The Prostate 75:1092^1101 (2015)
FGF19 Promotes Progression of Prostate Cancer Hirotaka Nagamatsu, Jun Teishima,* Keisuke Goto, Hiroyuki Shikuma, Hiroyuki Kitano, Koichi Shoji, Shogo Inoue, and Akio Matsubara Department of Urology, Integrated Health Sciences, Institute of Biomedical & Health Sciences, Hiroshima University, Hiroshima, Japan
BACKGROUND. Fibroblast growth factor (FGF) signaling pathways have been reported to play important roles in prostate cancer (PCa) progression. FGF19 is one of a subfamily of FGFs that circulate in serum and act in an endocrine manner. Our objective was to investigate its role in the progression of PCa. METHODS. The effect of FGF19 on the proliferation and epithelial–mesenchymal transition of LNCaP and PC3 cells was examined using MTT assay and Western blotting. Serum concentration of FGF19 was measured by ELISA in 209 patients with PCa, and the association between clinicopathological features and the presence of FGF19-positive cells in tissues derived from 155 patients who undergone radical prostatectomy was investigated. RESULTS. Under androgen-deprived conditions achieved by incubation in medium with FGF19, the expression of N-cadherin in LNCaP cells was enhanced, that of E-cadherin and caspase 3 was suppressed, and the viability of LNCaP and PC3 cells was significantly enhanced. Significantly higher levels of PSA were recorded in the group determined by immunohistochemistry staining to be FGF19-positive (P ¼ 0.0046). The 5-year biochemical recurrence-free survival rate after radical prostatectomy was 46.4% in the FGF19-positive group and 70.0% in the FGF19-negative group (P ¼ 0.0027). In multivariate analysis, the presence of FGF19-positive tissues was an independent factor for worse prognosis after radical prostatectomy (P ¼ 0.0052). Serum FGF19 levels in high Gleason grade group were higher than that in low Gleason grade group (P ¼ 0.0009). CONCLUSIONS. FGF19 might be associated with biochemical recurrence after radical prostatectomy by promoting cell proliferation and epithelial–mesenchymal transition of PCa. Prostate 75:1092–1101, 2015. # 2015 Wiley Periodicals, Inc. KEY WORDS:
FGF19; prostate cancer; endocrine
INTRODUCTION Prostate cancer (PCa) is one of the most common malignant neoplasms among men and is the second leading cause of male cancer deaths in the United States [1]. It has been a very important theme to clarify the molecular mechanism in the development and the progression of PCa. On the other hand, although several studies have suggested that obesity is associated with worse pathologic findings and a higher biochemical recurrence rate after radical prostatectomy, the molecular mechanism of these associations is unknown [2,3]. Fibroblast growth factors (FGFs), 22 structurally related polypeptides, are humoral local factors with diverse biological functions involved in the regulation of cell growth and differentiation. There are four receptors (FGFR1–4) that mediate FGF signalß 2015 Wiley Periodicals, Inc.
ing. Several previous studies have demonstrated that abnormalities of FGFs and FGFRs in the prostate lead to failure of tissue homeostasis and development of cancer [4,5]. FGF19, FGF21, and FGF23 comprise an “endocrine FGF subfamily” that circulate in serum and act in an endocrine manner. They regulate diverse physiologic processes, such as energy metabolism and bile acid (BA) homeostasis (FGF19), glucose and lipid
Correspondence to: Jun Teishima, Department of Urology, Integrated Health Sciences, Institute of Biomedical & Health Sciences, Hiroshima University, 1-2-3 Kasumi, Minamiku, Hiroshima 734-8551, Japan. E-mail: [email protected] Received 29 September 2014; Accepted 25 February 2015 DOI 10.1002/pros.22994 Published online 8 April 2015 in Wiley Online Library (wileyonlinelibrary.com).
