Am J Physiol Gastrointest Liver Physiol 306: G893–G902, 2014. First published April 3, 2014; doi:10.1152/ajpgi.00337.2013.
Fibroblast growth factor 15 deficiency impairs liver regeneration in mice Bo Kong,1* Jiansheng Huang,2* Yan Zhu,3 Guodong Li,4 Jessica Williams,6 Steven Shen,6 Lauren M. Aleksunes,1 Jason R. Richardson,5 Udayan Apte,6 David A. Rudnick,2 and Grace L. Guo1 1
Department of Pharmacology and Toxicology, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, New Jersey; 2Department of Pediatrics, Washington University School of Medicine, St. Louis, Missouri; 3 Department of General Surgery, Xuanwu Hospital, Capital Medical University, Beijing, Peoples Republic of China; 4 Department of Surgical Oncology, Cancer Treatment Center, Fourth Affiliated Hospital of Harbin Medical University, Harbin, Peoples Republic of China; 5Department of Environmental and Occupational Medicine, Rutgers Robert Wood Johnson Medical School, Rutgers, The State University of New Jersey, New Brunswick, New Jersey; and 6Department of Pharmacology, Toxicology, and Therapeutics, University of Kansas Medical Center, Kansas City, Kansas Submitted 7 October 2013; accepted in final form 30 March 2014
Kong B, Huang J, Zhu Y, Li G, Williams J, Shen S, Aleksunes LM, Richardson JR, Apte U, Rudnick DA, Guo GL. Fibroblast growth factor 15 deficiency impairs liver regeneration in mice. Am J Physiol Gastrointest Liver Physiol 306: G893–G902, 2014. First published April 3, 2014; doi:10.1152/ajpgi.00337.2013.—Fibroblast growth factor (FGF) 15 (human homolog, FGF19) is an endocrine FGF highly expressed in the small intestine of mice. Emerging evidence suggests that FGF15 is critical for regulating hepatic functions; however, the role of FGF15 in liver regeneration is unclear. This study assessed whether liver regeneration is altered in FGF15 knockout (KO) mice following 2/3 partial hepatectomy (PHx). The results showed that FGF15 KO mice had marked mortality, with the survival rate influenced by genetic background. Compared with wildtype mice, the KO mice displayed extensive liver necrosis and marked elevation of serum bile acids and bilirubin. Furthermore, hepatocyte proliferation was reduced in the KO mice because of impaired cell cycle progression. After PHx, the KO mice had weaker activation of signaling pathways that are important for liver regeneration, including signal transducer and activator of transcription 3, nuclear factor-B, and mitogen-activated protein kinase. Examination of the KO mice at early time points after PHx revealed a reduced and/or delayed induction of immediate-early response genes, including growth-control transcription factors that are critical for liver regeneration. In conclusion, the results suggest that FGF15 deficiency severely impairs liver regeneration in mice after PHx. The underlying mechanism is likely the result of disrupted bile acid homeostasis and impaired priming of hepatocyte proliferation. bile acids; farnesoid X receptor; partial hepatectomy THE LIVER HAS THE ABILITY to fully regenerate itself following partial resection or toxin-induced damage through a highly regulated process. Two-thirds partial hepatectomy (PHx) in rodents is the most common and established model to identify regulators of liver regeneration. Many factors have been shown to be critical to liver regrowth (21). Following PHx, a series of signals and pathways orchestrate the initiation, proliferation, and termination of cell cycle and cell proliferation, and liver mass is restored to the original size within a short period of time (21, 27). However, the exact mechanisms for the initiation and termination of proliferation remain unclear.
* B. Kong and J. Huang contributed equally to this work. Address for reprint requests and other correspondence: G. L. Guo, Dept. of Pharmacology and Toxicology, Ernest Mario School of Pharmacy, Rutgers, The State Univ. of New Jersey, Piscataway, NJ 08854 (e-mail: guo@eohsi. rutgers.edu). http://www.ajpgi.org
Bile acids have emerged as critical mediators of liver regeneration (11) and serve as signaling molecules and endogenous ligands for the farnesoid X receptor (FXR) (35). Bile acids promote liver regeneration, and whole body FXR deficiency enhances mortality and delays liver regeneration following PHx or chemical-induced liver injury in mice (11). FXR is highly expressed in the liver and intestine, and activation of FXR in both organs maintains bile acid homeostasis in a coordinated fashion (13, 15). Selective deletion of FXR in the livers of mice does not completely block, but delays, liver regeneration (5), suggesting that FXR signaling in the intestines and/or other organs may be critical in regulating liver regeneration. In the intestine, activation of FXR induces fibroblast growth factor 15 (FGF15), the mouse homolog of human FGF19. FGF15 is one of the endocrine FGFs, since it is highly expressed in the small intestine, but not in the liver. FGF15 travels to the liver, where it regulates a variety of hepatic functions. The best-known role of FGF15 in the liver is to suppress bile acid synthesis (12). Specifically, FGF15 accomplishes this suppression through activation of the fibroblast growth factor receptor 4 (FGFR4), a tyrosine kinase receptor expressed on the hepatocyte membrane that subsequently leads to the activation of downstream mitogen-activated protein kinase (MAPK) signaling cascades, including ERK1/2 and JNK1/2, and ultimately results in the suppression of the expression of bile-acid synthetic genes (15, 25). Emerging evidence suggests that FGF15/19 has dual functions. Indeed, besides regulating lipid and energy homeostasis, FGF19 promotes cell proliferation in vitro, and FGF19 transgenic mice develop liver tumors (23, 24). However, the functions and mechanisms of FGF15/19 in liver regeneration after PHx are not well defined. In the current study, we determined whether liver regeneration following PHx is altered by deficiency in FGF15 signaling in mice. MATERIALS AND METHODS
Materials. All chemicals were obtained from Sigma (St. Louis, MO). Antibodies were purchase from Cell Signaling (Danvers, MA). IL-6 Elisa kit was obtained from Abcam (Cambridge, MA). Animals. FGF15 knockout (KO) mice were provided by Dr. Curtis Klaassen at the University of Kansas Medical Center. Under pure C57BL/6J genetic background, FGF15 KO mice are mostly embryonic lethal (31), but they are viable and fertile under the 129SvJ genetic background. In detail, the heterozygous FGF15 KO mice
0193-1857/14 Copyright © 2014 the American Physiological Society
G893
G894
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
Table 1. Primers used for real-time quantitative PCR Gene
Forward Primer (5=-3=)
Reverse Primer (5=-3=)
Bsep Ccnd1 Ccne1 Cebp- c-Fos c-Jun c-Myc Cyp7a1 Egfr Fxr Gapdh Hgf IL-1 IL-6 NF-b Ntcp Ost Shp Socs3 Survivin Tgf-␣ Tgf-1 Tnf␣
TGAATGGACTGTCGGTATCTGTG GAAGGAGACCATTCCCTTGA TGTTACAGATGGCGCTTGCTC CGCCTTTAGACCCATGGAAG AATGGTGAAGACCGTGTCAGG TCTTCATTTTCTCACCAACTGCTT ACTTACAATCTGCGAGCCAGGACA AACAACCTGCCAGTACTAGATAGC TCAGCAACAACCCCATCCTC TCCGGACATTCAACCATCAC TGTGTCCGTCGTGGATCTGA GGTGTATCAGGAACAGGGGC ACTCAACTGTGAAATGCCACCTT CAACGATGATGCACTTGCAGA TGGCAGCTCTTCTCAAAGCA GGCCACAGACACTGCGCT GTATTTTCGTGCAGAAGATGCG CGATCCTCTTCAACCCAGATG GGAGATTTCGCTTCGGGACT CACCCCAGAGCGAATGGCGG TCCTGTTAGCTGTGTGCCAG CAACTTCTGTCTGGGACCCT ATGGCCTCCCTCTCATCAGT
CCACTGCTCCCAACGAATG GTTCACCAGAAGCAGTTCCA TTCAGCCAGGACACAATGGTC CCCGTAGGCCAGGCAGT CCCTTCGGATTCTCCGTTTCT CTCTCCAAATGCTCCCCAAA GCCCAAAGGAAATCCAGCCTTCAA GTGTAGAGTGAAGTCCTCCTTAGC GCTTGGATCACATTTGGGGC TCACTGCACATCCCAGATCTC CCTGCTTCACCACCTTCTTGAT GTCAAATTCATGGCCAAACCCT TGCTGCTGCGAGATTTGAAG GGTACTCCAGAAGACCAGAGG CCAAGAGTCGTCCAGGTCATAGA AGTGAGCCTTGATCTTGCTGAACT TTTCTGTTTGCCAGGATGCTC AGGGCTCCAAGACTTCACACA AAACTTGCTGTGGGTGACCA GTGAGGAAGGCGCAGCCAGG TGTGGGAATCTGGGCACTTG TAGTAGACGATGGGCAGTGG GCTCCTCCACTTGGTGGTTT
(50% C57BL/6J and 50% 129SvJ genetic background) were backcrossed for one generation to 129SvJ or C57BL/6J genetic background, respectively, and then the heterozygotes were intercrossed to generate homozygous KO and wild-type (WT) littermate control mice. These littermates with the same background were used to establish and expand the colony. The mouse genetic background was further validated by the Illumina Mouse MD Linkage Panel consisting of 1,440 single-nucleotide polymorphisms throughout the mouse genome. Briefly, two strains of the FGF15 KO mice on different genetic backgrounds were used in the current study, and the corresponding WT mice with the same genetic background served as controls: FGF15 KO (75% C57BL/6J and 25% 129SvJ) and FGF15 KO (A) (75% 129SvJ and 25% C57BL/6J). The mice were maintained on an ad libitum diet of rodent chow (Charles River Laboratories, Wilmington, MA) with a 12:12-h light-dark cycle. Procedures involving animals were conducted according to a protocol approved by the University of Kansas Medical Center Institutional Animal Care and Use Committee. All mice were closely monitored after surgery for general well-being, and experiments were terminated early when mice became moribund by showing significant decrease in activity, body weight, and body temperature. Two-thirds PHx. Eight-to-ten-week-old male mice (n ⫽ 4 –12 mice/group) underwent standard PHx surgery (10). FGF15 KO and FGF15 KO (A) mice were examined at various time points in separate experiments. Blood, liver, and small intestine were collected using a published method (15). Bromodeoxyuridine immunohistochemistry. One hour before livers were harvested, mice were injected intraperitoneally with 100 mg bromodeoxyuridine (BrdU)/kg body wt. BrdU staining was performed
by standard immunohistochemistry methods (20). The degree of BrdU incorporation into newly synthesized DNA was determined by counting positively stained hepatocyte nuclei in five microscopic fields per animal at ⫻400 magnification. The value was expressed as a percentage of BrdU-labeled nuclei vs. all nuclei in the fields. Serum parameter analysis. Serum activity of alanine aminotransferase (ALT), alkaline phosphatase (ALP), and bilirubin were analyzed using commercial kits (Pointe Scientific, Canton, MI). Total bile acids were measured by a kit from Bioquant (San Diego, CA). IL-6 levels were analyzed using an Elisa kit from Abcam. Quantification of hepatic mRNA expression. The mRNA levels were determined by RT-qPCR by a published method (28). Primer sequences are shown in Table 1. Western blot. Equal amounts of protein from individual mice were pooled in each group. Fifteen micrograms nuclei protein, 30 g total cell extract or cytosolic protein were load onto 10⬃15% PAGE gels. Western blot analysis was performed by standard techniques. The band’s densities from Western blot were semiquantified by using ImageJ software (http://rsb.info.nih.gov/ij/). The value indicated the relative density of that band compared with the density of the band of its loading control GAPDH or histone H3 (nuclear protein). Statistical analysis. The data are expressed as means ⫾ SE. All statistical tests used were done using Sigma Stat Statistical Software (Systat Software, Point Richmond, CA). The differences between the PHx and sham groups, or between WT and KO mice, were determined by t-test. The statistical significance between multiple groups was analyzed by one-way ANOVA followed by the Student-NewmanKeuls test. P ⬍ 0.05 is considered statistically significant.
