ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate...

10
Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone Deacetylase 3 Huating Li, 1,2 Zhanguo Gao, 3 Jin Zhang, 3 Xin Ye, 3 Aimin Xu, 4,5 Jianping Ye, 3 and Weiping Jia 1 Fibroblast growth factor 21 (FGF21) stimulates fatty acid oxidation and ketone body production in animals. In this study, we investigated the role of FGF21 in the metabolic activity of sodium butyrate, a dietary histone deacetylase (HDAC) inhibitor. FGF21 expression was examined in serum and liver after injection of sodium butyrate into dietary obese C57BL/6J mice. The role of FGF21 was determined using antibody neutralization or knockout mice. FGF21 transcription was investigated in liver and HepG2 hepatocytes. Trichostatin A (TSA) was used in the control as an HDAC inhibitor. Butyrate was compared with bezabrate and fenobrate in the induction of FGF21 expression. Butyrate induced FGF21 in the serum, enhanced fatty acid oxidation in mice, and stimulated ketone body production in liver. The butyrate activity was signicantly reduced by the FGF21 antibody or gene knockout. Butyrate induced FGF21 gene expression in liver and hepatocytes by inhibiting HDAC3, which suppresses peroxisome proliferatoractivated receptor-a func- tion. Butyrate enhanced bezabrate activity in the induction of FGF21. TSA exhibited a similar set of activities to butyrate. FGF21 mediates the butyrate activity to increase fatty acid use and ketogenesis. Butyrate induces FGF21 transcription by inhibi- tion of HDAC3. Diabetes 61:797806, 2012 T he broblast growth factor (FGF) superfamily contains at least 22 members with diverse bi- ological functions in the control of cell growth and development and wound healing (1). FGF21, a polypeptide with 210 amino acid residues, is abundantly expressed in the liver, although its expression is also reported in pancreata, adipose, and muscle (2). FGF21 plays an important role in the regulation of lipid metabo- lism (3). It promotes lipid oxidation, triglyceride clearance, and ketogenesis in liver (4). FGF21 knockout (FGF21 KO) mice exhibit deciency in ketogenesis and loss response to ketogenic diet, develop hepatic steatosis, and gain weight (4,5). FGF21 administration increased energy expenditure, decreased blood lipids, and reduced hepatic steatosis in dietary obese mice (6). Infusion of recombinant FGF21 also leads to glucose reduction in genetic and dietary obese mice (3,6). The physiological role of FGF21 remains to be investigated in humans. Several recent studies show that serum FGF21 levels are elevated in patients of meta- bolic syndrome (710). We reported that serum FGF21 was positively associated with the degree of nonalcoholic fatty liver disease in humans (11). FGF21 resistance may contribute to the association of FGF21 and nonalcoholic fatty liver disease (12,13). Although FGF21 has benecial activities in the regula- tion of lipid metabolism, application of FGF21 is limited by the route of FGF21 administration. Induction of FGF21 expression will be a feasible approach to enhance FGF21 activity in vivo. FGF21 expression is controlled at the transcriptional level by peroxisome proliferatoractivated receptor (PPAR)-a (14). PPAR-a agonist induces FGF21 expression in vitro and in vivo (4,8,15). Moreover, PPAR-g agonists, such as rosiglitazone and pioglitazone, induce FGF21 expression in hepatocytes (16). It is not clear if FGF21 expression is regulated by a bioactive component in the diet. We addressed this issue by investigating so- dium butyrate activity in the regulation of FGF21 in mice. Sodium butyrate (CH3CH2CH2COONa) is a fatty acid derivative found in foods, such as parmesan cheese and butter. It is also produced in large amounts from fermen- tation of dietary ber in the large intestine. We reported that butyrate supplementation prevented obesity, pro- tected insulin sensitivity, and ameliorated dyslipidemia in dietary obese mice (17). Increases in energy expenditure and fatty acid b-oxidation are important factors for the benecial effects of butyrate. However, the endocrine mechanism remains unknown for the elevated b-oxidation of fatty acids. Butyrate is an inhibitor of histone deacetylase (HDAC) (17,18), which removes the acetyl group from protein substrate, such as histone proteins (19,20). HDAC inhibitors regulate gene transcription through modication of histone protein acetylation (21). Studies from this and other groups suggest that HDAC inhibitors may be poten- tial therapeutics for metabolic syndrome (17,22). The mechanism of action deserves to be explored for HDAC inhibitors. As a dietary component, butyrate is more interesting to us. Although butyrate induces fatty acid b-oxidation by modifying gene transcription (17), it is unknown if FGF21 is involved in the metabolic activity of butyrate. In this study, FGF21 was examined to understand its role in the metabolic activities of butyrate. Our results suggest that FGF21 is induced by butyrate and involved in the stimulation of fatty acid b-oxidation in liver. Buty- rate enhances FGF21 transcription through inhibition of HDAC3. From the 1 Department of Endocrinology and Metabolism, Shanghai Jiao Tong University Afliated Sixth Peoples Hospital; Shanghai Diabetes Institute; Shanghai Clinical Center of Diabetes, Shanghai Key Laboratory of Diabetes Mellitus, Shanghai, China; the 2 Department of Medicine, Shanghai Jiao Tong University School of Medicine, Shanghai, China; the 3 Antioxidant and Gene Regulation Laboratory, Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, Louisiana; the 4 Departments of Med- icine and Pharmacology, University of Hong Kong, Hong Kong, China; and the 5 Research Centre of Heart, Brain, Hormone, and Healthy Aging, Univer- sity of Hong Kong, Hong Kong, China. Corresponding author: Weiping Jia, [email protected], or Jianping Ye, yej@ pbrc.edu. Received 21 June 2011 and accepted 4 January 2012. DOI: 10.2337/db11-0846 This article contains Supplementary Data online at http://diabetes .diabetesjournals.org/lookup/suppl/doi:10.2337/db11-0846/-/DC1. Ó 2012 by the American Diabetes Association. Readers may use this article as long as the work is properly cited, the use is educational and not for prot, and the work is not altered. See http://creativecommons.org/licenses/by -nc-nd/3.0/ for details. diabetes.diabetesjournals.org DIABETES, VOL. 61, APRIL 2012 797 ORIGINAL ARTICLE

Transcript of ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate...

