and immunohistochemistry of Ki67 in xenograft tumors. Growth factors were investigated using a human growth factor array, and certain factors were further confirmed by an immunoblot. The results indicated that the expression level of cytoplasmic VASH2 was higher in breast cancer tissues with a Ki67 (a proliferation marker) level of ≥14%, compared with tissues with a Ki67 level of <14%. VASH2 induced proliferation in vitro and in vivo. Four growth factors activated by VASH2 were identified as follows: Fibroblast growth factor 2 (FGF2), growth/differentiation factor-15 (GDF15), insulin-like growth factor-binding protein (IGFBP)3 and IGFBP6. FGF2 and GDF15 may contribute to VASH2-induced proliferation. The current study identified a novel role for VASH2 in human breast cancer, and this knowledge suggests that VASH2 may be a novel target in breast cancer treatment.


Introduction
The vasohibin (VASH) family consists of two members, VASH1 and VASH2 (1). VASH1 was initially identified as a regulator of negative feedback in angiogenesis induced by vascular endothelial growth factor (VEGF) or fibroblast growth factor 2 (FGF2) (2,3). VASH2 is a VASH1 homolog expressed in mononuclear cells that has been demonstrated to act as an angiogenesis stimulator in a mouse model of hypoxia-induced subcutaneous angiogenesis (3). VASH2 is also involved in the proliferation of hepatic (4) and ovarian (5,6) cancer.
It was previously demonstrated that there are two types of VASH2: Nuclear and cytoplasmic (7). In the present study, the focus was on cytoplasmic VASH2, thus all subsequent mention of VASH2 refers to the cytoplasmic form. VASH2 expression was investigated in human breast cancer in the current study; rabbit polyclonal anti-human VASH2 antibodies were produced and successfully used in immunoblotting and immunohistochemical analysis (7).
In the present study, VASH2 expression levels were indicated to be higher in grade 3 vs. grade 1-2 tissues, and in tissues with a level of Ki67 ≥14%. Ki67 is a marker for breast cancer proliferation. It was hypothesized that VASH2 is associated with cell proliferation in breast cancer, and in order to investigate the proliferative function of VASH2 in breast cancer cells and the underlying mechanism, VASH2 overexpression and knockdown in vitro and in vivo models were established. VASH2 produced a significant proliferative effect in vitro and in vivo. Human growth factor array demonstrated that VASH2 promoted proliferation in breast cancer cells via the upregulation of FGF2 and growth/differentiation factor-15 (GDF15) expression. The present study identified a novel role for VASH2 in human breast cancer, and this knowledge may lead to the possibility of VASH2 as a novel target in breast cancer treatment. Immunoblotting. Whole cell lysates were prepared in radioimmunoprecipitation assay buffer (Beyotime, Nantong, China) and blotted using the following primary antibodies: Rabbit polyclonal anti-VASH2 [prepared as described in (7)]; rabbit polyclonal anti-FGF2 (Sigma-Aldrich, St. Louis, MO, USA); goat polyclonal anti-GDF15 (Santa Cruz Biotechnology, Santa Cruz, CA, USA); and mouse monoclonal anti-GAPDH (Beyotime, Nantong, China). The secondary antibodies used for detection were horseradish peroxidase (HRP)-conjugated goat anti-mouse immunoglobulin G (IgG) (CWbio, Shanghai, China) and HRP-conjugated donkey anti-rabbit IgG (CWbio).

