IJC International Journal of Cancer
Ileal FGF15 contributes to fibrosis-associated hepatocellular carcinoma development Iker Uriarte1,2*, M. Ujue Latasa2*, Simone Carotti3, Maite G. Fernandez-Barrena1,2, Oihane Garcia-Irigoyen1, Maria Elizalde1, Raquel Urtasun1, Umberto Vespasiani-Gentilucci4, Sergio Morini3, Alvaro de Mingo5,6,7, Montserrat Mari5,6,7, Fernando J. Corrales1,2, Jesus Prieto1,2, Carmen Berasain1,2 and Matias A. Avila1,2 1
CIBEREHD Internal Medicine, University Clinic Navarra, Instituto de Salud Carlos III, Pamplona, Spain Division of Hepatology, CIMA, University of Navarra, Pamplona, Spain 3 Microscopic and Ultrastructural Anatomy, Centre for Integrated Biomedical Research—CIR, University Campus Bio-Medico of Rome, Italy 4 Hepatology Unit, University Campus Bio-Medico of Rome, Italy 5 Department of Cell Death and Proliferation, IIBB-CSIC, Barcelona, Spain 6 Liver Unit, Hospital Clinic, IDIBAPS, Barcelona, Spain 7 CIBEREHD IBB-CSIC, Barcelona, Spain 2
Key words: fibrosis, fibroblast growth factor 15/19, connective tissue growth factor Abbreviations: AFP: alpha-fetoprotein; AR: amphiregulin; a1(I)Col: collagen type I alpha 1; BEC: biliary epithelial cells; CTGF: connective tissue growth factor; DEN: diethylnitrosamine; ECM: extracellular matrix; EGFR: epidermal growth factor receptor; FGF: fibroblast growth factor; FGFR: fibroblast growth factor receptor; GSK3-b: glycogen synthase kinase-3-beta; HCC: hepatocellular carcinoma; HSC: hepatic stellate cells; KLb: klotho-beta; MAPK-ERK: mitogen activated protein kinase-extracellular signal-regulated kinase; TGFb: transforming growth factor-b; TIMP-1: tissue inhibitor of metalloproteases 1; a-SMA: alpha-smooth muscle actin Additional Supporting Information may be found in the online version of this article. Carmen Berasain and Matias A. Avila share senior authorship *I.U. and M.U.L. contributed equally to this work Grant sponsors: FIMA and the “UTE project CIMA”; Grant number: RTICC-RD06 00200061; Grant sponsor: CIBEREHD; Grant numbers: FIS PI10/02642, PI13/00359, PI10/00038, PI13/00385; Grant sponsor: Instituto de Salud Carlos III (ISCIII) co-financed by “Fondo Europeo de Desarrollo Regional” (FEDER) “Una manera de hacer Europa”; Grant number: PI13/00374; Grant sponsor: Gobierno de Navarra, “Torres Quevedo” and a “Ramon y Cajal” contract from Ministerio de Educacion, PFIS fellowship from ISCIII, FPU fellowship from Ministerio de Educacion, Cultura y Deporte, Spain; Grant sponsor: Fundacion Eugenio Rodrıguez Pascual DOI: 10.1002/ijc.29287 History: Received 13 June 2014; Accepted 25 Sep 2014; Online 23 Oct 2014 Correspondence to: Matias A. Avila, Division of Hepatology, University of Navarra, Avda, Pio XII, n55, 31008 Pamplona, Spain, Fax: 34948-194717, E-mail: [email protected] or Carmen Berasain, Division of Hepatology, University of Navarra, Avda, Pio XII, n55, 31008 Pamplona, Spain, Fax: 34-948-194717, E-mail: [email protected]
C 2014 UICC Int. J. Cancer: 00, 00–00 (2014) V
Short Report
Fibroblast growth factor 15 (FGF15), FGF19 in humans, is a gut-derived hormone and a key regulator of bile acids and carbohydrate metabolism. FGF15 also participates in liver regeneration after partial hepatectomy inducing hepatocellular proliferation. FGF19 is overexpressed in a significant proportion of human hepatocellular carcinomas (HCC), and activation of its receptor FGFR4 promotes HCC cell growth. Here we addressed for the first time the role of endogenous Fgf15 in hepatocarcinogenesis. Fgf151/1 and Fgf152/2 mice were subjected to a clinically relevant model of liver inflammation and fibrosis-associated carcinogenesis. Fgf152/2 mice showed less and smaller tumors, and histological neoplastic lesions were also smaller than in Fgf151/1 animals. Importantly, ileal Fgf15 mRNA expression was enhanced in mice undergoing carcinogenesis, but at variance with human HCC it was not detected in liver or HCC tissues, while circulating FGF15 protein was clearly upregulated. Hepatocellular proliferation was also reduced in Fgf152/2 mice, which also