Page 1: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

Sodium Butyrate Stimulates Expression of FibroblastGrowth Factor 21 in Liver by Inhibition of HistoneDeacetylase 3Huating Li,

1,2Zhanguo Gao,

3Jin Zhang,

3Xin Ye,

3Aimin Xu,

4,5Jianping Ye,

3and Weiping Jia

1

Fibroblast growth factor 21 (FGF21) stimulates fatty acidoxidation and ketone body production in animals. In this study,we investigated the role of FGF21 in the metabolic activity ofsodium butyrate, a dietary histone deacetylase (HDAC) inhibitor.FGF21 expression was examined in serum and liver afterinjection of sodium butyrate into dietary obese C57BL/6J mice.The role of FGF21 was determined using antibody neutralizationor knockout mice. FGF21 transcription was investigated in liverand HepG2 hepatocytes. Trichostatin A (TSA) was used in thecontrol as an HDAC inhibitor. Butyrate was compared withbezafibrate and fenofibrate in the induction of FGF21 expression.Butyrate induced FGF21 in the serum, enhanced fatty acidoxidation in mice, and stimulated ketone body production inliver. The butyrate activity was significantly reduced by theFGF21 antibody or gene knockout. Butyrate induced FGF21 geneexpression in liver and hepatocytes by inhibiting HDAC3, whichsuppresses peroxisome proliferator–activated receptor-a func-tion. Butyrate enhanced bezafibrate activity in the induction ofFGF21. TSA exhibited a similar set of activities to butyrate.FGF21 mediates the butyrate activity to increase fatty acid useand ketogenesis. Butyrate induces FGF21 transcription by inhibi-tion of HDAC3. Diabetes 61:797–806, 2012

The fibroblast growth factor (FGF) superfamilycontains at least 22 members with diverse bi-ological functions in the control of cell growthand development and wound healing (1). FGF21,

a polypeptide with 210 amino acid residues, is abundantlyexpressed in the liver, although its expression is alsoreported in pancreata, adipose, and muscle (2). FGF21plays an important role in the regulation of lipid metabo-lism (3). It promotes lipid oxidation, triglyceride clearance,and ketogenesis in liver (4). FGF21 knockout (FGF21 KO)mice exhibit deficiency in ketogenesis and loss response toketogenic diet, develop hepatic steatosis, and gain weight(4,5). FGF21 administration increased energy expenditure,

decreased blood lipids, and reduced hepatic steatosis indietary obese mice (6). Infusion of recombinant FGF21also leads to glucose reduction in genetic and dietaryobese mice (3,6). The physiological role of FGF21 remainsto be investigated in humans. Several recent studies showthat serum FGF21 levels are elevated in patients of meta-bolic syndrome (7–10). We reported that serum FGF21was positively associated with the degree of nonalcoholicfatty liver disease in humans (11). FGF21 resistance maycontribute to the association of FGF21 and nonalcoholicfatty liver disease (12,13).

Although FGF21 has beneficial activities in the regula-tion of lipid metabolism, application of FGF21 is limited bythe route of FGF21 administration. Induction of FGF21expression will be a feasible approach to enhance FGF21activity in vivo. FGF21 expression is controlled at thetranscriptional level by peroxisome proliferator–activatedreceptor (PPAR)-a (14). PPAR-a agonist induces FGF21expression in vitro and in vivo (4,8,15). Moreover, PPAR-gagonists, such as rosiglitazone and pioglitazone, induceFGF21 expression in hepatocytes (16). It is not clear ifFGF21 expression is regulated by a bioactive componentin the diet. We addressed this issue by investigating so-dium butyrate activity in the regulation of FGF21 in mice.

Sodium butyrate (CH3CH2CH2COONa) is a fatty acidderivative found in foods, such as parmesan cheese andbutter. It is also produced in large amounts from fermen-tation of dietary fiber in the large intestine. We reportedthat butyrate supplementation prevented obesity, pro-tected insulin sensitivity, and ameliorated dyslipidemia indietary obese mice (17). Increases in energy expenditureand fatty acid b-oxidation are important factors for thebeneficial effects of butyrate. However, the endocrinemechanism remains unknown for the elevated b-oxidationof fatty acids. Butyrate is an inhibitor of histone deacetylase(HDAC) (17,18), which removes the acetyl group fromprotein substrate, such as histone proteins (19,20). HDACinhibitors regulate gene transcription through modificationof histone protein acetylation (21). Studies from this andother groups suggest that HDAC inhibitors may be poten-tial therapeutics for metabolic syndrome (17,22). Themechanism of action deserves to be explored for HDACinhibitors. As a dietary component, butyrate is moreinteresting to us. Although butyrate induces fatty acidb-oxidation by modifying gene transcription (17), it isunknown if FGF21 is involved in the metabolic activityof butyrate.

In this study, FGF21 was examined to understand itsrole in the metabolic activities of butyrate. Our resultssuggest that FGF21 is induced by butyrate and involvedin the stimulation of fatty acid b-oxidation in liver. Buty-rate enhances FGF21 transcription through inhibition ofHDAC3.

From the 1Department of Endocrinology and Metabolism, Shanghai Jiao TongUniversity Affiliated Sixth People’s Hospital; Shanghai Diabetes Institute;Shanghai Clinical Center of Diabetes, Shanghai Key Laboratory of DiabetesMellitus, Shanghai, China; the 2Department of Medicine, Shanghai Jiao TongUniversity School of Medicine, Shanghai, China; the 3Antioxidant and GeneRegulation Laboratory, Pennington Biomedical Research Center, LouisianaState University System, Baton Rouge, Louisiana; the 4Departments of Med-icine and Pharmacology, University of Hong Kong, Hong Kong, China; andthe 5Research Centre of Heart, Brain, Hormone, and Healthy Aging, Univer-sity of Hong Kong, Hong Kong, China.

Corresponding author: Weiping Jia, [email protected], or Jianping Ye, [email protected].

