QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: mol.108.044826v1 74/2/403    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Muise, E. S. Articles by Wong, K. K. Search for Related Content PubMed PubMed Citation Articles by Muise, E. S. Articles by Wong, K. K.

Adipose Fibroblast Growth Factor 21 Is Up-Regulated by Peroxisome Proliferator-Activated Receptor and Altered Metabolic States

Departments of Molecular Profiling (E.S.M, J.R.T, K.K.W) and Metabolic Disorders (B.A., D.W.K., M.E.-S., X.Y., J.M., J.P.B.), Merck Research Laboratories, Rahway, New Jersey; and Department of Informatics, Rosetta Inpharmatics LLC, Seattle, Washington (Y.T.)

Received for publication January 16, 2008.

Accepted for publication May 7, 2008.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Adipose tissue is a metabolically responsive endocrine organ that secretes a myriad of adipokines. Antidiabetic drugs such as peroxisome proliferator-activated receptor (PPAR) agonists target adipose tissue gene expression and correct hyperglycemia via whole-body insulin sensitization. The mechanism by which altered gene expression in adipose tissue affects liver and muscle insulin sensitivity (and thus glucose homeostasis) is not fully understood. One possible mechanism involves the alteration in adipokine secretion, in particular the up-regulation of secreted factors that increase whole-body insulin sensitivity. Here, we report the use of transcriptional profiling to identify genes encoding for secreted proteins the expression of which is regulated by PPAR agonists. Of the 379 genes robustly regulated by two structurally distinct PPAR agonists in the epididymal white adipose tissue (EWAT) of db/db mice, 33 encoded for known secreted proteins, one of which was FGF21. Although FGF21 was recently reported to be up-regulated in cultured adipocytes by PPAR agonists and in liver by PPAR agonists and induction of ketotic states, we demonstrate that the protein is transcriptionally up-regulated in adipose tissue in vivo by PPAR agonist treatment and under a variety of physiological conditions, including fasting and high fat diet feeding. In addition, we found that circulating levels of FGF21 protein were increased upon treatment with PPAR agonists and under ketogenic states. These results suggest a role for FGF21 in mediating the antidiabetic activities of PPAR agonists.

The goal of the current study was to systematically identify potential secreted proteins regulated by PPAR in an animal model of diabetes using transcriptional profiling. We report the identification of 33 genes, coding for secreted proteins, robustly regulated by PPAR agonists in epididymal white adipose tissue (EWAT) of db/db mice, one of which was FGF21. Some of these have been previously reported to be regulated by PPAR, whereas others have not, although some have been implicated in metabolic processes. It is noteworthy that expression of FGF21 in vivo was found to be regulated by PPAR, suggesting a role for this protein in mediating the antidiabetic activities of PPAR agonists.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Compounds. Rosiglitazone (5-(4-{2-[methyl(pyridin-2-yl)amino]-ethoxy}benzyl)-1,3-thiazolidine-2,4-dione); the nonthiazolidinedione PPAR agonist, COOH [2-(2-(4-phenoxy-2-propylphenoxy)ethyl)-indole-5-acetic acid (Carley et al., 2004; Laplante et al., 2006)]; and fenofibrate (2-[4-(4-chlorobenzoyl)phenoxy]-2-methylpropanoic acid isopropyl ester) were synthesized by Merck Research Laboratory chemists. WY14643 was purchased from Chemsyn.

In Vivo Treatments for Transcriptional Profiling. All animal experiments and euthanasia protocols were conducted in strict accordance with the National Research Council's Guide for the Care and Use of Laboratory Animals. Animal experiment protocols were reviewed and approved by the Institutional Animal Care and Use Committee of Merck Research Laboratories. The laboratory animal facilities of Merck Research Laboratories are certified by the Association for Assessment and Accreditation of Laboratory Animal Care International. Animals were housed in temperature-, humidity-, and light-controlled rooms (21-23°C, 47-65%, 12-12 h light/dark cycle, respectively). All animals were male and were euthanized by CO₂ asphyxiation 6 h after the final dose (for drug treatments) and the tissues were harvested and flash-frozen in liquid nitrogen. Drug dosing was performed once daily in the morning by oral gavage for the number of days indicated.

db/db mice (lpr-/-, 10 weeks old, n = 3 per treatment group; Charles River Laboratories, Wilmington, MA) were housed eight animals per cage, provided food (LabDiet 5001; Purina St. Louis, MO) and water ad libitum, and dosed with vehicle (0.25% methylcellulose), rosiglitazone (30 mg/kg), or COOH [30 mg/kg (Carley et al., 2004; Laplante et al., 2006)] for 8 days (efficacious doses). Fasting tail-nick blood samples were collected 1 day before the start of dosing and 1 day after the last dose and plasma concentrations of glucose were determined (glucose Trinder assay kit; Sigma, St. Louis, MO). For serum FGF21 protein determination, 7-week-old animals were dosed once daily in the morning by oral gavage with vehicle (0.25% methylcellulose) or rosiglitazone (10 or 30 mg/kg) for 14 days.

To delineate the role of PPAR in the regulation of FGF21, PPAR KO mice [129Sv background; n = 3 per treatment group; Taconic Farms, Germantown, NY (Lee et al., 1995)] or their wild-type litter mates were dosed with vehicle (0.25% methylcellulose), rosiglitazone (100 mg/kg), or fenofibrate (200 mg/kg) for 7 days. Doses were chosen to ensure maximal receptor engagement because these animals were normoglycemic. PPAR KO mice were unavailable because of the embryonic lethality of the mutation.

To demonstrate FGF21 regulation by different PPAR agonists across multiple mouse models and fat depots, two additional strains (lean CD1 and C57BL/6) and a second fat depot (inguinal white adipose tissue, IWAT) were tested. CD1 mice (11 weeks old, n = 3 per treatment group; Charles River Laboratories) were dosed with vehicle (0.25% methylcellulose), rosiglitazone (100 mg/kg), or fenofibrate (300 mg/kg) for 14 days, whereas C57BL/6 mice (9-11 weeks old, n = 3 per treatment group; Charles River Laboratories) were fed standard rodent chow and were dosed with either vehicle (0.25% methylcellulose), the PPAR agonist WY14643 (30 mg/kg), or rosiglitazone (100 mg/kg) for 7 days. Doses were chosen to ensure maximal receptor engagement because these animals were normoglycemic. To determine the effects of ketogenic conditions on FGF21 expression, a second group of C57BL/6 mice (n = 4 per treatment group; Taconic Farms) were fed chow diet (LM-485 mouse/rat sterilizable diet; Harlan Teklad, Madison, WI) until 5 weeks of age, then either kept on chow diet or switched to a high-fat diet [DIO (VHFD) 60 kcal% fat; Research Diets, New Brunswick, NJ] for 10 weeks. A group of at least five mice on chow diet were fasted for 24 h before tissue harvest.