FGF19 in Prostate Cancer metabolism (FGF21), phosphate and vitamin D homeostasis (FGF23) [6–9]. Endocrine FGFs require a-Klotho and b-Klotho to act as coreceptors along with classic FGFR in order to mediate potent biologic activity, while canonical FGFs require only FGFRs [10]. Feng et al. [11] recently demonstrated that FGF19 promoted the proliferation and invasiveness in PCa cells. These previous studies indicate that FGF19 might act as both a growth factor and a metabolic regulator. In the work reported here, we investigated the impact of FGF19 in both PCa clinical samples and PCa cell lines. MATERIALS AND METHODS Cell Culture LNCaP and PC3 cells were purchased from the American Type Culture Collection (Manassas, VA). LNCaP cells were maintained in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS; Life Technologies, NY). PC3 cells were maintained in Opti-MEM reduced serum medium supplemented with 5% heat-inactivated FBS. Cells were maintained at 37°C in a humidified atmosphere of 5% CO2 and 95% air. Culture media were changed every 48 hr. MTT Assay LNCaP cells were plated in 96-well plates (2,000 cells per well) with 100 ml of RPMI-1640 medium containing 2% charcoal-stripped FBS. PC3 cells were plated in 96well plates (1,000 cells per well) with 100 ml of OptiMEM reduced serum medium not containing FBS. The cells were then incubated overnight at 37°C in a humidified atmosphere of 5% CO2 and 95% air. The medium of LNCaP cells was replaced with RPMI-1640 medium containing 2% charcoal-stripped FBS and 0, 0.1, or 1 ng/ml of recombinant FGF19 (R&D Systems, Minneapolis, MN). The media of PC3 cells was replaced with Opti-MEM reduced serum medium containing no FBS and 0, 0.1, or 1 ng/ml recombinant FGF19 (R&D Systems). MTT solution was added after the plates had been incubated for 3 days, and the plates were incubated at 37°C for another 3 hr. The plates were then analyzed using an ELISA plate reader (Bio-Rad, Hercules, CA) at 570 nm with the reference wavelength of 630 nm. Each experiment was repeated three times and the data were recorded as the mean and SD.
Immunohistochemistry Staining Tissue samples were obtained from patients undergoing radical prostatectomy at Hiroshima University
1093
Hospital between January 2005 and December 2012. All human tissues were obtained with informed consent. All sections were from formalin-fixed, paraffin-embedded tissue specimens. Samples derived from one representative section in every tissue were stained with anti-FGF19, anti-N-cadherin, and anti-KLB antibody (R&D Systems). Immunohistochemistry staining was performed with a Dako Envision þ Mouse Peroxidase Detection System (Dako Cytomation, Carpinteria, CA) as described previously [5,12]. The sections were deparaffinized in xylene before rehydration in 100%, 70%, and 50% ethanol and finally water. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide–methanol, and the slides were placed in water before antigen retrieval, in a 0.01-M citrate buffer (pH 6.0), using a microwave for 20 min. The sections were placed in normal goat serum (Dako Cytomation) for 1 hr to block non-specific antibody binding sites. The primary antibodies were diluted 1:100 and applied to the sections, which were then incubated overnight at 4°C. The sections were then washed in PBS before incubation with peroxidaselabeled anti-goat IgG (Dako Cytomation) for 1 hr at room temperature. The sections were washed in PBS, and immunoreactivity was visualized using a Dako Envision Kit (Dako Cytomation) according to the manufacturer’s instructions. They were then counterstained with 0.1% hematoxylin. When the components of PCa were stained more strongly than those of nonneoplastic prostatic tissue, the cases were defined as “FGF19-positive,” “KLB-positive,” and “N-cadherinpositive.” All slides were read independently by two investigators and classified according to the existence of FGF19, KLB, and N-cadherin-positive cells. Clinicopathological features and biochemical recurrence-free survival (BFS) in the two groups based on the staining of FGF19, KLB, and N-cadherin were investigated. Def|nition of Biochemical Recurrence Serum PSA was tested in every patient after radical prostatectomy by using the ECLIA method (Roche Diagnostics, Tokyo, Japan), and biochemical recurrence was defined as detectable PSA that increases >0.2 ng/ml over two subsequent measurements. Western Blotting Western blotting was performed as described previously [5,12]. To examine whether FGF19 affects the expression of N-cadherin, E-cadherin, and caspase 3, LNCaP and PC3 cells were cultured in RPMI-1640 medium containing 0.2% charcoal-stripped FBS for 24 hr and then incubated for 2 days in medium with or without FGF19 (0.1 ng/ml). Cells were scraped in The Prostate
1094
Nagamatsu et al.
