| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Lilly Research Laboratories, Lilly Corporate Center, Indianapolis, Indiana 46285
Address all correspondence and requests for reprints to: Alexei Kharitonenkov, Lilly Research Laboratories, Lilly Corporate Center, Indianapolis, Indiana 46285. E-mail: a.kharch{at}lilly.com.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Here we have addressed the hypothesis that exogenous FGF21 can mediate favorable effects on body adiposity in the context of established obesity and that it might do so via beneficial effects on energy metabolism. For this purpose, we studied diet-induced obese (DIO) mice, a specific animal model of obesity with nearly normal glucose and circulating lipids, and obese and overtly diabetic ob/ob mice. Systemic administration of FGF21 with a range of doses/delivery routes led to a profound dose-dependant weight loss, attenuated adiposity/water content, and improved hepatosteatosis in both of these animal models. We then explored the mechanisms of FGF21 action on energy balance via functional assays and at the molecular level by studying FGF21-induced changes in gene expression in various tissues and analyzing hormonal profiles in blood. This information provides novel mechanistic insights into FGF21 antiobesity and glucose/lipid-lowering effects. Taken together, our data suggest that FGF21 is a potent pharmacological agent that has the potential to effectively treat not only diabetes and dyslipidemia but obesity and fatty liver disease as well.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Protein
FGF21 was generated as described (2).
Analysis of metabolites and circulating factors
Glucose and plasma triglyceride levels were determined using Precision G Blood Glucose Testing System (Abbott Laboratories, Abbott Park, IL) and Hitachi 912 Clinical Chemistry analyzer (Roche Diagnostics, Indianapolis, IN), respectively. T3, T4, and TSH levels were measured by Ani Lytics Inc. (Gaithersburg, MD). Rodent multi-analyte profile (MAP) panel was performed by Rules Based Medicine (Austin, TX), and cytokines profiled were apolipoprotein A1, β2-microglobulin, calbindin, CD40, CD40 ligand, clusterin, C-reactive protein, cystatin-C, epidermal growth factor, endothelin1, eotaxin, factor VII, fibroblast growth factor (FGF)2, FGF9, fibrinogen, granulocyte-macrophage colony-stimulating factor, granulocyte chemotactic protein 2, GH, growth-regulated gene product 
, glutathione S-transferase- and -µ, haptoglobin, interferon-
, IgA, IL-10, IL-11, IL-12p70, IL-17, IL-18, IL-1, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, insulin, inducible protein 10, leptin, leukemia inhibitory factor, lymphotactin, monocyte chemoattractant protein 1 (MCP1), MCP3, MCP5, macrophage colony-stimulating factor, macrophage-derived chemokine, macrophage inflammatory protein 1 (MIP1), MIP1β, MIP1, MIP2, MIP3β, matrix metalloprotease 9, myoglobin, neutrophil gelatinase-associated lipocalin, oncostatin M, osteopontin, RANTES (Regulated upon Activation, Normal T Expressed, and Presumably Secreted), stem cell factor, serum glutamic oxalacetic transaminase, tissue inhibitor of metalloproteinases 1, tissue factor, TNF, thrombopoietin, vascular cell adhesion molecule 1, vascular endothelial growth factor, and von Willebrand factor.
Body composition analysis
Body composition of mice was determined using Quantitative Nuclear Magnetic Resonance analysis (ECHO MRI, 3-1 Composition Analyzer; Echo Medical Systems, Houston, TX) 1 d before initiation of treatment and on the last day of treatment. Fat-free mass was calculated by subtracting fat mass from total mass. Fat content of feces was determined using QNMR on the last day of treatment and calculated as fat mass normalized to total mass of feces (percent).
Energy homeostasis measurements
Indirect calorimeter chambers (OXYMAX; Columbus Instruments, Columbus, OH) were used to monitor various parameters of energy expenditure in treated animals over a 24-h period. Abdominally implanted telemetric temperature transmitters (model TA-F20; Data Science International, St. Paul, MN) were used to measure core body temperature every 5 min for 24 h after initiation of treatment. The beam breaker method was used to quantitate movement: ambulatory movement (new beam breaks per hour), fine movement (breaks within a beam per hour), and total movement (total beam breaks per hour).