In Vitro Treatments for Transcriptional Profiling. 3T3-L1 cells were grown to confluence in medium A (Dulbecco's modified Eagle's medium with 10% fetal calf serum, 100 units/ml penicillin, and 100 µg/ml streptomycin) at 37°C in 5% CO₂ and induced to differentiate as described previously (Zhang et al., 1996). In brief, differentiation was induced by incubating the cells with medium A supplemented with 3-isobutyl-1-methylxanthine, dexamethasone, and insulin for 2 days, followed by 2 days' incubation with medium A supplemented with insulin. The cells were further incubated in medium A for an additional 4 days to complete the adipocyte conversion. At day 8 after the initiation of differentiation, cells were incubated in medium A plus or minus compounds for 24 h at saturating concentrations: vehicle (0.1% dimethyl sulfoxide), rosiglitazone (10 µM), or COOH (10 µM).

RNA Isolation from Tissues and Cells. Total RNA was isolated from frozen tissues after homogenizing in UltraSpec RNA reagent (Biotecx Laboratories, Inc., Houston, TX) or TRIzol reagent (Invitrogen, Carlsbad, CA) with a PT 10/35 Polytron (Kinematica Inc., Bohemia, NY) and processed using RNeasy kits (QIAGEN, Valencia, CA) according to the manufacturers' instructions. For 3T3-L1 adipocytes, total RNA was prepared using TRIzol reagent (Invitrogen) and RNeasy kits (QIAGEN). RNA concentrations were estimated from absorbance at 260 nm.

Microarray Analysis. Microarray processing was performed as described previously (Hughes et al., 2001). In brief, labeled cRNA was hybridized for 48 h onto 60-mer two-color spotted microarrays (Agilent Technologies, Santa Clara, CA). Individual treatment samples (including individual vehicle treatment samples) were hybridized against a vehicle treatment pool. LogRatio and P values were generated by averaging replicates (3-5 replicates per treatment) and using the AHEM algorithm in the Rosetta Resolver version 6.0 software (Rosetta Inpharmatics LLC, Seattle, WA). LogRatio values represent the difference in regulation of the compound treated samples versus the vehicle-treated sample, where a positive value signifies up-regulation after compound treatment and a negative value signifies down-regulation after compound treatment. Genes coding for secreted proteins were identified using keyword searches such as "secreted" and "extracellular" from the SwissProt, TrEMBL, and Proteome databases.

Cloning of FGF21 into The Yeast Expression Vector pPICZB. The mouse FGF21 cDNA was cloned into the pPICZB vector by polymerase chain reaction from liver cDNA using standard molecular biology techniques and the primers 5'-CTCGAGAAAAGATACCCCATCCCTGACTCCAGCCCCCTCCTC-3' (sense) and 5'-GCGGCCGCTCACTTGTCATCGTCGTCCTTGTAATCGGACGCATAGCTGGGGCTTCGGCCCTGT-3' (antisense). A FLAG tag sequence was introduced at the 3' end.

Expression and Purification of Recombinant FGF21. The recombinant plasmid pPICZ-FGF21 was linearized with SalI to favor the integration at the his4 locus of Pichia pastoris genome and transformed into yeast by electroporation. Transformants were selected by incubation at 30°C for 2 to 3 days on yeast extract/peptone/glucose agar plates containing 100 µg/ml Zeocin. A total of 16 Zeocin-resistant colonies were analyzed for production of FGF21 through small scale expression trials in 10 ml of buffered minimal glycerol-complex medium. The cultures were incubated at 30°C until they reached OD₆₀₀ = 2 (approximately 24 h). Subsequently, the cells were pelleted by centrifugation (3000g for 5 min) and resuspended in 2 ml of buffered minimal methanol-complex medium for induction of the AOX I promoter. For expression of FGF21, methanol was added to the culture every 24 h at a final concentration of 0.5%, and incubation was carried out at 30°C for 3 days with vigorous shaking. The culture medium was cleared of yeast cells by centrifugation at 3000g for 10 min. Presence of recombinant FLAG-tagged FGF21 in the culture media was monitored by SDS-PAGE followed by silver staining and by Western blot analysis using anti-FLAG antibodies. The estimated full-length protein yield was 1 mg/ml of media. A yeast clone producing a protein species of the same size as FGF21 was chosen for large-scale production, where the volume of growth medium was increased to 1000 ml. The purification procedure was as described above for small-scale production but increased proportionately.

FGF21 was purified via affinity column chromatography using anti-FLAG M2 monoclonal antibody affinity gel (Sigma) from the media containing secreted recombinant FLAG-tagged mouse FGF21. Fractions were eluted with 250 µg/ml FLAG peptide. The majority of the purified protein eluted in fraction 2. The peak fractions were concentrated using the Biomax-5 ultrafiltration device (Millipore, Billerica, MA) and washed three times with PBS to remove any residual peptide. Final protein concentration was determined using BCA protein assay (Pierce, Rockford, IL) according to the manufacturers' instructions, using bovine serum albumin protein dilutions for standard curve.

Generation of Antipeptide Antibodies. Three different antigenic peptides were identified, synthesized and conjugated with keyhole limpet hemocyanin (GGQVRQRYLYTDDDQ = peptide A; LEDGYNVYQSEAHG = peptide B; and LPPEPPDVGSSDPL = peptide C). They were used in combination to challenge rabbits (Covance Research Products, Princeton, NJ). The antisera were tested by enzyme-linked immunosorbent assay (Covance Research Products) with each antigenic peptide to give the individual relative antigenic response.