Tris–glycine SDS sample buffer (Invitrogen, Carlsbad, CA). Supernatant protein was electrophoresed on sodium dodecylsulfate–polyacrylamide gel electrophoresis gels and electrotransferred onto nitrocellulose filters. Filters were incubated for 1 hr at room temperature with N-cadherin (Abcam, Cambridge, UK), E-cadherin (Neomarkers, Fremont, CA), and caspase 3 (Cell Signaling Technology, Danvers, MA), and peroxidaseconjugated anti-mouse IgG and anti-rabbit IgG (Santa Cruz Biotechnology, Santa Cruz, CA) were used in the secondary reaction. Immunocomplexes were visualized with an ECL Western Blot Detection System (Amersham Biosciences, Piscataway, NJ). b-Actin (Sigma, St Louis, MO) was also stained as a loading control. Blood Sample and FGF19 Measurement Blood samples were drawn into clot activator tubes and centrifuged at 3,000g for 10 min. The sera were stored at 80°C, and the FGF19 and KLB in them were measured in duplicate by sandwich enzyme-linked immunosorbent assay (Phoenix Pharmaceuticals, Inc, USA; R&D Systems). Statistical Analysis The MTT assay results were analyzed using the Mann–Whitney U-test, the relationship between FGF19 immunohistochemical staining and clinicopathological findings was analyzed using the x2-test,
and the relationship between the staining and BFS rates was analyzed using the log-rank test. Postoperative predictive factors were analyzed using a Cox hazard analysis. The relationship between FGF19 serum level and clinicopathological findings was analyzed using the x2-test. All statistical analyses were conducted using JMP version 10, and a P-value less than 0.05 was considered statistically significant.
RESULTS FGF19 Promotes the Proliferation of PCa Cell Lines An MTT assay was performed to determine the effect of FGF19 on the proliferation of LNCaP cells and PC3 cells. Figure 1a and b shows that the viability of LNCaP and PC3 cells was significantly enhanced by treatment with FGF19. In both of these cell lines, viability was highest with 0.1 ng/ml of FGF19. FGF19 Induces Epithelial^Mesenchymal Transition and Inhibits Apoptosis in PCa Cells Figure 2 shows protein expression in LNCaP and PC3 cells under androgen-deprived conditions achieved by incubation in medium with FGF19 for 2 days. PC3 already expresses as N-cadherin and not E-cadherin. So, there was no significant change of the protein expression. PC3 acts a positive control. The expression of N-cadherin was enhanced and that
Fig. 1. a: A culture of 2,000 LNCap cells was incubated in RPMI-1640 medium with 2.0% charcoal-stripped FBS and either 0, 0.1, or 1.0 ng/ml of fibroblast growth factor19 (FGF19). b: A culture of1,000 PC3 cells was incubated in Opti-MEM medium with no FBS and either 0, 0.1, or1.0 ng/ml FGF19. FGF19 stimulated the proliferation of LNCaP and PC3 cells. Representative results of three trials are shown. Data are presented asmean SD of the OD 570 nm. The Prostate
FGF19 in Prostate Cancer
1095
FGF19-positive group was significantly higher than in the FGF19-negative group. There was no significant difference between two groups in the age, BMI, Tstage, extraprostatic extension (EPE), and resection margin (RM). Serum PSA level in the KLB-positive group and N-cadherin-positive group was significantly higher than in the KLB-negative group and N-cadherinnegative group. The rate of Gleason score 8 in the Ncadherin-positive group was higher than that in the Ncadherin-negative group. (Tables II and III). Fig. 2. N-cadherin, E-cadherin, and caspase 3 proteins as determined by Western blotting. FGF19 enhanced the expression of N-cadherin and suppressed the expression of E-cadherin and caspase 3 in LNCaPcells.