Vasohibin-2 promotes proliferation in human breast cancer cells via upregulation of fibroblast growth factor-2 and growth/differentiation factor-15 expression
Immunohistochemistry. Immunohistochemical staining of the clinical samples was performed as previously described (7). Xenograft tumors were harvested from mice and stained with the following primary antibodies: Rabbit polyclonal anti-VASH2 [prepared as previously described (7)], rabbit polyclonal anti-estrogen receptor (ER)α and Ki67 (Maixin Biotech, Fuzhou, China), and mouse monoclonal antibodies targeting the progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2; Maixin Biotech). VASH2 staining intensity was classified as weak or strong. ERα, PR and HER2 staining were classified as positive or negative. HER2 +/++ was also classified as negative; only HER2 +++ was classified as positive. Ki67 staining was classified as <14% and ≥14%. VASH2 staining intensity was classified as Low (negative or weak staining) and high (middle or strong staining).
In vivo tumorigenesis. MCF7-EGFP or MCF7-VASH2 cells (2x10 6 ) were bilaterally injected subcutaneously into the flanks of eight mice. Eighty days later, the mice were sacrificed and the xenograft tumors harvested. In addition, BT474-scramble or BT474-shVASH2 cells (1x10 6 ) were bilaterally injected subcutaneously into the flanks of seven mice and the xenograft tumors were harvested at 60 days post-inoculation. Tumor volume was calculated as follows: (Width 2 x length)/2.
Antibody array. Quantitative sandwich-based antibody array (#QAH-GF-1; RayBiotech, Guangzhou, China) was used to detect 40 human growth factors in lysates of the MCF7-EGFP, MCF7-VASH2, BT474-scramble and BT474-shVASH2 cells. All detection services were provided by RayBiotech (Norcross, GA, USA). Each antibody produced four dots, and the averages of the median signal intensities were used for all calculations. Fold change of >1.5 or <0.66 compared with controls was selected as the distinction between overexpression or knockdown, respectively.
Statistical analysis. Statistical analysis was performed using SPSS, version 13.0 (SPSS, Inc., Chicago, IL, USA). Comparisons between treated and control groups were conducted using Student's t-test, and P<0.05 was considered to indicate a statistically significant difference. Pearson's χ 2 test was used to compare rates of the data in Table I.

Results
VASH2 staining and the clinicopathological characteristics of breast cancer. VASH2 expression levels in 99 human breast cancer tissue samples were assessed by immunohistochemical analysis. Table I shows the association between VASH2 staining and the clinicopathological characteristics of breast cancer. VASH2 expression was high in 70/99 (70.7%) breast cancer tissues. VASH2 staining was generally higher in grade 3 tissues and those with Ki67 ≥14% (Pearson χ 2 , P<0.001). These findings indicate that VASH2 may promote proliferation in human breast cancer cells. Fig. 1 displays the representative images of high and low VASH2 expression levels in human breast cancer tissues.

VASH2 promotes proliferation in human breast cancer cells in vitro and in vivo.
The VASH2 expression levels in VASH2-overexpressing (MCF7-VASH2) and VASH2 knockdown cells (BT474-shVASH2) were confirmed using immunoblotting (Fig. 2), indicating the successful establishment of in vitro models of VASH2 overexpression and knockdown. The proliferative function of VASH2 was investigated in vitro using cell proliferation ELISAs. Results indicated that the optical density at 450 nm (OD450) of MCF7-VASH2 cells was significantly higher than that of MCF7-EGFP cells, while the OD450 of BT474-shVASH2 cells was significantly lower than that of BT474-scramble cells (Fig. 3A, P<0.05). These data indicate that VASH2 induced cell proliferation in vitro. MCF7-EGFP, MCF7-VASH2, BT474-scramble or BT474-shVASH2 cells were injected into the flanks of nude mice. At 80 days post-inoculation, mice that had been injected with MCF7-VASH2 cells had developed significantly larger tumors than mice injected with MCF7-EGFP cells (Fig. 3B, P<0.05). At 60 days post-inoculation, mice that had been injected with BT474-shVASH2 cells had developed significantly smaller tumors than mice injected with     BT474-scramble cells (Fig. 3B, P<0.05). The levels of Ki67 staining in MCF7-VASH2 xenograft tumors were significantly higher than in MCF7-EGFP xenograft tumors (Fig. 3C, P<0.05), and the levels in BT474-shVASH2 xenograft tumors were significantly lower than in BT474-scramble xenograft tumors (Fig. 3C, P<0.05). These findings indicate that VASH2 also induces proliferation in vivo.

