Peroxisome Proliferator-activated Receptor- Co-activator 1-mediated Metabolic Remodeling of Skeletal Myocytes Mimics Exercise Training and Reverses Lipid-induced Mitochondrial Inefficiency
-
首席医学网
2008年08月15日 04:53:20 Friday
-
作者：Timothy R. Koves,Ping Li,Jie An,Takayuki Akimoto,Dorothy Slentz,Olga Ilkayeva,G. Lynis Dohm,Zhen Yan,Christopher B. Newgard, Deborah M. Muoio 作者单位：Departments of Medicine and Pharmacology & Cancer Biology, and the Sarah W. Stedman Nutrition and Metabolism Center, Duke University, Durham, North Carolina 2771 the Department of Medicine, Duke University, Durham, North Carolina 2771 and the Department of Physiology, Brody Medical School, East
加入收藏夹
【摘要】 Peroxisome proliferator-activated receptor- co-activator 1 (PGC1) is a promiscuous co-activator that plays a key role in regulating mitochondrial biogenesis and fuel homeostasis. Emergent evidence links decreased skeletal muscle PGC1 activity and coincident impairments in mitochondrial performance to the development of insulin resistance in humans. Here we used rodent models to demonstrate that muscle mitochondrial efficiency is compromised by diet-induced obesity and is subsequently rescued by exercise training. Chronic high fat feeding caused accelerated rates of incomplete fatty acid oxidation and accumulation of -oxidative intermediates. The capacity of muscle mitochondria to fully oxidize a heavy influx of fatty acid depended on factors such as fiber type and exercise training and was positively correlated with expression levels of PGC1. Likewise, an efficient lipid-induced substrate switch in cultured myocytes depended on adenovirus-mediated increases in PGC1 expression. Our results supported a novel paradigm in which a high lipid supply, occurring under conditions of low PGC1, provokes a disconnect between mitochondrial -oxidation and tricarboxylic acid cycle activity. Conversely, the metabolic remodeling that occurred in response to PGC1 overexpression favored a shift from incomplete to complete -oxidation. We proposed that PGC1 enables muscle mitochondria to better cope with a high lipid load, possibly reflecting a fundamental metabolic benefit of exercise training.
【关键词】 Peroxisome Proliferatoractivated Receptor Coactivator mediated Metabolic Remodeling Skeletal Myocytes Exercise Training Reverses Lipidinduced Mitochondrial Inefficiency
INTRODUCTION
Obesity and type 2 diabetes are two closely associated diseases that continue to affect Westernized societies at epidemic rates (1). Results from both epidemiological and animal studies suggest that the risk of developing these diseases is increased by consumption of a high fat diet (2), which has been attributed in part to increased intramuscular triacylglycerol storage and adverse changes in skeletal muscle energy metabolism (3). Moreover, mounting evidence from animal and cell-based models indicates that lipid oversupply to skeletal muscle results in functional perturbations that eventually manifest as impaired insulin action (2, 4, 5). Together, these and other reports have established a compelling connection between muscle lipid dysregulation and impaired metabolic control at both the cellular and systemic levels. Despite intense investigation, a clear understanding of the biochemical and molecular mechanisms that mediate lipid-induced muscle dysfunction has remained elusive.
Similar to high fat feeding, exercise training increases skeletal muscle supply and storage of lipid substrates (6). However, in contrast to the adverse events provoked by a high fat diet, exercise is known to promote muscle as well as whole body metabolic fitness (7). These observations suggest that physical activity somehow enhances the capacity of the muscle to tolerate and/or adapt to a high lipid load. Indeed, contractile activity leads to dramatic adjustments in oxidative fuel metabolism that are largely mediated by an increase in muscle mitochondrial biogenesis (8). Most interestingly, emergent evidence links muscle mitochondrial insufficiencies to the development of both age-related and inherited insulin resistance (9-14). These reports prompted us to consider whether compromised muscle mitochondrial performance might be a central component of diet-induced obesity. We further reasoned that the anticipated mitochondrial abnormalities caused by lipid overload might be reversed by exercise training.