of E-cadherin and caspase 3 was suppressed in LNCaP cells. These results show that FGF19 induces epithelial–mesenchymal transition and inhibits apoptosis in PCa cells. Accumulation of FGF19 in Cytoplasm Increased in Poorly Differentiated PCa Cells One hundred fifty-five tissue samples derived from radical prostatectomy specimens of patients with no prior therapy were examined. The immunohistochemistry staining shown in Figure 3 revealed that the components of PCa were stained more strongly than those of non-neoplastic prostatic tissue. Thirty-nine samples were classified FGF19-positive. Table I shows the clinicopathological features of the FGF19-positive and FGF19-negative groups. Serum PSA level in the
FGF19-Positive PCa Showed Higher Risk of Biochemical Recurrence After Radical Prostatectomy Figure 4 shows BFS in the FGF19-positive and FGF19-negative groups. The 5-year BFS after radical prostatectomy in the FGF19-positive group was significantly lower than that in the FGF19-negative group (46.4% vs. 70.0%, P ¼ 0.0027). A univariate analysis showed that PSA (P ¼ 0.0052), Gleason score (P ¼ 0.0061), EPE (P < 0.0001), RM (P < 0.0001), and FGF19 positivity (P ¼ 0.0058) were significantly prognostic factors for BFS. In multivariate analysis, FGF19 positivity, PSA, and RM were independent predictors of recurrence after radical prostatectomy (Table IV). Presence of FGF19-Positive Cells Correlated With Positive Immunohistochemistry With N-Cadherin in PCa Tissues The relationship between the positivity of N-cadherin and that of FGF19-positive cells in prostate tissues is shown in Table V. Figure 5 shows the representative finding of FGF19 and N-cadherin expressed in the same tissue. The prevalence of FGF19positive cells was significantly higher in N-cadherinpositive cases than in the cases where N-cadherin was negative. These results show that FGF19 induces epithelial–mesenchymal transition as with Western blotting. Serum FGF19 Level in the BMI < 25 Group was Signif|cantly Higher Than That in the BMI 25 Group
Fig. 3. Immunohistochemical staining with anti-FGF19 antibody in a radical prostatectomy specimen. The components of cancer cells were stained more strongly than those of non-neoplastic prostate cells.
Table VI shows the serum level of FGF19 and clinical characteristics of 209 patients. Serum FGF19 was measured by ELISA in 209 patients who had undergone a prostate biopsy and had detected PCa. Table VII shows serum FGF19 levels in the BMI 25 group were significantly lower than those in the BMI < 25 group (P ¼ 0.0490). On the other hand, there was no relationship between serum FGF19 levels and other clinicopathological parameters (Table VIII). The Prostate
1096
Nagamatsu et al.