RNA isolation, RT, and real-time quantitative PCR
RNA was isolated from tissues using TRIzol reagent (Invitrogen, Carlsbad, CA) or by homogenization of frozen samples in Lysing Matrix D shaker tubes (MP Biomedicals, Santa Ana, CA) and was reverse transcribed into cDNA using a High-Capacity cDNA Reverse Transcription Kit (PE Applied Biosystems, Foster City, CA). Reactions were performed in triplicate on an ABI Prism 7900HT (PE Applied Biosystems) and were normalized to either 36B4 mRNA or 18S rRNA. Results are expressed as fold induction relative to vehicle-treated samples. Primers and probes were from Biosearch Technologies Inc. (Novato, CA) or PE Applied Biosystems. Sequences for mouse 36B4 were forward primer 5'-GGCCCGAGAAGACCTCCTT-3', reverse primer 5'-TCAATGGTGCCTCTGGAGATT-3', and probe 5'-CCAGGCTTTGGGCATCACCACG-3'. Assays-on-Demand Gene Expression Products (PE Applied Biosystems) were as follows: ABCA1, Mm00442646_m1; ABCG5, Mm00446249_m1; ABCG8, Mm00445970_m1; ACC1, Mm01304285_m1; ACC2, Mm01204683_m1; AGRP, Mm00475829_g1; aP2, Mm00445880_m1; ApoA1, Mm00437569_m1; ApoA3, Mm00445670_m1; ATGL, Mm00503040_m1; βKlotho, Mm00473122_m1; CART, Mm00489086_m1; CCK, Mm00446170_m1; CD36, Mm00432403_m1; ChREBP, Mm00498811_m1; CIDEA, Mm00432554_m1; CPT1, Mm00550438_m1; CYP7A1, Mm00484152_m1; CYP8B1, Mm00501637_s1; D2, Mm00515664_m1; ERR, Mm00433143_m1; FAS, Mm00433237_m1; FGF21, Mm00840165_g1; FoxA2, Mm00839704_mH; FXR, Mm00436419_m1; Glucagon, Mm00801712_m1; GPAT, Mm00833328_m1; HMGCS2, Mm00550050_m1; HMGR, Mm01282499_m1; HNF1, Mm00493434_m1; HSL, Mm00495359_m1; INSIG1, Mm00463389_m1; INSIG2, Mm00460119_m1; Insulin, Mm01259683_g1; Insulin receptor, Mm00439693_m1; Leptin, Mm00434759_m1; Leptin receptor, Mm00440174_m1; LPAAT, Mm00479700_m1; LXR, Mm0443454_m1; NPY, Mm00445771_m1; PEPCK, Mm00440636_m1; PGC1, Mm00447183_m1; POMC, Mm00445771_m1; PON1, Mm00599936_m1; PPAR, Mm00440939_m1; PPAR, Mm00440945_m1; PPAR
, Mm00803186_g1; SCD1, Mm00772290_m1; SHP, Mm00442278_m1; SREBP1, Mm00550338_m1; UCP1, Mm01244860_m1; UCP2, Mm00495907_m1; and UCP3, Mm00494074_m1.
Statistical analysis
Data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA, followed by Dunnett’s multiple comparisons test. Significant differences of P < 0.05 are identified with an asterisk in the figures.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Body/tissue composition measurements showed a profound FGF21-induced reduction of total body fat mass (Fig. 1B), liver weights (Fig. 1C), and liver fat content (Fig. 1D), whereas changes in total body fat-free mass were only minor, primarily due to reduction in water content (Fig. 1B). No significant modulation in total food consumption was observed in FGF21-treated DIO mice (Fig. 1E). However, when caloric intake was normalized to animal body weights, FGF21-treated animals consumed more food (Fig. 1F).
|
).