expressed lower levels of the HCC marker alphafetoprotein (AFP). Interestingly, lack of FGF15 resulted in attenuated fibrogenesis. However, in vitro experiments showed that liver fibrogenic stellate cells were not direct targets for FGF15/FGF19. Conversely we demonstrate that FGF15/FGF19 induces the expression of the pro-fibrogenic and pro-tumorigenic connective tissue growth factor (CTGF) in hepatocytes. These findings suggest the existence of an FGF15-triggered CTGF-mediated paracrine action on stellate cells, and an amplification mechanism for the hepatocarcinogenic effects of FGF15 via CTGF production. In summary, our observations indicate that ileal FGF15 may contribute to HCC development in a context of chronic liver injury and fibrosis.
2
Gut-derived Fgf15 in hepatocarcinoma development
Short Report
What’s new? Fibroblast growth factor-19 (FGF19), in rodents called FGF15, is a gut-derived hormone recently implicated as a driver gene in liver carcinogenesis. Here, the authors show that Fgf152/2 mice develop less hepatocellular carcinoma and less liver fibrosis as compared to Fgf151/1 littermates, underscoring the important role of the factor in liver damage and cancer development. Interestingly, Fgf15 expression is not detected in injured liver or carcinoma tissue, but is upregulated in the ileum and blood, pointing to a new gut-liver axis involved in hepatocarcinogenesis.
Hepatocellular carcinoma (HCC) is one of the most frequent types of cancers worldwide and the third cause of cancerrelated deaths. HCC-associated mortality has rapidly increased globally over the past 20 years, particularly in the United States, while China still accounts for about half of the total liver-cancer related deaths and new cases declared annually. From the molecular perspective HCC is a complex type of tumor, with marked deregulation of several signaling pathways and accumulated genetic alterations.1 Usually HCC develops stepwise on a background of chronic liver injury, fibrosis and inflammation, generally caused by viral infections (hepatitis B and C viruses), alcoholic cirrhosis, exposure to environmental toxins and nonalcoholic fatty liver disease.1 Its molecular complexity and the underlying chronic liver disease, make HCC a very difficult to treat neoplasia, particularly when diagnosed at advanced stages. Great efforts have been made in recent years to identify critical pathways in tumor progression that could be inhibited with systemic therapies and yet cause minimal hepatotoxicity. However, albeit several targeted drugs have been tested in clinical trials, only sorafenib, a multikinase inhibitor, has shown some clinical benefit.1 On one hand this situation indicates that targeted therapies may be useful in HCC treatment, but at the same time it calls for the identification of other key molecular targets that could be drugged alone or in combination with sorafenib. Fibroblast growth factor 19 (FGF19) is a ligand of the FGF receptor 4 (FGFR4), which is highly expressed in hepatocytes, and it can also bind and activate FGFR1c, 2c and 3c.2 Efficient FGF19 signaling through these tyrosine kinase receptors relies on the expression of the klotho-beta (KLb) trans-membrane co-receptor, abundantly expressed in hepatocytes.2 Physiologically FGF19, and its murine orthologue FGF15, behaves as an enterohepatic hormone regulating bile acid synthesis and intermediary metabolism.3 Besides, FGF19/FGFR4 signaling has potent effects on proliferation, survival and motility in cancer cells.2 In humans, FGF19 and FGFR4 expression have been found up-regulated in a significant proportion of HCCs, FGF19 levels correlating with poor prognosis, and genetic or pharmacological inhibition of FGF19/FGFR4 interaction and signaling have proven efficacious experimental anti-HCC strategies.4–8 Interestingly, ectopic expression of FGF19 in skeletal muscle of transgenic mice resulted in enhanced hepatocellular proliferation and in the development of HCC,9 a process that was accelerated in mice treated with the carcinogen diethylnitrosamine (DEN) and was dependent on hepatocellular FGFR4 expression.8 In