Fig. 1. Mortality rate, liver growth, and liver injury in mice with fibroblast growth factor (FGF) 15 deficiency after partial hepatectomy (PHx). A: Kaplan-Meier analysis of the survival of wild-type (WT), FGF15 knockout (KO), and FGF15 KO (A) mice after PHx; and changes in the ratio of remnant to original liver weight during liver regeneration in WT and FGF15 KO (A) mice. Data are expressed as a percentage of the initial value before PHx. B: serum alanine aminotransferase (ALT), alkaline phosphatase (ALP), bilirubin, and bile acid levels in WT, FGF15 KO, and FGF15 KO (A) mice following PHx. C: representative hematoxylin and eosin staining (⫻400) of liver sections from WT, FGF15 KO, and FGF15 KO (A) mice after PHx. Inset, the necrotic area in FGF15 KO (A) mice. *P ⬍ 0.05, significant difference between WT and FGF15 KO mice at the same time point. FGF15 KO (75% C57BL/6J and 25% 129SvJ; PHx 0, 0.5, 3, 6, 12, 24, 36, and 48 h, n ⫽ 7, 5, 4, 4, 5, 5, 9, and 10 for WT, respectively; n ⫽ 4, 5, 4, 4, 4, 4, 12, and 10 for FGF15 KO, respectively) and FGF15 KO (A) (75% 129SvJ and 25% C57BL/6J; PHx 0, 6, 24, 36, 72, 96, 168, and 336 h, n ⫽ 3, 4, 7, 8, 6, 8, 5, and 5 for WT, respectively; n ⫽ 4, 4, 10, 10, 7, 11, 5, and 7 for FGF15 KO, respectively). AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
G895
G896
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
RESULTS
Increased liver necrosis and mortality rate of FGF15 KO mice after PHx. FGF15 KO mice on two genetic backgrounds [FGF15 KO and FGF15 KO (A)] and their corresponding WT controls were subjected to PHx. Before surgery, liver-to-body weight ratios were similar between WT and either FGF15 KO or FGF15 KO (A) mice (about 4.5%). In addition, the small intestine of WT mice on either C57BL/6J or 129SvJ background had similar baseline FGF15 mRNA levels (data not shown). After PHx, FGF15 KO mice showed significant mortality compared with WT controls, which was strain dependent. KO mice that survived the PHx also displayed a reduced recovery of liver mass. Specifically, 24 – 48 h after PHx, the mortality rate was 90% in FGF15 KO mice and 60 – 65% in FGF15 KO (A) mice, whereas only ⬃10% WT mice died (Fig. 1A). However, the surviving FGF15 KO (A) mice had reduced liver growth for the entire duration following PHx (Fig. 1A). The increased mortality in FGF15 KO mice was not likely because of basal liver injury, because before surgery, serum levels of ALT, ALP, bilirubin, and bile acids were similar between WT and FGF15 KO mice (data not shown). After PHx, these parameters were transiently and moderately increased in WT mice and returned to normal ranges within 72 h. In contrast, FGF15 KO mice showed a significant elevation of these parameters at 24 and 36 h (Fig. 1B). Furthermore, the FGF15 KO mice, which suffered more mortality, showed a marked increase in these parameters at 48 h, whereas the FGF15 KO (A) mice, which had less mortality, reduced these parameters by 72 h. Severe hepatocyte necrosis was observed in FGF15 KO mice, but not in WT mice, at 24 h after PHx (Fig. 1C), and the necrotic area further increased in the KO mice at 36 and 48 h after PHx (Fig. 1C). In FGF15 KO (A) mice, the region of necrosis was less extensive than observed in FGF15 KO mice following PHx (Fig. 1C). Impaired hepatocyte DNA replication after PHx in FGF15 KO mice. Staining of BrdU incorporated into DNA 24 and 36 h after PHx is routinely used to assess DNA replication in hepatocytes. In this study, a significant decrease in BrdU labeling was observed in FGF15 KO mice at 24 to 48 h after PHx (Fig. 2A). At 36 h, the time point of peak DNA replication, the percentage of hepatocytes stained with BrdU, was close to 40% in WT mice but was only 18% in FGF15 KO mice (Fig. 2B). A similar decrease in BrdU incorporation was observed in FGF15 KO (A) mice (data not shown). Expression of genes involved in cell-cycle progression, bile acid homeostasis, and inflammation. Because most of the FGF15 KO mice suffered mortality 24 – 48 h after PHx, the mRNA levels of genes Ccnd1 and Ccne1 that encode Cyclin D1 and Cyclin E1 were determined at 6, 12, and 24 h after PHx. Both Cyclin D1 and Cyclin E1 mRNA levels were low and similar between WT and KO mice at 6 and 12 h following PHx but were significantly upregulated at 24 h (Fig. 3, A and B), However, Ccnd1 mRNA levels were much lower in FGF15 KO and FGF15 KO (A) compared with WT mice 24 h after PHx. Interestingly, Ccne1 mRNA levels were upregulated in mice with low Ccnd1 expression level, indicating a compensatory response to the lack of Ccnd1 induction. The protein levels of several critical regulators for hepatocyte proliferation following PHx were then determined (Fig. 3C). In WT mice, Cyclin D1 was increased markedly at 36 h
Fig. 2. Impaired hepatocyte proliferation after PHx in FGF15-deficient mice. A: representative bromodeoxyuridine (BrdU) staining (⫻400) of liver sections from WT and FGF15 KO mice after PHx. B: the percentage of BrdU-positive hepatocytes in WT and FGF15 KO mice. *P ⬍ 0.05, significant difference between WT and FGF15 KO mice at the same time point.
and then decreased at 48 h. In comparison, the KO mice had much lower Cyclin D1 levels at 36 h, but the levels were increased at 48 h. Proliferating cell nuclear antigen was markedly induced in WT mice at both 36 and 48 h but was undetectable in the KO mice at any time point. Similar changes were observed for CDK2 and CDK4. Transcription factors CDC2, CDC25b, and FoxM1b are critical regulators of G2/M transition and mitosis and were upregulated at 36 and 48 h in WT mice, but their levels were much lower in FGF15 KO mice (Fig. 3D). Because FGF15 is known to suppress bile acid synthesis and the impaired liver regeneration in FGF15 KO mice is accompanied by increased serum bile acids, the mRNA levels of genes encoding proteins involved in bile acid homeostasis were measured at 0, 0.5, 3, 36, and 48 h following PHx. Additionally, levels of inflammatory cytokines involved in liver regeneration and cell-cycle regulators were determined. The mRNA levels of FXR were not changed dramatically by PHx. The mRNA levels of sodium taurocholate cotransporting polypeptide (for hepatic bile acid uptake) and bile salt efflux pump (for canalicular bile acid excretion) gradually decreased with time after PHx, but without statistically significant differences between WT and FGF15 KO mice (Fig. 3E). Cyp7a1 gene encodes the rate-limiting enzyme in bile acid synthesis, and its expression is suppressed by FGF15 (12, 15).
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
G897
Fig. 3. Expression of genes in bile acid homeostasis, inflammation, and cell cycle progression. mRNA levels of Ccnd1 (A) and Ccne1 (B) in FGF15 KO and FGF15 KO (A) mice. C: protein levels of cell cycle-related proteins after PHx. D: mRNA levels of Cdc25b and FoxM1b, and protein levels of phospho-CDC2 in cytosol and nuclei. E and F: levels of mRNA of genes involved in bile acids homeostasis (E) and inflammation (F) and PHx 0, 0.5, 3, 36, and 48 h, n ⫽ 7, 5, 4, 9, and 10 for WT, respectively; n ⫽ 4, 5, 4, 12, and 10 for FGF15 KO, respectively. *P ⬍ 0.05, significant difference between WT and FGF15 KO mice at the same time point.
In WT mice, the mRNA levels of Cyp7a1 did not change at 0.5 and 3 h but increased ⬃6- and 2-fold at 36 and 48 h following PHx (Fig. 3E). In FGF15 KO mice, the basal hepatic Cyp7a1 mRNA levels were 20-fold higher than those in WT mice, but dropped to 10-fold at 0.5 h after PHx, and returned to WT levels at 3 h. At 36 and 48 h, the levels of Cyp7a1 mRNA were almost undetectable in FGF15 KO mice. The mRNA levels of
small heterodimer partner (SHP) were induced at 3 h in FGF15 KO mice, which were fivefold higher than those in WT mice. The mRNA levels of organic solute transporter- (Ost) were robustly upregulated (⬃25-fold) in FGF15 KO mice at 3 h following PHx, and further increased at 36 and 48 h (⬃150- to 180-fold) in FGF15 KO mice. Induction of SHP and Ost indicates FXR activation in the liver.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
G898
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
The mRNA levels of the inflammatory factors, IL-1, IL-6, tumor necrosis factor (Tnf)-␣, and transforming growth factor (Tgf)-1, known to be critical for liver regeneration, were measured (Fig. 3F). The expression of IL-1, IL-6, and Tnf␣ was increased at 0.5 and 3 h post-PHx, but no difference was observed between the WT and KO mice, except for Tnf␣ mRNA, which was higher in KO mice than in WT mice at 3 h (⬃2-fold). However, at 36 and 48 h, a marked increase was observed for IL-1, IL-6, and Tnf␣ in KO mice but not in WT mice. No significant difference of Tgf-1 expression was found between WT and KO mice. Impaired activation of mitogen-activated protein kinase, signal transducer and activator of transcription 3, and NF-B signaling in FGF15 KO mice after PHx. Because FGF15 deficiency causes severe inhibition of DNA synthesis and hepatocyte proliferation, the activation status of signaling pathways involved in “survival” at early stages of liver regeneration were assessed. The MAPK pathways are important in liver regeneration because c-Jun- (4, 26) and Egr1-deficient (19) mice exhibit severe defects in liver regeneration following PHx. FGF15 is known to activate the JNK/ERK signaling pathways (15). In the FGF15 KO mice following PHx, JNK and p38 were activated to a lesser degree than in WT (Fig. 4). Interestingly, ERK activation appeared to be stronger in FGF15 KO mice (Fig. 4). Signal transducer and activator of transcription 3 is critical to several cytokine-controlled cellular processes. A cohort of cytokines and growth factors activate signal transducer and activator of transcription 3 (STAT3) proteins downstream of JAK2 and JAK1; upon stimulation, activated STAT3s dimerize and translocate into the nucleus, where they bind to target genes and induce transcription (30). STAT3 has been shown to be critical in liver regeneration (7, 18). In this study, total JAK1, JAK2, and c-Myc were lower in KO mice before and after PHx (Fig. 5A). In WT mice, STAT3 was markedly activated, as measured by Y705 phosphorylation as early as 0.5 h following PHx, and by S727 phosphorylation at 36 h (Fig. 5B). In contrast, FGF15 KO mice had much lower Y705 phosphorylation and S727 phosphorylation after PHx levels. In addition, the total STAT3 levels in nuclear extracts were decreased in the KO mice (Fig. 5B).