Consistent with previously published reports (2, 3, 4, 5, 6, 7), FGF21 administration in DIO mice also led to blood glucose lowering. However, in contrast to body weight reduction which was dose dependent, the full glycemic effect was already observed at the lowest tested dose of FGF21 (Fig. 1H). No significant decrease in plasma triglycerides was detected in DIO mice (not shown).
FGF21 also induced dose-dependent weight loss in DIO mice when administered via once daily sc injections. In this case, however, approximately a 10-fold greater dose of FGF21 was required to achieve an equivalent weight reduction compared with FGF21 administration via Alzet pumps (Fig. 1, A vs. I). Nevertheless, complete glucose lowering was again achieved with the lowest employed 1 mg/kg daily sc dose of FGF21 (Fig. 1J).
Finally, FGF21 also exerted antiobesity effects in leptin-deficient ob/ob mice. When a dose of 1 mg/kg · d was delivered via continuous infusion, animals lost approximately 3 g of their body weight (Fig. 2A). As in DIO mice, this effect was primarily due to a reduction of total fat mass and water (Fig. 2B). Decreases in liver weights mirrored reductions in liver fat content (Fig. 2, C and D), and no modulation of total caloric intake was detected (Fig. 2E).
|
), had a significantly higher energy expenditure rate (Fig. 3B) during both light and dark cycles, and a had lower respiratory quotient (RQ) in the dark cycle (Fig. 3C) as well as a small but significant elevation of core body temperature beginning 4 h after initiation of treatment and lasting for approximately 10 h (Fig. 3D), compared with the vehicle-treated mice.
|
), whereas no significant changes in levels of 65 other circulating polypeptides examined, including GH, were observed (not shown).
FGF21 triggers gene expression changes
Because FGF21 is believed to exert its net effects through regulation of gene transcription (2), we examined gene expression changes in tissues from control and FGF21-dosed DIO mice after 2 wk administration. Animals were not fasted at the time of tissue collection, and expression was measured using quantitative PCR. We focused this line of research on liver, fat, and pancreas, which represent primary target tissues for FGF21 action because they express βKlotho (8) and various FGF receptors (9), the required components of FGF21 receptor machinery (10, 11, 12), and respond to FGF21 stimulation in a direct way (2, 3, 5, 10, 13, 14). Expression profiles of genes involved in the regulation of lipid and glucose metabolism as well as selected genes that have been implicated as part of the FGF21 signaling pathway were profoundly altered in all these tissues. We also profiled gene changes in hypothalamus because FGF21-dosed animals appeared to be mildly hyperphagic (Fig. 1F).
In liver (Table 1), the transcripts for peroxisome proliferator-activated receptor (PPAR) and PPAR were reduced, whereas hepatocyte nuclear factor 4 (HNF4) expression was increased at each dose of FGF21. In contrast, the mRNA levels of other profiled nuclear hormone receptors, HNF1, PPAR, liver X receptor (LXR), farnesoid X receptor (FXR), and estrogen-related receptor (ERR) were not changed.
|
coactivator 1 (PGC1) mRNAs were observed.
FGF21 also induced a modulation of various genes involved in cholesterol/bile acid metabolism. A trend toward lowering of a rate-limiting enzyme within the cholesterol synthesis pathway, 3-hydroxy-3-methylglutaryl CoA reductase (HMGR), was observed. This also coincided with an elevation in insulin-induced gene 2 (INSIG2) that controls HMGR expression; however, no change in INSIG1 was detected. The mRNA levels of cholesterol 12-hydroxylase (CYP8B1) were reduced in a dose-dependent manner. Changes in mRNAs encoding other critical molecules within these pathways, cholesterol 7-hydroxylase (CYP7A1) and small heterodimer partner (SHP), cholesterol transporters, ATP-binding cassette transporter, subfamily A, member 1 (ABCA1), ATP-binding cassette, subfamily G, member 5 (ABCG5) and 8 (ABCG8), as well as sterol regulatory element-binding protein 1 (SREBP1) and 3-hydroxy-3-methylglutaryl CoA synthase (HMGCS2) were not observed.