TABLE I. Patients Characteristics of FGF19-Positive and FGF19-Negative Group FGF19-positive (N ¼ 39)
Age (years) PSA (ng/ml) BMI (kg/m2)
Tstage pT2 pT3 Gleason score 10) Gleason score (38) Positive surgical margin Capsular invasion FGF19-positive
Multivariate analysis
HR
95%
CI
P-value
HR
95%
CI
P-value
1.542 0.67 2.313 2.315 4.926 3.719 2.372
0.835 0.314 1.285 1.277 2.630 1.911 1.295
2.780 1.303 4.239 4.167 9.783 6.885 4.268
0.162 0.2479 0.0052 0.0061
FGF19 Promotes Progression of Prostate Cancer Hirotaka Nagamatsu, Jun Teishima,* Keisuke Goto, Hiroyuki Shikuma, Hiroyuki Kitano, Koichi Shoji, Shogo Inoue, and Akio Matsubara Department of Urology, Integrated Health Sciences, Institute of Biomedical & Health Sciences, Hiroshima University, Hiroshima, Japan
BACKGROUND. Fibroblast growth factor (FGF) signaling pathways have been reported to play important roles in prostate cancer (PCa) progression. FGF19 is one of a subfamily of FGFs that circulate in serum and act in an endocrine manner. Our objective was to investigate its role in the progression of PCa. METHODS. The effect of FGF19 on the proliferation and epithelial–mesenchymal transition of LNCaP and PC3 cells was examined using MTT assay and Western blotting. Serum concentration of FGF19 was measured by ELISA in 209 patients with PCa, and the association between clinicopathological features and the presence of FGF19-positive cells in tissues derived from 155 patients who undergone radical prostatectomy was investigated. RESULTS. Under androgen-deprived conditions achieved by incubation in medium with FGF19, the expression of N-cadherin in LNCaP cells was enhanced, that of E-cadherin and caspase 3 was suppressed, and the viability of LNCaP and PC3 cells was significantly enhanced. Significantly higher levels of PSA were recorded in the group determined by immunohistochemistry staining to be FGF19-positive (P ¼ 0.0046). The 5-year biochemical recurrence-free survival rate after radical prostatectomy was 46.4% in the FGF19-positive group and 70.0% in the FGF19-negative group (P ¼ 0.0027). In multivariate analysis, the presence of FGF19-positive tissues was an independent factor for worse prognosis after radical prostatectomy (P ¼ 0.0052). Serum FGF19 levels in high Gleason grade group were higher than that in low Gleason grade group (P ¼ 0.0009). CONCLUSIONS. FGF19 might be associated with biochemical recurrence after radical prostatectomy by promoting cell proliferation and epithelial–mesenchymal transition of PCa. Prostate 75:1092–1101, 2015. # 2015 Wiley Periodicals, Inc. KEY WORDS:
FGF19; prostate cancer; endocrine
INTRODUCTION Prostate cancer (PCa) is one of the most common malignant neoplasms among men and is the second leading cause of male cancer deaths in the United States [1]. It has been a very important theme to clarify the molecular mechanism in the development and the progression of PCa. On the other hand, although several studies have suggested that obesity is associated with worse pathologic findings and a higher biochemical recurrence rate after radical prostatectomy, the molecular mechanism of these associations is unknown [2,3]. Fibroblast growth factors (FGFs), 22 structurally related polypeptides, are humoral local factors with diverse biological functions involved in the regulation of cell growth and differentiation. There are four receptors (FGFR1–4) that mediate FGF signalß 2015 Wiley Periodicals, Inc.
ing. Several previous studies have demonstrated that abnormalities of FGFs and FGFRs in the prostate lead to failure of tissue homeostasis and development of cancer [4,5]. FGF19, FGF21, and FGF23 comprise an “endocrine FGF subfamily” that circulate in serum and act in an endocrine manner. They regulate diverse physiologic processes, such as energy metabolism and bile acid (BA) homeostasis (FGF19), glucose and lipid
Correspondence to: Jun Teishima, Department of Urology, Integrated Health Sciences, Institute of Biomedical & Health Sciences, Hiroshima University, 1-2-3 Kasumi, Minamiku, Hiroshima 734-8551, Japan. E-mail: [email protected] Received 29 September 2014; Accepted 25 February 2015 DOI 10.1002/pros.22994 Published online 8 April 2015 in Wiley Online Library (wileyonlinelibrary.com).
