Stromal fibroblast activation protein alpha promotes gastric cancer progression via epithelial-mesenchymal transition through Wnt/ β-catenin pathway

There were 60 GC cases included in this study, all of which have received radical operation at the Department of Gastrointestinal Surgery, Zhongnan Hospital of Wuhan University (Wuhan, China) from February 2009 to April 2011. Major clinicopathological characteristics including age, gender, tumor diameter, and TNM stages were collected. In addition, the information of follow up was available. TNM stages were determined according to the UICC/AJCC 7th TNM staging system of GC. The primary endpoint for this study was overall survival (OS), which was defined as the interval from the date of surgery to GC related death. Written informed consent was obtained from the patients with the study protocol approved by the ethics committee of Zhongnan Hospital of Wuhan University. The study was undertaken in accordance with the ethical standards of the World Medical Association Declaration of Helsinki.

Routine IHC method was performed for the staining of FAP. The primary antibody was rabbit anti-human monoclonal antibody against FAP (ab227703, Abcam, UK, dilution 1/200), with corresponding horseradish peroxidase (HRP) conjugated secondary antibody (ab6721, Abcam, UK, dilution 1/200). The FAP positive CAFs were indicated by both morphological features and the IHC reaction results. The reaction products were visualized with diaminobenzidine (DAB, DAKO, Denmark). Then the slides were evaluated by two senior pathologists, who were blinded to the patients’ clinical features and outcomes. A consensus was achieved using a multi-headed microscope in case of discrepancy. In brief, at least 4 standard-compliant vision fields of FAP expression (magnification, × 200) per patient was considered to be adequate, with no focus on hotspots. The digital images were captured under Olympus BX51 fluorescence microscope equipped with Olympus DP72 camera (Olympus Optical Co., Ltd., Tokyo, Japan). Identical settings were used for every photograph, so as to minimize the selection bias.

The SGC7901 cell line (human gastric cancer cell lines), GES1 cell line (normal mucosal epithelium cells), and HELF cell line (human embryonic lung fibroblasts; Cat NO.: CL-0281) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% Fetal Bovine Serum (FBS), 100 IU/ml penicillin and 100 mg/ml streptomycin in a humidified atmosphere with 5% CO₂ at 37 °C.

The lentivirus FAP-copGFP (1 × 10⁸ TU/ml) and a negative control (NC) were purchased from GenePharma (Shanghai, China). HELF cells seeded in six-well plates were transfected with control or lentivirus FAP-copGFP according to the manufacturer’s instructions. The multiplicity of infection (MOI) in this study was 50:1. Then puromycin was used to establish the stable transfected HELF cell line (HELF^FAP). SGC7901 co-cultured with HELF^FAP and HELF^NC cells were used for further experiments.

Cholecystokinin-8 (CCK-8) assay (Dojindo, Japan) was performed to detect the cell viability and cell growth. Briefly, 6000 viable gastric cancer cells were seeded in 96-well plates. After specific treatment, each well was mixed with 10 μL CCK-8 and incubated for additional 1 h. The OD values were detected at an absorbance of 450 nm.

A colony formation assay was used to detect cells survival. For clonogenicity analysis, 1000 viable co-cultured SGC7901 cells were placed in six-well plates. Culture medium was changed every two days. After two weeks of incubation, colonies were fixed with 4% paraformaldehyde and stained with crystal violet. The cells were photographed and the numbers of colonies were scored.

SGC7901 cells seeded in 6-well plates were scratched, washed with PBS supplemented with 1% FBS and treated as indicated. The cells were photographed by phase contrast microscope at 24 h in several pre-marked spots. Then the mean distance between both edges of cell free area was calculated.

The polycarbonate membrane in the transwell chambers were precoated with Matrigel with 1:40 dilution (Corning, USA) in 37 °C and air dried. There were 15,000 cells seeded and adhered in each chamber. After 24 h, the cells were fixed with 4% paraformaldehyde (PFA) and stained by 0.1% crystal violet, the number of migrated cells were counted and statistically analyzed. For migration assay, no Matrix gel was required.

SGC7901 cells were placed in 12-well plates overnight, and then treated with compounds according to the manufacturer. Cells were then harvested, washed twice with pre-cold PBS, and evaluated for apoptosis by double staining with FITC-conjugated annexin V and propidium iodide (PI) (MultiSciences, Hangzhou, China) for 30 min in the dark. To assess the cell cycle, harvested cells were labeled with PI (5 mg/ml) in the presence of binding buffer (MultiSciences, Hangzhou, China) in darkness for 30 min.