Scavenger receptor CD36 and paraoxonase 1 (PON1) mRNAs, both involved in the regulation of lipoprotein metabolism, were profoundly modulated; however, the levels of apolipoproteins A3 (ApoA3) and A1 (ApoA1) were unchanged.
Both insulin and leptin receptor mRNA was elevated at each dose and reached 40 and 325% increases vs. control, respectively, at the highest concentration of FGF21. No change in βKlotho mRNA was detected, whereas FGF21 transcript levels were dramatically decreased.
In white adipose tissue (WAT) (Table 2), FGF21 administration also led to alterations in the mRNA profiles of several genes involved in lipid metabolism. Among these, a dramatic up-regulation of uncoupling protein 1 (UCP1) was observed. Consistent with this observation, the transcript for PGC1 was also increased at each dose of FGF21. No change in mRNA levels of UCP2 and cell death-inducing DFFA-like effector A (CIDEA) were observed, whereas UCP3 trended downward. The transcripts for adipogenic fatty acid-binding protein 2 (aP2), carbohydrate-responsive element-binding protein (ChREBP), SCD1, DGAT1, ACC1, ACC2, PPAR, and PEPCK were all induced upon FGF21 treatment. No change in PPAR, PPAR, CPT1, and CD36 mRNAs was detected. The mRNA levels of hormone-sensitive lipase (HSL) and adipose triglyceride lipase (ATGL) were elevated.
|
In brown adipose (BAT) (Table 3), FGF21 administration led to an elevation of UCP1 and ACC2 transcripts, whereas ACC1, CPT1, and DGAT1 mRNAs were unchanged. The level of deiodinase type 2 (D2), the critical enzyme in activation of thyroid hormone function, was elevated by almost 4-fold. Nevertheless, no significant increase in TSH, but rather a reduction in the circulating levels of T3 and T4 was observed (Fig. 3F).
|
), FGF21 administration led to a significant reduction in FGF21 mRNA, whereas βKlotho transcript was increased at every dose of FGF21. No FGF21-induced changes in glucagon, insulin, and insulin receptor mRNAs were detected.
Because FGF21-dosed animals are hyperphagic (Fig. 1F), we determined whether FGF21 administration might affect hypothalamic neuropeptides involved in the regulation of appetite and satiety. Notably, agouti-related peptide (AGRP) and neuropeptide Y (NPY) mRNAs were significantly elevated in FGF21-administered animals, whereas cocaine-and amphetamine-regulated transcript (CART), proopiomelanocortin (POMC), and cholecystokinin (CCK) did not change. High levels of βKlotho mRNA (Table 3) as well as transcripts for various FGF receptors (not shown) were also expressed in hypothalamus, but their mRNA levels were not modulated by FGF21 administration.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
When administered to obese mice via continuous infusion, FGF21 induced dramatic and dose-dependent weight lowering that was primarily due to a reduction in total adiposity coupled with a minor loss of total body water (Fig. 1, A and B). The latter observation provides preclinical evidence to suggest that future attempts to develop FGF21 as a therapy for type 2 diabetes will not be hampered by edema as a side effect, a common adverse consequence seen with some antidiabetic therapies (15).
FGF21 when administered via daily bolus sc injections also induced a substantial weight-loss effect. However, the protein is significantly more efficacious when delivered via miniosmotic (Alzet) pumps (compare Fig. 1, A and I). Administration via infusion allows for the continuous presence of circulating bioactive FGF21 throughout the course of the study, whereas daily bolus sc delivery leads to a temporary spike in FGF21 blood levels because FGF21 half-life in mice is only about 1 h (7). Thus, the uninterrupted activation of FGF21 signaling is required to achieve maximal weight-lowering effect.