IL-6 is a strong STAT3 activator, and serum IL-6 was increased as early as 3 h after PHx in both WT and KO mice. However, IL-6 levels subsequently decreased in WT mice at 36 and 48 h. In contrast, serum IL-6 remained elevated in KO mice after PHx (Fig. 5C). Furthermore, the mRNA levels of STAT3 target genes were higher in WT mice at 36 h (c-Myc and survivin) and at 48 h (survivin). Socs3, which is a STAT3 inhibitor and a STAT3 target gene, was induced equally in WT and KO mice at 0.5 and 3 h. However, the degree of induction was less at 36 and 48 h in WT mice and markedly increased in KO mice (Fig. 5D). NF-B is highly activated within hours after PHx; therefore, NF-B activation has been considered a critical initiation signal in liver regeneration (8). Moreover, NF-B activation prevents apoptosis and promotes hepatocyte survival (3). In the current study, increased nuclear p65 protein was apparent in WT mice, but this was absent in the FGF15 KO mice (Fig. 6), indicating that FGF15 deficiency disrupts NF-B signaling following PHx. The other well-known survival signaling pathways, such as phosphatidylinositol 3-kinase/protein kinase B, mammalian target of rapamycin, and insulin receptor signaling pathway [insulin-like growth factor-I-binding protein, insulinlike growth factor (IGF)-I, and IGF-II], showed no significant difference between WT and FGF15 KO mice after PHx (data not shown). Impaired gene expression in the early stage of liver regeneration in FGF15 KO mice after PHx. MAPK, STAT3, and NF-B activation are crucial for the priming phase of liver regeneration and may be responsible for the early response following PHx (1, 2, 33). The results shown above suggested that the FGF15 mice were deficient in initiating liver regeneration after PHx. c-Jun and c-Fos are components of activator protein-1 (AP-1), which is important for initiation of regeneration. The FGF15 KO mice showed a delay in induction of the mRNAs of c-Jun and c-Fos. At 10 min after PHx, the induction of c-Jun and c-Fos was markedly higher in WT mice than that in FGF15 KO mice. Finally, at 30 min, FGF15 KO mice had equal (c-Jun) or higher (c-Fos) expression than the WT mice had (Fig. 7A). A number of additional growth factors and transcription factors, including hepatocyte growth factor (HGF), epidermal growth factor receptor (EGFR), TGF-␣, IL-6, NF-B, and
Fig. 4. Disturbed mitogen-activated protein kinase (MAPK) signaling pathways after PHx in FGF15 KO mice. The mammalian MAPK family includes JNK, ERK, and p38. Signaling cascades induce phosphorylation/ activation of the MAPK signaling pathway and play a central role in the regulation of genes involved in hepatocyte proliferation and in protecting against apoptotic cell death.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
G899
Fig. 5. Impaired activation of signal transducer and activator of transcription (STAT) 3 in FGF15 KO mice after PHx. A: total JAK1/2 and c-Myc protein level. B: total and phospho-proteins associated with STAT3 signaling pathways. C: serum IL-6 concentration in WT and FGF15 KO mice after PHx. D: gene expression of Survivin and Socs3. PHx 0, 0.5, 3, 36, and 48 h, n ⫽ 7, 5, 4, 9, and 10 for WT, respectively; n ⫽ 4, 5, 4, 12, and 10 for FGF15 KO, respectively; *P ⬍ 0.05 between WT and FGF15 KO mice at the same time point.
CCAAT enhancer-binding protein- (CEBP-), are involved in initiating the hepatic regenerative process. The mRNA levels of Hgf and Egfr at 10 min and NF-B and Tgf-␣ at 30 min after PHx were higher in WT mice than in FGF15KO mice, but mRNA levels of other genes examined did not show significant differences across genotypes (Fig. 7B).
DISCUSSION
FGF15 is the mouse homolog of human FGF19 and highly expressed in the small intestine. Emerging evidence suggests that FGF15/19 is an endocrine FGF and is critical for a variety of liver functions, including suppressing bile acid synthesis,
Fig. 6. NF-B signaling deactivated after PHx. Activation of the NF-B signaling pathway is associated with nuclear translocation of the p65 component of the complex. The translocation of the cytosolic p65 to the nucleus decreased in FGF15 KO mice after PHx.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
G900
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
Fig. 7. FGF15 deficiency affected the expression of immediate-early response genes (A) and growth factors and transcription factors (B) after PHx. WT and FGF15 KO mice underwent Sham or PHx surgery (n ⫽ 5/group at each time point). Liver tissue samples were collected at 10 and 30 min after surgery. mRNA levels were determined. *P ⬍ 0.05 between WT and FGF15 KO mice at the same time point.
promoting protein synthesis, and improving insulin sensitivity (12, 14, 15, 32). However, the role of FGF15 in liver regeneration is not fully defined. The current study demonstrates that mice deficient in FGF15 have severely impaired liver regeneration following PHx. The FGF15 KO mice exhibited severe liver necrosis, elevated levels of bile acids, and high mortality. Markers indicating successful hepatocyte proliferation were diminished in FGF15 KO mice after PHx. Furthermore, activation of several transcription factors critical for liver regeneration, including NF-B and STAT3, was abolished at early time points in FGF15 KO mice following PHx. This study clearly demonstrates that FGF15 deficiency causes acute liver failure in mice after PHx, and the KO mice failed to regenerate their livers. Our results are consistent with the recently published study of Uriarte and coworkers (29). In that study, increased bile acids and reduced FGF15 both contributed to liver failure following PHx, as evidenced by finding that both administration of cholestyramine to reduce bile acid levels and exogenous Fgf15 protein administration reduced liver necrosis and mortality and promoted liver regeneration in the FGF15 KO mice after PHx. In the current study, we confirmed and extended those findings by demonstrating that mouse genetic background is crucial for the severity of liver failure in FGF15 KO mice after PHx. Under the C57BL/6J background, FGF15 KO mice are mostly embryonic lethal, since ⬍2% of the homozygous mice can survive (31), but under 129SvJ background, FGF15 KO mice are viable and fertile (unpublished data). Our results showed that, under 75% C57BL/6J and 25% 129SvJ background, FGF15 KO suffered more liver damage and significantly higher mortality rate compared with KO mice under 25% C57BL/6J and 75%
Fig. 8. Schematic representation of effects on bile acid homeostasis and liver regeneration after FGF15 deletion. Activation of FXR in the intestine induces FGF15 protein, which could repress Cyp7a1 gene transcription by the activation of JNK/ERK pathways in the liver through its receptor FGFR4. Lack of FGF15 feedback suppression after FGF15 deletion leads to increased bile acid biosynthesis, and accumulation of bile acids would cause toxicity and damage to hepatocytes. In addition to feedback regulation of Cyp7a1 transcription, reduced JNK/ERK activation could deactivate STAT3/NF-B signaling pathways, which are crucial for the priming phase of liver regeneration.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
129SvJ background. Uriarte et al. (29) also used the KO mice on a mixed C57BL/6J/129/SvJ background, but the percentage of each background was not reported. The role of FGF15 in liver regeneration after PHx is not without controversy, since mice deficient in FGFR4, which is believed to be the major receptor of FGF15 in the liver, showed only slight delay in liver regeneration after PHx (34). Because the FGFR4 KO mice used for PHx were on 129Sv background, and these mice only showed slight delay in liver regeneration, it is likely that the difference between FGF15 KO mice and FGFR4 KO mice in response to PHx is partly due to the genetic background. Indeed, strain-dependent effects on liver proliferation and carcinogenesis have been reported (17, 22). However, it cannot be excluded that FGF15 may activate other receptors in the FGFR4 KO mice, since FGF19 was reported to activate not only FGFR4 receptor but also several other FGFRs that are expressed in the liver (15, 16). The mechanism of the strain-dependent difference of the FGF15 pathway is not known. However, our initial analysis indicates that the mRNA levels of FGF15 in mice with pure C57BL/6J or 129SvJ background do not differ, suggesting that levels of expressed FGF15 mRNA do not explain the impact of genetic background on severity of liver injury following PHx in the FGF15 KO. Variation in certain genetic loci(s) between these two stains might be responsible for the differential response to FGF15 deletion after PHx, so future investigation of the genetic loci(s) that is responsible for the strain-specific differences may help to determine novel pathways in regulating liver growth. FGF15 is a major suppressor for bile acid synthesis, and high bile acids were persistently present in KO mice following PHx, whereas the WT mice only showed a modest and transient increase in serum bile acids following PHx. The basal Cyp7a1 expression was markedly higher in the KO mice (20⫻), but this was sharply suppressed within 3 h following PHx. The other mechanisms would include the induction of SHP and inflammation in the KO mice starting at 3 h post-PHx because SHP and inflammation are known to suppress Cyp7a1 gene expression (6, 9). However, the acute suppression of Cyp7a1 did not reduce bile acid levels. Furthermore, a dramatic induction of Ost mRNA in the KO mice following PHx indicates enhanced bile acid burden, and the increase in bile acids is likely a mechanism responsible for more extensive liver necrosis and mortality in FGF15 KO mice after PHx. We did not define the effects of reducing bile acids on liver regeneration in FGF15 KO mice. However, recent studies reported that reducing the bile acid pool decreased mortality after PHx in FGF15 KO or FXR KO mice (11, 29). Most FGF15 KO mice died within 36 – 48 h after PHx, and these mice had increased bile acid levels. In the mice that survived, there was no evidence of normal liver regeneration after PHx as evaluated by commonly used markers of regeneration (1–3). The activation of immediate-early response genes is believed to be driven by inflammation and, in particular, increased TNF-␣ and IL-6, classical activators of NF-B and STAT3, respectively. Although we did not observe a reduction of the expression of these cytokines in the KO mouse liver following PHx at early time points (0, 0.5, and 3 h), activation of both NF-B and STAT3 was much lower in the KO mice. AP-1 expression was also decreased, and several growth factors and transcription factors were reduced at 10 min