Received 21 June 2011 and accepted 4 January 2012.DOI: 10.2337/db11-0846This article contains Supplementary Data online at http://diabetes

.diabetesjournals.org/lookup/suppl/doi:10.2337/db11-0846/-/DC1.� 2012 by the American Diabetes Association. Readers may use this article as

long as the work is properly cited, the use is educational and not for profit,and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.

diabetes.diabetesjournals.org DIABETES, VOL. 61, APRIL 2012 797

ORIGINAL ARTICLE

Page 2: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

RESEARCH DESIGN AND METHODS

Cells and reagents. Human HepG2 hepatocytes from the American TypeCulture Collection (Manassas, VA) were maintained in Dulbecco’s modifiedEagle’s medium (DMEM) supplemented with 10% FCS. Butyrate (19–137) wasobtained from Millipore (Billerica, MA). Trichostatin A (TSA) (58880–19–6)was obtained from A.G. Scientific (San Diego, CA). Bezafibrate (B7273),fenofibrate (F6020), hexanoate (C4026), and sodium 3-hydroxybutyrate (BOH)(54965) were purchased from Sigma-Aldrich (St. Louis, MO).Animals. Dietary obese mice were generated by feeding male C57BL/6J mice(The Jackson Laboratory, Bar Harbor, ME) a high-fat diet (D12331, 58% caloriesin fat; Research Diets, New Brunswick, NJ) for 20 weeks. The lean control micewere fed chow diet (5001, 13% calories in fat; LabDiet, Richmond, IN). Theexperiments on FGF21 KO mice (23) were conducted at the University of HongKong laboratory of A.X. All of the mice were housed in the animal facility witha 12-h light/dark cycle and constant temperature (22–24°C). The mice werehoused at three to four mice per cage with free access to water and diet. Allthe procedures were approved by the institutional animal care and use com-mittee at the Pennington Biomedical Research Center.

Butyrate treatment. Butyrate or control agent was delivered by intraper-itoneal injections in mice after overnight fasting at the following dosages:butyrate (500 mg/kg body wt), TSA (0.8 mg/kg body wt), hexanoate (625 mg/kgbody wt), and bezafibrate (100 mg/kg body wt). Mice were assigned randomlyto the treatment or control groups at five to eight mice per group. In the re-peating experiments, the mice were allowed to recover for at least 7 days beforethe next injection. Blood and tissue samples were stored at 280°C.FGF21 enzyme-linked immunosorbent assay and ketone body

measurement. Concentrations of FGF21 in serum and liver tissue werequantified using ELISA kits (Antibody & Immunoassay Services, University ofHong Kong). The assay was proven to be highly specific to mouse FGF21, withno cross-reaction to other members of the FGF family. The intra- and inter-assay variations were 4.2 and 7.6%, respectively. Ketone body in the serumwas determined with b-hydroxybutyrate concentration using a ketone bodyassay kit (BioAssay Systems, Hayward, CA).Quantitative real-time PCR. Total RNA was extracted from frozen tissuesusing Tri-Reagent (T9424, Sigma-Aldrich). RNA was used for generation ofcDNA (Bio-Rad, Hercules, CA). Quantitative real-time PCR (qRT-PCR) was

FIG. 1. Butyrate increases FGF21 mRNA and protein in HepG2 cells. A: Increase of FGF21 mRNA expression by butyrate. The cells were serumstarved in DMEM supplemented with 0.25% BSA overnight and treated with bezafibrate (100 mmol/L), fenofibrate (500 mmol/L), butyrate (1 mmol/L),BOH (1 mmol/L), and hexanoate (1 mmol/L) for 2 h. The total RNA was extracted and subjected to qRT-PCR analysis for FGF21 mRNA. B: Increaseof FGF21 protein by butyrate. The cells were serum starved in DMEM supplemented with 0.25% BSA overnight and treated with bezafibrate (100mmol/L), fenofibrate (500 mmol/L), butyrate (1 mmol/L), BOH (1 mmol/L), and hexanoate (1 mmol/L) for 6 h. The FGF21 protein was determined inthe whole-cell lysate by enzyme-linked immunosorbent assay. C: Butyrate increases FGF21 mRNA in a dose-dependent manner. D: Butyrateinduces FGF21 protein in a dose-dependent manner. E: TSA increases FGF21 mRNA in a dose-dependent manner. F: TSA induces FGF21 protein ina dose-dependent manner. Data are mean 6 SEM, n = 4. *P < 0.05 vs. control.

FGF21 REGULATION BY BUTYRATE

798 DIABETES, VOL. 61, APRIL 2012 diabetes.diabetesjournals.org

Page 3: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

performed in a 7900HT Fast Real-Time PCR System (Applied Biosystems,Foster City, CA) using SYBR Green Master Mix (Applied Biosystems). Primerswere made according to published studies (7,14,24). All samples were ana-lyzed in duplicate, and FGF21 signal was normalized with cyclophilin signal(the internal control).Metabolic chamber. Respiratory exchange ratio (RER) was monitored withthe Comprehensive Laboratory Animal Monitoring System (Columbus Instru-ments, Columbus, OH) as described previously (17). To inhibit FGF21 activity,a monoclonal rabbit–anti-mouse FGF21 antibody (Antibody & ImmunoassayServices) was injected at 3 mg per mouse. A rabbit IgG was injected at thesame dose in the control mice.Transfection and luciferase assay. HEK 293 cells were plated in 24-wellplates (4 3 105 cells per well) and transfected with FGF21 luciferase reporter(0.2 mg DNA per well) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA).The FGF21 luciferase reporter was a gift from Dr. Steven Kliewer at theSouthwestern Medical Center, University of Texas, Dallas, Texas (14). Theluciferase assay was conducted using the dual luciferase substrate system(E1501; Promega, Madison, WI), and the result was normalized with the

internal control Renilla luciferase. Each experiment was repeated at leastthree times with consistent results.Chromatin immunoprecipitation. Chromatin immunoprecipitation (ChIP)was conducted in fresh liver tissues collected at 2 h after butyrate injection inmice. Protein-DNA cross-linking was performed in the samples with 1%formaldehyde at room temperature for 15 min and terminated by glycine. Livernuclei were isolated, and chromosome DNA was broken into fragments of 400–1,200 base pair (bp) by sonication. Immunoprecipitation was conducted withantibodies to PPAR-a (sc-9000; Santa Cruz Biotechnology), HDAC3 (ab7030;Abcam), or RNA polymerase II (sc-9001; Santa Cruz Biotechnology). IgG wasused in controls for the nonspecific signal. DNA signal was quantified usingSYBR green qRT-PCR. The PCR primers (Forward: 59-AGGGCCCGAATGC-TAAGC-39; Reverse: 59-AGCCAAGCAGGTGGAAGTCT-39) cover the PPAR-abinding site (21,119/21,044) in the mouse FGF21 gene promoter (14).Statistical analysis. In this study, data are presented as the mean6 SEM frommultiple samples or repeats. All of the in vitro experiments were conducteda minimum of three times. Student t test or two-way ANOVA was used in thestatistical analysis with significance P # 0.05.