spite of this compelling evidence, the role of endogenous FGF15/FGF19 in hepatocarcinogenesis has not been directly evaluated so far. We recently identified an important function for FGF15 in liver regeneration and hepatocellular proliferation in mice.10 In this study we demonstrate that FGF15 contributes to HCC progression in a clinically relevant model of liver cancer.11
Material and Methods Induction of HCC and acute CCl4 treatment
Fgf152/2 mice and their wild type littermate controls (Fgf151/1) are on a mixed C57BL/6129/Sv background and have been backcrossed for more than 16 generations. These mice have been described before.10 HCC was induced essentially as previously described with minor modifications.11 Briefly, 15-day-old mice (n 5 13 Fgf152/2 and n 5 21 Fgf151/1) were injected with DEN (25 mg/Kg, i.p.) followed by biweekly i.p. injections of CCl4 (0.5 mL/Kg) starting at 4 weeks of age. Mice were weighted and killed at 27 weeks of age, after 12 CCl4 injections. Livers were excised, weighted and macroscopically examined for gross lesions. Liver index, the ratio between liver and body weight (expressed as %), was calculated. Tumors size was estimated with a Vernier caliper. Tumoral lesions and tumor-free tissues were excised and either snapfrozen in liquid N2, or paraffin embedded for histological analyses. Acute CCl4 administration was performed as described.12 Histological analyses
Liver tissue sections from all mice were stained with H&E and examined by light microscopy (Olympus BX-51) by two pathologists (S.C. and S.M.) blinded to the identity of samples. Lesions (dysplastic foci, adenomas and HCCs) were classified according to published criteria.13 a2Smooth muscle actin (a2Sma), Ctgf (antibody sc-14939, Santa Cruz Biotechnology) and Ki67 immunodetection and Sirius Red staining were performed essentially as described.14 Cell isolation, culture and transfections
Primary HSCs and hepatocytes were isolated from Fgf151/1 mice essentially as described.15,16 The origin and culture conditions of mouse AML12 hepatocytes, human Hep3B and HepG2 cells have been previously reported.16 Transfections with CTGF-promoter luciferase reporter construct (nucleotides 2485 to 115 of human CTGF 5’ region), and with b-catenin, glycogen synthase kinase-b (GSK3-b) and control siGL siRNAs (all from Dharmacon), were performed as described.17
C 2014 UICC Int. J. Cancer: 00, 00–00 (2014) V
3
Uriarte et al.
Figure 1. HCC development is attenuated in Fgf152/2 mice. Mice were injected with DEN followed by CCl4 as indicated and sacrificed at 27 weeks of age (a). Shown are liver index (b), representative liver images (c), tumor number (d), mean size of tumors (e), representative H&E sections highlighting different types of lesions (f), and mean diameter of histological lesions (g). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Total RNA from liver tissue and cell lines was extracted using the automated Maxwell system from Promega. Reverse transcription was performed as reported.14 Real-time PCRs were performed with iQ SYBR Green supermix (BioRad) in a CFX96 system from BioRad as previously described.14 Primers used are described in Supporting Information Table 1. Western blot analyses
Cells and liver tissues were lysed and homogenates were subjected to Western blot analysis as reported.14 CTGF and amphiregulin (AR) were detected as described.12,17 FGF15 determination
FGF15 levels were measured in mouse serum and liver tissues by ELISA (MyBiosource, San Diego). Sera from mice fed a 1% cholate-supplemented diet (CA1%) for 5 days and mice infected with an FGF15 expressing adenovirus were used as controls.10 Background signals from Fgf152/2 mice sera and liver extracts were substracted to all values. Statistical analysis
Data are means 6 SEM. Analyses were performed using GraphPad Prism version 5.00 (GraphPad Software, San C 2014 UICC Int. J. Cancer: 00, 00–00 (2014) V
Diego). Data were compared among groups using the Student’s t-test. A p value of
