Fibroblast growth factor receptor 4 (FGFR4) belongs to the tyrosine kinase receptor family. Little is known about the effect of FGFR4 on gastric cancer (GC). Therefore, the objective of the current study was to elucidate the role of FGFR4 in the tumorigenesis and progression of GC.
FGFR4 and some common prognosis markers, including p53, neu, and proliferating cell nuclear antigen (PCNA), were detected in 71 tissue samples from patients with GC using immunohistochemical analysis. In addition, a series of functional assays were carried out using small interfering RNA (siRNA) and included proliferation assays, clone assays, and apoptosis detection.
Cytoplasmic FGFR4 expression in GC tissues was negative (7% of samples), low (14.1% of samples), intermediate (40.8%), and high (38% of samples). FGFR4 expression was associated with lymph node status and with PCNA and neu expression (P < .05). The 5-year relative survival rate was 61.5% in patients who had GC with low FGFR4 expression but was only 42% in patients who had high FGFR4 expression (P = .058). A subgroup analysis of the patients who had high FGFR4 expression revealed that those with stage III and IV disease had a worse prognosis (P = .044). Moreover, knockdown of FGFR4 expression led to decreased proliferation and an increased rate of apoptosis in the MKN45 and SGC7901 GC cell lines (P < .05). Western blot analysis demonstrated that the expression of caspase 3 increased, whereas the expression of extra-large B-cell lymphoma (Bcl-xL) decreased in MKN45 and SGC7901 cells after FGFR4-siRNA transfection.
Gastric cancer is the fourth most frequent malignant tumor worldwide.1 Recently, more and more studies have indicated that growth factor receptors are involved in the tumorigenesis and progression of malignant tumors. Among them, the fibroblast growth factor (FGF)/fibroblast growth factor-receptor (FGFR) system plays a pivotal role in cancer development through its effects in angiogenesis, differentiation, survival, and motility.2, 3 Some reports in the literature have demonstrated that FGFR4, as a member of the FGFR family, plays a crucial role in tissue repair, embryonic development, and so on.4, 5 However, the role of FGFR4 in human cancer, and especially in gastric cancer, has not been explicitly clarified.
Our previous research using real-time polymerase chain reaction (PCR) analysis revealed that FGFR4 messenger RNA (mRNA) levels were increased remarkably in gastric cancer tissues compared with the levels in corresponding normal tissues.6 Furthermore, high expression levels of FGFR4 were verified in renal cell carcinoma and pancreatic carcinoma.7, 8 It also was reported that liver tissues has the highest transcript expression of FGFR4 compared with tissues from other major organs.9 Recently, Ho et al10 reported that FGFR4 contributed significantly to hepatocellular carcinoma (HCC) progression by modulating α-fetoprotein secretion, proliferation, and antiapoptosis, and Roidl et al11 reported that the up-regulation of FGFR4 was associated with resistance to chemotherapy in breast cancer cell lines. To our knowledge, there has been no related research to date about the function of FGFR4 in the progression of gastric cancer.
To identify the vital role of FGFR4 and to explore the mechanism of FGFR4 in the tumorigenesis and development of gastric cancer, we conduced immunohistochemical analyses of FGFR4 and of some common prognosis markers, such as p53, neu, and proliferating cell nuclear antigen (PCNA), in tissue samples from patients with gastric cancer. MKN45 and SGC7901, the 2 most common gastric cancer cell lines in our experiment center, have different malignant potential. In addition, our preliminary study indicated that expression levels of FGFR4 in MKN45 and SGC7901 cells were significantly higher than the levels in 2 other gastric cancer cell lines, SNU-1 and SNU-16. Therefore, we choose MKN45 and SGC7901 cells for FGFR4 small interfering RNA (siRNA) assays to investigate the knockdown of FGFR4 expression. In addition, we performed a series of functional assays in the MKN45 and SGC7901 cell lines, including proliferation assays, clone assays, and apoptosis detection. The findings from this study indicate that FGFR4 may play an extremely crucial role in gastric cancer growth, apoptosis, and progression, pointing to its potential as a novel drug target against gastric cancer.