FIG. 2. Butyrate administration increases serum and hepatic FGF21 in obese mice. A: Serum FGF21 concentrations at basal level and 7 h afterbutyrate injection in obese mice (n = 8). B: Serum FGF21 concentrations at basal levels and 7 h after TSA injection in obese mice (n = 5–6).C: Serum FGF21 concentrations at basal levels and 7 h after hexanoate injection in obese mice (n = 5–6). D: FGF21 mRNA expressions in the liverof obese mice at 2 h after butyrate injection (n = 4). E: FGF21 protein levels in the liver of obese mice at 2 h after butyrate injection (n = 4). Dataare mean 6 SEM. In A and B, *P < 0.05 vs. basal. In D and E, *P < 0.05 vs. PBS.

H. LI AND ASSOCIATES

diabetes.diabetesjournals.org DIABETES, VOL. 61, APRIL 2012 799

Page 4: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

RESULTS

Butyrate induces FGF21 expression in cellular models.In hepatocytes, FGF21 secretion is controlled at thetranscriptional level, and PPAR-a is an activator of FGF21gene promoter. PPAR-a ligand induces FGF21 tran-scription and protein expression in hepatocytes. It isnot known if sodium butyrate is able to induce FGF21expression. To address this question, we investigatedregulation of FGF21 expression by butyrate in humanHepG2 hepatocytes. In the study, bezafibrate and fenofibratewere used as positive controls for the induction of FGF21expression. FGF21 was induced in mRNA and protein byeither positive control (Fig. 1A and B). In the absence ofpositive controls, butyrate induced FGF21 expression inmRNA and protein dramatically (Fig. 1A and B). The in-duction was in a dose-dependent manner for both mRNAand protein (Fig. 1C and D). Butyrate is a sodium salt of

short-chain fatty acid with HDAC inhibitor activity. Todifferentiate the activities of fatty acid and HDAC in-hibitor, we used short-chain fatty acids, BOH (four car-bon fatty acid) and hexanoate (six carbon fatty acid), inthe control. At the same concentration to butyrate, thetwo fatty acids did not exhibit any significant activity inthe regulation of FGF21 mRNA or protein (Fig. 1A and B).Since these two fatty acids are not HDAC inhibitors, thedata suggest that butyrate may act through inhibition ofHDACs. To test this possibility, we used TSA, a classicalHDAC inhibitor. Like butyrate, TSA increased FGF21 inmRNA and protein in a dose-dependent manner (Fig. 1Eand F). These data suggest that in the absence of PPAR-aligand, butyrate increases FGF21 expression in hepa-tocytes. This FGF21 response is related to the inhibitionof HDACs by butyrate, but not the fatty acid nature ofbutyrate.

FIG. 3. Butyrate administration enhances ketogenesis via induction of FGF21. A: Serum b-hydroxybutyrate concentrations at 7 h after butyrateinjection in obese mice. B: Ketogenic gene (HMGCS2 and CPT1a) mRNA expressions in the liver of obese mice. *P < 0.05 vs. PBS. C: Correlationsof increasing values of serum b-hydroxybutyrate with increasing values of serum FGF21 in obese mice treated with butyrate. D: Correlations ofincreasing values of serum b-hydroxybutyrate with increasing values of serum FGF21 in obese mice treated with TSA. Data are mean 6 SEM (n =6–8). E: Serum b-hydroxybutyrate concentrations at basal levels and 7 h after butyrate injection in the FGF21 KO and WT mice. F: Ketogenic gene(HMGCS2 and CPT1a) mRNA expressions after butyrate injection compared with baseline in the liver of FGF21 KO and WT mice. Data are mean6SEM (n = 5). *P < 0.05 vs. WT.

FGF21 REGULATION BY BUTYRATE

800 DIABETES, VOL. 61, APRIL 2012 diabetes.diabetesjournals.org

Page 5: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

Butyrate administration increases serum FGF21 inobese mice. To test the butyrate activity in vivo, we ex-amined serum FGF21 in obese mice after butyrate treat-ment, which was administrated through intraperitonealinjection. In the mouse study, the same controls wereused, such as PBS in the vehicle control, bezafibrate asa positive control, TSA for HDAC inhibition, and hex-anoate as the fatty acid control. Bezafibrate increased se-rum FGF21 by onefold as expected. Butyrate exhibited thesame activity in the induction of serum FGF21 (Fig. 2A).TSA also induced the FGF21 by onefold (Fig. 2B). Nosignificant change was observed for PBS (Fig. 2A–E) orhexanoate (Fig. 2C). To determine the source of FGF21,we examined FGF21 expression in the liver at 2 h afterinjection. The induction in FGF21 was observed for beza-fibrate and butyrate (Fig. 2D and E). These data suggestthat butyrate induces serum FGF21 in mice by activationof FGF21 expression in liver. The activity is related to theHDAC inhibition by butyrate.Butyrate enhances ketogenesis via FGF21 in mice.Induction of ketone body production is a major activityof FGF21, which induces gene expression in the keto-genesis pathway. To determine the biological activity,we examined the serum ketone bodies by measuringb-hydroxybutyrate in the butyrate-treated mice. Butyratesignificantly increased b-hydroxybutyrate, and TSA ex-hibited a similar activity (Fig. 3A). Carnitine palmitoyl-transferase 1a (CPT1a) and 3-hydroxy-3-methylglutaryl-CoAsynthase 2 (HMGCS2) are FGF21 responsive genes and arerequired for ketogenesis by FGF21 (25,26). In the liver, thetwo genes were increased in mRNA by more than onefoldin response to butyrate (Fig. 3B). The increase in bloodketone body was positively associated with the change inserum FGF21 (Fig. 3C and D). To test whether butyrateenhances ketogenesis via FGF21, we conducted the studyin FGF21 KO mice. KO mice and wild-type (WT) miceexhibited similar levels of ketone body at the baseline. Inresponse to butyrate, the ketone body level was significantlylower in the KO mice (Fig. 3E). The WT mice exhibited1.8 mmol/L increase in b-hydroxybutyrate, while KO miceexhibited only 0.5 mmol/L increase in b-hydroxybutyrate.The responses in CPT1a and HMGCS2 were also signifi-cantly lower in the KO mice (Fig. 3F). This group of datasuggests that butyrate induces ketogenesis. FGF21 is re-quired for the butyrate activity.FGF21 mediates butyrate activity to promote fattyacid use in obese mice. The increase in ketone bodyproduction suggests elevation in fatty acid use by the liver,

