Induction of Fibroblast Growth Factor Receptor 4 by Helicobacter pylori via Signal Transducer and Activator of Transcription 3 With a Feedforward Activation Loop Involving SRC Signaling in Gastric Cancer

Abstract

Background & aims: Helicobacter pylori (H pylori) infection is the main risk factor for gastric cancer. The role of fibroblast growth factor receptors (FGRFs) in H pylori-mediated gastric tumorigenesis remains largely unknown. This study investigated the molecular and mechanistic links between H pylori, inflammation, and FGFR4 in gastric cancer.

Methods: Cell lines, human and mouse gastric tissue samples, and gastric organoids models were implemented. Infection with H pylori was performed using in vitro and in vivo models. Western blot, real-time quantitative reverse-transcription polymerase chain reaction, flow cytometry, immunofluorescence, immunohistochemistry, chromatin immunoprecipitation, and luciferase reporter assays were used for molecular, mechanistic, and functional studies.

Results: Analysis of FGFR family members using The Cancer Genome Atlas data, followed by validation, indicated that FGFR4 messenger (m)RNA was the most significantly overexpressed member in human gastric cancer tissue samples (P < .001). We also detected high levels of Fgfr4 mRNA and protein in gastric dysplasia and adenocarcinoma lesions in mouse models. Infection with J166, 7.13, and PMSS1 cytotoxin-associated gene A (CagA)+ H pylori strains induced FGFR4 mRNA and protein expression in in vitro and in vivo models. This was associated with a concordant activation of signal transducer and activator of transcription 3 (STAT3). Analysis of the FGFR4 promoter suggested several putative binding sites for STAT3. Using chromatin immunoprecipitation assay and an FGFR-promoter luciferase reporter containing putative STAT3 binding sites and their mutants, we confirmed a direct functional binding of STAT3 on the FGFR4 promoter. Mechanistically, we also discovered a feedforward activation loop between FGFR4 and STAT3 where the fibroblast growth factor 19–FGFR4 axis played an essential role in activating STAT3 in a SRC proto-oncogene non-receptor tyrosine kinase dependent manner. Functionally, we found that FGFR4 protected against H pylori-induced DNA damage and cell death.

Conclusions: Our findings demonstrated a link between infection, inflammation, and FGFR4 activation, where a feedforward activation loop between FGFR4 and STAT3 is established via SRC proto-oncogene non-receptor tyrosine kinase in response to H pylori infection. Given the relevance of FGFR4 to the etiology and biology of gastric cancer, we propose FGFR4 as a druggable molecular vulnerability that can be tested in patients with gastric cancer.

Keywords: FGFR4; Gastric Cancer; H pylori; SRC; STAT3.

Figure 1.

FGFR4 is upregulated in gastric cancer tissues. (A) TCGA database analysis of FGFR family members’ mRNA expression. FGFR4 was the most significantly up-regulated FGFR in gastric cancer (GC) compared with normal gastric (NG) human tissue. (B) qRT-PCR analysis of FGFR4 mRNA in paired NG and GC human tissues (n = 45). (C) Immunohistochemistry (IHC) analysis shows the expression of FGFR4 in normal and human gastric cancer tissue samples (original magnification ×10 is shown in the upper panels; insets at original magnification ×40 are shown in the lower panels). Quantification scores are shown on the right panel. (D) qRT-PCR data demonstrate the expression of Fgfr4 mRNA in WT (n = 11) and TFF1-KO (n = 22) mouse tissues. (E) Immunohistochemistry analysis shows the expression of FGFR4 in neoplastic stomach tissues of the TFF1-KO mice compared with normal (TFF1-WT) (original magnification ×40). H&E staining of the corresponding tissues is shown on the right panels (original magnification ×10). (F) Kaplan-Meier survival curve shows poor overall survival (OS) and disease-free survival (DFS) in patients with gastric cancer with high expression of FGFR4. (G) Kaplan-Meier survival curve demonstrates an association between high expression of FGFR4 and poor overall survival in patients with intestinal-type gastric cancer. HR, hazard ratio. Data are expressed as the means ± standard deviations. Box and whisker plot: The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively, and the whiskers mark the minimum and maximum of all the data. P < .05, P < .01, and P < .001 are considered significant.

Figure 2.