Ileal FGF15 contributes to fibrosis-associated hepatocellular carcinoma development Iker Uriarte1,2*, M. Ujue Latasa2*, Simone Carotti3, Maite G. Fernandez-Barrena1,2, Oihane Garcia-Irigoyen1, Maria Elizalde1, Raquel Urtasun1, Umberto Vespasiani-Gentilucci4, Sergio Morini3, Alvaro de Mingo5,6,7, Montserrat Mari5,6,7, Fernando J. Corrales1,2, Jesus Prieto1,2, Carmen Berasain1,2 and Matias A. Avila1,2 1
CIBEREHD Internal Medicine, University Clinic Navarra, Instituto de Salud Carlos III, Pamplona, Spain Division of Hepatology, CIMA, University of Navarra, Pamplona, Spain 3 Microscopic and Ultrastructural Anatomy, Centre for Integrated Biomedical Research—CIR, University Campus Bio-Medico of Rome, Italy 4 Hepatology Unit, University Campus Bio-Medico of Rome, Italy 5 Department of Cell Death and Proliferation, IIBB-CSIC, Barcelona, Spain 6 Liver Unit, Hospital Clinic, IDIBAPS, Barcelona, Spain 7 CIBEREHD IBB-CSIC, Barcelona, Spain 2
Key words: fibrosis, fibroblast growth factor 15/19, connective tissue growth factor Abbreviations: AFP: alpha-fetoprotein; AR: amphiregulin; a1(I)Col: collagen type I alpha 1; BEC: biliary epithelial cells; CTGF: connective tissue growth factor; DEN: diethylnitrosamine; ECM: extracellular matrix; EGFR: epidermal growth factor receptor; FGF: fibroblast growth factor; FGFR: fibroblast growth factor receptor; GSK3-b: glycogen synthase kinase-3-beta; HCC: hepatocellular carcinoma; HSC: hepatic stellate cells; KLb: klotho-beta; MAPK-ERK: mitogen activated protein kinase-extracellular signal-regulated kinase; TGFb: transforming growth factor-b; TIMP-1: tissue inhibitor of metalloproteases 1; a-SMA: alpha-smooth muscle actin Additional Supporting Information may be found in the online version of this article. Carmen Berasain and Matias A. Avila share senior authorship *I.U. and M.U.L. contributed equally to this work Grant sponsors: FIMA and the “UTE project CIMA”; Grant number: RTICC-RD06 00200061; Grant sponsor: CIBEREHD; Grant numbers: FIS PI10/02642, PI13/00359, PI10/00038, PI13/00385; Grant sponsor: Instituto de Salud Carlos III (ISCIII) co-financed by “Fondo Europeo de Desarrollo Regional” (FEDER) “Una manera de hacer Europa”; Grant number: PI13/00374; Grant sponsor: Gobierno de Navarra, “Torres Quevedo” and a “Ramon y Cajal” contract from Ministerio de Educacion, PFIS fellowship from ISCIII, FPU fellowship from Ministerio de Educacion, Cultura y Deporte, Spain; Grant sponsor: Fundacion Eugenio Rodrıguez Pascual DOI: 10.1002/ijc.29287 History: Received 13 June 2014; Accepted 25 Sep 2014; Online 23 Oct 2014 Correspondence to: Matias A. Avila, Division of Hepatology, University of Navarra, Avda, Pio XII, n55, 31008 Pamplona, Spain, Fax: 34948-194717, E-mail: [email protected] or Carmen Berasain, Division of Hepatology, University of Navarra, Avda, Pio XII, n55, 31008 Pamplona, Spain, Fax: 34-948-194717, E-mail: [email protected]
C 2014 UICC Int. J. Cancer: 00, 00–00 (2014) V
Short Report
Fibroblast growth factor 15 (FGF15), FGF19 in humans, is a gut-derived hormone and a key regulator of bile acids and carbohydrate metabolism. FGF15 also participates in liver regeneration after partial hepatectomy inducing hepatocellular proliferation. FGF19 is overexpressed in a significant proportion of human hepatocellular carcinomas (HCC), and activation of its receptor FGFR4 promotes HCC cell growth. Here we addressed for the first time the role of endogenous Fgf15 in hepatocarcinogenesis. Fgf151/1 and Fgf152/2 mice were subjected to a clinically relevant model of liver inflammation and fibrosis-associated carcinogenesis. Fgf152/2 mice showed less and smaller tumors, and histological neoplastic lesions were also smaller than in Fgf151/1 animals. Importantly, ileal Fgf15 mRNA expression was enhanced in mice undergoing carcinogenesis, but at variance with human HCC it was not detected in liver or HCC tissues, while circulating FGF15 protein was clearly upregulated. Hepatocellular proliferation was also reduced in Fgf152/2 mice, which also expressed lower