Total RNA was extracted using RNA simple Total RNA kit (TIANGEN, Beijing, China). cDNA was generated with a first-strand cDNA synthesis Kit (Thermo, Waltham, MA) using the protocol recommended by the manufacturer.

The one-step real-time quantitative PCR were carried out in a 20 μl reaction mixture containing 10 μl 2 × SYBR Premix EX Taq II (Takara, Tokyo, Japan), 0.4 μM primers, and 1 μl of template cDNA. The primers were listed in Additional file 1: Table S1. All real-time RT-PCRs were performed at CFX96 real-time PCR detection system (Bio-Rad, Hercules, CA). Fluorescence was measured at the end of the annealing period of each cycle to monitor amplification. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as internal reference.

Cells were washed with cold PBS twice and prepared in RIPA lysis buffer, and western blot analysis was performed as described previously [18]. Specific primary antibodies used were the following: DKK1 (ab61275, Abcam); LEF1 (ab217378, Abcam); ZO-1 (61–7300, Thermo); Vimentin (ab92547, Abcam); N-cadherin (ab76011, Abcam); E-cadherin (ab1416, Abcam). Anti-GAPDH was purchased from Aspen (Wuhan, China). After incubating with a fluorescein-conjugated secondary antibody (Li-Cor, Lincoln, NE, USA), the membranes were analyzed using an Odyssey fluorescence scanner (Li-Cor, Lincoln, NE, USA).

SGC7901 cells were seeded on 24 mm coverslips, fixed with 4% PFA for 30 min, treated by 0.1% Triton X-100 and blocked in 5% BSA for 1 h at room temperature. Sequentially the fixed cells were incubated with primary antibody at 4 °C overnight (E-cadherin, ab1416, Abcam, dilution 1/50; α-SMA, ab32575, Abcam, dilution 1/300), washed with PBS and incubated with Cy3-labelled or FITC-labelled secondary antibody for 1 h at room temperature. The nuclei were labelled with DAPI (2 mg/ml), and the immunofluorescence staining was analyzed using a fluorescence microscope (Olympus BX5, Olympus Optical Co., Ltd., Tokyo, Japan).

Six-week-old female BALB/cA nu/nu mice were purchased from Vital River Laboratory Animal Technology Company (Beijing, China) and maintained in an Animal Biosafety Level 3 Laboratory at the Animal Experimental Center of Wuhan University. All animal experiments were performed according to the Wuhan University Animal Care Facility and National Institutes of Health guidelines. Approximately 3 × 10⁶ SGC7901 cells and 1 × 10⁶ HELF^FAP cells (Group I, n = 5), 3 × 10⁶ SGC7901 cells and 1 × 10⁶ HELF^NC cells (Group II, n = 5) were harvested and suspended in 200 ml of PBS and Matrigel (BD Bio-science, USA) (1:1) and injected subcutaneously into the right flank of each mouse. The size of subcutaneous tumors was recorded every two days. Five weeks later, mice were sacrificed, and the tumors were removed. The weight of tumors was recorded and statistically analyzed. The xenograft tumor slides were incubated with the following primary antibodies: anti-CD31 was purchased from ABclonal (Boston, USA) and anti-Ki67 from Cell Signaling Technology (Boston, USA). Anti-rabbit or anti-mouse peroxidaseconjugated secondary antibody (ABclonal, Boston, USA) and diaminobenzidine colorimetric reagent solution (Dako, Carpinteria, CA) were used. The staining processes were performed according to standard methods.

A total of 60 patients were included in this study, detailed information about patients’ demographics, clinicopathological characteristics was shown in Table 1. There were 4 groups including I (n = 12), II (n = 13), III (n = 27), and IV (n = 8). FAP was mainly expressed in cancer cells or CAFs (Fig. 1a, b). The positive ratio of FAP was 91.7% in GC tissues (n = 55). The FAP positive CAFs in GC tissues (32.80 ± 19.3) was much higher than that in peritumoral tissues (0.41 ± 0.21), the difference was statically different (P < 0.01).