Evaluation of Mouse Plasma/Serum and Sample Preparation. Approximately 1 ml of whole blood was removed via cardiac puncture and immediately aliquoted into two 500-µl Microtainer tubes with EDTA (BD Biosciences, San Jose, CA) and mixed by slow inversion. The blood samples were then spun in a Hermle refrigerated microcentrifuge (with adapters; MIDSCI, St. Louis, MO) for 5 min at 14,000 rpm at 4°C. The supernatant (plasma) was removed carefully and aliquoted into 1.5-ml Eppendorf tubes and flash-frozen in liquid nitrogen. Approximately 250 to 300 µl of plasma per animal was collected by this method. To load the maximum volume of plasma/serum on the gel, albumin was depleted using the Cibacron Blue spin column albumin removal kit from Genomic Solutions, Inc (Ann Arbor, MI). Fifty microliters of serum was diluted to 200 µl with binding buffer provided and passed over the column. The albumin-depleted serum was passed over a second time, and this 200 µl of diluted serum was mixed with 5x sample buffer (250 mM Tris, pH 6.8, 10% SDS, 0.5% bromphenol blue, 50% glycerol, and 100 mM β-mercaptoethanol). An aliquot of 30 µl was boiled for 25 min, and 25 µl of boiling sample was loaded onto a 10 to 20% acrylamide gel for Western blot analysis.

Quantitative Western Blot Analysis. Protein samples were separated on SDS-PAGE 10 to 20% gradient gels (Ready gel; Bio-Rad Laboratories, Hercules, CA) and were transferred to nitrocellulose (0.2 µm pore size; Invitrogen) using the Bio-Rad Western transfer unit. The membranes were blocked in 3% skim milk in PBS at 4°C overnight. Blocking solution was replaced with fresh blocking solution containing 0.1% Tween 20 and antisera at the appropriate dilution. Blots were incubated at room temperature with shaking for 2 h. The membranes were rinsed in water several times and washed four times with PBS + 0.1% Tween 20 for a total of 2 h. Secondary antibody (IR Dye 800 conjugated anti-rabbit IgG from Rockland Immunochemicals (Gilbertsville, PA) was incubated with the membranes at a dilution of 1:5000 in blocking buffer + 0.1% Tween 20. Blots were incubated at room temperature with shaking for 45 min, covered with aluminum foil. All procedures from this step forward were kept covered to protect the blot from light. The secondary antibody was removed and filters were washed three times in PBS + 0.1% Tween 20 with water rinses in between. The blots were then washed in PBS (no Tween) for a final wash. The filters were scanned on the Odyssey Infrared Imaging system by LI-COR (Lincoln, NE). Blots were analyzed using Odyssey Application software and interfaced with Excel software (Microsoft Corp., Redmond, WA). Statistical analysis was performed using Excel and the Student's t test (two-sample, two-tail) to determine whether there were significant differences between the diets. Data were expressed as mean ± S.D. Purified recombinant mouse FGF21 was used as standard to quantitate plasma levels of FGF21.

Determination of Specific Bands by Peptide Blocking. Plasma samples have several bands that are visible after developing with antibody. Evaluation of albumin depleted db/db mouse plasma by Western blot was compared with blots developed using prebleed (naive) antisera from the same rabbit (MS2692) that was later challenged with three peptide antigens. Specificity was verified by peptide blocking using a mixture of the three peptides used to challenge the rabbit. In two separate tubes, 10 µl of antiserum from MS2692 were added. To one of the tubes, 100 µg of peptide A, 100 µg of peptide B, and 50 µg of peptide C were added in a total of 250 µl of PBS. To the other tube, 250 µl of PBS was added. Both tubes were incubated at 37°C for 1 h and then 4°C overnight. Both tubes were centrifuged at 11,000 rpm for 15 min at 4°C. The supernatants were diluted to 10 ml in 3% milk powder and 0.2% Tween 20 in PBS. This was used as antisera for duplicate Western blots. Blots were developed as described under Quantitative Western Blot Analysis.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Identification of Genes Encoding for Secreted Proteins Regulated by PPAR Agonists in Adipose Tissue. To identify novel secreted proteins regulated by PPAR agonists, global gene expression profiling experiments were conducted in adipose tissue of diabetic db/db mice treated with either vehicle or one of two structurally different PPAR agonists (rosiglitazone or COOH at 30 mg/kg) for 8 days. This treatment paradigm resulted in an improvement in plasma glucose level from approximately 450 mg/dl to approximately 150 mg/dl in these animals (data not shown). Analysis of EWAT gene expression identified 379 genes that were regulated at least 1.5-fold (with p < 0.01) by both PPAR agonists (Fig. 1A), representing roughly a third of the total signatures of each compound in terms of number of probes meeting that cutoff (data not shown). Because of the significant structural differences between these two ligands, we believe that the intersection, using these cutoffs, represents a very robust PPAR target gene signature (both directly and indirectly regulated genes) in the adipose tissue. Of the 379 genes, 33 were annotated as coding for secreted proteins from public databases (see Materials and Methods) (Fig. 1B). A subset of the 33 genes has previously been shown to be regulated at the transcript level by PPAR agonists [e.g., adiponectin (Adipoq; Combs et al., 2002), FGF21 (Fgf21; Wang et al., 2008), lipoprotein lipase (Lpl; Auwerx et al., 1996), Laminin b3 (Lamb3; Routh et al., 2002), and TNF (Tnf; Singh Ahuja et al., 2001)] (Table 1). Adiponectin, FGF21, and lipoprotein lipase were up-regulated by PPAR agonist treatment, whereas Laminin b3 and TNF were down-regulated (Table 1). Adiponectin and TNF are two well established proteins that oppositely regulate insulin sensitivity (Boyle, 2004). The expression profile in adipose tissue shows that in db/db mice, adiponectin mRNA is decreased 3.8-fold and TNF is increased 1.7-fold compared with lean db/+ animals. Pharmacological treatment with PPAR agonists increased adiponectin mRNA expression 2.7-fold and decreased TNF expression 1.5-fold, consistent with their role in regulating insulin sensitization (Table 1). Adiponectin gene expression changes also translate into changes in plasma adiponectin, which is decreased in obese and diabetic humans and increased with PPAR agonist treatment (Yu et al., 2002). Similar results have been reported in rodent models (Nawrocki et al., 2006).

View larger version (50K): [in this window] [in a new window]   Fig. 1. PPAR agonists alter the gene signature of db/db EWAT. Three hundred and seventy-nine genes were regulated by both rosiglitazone (RSG) and COOH at least 1.5-fold with p < 0.01 (represented by 452 probes on the array as shown in A). The regulation of these genes in db/db EWAT compared with lean EWAT is included (Lean versus db/db). Of these 379 PPAR responsive genes, 33 code for known secreted proteins (represented by 44 probes on the array as shown in B). Shown are logRatio values colored magenta for up-regulation or cyan for down-regulation across a 4x range (± 0.6 logRatio). Data points with p > 0.05 are shaded gray.