levels of the HCC marker alphafetoprotein (AFP). Interestingly, lack of FGF15 resulted in attenuated fibrogenesis. However, in vitro experiments showed that liver fibrogenic stellate cells were not direct targets for FGF15/FGF19. Conversely we demonstrate that FGF15/FGF19 induces the expression of the pro-fibrogenic and pro-tumorigenic connective tissue growth factor (CTGF) in hepatocytes. These findings suggest the existence of an FGF15-triggered CTGF-mediated paracrine action on stellate cells, and an amplification mechanism for the hepatocarcinogenic effects of FGF15 via CTGF production. In summary, our observations indicate that ileal FGF15 may contribute to HCC development in a context of chronic liver injury and fibrosis.
2
Gut-derived Fgf15 in hepatocarcinoma development
Short Report
What’s new? Fibroblast growth factor-19 (FGF19), in rodents called FGF15, is a gut-derived hormone recently implicated as a driver gene in liver carcinogenesis. Here, the authors show that Fgf152/2 mice develop less hepatocellular carcinoma and less liver fibrosis as compared to Fgf151/1 littermates, underscoring the important role of the factor in liver damage and cancer development. Interestingly, Fgf15 expression is not detected in injured liver or carcinoma tissue, but is upregulated in the ileum and blood, pointing to a new gut-liver axis involved in hepatocarcinogenesis.
Hepatocellular carcinoma (HCC) is one of the most frequent types of cancers worldwide and the third cause of cancerrelated deaths. HCC-associated mortality has rapidly increased globally over the past 20 years, particularly in the United States, while China still accounts for about half of the total liver-cancer related deaths and new cases declared annually. From the molecular perspective HCC is a complex type of tumor, with marked deregulation of several signaling pathways and accumulated genetic alterations.1 Usually HCC develops stepwise on a background of chronic liver injury, fibrosis and inflammation, generally caused by viral infections (hepatitis B and C viruses), alcoholic cirrhosis, exposure to environmental toxins and nonalcoholic fatty liver disease.1 Its molecular complexity and the underlying chronic liver disease, make HCC a very difficult to treat neoplasia, particularly when diagnosed at advanced stages. Great efforts have been made in recent years to identify critical pathways in tumor progression that could be inhibited with systemic therapies and yet cause minimal hepatotoxicity. However, albeit several targeted drugs have been tested in clinical trials, only sorafenib, a multikinase inhibitor, has shown some clinical benefit.1 On one hand this situation indicates that targeted therapies may be useful in HCC treatment, but at the same time it calls for the identification of other key molecular targets that could be drugged alone or in combination with sorafenib. Fibroblast growth factor 19 (FGF19) is a ligand of the FGF receptor 4 (FGFR4), which is highly expressed in hepatocytes, and it can also bind and activate FGFR1c, 2c and 3c.2 Efficient FGF19 signaling through these tyrosine kinase receptors relies on the expression of the klotho-beta (KLb) trans-membrane co-receptor, abundantly expressed in hepatocytes.2 Physiologically FGF19, and its murine orthologue FGF15, behaves as an enterohepatic hormone regulating bile acid synthesis and intermediary metabolism.3 Besides, FGF19/FGFR4 signaling has potent effects on proliferation, survival and motility in cancer cells.2 In humans, FGF19 and FGFR4 expression have been found up-regulated in a significant proportion of HCCs, FGF19 levels correlating with poor prognosis, and genetic or pharmacological inhibition of FGF19/FGFR4 interaction and signaling have proven efficacious experimental anti-HCC strategies.4–8 Interestingly, ectopic expression of FGF19 in skeletal muscle of transgenic mice resulted in enhanced hepatocellular proliferation and in the development of HCC,9 a process that was accelerated in mice treated with the carcinogen diethylnitrosamine (DEN) and was dependent on hepatocellular FGFR4 expression.8 In