G901
after PHx in FGF15 KO mice, indicating that deletion of FGF15 may have affected the initiation of liver regeneration, which could result in impaired activation of genes required for cell survival. However, it is unclear how FGF15 facilitates the activation of hepatic NF-B and STAT3 signaling pathways. One possibility is that FGF15 binds and activates its receptor FGFR4, which, through a tyrosine phosphorylation cascade, activates STAT3 and NF-B signaling pathways. The results reported in the current study further confirm that intestinal FGF15 could promote liver regeneration in mice, and FGF15 gene deletion in mice markedly increases mortality after PHx (29). Moreover, we have shown that mouse genetic background affects FGF15-regulated liver regeneration. More importantly, we showed that, after PHx, the FGF15 KO mice had impaired signaling pathways that are critical in liver regeneration, including STAT3, NF-B, and MAPK. In summary, the current study suggests two roles that FGF15 may play in promoting liver regeneration (Fig. 8): 1) FGF15 deletion disturbs bile acid homeostasis, and excessive bile acid accumulation in the regenerating liver can cause hepatocyte damage and disrupt hepatic regeneration and 2) deletion of FGF15 may lead to a defect in the initiation of liver regeneration, which might contribute to liver failure in the long term. These two roles might interact with each other, and future studies will be needed to clarify the underlying mechanism(s). ACKNOWLEDGMENTS We greatly appreciate the suggestions from Dr. George K. Michalopoulos at the University of Pittsburgh. GRANTS This study was supported by the National Institutes of Health fund [DK081343, DK-090036, and GM-104037 (G. L. Guo), as well as ES-005022]. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the authors. AUTHOR CONTRIBUTIONS Author contributions: B.K., J.H., and G.L.G. conception and design of research; B.K., J.H., Y.Z., G.L., J.A.W., S.S., and G.L.G. performed experiments; B.K., J.H., and G.L.G. analyzed data; B.K., J.H., L.M.A., J.R.R., U.A., D.A.R., and G.L.G. interpreted results of experiments; B.K., J.H., and G.L.G. prepared figures; B.K., J.H., and G.L.G. drafted manuscript; B.K., J.H., Y.Z., G.L., J.A.W., S.S., L.M.A., J.R.R., U.A., D.A.R., and G.L.G. edited and revised manuscript; B.K., J.H., Y.Z., G.L., J.A.W., S.S., L.M.A., J.R.R., U.A., D.A.R., and G.L.G. approved final version of manuscript. REFERENCES 1. Akerman P, Cote P, Yang SQ, McClain C, Nelson S, Bagby GJ, Diehl AM. Antibodies to tumor necrosis factor-alpha inhibit liver regeneration after partial hepatectomy. Am J Physiol Gastrointest Liver Physiol 263: G579 –G585, 1992. 2. Aldeguer X, Debonera F, Shaked A, Krasinkas AM, Gelman AE, Que X, Zamir GA, Hiroyasu S, Kovalovich KK, Taub R, Olthoff KM. Interleukin-6 from intrahepatic cells of bone marrow origin is required for normal murine liver regeneration. Hepatology 35: 40 –48, 2002. 3. Beg AA, Sha WC, Bronson RT, Ghosh S, Baltimore D. Embryonic lethality and liver degeneration in mice lacking the RelA component of NF-kappa B. Nature 376: 167–170, 1995. 4. Behrens A, Sibilia M, David JP, Mohle-Steinlein U, Tronche F, Schutz G, Wagner EF. Impaired postnatal hepatocyte proliferation and liver regeneration in mice lacking c-jun in the liver. EMBO J 21: 1782–1790, 2002. 5. Borude P, Edwards G, Walesky C, Li F, Ma X, Kong B, Guo GL, Apte U. Hepatocyte-specific deletion of farnesoid X receptor delays but does
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
G902
6. 7. 8. 9.
10. 11. 12.
13.
14.
15.
16.
17. 18. 19.
20.
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
not inhibit liver regeneration after partial hepatectomy in mice. Hepatology 56: 2344 –2352, 2012. Chiang JYL. Regulation of bile acid synthesis: pathways, nuclear receptors, and mechanisms (Abstract). J Hepatol 40: 539, 2004. Cressman DE, Diamond RH, Taub R. Rapid activation of the Stat3 transcription complex in liver regeneration. Hepatology 21: 1443–1449, 1995. FitzGerald MJ, Webber EM, Donovan JR, Fausto N. Rapid DNA binding by nuclear factor kappa B in hepatocytes at the start of liver regeneration. Cell Growth Differ 6: 417–427, 1995. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, Galardi C, Wilson JG, Lewis MC, Roth ME, Maloney PR, Willson TM, Kliewer SA. A regulatory cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol Cell 6: 517–526, 2000. Huang J, Barr E, Rudnick DA. Characterization of the regulation and function of zinc-dependent histone deacetylases during rodent liver regeneration. Hepatology 57: 1742–1751, 2013. Huang W, Ma K, Zhang J, Qatanani M, Cuvillier J, Liu J, Dong B, Huang X, Moore DD. Nuclear receptor-dependent bile acid signaling is required for normal liver regeneration. Science 312: 233–236, 2006. Inagaki T, Choi M, Moschetta A, Peng L, Cummins CL, McDonald JG, Luo G, Jones SA, Goodwin B, Richardson JA, Gerard RD, Repa JJ, Mangelsdorf DJ, Kliewer SA. Fibroblast growth factor 15 functions as an enterohepatic signal to regulate bile acid homeostasis. Cell Metab 2: 217–225, 2005. Kim I, Ahn SH, Inagaki T, Choi M, Ito S, Guo GL, Kliewer SA, Gonzalez FJ. Differential regulation of bile acid homeostasis by the farnesoid X receptor in liver and intestine. J Lipid Res 48: 2664 –2672, 2007. Kir S, Beddow SA, Samuel VT, Miller P, Previs SF, Suino-Powell K, Xu HE, Shulman GI, Kliewer SA, Mangelsdorf DJ. FGF19 as a postprandial, insulin-independent activator of hepatic protein and glycogen synthesis. Science 331: 1621–1624, 2011. Kong B, Wang L, Chiang JY, Zhang Y, Klaassen CD, Guo GL. Mechanism of tissue-specific farnesoid X receptor in suppressing the expression of genes in bile-acid synthesis in mice. Hepatology 56: 1034 – 1043, 2012. Kurosu H, Choi M, Ogawa Y, Dickson AS, Goetz R, Eliseenkova AV, Mohammadi M, Rosenblatt KP, Kliewer SA, Kuro-o M. Tissuespecific expression of Klotho and fibroblast growth factor (FGF) receptor isoforms determines metabolic activity of FGF19 and FGF21. J Biol Chem 282: 26687–26695, 2007. Lee GH. Genetic dissection of murine susceptibilities to liver and lung tumors based on the two-stage concept of carcinogenesis. Pathol Int 48: 925–933, 1998. Li W, Liang X, Kellendonk C, Poli V, Taub R. STAT3 contributes to the mitogenic response of hepatocytes during liver regeneration. J Biol Chem 277: 28411–28417, 2002. Liao Y, Shikapwashya ON, Shteyer E, Dieckgraefe BK, Hruz PW, Rudnick DA. Delayed hepatocellular mitotic progression and impaired liver regeneration in early growth response-1-deficient mice. J Biol Chem 279: 43107–43116, 2004. Maran RRM, Thomas A, Roth M, Sheng Z, Esterly N, Pinson D, Gao X, Zhang Y, Ganapathy V, Gonzalez FJ, Guo GL. Farnesoid X receptor
21. 22. 23.
24.
25.
26.
27. 28.
29.
30.
31.
32.
33.
34.
35.
deficiency in mice leads to increased intestinal epithelial cell proliferation and tumor development. J Pharmacol Exp Ther 328: 469 –477, 2009. Michalopoulos GK. Liver regeneration after partial hepatectomy: critical analysis of mechanistic dilemmas. Am J Pathol 176: 2–13, 2010. Montagutelli X. Effect of the genetic background on the phenotype of mouse mutations. J Am Soc Nephrol 11, Suppl 16: S101–S105, 2000. Nicholes K, Guillet S, Tomlinson E, Hillan K, Wright B, Frantz GD, Pham TA, Dillard-Telm L, Tsai SP, Stephan JP, Stinson J, Stewart T, French DM. A mouse model of hepatocellular carcinoma: ectopic expression of fibroblast growth factor 19 in skeletal muscle of transgenic mice. Am J Pathol 160: 2295–2307, 2002. Pai R, Dunlap D, Qing J, Mohtashemi I, Hotzel K, French DM. Inhibition of fibroblast growth factor 19 reduces tumor growth by modulating -catenin signaling. Cancer Res 68: 5086 –5095, 2008. Song KH, Li T, Owsley E, Strom S, Chiang JY. Bile acids activate fibroblast growth factor 19 signaling in human hepatocytes to inhibit cholesterol 7alpha-hydroxylase gene expression. Hepatology 49: 297–305, 2009. Stepniak E, Ricci R, Eferl R, Sumara G, Sumara I, Rath M, Hui L, Wagner EF. c-Jun/AP-1 controls liver regeneration by repressing p53/p21 and p38 MAPK activity. Genes Dev 20: 2306 –2314, 2006. Taub R. Liver regeneration: from myth to mechanism. Nat Rev Mol Cell Biol 5: 836 –847, 2004. Thomas A, Hart SN, Kong B, Fang J, Zhong X, Guo GL. Genome-wide tissue specific farnesoid X receptor binding in mouse liver and intestine.. Hepatology 51: 1410 –1419, 2010. Uriarte I, Fernandez-Barrena MG, Monte MJ, Latasa MU, Chang HC, Carotti S, Vespasiani-Gentilucci U, Morini S, Vicente E, Concepcion AR, Medina JF, Marin JJ, Berasain C, Prieto J, Avila MA. Identification of fibroblast growth factor 15 as a novel mediator of liver regeneration and its application in the prevention of post-resection liver failure in mice. Gut 62: 899 –910, 2013. Wang H, Lafdil F, Kong X, Gao B. Signal transducer and activator of transcription 3 in liver diseases: a novel therapeutic target. Int J Biol Sci 7: 536 –550, 2011. Wright TJ, Ladher R, McWhirter J, Murre C, Schoenwolf GC, Mansour SL. Mouse FGF15 is the ortholog of human and chick FGF19, but is not uniquely required for otic induction. Dev Biol 269: 264 –275, 2004. Wu AL, Coulter S, Liddle C, Wong A, Eastham-Anderson J, French DM, Peterson AS, Sonoda J. FGF19 regulates cell proliferation, glucose and bile acid metabolism via FGFR4-dependent and independent pathways. PloS one 6: e17868, 2011. Yamada Y, Kirillova I, Peschon JJ, Fausto N. Initiation of liver growth by tumor necrosis factor: deficient liver regeneration in mice lacking type I tumor necrosis factor receptor. Proc Natl Acad Sci USA 94: 1441–1446, 1997. Yu C, Wang F, Kan M, Jin C, Jones RB, Weinstein M, Deng CX, McKeehan WL. Elevated cholesterol metabolism and bile acid synthesis in mice lacking membrane tyrosine kinase receptor FGFR4. J Biol Chem 275: 15482–15489, 2000. Zhu Y, Li F, Guo GL. Tissue-specific function of farnesoid X receptor in liver and intestine. Pharmacol Res 63: 259 –265, 2011.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
Fibroblast growth factor 15 deficiency impairs liver regeneration in mice Bo Kong,1* Jiansheng Huang,2* Yan Zhu,3 Guodong Li,4 Jessica Williams,6 Steven Shen,6 Lauren M. Aleksunes,1 Jason R. Richardson,5 Udayan Apte,6 David A. Rudnick,2 and Grace L. Guo1 1