No evidence of decreased total caloric intake (Fig. 1E), changes in physical activity, or overt impairment of nutrient absorption was noted, suggesting that FGF21 is likely to exert its antiobesity effects primarily through the modulation of basal metabolic rate. Indeed, the systemic delivery of FGF21 in DIO mice increased oxygen consumption and overall energy expenditure (Fig. 3, A and B). Importantly, the magnitude of FGF21 effects on energy expenditure and oxygen consumption were comparable during both dark and light cycles, reflecting continuous actions of FGF21, which argues against the dependence of FGF21 actions on diurnal variations in the animals’ metabolic status.
Consistent with its effects to increase net energy expenditure, FGF21 administration also led to a small but significant elevation of core body temperature (Fig. 3D). It is important to note that this observation contrasts with an earlier report (5) where body temperature lowering was observed, which is likely due to different experimental approaches employed by Inagaki et al. (5). In that report, lowering of core body temperature was observed only in lean, starved, and therefore energy-deprived, FGF21 transgenic mice; no such effects were reported in fed animals. Our studies were conducted using obese animals with ad libitum access to food, and body temperatures were measured in the fed state.
Decreased total adiposity, improved hepatosteatosis, and modest increase in fecal fat content in FGF21-dosed obese animals indicated that FGF21’s actions were mediated by underlying regulation of lipid metabolism, as also suggested by prior studies employing different experimental models (2, 4, 5, 7, 16). This concept is further supported by our finding of lowered RQ in treated obese mice (Fig. 3C), which reflects a relative increase in lipid oxidation rate and the preferential use of fat as a fuel source.
Even though FGF21 did not alter total food consumption, FGF21 administration led to a steady reduction in body weights during the course of the study (Fig. 1A). Thus, FGF21-dosed mice revealed the development of mild hyperphagia as noted when their caloric intake was normalized to animals’ body weights (Fig. 1F). This is reminiscent of earlier observations made in FGF21 transgenic mice (2).
In harmony with the earlier reported ability of FGF21 to regulate blood glucose and its disposal in tolerance tests (2, 3, 4, 5, 6, 7), we observed normalization of hyperglycemia in DIO mice even at the lowest tested dose of FGF21 (Fig. 1H). A dose of 0.1 mg/kg · d administered via miniosmotic (Alzet) pumps achieved steady-state blood concentrations of about 7.4 ng/ml; thus, levels of about 5–10 ng/ml appear to be sufficient for nearly complete glucose normalization in DIO mice. In contrast, FGF21-induced weight loss was further potentiated with doses greater than 0.1 mg/kg · d, indicating that maximal antiobesity effects require greater FGF21 exposure (Fig. 1A).
To understand the mechanistic determinants of FGF21 effects to induce energy expenditure and promote weight loss in obese mice, we performed extensive studies using blood and tissues derived from FGF21-dosed and control animals. In these studies, we demonstrated alterations in levels of circulating hormones and key changes in gene expression that shed important new light on the mechanisms underlying FGF21 actions.
Of 67 total hormones and cytokines profiled in blood, the levels of only two polypeptides were significantly modulated by FGF21 administration (Fig. 3E). As expected based on previously reported results (2, 4, 5, 7), insulin levels were suppressed, and this effect reached its maximum at the lowest dose of FGF21. Leptin levels in these obese mice were dramatically lowered in proportion to progressive weight loss and reduced adiposity. Given the lack of effects on a variety of additional circulating factors and the likelihood that these effects on insulin and leptin could be implicated as secondary to interrelated changes in insulin/leptin sensitivity and adiposity, primary effects on hormones and cytokines do not appear to represent major mechanisms for FGF21 antiobesity actions. FGF21-dependent increases in the transcript levels of insulin and leptin receptors observed in liver (Table 1) as well as a decrease in leptin mRNA in WAT (Table 2) could also represent secondary consequences of changes in adiposity and insulin sensitivity.
FGF21 signaling requires the presence of FGF receptors and a critical cofactor, βKlotho, which is selectively expressed in tissues implicated as direct targets for FGF21 action (10, 11, 12). A careful analysis of gene expression profiles in these target organs revealed key changes that are likely to represent the primary mechanisms underlying the antiobesity effects of FGF21.