View this table: [in this window] [in a new window]   TABLE 1 Thirty-three secreted proteins regulated by PPAR agonists in adipose tissue of db/db mice Shown are the mRNA ratio values obtained by Agilent microarray analysis for the 33 secreted proteins significantly regulated by both PPAR agonists in EWAT of db/db mice.

Other genes identified here have been implicated in the diabetes phenotype [Retnla (Blagoev et al., 2002), IL13 (Kretowski et al., 2000), Sorbs1 (Kimura et al., 2001), Igf1 (Yakar et al., 2004), Npy (Woods et al., 1998)], in lipid metabolism [ApoC4 (Allan and Taylor, 1996), Angptl3 (Koishi et al., 2002)], and in the inflammatory response [Csf1 (Jadus et al., 1996), Ptpro (Teshima et al., 1998), Tnf (Hata et al., 2004), Ccl11 (Rokudai et al., 2006), Ccl8 (Heinemann et al., 2000), IL13 (Lloyd et al., 1997), Ccl12/Ccl2 (Moore et al., 2006), Ccl7 (Combadiere et al., 1995)]. Retnla, IL13, Sorbs1, Igf1, Apoc4, Angptl3, Ccl11, and Ccl8 were up-regulated in EWAT by PPAR agonist treatment, whereas Npy, Csf1, Ptpro, Tnf, Ccl12/Ccl2, and Ccl7 were down-regulated (Table 1).

Other genes not previously associated with PPAR agonist treatment were also identified; Fgf13, Postn, and Serpina3g mRNAs were up-regulated in EWAT by PPAR agonist treatment, whereas Ctgf, Lox, Lfng, Wisp1, Timp1, Lrg1, Adamts, Wisp2, Hp, Tnc, and Ngp mRNAs were down-regulated (Table 1). Although the individual effects of modulating the expression levels of these genes is unknown at this time, collectively they must contribute to the biological activity of PPAR agonists. Here, we focused on FGF21 for follow-up analysis as this protein has therapeutic potential to treat diabetes.

Recombinant FGF21 Expressed in P. pastoris and Production of anti-FGF21 Antibodies. Murine FGF21 was cloned and expressed in P. pastoris with a FLAG tag at its 3'-end. The protein was purified from the media using an anti-FLAG column and eluted with flag peptide (Fig. 2). This protein increased glucose uptake in 3T3-L1 adipoctyes (at 5 µg/ml) as reported previously (Kharitonenkov et al., 2005), but this activity was insulin-dependent (data not shown). To ascertain whether FGF21 was a circulating factor, polyclonal antibodies raised against 3 peptide sequences of FGF21 were used to detect circulating levels of this protein. The limit of detection for the FGF21 antibodies was 5 ng/ml determined by quantitative Western Blot analysis using recombinant P. pastoris-expressed protein as a standard. Endogenous levels of FGF21 protein in db/db plasma were detected and estimated at approximately 100 ng/ml. An expected band at 23 kDa was visible from db/db mouse plasma, and this was blocked by preincubating the antibody with peptide. This endogenous band migrated close to the recombinant FGF21 protein. A higher molecular mass band at 30 kDa was observed that was also blocked by peptide (Fig. 3). This extra band could represent a longer transcript (alternative start site), a higher molecular mass protein complex, or cross-reactivity to a related or unrelated protein. The peptides did not match any other mouse protein by BLAST search, however, and we could not identify a different start site or alternative transcript within the genomic region of Fgf21. Additional studies, such as mass spectrometry, are required to confirm the identity of this extra band.

View larger version (83K): [in this window] [in a new window]   Fig. 2. Purification of mouse FGF21. The FGF21 gene was expressed in P. pastoris as a C-terminal FLAG tag fusion. The protein was purified from the media using anti-Flag column and eluted with FLAG peptide. Eluted material was dialyzed in PBS. Aliquots from different stages of the purification process were analyzed by SDS-PAGE and silver staining. A single protein at the expected size (23 kDa plus FLAG tag) was produced. Lanes are: S, induction supernatant; F, FLAG column flow through; 1-4, FLAG column fractions 1 to 4; C, control FGF21 from a previous preparation; M, Bio-Rad Precision molecular mass markers.

View larger version (34K): [in this window] [in a new window]   Fig. 3. Plasma FGF21 from db/db mice is up-regulated by the PPAR agonist rosiglitazone. Immunodetection of pooled plasma from db/db mice treated with rosiglitazone showed an approximately 3- to 5-fold increase in FGF21 protein levels compared with vehicle treatment. Shown in A are representative Western blots with or without pretreatment with blocking peptides showing the immunoreactive 23- and 30-kDa bands representing FGF21. Included on the far right of each blot is recombinant FGF21 protein expressed and purified from P. pastoris as a C-terminal flag tag fusion and detected at approximately 25 kDa (arrow). Shown in B is the quantitation of the Western blot in A, right panel, either for the 23-kDa band alone, or the combination of the 23- and 30-kDa bands. Units are relative intensity levels from the densitometer. There was no change in plasma FGF21 between lean db/+ and db/db mice when using only the 23-kDa band, but there was a decrease in db/db compared with lean db/+ mice when using both the 23- and 30-kDa bands.

PPAR Agonists Increase Circulating Levels of FGF21 in db/db Mice. FGF21 was among the top five genes, based on degree of regulation, whose mRNA was up-regulated by PPAR agonists in adipose tissue and annotated as a secreted protein (Table 1). During the preparation of this manuscript, Wang et al. (2008) reported similar results in 3T3-L1 adipocytes. Using anti-FGF21 antibodies, pooled plasma samples from db/db mice were analyzed by Western blot analysis. Interestingly, upon treatment of db/db mice with the PPAR agonist rosiglitazone, a dose-dependent increase (approximately 3- to 5-fold) in plasma FGF21 was observed, whether only the 23-kDa band or both the 23- and 30-kDa bands were used in the quantitation (Fig. 3). There was either no difference in the plasma concentration of FGF21 between lean db/+ and db/db mice, or there was a decrease in db/db mice compared with lean db/+ mice, depending on whether only the 23-kDa band or both the 23- and 30-kDa bands were used in the quantitation, respectively (Fig. 3).