Department of Pharmacology and Toxicology, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, New Jersey; 2Department of Pediatrics, Washington University School of Medicine, St. Louis, Missouri; 3 Department of General Surgery, Xuanwu Hospital, Capital Medical University, Beijing, Peoples Republic of China; 4 Department of Surgical Oncology, Cancer Treatment Center, Fourth Affiliated Hospital of Harbin Medical University, Harbin, Peoples Republic of China; 5Department of Environmental and Occupational Medicine, Rutgers Robert Wood Johnson Medical School, Rutgers, The State University of New Jersey, New Brunswick, New Jersey; and 6Department of Pharmacology, Toxicology, and Therapeutics, University of Kansas Medical Center, Kansas City, Kansas Submitted 7 October 2013; accepted in final form 30 March 2014
Kong B, Huang J, Zhu Y, Li G, Williams J, Shen S, Aleksunes LM, Richardson JR, Apte U, Rudnick DA, Guo GL. Fibroblast growth factor 15 deficiency impairs liver regeneration in mice. Am J Physiol Gastrointest Liver Physiol 306: G893–G902, 2014. First published April 3, 2014; doi:10.1152/ajpgi.00337.2013.—Fibroblast growth factor (FGF) 15 (human homolog, FGF19) is an endocrine FGF highly expressed in the small intestine of mice. Emerging evidence suggests that FGF15 is critical for regulating hepatic functions; however, the role of FGF15 in liver regeneration is unclear. This study assessed whether liver regeneration is altered in FGF15 knockout (KO) mice following 2/3 partial hepatectomy (PHx). The results showed that FGF15 KO mice had marked mortality, with the survival rate influenced by genetic background. Compared with wildtype mice, the KO mice displayed extensive liver necrosis and marked elevation of serum bile acids and bilirubin. Furthermore, hepatocyte proliferation was reduced in the KO mice because of impaired cell cycle progression. After PHx, the KO mice had weaker activation of signaling pathways that are important for liver regeneration, including signal transducer and activator of transcription 3, nuclear factor-B, and mitogen-activated protein kinase. Examination of the KO mice at early time points after PHx revealed a reduced and/or delayed induction of immediate-early response genes, including growth-control transcription factors that are critical for liver regeneration. In conclusion, the results suggest that FGF15 deficiency severely impairs liver regeneration in mice after PHx. The underlying mechanism is likely the result of disrupted bile acid homeostasis and impaired priming of hepatocyte proliferation. bile acids; farnesoid X receptor; partial hepatectomy THE LIVER HAS THE ABILITY to fully regenerate itself following partial resection or toxin-induced damage through a highly regulated process. Two-thirds partial hepatectomy (PHx) in rodents is the most common and established model to identify regulators of liver regeneration. Many factors have been shown to be critical to liver regrowth (21). Following PHx, a series of signals and pathways orchestrate the initiation, proliferation, and termination of cell cycle and cell proliferation, and liver mass is restored to the original size within a short period of time (21, 27). However, the exact mechanisms for the initiation and termination of proliferation remain unclear.
* B. Kong and J. Huang contributed equally to this work. Address for reprint requests and other correspondence: G. L. Guo, Dept. of Pharmacology and Toxicology, Ernest Mario School of Pharmacy, Rutgers, The State Univ. of New Jersey, Piscataway, NJ 08854 (e-mail: guo@eohsi. rutgers.edu). http://www.ajpgi.org
Bile acids have emerged as critical mediators of liver regeneration (11) and serve as signaling molecules and endogenous ligands for the farnesoid X receptor (FXR) (35). Bile acids promote liver regeneration, and whole body FXR deficiency enhances mortality and delays liver regeneration following PHx or chemical-induced liver injury in mice (11). FXR is highly expressed in the liver and intestine, and activation of FXR in both organs maintains bile acid homeostasis in a coordinated fashion (13, 15). Selective deletion of FXR in the livers of mice does not completely block, but delays, liver regeneration (5), suggesting that FXR signaling in the intestines and/or other organs may be critical in regulating liver regeneration. In the intestine, activation of FXR induces fibroblast growth factor 15 (FGF15), the mouse homolog of human FGF19. FGF15 is one of the endocrine FGFs, since it is highly expressed in the small intestine, but not in the liver. FGF15 travels to the liver, where it regulates a variety of hepatic functions. The best-known role of FGF15 in the liver is to suppress bile acid synthesis (12). Specifically, FGF15 accomplishes this suppression through activation of the fibroblast growth factor receptor 4 (FGFR4), a tyrosine kinase receptor expressed on the hepatocyte membrane that subsequently leads to the activation of downstream mitogen-activated protein kinase (MAPK) signaling cascades, including ERK1/2 and JNK1/2, and ultimately results in the suppression of the expression of bile-acid synthetic genes (15, 25). Emerging evidence suggests that FGF15/19 has dual functions. Indeed, besides regulating lipid and energy homeostasis, FGF19 promotes cell proliferation in vitro, and FGF19 transgenic mice develop liver tumors (23, 24). However, the functions and mechanisms of FGF15/19 in liver regeneration after PHx are not well defined. In the current study, we determined whether liver regeneration following PHx is altered by deficiency in FGF15 signaling in mice. MATERIALS AND METHODS
Materials. All chemicals were obtained from Sigma (St. Louis, MO). Antibodies were purchase from Cell Signaling (Danvers, MA). IL-6 Elisa kit was obtained from Abcam (Cambridge, MA). Animals. FGF15 knockout (KO) mice were provided by Dr. Curtis Klaassen at the University of Kansas Medical Center. Under pure C57BL/6J genetic background, FGF15 KO mice are mostly embryonic lethal (31), but they are viable and fertile under the 129SvJ genetic background. In detail, the heterozygous FGF15 KO mice
0193-1857/14 Copyright © 2014 the American Physiological Society
G893
G894
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
Table 1. Primers used for real-time quantitative PCR Gene
Forward Primer (5=-3=)
Reverse Primer (5=-3=)
Bsep Ccnd1 Ccne1 Cebp- c-Fos c-Jun c-Myc Cyp7a1 Egfr Fxr Gapdh Hgf IL-1 IL-6 NF-b Ntcp Ost Shp Socs3 Survivin Tgf-␣ Tgf-1 Tnf␣
TGAATGGACTGTCGGTATCTGTG GAAGGAGACCATTCCCTTGA TGTTACAGATGGCGCTTGCTC CGCCTTTAGACCCATGGAAG AATGGTGAAGACCGTGTCAGG TCTTCATTTTCTCACCAACTGCTT ACTTACAATCTGCGAGCCAGGACA AACAACCTGCCAGTACTAGATAGC TCAGCAACAACCCCATCCTC TCCGGACATTCAACCATCAC TGTGTCCGTCGTGGATCTGA GGTGTATCAGGAACAGGGGC ACTCAACTGTGAAATGCCACCTT CAACGATGATGCACTTGCAGA TGGCAGCTCTTCTCAAAGCA GGCCACAGACACTGCGCT GTATTTTCGTGCAGAAGATGCG CGATCCTCTTCAACCCAGATG GGAGATTTCGCTTCGGGACT CACCCCAGAGCGAATGGCGG TCCTGTTAGCTGTGTGCCAG CAACTTCTGTCTGGGACCCT ATGGCCTCCCTCTCATCAGT
CCACTGCTCCCAACGAATG GTTCACCAGAAGCAGTTCCA TTCAGCCAGGACACAATGGTC CCCGTAGGCCAGGCAGT CCCTTCGGATTCTCCGTTTCT CTCTCCAAATGCTCCCCAAA GCCCAAAGGAAATCCAGCCTTCAA GTGTAGAGTGAAGTCCTCCTTAGC GCTTGGATCACATTTGGGGC TCACTGCACATCCCAGATCTC CCTGCTTCACCACCTTCTTGAT GTCAAATTCATGGCCAAACCCT TGCTGCTGCGAGATTTGAAG GGTACTCCAGAAGACCAGAGG CCAAGAGTCGTCCAGGTCATAGA AGTGAGCCTTGATCTTGCTGAACT TTTCTGTTTGCCAGGATGCTC AGGGCTCCAAGACTTCACACA AAACTTGCTGTGGGTGACCA GTGAGGAAGGCGCAGCCAGG TGTGGGAATCTGGGCACTTG TAGTAGACGATGGGCAGTGG GCTCCTCCACTTGGTGGTTT
(50% C57BL/6J and 50% 129SvJ genetic background) were backcrossed for one generation to 129SvJ or C57BL/6J genetic background, respectively, and then the heterozygotes were intercrossed to generate homozygous KO and wild-type (WT) littermate control mice. These littermates with the same background were used to establish and expand the colony. The mouse genetic background was further validated by the Illumina Mouse MD Linkage Panel consisting of 1,440 single-nucleotide polymorphisms throughout the mouse genome. Briefly, two strains of the FGF15 KO mice on different genetic backgrounds were used in the current study, and the corresponding WT mice with the same genetic background served as controls: FGF15 KO (75% C57BL/6J and 25% 129SvJ) and FGF15 KO (A) (75% 129SvJ and 25% C57BL/6J). The mice were maintained on an ad libitum diet of rodent chow (Charles River Laboratories, Wilmington, MA) with a 12:12-h light-dark cycle. Procedures involving animals were conducted according to a protocol approved by the University of Kansas Medical Center Institutional Animal Care and Use Committee. All mice were closely monitored after surgery for general well-being, and experiments were terminated early when mice became moribund by showing significant decrease in activity, body weight, and body temperature. Two-thirds PHx. Eight-to-ten-week-old male mice (n ⫽ 4 –12 mice/group) underwent standard PHx surgery (10). FGF15 KO and FGF15 KO (A) mice were examined at various time points in separate experiments. Blood, liver, and small intestine were collected using a published method (15). Bromodeoxyuridine immunohistochemistry. One hour before livers were harvested, mice were injected intraperitoneally with 100 mg bromodeoxyuridine (BrdU)/kg body wt. BrdU staining was performed
by standard immunohistochemistry methods (20). The degree of BrdU incorporation into newly synthesized DNA was determined by counting positively stained hepatocyte nuclei in five microscopic fields per animal at ⫻400 magnification. The value was expressed as a percentage of BrdU-labeled nuclei vs. all nuclei in the fields. Serum parameter analysis. Serum activity of alanine aminotransferase (ALT), alkaline phosphatase (ALP), and bilirubin were analyzed using commercial kits (Pointe Scientific, Canton, MI). Total bile acids were measured by a kit from Bioquant (San Diego, CA). IL-6 levels were analyzed using an Elisa kit from Abcam. Quantification of hepatic mRNA expression. The mRNA levels were determined by RT-qPCR by a published method (28). Primer sequences are shown in Table 1. Western blot. Equal amounts of protein from individual mice were pooled in each group. Fifteen micrograms nuclei protein, 30 g total cell extract or cytosolic protein were load onto 10⬃15% PAGE gels. Western blot analysis was performed by standard techniques. The band’s densities from Western blot were semiquantified by using ImageJ software (http://rsb.info.nih.gov/ij/). The value indicated the relative density of that band compared with the density of the band of its loading control GAPDH or histone H3 (nuclear protein). Statistical analysis. The data are expressed as means ⫾ SE. All statistical tests used were done using Sigma Stat Statistical Software (Systat Software, Point Richmond, CA). The differences between the PHx and sham groups, or between WT and KO mice, were determined by t-test. The statistical significance between multiple groups was analyzed by one-way ANOVA followed by the Student-NewmanKeuls test. P ⬍ 0.05 is considered statistically significant.