MATERIALS AND METHODS
Paraffin-embedded tissue specimens of 71 consecutive, primary gastric cancers were collected from the tissue bank at Shanghai Cancer Center, Fu Dan University (Shanghai, China). All patients were diagnosed and underwent radical gastrectomy at the Department of Abdominal Surgery in 2003. Treatment decisions were based on consensus recommendations at the time. All diagnoses were documented pathologically, and personal files were retrieved to obtain clinical data (Table 1) with approval of the hospital's ethics committee. Follow-up was performed routinely, and the median follow-up of patients who remained alive at the time of analysis was 62 months (range, 4-83 months).
Table 1. Relation Between Fibroblast Growth Factor Receptor 4 Expression and Clinicopathologic Features of Patients With Gastric Cancer
High FGFR4 expression indicates a cytoplasmic immunostaining score of +++.
Low FGFR4 expression indicates cytoplasmic immunostaining scores of −, +, and ++.
Tumor size, cm
Lymph node status
Rabbit polyclonal anti-FGFR4 antibody was obtained from Santa Cruz Biotechnology (Santa Cruz, Calif). Rabbit monoclonal antibody against extra-large B-cell lymphoma (anti-Bcl-xL), rabbit monoclonal anticaspase 3, and mouse monoclonal antibody against glyceraldehyde 3-phosphate dehydrogenase all were purchased from Cell Signaling Technology (Beverly, Mass). Mouse monoclonal anti-p53, anti-neu, anti-c-myc, and anti-PCNA antibodies all were obtained from Dako (Hamburg, Germany). Secondary horseradish peroxidase-conjugated antibodies goat antimouse and goat antirabbit (Sigma-Aldrich Corporation, St. Louis, Mo) were used. Goat antirabbit immunoglobulin G (IgG)-fluorescein isothiocyanate (FITC) antibody (catalog no. T6778) also was purchased from Sigma-Aldrich Corporation.
Expression levels of FGFR4, p53, neu, and PCNA in postoperative paraffin-embedded tumor specimens from all selected patients were detected with immunohistochemistry. The concentrations of antibody and the sites of positive expression were as follows: anti-FGFR4, 1:500 dilution (positive site, cytoplasm); anti-p53, 1:100 dilution (positive site, nucleus); anti-neu, 1:100 dilution (positive site, membrane); and anti-PCNA, 1:100 dilution (positive site, nucleus). The detailed staining procedures strictly followed the supplier's recommendation. Negative controls were obtained by incubating parallel slides with the primary antibodies omitted. In addition, sections with confirmed positive staining from each run served as positive controls.
Immunohistochemical Staining Scores
All slides were evaluated semiquantitatively by 2 independent pathologists (X.Y.Z. and L.Y.) who were blinded to patients' clinical data when scoring immunohistochemical results in archival tissue samples. Cytoplasmic FGFR4 immunostaining was scored on a 4-point scale as negative (−), low (+), intermediate (++), or high (+++).12 In the final analysis, the immunostaining for each protein was determined as either positive or negative by using the following cutoff values: Staining for p53 was interpreted as positive when >10% of the tumor nucleus was stained; PCNA expression was interpreted as strongly positive when >60% of the tumor nucleus was stained; and neu expression was interpreted positive when >10% of the tumor membrane was stained.
Cell Lines and Cell Culture
The human gastric cancer cell lines MKN45 and SGC7901 were purchased from Chinese Academy of Sciences Cell Bank of Type Culture Collection (Shanghai, China). Cell lines were cultivated in RPMI-1640 medium (Gibco, Grand Island, NY) with 10% fetal bovine serum (Gibco), 100 μg/mL penicillin, and 100 μg/mL streptomycin (Caisson Laboratories, Inc., Logan, Utah) at 37°C in a humidified 5% CO2 atmosphere, as described previously.13
FGFR4 Silencing Using siRNAs
The FGFR4 siRNAs and 6-carboxyfluorescein (FAM) negative control siRNAs were purchased from Shanghai Genepharma RNAi Company (Shanghai, China). We used the following siRNA sequences: For FGFR4 siRNA (si1792), the sense sequence was 5′-GCCGACACAA GAACAUCAUtt-3′, and the antisense sequence was 5′-AUGAUGUUCUUGUGUCGGCtt-3′. For the negative control (control siRNA [siCtrl]), the sense sequence was 5′-UUCUCCGAACGUGUCACGUtt-3′, and the antisense sequence was 5′-ACGUGACACGUUCGGA GAAtt-3′. Each siRNA was transfected into gastric cancer cells using Lipofectamine 2000 (Invitrogen Life Technologies, Carlsbad, Calif) according to the manufacturer's instructions. The volume ratio of lipofectamine 2000 and siRNA was 1:2. Transfected cells were assayed at indicated time points. Seventy-two hours after transfection, the treated cells were harvested, and specific silencing of FGFR4 was confirmed at the RNA and protein levels by Western blot and immunofluorescence analyses.