which produces ketone bodies through b-oxidation oflong-chain fatty acids. Fatty acids and glucose are sub-strates in mitochondria for the production of ATP in theperipheral tissues. There is competition between the twosubstrates. When fatty acid usage is enhanced, carbohy-drate use will be reduced. To test fatty acid oxidation inthe whole body, we determined RER using the metabolicchamber. RER is a volume ratio of CO2 exhaled versus O2inhaled in the body. A decrease in RER suggests an in-crease in fatty acid use. In response to the butyrate in-jection, the mice exhibited a 6% reduction in RER value(Fig. 4A). The reduction was blocked when the serumFGF21 was neutralized by the FGF21 antibody, whichdecreased FGF21 protein in the serum (Fig. 4B). In thecontrol, a nonspecific IgG was used. The butyrate activitywas not affected (Fig. 4A), and FGF21 protein was not re-duced. In the study, butyrate did not change the O2 con-sumption and spontaneous physical activity in the mice (datanot shown). These results suggest that butyrate increasesb-oxidation of fatty acids in the whole body and that theactivity requires FGF21. The acute treatment by butyratedid not change energy expenditure in the obese mice.Butyrate activates FGF21 gene transcription byinhibiting HDAC3. FGF21 expression is controlled atthe transcriptional level. To understand the molecularmechanism of butyrate action, we investigated FGF21gene transcription. The butyrate response element wasinvestigated in the FGF21 gene promoter. The elementwas searched through screening the promoter fragmentsof different length spanning 21,497 and 5 (14). The ac-tivity of the longest promoter (21,497/5 bp) was inducedsignificantly by butyrate or bezafibrate as indicated by theluciferase activity (Fig. 5A). In the same condition, theshorter (,977 bp) promoters did not respond to butyrate.There are two PPAR-a response elements in the longpromoter, and the distal element is absent in the shorterpromoter. The data suggest that the distal element is re-sponsive to butyrate. When butyrate was compared withTSA in the induction of the long promoter, both agentsexhibited dose-dependent activity in the activation of thereporter with a similar strength (Fig. 5B). This group ofdata suggests that butyrate induces FGF21 promoter ac-tivity by inhibition of HDACs. The distal PPAR-a responseelement may be activated by butyrate.

Butyrate inhibits multiple HDACs in class I and II, whichcontain .10 isoforms. Butyrate inhibits HDAC1, HDAC2,and HDAC3. To identify the HDAC isoform that is re-sponsible for FGF21 promoter activation, we used gene

FIG. 4. Butyrate administration promotes fatty acid oxidation via FGF21 in obese mice. Substrate use, O2 consumption, and spontaneous physicalactivity were examined using the metabolic chamber in obese mice (n = 8). Butyrate was injected at 500 mg/kg body wt at 12 A.M., and data wererecorded during 7 P.M. to 7 A.M. after the injection. A: Substrate use in mice is expressed by RER. B: Serum FGF21 was determined after injection of3 mg rabbit–anti-mouse FGF21 antibody (Ab) or normal IgG. Data are mean 6 SEM. *P < 0.05 vs. basal.

H. LI AND ASSOCIATES

diabetes.diabetesjournals.org DIABETES, VOL. 61, APRIL 2012 801

Page 6: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

knockdown. Vector-based small interfering RNA (siRNA)was used to knock down each of HDAC1, HDAC2, andHDAC3 in transient transfection (Supplementary Data 1).The FGF21 reporter was used to monitor the knockdowneffect. HDAC3 knockdown increased the promoter activityby fivefold (Fig. 5C), while knock down of HDAC1 andHDAC2 generated only weaker responses in the promoter.The data suggest that HDAC3 is the major HDAC isoform

in the FGF21 gene promoter. To understand the relation-ship of HDAC3 and PPAR-a element, we overexpressedPPAR-a to activate the element specifically and knockeddown HDAC3. In this model, the PPAR-a activity wasenhanced by 26-fold in the presence of HDAC3 knock-down (Fig. 5D). The knockdown effect was significantlyreduced in the shorter promoters of 2977/5 or 298/5(Fig. 6D). In the shortest promoter (266/5) that does not

FIG. 5. Butyrate activates FGF21 gene transcription by targeting HDAC3. Transient transfection was conducted with the FGF21 luciferasereporters in HEK 293 cells. In cotransfection, the expression plasmids for PSG5–PPAR-a, retinoid X receptor, siRNA for HDACs (siHDAC1,siHDAC2, and siHDAC3), and scrambled siRNA were used. The cells were treated with bezafibrate, butyrate, or TSA at 24 h after transfection andharvested 16 h later for luciferase assay. In all of the transient transfection experiments, the internal control was 0.1 mg/well of simian virus 40(SV40) R. luciferase reporter plasmid, and the total DNA concentration was corrected in each well with a control plasmid. A: Butyrate responseelement. Cells were transfected with the FGF21 luciferase reporter plasmids in the presence of SV40. Cells were treated with 100 mmol/L beza-fibrate or 0.5 mmol/L butyrate for 16 h. B: Dose-dependent activity of butyrate. Cells were transfected with the long promoter (21,497/5) reporterin the presence of PPAR-a expression plasmid or control plasmid. Cells were treated with bezafibrate (100 mmol/L), butyrate (0.5, 1.5, and2.0 mmol/L), or TSA (100, 200, and 400 nmol/L) for 16 h. C: HDAC3 knockdown. In the FGF21 (21,497/5) reporter system, HDACs were inhibitedwith siRNA of HDAC1, HDAC2, and HDAC3. Scrambled siRNA was used as a negative control. D: HDAC3 interaction with PPAR-a. In the FGF21luciferase reporter system, the reporter activity was induced by cotransfection of PPAR-a expression vector. HDAC3 was knocked down by siRNA.Data are mean 6 SEM. *P < 0.05 vs. control; #P < 0.05 vs. shorter promoters. TATA, TATA box.