Fig. 1. Mortality rate, liver growth, and liver injury in mice with fibroblast growth factor (FGF) 15 deficiency after partial hepatectomy (PHx). A: Kaplan-Meier analysis of the survival of wild-type (WT), FGF15 knockout (KO), and FGF15 KO (A) mice after PHx; and changes in the ratio of remnant to original liver weight during liver regeneration in WT and FGF15 KO (A) mice. Data are expressed as a percentage of the initial value before PHx. B: serum alanine aminotransferase (ALT), alkaline phosphatase (ALP), bilirubin, and bile acid levels in WT, FGF15 KO, and FGF15 KO (A) mice following PHx. C: representative hematoxylin and eosin staining (⫻400) of liver sections from WT, FGF15 KO, and FGF15 KO (A) mice after PHx. Inset, the necrotic area in FGF15 KO (A) mice. *P ⬍ 0.05, significant difference between WT and FGF15 KO mice at the same time point. FGF15 KO (75% C57BL/6J and 25% 129SvJ; PHx 0, 0.5, 3, 6, 12, 24, 36, and 48 h, n ⫽ 7, 5, 4, 4, 5, 5, 9, and 10 for WT, respectively; n ⫽ 4, 5, 4, 4, 4, 4, 12, and 10 for FGF15 KO, respectively) and FGF15 KO (A) (75% 129SvJ and 25% C57BL/6J; PHx 0, 6, 24, 36, 72, 96, 168, and 336 h, n ⫽ 3, 4, 7, 8, 6, 8, 5, and 5 for WT, respectively; n ⫽ 4, 4, 10, 10, 7, 11, 5, and 7 for FGF15 KO, respectively). AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
G895
G896
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
RESULTS
Increased liver necrosis and mortality rate of FGF15 KO mice after PHx. FGF15 KO mice on two genetic backgrounds [FGF15 KO and FGF15 KO (A)] and their corresponding WT controls were subjected to PHx. Before surgery, liver-to-body weight ratios were similar between WT and either FGF15 KO or FGF15 KO (A) mice (about 4.5%). In addition, the small intestine of WT mice on either C57BL/6J or 129SvJ background had similar baseline FGF15 mRNA levels (data not shown). After PHx, FGF15 KO mice showed significant mortality compared with WT controls, which was strain dependent. KO mice that survived the PHx also displayed a reduced recovery of liver mass. Specifically, 24 – 48 h after PHx, the mortality rate was 90% in FGF15 KO mice and 60 – 65% in FGF15 KO (A) mice, whereas only ⬃10% WT mice died (Fig. 1A). However, the surviving FGF15 KO (A) mice had reduced liver growth for the entire duration following PHx (Fig. 1A). The increased mortality in FGF15 KO mice was not likely because of basal liver injury, because before surgery, serum levels of ALT, ALP, bilirubin, and bile acids were similar between WT and FGF15 KO mice (data not shown). After PHx, these parameters were transiently and moderately increased in WT mice and returned to normal ranges within 72 h. In contrast, FGF15 KO mice showed a significant elevation of these parameters at 24 and 36 h (Fig. 1B). Furthermore, the FGF15 KO mice, which suffered more mortality, showed a marked increase in these parameters at 48 h, whereas the FGF15 KO (A) mice, which had less mortality, reduced these parameters by 72 h. Severe hepatocyte necrosis was observed in FGF15 KO mice, but not in WT mice, at 24 h after PHx (Fig. 1C), and the necrotic area further increased in the KO mice at 36 and 48 h after PHx (Fig. 1C). In FGF15 KO (A) mice, the region of necrosis was less extensive than observed in FGF15 KO mice following PHx (Fig. 1C). Impaired hepatocyte DNA replication after PHx in FGF15 KO mice. Staining of BrdU incorporated into DNA 24 and 36 h after PHx is routinely used to assess DNA replication in hepatocytes. In this study, a significant decrease in BrdU labeling was observed in FGF15 KO mice at 24 to 48 h after PHx (Fig. 2A). At 36 h, the time point of peak DNA replication, the percentage of hepatocytes stained with BrdU, was close to 40% in WT mice but was only 18% in FGF15 KO mice (Fig. 2B). A similar decrease in BrdU incorporation was observed in FGF15 KO (A) mice (data not shown). Expression of genes involved in cell-cycle progression, bile acid homeostasis, and inflammation. Because most of the FGF15 KO mice suffered mortality 24 – 48 h after PHx, the mRNA levels of genes Ccnd1 and Ccne1 that encode Cyclin D1 and Cyclin E1 were determined at 6, 12, and 24 h after PHx. Both Cyclin D1 and Cyclin E1 mRNA levels were low and similar between WT and KO mice at 6 and 12 h following PHx but were significantly upregulated at 24 h (Fig. 3, A and B), However, Ccnd1 mRNA levels were much lower in FGF15 KO and FGF15 KO (A) compared with WT mice 24 h after PHx. Interestingly, Ccne1 mRNA levels were upregulated in mice with low Ccnd1 expression level, indicating a compensatory response to the lack of Ccnd1 induction. The protein levels of several critical regulators for hepatocyte proliferation following PHx were then determined (Fig. 3C). In WT mice, Cyclin D1 was increased markedly at 36 h
Fig. 2. Impaired hepatocyte proliferation after PHx in FGF15-deficient mice. A: representative bromodeoxyuridine (BrdU) staining (⫻400) of liver sections from WT and FGF15 KO mice after PHx. B: the percentage of BrdU-positive hepatocytes in WT and FGF15 KO mice. *P ⬍ 0.05, significant difference between WT and FGF15 KO mice at the same time point.
and then decreased at 48 h. In comparison, the KO mice had much lower Cyclin D1 levels at 36 h, but the levels were increased at 48 h. Proliferating cell nuclear antigen was markedly induced in WT mice at both 36 and 48 h but was undetectable in the KO mice at any time point. Similar changes were observed for CDK2 and CDK4. Transcription factors CDC2, CDC25b, and FoxM1b are critical regulators of G2/M transition and mitosis and were upregulated at 36 and 48 h in WT mice, but their levels were much lower in FGF15 KO mice (Fig. 3D). Because FGF15 is known to suppress bile acid synthesis and the impaired liver regeneration in FGF15 KO mice is accompanied by increased serum bile acids, the mRNA levels of genes encoding proteins involved in bile acid homeostasis were measured at 0, 0.5, 3, 36, and 48 h following PHx. Additionally, levels of inflammatory cytokines involved in liver regeneration and cell-cycle regulators were determined. The mRNA levels of FXR were not changed dramatically by PHx. The mRNA levels of sodium taurocholate cotransporting polypeptide (for hepatic bile acid uptake) and bile salt efflux pump (for canalicular bile acid excretion) gradually decreased with time after PHx, but without statistically significant differences between WT and FGF15 KO mice (Fig. 3E). Cyp7a1 gene encodes the rate-limiting enzyme in bile acid synthesis, and its expression is suppressed by FGF15 (12, 15).
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
G897
Fig. 3. Expression of genes in bile acid homeostasis, inflammation, and cell cycle progression. mRNA levels of Ccnd1 (A) and Ccne1 (B) in FGF15 KO and FGF15 KO (A) mice. C: protein levels of cell cycle-related proteins after PHx. D: mRNA levels of Cdc25b and FoxM1b, and protein levels of phospho-CDC2 in cytosol and nuclei. E and F: levels of mRNA of genes involved in bile acids homeostasis (E) and inflammation (F) and PHx 0, 0.5, 3, 36, and 48 h, n ⫽ 7, 5, 4, 9, and 10 for WT, respectively; n ⫽ 4, 5, 4, 12, and 10 for FGF15 KO, respectively. *P ⬍ 0.05, significant difference between WT and FGF15 KO mice at the same time point.
In WT mice, the mRNA levels of Cyp7a1 did not change at 0.5 and 3 h but increased ⬃6- and 2-fold at 36 and 48 h following PHx (Fig. 3E). In FGF15 KO mice, the basal hepatic Cyp7a1 mRNA levels were 20-fold higher than those in WT mice, but dropped to 10-fold at 0.5 h after PHx, and returned to WT levels at 3 h. At 36 and 48 h, the levels of Cyp7a1 mRNA were almost undetectable in FGF15 KO mice. The mRNA levels of
small heterodimer partner (SHP) were induced at 3 h in FGF15 KO mice, which were fivefold higher than those in WT mice. The mRNA levels of organic solute transporter- (Ost) were robustly upregulated (⬃25-fold) in FGF15 KO mice at 3 h following PHx, and further increased at 36 and 48 h (⬃150- to 180-fold) in FGF15 KO mice. Induction of SHP and Ost indicates FXR activation in the liver.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
G898
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
The mRNA levels of the inflammatory factors, IL-1, IL-6, tumor necrosis factor (Tnf)-␣, and transforming growth factor (Tgf)-1, known to be critical for liver regeneration, were measured (Fig. 3F). The expression of IL-1, IL-6, and Tnf␣ was increased at 0.5 and 3 h post-PHx, but no difference was observed between the WT and KO mice, except for Tnf␣ mRNA, which was higher in KO mice than in WT mice at 3 h (⬃2-fold). However, at 36 and 48 h, a marked increase was observed for IL-1, IL-6, and Tnf␣ in KO mice but not in WT mice. No significant difference of Tgf-1 expression was found between WT and KO mice. Impaired activation of mitogen-activated protein kinase, signal transducer and activator of transcription 3, and NF-B signaling in FGF15 KO mice after PHx. Because FGF15 deficiency causes severe inhibition of DNA synthesis and hepatocyte proliferation, the activation status of signaling pathways involved in “survival” at early stages of liver regeneration were assessed. The MAPK pathways are important in liver regeneration because c-Jun- (4, 26) and Egr1-deficient (19) mice exhibit severe defects in liver regeneration following PHx. FGF15 is known to activate the JNK/ERK signaling pathways (15). In the FGF15 KO mice following PHx, JNK and p38 were activated to a lesser degree than in WT (Fig. 4). Interestingly, ERK activation appeared to be stronger in FGF15 KO mice (Fig. 4). Signal transducer and activator of transcription 3 is critical to several cytokine-controlled cellular processes. A cohort of cytokines and growth factors activate signal transducer and activator of transcription 3 (STAT3) proteins downstream of JAK2 and JAK1; upon stimulation, activated STAT3s dimerize and translocate into the nucleus, where they bind to target genes and induce transcription (30). STAT3 has been shown to be critical in liver regeneration (7, 18). In this study, total JAK1, JAK2, and c-Myc were lower in KO mice before and after PHx (Fig. 5A). In WT mice, STAT3 was markedly activated, as measured by Y705 phosphorylation as early as 0.5 h following PHx, and by S727 phosphorylation at 36 h (Fig. 5B). In contrast, FGF15 KO mice had much lower Y705 phosphorylation and S727 phosphorylation after PHx levels. In addition, the total STAT3 levels in nuclear extracts were decreased in the KO mice (Fig. 5B).