VASH2 induces FGF2, GDF15, insulin-like growth factor-binding protein (IGFBP)3 and IGFBP6 expression.
Using sandwich-based antibody array, 40 human growth factors were detected in MCF7-EGFP, MCF7-VASH2, BT474-scramble and BT474-shVASH2 cell lysate samples. Table II denotes the growth factor array results. A fold change of >1.5 or <0.66 compared with controls was selected as the definition of up-and downregulation, respectively. FGF2, GDF15, IGFBP3 and IGFBP6 expression levels were elevated in the VASH2-overexpressing MCF7-VASH2 cells and were reduced in the VASH2-knockdown BT474-shVASH2 cells; this indicates that they may have contributed to the specific reaction induced by VASH2 (Fig. 4). The protein expression levels of FGF2 and GDF15 were then detected by immunoblot, which confirmed the results of the antibody array (Fig. 4D). Androgen receptor, brain-derived neurotrophic factor, bone morphogenetic protein-4, epidermal growth factor (EGF), heparin-binding EGF, IGFBP2, and VEGF receptor-2 fold changes were also >1.5 (Table II) in the VASH2-overexpressing MCF7-VASH2 cells, but were not altered in VASH2-knockdown BT474-shVASH2 cells.
Notably, the fold changes for platelet-derived growth factor-AA and placental growth factor in the VASH2-overexpressing and knockdown cells were <0.66 (Table II), indicating that they may have contributed to a non-specific reaction. These data suggest that VASH2 upregulates FGF2, GDF15, IGFBP3 and IGFBP6 expression.

Discussion
VASH2 is involved in tumor proliferation (4,5). Rabbit anti-human VASH2 polyclonal antibodies were generated and used in immunohistochemical analysis of VASH2 expression

A B
C levels (7). In the present study, VASH2 expression in clinical human breast cancer tissues was investigated, and significantly higher levels of VASH2 in grade 3 and Ki67 ≥14% breast cancer tissues were detected. Ki67 expression levels vary during the cell cycle; the levels are low during the G 1 and early S phases, and high during mitosis, followed by a sharp reduction during anaphase and telophase. Ki67 cannot be detected during the G 0 resting phase (8,9). Ki67 inhibition leads to the arrest of cell proliferation (10,11). Ki67 expression levels increase progressively from benign breast disease to ductal carcinoma in situ to invasive breast cancer (12)(13)(14). In addition, Ki67 is considered to be a good proliferation marker in clinical practice (15). In the current study, it was hypothesized that VASH2 is associated with cell proliferation, and to confirm the possible function of VASH2 in proliferation, in vitro and in vivo models of VASH2 overexpression and knockdown were developed. Analysis of the models indicated that VASH2 promotes the proliferation of breast cancer cells in vitro and in vivo. Various cancer cells synthesize growth factors to which they are responsive (16), and these growth factors are important in the processes of tumor cell clonal expansion, angiogenesis, invasion and metastasis (17). It was hypothesized that VASH2 may induce proliferation via activation of growth factor expression. To confirm this, human growth factor array analysis was performed using VASH2-overexpression and knockdown in vitro models. A total of 40 common proliferation-related growth factors in four cell lysate samples (MCF7-VASH2, MCF7-EGFP, BT474-shVASH2 and BT474-scramble) were investigated. VASH2 increased the expression of four growth factors: FGF2, GDF15, IGFBP3 and IGFBP6. FGF2 (18) induces cell proliferation in various types of cancer. GDF15 serves a function in cell proliferation, apoptosis, metastasis and angiogenesis, through autocrine and paracrine signaling (19).
IGFBP3 and IGFBP6 are IGF-binding proteins that inhibit IGFs, therefore functioning as tumor suppressors (20,21). However, IGFBP3 overexpression in breast cancer is linked to poor prognosis (22,23). Previously, it has been reported that IGFBP3 promotes cancer cell growth via an IGF-independent manner (24). It was also reported that IGFBP6 promoted cancer cell migration in an IGF-independent manner (21). Therefore, the function of VASH2-regulated IGFBP3 and IGFBP6 expression remains unclear. It is possible that the VASH2-induced proliferation occurred via upregulation of the expression of FGF2 and GDF15.
The present study demonstrated a high level of VASH2 expression in breast cancer cells, and that VASH2 functions as an inducer of growth factor expression, promoting cell proliferation in breast cancer. In conclusion, the current study indicated that VASH2 may have potential as a novel anticancer target.