Cell Lysis and Western Blot Analysis
Cells were harvested 72 hours after transfection, and whole-cell lysates were prepared using the Mammalian Protein Extraction Reagent (Merck, Germany) in accordance with the manufacturer's instructions. Protein concentrations of samples were determined by using a bicinchoninic acid protein assay (Pierce, Rockford, Ill). Protein samples (40 μg of each protein) boiled for 5 minutes were separated in 10% SDS-polyacrylamide gels and transferred onto polyvinylidine fluoride membranes. Membranes were blocked for 1 hour at room temperature with phosphate-buffered saline (PBS) containing 0.05% Tween-20 and 5% nonfat dried milk and then were incubated overnight at 4°C with primary antibodies under manufacturer-recommended conditions. Immunoblots were washed 3 times with PBS containing 0.05% Tween-20 and 1% nonfat milk and were incubated with secondary antibodies conjugated with horseradish peroxidase against mouse IgG or rabbit IgG for 1 hour at room temperature. Immunoreactive proteins were visualized using the ECL detection system (Image Quant LAS-3000; General Electric Company, Fairfield, Conn). Three independent Western blot assays were performed for each sample.
Cells (2 × 104 cells per well) that were transfected with si1792 and siCtrl for 48 hours were incubated in 96-well culture plates in 100 μL medium. Twenty-four hours later, transfected cells were fixed with 4% paraformaldehyde, permeabilized with Triton X-100, and stained with FGFR4 antibody at 4°C overnight. The cells were incubated with FITC secondary antibody solution at room temperature for 1 hour. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI), and the expression of FGFR4 was observed and analyzed through a converted fluorescence microscope.
Cell Proliferation and Clonogenic Assays
Cells that were transfected with si1792 and siCtrl for 48 hours were detached using trypsin and counted. Then, 2 × 103 cells per well were incubated in 96-well culture plates (Corning Inc., Corning, NY) in 100 μL medium. After culturing for 1 day, 2 days, 3 days, 4 days, and 5 days, the supernatant was removed, and cell growth was detected using Cell Counting Kit-8 (CCK-8) (Dojindo Laboratories, Kumamoto, Japan) according to the manufacturer's instructions. Absorbance was measured at 450 nm using a microplate reader. After cell transfection for 48 hours, an equal number of cells from the negative and mock groups and from the FGFR4-knocked down group were harvested in triplicate and replated onto 60-mm Petri dishes. After 12 to 14 days of incubation, the plates were gently washed and stained with crystal violet. Viable colonies that contained >50 cells were counted. All proliferation and clonogenic assays were performed independently at least 3 times, and all experimental data were analyzed using the SPSS statistical software package (version 13.0; SPSS, Inc., Chicago, Ill).
Cells were transfected with 6 μL of si1792 or siCtrl using 3 μL lipofectamine 2000 according to the manufacturer's protocol. After 48 hours, the cells were harvested. The annexin V and propidium iodide (PI) for flow cytometry were purchased from Invitrogen (catalog no. V13241; Invitrogen Life Technologies) for detecting apoptosis, and 5 μL annexin V and 1 μL 100 μg/mL PI working solution were added to each 100 μL of cell suspension. The cells were incubated at room temperature for 15 minutes; then, 400 μL of 1 times annexin-binding buffer were added and mixed gently, and the samples were kept on ice according to the manufacturer's instructions. Thereafter, all samples were analyzed by a FACSCalibur flow cytometer with CellQuest software (BD Biosciences, Franklin Lakes, NJ). All experiments to detect apoptosis were performed independently at least 3 times, and all experimental data were entered into the SPSS 13.0 software program for analysis.