H pylori infection induces FGFR4 expression and STAT3 activation. (A) GSEA based on gene expression analysis of gastric cancers (GCs) in the TCGA database indicated that FGFR4 has a significant correlation with H pylori-induced signaling in gastric cancer. NES, normalized enrichment score. (B) The qRT-PCR analysis demonstrates progressive induction of FGFR4 mRNA expression at different times after H pylori infection in AGS cells. Ctrl, control. (C) Western blot analysis demonstrates total and p-FGFR4 and p-STAT3 at 3, 6, and 24 hours after H pylori infection in AGS cells. Cag A, cytotoxin-associated gene A. (D) Western blot analysis shows total and p-FGFR4 and p-STAT3 in gastric tissues of WT mice infected with PMSS1 H pylori for 2 weeks compared with noninfected mice. (E) Immunofluorescence data indicate strong expression of FGFR4 and nuclear STAT3 staining in WT gastric mouse tissues infected with PMSS1 H pylori (lower panels), and weak to absent staining of FGFR4 and p-STAT3 was detected in noninfected mice (upper panels) (scale bar of 10 μm is shown in the merge image). H&E staining and bright-field images of representative organoids are shown on the right side of the panels. DAPI, 4′,6-diamidino-2-phenylindole. Quantification of positive nuclear p-STAT3 staining was performed in at least 200 cells from 3 different fields. Data are presented as a percentage ± standard error of the mean (right panel). (F) Immunofluorescence data of FGFR4 and nuclear STAT3 staining in uninfected and H pylori 7.13–infected organoids derived from normal human gastric tissues. H pylori–infected organoids showed strong nuclear p-STAT3 and FGFR4 staining (lower panels) compared with noninfected organoids (upper panels). Quantification of positive nuclear STAT3 cells was performed in at least 100 counted cells from 4 organoids (scale bar of 50 μm is shown in the merge image). Arrowheads indicate cells with nuclear localization of STAT3. Data are presented as a percentage ± standard error of the mean (right panel). H&E staining and bright-field images of representative organoids are shown on the right side of the panels. Data are expressed as the means ± standard deviations. P < .05, P < .01, and P < .001 are considered significant.

Figure 3.

FGFR4 expression coincides with STAT3 activation in vivo and in vitro. (A) Immunofluorescence staining demonstrates an increase in FGFR4 and p-STAT3 (nuclear staining) in gastric tissues from the GP130^F/F mouse model (1.5 and 3.5 months old) compared with WT (scale bar of 50 μm is shown is the merge image). DAPI, 4′,6-diamidino-2-phenylindole. (B) Immunofluorescence staining demonstrates an increase in FGFR4 and p-STAT3 (nuclear staining) in neoplastic gastric tissues from the TFF1-KO mice compared with normal stomach (scale bar of 50 μm is shown is the merge image). Arrowheads indicate cells with nuclear localization of STAT3. (C) qRT-PCR data (left panel) show an increase of FGFR4 after IL6 treatment (100 ng/mL) at 30, 60, and 120 minutes in AGS cells. Western blot analysis (right panel) confirms the increase in p-FGFR4 and p-STAT3 after IL6 treatment. NC, negative control. (D) qRT-PCR data indicate an increase of FGFR4 expression after H pylori infection (J166 or 7.13 strains) in AGS cells. This effect is reversed after STAT3 inhibitor treatment (0.4 μmol/L napabucasin, overnight). (E) Western blot data confirm the induction of p-FGFR4 and p-STAT3 after H pylori infection in AGS cells, an effect that is abrogated by napabucasin treatment (STAT3 inhibitor). (F) Immunofluorescence indicates that both IL6 treatment and H pylori infection induce expression of FGFR4 and nuclear p-STAT3 staining in AGS, whereas napabucasin inhibits the expression of both FGFR4 and p-STAT3 (scale bar of 50 μm is shown is the merge image). Data are expressed as the means ± standard deviations. P < .05, P < .01 and P < .001 are considered significant. (G) qRT-PCR data show an increase of Fgfr4 mRNA expression in WT mice stomach after H pylori (PMSS1) infection. This increase was abolished after treatment with napabucasin (40 mg/kg). Box and whisker plot: The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively, and the whiskers mark the minimum and maximum of all the data. (H) Western blot analysis confirms the expression of FGFR4 and p-STAT3 after H pylori PMSS1 infection in mice. This effect was abrogated by napabucasin treatment.

Figure 4.

FGFR4 expression is induced via direct binding of STAT3 on the FGFR4 promoter. (A) A schematic diagram shows the location of STAT3 putative binding regions on the FGFR4 promoter. ChIP assay primers were designed to cover regions P1 to P6. UTR, untranslated region. FGFR4 luciferase promoter-reporter assay demonstrates an increase in the luciferase activity in (B) AGS and (C) MKN-28 cells after H pylori infection (8 hours) or IL6 treatment (100 ng/mL, 6 hours). NC, negative control. Inhibition of STAT3 by napabucasin abrogated the luciferase induction under these conditions. (D) ChIP assay showing amplification of DNA fragments quantified by qRT-PCR and demonstrating the binding of STAT3 on the FGFR4 promotor in both P2 and P4 regions. Data are expressed as the means ± standard deviations. P < .05, P < .01 and P < .001 are considered significant.