FAP expression correlated with the tumor diameter (P = 0.024), tumor differentiation degrees (P = 0.002), and TNM stage (P = 0.001), but not correlated with age and gender (P > 0.05 for all). According to the median value of FAP positive CAFs, GC cases were divided into high expression of FAP group (n = 30) and low expression of FAP group (n = 30). The median OS of GC cases with high expression of FAP (30.2 months) was shorter than that with low expression of FAP (37.8 months), the difference was statically different (P < 0.01, Fig. 1c).

The infection efficiency of FAP-copGFP was 100% at 72 h after infection and puromycin-based screening, which was indicated by green fluorescence. FAP expression was significantly elevated in HELF^FAP cells by nearly sixtyfold than HELF^NC cells. The IF results also indicate significantly higher FAP protein expression within HELF^FAP cells (Additional file 2: Figure S1). Therefore, HELF^FAP cells with overexpression of FAP were constructed for further studies, including cell proliferation, migration, invasion, as well as apoptosis.

The proliferation and migration abilities of SGC7901 were significantly elevated by exogenous FAP in dose-dependent manner, as shown in Fig. 2a-c. The co-culture system went a further step to confirm this phenomenon. After co-cultured with HELF^FAP, HELF^NC and HELF cells for 72 h, SGC7901 cells were harvested for CCK8 assays. The OD (450) value was recorded every 24 h to draw the proliferation curve, which indicated that the OD value of SGC7901^FAP was much higher (Fig. 2d). The number of SGC7901^FAP cells colony was also much higher than that of SGC7901^NC (Fig. 2e). Western blot assay indicated that the expression of PCNA and MMP9 protein in SGC7901^FAP were highest (Fig. 2f). The width of injury was lower in SGC7901^FAP cells in 24 h by wound healing assay (Fig. 2g). Cell migration and invasion abilities were determined by Transwell assay. The number of migrated and invasive SGC7901 cells in FAP group was much higher than that in NC group (Fig. 2h).

The cocultured GES1 and SGC7901 cells were treated with cis-platinum meanwhile, then the apoptosis effect was detected by flow cytometry (FCM). Apoptotic SGC7901 cells in NC group was higher than that in FAP group, whereas the result turned to the opposite in GES1 cells (Fig. 3a). The cellular circle of SGC7901 was also detected by FCM to evaluate the potential reasons of apoptosis. However, no differences could be observed between FAP and NC groups (Fig. 3b). Then we hypothesized the potential correlation between FAP and Caspase family considering the apoptosis effect, whereas Western blot assay indicated that no significant differences regarding the expression of Caspase3, Caspase 9, Bax and Bcl-2 between FAP and NC groups (Additional file 3: Figure S2).

Exogenous FAP promotes EMT in dose-dependent manner. The expression of E-cadherin and ZO-1 were reduced, while that of N-cadherin and Vimentin were increased by qRT-PCR assay (Fig. 4a), and Western blotting assay (Fig. 4b). In addition, the DKK1 and LEF-1 protein, which could be participated in Wnt/ β-catenin pathway, were also increased with more exogenous FAP. The result also accompanied in SGC7901 cells co-cultured in NC and FAP groups.

The morphology of SGC7901 cells in FAP group tended to be fibroblast-like, long fusiform, which was indicated by red arrows in Fig. 4c. The expression of E-cadherin and ZO-1 were reduced, while that of N-cadherin and Vimentin were increased in SGC cells of SF group by Western blotting assay. Similarly, the DKK1 and LEF-1 protein were also increased (Fig. 4d). The expression of E-cadherin was reduced, while that of α-SMA was increased both in GES1 cells and SGC7901 cells of FAP group by the IF staining (Fig. 4e).

To investigate the in vivo effects of stromal FAP, we examined the tumor promoting effect of FAP in a xenograft gastric cancer nude mouse model. SGC7901 cells (3 × 10⁶) were implanted subcutaneously in the right flank of nude mice, accompanied with HELF^FAP (1 × 10⁶) (n = 5) and HELF^NC (1 × 10⁶) (n = 5) cells, respectively. The combination of SGC7901 and HELF^FAP was much more effective in elevating tumor burden (Fig. 5a). The tumor volume and weight in the NC group were significantly lower than FAP group (Fig. 5b, c). Ki67 and CD31were examined by immunohistochemistry in the tumor sections. Both Ki67 and CD31 expression were elevated in FAP group. Taken together, stromal FAP promotes GC progression in a xenograft gastric cancer nude mouse model.