Statistical analyses were performed using the statistical software package SPSS 13.0 (SPSS, Inc.). Molecular analyses were carried out with the investigators blinded to clinical data. Associations between categorical variables were analyzed using the Pearson chi-square test. Survival curves were plotted according to the Kaplan-Meier method, and comparisons between groups were performed using the log-rank test. Data are expressed as the mean ± standard deviation of 3 individual experiments done in triplicate. The Student t test was used to compare data between 2 groups. One-way analysis of variance (ANOVA) and the Dunnett test were used to compare data between 3 or more groups. P values <.05 were considered statistically significant.
Association Between FGFR4 Expression and Clinicopathologic Characteristics
Cytoplasmic FGFR4 immunostaining in gastric cancer tissue samples was 7% negative (−), 14.1% low (+), 40.8% intermediate (++), and 38% high (+++) (Fig. 1). To study correlations between FGFR4 expression and clinicopathologic characteristics, the sample were divided into 2 groups for analysis: a low FGFR4 expression group (including −, +, and ++ samples) and a high FGFR4 expression group (including +++ samples). FGFR4 expression was associated positively with lymph node status, neu expression, and PCNA expression, as indicated in Table 1, and the differences were statistically significant (P = .020, P = .026, and P = .021, respectively).
An analysis of prognosis indicated that the 5-year relative survival rate (RSR) was 61.5% in patients who had gastric cancer with low FGFR4 expression but only 42% in patients who had gastric cancer with high FGFR4 expression, although the difference was not statistically significant (P = .058) (Fig. 2A). A subgroup analysis revealed that the prognosis for both groups was similar among patients who had stage I/II disease (Fig. 2B). However, among the patients who had high FGFR4 expression, patients with stage III/IV disease had a worse prognoses (P = .044) (Fig. 2C). Furthermore, between the 2 groups, significant difference in survival was observed for patients with N1 through N3 lymph node status (Fig. 2D).
FGFR4-siRNA Knocks Down FGFR4 Expression in Gastric Cancer Cells
First, we performed an immunofluorescence assay to verify the efficiency of FGFR4 si1792. Green fluorescence cytoplasmic staining was considered FGFR4-positive expression. Compared with the negative and mock groups, green fluorescence staining in MKN45 and SGC7901 cells that were treated with FGFR4 si1792 for 72 hours was weakened significantly, which meant that si1792 could knock down FGFR4 expression (Fig. 3A-D). Furthermore, the results from Western blot analysis revealed that FGFR4 protein expression in the si1792 group decreased remarkably comparing with that in the mock and negative groups, in accordance with our immunofluorescence assay results (Fig. 3E).
FGFR4 Expression Affects the Proliferation Ability of Gastric Cancer Cells
To study the role of FGFR4 in the ability of cells to proliferate, we used MKN45 and SGC7901 cells for the transient knockdown of FGFR4 by FGFR4 siRNA (si1792). After 48 hours of transfection, the CCK-8 kit was used to evaluate the proliferative ability of MKN45 and SGC7901 cells with different treatments. In MKN45 cells, absorbency values in the si1792 group on Days 5, 6, and 7 after transfection were 0.606 ± 0.072, 1.009 ± 0.098, and 1.349 ± 0.081, respectively; whereas the absorbency values in the negative group were 0.957 ± 0.092, 1.411 ± 0.096, and 1.886 ± 0.095, respectively (Fig. 4A). Similar results were observed in SGC7901 cells, with absorbency values in the si1792 group on Days 5, 6, and Day 7 after transfection of 0.891 ± 0.149, 1.333 ± 0.108, and 1.760 ± 0.108, respectively, and absorbency values in negative group of 1.306 ± 0.062, 1.792 ± 0.064, and 2.216 ± 0.066, respectively (Fig. 4B). Obviously, transfection of si1792 efficiently suppressed the proliferation potential of MKN45 and SGC7901 cells compared with negative siRNA-transfected cells from Days 5, 6, and 7 after transfection (P < .05; Student t test) (Fig. 4A,B).