Figure 5.

The FGF19-FGFR4 axis promotes a feedforward activation loop of STAT3 via SRC. (A) qRT-PCR analysis shows FGF1, FGF3, and FGF19 expressions in human gastric cancer (GC) compared with normal glandular (NG) stomach tissues. (B) TCGA database analysis indicated that FGF19 mRNA was significantly upregulated in GC compared with NG tissues. Kaplan-Meier survival curve indicates that patients with GC with high expression of (C) FGF19 and (D) FGFR4+FGF19 have poor overall survival (OS) and poor disease-free survival (DFS) compared with patients with GC with low expression. HR, hazard ratio. (E) qRT-PCR analysis demonstrates upregulation of FGF19 at low-grade dysplasia (LGD) and high-grade dysplasia (HGD)/cancer tissue samples from the TFF1-KO mice compared with normal gastric tissues. (F) qRT-PCR analysis demonstrates upregulation of Fgf15 mRNA expression after 1 and 2 weeks of PMSS1 H pylori infection in WT mice stomachs. Box and whisker plot: The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively, and the whiskers mark the minimum and maximum of all the data. (G) qRT-PCR analysis demonstrates FGF19 mRNA expression is induced at different times after H pylori infection in MKN28 cell lines. Data are expressed as the means ± standard deviations. P < .05, P < .01 and P < .001 are considered significant. (H) Western blot analysis is shown in MKN28 cells transfected with control or FGFR4 siRNA. After stimulation with FGF19 (200 ng/mL), MKN28 cells demonstrate a remarkable increase of p-FGFR4, p-STAT3, p-SRC, and p-extracellular signal–regulated kinase (ERK). In contrast, all of these proteins are significantly reduced after FGFR4 siRNA knockdown, with or without FGF19 stimulation. (I) Western blot analysis indicates that FGF19 neutralizing antibody blocks the H pylori-mediated increase in p-FGFR4, pSTAT3, p-SRC, and p-ERK compared with conditioned media without neutralizing antibody. The secretion of FGF19 is confirmed in the conditioned media of MKN28 cells.

Figure 6.

The FGFR4-SRC axis is crucial for the H pylori-induced activation of STAT3 signaling. (A) Pearson’s correlation of FGFR4 and SRC mRNA levels using the TCGA database. GSEA analysis indicates a positive correlation between FGFR4 and the (B) SRC and (C) STAT3 signaling pathways in gastric cancer. NES, normalized enrichment score. (D and E) Western blot analysis shows an increase in FGFR4, SRC, and STAT3 phosphoproteins after H pylori infection. This increase is abolished after the knockdown of FGFR4 using siRNA in AGS, MKN28, and GES-1 cells. Cag A, cytotoxin-associated gene A; NC, negative control. (F) Western blot analysis indicates that SRC knockdown after H pylori infection decreases the expression p-FGFR4, p-SRC, and p-STAT3 compared with control siRNA in AGS and MKN28 cells. (G) Western blot analysis, after inhibition of SRC using dasatinib, demonstrates a decrease in the expression of p-FGFR4, p-SRC, and p-STAT3 after H pylori infection compared with control siRNA in AGS and MKN28 cells.

Figure 7.

FGFR4 protects gastric cancer cells from apoptosis during H pylori infection. (A) Western blot analysis shows an increase of cleaved poly (adenosine-5′-diphosphate–ribose) polymerase (PARP) and cleaved-caspase 3, apoptotic markers, after knockdown of FGFR4 in H pylori-infected MKN28 cells. Cag A, cytotoxin-associated gene A; NC, negative control. (B and C) ApoTox-Glo assay data show decreased cell viability and increased caspase 3/7 activity after knockdown of FGFR4 in H pylori-infected MKN28 cells. ED₅₀, 50% effective dose. RFU, relative fluorescence units; RLU, relative luminescence unit. (D) Annexin V/propidium iodide (PI) staining flow cytometry analysis demonstrates a significant increase in apoptotic cells after knockdown of FGFR4 in H pylori-infected MKN28 cells compared with control cells. The apoptosis quantification ratio is presented on the right panel. (E) Pearson’s correlation test is shown for FGFR4 and BCL2L1 mRNAs (left panel) and FGFR4 and BIRC5 mRNAs, using TCGA data. (F) A qRT-PCR analysis is shown for prosurvival genes (BCL-XL, BIRC5, and XIAP) in control and FGFR4 siRNA conditions, with or without infection with J166 or 7.13 H pylori strains for 24 hours in MKN28 cells. All genes displayed an increase in mRNA expression levels in response to H pylori infection. This increase is attenuated after the knockdown of FGFR4. Data are expressed as the means ± standard deviations. P < .05, P < .01 and P < .001 are considered significant.