FGF19 in Prostate Cancer metabolism (FGF21), phosphate and vitamin D homeostasis (FGF23) [6–9]. Endocrine FGFs require a-Klotho and b-Klotho to act as coreceptors along with classic FGFR in order to mediate potent biologic activity, while canonical FGFs require only FGFRs [10]. Feng et al. [11] recently demonstrated that FGF19 promoted the proliferation and invasiveness in PCa cells. These previous studies indicate that FGF19 might act as both a growth factor and a metabolic regulator. In the work reported here, we investigated the impact of FGF19 in both PCa clinical samples and PCa cell lines. MATERIALS AND METHODS Cell Culture LNCaP and PC3 cells were purchased from the American Type Culture Collection (Manassas, VA). LNCaP cells were maintained in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS; Life Technologies, NY). PC3 cells were maintained in Opti-MEM reduced serum medium supplemented with 5% heat-inactivated FBS. Cells were maintained at 37°C in a humidified atmosphere of 5% CO2 and 95% air. Culture media were changed every 48 hr. MTT Assay LNCaP cells were plated in 96-well plates (2,000 cells per well) with 100 ml of RPMI-1640 medium containing 2% charcoal-stripped FBS. PC3 cells were plated in 96well plates (1,000 cells per well) with 100 ml of OptiMEM reduced serum medium not containing FBS. The cells were then incubated overnight at 37°C in a humidified atmosphere of 5% CO2 and 95% air. The medium of LNCaP cells was replaced with RPMI-1640 medium containing 2% charcoal-stripped FBS and 0, 0.1, or 1 ng/ml of recombinant FGF19 (R&D Systems, Minneapolis, MN). The media of PC3 cells was replaced with Opti-MEM reduced serum medium containing no FBS and 0, 0.1, or 1 ng/ml recombinant FGF19 (R&D Systems). MTT solution was added after the plates had been incubated for 3 days, and the plates were incubated at 37°C for another 3 hr. The plates were then analyzed using an ELISA plate reader (Bio-Rad, Hercules, CA) at 570 nm with the reference wavelength of 630 nm. Each experiment was repeated three times and the data were recorded as the mean and SD.
Immunohistochemistry Staining Tissue samples were obtained from patients undergoing radical prostatectomy at Hiroshima University
1093
Hospital between January 2005 and December 2012. All human tissues were obtained with informed consent. All sections were from formalin-fixed, paraffin-embedded tissue specimens. Samples derived from one representative section in every tissue were stained with anti-FGF19, anti-N-cadherin, and anti-KLB antibody (R&D Systems). Immunohistochemistry staining was performed with a Dako Envision þ Mouse Peroxidase Detection System (Dako Cytomation, Carpinteria, CA) as described previously [5,12]. The sections were deparaffinized in xylene before rehydration in 100%, 70%, and 50% ethanol and finally water. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide–methanol, and the slides were placed in water before antigen retrieval, in a 0.01-M citrate buffer (pH 6.0), using a microwave for 20 min. The sections were placed in normal goat serum (Dako Cytomation) for 1 hr to block non-specific antibody binding sites. The primary antibodies were diluted 1:100 and applied to the sections, which were then incubated overnight at 4°C. The sections were then washed in PBS before incubation with peroxidaselabeled anti-goat IgG (Dako Cytomation) for 1 hr at room temperature. The sections were washed in PBS, and immunoreactivity was visualized using a Dako Envision Kit (Dako Cytomation) according to the manufacturer’s instructions. They were then counterstained with 0.1% hematoxylin. When the components of PCa were stained more strongly than those of nonneoplastic prostatic tissue, the cases were defined as “FGF19-positive,” “KLB-positive,” and “N-cadherinpositive.” All slides were read independently by two investigators and classified according to the existence of FGF19, KLB, and N-cadherin-positive cells. Clinicopathological features and biochemical recurrence-free survival (BFS) in the two groups based on the staining of FGF19, KLB, and N-cadherin were investigated. Def|nition of Biochemical Recurrence Serum PSA was tested in every patient after radical prostatectomy by using the ECLIA method (Roche Diagnostics, Tokyo, Japan), and biochemical recurrence was defined as detectable PSA that increases >0.2 ng/ml over two subsequent measurements. Western Blotting Western blotting was performed as described previously [5,12]. To examine whether FGF19 affects the expression of N-cadherin, E-cadherin, and caspase 3, LNCaP and PC3 cells were cultured in RPMI-1640 medium containing 0.2% charcoal-stripped FBS for 24 hr and then incubated for 2 days in medium with or without FGF19 (0.1 ng/ml). Cells were scraped in The Prostate
1094
Nagamatsu et al.