PPAR Agonists Up-Regulate FGF21 mRNA in Adipose Tissue and Adipocytes but Not in Liver. The transcript for FGF21 was significantly up-regulated in adipose tissue after PPAR agonist treatment in db/db mice (2.2- and 2.8-fold in EWAT by rosiglitazone and COOH, respectively, both with p < 0.01), lean C57BL/6 mice (2.7-fold in EWAT by rosiglitazone with p < 0.01), and lean CD1 mice (2.7 and 2.5-fold in EWAT and IWAT, respectively by rosiglitazone with p < 0.01). Taqman analysis also confirmed the db/db microarray results (data not shown). We tested several mouse strains and a second fat depot to determine whether the regulation of FGF21 by PPAR was specific to db/db EWAT or if the response was more general. To further confirm that PPAR agonists increase FGF21 expression in fat cells, we treated 3T3-L1 adipocytes with rosiglitazone or COOH and observed significant 6.6- and 8.6-fold increases in FGF21 expression, respectively, by microarray analysis (Table 2), in agreement with recent findings (Wang et al., 2008). Examination of liver gene expression of db/db, lean C57BL/6 and 129Sv mice treated with rosiglitazone showed no significant up-regulation of FGF21 mRNA (Table 2). Finally, there was also no difference in FGF21 mRNA levels in EWAT and liver of vehicle treated lean db/+ and db/db mice at baseline. This is consistent with the circulating protein levels if only the 23 kDa immunoreactive band is used to quantitate plasma FGF21 levels (Fig. 3).

PPAR Agonists Regulate FGF21 mRNA Expression Levels in Liver but Not in Adipocytes. The FGF21 transcript was up-regulated in livers of wild-type 129Sv and C57BL/6 mice after treatment with structurally distinct PPAR agonists (3.8 and 4.1-fold by fenofibrate (129Sv) and WY14643 (C57BL/6), respectively, both with p < 0.01), and this induction was abolished in PPAR knockout mice (Table 2). In fact, the basal level of this transcript in PPAR KO mice was 5 times lower compared with their wild-type 129Sv littermates (Table 2). These results are consistent with recently published reports (Badman et al., 2007; Inagaki et al., 2007). In terms of adipose tissue, PPAR agonist treatment significantly up-regulated FGF21 mRNA in EWAT of lean C57BL/6 mice (1.7-fold with p < 0.01 with WY14643) but had no effect in IWAT of lean CD1 mice (1.1-fold with p = 0.61 with fenofibrate) (Table 2). We did not profile each PPAR agonist (WY14643 and fenofibrate) in both adipose depots (EWAT and IWAT) therefore more studies are needed to confirm these results. There was also no effect on FGF21 transcript levels in 3T3-L1 adipocytes after treatment with the PPAR agonist WY14643 (Table 2).

FGF21 Is Up-Regulated by Fasting and High Fat Feeding in Both Liver and Adipose Tissue. In agreement with recently published results (Badman et al., 2007; Inagaki et al., 2007), fasting and high fat diet feeding, both ketogenic conditions, resulted in approximately 1.5- to 3.5-fold increased levels of FGF21 protein in the circulation (Figs. 4 and 5 respectively, depending on which immunoreactive bands are used in the quantitation) and increased levels of FGF21 mRNA in the liver (8.1- and 8.7-fold, respectively, both with p < 0.01) (Table 2). The level of FGF21 mRNA in adipose was also increased after fasting and high fat diet feeding, although not as robustly as in liver (2.1- and 1.8-fold, respectively, both with p < 0.01) (Table 2).

View larger version (36K): [in this window] [in a new window]   Fig. 4. Plasma FGF21 is increased upon fasting. C57BL/6 mice were fasted for 48 h, and plasma was collected and analyzed by Western blot relative to fed mice. The fasted cohort (B1-4) exhibited an increase in immunopositive bands (at 23- and 30-kDa) compared with the fed cohort (A1-4) (A). Shown in B are the quantitation of either the 23-kDa band alone or the 23- + 30-kDa bands together by densitometry showing a 1.5- (not significant) and 3-fold increase of plasma FGF21 with fasting, respectively. Data are expressed as mean ± S.D. Statistical analysis was performed by the Student's t test (two-sample, two-tail) to determine whether there were significant differences between the diets. **, P < 0.01.

View larger version (36K): [in this window] [in a new window]   Fig. 5. Plasma FGF21 is increased on high-fat diet. C57BL/6 mice were fed a high-fat diet for 4 weeks and plasma samples were collected and analyzed by Western blot relative to chow fed mice. The high-fat diet cohort (D1-4) exhibited an increase in immunopositive bands (at 23- and 30-kDa) compared with the chow diet cohort (C1-4) (A). Shown in B are the quantitation of either the 23-kDa band alone or the 23- + 30-kDa bands together by densitometry showing an approximately 3.5- and 2.5- fold increase of plasma FGF21 with high-fat diet feeding, respectively. Data are expressed as mean ± S.D. Statistical analysis was performed by the Student's t test (two-sample, two-tail) to determine whether there were significant differences between the diets. **, P < 0.01; *, P < 0.05.

	Discussion

Top Abstract Materials and Methods Results Discussion References

At the time of this work, FGF21 was not linked to PPAR regulation. We focused on FGF21, which is a newly identified molecule with therapeutic potential for diabetes for further follow-up studies (Nishimura et al., 2000; Kharitonenkov et al., 2005, 2007).

The expression of FGF21 in vivo was found to be regulated by PPAR. The protein was transcriptionally up-regulated in both liver and adipose tissue under a variety of pharmacological and physiological conditions, including fasting and high-fat diet feeding. In the case of PPAR agonist treatment of db/db mice, mRNA levels for FGF21 were elevated 2- to 3-fold in adipose tissue coincident with an elevation of plasma FGF21 levels. Circulating FGF21 was estimated to be approximately 100 ng/ml in db/db mice and increased up to 500 ng/ml when treated with PPAR agonists. This estimated increase in circulating FGF21 levels by PPAR in our experiment is within range with the reported in vitro dosing concentration of 1 µg/ml reported to elicit the insulin sensitization effects in adipocytes (Kharitonenkov et al., 2005). Whether FGF21 plays a role in the insulin sensitization activity of PPAR agonist is not known, but it is tempting to speculate that PPAR activation results in an increase in FGF21 production by the adipose tissue, which then acts locally in an autocrine fashion and/or circulates to the liver, muscle, or islets to improve insulin action. In fact, recent evidence suggests that PPAR agonism enhances the insulin sensitization effects of FGF21 in 3T3-L1 adipocytes and may suggest an amplification of the FGF21 effect to increase glucose uptake in adipose tissue via the increase in FGF21 circulating levels and receptor signaling via an undefined mechanism (Moyers et al., 2007). The fact that FGF21 mRNA was also up-regulated in adipose tissue (EWAT and IWAT) of lean mice (CD1, C57BL/6) after PPAR agonist treatment suggests that changes in glucose and other metabolites are not regulating FGF21 mRNA levels, because these are not altered by PPAR agonists in lean animals, and that FGF21 requires some other factor or condition to affect glucose/lipid control.