Here we demonstrate that a high fat diet imposes a persistent lipid burden on muscle mitochondria, resulting in a disproportionate increase in incomplete compared with complete -oxidation. The capacity of isolated mitochondria to fully oxidize a heavy influx of fatty acid depended on factors such as fiber type and exercise training and was positively correlated with expression levels of PPAR co-activator- (PGC1),2 a promiscuous co-activator of most nuclear hormone receptors as well as several other transcription factors (15). In skeletal muscle, PGC1 has been shown to stimulate mitochondrial biogenesis via co-activation of the nuclear respiratory factor (16) and to regulate genes involved in oxidative phosphorylation through interactions with estrogen-related receptor (17). PGC1 also co-activates the peroxisome proliferator-activated receptors (PPARs), a family of lipid activated nuclear hormone receptors that play a key role in mediating adaptive regulation of muscle fatty acid oxidation (18). To test the role of PGC1 in permitting an efficient lipid-induced substrate switch, we evaluated metabolic adaptation to a high lipid supply in L6 myotubes that expressed high levels compared with low levels of the co-activator. In a manner that resembled exercise training, PGC1 coordinated the induction of -oxidative enzymes with the activation of downstream metabolic pathways (e.g. tricarboxylic acid cycle), and thereby favored a shift from incomplete to complete oxidation of lipid substrates. The implications of these findings with respect to insulin resistance and its reversal by physical activity are discussed.
EXPERIMENTAL PROCEDURES
Animal Studies-Wistar rats (300 g; Charles River Breeding Laboratories, Wilmington, MA) or male C57/Bl6J mice were housed in a temperature-controlled environment with a 12:12 h light/dark cycle and were provided ad libitum access to Purina standard chow (SC) or high fat diet (HF; Research Diets 45% fat) and water. On the day of the experiments, food was removed either 4 (fed) or 24 h (starved) before anesthesia was administered. Studies were conducted in accordance with Duke University Institutional Animal Care and Use Committee.
Exercise and Denervation-Rats (n = 7) were habituated and treadmill-trained for 4 weeks as described previously (19). Untrained rats (n = 7) served as controls. Effects of acute exercise were tested in mice (n = 10) that were subjected to 2-3-h bouts of treadmill running (0% incline, 20-28 m·min-1) separated by 1 h of rest and compared against rested controls (n = 10). Muscles were excised 12 h after the second bout of exercise. Effects of muscle inactivity were tested in mice that underwent right hind limb denervation by severing the sciatic nerve (n = 12) (20), compared against the contralateral sham-operated hind limb as a control. Muscles were excised 72 h after surgery. Mouse exercise training studies were performed in animals fed on either a SC or HF diet for 14 weeks. During the final 2 weeks of the diet, mice were kept sedentary or exercise-trained by voluntary wheel running (21). Intraperitoneal glucose tolerance tests were performed 48 h after the last bout of voluntary running in mice that were fasted overnight (22).
Mitochondrial Isolation and Fatty Acid Oxidation-Muscle mitochondria were isolated by differential centrifugation, and mitochondrial integrity was assessed as described previously (19, 23). The final mitochondrial pellet was resuspended in 1 ml of suspension buffer consisting of (in mM) 250 sucrose, 10 Tris-HCl, 1 EDTA, and 2 ATP, pH 7.4. Fatty acid oxidation was conducted using 40 µl of the final mitochondrial suspensions incubated in the presence of 160 µl of an oxidation buffer containing (in mM) 0.150 oleate, ([1-14C]oleate, at 0.5 µCi ml-1), 100 sucrose, 10 Tris-HCl, 10 KH2PO4, 100 KCl, 1 MgCl2, 1 L-carnitine, 0.1 malate, 2 ATP, 0.05 coenzyme A, 1 dithiothreitol, and 0.3% bovine serum albumin for 30 min at 37 °C (19). The reaction was terminated by the addition of 100 µl of 70% perchloric acid. 14CO2 was trapped in 200 µl of 1 N NaOH. 14C-Labeled acid-soluble metabolites (ASM) and 14CO2 fractions were quantified by liquid scintillation counting, and data were normalized to mitochondrial protein.