In liver, the most prominently affected gene was SCD1, a key lipogenic enzyme involved in the conversion of saturated to monounsaturated fatty acids (17); hepatic SCD1 mRNA expression was dramatically suppressed by FGF21 in a dose-dependent manner. Importantly, SCD1-null animals are lean, metabolically fit, and have lower liver fat content (18), similar to the phenotype of FGF21-treated DIO animals. Furthermore, FGF21’s ability to suppress SCD1 mRNA expression was compromised in leptin-deficient ob/ob mice compared with DIO animals with normal leptin signaling (not shown). These data further suggest that a reduction in SCD1 contributes to the antiobesity efficacy of FGF21.
In addition to a reduction in SCD1 mRNA, we observed transcript lowering for ACC1/2 as well as of GPAT and FAS and a prominent decrease in PPAR expression (Table 1). This was suggestive of a substantial net reduction in hepatic de novo lipogenesis coupled with a potential increase in long-chain fatty acid oxidation (via suppression of ACC1/2 expression, which would be predicted to decrease malonyl CoA levels). Along with PPAR lowering, we also noted modest reductions in PPAR mRNA levels. Together, these effects further implicate these nuclear receptors as important parts of FGF21 signaling pathways as also suggested by other reported results (4, 5, 14, 16, 19).
Two additional key hepatic transcriptional networks were also affected by FGF21, HNF4 and Foxa2, which were induced by up to 1.5- and 2.5-fold, respectively. HNF4 has an important role in the maintenance of glucose metabolism in liver and is also able to regulate ApoA1 gene expression (20). The latter may explain the ability of FGF21 to induce high-density lipoprotein cholesterol that was previously reported in primates (7). Foxa2 is a master regulator of glucose and lipid homeostasis, which when overexpressed as a constitutively active variant in mice can augment insulin sensitivity, normalize plasma glucose, and improve hepatosteatosis (21). Thus, Foxa2 induction in liver is also a likely contributor to the observed net effects of FGF21 actions.
Additional hepatic actions of FGF21 treatment were consistent with the proposed role that βKlotho may have in cholesterol/bile acid metabolism (22). These effects included a downward trend in HMGR mRNA levels as well as a parallel increase in INSIG2, the HMGR expression inhibitor. In addition, mRNA for CYP8B1, which catalyzes cholic acid synthesis downstream of HMGR, was lowered in a dose-dependant manner (Table 1). Such an effect might contribute to elevated fat content in the feces of FGF21-treated mice via possible changes in individual pools of bile acids. Further studies will be required to address this question.
Taken together, gene expression data derived from the analysis of livers from treated mice are indicative of specific FGF21 actions within a broad network of metabolic regulatory pathways. Effects mediated by these pathways would be predicted to reduce de novo lipogenesis, lower triglyceride secretion, augment hepatic fat utilization, enhance hepatic insulin sensitivity, and reverse hepatosteatosis.
In WAT, FGF21 induced the up-regulation of PGC1, a master regulator of mitochondrial biogenesis/oxidative metabolism and its downstream target UCP1, a critical mediator of nonshivering thermogenesis (23). In contrast to the down-regulation of lipogenic pathway genes in liver, several of the same genes were induced by FGF21 in WAT. These effects included the induction of transcripts encoding SCD1, DGAT1, ACC1, ACC2, ChREBP, aP2, and phosphoenolpyruvate carboxykinase (PEPCK). Given that some of these genes are also PPAR targets, an increase in PPAR activity can be implicated, consistent with the earlier finding (14). Because HSL gene expression in WAT was also up-regulated, a parallel increase in both lipogenesis and lipid mobilization can be inferred. Taken together with the increase in UCP1 and PGC1, it appears likely that FGF21 treatment may induce a state of increased futile cycling and energy expenditure in this, metabolically activated, adipose tissue.