It is possible that the origin of circulating FGF21 is liver as a result of PPAR activation in adipose tissue, because FGF21 is normally highly expressed in the liver. However we could not detect any significant up-regulation of FGF21 mRNA in the livers of db/db, C57BL/6, or 129Sv mice after PPAR agonist treatment. These data suggest that the increase in plasma FGF21 protein after PPAR agonist treatment is of adipose origin. Consistent with our in vivo findings, Wang et al. (2008) recently reported the identification of PPAR response elements in the promoter region of FGF21 and showed that PPAR can directly up-regulate this gene in 3T3-L1 adipocytes, further supporting the hypothesis that adipose can secrete FGF21.

Basal expression levels of FGF21 mRNA in adipose and liver of db/db mice were similar to those of lean db/+ mice. This was in agreement with plasma protein levels if only the expected 23-kDa immunoreactive band was used in the quantitation. If both the 23- and 30-kDa bands were used, the circulating FGF21 protein levels would be lower in db/db mice compared with lean db/+ animals, which would be consistent with the decreased insulin sensitivity of this model. But we cannot conclude that the 30-kDa band is indeed FGF21; therefore, we can not draw any conclusion regarding circulating protein levels of FGF21 in lean db/+ versus db/db mice.

Treatment with PPAR agonists up-regulated FGF21 mRNA levels in liver and this was abolished in PPAR knockout mice, consistent with published results (Badman et al., 2007; Inagaki et al., 2007). PPAR agonists also upregulated FGF21 mRNA in adipose tissue of lean C57BL/6 mice (EWAT) but not that of lean CD1 mice (IWAT) or in 3T3-L1 adipocytes. Additional experiments are needed to confirm the regulation of adipose FGF21 by PPAR.

FGF21 mRNA levels were also significantly up-regulated by fasting and high-fat diet conditions in not only liver, as demonstrated recently (Badman et al., 2007; Inagaki et al., 2007), but also in EWAT of normal mice. In the case of a 24-h fast, liver expression was increased 8.1-fold, whereas adipose expression was increased 2.1-fold (both with p < 0.01) (Table 2). The gene expression changes were substantiated by increases in plasma protein levels as high as approximately 3-fold (Fig. 4). A similar effect was observed under 10 weeks of high-fat feeding conditions; liver and adipose gene expressions were up 8.7- and 1.8-fold, respectively (both with p < 0.01, Table 2), and plasma protein levels were increased up to approximately 3.5-fold (Fig. 5).

In conclusion, we have used transcriptional profiling to identify 33 genes coding for secreted proteins that were robustly regulated by PPAR agonists in adipose tissue of diabetic db/db mice. Several have previously been implicated in the diabetes phenotype (glucose and lipid metabolism and inflammation). Here, FGF21 was further characterized in terms of mRNA and protein expression regulation across several pharmacological and metabolic perturbations, including fasting and high-fat diet feeding. Although FGF21 was recently reported to be regulated by PPAR agonists in liver and by PPAR agonists in an adipocytic cell line, we show that this protein is regulated by PPAR agonists in adipose tissue in vivo and circulates in the plasma. We also report the first soluble expression of recombinant FGF21 using the P. pastoris expression system, and this will be useful in future biochemical analysis of this very interesting protein. Other genes in this list were described previously to be regulated by PPAR and have been implicated in the molecular pathway of this transcription factor. The identification of novel PPAR-regulated genes enhances our knowledge of the mechanisms of action of this master regulator and could point to novel drug targets for the metabolic syndrome.

	Acknowledgements

 
We thank Tomas Doebber, Margaret Wu, and Roger Meurer for the db/db, PPAR KO, and CD1 animal studies; Ching Chang and Margaret McCann for the C57BL/6 PPAR agonist study; Marie Thompson and Gino Castriota for the 3T3-L1 PPAR agonist study; Merck Research Laboratory chemists for synthesizing rosiglitazone, COOH, and fenofibrate; Robert Kleinhanz, Mark Ferguson, and Michelle Thorton for project management; Ruojing Wang for critical reading of the manuscript; and the Rosetta Gene Expression Laboratory for the microarray processing.

	Footnotes

 
E.S.M. and B.A. contributed equally to this work.

ABBREVIATIONS: PPAR, peroxisome proliferator-activated receptor; EWAT, epididymal white adipose tissue; WY14643, 4-chloro-6-(2,3-xylidino)-2-pyrimidinylthioacetic acid; COOH, 2-(2-(4-phenoxy-2-propylphenoxy)ethyl)indole-5-acetic acid; KO, knockout; IWAT, inguinal white adipose tissue; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate-buffered saline.

Address correspondence to: Kenny K. Wong, Merck Research Laboratories, P.O. Box 2000, Rahway, NJ 07065, USA. E-mail: kenny_wong{at}merck.com

	References

Top Abstract Materials and Methods Results Discussion References

 
Allan CM and Taylor JM (1996) Expression of a novel human apolipoprotein (apoC-IV) causes hypertriglyceridemia in transgenic mice. J Lipid Res 37: 1510-1518.[Abstract]

Auwerx J, Schoonjans K, Fruchart JC, and Staels B (1996) Transcriptional control of triglyceride metabolism: fibrates and fatty acids change the expression of the LPL and apo C-III genes by activating the nuclear receptor PPAR. Atherosclerosis 124: S29-S37.[CrossRef][Medline]

Badman MK, Pissios P, Kennedy AR, Koukos G, Flier JS, and Maratos-Flierl E (2007) Hepatic fibroblast growth factor 21 is regulated by PPAR alpha and is a key mediator of hepatic lipid metabolism in ketotic states. Cell Metabolism 5: 426-437.[CrossRef][Medline]

Berger JP, Akiyama TE, and Meinke PT (2005) PPARs: therapeutic targets for metabolic disease. Trends Pharmacol Sci 26: 244-251.[CrossRef][Medline]

Blagoev B, Kratchmarova I, Nielsen MM, Fernandez MM, Voldby J, Andersen JS, Kristiansen K, Pandey A, and Mann M (2002) Inhibition of adipocyte differentiation by resistin-like molecule - biochemical characterization of its oligomeric nature. J Biol Chem 277: 42011-42016.[Abstract/Free Full Text]

Boyle PJ (2004) What are the effects of peroxisome proliferator-activated receptor agonists on adiponectin, tumor necrosis factor-alpha, and other cytokines in insulin resistance? Clin Cardiol 27: 11-16.