L6 Muscle Cells-L6 myoblasts (ATCC, Manassas, VA) were grown on 6- or 12-well collagen-coated plates in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 50 mg/ml gentamycin in a humidified incubator at 37 °C with 5% CO2 until 80% confluent. Cells were induced to differentiate into myotubes in Dulbecco's modified Eagle's medium with 2% horse serum and 50 µg/ml gentamycin (differentiation medium). On day 2, cells were treated with 20 µl (4 x 109 plaque-forming units/ml) of purified adenovirus per 500,000 cells and then 24 h later with vehicle or fatty acids. Myotubes were harvested 24 h after fatty acid treatment for isolation of protein and RNA. Fatty acid oxidation assays were performed following incubations in 500 µl of differentiation medium plus 12.5 mM HEPES, 0.5% bovine serum albumin, 1.0 mM carnitine, 100-500 µM sodium oleate, and 1.0 µCi/ml [14C]oleate (Sigma) for 3 h at 37 °C. Medium was transferred to new dishes and assayed for labeled oxidation products (CO2 and ASM) as described previously (24). Cells washed twice with ice-cold phosphate-buffered saline were harvested in 200 µl of 0.05% SDS, and lysates were stored at -80 °C for protein determination.
Western Blotting-L6 cell and gastrocnemius tissue lysates were homogenized directly in loading buffer consisting of 50 mM Tris-HCl, 10% glycerol, 0.01% bromphenol blue, 20 mM dithiothreitol, 143 mM -mercaptoethanol, and 1% SDS supplemented with phosphatase and protease inhibitor mixtures (Sigma). Protein concentrations were determined from cell and tissue extracts using the RC/DC protein assay (Bio-Rad). Twenty µg of cellular protein or 40 µg of gastrocnemius lysate were separated by SDS-PAGE using NuPAGE 4-12% BisTris gradient gel and MOPS electrophoresis buffers (Invitrogen), transferred to polyvinylidene difluoride membranes, and probed overnight with either PGC1 (1:1000; Chemicon, Temecula, CA), COXIV (1:1000; MitoSciences, Eugene, OR), or -tubulin (1:20,000; Sigma) antibodies. Proteins were visualized by horseradish peroxidase-conjugated immunoglobulin G antibodies and ECL Plus chemiluminescence detection kit and were quantified by using ImageQuant software (Amersham Biosciences).
Recombinant Adenovirus-Recombinant adenovirus encoding mouse PGC1 (AdCMV-PGC1) was a kind gift from Dr. Bruce Spiegelman. Recombinant adenoviruses encoding -galactosidase (AdCMV--gal) or green fluorescent protein (AdCMV-GFP) were used to control for nonspecific effects of virus treatment. The adenoviruses were amplified and purified using methods published previously (25).
Acylcarnitine Profiling-Ten mg of powdered muscle tissue was transferred to a chilled 1-ml Potter-Elvehjem homogenizer containing 0.3 ml of distilled, de-ionized water. Samples were subjected to 20 strokes and immediately transferred to 1.5-ml tubes and sonicated at 2.5 watts for 5 s and then subsequently processed as described previously (26). Acylcarnitines were measured by direct-injection electrospray tandem mass spectrometry, using a Micromass Quattro MicroTM system equipped with a model 2777 autosampler, a model 1525 high pressure liquid chromatography solvent delivery system, and a data system running 4.0 MassLynx software (Waters, Milford, MA).
Whole-cell Polarographic Measurements-L6 myotubes grown on 15-cm collagen-coated plates were treated with control (-galactosidase) or PGC1 recombinant adenoviruses and harvested after 48 h. Myocytes were washed twice with phosphate-buffered saline, trypsinized, and centrifuged at 800 x g. Cell pellets were washed twice and resuspended in 700 µl of phosphate-buffered saline containing 10 mM HEPES, 0.2% bovine serum albumin, and either no additional substrate, 10 mM glucose, or 100 µM palmitate. Small aliquots of the cell suspension (50-200 µl) were added to a final volume of 500 µl, and oxygen consumption was determined using an OxyTherm system (Hansatech Instruments, Norfolk, UK) calibrated at 37 °C. Uncoupled respiration was stimulated by the addition of 2 µM FCCP.