IL-6 is a strong STAT3 activator, and serum IL-6 was increased as early as 3 h after PHx in both WT and KO mice. However, IL-6 levels subsequently decreased in WT mice at 36 and 48 h. In contrast, serum IL-6 remained elevated in KO mice after PHx (Fig. 5C). Furthermore, the mRNA levels of STAT3 target genes were higher in WT mice at 36 h (c-Myc and survivin) and at 48 h (survivin). Socs3, which is a STAT3 inhibitor and a STAT3 target gene, was induced equally in WT and KO mice at 0.5 and 3 h. However, the degree of induction was less at 36 and 48 h in WT mice and markedly increased in KO mice (Fig. 5D). NF-B is highly activated within hours after PHx; therefore, NF-B activation has been considered a critical initiation signal in liver regeneration (8). Moreover, NF-B activation prevents apoptosis and promotes hepatocyte survival (3). In the current study, increased nuclear p65 protein was apparent in WT mice, but this was absent in the FGF15 KO mice (Fig. 6), indicating that FGF15 deficiency disrupts NF-B signaling following PHx. The other well-known survival signaling pathways, such as phosphatidylinositol 3-kinase/protein kinase B, mammalian target of rapamycin, and insulin receptor signaling pathway [insulin-like growth factor-I-binding protein, insulinlike growth factor (IGF)-I, and IGF-II], showed no significant difference between WT and FGF15 KO mice after PHx (data not shown). Impaired gene expression in the early stage of liver regeneration in FGF15 KO mice after PHx. MAPK, STAT3, and NF-B activation are crucial for the priming phase of liver regeneration and may be responsible for the early response following PHx (1, 2, 33). The results shown above suggested that the FGF15 mice were deficient in initiating liver regeneration after PHx. c-Jun and c-Fos are components of activator protein-1 (AP-1), which is important for initiation of regeneration. The FGF15 KO mice showed a delay in induction of the mRNAs of c-Jun and c-Fos. At 10 min after PHx, the induction of c-Jun and c-Fos was markedly higher in WT mice than that in FGF15 KO mice. Finally, at 30 min, FGF15 KO mice had equal (c-Jun) or higher (c-Fos) expression than the WT mice had (Fig. 7A). A number of additional growth factors and transcription factors, including hepatocyte growth factor (HGF), epidermal growth factor receptor (EGFR), TGF-␣, IL-6, NF-B, and
Fig. 4. Disturbed mitogen-activated protein kinase (MAPK) signaling pathways after PHx in FGF15 KO mice. The mammalian MAPK family includes JNK, ERK, and p38. Signaling cascades induce phosphorylation/ activation of the MAPK signaling pathway and play a central role in the regulation of genes involved in hepatocyte proliferation and in protecting against apoptotic cell death.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
G899
Fig. 5. Impaired activation of signal transducer and activator of transcription (STAT) 3 in FGF15 KO mice after PHx. A: total JAK1/2 and c-Myc protein level. B: total and phospho-proteins associated with STAT3 signaling pathways. C: serum IL-6 concentration in WT and FGF15 KO mice after PHx. D: gene expression of Survivin and Socs3. PHx 0, 0.5, 3, 36, and 48 h, n ⫽ 7, 5, 4, 9, and 10 for WT, respectively; n ⫽ 4, 5, 4, 12, and 10 for FGF15 KO, respectively; *P ⬍ 0.05 between WT and FGF15 KO mice at the same time point.
CCAAT enhancer-binding protein- (CEBP-), are involved in initiating the hepatic regenerative process. The mRNA levels of Hgf and Egfr at 10 min and NF-B and Tgf-␣ at 30 min after PHx were higher in WT mice than in FGF15KO mice, but mRNA levels of other genes examined did not show significant differences across genotypes (Fig. 7B).
DISCUSSION
FGF15 is the mouse homolog of human FGF19 and highly expressed in the small intestine. Emerging evidence suggests that FGF15/19 is an endocrine FGF and is critical for a variety of liver functions, including suppressing bile acid synthesis,
Fig. 6. NF-B signaling deactivated after PHx. Activation of the NF-B signaling pathway is associated with nuclear translocation of the p65 component of the complex. The translocation of the cytosolic p65 to the nucleus decreased in FGF15 KO mice after PHx.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
G900
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
Fig. 7. FGF15 deficiency affected the expression of immediate-early response genes (A) and growth factors and transcription factors (B) after PHx. WT and FGF15 KO mice underwent Sham or PHx surgery (n ⫽ 5/group at each time point). Liver tissue samples were collected at 10 and 30 min after surgery. mRNA levels were determined. *P ⬍ 0.05 between WT and FGF15 KO mice at the same time point.
promoting protein synthesis, and improving insulin sensitivity (12, 14, 15, 32). However, the role of FGF15 in liver regeneration is not fully defined. The current study demonstrates that mice deficient in FGF15 have severely impaired liver regeneration following PHx. The FGF15 KO mice exhibited severe liver necrosis, elevated levels of bile acids, and high mortality. Markers indicating successful hepatocyte proliferation were diminished in FGF15 KO mice after PHx. Furthermore, activation of several transcription factors critical for liver regeneration, including NF-B and STAT3, was abolished at early time points in FGF15 KO mice following PHx. This study clearly demonstrates that FGF15 deficiency causes acute liver failure in mice after PHx, and the KO mice failed to regenerate their livers. Our results are consistent with the recently published study of Uriarte and coworkers (29). In that study, increased bile acids and reduced FGF15 both contributed to liver failure following PHx, as evidenced by finding that both administration of cholestyramine to reduce bile acid levels and exogenous Fgf15 protein administration reduced liver necrosis and mortality and promoted liver regeneration in the FGF15 KO mice after PHx. In the current study, we confirmed and extended those findings by demonstrating that mouse genetic background is crucial for the severity of liver failure in FGF15 KO mice after PHx. Under the C57BL/6J background, FGF15 KO mice are mostly embryonic lethal, since ⬍2% of the homozygous mice can survive (31), but under 129SvJ background, FGF15 KO mice are viable and fertile (unpublished data). Our results showed that, under 75% C57BL/6J and 25% 129SvJ background, FGF15 KO suffered more liver damage and significantly higher mortality rate compared with KO mice under 25% C57BL/6J and 75%
Fig. 8. Schematic representation of effects on bile acid homeostasis and liver regeneration after FGF15 deletion. Activation of FXR in the intestine induces FGF15 protein, which could repress Cyp7a1 gene transcription by the activation of JNK/ERK pathways in the liver through its receptor FGFR4. Lack of FGF15 feedback suppression after FGF15 deletion leads to increased bile acid biosynthesis, and accumulation of bile acids would cause toxicity and damage to hepatocytes. In addition to feedback regulation of Cyp7a1 transcription, reduced JNK/ERK activation could deactivate STAT3/NF-B signaling pathways, which are crucial for the priming phase of liver regeneration.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
129SvJ background. Uriarte et al. (29) also used the KO mice on a mixed C57BL/6J/129/SvJ background, but the percentage of each background was not reported. The role of FGF15 in liver regeneration after PHx is not without controversy, since mice deficient in FGFR4, which is believed to be the major receptor of FGF15 in the liver, showed only slight delay in liver regeneration after PHx (34). Because the FGFR4 KO mice used for PHx were on 129Sv background, and these mice only showed slight delay in liver regeneration, it is likely that the difference between FGF15 KO mice and FGFR4 KO mice in response to PHx is partly due to the genetic background. Indeed, strain-dependent effects on liver proliferation and carcinogenesis have been reported (17, 22). However, it cannot be excluded that FGF15 may activate other receptors in the FGFR4 KO mice, since FGF19 was reported to activate not only FGFR4 receptor but also several other FGFRs that are expressed in the liver (15, 16). The mechanism of the strain-dependent difference of the FGF15 pathway is not known. However, our initial analysis indicates that the mRNA levels of FGF15 in mice with pure C57BL/6J or 129SvJ background do not differ, suggesting that levels of expressed FGF15 mRNA do not explain the impact of genetic background on severity of liver injury following PHx in the FGF15 KO. Variation in certain genetic loci(s) between these two stains might be responsible for the differential response to FGF15 deletion after PHx, so future investigation of the genetic loci(s) that is responsible for the strain-specific differences may help to determine novel pathways in regulating liver growth. FGF15 is a major suppressor for bile acid synthesis, and high bile acids were persistently present in KO mice following PHx, whereas the WT mice only showed a modest and transient increase in serum bile acids following PHx. The basal Cyp7a1 expression was markedly higher in the KO mice (20⫻), but this was sharply suppressed within 3 h following PHx. The other mechanisms would include the induction of SHP and inflammation in the KO mice starting at 3 h post-PHx because SHP and inflammation are known to suppress Cyp7a1 gene expression (6, 9). However, the acute suppression of Cyp7a1 did not reduce bile acid levels. Furthermore, a dramatic induction of Ost mRNA in the KO mice following PHx indicates enhanced bile acid burden, and the increase in bile acids is likely a mechanism responsible for more extensive liver necrosis and mortality in FGF15 KO mice after PHx. We did not define the effects of reducing bile acids on liver regeneration in FGF15 KO mice. However, recent studies reported that reducing the bile acid pool decreased mortality after PHx in FGF15 KO or FXR KO mice (11, 29). Most FGF15 KO mice died within 36 – 48 h after PHx, and these mice had increased bile acid levels. In the mice that survived, there was no evidence of normal liver regeneration after PHx as evaluated by commonly used markers of regeneration (1–3). The activation of immediate-early response genes is believed to be driven by inflammation and, in particular, increased TNF-␣ and IL-6, classical activators of NF-B and STAT3, respectively. Although we did not observe a reduction of the expression of these cytokines in the KO mouse liver following PHx at early time points (0, 0.5, and 3 h), activation of both NF-B and STAT3 was much lower in the KO mice. AP-1 expression was also decreased, and several growth factors and transcription factors were reduced at 10 min
G901
after PHx in FGF15 KO mice, indicating that deletion of FGF15 may have affected the initiation of liver regeneration, which could result in impaired activation of genes required for cell survival. However, it is unclear how FGF15 facilitates the activation of hepatic NF-B and STAT3 signaling pathways. One possibility is that FGF15 binds and activates its receptor FGFR4, which, through a tyrosine phosphorylation cascade, activates STAT3 and NF-B signaling pathways. The results reported in the current study further confirm that intestinal FGF15 could promote liver regeneration in mice, and FGF15 gene deletion in mice markedly increases mortality after PHx (29). Moreover, we have shown that mouse genetic background affects FGF15-regulated liver regeneration. More importantly, we showed that, after PHx, the FGF15 KO mice had impaired signaling pathways that are critical in liver regeneration, including STAT3, NF-B, and MAPK. In summary, the current study suggests two roles that FGF15 may play in promoting liver regeneration (Fig. 8): 1) FGF15 deletion disturbs bile acid homeostasis, and excessive bile acid accumulation in the regenerating liver can cause hepatocyte damage and disrupt hepatic regeneration and 2) deletion of FGF15 may lead to a defect in the initiation of liver regeneration, which might contribute to liver failure in the long term. These two roles might interact with each other, and future studies will be needed to clarify the underlying mechanism(s). ACKNOWLEDGMENTS We greatly appreciate the suggestions from Dr. George K. Michalopoulos at the University of Pittsburgh. GRANTS This study was supported by the National Institutes of Health fund [DK081343, DK-090036, and GM-104037 (G. L. Guo), as well as ES-005022]. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the authors. AUTHOR CONTRIBUTIONS Author contributions: B.K., J.H., and G.L.G. conception and design of research; B.K., J.H., Y.Z., G.L., J.A.W., S.S., and G.L.G. performed experiments; B.K., J.H., and G.L.G. analyzed data; B.K., J.H., L.M.A., J.R.R., U.A., D.A.R., and G.L.G. interpreted results of experiments; B.K., J.H., and G.L.G. prepared figures; B.K., J.H., and G.L.G. drafted manuscript; B.K., J.H., Y.Z., G.L., J.A.W., S.S., L.M.A., J.R.R., U.A., D.A.R., and G.L.G. edited and revised manuscript; B.K., J.H., Y.Z., G.L., J.A.W., S.S., L.M.A., J.R.R., U.A., D.A.R., and G.L.G. approved final version of manuscript. REFERENCES 1. Akerman P, Cote P, Yang SQ, McClain C, Nelson S, Bagby GJ, Diehl AM. Antibodies to tumor necrosis factor-alpha inhibit liver regeneration after partial hepatectomy. Am J Physiol Gastrointest Liver Physiol 263: G579 –G585, 1992. 2. Aldeguer X, Debonera F, Shaked A, Krasinkas AM, Gelman AE, Que X, Zamir GA, Hiroyasu S, Kovalovich KK, Taub R, Olthoff KM. Interleukin-6 from intrahepatic cells of bone marrow origin is required for normal murine liver regeneration. Hepatology 35: 40 –48, 2002. 3. Beg AA, Sha WC, Bronson RT, Ghosh S, Baltimore D. Embryonic lethality and liver degeneration in mice lacking the RelA component of NF-kappa B. Nature 376: 167–170, 1995. 4. Behrens A, Sibilia M, David JP, Mohle-Steinlein U, Tronche F, Schutz G, Wagner EF. Impaired postnatal hepatocyte proliferation and liver regeneration in mice lacking c-jun in the liver. EMBO J 21: 1782–1790, 2002. 5. Borude P, Edwards G, Walesky C, Li F, Ma X, Kong B, Guo GL, Apte U. Hepatocyte-specific deletion of farnesoid X receptor delays but does
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
G902
6. 7. 8. 9.