FGF21 REGULATION BY BUTYRATE

802 DIABETES, VOL. 61, APRIL 2012 diabetes.diabetesjournals.org

Page 7: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

have a peroxisome proliferator response element (PPRE),the effect of HDAC3 knockdown was not observed. Thedata suggest that HDAC3 may regulate both PPREs in theFGF21 gene promoter. However, the distal PPRE is moresensitive to HDAC3. This group of data suggests that buty-rate activates the FGF21 gene promoter through the distalPPRE by inhibiting HDAC3, which inhibits the gene pro-moter at the PPREs.HDAC3 interacts with PPAR-a in the FGF21 promoter.The knockdown data suggest that HDAC3 suppressesPPAR-a in the FGF21 gene promoter. HDAC3 is likelya component of PPAR-a corepressor that inhibits thetranscription activity in the absence of PPAR-a ligand toblock recruitment of RNA polymerase II. In response toPPAR-a activation, the corepressor will be disabled, leadingto recruitment of RNA polymerase II in the initiation of genetranscription. Those possibilities are supported by thefunctional assay of the FGF21 gene promoter. To prove theHDAC3–PPAR-a interaction at the protein level, we per-formed ChIP assay at the distal PPRE in the FGF21 genepromoter in liver tissues. Bezafibrate increased PPAR-abinding to the FGF21 promoter DNA (Fig. 6A). The PPAR-aactivity was associated with an increase in RNA polymeraseII signal (Fig. 6B) and a decrease in HDAC3 signal (Fig. 6C).Butyrate did not increase PPAR-a binding to the FGF21gene but did increase the polymerase II signal and reducethe HDAC3 signal (Fig. 6A–C). These data suggest thatHDAC3 interacts with PPAR-a in the FGF21 gene promoter.Butyrate activates PPAR-a without increasing its DNA-binding activity. Bezafibrate increases the DNA binding ofPPAR-a.Butyrate exhibits additive effects with bezafibrate inthe induction of FGF21. The above data suggest thatbutyrate and bezafibrate induce FGF21 expression throughthe PPRE. However, they act through different mecha-nisms, as suggested by the ChIP assays. The two agentsmay have an additive or synergistic interaction in the in-duction of FGF21 expression. To test the possibility, wecombined the two agents in the induction of FGF21 expres-sion. In the HepG2 cell line, bezafibrate induced FGF21protein expression, as indicated by the protein abundancein the whole-cell lysate. In the presence of butyrate, thebezafibrate effect was enhanced significantly (Fig. 7A). Inobese mice, the bezafibrate activity was also enhanced bybutyrate, as indicated by the change in serum FGF21

protein (Fig. 7B). In liver tissue, butyrate enhanced thebezafibrate activity with an increase in FGF21 mRNAand protein (Fig. 7C and D). Butyrate also enhancedbezafibrate activity in the induction of ketone body pro-duction. After treatment by bezafibrate, the obese mice didnot exhibit a significant increase in b-hydroxybutyrate(Fig. 7E). When bezafibrate was administrated togetherwith butyrate, the serum ketone body was elevated morethan onefold (Fig. 7E). To understand the mechanism,we examined several PPAR-a target genes that are in-volved in ketogenesis and fatty acid oxidation, such asHMGCS2, CPT1a, very long-chain acyl-CoA dehydroge-nase (ACADVL), acyl-CoA synthase long-chain familymember 1 (ACSL1), acyl-CoA oxidase (ACO), and liver fattyacid–binding protein (L-FABP). Bezafibrate induced allof these genes (Fig. 7F). In the presence of butyrate,bezafibrate-induced responses were significantly enhancedin most of these genes (Fig. 7F). These data suggest thatbutyrate enhances the activity of PPAR-a agonist in thestimulation of FGF21 secretion and ketone body pro-duction and the expressions of other PPAR-a targetgenes.

DISCUSSION

Our study suggests that butyrate is a new inducer ofFGF21. FGF21 is a cytokine/hormone that stimulates useof long-chain fatty acids through b-oxidation in liver in theproduction of ketone bodies. FGF21 is required for theantiobesity effect of ketogenic diet as shown in the phe-notype of FGF21 KO mice (4,5). The studies suggest thatFGF21 is required for b-oxidation of the dietary long-chainfatty acids. Induction of FGF21 expression may be an ap-proach to treat obesity. However, there is not much optionavailable in the induction of FGF21. PPAR-a ligand/agonistare able to induce FGF21 expression (4,14,15), but theycannot apply to every patient. As a precursor of PPAR-aligand, the long-chain fatty acids also induce FGF21 ex-pression (27). Serum FGF21 correlates to the concentra-tion of long-chain fatty acid in humans in a 24-h oscillatorypattern, as shown in our previous study (28). PPAR-gligands recently were reported to induce FGF21 (16). He-patic FGF21 is also induced by triiodothyronine via aPPAR-a–dependent mechanism (29). The current study sug-gests that butyrate is a powerful inducer of FGF21 in liver.

FIG. 6. HDAC3 interacts with PPAR-a in the FGF21 promoter. ChIP assays were performed using liver tissues collected after bezafibrate andbutyrate injection. Immunoprecipitation was performed with antibodies to PPAR-a (A), polymerase II (B), and HDAC3 (C). Rabbit IgG was used inthe negative control. The specific chromatin DNA was quantified in qRT-PCR with the primer against the distal PPRE (21,119/21,044) in themouse FGF21 gene promoter. Data are mean 6 SEM, n = 4. *P < 0.05 vs. IgG control.

H. LI AND ASSOCIATES

diabetes.diabetesjournals.org DIABETES, VOL. 61, APRIL 2012 803

Page 8: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

We examined FGF21 in epididymal fat and did not observethe effect (data not shown), suggesting a tissue-specific ef-fect of butyrate. As a dietary component and a product ofa dietary fiber, butyrate is an outstanding bioactive agent inthe dietary intervention of obesity.

In this study, we investigated the role of FGF21 in thebutyrate regulation of fatty acid oxidation. FGF21 activitywas tested by antibody neutralization or in FGF21 KOmice. In mice treated with butyrate, FGF21 stimulatesketone body production in liver and increases fatty acid

FIG. 7. Butyrate enhances PPAR-a agonist activity in the stimulation of FGF21 expression. A: FGF21 protein expression in HepG2 cells. FGF21protein was determined in the whole-cell lysate by enzyme-linked immunosorbent assay (ELISA) after HepG2 cells were treated with bezafibrateand butyrate. *P < 0.05 vs. control; #P < 0.05 vs. bezafibrate. B: Serum FGF21 concentration. Serum FGF21 was determined in obese mice at 7 hafter treatment with bezafibrate (100 mg/kg body wt) and butyrate (500 mg/kg body wt) (n = 5–6). *P < 0.05 vs. basal; #P < 0.05 vs. bezafibrate.C: FGF21 mRNA in the liver of obese mice. The liver tissue was collected at 2 h after the mice were treated with bezafibrate and butyrate (n = 4).The mRNA expression was determined in qRT-PCR. D: Protein level in the liver. FGF21 protein was determined in the liver tissue by ELISA. E:Serum b-hydroxybutyrate concentration at 7 h after butyrate and bezafibrate injection in obese mice (n = 5–6). F: Genes associated with keto-genesis and fatty acid oxidation (HMGCS2, CPT1a, ACADVL, ACSL1, ACO, and L-FABP) mRNA expressions in the liver. Data are mean 6 SEM.*P < 0.05 vs. PBS; #P < 0.05 vs. bezafibrate.