Several effects similar to transcriptional readouts in WAT were observed in BAT, including the induction of UCP1, ACC1, and ACC2 transcripts. Interestingly, mRNA levels for deiodinase type 2 (D2), the enzyme controlling local thyroid hormone activation, were also substantially elevated. Taking into account a recent concept on bile acids serving as energy expenditure regulators via D2 activation (24), the potential for FGF21/βKlotho to modulate bile acid synthesis (22) (Table 1) as well as FGF21 effects on energy expenditure seen in this study (Fig. 3B), we suggest that FGF21 may coordinately regulate BAT activity via both direct and indirect actions that could involve bile acids, thyroid hormone action, and UCP1 activity. Although the mechanistic details of these interactions are currently unclear, it is important to mention that the effect of FGF21 on thyroid hormone action is likely to take place only locally, in BAT, as rather a modest reduction in circulating T3 and T4 levels was observed (Fig. 3F).
Modest hyperphagia (Fig. 1F) induced in animals by FGF21 administration was indicative of FGF21-mediated effects on central regulation of feeding. Therefore, we studied gene expression changes in hypothalamus of FGF21-dosed DIO mice. The observed elevation in mRNAs for appetite-promoting hypothalamic neuropeptides, AGRP and NPY, suggests that the observed hyperphagia may occur via a compensatory nutrient sensing-mediated process (25) in response to FGF21-induced increases in energy expenditure. Alternatively, the elevation in orexigenic AGRP and NPY in hypothalamus raises an intriguing possibility of direct FGF21 actions in brain. Indeed, FGF receptors (not shown) and βKlotho (Table 3), the required components of FGF21 receptor, are expressed in the hypothalamus, and FGF21 has recently been shown to penetrate the blood-brain barrier (26).
Finally, we also detected a prominent effect of exogenous FGF21 treatment on endogenous FGF21 mRNA expression in liver, adipose, and pancreas (6, 19). Notably, FGF21 transcript levels were substantially lowered in each of these tissues after FGF21 administration. These results indicate that FGF21 in vivo, either directly or through an indirect mechanism, is able to feed back and down-regulate its own expression.
In conclusion, this report is the first to reveal a profound antiobesity effect of exogenous FGF21 when administered systemically in murine models of obesity. The antiobesity efficacy of FGF21 primarily derives from reduced total adiposity and a modest decrease in total body water, and it occurs despite mild hyperphagia, implicating a primary impact on nutrient metabolism. Although beneficial effects on glucose homeostasis and hepatic steatosis were also observed, the dose-response relationships we established in the course of these studies also suggest that FGF21 effects on glucose-, lipid-, and body-weight-regulating functions can be achieved independently. Our data clearly indicate that FGF21 actions to decrease body weight and adiposity in obese mice were likely to be mediated by an increase in energy expenditure and preferential fat utilization as evidenced by elevated oxygen consumption, increased core body temperature, and a reduction in RQ. A careful analysis of FGF21-mediated changes in gene expression in tissues implicated as potential direct targets of FGF21 suggests that potentiation of fatty acid oxidation and suppression of de novo lipogenesis may occur in liver. In WAT, predicted effects of FGF21 include the induction of futile cycling because genes affecting lipogenesis, lipid mobilization, and uncoupling were all coordinately induced. The effects demonstrated in these studies lend further support to the concept that FGF21 might have potential therapeutic benefits in obese humans.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online August 7, 2008
Abbreviations: AGRP, Agouti-related peptide; aP2, adipogenic fatty acid-binding protein 2; BAT, brown adipose tissue; ChREBP, carbohydrate-responsive element-binding protein; CoA, coenzyme A; DGAT, diglyceride acyltransferase-1; DIO, diet-induced obese; FGF21, fibroblast growth factor 21; HNF4, hepatocyte nuclear factor 4; NPY, neuropeptide Y; PEPCK, phosphoenolpyruvate carboxykinase; PGC1, PPAR coactivator 1; PPAR, peroxisome proliferator-activated receptor ; RQ, respiratory quotient; SCD1, stearoyl-CoA desaturase 1; UCP1, uncoupling protein 1; WAT, white adipose tissue.