Tris–glycine SDS sample buffer (Invitrogen, Carlsbad, CA). Supernatant protein was electrophoresed on sodium dodecylsulfate–polyacrylamide gel electrophoresis gels and electrotransferred onto nitrocellulose filters. Filters were incubated for 1 hr at room temperature with N-cadherin (Abcam, Cambridge, UK), E-cadherin (Neomarkers, Fremont, CA), and caspase 3 (Cell Signaling Technology, Danvers, MA), and peroxidaseconjugated anti-mouse IgG and anti-rabbit IgG (Santa Cruz Biotechnology, Santa Cruz, CA) were used in the secondary reaction. Immunocomplexes were visualized with an ECL Western Blot Detection System (Amersham Biosciences, Piscataway, NJ). b-Actin (Sigma, St Louis, MO) was also stained as a loading control. Blood Sample and FGF19 Measurement Blood samples were drawn into clot activator tubes and centrifuged at 3,000g for 10 min. The sera were stored at 80°C, and the FGF19 and KLB in them were measured in duplicate by sandwich enzyme-linked immunosorbent assay (Phoenix Pharmaceuticals, Inc, USA; R&D Systems). Statistical Analysis The MTT assay results were analyzed using the Mann–Whitney U-test, the relationship between FGF19 immunohistochemical staining and clinicopathological findings was analyzed using the x2-test,
and the relationship between the staining and BFS rates was analyzed using the log-rank test. Postoperative predictive factors were analyzed using a Cox hazard analysis. The relationship between FGF19 serum level and clinicopathological findings was analyzed using the x2-test. All statistical analyses were conducted using JMP version 10, and a P-value less than 0.05 was considered statistically significant.
RESULTS FGF19 Promotes the Proliferation of PCa Cell Lines An MTT assay was performed to determine the effect of FGF19 on the proliferation of LNCaP cells and PC3 cells. Figure 1a and b shows that the viability of LNCaP and PC3 cells was significantly enhanced by treatment with FGF19. In both of these cell lines, viability was highest with 0.1 ng/ml of FGF19. FGF19 Induces Epithelial^Mesenchymal Transition and Inhibits Apoptosis in PCa Cells Figure 2 shows protein expression in LNCaP and PC3 cells under androgen-deprived conditions achieved by incubation in medium with FGF19 for 2 days. PC3 already expresses as N-cadherin and not E-cadherin. So, there was no significant change of the protein expression. PC3 acts a positive control. The expression of N-cadherin was enhanced and that
Fig. 1. a: A culture of 2,000 LNCap cells was incubated in RPMI-1640 medium with 2.0% charcoal-stripped FBS and either 0, 0.1, or 1.0 ng/ml of fibroblast growth factor19 (FGF19). b: A culture of1,000 PC3 cells was incubated in Opti-MEM medium with no FBS and either 0, 0.1, or1.0 ng/ml FGF19. FGF19 stimulated the proliferation of LNCaP and PC3 cells. Representative results of three trials are shown. Data are presented asmean SD of the OD 570 nm. The Prostate
FGF19 in Prostate Cancer
1095
FGF19-positive group was significantly higher than in the FGF19-negative group. There was no significant difference between two groups in the age, BMI, Tstage, extraprostatic extension (EPE), and resection margin (RM). Serum PSA level in the KLB-positive group and N-cadherin-positive group was significantly higher than in the KLB-negative group and N-cadherinnegative group. The rate of Gleason score 8 in the Ncadherin-positive group was higher than that in the Ncadherin-negative group. (Tables II and III). Fig. 2. N-cadherin, E-cadherin, and caspase 3 proteins as determined by Western blotting. FGF19 enhanced the expression of N-cadherin and suppressed the expression of E-cadherin and caspase 3 in LNCaPcells.