spite of this compelling evidence, the role of endogenous FGF15/FGF19 in hepatocarcinogenesis has not been directly evaluated so far. We recently identified an important function for FGF15 in liver regeneration and hepatocellular proliferation in mice.10 In this study we demonstrate that FGF15 contributes to HCC progression in a clinically relevant model of liver cancer.11
Material and Methods Induction of HCC and acute CCl4 treatment
Fgf152/2 mice and their wild type littermate controls (Fgf151/1) are on a mixed C57BL/6129/Sv background and have been backcrossed for more than 16 generations. These mice have been described before.10 HCC was induced essentially as previously described with minor modifications.11 Briefly, 15-day-old mice (n 5 13 Fgf152/2 and n 5 21 Fgf151/1) were injected with DEN (25 mg/Kg, i.p.) followed by biweekly i.p. injections of CCl4 (0.5 mL/Kg) starting at 4 weeks of age. Mice were weighted and killed at 27 weeks of age, after 12 CCl4 injections. Livers were excised, weighted and macroscopically examined for gross lesions. Liver index, the ratio between liver and body weight (expressed as %), was calculated. Tumors size was estimated with a Vernier caliper. Tumoral lesions and tumor-free tissues were excised and either snapfrozen in liquid N2, or paraffin embedded for histological analyses. Acute CCl4 administration was performed as described.12 Histological analyses
Liver tissue sections from all mice were stained with H&E and examined by light microscopy (Olympus BX-51) by two pathologists (S.C. and S.M.) blinded to the identity of samples. Lesions (dysplastic foci, adenomas and HCCs) were classified according to published criteria.13 a2Smooth muscle actin (a2Sma), Ctgf (antibody sc-14939, Santa Cruz Biotechnology) and Ki67 immunodetection and Sirius Red staining were performed essentially as described.14 Cell isolation, culture and transfections
Primary HSCs and hepatocytes were isolated from Fgf151/1 mice essentially as described.15,16 The origin and culture conditions of mouse AML12 hepatocytes, human Hep3B and HepG2 cells have been previously reported.16 Transfections with CTGF-promoter luciferase reporter construct (nucleotides 2485 to 115 of human CTGF 5’ region), and with b-catenin, glycogen synthase kinase-b (GSK3-b) and control siGL siRNAs (all from Dharmacon), were performed as described.17
C 2014 UICC Int. J. Cancer: 00, 00–00 (2014) V
3
Uriarte et al.
Figure 1. HCC development is attenuated in Fgf152/2 mice. Mice were injected with DEN followed by CCl4 as indicated and sacrificed at 27 weeks of age (a). Shown are liver index (b), representative liver images (c), tumor number (d), mean size of tumors (e), representative H&E sections highlighting different types of lesions (f), and mean diameter of histological lesions (g). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Total RNA from liver tissue and cell lines was extracted using the automated Maxwell system from Promega. Reverse transcription was performed as reported.14 Real-time PCRs were performed with iQ SYBR Green supermix (BioRad) in a CFX96 system from BioRad as previously described.14 Primers used are described in Supporting Information Table 1. Western blot analyses
Cells and liver tissues were lysed and homogenates were subjected to Western blot analysis as reported.14 CTGF and amphiregulin (AR) were detected as described.12,17 FGF15 determination
FGF15 levels were measured in mouse serum and liver tissues by ELISA (MyBiosource, San Diego). Sera from mice fed a 1% cholate-supplemented diet (CA1%) for 5 days and mice infected with an FGF15 expressing adenovirus were used as controls.10 Background signals from Fgf152/2 mice sera and liver extracts were substracted to all values. Statistical analysis
Data are means 6 SEM. Analyses were performed using GraphPad Prism version 5.00 (GraphPad Software, San C 2014 UICC Int. J. Cancer: 00, 00–00 (2014) V
Diego). Data were compared among groups using the Student’s t-test. A p value of