Furthermore, in MKN45 cells, cloning efficiency was 11.57% ± 1.31% in the si1792 group, 20.47% ± 1.00% in the negative group, and 18.57% ± 1.46% in the mock group; whereas the cloning efficiency of SGC7901 cells was 8.52% ± 1.45%, 21.25% ± 1.21%, and 19.07% ± 1.32%, respectively, which revealed that cloning efficiency in the si1792 group was remarkably lower than that in the negative and mock groups with a difference that was statistically significant (P < .05; 1-way ANOVA) (Fig. 4C,D). In other words, transfection of si1792 observably inhibited the proliferative ability of MKN45 and SGC7901 cells, in accordance with our CCK-8 assay results.
Knockdown of FGFR4 Increases Cell Apoptosis and Caspase 3 Expression in Gastric Cancer Cells
To study the role of FGFR4 in cell apoptosis, MKN45 and SGC7901 cells were transiently transfected with FGFR4 siRNA (si1792) for 48 hours, and the apoptotic cells were detected by flow cytometry. In MKN45 cells, the apoptotic rate was 25.37% ± 0.67% in the si1792 group, 9.57% ± 0.86% in the negative group, and 10.62% ± 0.86% in the mock group; whereas, in SGC7901 cells, the apoptotic rate was 23.31% ± 0.65%, 10.01% ± 0.89%, and 9.61% ± 0.65%, respectively. The rate of apoptosis in MKN45 and SGC7901 cells increased remarkably in the si1792 group compared with the rates in the negative and mock groups, as illustrated in Figure 5A-F. There was a statistically significant difference in the 2 cell lines with different treatments (P < .05; 1-way ANOVA).
We also used the apoptosis-related molecules Bcl-xL and caspase 3 were to evaluate the apoptotic ability of MKN45 and SGC7901 cells with under different treatment conditions. Figure 5G indicates that, compared with negative and mock groups, caspase 3 expression increased while Bcl-xL expression decreased in MKN45 and SGC7901 cells that were treated with si1792.
FGFR4, which belongs to the FGFR family, reportedly contributes to oncogenic cell transformation in different human cancers. FGFR4 plays a vital role in influencing the biologic characteristics of tumor cells, as reported mainly with regard to disease progression in patients with hepatocellular carcinoma, breast cancer, lung cancer, and other tumors.10, 11 However, to our knowledge, there is very little related research in terms of the role of FGFR4 in the tumorigenesis and development of gastric cancer. Therefore, we investigated the expression of FGFR4 in gastric cancer tissues and performed a series of functional assays to identify the vital role of FGFR4 and to explore the mechanism of FGFR4 in gastric cancer using siRNA-mediated down-regulation of FGFR4 expression.
Our previous study indicated that FGFR4 mRNA expression was increased remarkably in gastric cancer tissues.6 In the current study, immunohistochemical analysis indicated that there was 95% cytoplasmic staining of FGFR4 in gastric cancer tissues (including +, ++, and +++ samples), in accordance with our previous results. Furthermore, the probability of lymph node metastasis in patients who had high FGFR4 expression was much greater than that in patients who had low FGFR4 expression (P = .05), indicating high FGFR4 expression may be associated with more malignant biologic behavior and a worse prognosis. When combined with follow-up data, the 5-year relative survival rate was much higher in patients with gastric cancer who had low FGFR4 expression, although the difference was not statistically significant (P = .058), possibly because of the small sample size of patients with gastric cancer in our study. A subgroup analysis indicated that, among the patients who had high FGFR4 expression, those with stage III/IV disease had a worse prognosis, in accordance with a previous report that FGFR4 expression influenced disease progression in patients with breast cancer.11
The p53 protein is the product of a tumor suppressor gene, which modulates cell proliferation though control of the G1 arrest checkpoint in the cell cycle, as reported previously.14 Patients with nonfunctional p53 who have gastric cancer usually have more positive lymph nodes and tend to have a dismal prognosis.15 The HER-2/neu protein, which is extensively homologous to the epidermal growth factor receptor, is intimately involved in normal cell proliferation and tissue growth, has been studied most in the field of breast carcinoma, and is correlated with a poor prognosis. Moreover, some studies have demonstrated that neu amplification may be an independent prognostic factor in patients with gastric cancer.16 PCNA originally was identified as an antigen that was expressed in the nuclei of cells during the DNA synthesis phase of the cell cycle. PCNA is important for both DNA synthesis and DNA repair. Patients who have gastric cancer with high PCNA expression may have a worse prognosis.17 In the current study, FGFR4 expression was associated positively with neu expression and PCNA expression, suggesting that patients who have gastric cancer with high FGFR4 expression may have a more malignant phenotype and a worse prognosis.