Transcriptional Profiling by Microarray and Real Time Quantitative PCR-High quality RNA was isolated using the RNeasy total RNA isolation kit (Qiagen, Valencia, CA). For muscle tissues, RNA was isolated from 10 to 20 mg of soleus, extensor digitorum longus, or gastrocnemius muscle using the RNeasy fibrous tissue kit (Qiagen, Valencia, CA). RNA quality was verified using an Agilent Bioanalyzer (Agilent Technologies, Palo Alto, CA) and quantified using RiboGreen reagent (Molecular Probes, Eugene, OR). For cell studies, L6 myotubes were grown on collagen-coated 6-well plates and treated for 48 h with control (-galactosidase) or PGC1 in the presence or absence of 500 µM sodium oleate for 24 h. RNA from day 4 myotubes was harvested during two separate experiments that were performed in triplicate. Pooled samples from each treatment group were analyzed at the Duke University DNA Microarray Center using the Rat Genome Oligo Set version 1.1 custom-printed microarray (Operon, Huntsville, AL) that was prepared with a GeneMachines OmniGrid arrayer (Genomic Solutions, Ann Arbor, MI). Samples were analyzed against a reference standard, and the raw microarray data were imported into GeneSpring 5.1 software (Silicon Genetics, Redwood City, CA). After Lowess transformation normalization, the data were screened for genes having spot intensities of "present" or "marginal." Cross-gene error model was set for deviation from 1.0, and relevant genes were limited to those having changed at least 1.5-fold. The resulting annotated list was used for GeneSpring pathway analysis to detect coordinated changes in lipid-related metabolic pathways. cDNA for real time PCR was synthesized from 1 µg of RNA using the IScript cDNA synthesis kit (Bio-Rad) in a 20-µl reaction volume and diluted 5-fold for subsequent reverse transcription-PCR. Results from the pooled-sample fluorescent microarray analysis were validated by real time quantitative PCR using a Prism 7000 with TaqMan chemistry (Applied Biosystems, Foster City, CA) to measure differentially expressed genes in individual samples. Sequence-specific primers and probes were designed using Primer Express software (Applied Biosystems, Foster City, CA). The sequences for the mouse genes were as follows: PGC1, forward, 5'-GCACCAGCCAACACTCA-3', reverse, 5'-TGGGTGTGGTTTGCTGCA-3', and probe, 5'-FAM-ACAATGAATGCAGCGGTC-3'; PGC1, forward, 5'-GCTCTCTTGGCTGCGCTTA-3', reverse, 5'-ACGATGTGGGGCTG-3', and probe, 5'-FAM-AAGACCCTGGATGACATCC. All other genes were assessed using predesigned/prevalidated FAM-labeled Assays-on-Demand from Applied Biosystems and normalized using values from a duplexed reaction with VIC-labeled 18 S endogenous control gene (Applied Biosystems, Foster City, CA).
FIGURE 1.
Fatty acid metabolism in rat muscle mitochondria. Mitochondria were isolated from deep red or superficial white rat gastrocnemius muscle (A), whole rat gastrocnemius muscles from sedentary and 4-week exercise-trained rats (B), and whole gastrocnemius muscles harvested in the ad libitum fed or 24-h starved state from rats fed on a either a SC or HF diet for 12 weeks (C). Mitochondria were incubated for 30 min at 37 °C in the presence of 150 µM [1-14C]palmitate, and radiolabel incorporation in CO2 was determined as a measure of complete oxidation. D, label incorporation into ASM was measured to assess incomplete fatty acid oxidation. Complete and incomplete oxidation rates were normalized to total mitochondrial protein. E, gastrocnemius muscles were harvested from rats that were allowed to feed ad libitum (fed) or starved 24 h after 12 weeks on either an SC or HF diet. Muscle acylcarnitine profiles were evaluated by tandem mass spectrometry and are expressed as a percent of SC-fed controls. Data represent means ± S.E. from 5 to 8 animals per group. Statistical differences between diet groups were analyzed by Student's t test or ANOVA. * indicates p ** indicates p versus HF; indicates p versus starved groups. ANOVA revealed a starvation effect (p p FAO, fatty acid oxidation.
Statistics-Data were analyzed by Student's t test or by analysis of variance (ANOVA). ANOVA was performed on data at a minimum p post hoc test was performed to evaluate differences between exercise and/or dietary treatments with JMP software version 5 (SAS Institute Inc., Cary, NC). Data are shown as means ± S.E.
FIGURE 2.