FGF21 REGULATION BY BUTYRATE

804 DIABETES, VOL. 61, APRIL 2012 diabetes.diabetesjournals.org

Page 9: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

use in whole body. FGF21 represents an endocrinemechanism by which butyrate enhances fatty acid use, asobserved in our previous study (17). In that study, buty-rate enhanced energy expenditure and prevented dietaryobesity in mice. The mechanism is related to PPAR-gcoactivator 1-a activation, which enhances mitochondrialbiogenesis. Before this study, it was unknown whether ahormone is involved in the butyrate activity. In the currentstudy, FGF21 was found as a butyrate-responsive cytokine/hormone in the stimulation of ketone body production andfatty acid use. These butyrate activities were blocked byFGF21 antibody or FGF21 gene knockout. These data sug-gest that FGF21 mediates butyrate activities in the regula-tion of fatty acid metabolism. FGF21 may also contribute toglucose reduction by butyrate (Supplementary Data 2).

The current study suggests that butyrate stimulatesFGF21 expression through activation of PPAR-a. ThePPAR-a activation is a result of HDAC3 inhibition by bu-tyrate. We found that butyrate activates the distal PPRE inthe FGF21 gene promoter. The mechanism is inhibition ofHDAC3. This conclusion is supported by several lines ofevidence, such as the gene promoter analysis, HDAC in-hibitor activity, HDAC3 knockdown effect, and ChIP assay.Although PPAR-a activates FGF21 transcription, as repor-ted by several groups, it is not clear how PPAR-a is regu-lated by the nuclear corepressor. We found that HDAC3 isa component of the corepressor. Inhibition of HDAC3 bybutyrate, TSA, or gene knockdown led to activation ofPPAR-a in the FGF21 gene promoter. In the ChIP assay,HDAC3 interacted with PPAR-a at the PPRE. These datasuggest that HDAC3 regulates PPAR-a in the FGF21 pro-moter. Inhibition of HDAC3 represents a new approach inthe induction of FGF21.

Our data suggest that butyrate enhances bezafibrateactivity in the induction of FGF21. PPAR-a agonist is themajor medicine in the induction of FGF21 (8,15). Weobserved that butyrate and bezafibrate activated thesame set of PPAR-a target genes. In ChIP assay, beza-fibrate and butyrate all reduced the HDAC3 activity inthe FGF21 gene promoter. However, they act through dif-ferent mechanisms. Bezafibrate enhanced PPAR-a bindingto the promoter DNA, but butyrate did not. Butyrate inhi-bited HDAC3 enzyme activity in the promoter DNA in theactivation of PPAR-a. PPAR-a is able to bind to target DNAin the absence of ligands (14). The difference in mechanismof action suggests why butyrate enhances the bezafibrateactivity. The HDAC inhibitors may promote the therapeuticactivity of PPAR-a agonist.

In summary, we provide evidence that butyrate increa-ses hepatic production of FGF21 to stimulate b-oxidationof long-chain fatty acids and production of ketone bodies.Our data support that fatty acid oxidation is induced byPPAR-a activation as reported for PPAR-a agonists (30–32).Mechanically, butyrate inhibits HDAC3 activity in theFGF21 gene promoter to enhance PPAR-a function inthe transcriptional activation. This butyrate activity is notdependent on the PPAR-a ligand. The study suggests thatbutyrate is a new chemical inducer of FGF21 in the body.Butyrate is likely to enhance the therapeutic actions ofPPAR-a activators and reduce the side effect of the PPAR-aligands through a dosage reduction.

ACKNOWLEDGMENTS

This study was supported by National 973 project of China(2011CB504001); Major Program of Shanghai Municipality

for Basic Research (08dj1400601); and Program for Out-standing Academic Leader (LJ06010) to W.J.; the NationalInstitutes of Health (NIH project grant R01-DK068036-06)to J.Y.; an American Diabetes Association award (1-09-JF-17) to Z.G.; and a Scholarship Award for Excellent DoctoralStudent granted by Ministry of Education, State-SponsoredStudy Abroad Program, and National Key Basic ResearchProgram of China (973 Program) (2012CB524900) to H.L.The qRT-PCR test and metabolic phenotyping and imag-ing studies were conducted in the genomic core, pheno-typing core, and imaging core of the Pennington BiomedicalResearch Center, which are supported by the NIH grants(2P30-DK072476-06 and P20-RR021945).

No potential conflicts of interest relevant to this articlewere reported.

H.L. conducted the experiments and wrote the manu-script. Z.G. and J.Z. contributed to the experiment designand data analysis. X.Y. conducted some experiments. A.X.was involved in experiment design in the study of FGF21KO mice and the FGF21 assay. J.Y. and W.J. designed thestudy and were involved in data interpretation and writingthe manuscript. W.J. is the guarantor of this work and, assuch, had full access to all the data in the study and takesresponsibility for the integrity of the data and the accuracyof the data analysis.

The authors thank Dr. Steven A. Kliewer (University ofTexas Southwestern Medical Center) for the FGF21luciferase constructs. The authors also thank Dr. TaraM. Henagan, Zhong Wang, Xian Zhang, and Yongmei Yu(Pennington Biomedical Research Center) for excellenttechnical support.

REFERENCES

1. Itoh N, Ornitz DM. Evolution of the Fgf and Fgfr gene families. TrendsGenet 2004;20:563–569

2. Kharitonenkov A, Larsen P. FGF21 reloaded: challenges of a rapidlygrowing field. Trends Endocrinol Metab 2011;22:81–86

3. Kharitonenkov A, Shiyanova TL, Koester A, et al. FGF-21 as a novel met-abolic regulator. J Clin Invest 2005;115:1627–1635

4. Badman MK, Pissios P, Kennedy AR, Koukos G, Flier JS, Maratos-Flier E.Hepatic fibroblast growth factor 21 is regulated by PPARalpha and is a keymediator of hepatic lipid metabolism in ketotic states. Cell Metab 2007;5:426–437

5. Badman MK, Koester A, Flier JS, Kharitonenkov A, Maratos-Flier E. Fi-broblast growth factor 21-deficient mice demonstrate impaired adaptationto ketosis. Endocrinology 2009;150:4931–4940