In our study, using the CCK-8 kit to conduct a proliferation assay revealed that the transfection of si1792 efficiently suppressed the proliferative potential of MKN45 and SGC7901 cells compared with negative siRNA-transfected cells. Obviously, influencing the ability of cells to proliferate is an effect of FGFR4 signaling in gastric cancer. These findings suggest that the inactivation of FGFR4 may reduce tumor growth in gastric cancer. Wang et al reported that FGFR4 stabilization was associated with enhanced proliferation and anchorage-independent growth in vitro in human prostate cancers.18 Furthermore, FGFR4 knockdown using inducible short hairpin RNA significantly reduced colony-forming ability in vitro and tumor growth in vivo.19 Cloning efficiency also is an index that reflects cell proliferation. Our results indicated that cloning efficiency in the si1792 transfection group was remarkably less than that in the negative and mock groups, in accordance with the results from our proliferation assay.
Antiapoptotic effects often contribute to cancer cell survival and chemoresistance of cancer cells (eg, doxorubicin resistance).20 Antiapoptotic effects have been reported in FGFR1 and FGFR3.21, 22 Recently, Roidl et al observed that an antiapoptotic signaling pathway was initiated by FGFR4, regulating the expression of Bcl-xL through the mitogen-activated protein kinase cascade.11 Bcl-xL, a prominent downstream effector molecule of antiapoptosis, is up-regulated in different tumors, such as breast and liver cancer, frequently leading to chemoresistance. In the current study, the apoptotic rate in MKN45 and SGC7901 cells was elevated remarkably and Bcl-xL expression was weakened in the si1792 group compared with the negative and mock groups. In other words, the up-regulation of FGFR4 may contribute to an antiapoptotic effect in MKN45 and SGC7901 cells, which is in line with the report by Roidl et al in breast cancer cells. Furthermore, the expression by MKN45 and SGC7901 cells of caspase 3, which is a pivotal molecule during cell apoptosis, increased in the si1792 group compare with expression the negative and mock groups, which may establish more adequately the vital role of FGFR4 in regulating apoptosis. Similar results were reported from a study in hepatocellular carcinoma in which caspase 3 expression gradually increased after an increase in the dose of FGFR4 inhibitor.10
In conclusion, to our knowledge, this study is the first detailed exploration of the pivotal role of FGFR4 in the tumorigenesis and progression of gastric cancer. Our findings demonstrate that FGFR4 expression is high in gastric cancer tissues, that it is associated with lymph node status and with neu and PCNA expression, and that it appears to be related to prognosis in patients with gastric cancer. Subgroup analysis revealed that, among patients with gastric cancer who have high FGFR4 expression levels, those with stage III/IV disease may have a worse prognosis. Furthermore, we observed that knockdown of FGFR4 expression contributed to reducing the proliferative ability and increasing the apoptotic rate in MKN45 and SGC7901 cells with high FGFR4 expression. Consequently, we believe that FGFR4 may serve as a target for novel therapies in patients with gastric cancer.
We thank Dr. Yuhu Xin, Dr. Mengyun Wang, Dr. Yingyi Li, Miss Zhen Wang, and Miss Yue Cao at the Laboratory Center, Shanghai Cancer Center, Fu Dan University for their excellent expert assistance and technical support.
This work was supported by Department of Abdominal Surgery, Fu Dan University Shanghai Cancer Center, Shanghai, China.