Carley AN, Semeniuk LM, Shimoni Y, Aasum E, Larsen TS, Berger JP, and Severson DL (2004) Treatment of type 2 diabetic db/db mice with a novel PPAR gamma agonist improves cardiac metabolism but not contractile function. Am J Physiol Endocrinol Metab 286: E449-E455.[Abstract/Free Full Text]

Combadiere C, Ahuja SK, Vandamme J, Tiffany HL, Gao JL, and Murphy PM (1995) Monocyte chemoattractant protein-3 is a functional ligand for CC chemokine receptor-1 and receptor-2B. J Biol Chem 270: 29671-29675.[Abstract/Free Full Text]

Combs TP, Wagner JA, Berger J, Doebber T, Wang WJ, Zhang BB, Tanen M, Berg AH, O'Rahilly S, Savage DB, et al. (2002) Induction of adipocyte complement-related protein of 30 kilodaltons by PPAR gamma agonists: a potential mechanism of insulin sensitization. Endocrinology 143: 998-1007.[Abstract/Free Full Text]

Hata H, Sakaguchi N, Yoshitomi H, Iwakura Y, Sekikawa K, Azuma Y, Kanai C, Moriizumi E, Nomura T, Nakamura T, et al. (2004) Distinct contribution of IL-6, TNF-alpha, IL-1, and IL-10 to T cell-mediated spontaneous autoimmune arthritis in mice. J Clin Invest 114: 582-588.[CrossRef][Medline]

Heinemann A, Hartnell A, Stubbs VEL, Murakami K, Soler D, LaRosa G, Askenase PW, Williams TJ, and Sabroe I (2000) Basophil responses to chemokines are regulated by both sequential and cooperative receptor signaling. J Immunol 165: 7224-7233.[Abstract/Free Full Text]

Hughes TR, Mao M, Jones AR, Burchard J, Marton MJ, Shannon KW, Lefkowitz SM, Ziman M, Schelter JM, Meyer MR, et al. (2001) Expression profiling using microarrays fabricated by an ink-jet oligonucleotide synthesizer. Nat Biotechnol 19: 342-347.[CrossRef][Medline]

Inagaki T, Dutchak P, Zhao GX, Ding XS, Gautron L, Parameswara V, Li Y, Goetz R, Mohammadi M, Esser V, et al. (2007) Endocrine regulation of the fasting response by PPAR alpha-mediated induction of fibroblast growth factor 21. Cell Metab 5: 415-425.[CrossRef][Medline]

Jadus MR, Irwin MCN, Irwin MR, Horansky RD, Sekhon S, Pepper KA, Kohn DB, and Wepsic HT (1996) Macrophages can recognize and kill tumor cells bearing the membrane isoform of macrophage colony-stimulating factor. Blood 87: 5232-5241.[Abstract/Free Full Text]

Kharitonenkov A, Shiyanova TL, Koester A, Ford AM, Micanovic R, Galbreath EJ, Sandusky GE, Hammond LJ, Moyers JS, Owens RA, et al. (2005) FGF-21 as a novel metabolic regulator. J Clin Invest 115: 1627-1635.[CrossRef][Medline]

Kharitonenkov A, Wroblewski VJ, Koester A, Chen YF, Clutinger CK, Tigno XT, Hansen BC, Shanafelt AB, and Etgen GJ (2007) The metabolic state of diabetic monkeys is regulated by fibroblast growth factor-21. Endocrinology 148: 774-781.[Abstract/Free Full Text]

Kimura A, Baumann CA, Chiang SH, and Saltiel AR (2001) The sorbin homology domain: a motif for the targeting of proteins to lipid rafts. Proc Natl Acad Sci U S A 98: 9098-9103.[Abstract/Free Full Text]

Klimcakova E, Moro C, Mazzucotelli A, Lolmede K, Viguerie N, Galitzky J, Stich V, and Langin D (2007) Profiling of adipokines secreted from human subcutaneous adipose tissue in response to PPAR agonists. Biochem Biophys Res Commun 358: 897-902.[CrossRef][Medline]

Koishi R, Ando Y, Ono M, Shimamura M, Yasumo H, Fujiwara T, Horikoshi H, and Furukawa H (2002) Angptl3 regulates lipid metabolism in mice. Nature Genetics 30: 151-157.[CrossRef][Medline]

Kretowski A, Mysliwiec J, and Kinalska I (2000) In vitro interleukin-13 production by peripheral blood in patients with newly diagnosed insulin-dependent diabetes mellitus and their first degree relatives. Scand J Immunol 51: 321-325.[CrossRef][Medline]

Laplante M, Festuccia WT, Soucy G, Gelinas Y, Lalonde J, Berger JP, and Deshaies Y (2006) Mechanisms of the depot specificity of peroxisome proliferator-activated receptor gamma action on adipose tissue metabolism. Diabetes 55: 2771-2778.[Abstract/Free Full Text]

Lecka-Czernik B, Ackert-Bicknell C, Adamo ML, Marmolejos V, Churchill GA, Shockley KR, Reid IR, Grey A, and Rosen CJ (2007) Activation of peroxisome proliferator-activated receptor gamma (PPAR gamma) by rosiglitazone suppresses components of the insulin-like growth factor regulatory system in vitro and in vivo. Endocrinology 148: 903-911.[Abstract/Free Full Text]

Lee SST, Pineau T, Drago J, Lee EJ, Owens JW, Kroetz DL, Fernandezsalguero PM, Westphal H, and Gonzalez FJ (1995) Targeted disruption of the alpha-isoform of the peroxisome proliferator-activated receptor gene in mice results in abolishment of the pleiotropic effects of peroxisome proliferators. Mol Cell Biol 15: 3012-3022.[Abstract]

Lloyd GM, Minto AW, Dorf ME, Proudfoot A, Wells TNC, Salant DJ, and GutierrezRamos JC (1997) RANTES and monocyte chemoattractant protein-1 (MCP-1) play an important role in the inflammatory phase of crescentic nephritis, but only MCP-1 is involved in crescent formation and interstitial fibrosis. J Exp Med 185: 1371-1380.[Abstract/Free Full Text]

Moore BB, Murray L, Das A, Wilke CA, Herrygers AB, and Toews GB (2006) The role of CCL12 in the recruitment of fibrocytes and lung fibrosis. Am J Respir Cell Mol Biol 35: 175-181.[Abstract/Free Full Text]