6. Coskun T, Bina HA, Schneider MA, et al. Fibroblast growth factor 21corrects obesity in mice. Endocrinology 2008;149:6018–6027

7. Zhang X, Yeung DC, Karpisek M, et al. Serum FGF21 levels are increased inobesity and are independently associated with the metabolic syndrome inhumans. Diabetes 2008;57:1246–1253

8. Gälman C, Lundåsen T, Kharitonenkov A, et al. The circulating metabolicregulator FGF21 is induced by prolonged fasting and PPARalpha activa-tion in man. Cell Metab 2008;8:169–174

9. Li H, Bao Y, Xu A, et al. Serum fibroblast growth factor 21 is associatedwith adverse lipid profiles and gamma-glutamyltransferase but not in-sulin sensitivity in Chinese subjects. J Clin Endocrinol Metab 2009;94:2151–2156

10. Chavez AO, Molina-Carrion M, Abdul-Ghani MA, Folli F, Defronzo RA,Tripathy D. Circulating fibroblast growth factor-21 is elevated in impairedglucose tolerance and type 2 diabetes and correlates with muscle andhepatic insulin resistance. Diabetes Care 2009;32:1542–1546

11. Li H, Fang Q, Gao F, et al. Fibroblast growth factor 21 levels are increasedin nonalcoholic fatty liver disease patients and are correlated with hepatictriglyceride. J Hepatol 2010;53:934–940

12. Berglund ED, Li CY, Bina HA, et al. Fibroblast growth factor 21 controlsglycemia via regulation of hepatic glucose flux and insulin sensitivity.Endocrinology 2009;150:4084–4093

13. Fisher FM, Chui PC, Antonellis PJ, et al. Obesity is a fibroblast growthfactor 21 (FGF21)-resistant state. Diabetes 2010;59:2781–2789

H. LI AND ASSOCIATES

diabetes.diabetesjournals.org DIABETES, VOL. 61, APRIL 2012 805

Page 10: ORIGINAL ARTICLE Sodium Butyrate Stimulates Expression of … · 2012. 3. 21. · Sodium Butyrate Stimulates Expression of Fibroblast Growth Factor 21 in Liver by Inhibition of Histone

14. Inagaki T, Dutchak P, Zhao G, et al. Endocrine regulation of the fastingresponse by PPARalpha-mediated induction of fibroblast growth factor 21.Cell Metab 2007;5:415–425

15. Christodoulides C, Dyson P, Sprecher D, Tsintzas K, Karpe F. Circulatingfibroblast growth factor 21 is induced by peroxisome proliferator-activatedreceptor agonists but not ketosis in man. J Clin Endocrinol Metab 2009;94:3594–3601

16. Oishi K, Tomita T. Thiazolidinediones are potent inducers of fibroblastgrowth factor 21 expression in the liver. Biol Pharm Bull 2011;34:1120–1121

17. Gao Z, Yin J, Zhang J, et al. Butyrate improves insulin sensitivity and in-creases energy expenditure in mice. Diabetes 2009;58:1509–1517

18. Davie JR. Inhibition of histone deacetylase activity by butyrate. J Nutr2003;133(Suppl.):2485S–2493S

19. Bryzgalova G, Effendic S, Khan A, et al. Anti-obesity, anti-diabetic, andlipid lowering effects of the thyroid receptor beta subtype selective agonistKB-141. J Steroid Biochem Mol Biol 2008;111:262–267

20. Grover GJ, Mellström K, Malm J. Therapeutic potential for thyroid hor-mone receptor-beta selective agonists for treating obesity, hyperlipidemiaand diabetes. Curr Vasc Pharmacol 2007;5:141–154

21. Sewack GF, Ellis TW, Hansen U. Binding of TATA binding protein toa naturally positioned nucleosome is facilitated by histone acetylation. MolCell Biol 2001;21:1404–1415

22. Mihaylova MM, Vasquez DS, Ravnskjaer K, et al. Class IIa histone deacetylasesare hormone-activated regulators of FOXO and mammalian glucose homeo-stasis. Cell 2011;145:607–621

23. Hotta Y, Nakamura H, Konishi M, et al. Fibroblast growth factor 21 regu-lates lipolysis in white adipose tissue but is not required for ketogenesisand triglyceride clearance in liver. Endocrinology 2009;150:4625–4633

24. Clemenz M, Frost N, Schupp M, et al. Liver-specific peroxisome pro-liferator-activated receptor alpha target gene regulation by the angiotensintype 1 receptor blocker telmisartan. Diabetes 2008;57:1405–1413

25. Drynan L, Quant PA, Zammit VA. Flux control exerted by mitochondrialouter membrane carnitine palmitoyltransferase over beta-oxidation, ke-togenesis and tricarboxylic acid cycle activity in hepatocytes isolated fromrats in different metabolic states. Biochem J 1996;317:791–795

26. Hegardt FG. Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase:a control enzyme in ketogenesis. Biochem J 1999;338:569–582

27. Mai K, Andres J, Biedasek K, et al. Free fatty acids link metabolism andregulation of the insulin-sensitizing fibroblast growth factor-21. Diabetes2009;58:1532–1538

28. Yu H, Xia F, Lam KS, et al. Circadian rhythm of circulating fibroblastgrowth factor 21 is related to diurnal changes in fatty acids in humans. ClinChem 2011;57:691–700

29. Adams AC, Astapova I, Fisher FM, et al. Thyroid hormone regulates he-patic expression of fibroblast growth factor 21 in a PPARalpha-dependentmanner. J Biol Chem 2010;285:14078–14082

30. Litherland NB, Bionaz M, Wallace RL, Loor JJ, Drackley JK. Effects of theperoxisome proliferator-activated receptor-alpha agonists clofibrate andfish oil on hepatic fatty acid metabolism in weaned dairy calves. J Dairy Sci2010;93:2404–2418

31. Muoio DM, Way JM, Tanner CJ, et al. Peroxisome proliferator-activatedreceptor-alpha regulates fatty acid utilization in primary human skeletalmuscle cells. Diabetes 2002;51:901–909

32. Cree MG, Newcomer BR, Herndon DN, et al. PPAR-alpha agonism im-proves whole body and muscle mitochondrial fat oxidation, but does notalter intracellular fat concentrations in burn trauma children in a ran-domized controlled trial. Nutr Metab (Lond) 2007;4:9

FGF21 REGULATION BY BUTYRATE

806 DIABETES, VOL. 61, APRIL 2012 diabetes.diabetesjournals.org