Received June 2, 2008.
Accepted for publication July 31, 2008.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and is a key mediator of hepatic lipid metabolism in ketotic states. Cell Metab 6:426–437-mediated induction of fibroblast growth factor 21. Cell Metab 5:415–425[CrossRef][Medline] signaling. J Cell Physiol 210:1–6[CrossRef][Medline] is a key regulator of hepatic FGF21. Biochem Biophys Res Commun 360:437–440[CrossRef][Medline] regulating expression of a group of genes containing fibroblast growth factor 21 that are selectively repressed by SIRT1 in adipocytes. Mol Cell Biol 28:188–200 coactivators, energy homeostasis, and metabolism. Endocr Rev 27:728–735This article has been cited by other articles:
 |
|
 |
|
  S. A Kliewer and D. J Mangelsdorf Fibroblast growth factor 21: from pharmacology to physiology Am. J. Clinical Nutrition, January 1, 2010; 91(1): 254S - 257S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Sanchez, A. Palou, and C. Pico Response to Carbohydrate and Fat Refeeding in the Expression of Genes Involved in Nutrient Partitioning and Metabolism: Striking Effects on Fibroblast Growth Factor-21 Induction Endocrinology, December 1, 2009; 150(12): 5341 - 5350. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Hojman, M. Pedersen, A. R. Nielsen, R. Krogh-Madsen, C. Yfanti, T. Akerstrom, S. Nielsen, and B. K. Pedersen Fibroblast Growth Factor-21 Is Induced in Human Skeletal Muscles by Hyperinsulinemia Diabetes, December 1, 2009; 58(12): 2797 - 2801. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
 |
|
 M. K. Badman, A. R. Kennedy, A. C. Adams, P. Pissios, and E. Maratos-Flier A very low carbohydrate ketogenic diet improves glucose tolerance in ob/ob mice independently of weight loss Am J Physiol Endocrinol Metab, November 1, 2009; 297(5): E1197 - E1204. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Hotta, H. Nakamura, M. Konishi, Y. Murata, H. Takagi, S. Matsumura, K. Inoue, T. Fushiki, and N. Itoh Fibroblast Growth Factor 21 Regulates Lipolysis in White Adipose Tissue But Is Not Required for Ketogenesis and Triglyceride Clearance in Liver Endocrinology, October 1, 2009; 150(10): 4625 - 4633. [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]
|
|
|
|
 |
|
 X. Wu, H. Ge, B. Lemon, J. Weiszmann, J. Gupte, N. Hawkins, X. Li, J. Tang, R. Lindberg, and Y. Li Selective activation of FGFR4 by an FGF19 variant does not improve glucose metabolism in ob/ob mice PNAS, August 25, 2009; 106(34): 14379 - 14384. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Li, Y. Bao, A. Xu, X. Pan, J. Lu, H. Wu, H. Lu, K. Xiang, and W. Jia Serum Fibroblast Growth Factor 21 Is Associated with Adverse Lipid Profiles and {gamma}-Glutamyltransferase But Not Insulin Sensitivity in Chinese Subjects J. Clin. Endocrinol. Metab., June 1, 2009; 94(6): 2151 - 2156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Nelson, J. P. Otis, and H. V. Carey A role for nuclear receptors in mammalian hibernation J. Physiol., May 1, 2009; 587(9): 1863 - 1870. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bhatnagar, H. A. Damron, and F. B. Hillgartner Fibroblast Growth Factor-19, a Novel Factor That Inhibits Hepatic Fatty Acid Synthesis J. Biol. Chem., April 10, 2009; 284(15): 10023 - 10033. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Schibler The 2008 Pittendrigh/Aschoff Lecture: Peripheral Phase Coordination in the Mammalian Circadian Timing System J Biol Rhythms, February 1, 2009; 24(1): 3 - 15. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