Moyers JS, Shiyanova TL, Mehrbod F, Dunbar JD, Noblitt TW, Otto KA, Reifel-Miller A, and Kharitonenkov A (2007) Molecular determinants of FGF-21 activity-synergy and cross-talk with PPAR gamma signaling. J Cell Physiol 210: 1-6.[CrossRef][Medline]

Nawrocki AR, Rajala MW, Tomas E, Pajvani UB, Saha AK, Trumbauer ME, Pang Z, Chen AS, Ruderman NB, Chen H, et al. (2006) Mice lacking adiponectin show decreased hepatic insulin sensitivity and reduced responsiveness to peroxisome proliferator-activated receptor gamma agonists. J Biol Chem 281: 2654-2660.[Abstract/Free Full Text]

Nishimura T, Nakatake Y, Konishi M, and Itoh N (2000) Identification of a novel FGF, FGF-21, preferentially expressed in the liver. Biochim Biophys Acta 1492: 203-206.[Medline]

Rokudai A, Terui Y, Kuniyoshi R, Mishima Y, Mishima Y, Aizu-Yokota E, Sonoda Y, Kasahara T and Hatake K (2006) Differential regulation of eotaxin-1/CCL11 and eotaxin-3/CCL26 production by the TNF-alpha and IL-4 stimulated human lung fibroblast. Biol Pharm Bull 29: 1102-1109.[CrossRef][Medline]

Routh RE, Johnson JH, and McCarthy KJ (2002) Troglitazone suppresses the secretion of type i collagen by mesangial cells in vitro. Kidney Int 61: 1365-1376.[CrossRef][Medline]

Scherer PE, Williams S, Fogliano M, Baldini G, and Lodish HF (1995) A novel serum-protein similar to C1Q, produced exclusively in adipocytes. J Biol Chem 270: 26746-26749.[Abstract/Free Full Text]

Singh Ahuja H, Liu S, Crombie DL, Boehm M, Leibowitz MD, Heyman RA, Depre C, Nagy L, Tontonoz P, and Davies PJA (2001) Differential effects of rexinoids and thiazolidinediones on metabolic gene expression in diabetic rodents. Mol Pharmacol 59: 765-773.[Abstract/Free Full Text]

Steppan CM, Bailey ST, Bhat S, Brown EJ, Banerjee RR, Wright CM, Patel HR, Ahima RS, and Lazar MA (2001) The hormone resistin links obesity to diabetes. Nature 409: 307-312.[CrossRef][Medline]

Teshima S, Rokutan K, Nikawa T, and Kishi K (1998) Macrophage colony-stimulating factor stimulates synthesis and secretion of a mouse homolog of a human ige-dependent histamine-releasing factor by macrophages in vitro and in vivo. J Immunol 161: 6356-6366.[Abstract/Free Full Text]

Wang H, Qiang L and Farmer SR (2008) Identification of a domain within PPAR{gamma} regulating expression of a group of genes containing FGF21 that are selectively repressed by SIRT1 in adipocytes. Mol Cell Biol 28: 188-200.[Abstract/Free Full Text]

Wild S, Roglic G, Green A, Sicree R, and King H (2004) Global prevalence of diabetes—estimates for the year 2000 and projections for 2030. Diabetes Care 27: 1047-1053.[Abstract/Free Full Text]

Woods SC, Seeley RJ, Porte D, and Schwartz MW (1998) Signals that regulate food intake and energy homeostasis. Science 280: 1378-1383.[Abstract/Free Full Text]

Yakar S, Setser J, Zhao H, Stannard B, Haluzik M, Glatt V, Bouxsein ML, Kopchick JJ, and LeRoith D (2004) Inhibition of growth hormone action improves insulin sensitivity in liver IGF-1-deficient mice. J Clin Invest 113: 96-105.[CrossRef][Medline]

Yu JG, Javorschi S, Hevener AL, Kruszynska YT, Norman RA, Sinha M, and Olefsky JM (2002) The effect of thiazolidinediones on plasma adiponectin levels in normal, obese, and type 2 diabetic subjects. Diabetes 51: 2968-2974.[Abstract/Free Full Text]

Zhang B, Berger J, Hu EI, Szalkowski D, WhiteCarrington S, Spiegelman BM, and Moller DE (1996) Negative regulation of peroxisome proliferator-activated receptor-gamma gene expression contributes to the antiadipogenic effects of tumor necrosis factor-alpha. Mol Endocrinol 10: 1457-1466.[Abstract/Free Full Text]

This article has been cited by other articles:

				M. K. Badman, A. Koester, J. S. Flier, A. Kharitonenkov, and E. Maratos-Flier Fibroblast Growth Factor 21-Deficient Mice Demonstrate Impaired Adaptation to Ketosis Endocrinology, November 1, 2009; 150(11): 4931 - 4940. [Abstract] [Full Text] [PDF]

				K. Li, L. Li, M. Yang, H. Zong, H. Liu, and G. Yang Effects of rosiglitazone on fasting plasma fibroblast growth factor-21 levels in patients with type 2 diabetes mellitus Eur. J. Endocrinol., September 1, 2009; 161(3): 391 - 395. [Abstract] [Full Text] [PDF]

				C. Christodoulides, P. Dyson, D. Sprecher, K. Tsintzas, and F. Karpe Circulating Fibroblast Growth Factor 21 Is Induced by Peroxisome Proliferator-Activated Receptor Agonists But Not Ketosis in Man J. Clin. Endocrinol. Metab., September 1, 2009; 94(9): 3594 - 3601. [Abstract] [Full Text] [PDF]

				E. D. Berglund, C. Y. Li, H. A. Bina, S. E. Lynes, M. D. Michael, A. B. Shanafelt, A. Kharitonenkov, and D. H. Wasserman Fibroblast Growth Factor 21 Controls Glycemia via Regulation of Hepatic Glucose Flux and Insulin Sensitivity Endocrinology, September 1, 2009; 150(9): 4084 - 4093. [Abstract] [Full Text] [PDF]

				K. Mai, J. Andres, K. Biedasek, J. Weicht, T. Bobbert, M. Sabath, S. Meinus, F. Reinecke, M. Mohlig, M. O. Weickert, et al. Free Fatty Acids Link Metabolism and Regulation of the Insulin-Sensitizing Fibroblast Growth Factor-21 Diabetes, July 1, 2009; 58(7): 1532 - 1538. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: mol.108.044826v1 74/2/403    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Muise, E. S. Articles by Wong, K. K. Search for Related Content PubMed PubMed Citation Articles by Muise, E. S. Articles by Wong, K. K.