of E-cadherin and caspase 3 was suppressed in LNCaP cells. These results show that FGF19 induces epithelial–mesenchymal transition and inhibits apoptosis in PCa cells. Accumulation of FGF19 in Cytoplasm Increased in Poorly Differentiated PCa Cells One hundred fifty-five tissue samples derived from radical prostatectomy specimens of patients with no prior therapy were examined. The immunohistochemistry staining shown in Figure 3 revealed that the components of PCa were stained more strongly than those of non-neoplastic prostatic tissue. Thirty-nine samples were classified FGF19-positive. Table I shows the clinicopathological features of the FGF19-positive and FGF19-negative groups. Serum PSA level in the
FGF19-Positive PCa Showed Higher Risk of Biochemical Recurrence After Radical Prostatectomy Figure 4 shows BFS in the FGF19-positive and FGF19-negative groups. The 5-year BFS after radical prostatectomy in the FGF19-positive group was significantly lower than that in the FGF19-negative group (46.4% vs. 70.0%, P ¼ 0.0027). A univariate analysis showed that PSA (P ¼ 0.0052), Gleason score (P ¼ 0.0061), EPE (P < 0.0001), RM (P < 0.0001), and FGF19 positivity (P ¼ 0.0058) were significantly prognostic factors for BFS. In multivariate analysis, FGF19 positivity, PSA, and RM were independent predictors of recurrence after radical prostatectomy (Table IV). Presence of FGF19-Positive Cells Correlated With Positive Immunohistochemistry With N-Cadherin in PCa Tissues The relationship between the positivity of N-cadherin and that of FGF19-positive cells in prostate tissues is shown in Table V. Figure 5 shows the representative finding of FGF19 and N-cadherin expressed in the same tissue. The prevalence of FGF19positive cells was significantly higher in N-cadherinpositive cases than in the cases where N-cadherin was negative. These results show that FGF19 induces epithelial–mesenchymal transition as with Western blotting. Serum FGF19 Level in the BMI < 25 Group was Signif|cantly Higher Than That in the BMI 25 Group
Fig. 3. Immunohistochemical staining with anti-FGF19 antibody in a radical prostatectomy specimen. The components of cancer cells were stained more strongly than those of non-neoplastic prostate cells.
Table VI shows the serum level of FGF19 and clinical characteristics of 209 patients. Serum FGF19 was measured by ELISA in 209 patients who had undergone a prostate biopsy and had detected PCa. Table VII shows serum FGF19 levels in the BMI 25 group were significantly lower than those in the BMI < 25 group (P ¼ 0.0490). On the other hand, there was no relationship between serum FGF19 levels and other clinicopathological parameters (Table VIII). The Prostate
1096
Nagamatsu et al.
TABLE I. Patients Characteristics of FGF19-Positive and FGF19-Negative Group FGF19-positive (N ¼ 39)
Age (years) PSA (ng/ml) BMI (kg/m2)
Tstage pT2 pT3 Gleason score 10) Gleason score (38) Positive surgical margin Capsular invasion FGF19-positive
Multivariate analysis
HR
95%
CI
P-value
HR
95%
CI
P-value
1.542 0.67 2.313 2.315 4.926 3.719 2.372
0.835 0.314 1.285 1.277 2.630 1.911 1.295
2.780 1.303 4.239 4.167 9.783 6.885 4.268
0.162 0.2479 0.0052 0.0061