10. 11. 12.
13.
14.
15.
16.
17. 18. 19.
20.
INTESTINAL FGF15 PROMOTE LIVER REGENERATION
not inhibit liver regeneration after partial hepatectomy in mice. Hepatology 56: 2344 –2352, 2012. Chiang JYL. Regulation of bile acid synthesis: pathways, nuclear receptors, and mechanisms (Abstract). J Hepatol 40: 539, 2004. Cressman DE, Diamond RH, Taub R. Rapid activation of the Stat3 transcription complex in liver regeneration. Hepatology 21: 1443–1449, 1995. FitzGerald MJ, Webber EM, Donovan JR, Fausto N. Rapid DNA binding by nuclear factor kappa B in hepatocytes at the start of liver regeneration. Cell Growth Differ 6: 417–427, 1995. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, Galardi C, Wilson JG, Lewis MC, Roth ME, Maloney PR, Willson TM, Kliewer SA. A regulatory cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol Cell 6: 517–526, 2000. Huang J, Barr E, Rudnick DA. Characterization of the regulation and function of zinc-dependent histone deacetylases during rodent liver regeneration. Hepatology 57: 1742–1751, 2013. Huang W, Ma K, Zhang J, Qatanani M, Cuvillier J, Liu J, Dong B, Huang X, Moore DD. Nuclear receptor-dependent bile acid signaling is required for normal liver regeneration. Science 312: 233–236, 2006. Inagaki T, Choi M, Moschetta A, Peng L, Cummins CL, McDonald JG, Luo G, Jones SA, Goodwin B, Richardson JA, Gerard RD, Repa JJ, Mangelsdorf DJ, Kliewer SA. Fibroblast growth factor 15 functions as an enterohepatic signal to regulate bile acid homeostasis. Cell Metab 2: 217–225, 2005. Kim I, Ahn SH, Inagaki T, Choi M, Ito S, Guo GL, Kliewer SA, Gonzalez FJ. Differential regulation of bile acid homeostasis by the farnesoid X receptor in liver and intestine. J Lipid Res 48: 2664 –2672, 2007. Kir S, Beddow SA, Samuel VT, Miller P, Previs SF, Suino-Powell K, Xu HE, Shulman GI, Kliewer SA, Mangelsdorf DJ. FGF19 as a postprandial, insulin-independent activator of hepatic protein and glycogen synthesis. Science 331: 1621–1624, 2011. Kong B, Wang L, Chiang JY, Zhang Y, Klaassen CD, Guo GL. Mechanism of tissue-specific farnesoid X receptor in suppressing the expression of genes in bile-acid synthesis in mice. Hepatology 56: 1034 – 1043, 2012. Kurosu H, Choi M, Ogawa Y, Dickson AS, Goetz R, Eliseenkova AV, Mohammadi M, Rosenblatt KP, Kliewer SA, Kuro-o M. Tissuespecific expression of Klotho and fibroblast growth factor (FGF) receptor isoforms determines metabolic activity of FGF19 and FGF21. J Biol Chem 282: 26687–26695, 2007. Lee GH. Genetic dissection of murine susceptibilities to liver and lung tumors based on the two-stage concept of carcinogenesis. Pathol Int 48: 925–933, 1998. Li W, Liang X, Kellendonk C, Poli V, Taub R. STAT3 contributes to the mitogenic response of hepatocytes during liver regeneration. J Biol Chem 277: 28411–28417, 2002. Liao Y, Shikapwashya ON, Shteyer E, Dieckgraefe BK, Hruz PW, Rudnick DA. Delayed hepatocellular mitotic progression and impaired liver regeneration in early growth response-1-deficient mice. J Biol Chem 279: 43107–43116, 2004. Maran RRM, Thomas A, Roth M, Sheng Z, Esterly N, Pinson D, Gao X, Zhang Y, Ganapathy V, Gonzalez FJ, Guo GL. Farnesoid X receptor
21. 22. 23.
24.
25.
26.
27. 28.
29.
30.
31.
32.
33.
34.
35.
deficiency in mice leads to increased intestinal epithelial cell proliferation and tumor development. J Pharmacol Exp Ther 328: 469 –477, 2009. Michalopoulos GK. Liver regeneration after partial hepatectomy: critical analysis of mechanistic dilemmas. Am J Pathol 176: 2–13, 2010. Montagutelli X. Effect of the genetic background on the phenotype of mouse mutations. J Am Soc Nephrol 11, Suppl 16: S101–S105, 2000. Nicholes K, Guillet S, Tomlinson E, Hillan K, Wright B, Frantz GD, Pham TA, Dillard-Telm L, Tsai SP, Stephan JP, Stinson J, Stewart T, French DM. A mouse model of hepatocellular carcinoma: ectopic expression of fibroblast growth factor 19 in skeletal muscle of transgenic mice. Am J Pathol 160: 2295–2307, 2002. Pai R, Dunlap D, Qing J, Mohtashemi I, Hotzel K, French DM. Inhibition of fibroblast growth factor 19 reduces tumor growth by modulating -catenin signaling. Cancer Res 68: 5086 –5095, 2008. Song KH, Li T, Owsley E, Strom S, Chiang JY. Bile acids activate fibroblast growth factor 19 signaling in human hepatocytes to inhibit cholesterol 7alpha-hydroxylase gene expression. Hepatology 49: 297–305, 2009. Stepniak E, Ricci R, Eferl R, Sumara G, Sumara I, Rath M, Hui L, Wagner EF. c-Jun/AP-1 controls liver regeneration by repressing p53/p21 and p38 MAPK activity. Genes Dev 20: 2306 –2314, 2006. Taub R. Liver regeneration: from myth to mechanism. Nat Rev Mol Cell Biol 5: 836 –847, 2004. Thomas A, Hart SN, Kong B, Fang J, Zhong X, Guo GL. Genome-wide tissue specific farnesoid X receptor binding in mouse liver and intestine.. Hepatology 51: 1410 –1419, 2010. Uriarte I, Fernandez-Barrena MG, Monte MJ, Latasa MU, Chang HC, Carotti S, Vespasiani-Gentilucci U, Morini S, Vicente E, Concepcion AR, Medina JF, Marin JJ, Berasain C, Prieto J, Avila MA. Identification of fibroblast growth factor 15 as a novel mediator of liver regeneration and its application in the prevention of post-resection liver failure in mice. Gut 62: 899 –910, 2013. Wang H, Lafdil F, Kong X, Gao B. Signal transducer and activator of transcription 3 in liver diseases: a novel therapeutic target. Int J Biol Sci 7: 536 –550, 2011. Wright TJ, Ladher R, McWhirter J, Murre C, Schoenwolf GC, Mansour SL. Mouse FGF15 is the ortholog of human and chick FGF19, but is not uniquely required for otic induction. Dev Biol 269: 264 –275, 2004. Wu AL, Coulter S, Liddle C, Wong A, Eastham-Anderson J, French DM, Peterson AS, Sonoda J. FGF19 regulates cell proliferation, glucose and bile acid metabolism via FGFR4-dependent and independent pathways. PloS one 6: e17868, 2011. Yamada Y, Kirillova I, Peschon JJ, Fausto N. Initiation of liver growth by tumor necrosis factor: deficient liver regeneration in mice lacking type I tumor necrosis factor receptor. Proc Natl Acad Sci USA 94: 1441–1446, 1997. Yu C, Wang F, Kan M, Jin C, Jones RB, Weinstein M, Deng CX, McKeehan WL. Elevated cholesterol metabolism and bile acid synthesis in mice lacking membrane tyrosine kinase receptor FGFR4. J Biol Chem 275: 15482–15489, 2000. Zhu Y, Li F, Guo GL. Tissue-specific function of farnesoid X receptor in liver and intestine. Pharmacol Res 63: 259 –265, 2011.
AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00337.2013 • www.ajpgi.org