Regulation of muscle PGC1 and PGC1 mRNA expression in response to energy stress. Total RNA was harvested from rat soleus, whole gastrocnemius (gastroc), or extensor digitorum longus (Edl) muscles (A); gastrocnemius muscles from sedentary (Sed) or acutely exercised mice (B); sham-operated or denervated mouse gastrocnemius muscles (C); and soleus muscles from rats fed on a SC or HF diet for 12 weeks, harvested in the ad libitum fed or 24-h starved state (D). Gene expression was normalized using 18 S and an endogenous control, and data are presented as fold change and represent mean ± S.E. of 5-10 animals per group. Differences between treatments were analyzed by Student's t test. * indicates p
RESULTS
Physiological Regulation of Fatty Acid Oxidation in Skeletal Muscle Mitochondria-Muscles that are exercise-trained or enriched in type I (red) myofibers exhibit enhanced fatty acid oxidative capacity compared with their sedentary or type II (white) muscle counterparts. To determine whether these properties are due strictly to differences in mitochondrial content or, alternatively, could reflect distinctions in mitochondrial performance, we evaluated [1-14C]palmitate oxidation in isolated rat muscle mitochondria. Most strikingly, we found that oxidation rates (expressed per g mitochondrial protein) in mitochondria from the deep red gastrocnemius were 8-fold higher than those from the superficial white gastrocnemius (Fig. 1A). Similarly, exercise training, which causes a shift in muscle fiber composition, increased mitochondrial fatty acid oxidative capacity 2-fold over the sedentary condition (Fig. 1B).
FIGURE 3.
Adenovirus-mediated PGC1 overexpression in rat L6 myocytes. Rat L6 myocytes were treated with recombinant adenoviruses encoding -galactosidase (-gal), green fluorescent protein (GFP), or PGC1 (4 x 109 plaque-forming units/ml), and protein expression of the transgene was measured 48 h later. A shows high efficiency transduction of -galactosidase (left) and green fluorescent protein (right) in differentiated myotubes. PGC1 mRNA (B) and protein (C) levels increased dose-dependently with increasing viral titer. D, COX IV protein levels in no virus (NVC) and -galactosidase-treated control cells compared against cells treated with increasing doses of rAd-CMV-PGC1.
Exercise-trained and/or type I skeletal muscles are also characterized by increased lipid uptake and elevated intramuscular triacylglycerol storage (6, 27). Thus, lipid-induced increases in PPAR transcriptional activity could contribute to enhanced mitochondrial -oxidative capacity. In order to assess whether a change in fatty acid supply would be sufficient to alter mitochondrial lipid metabolism, we next evaluated muscle mitochondria from rats that were starved 24 h or fed a high fat diet for 10 weeks. Similar to previous studies by our laboratory (26, 28) and others (29), both manipulations increased circulating nonesterified free fatty acids by 2-fold and coincidentally increased mRNA expression of several known PPAR target genes (not shown). However, unlike the exercise and fiber type models, neither of the nutritional maneuvers enhanced the capacity of the muscle mitochondria to fully oxidize palmitate to CO2 (Fig. 1C). In these studies we also measured label incorporation into chain-shortened ASM, which provide an index of incomplete -oxidation. Compared with the SC-fed condition, both starvation and HF feeding increased 14C label incorporation into the ASM fraction by 36 and 27%, respectively (Fig. 1D). ASM production was greatest in muscle mitochondria from HF rats harvested in the starved condition.
In Vivo Metabolic Analysis by Acylcarnitine Profiling-To gain further insight into diet-induced changes in mitochondrial function, we employed tandem mass spectrometry to profile mitochondrially derived acylcarnitine esters in muscles from rats fed either a SC or a HF diet. The acylcarnitine analysis provides a comprehensive snapshot of in vivo substrate flux through specific steps of the -oxidative spiral and is thus commonly used as a diagnostic tool to detect inborn metabolic errors (30). Here and in previous studies we have found this to be an effective method for screening diet and/or obesity-related metabolic abnormalities (26).
In rats fed the SC diet, starvation increased several medium and long chain acylcarnitine intermediates, consistent with increased fatty acid delivery and high rates of -oxidation that occur in the fasted condition (Fig. 1G). In muscles from HF-fed animals, most lipid-derived (even chain) acylcarnitine intermediates were persistently elevated (compared with the SC-fed group), but increased only modestly during the fed-to-starved transition. In contrast, muscle levels of isovaleryl-carnitine (C5), which is derived from the catabolism of leucine, were higher in the SC compared with the HF group but were unaffected by fasting. The lowering of this metabolite by the HF diet most likely reflects a mitochondrial substrate switch from branched chain amino acids to fatty acids. In general, the nutritionally induced changes in the acylcarnitine profile are consistent with the notion that starvation and HF feeding increase mitochondrial rates of incomplete -oxidation.
Transcriptional Regulation of PGC1 and PGC1 in Skeletal Muscle-The results presented in Fig. 1 suggested that under some circumstances PPAR activation alone is insufficient to enhance the potential of muscle mitochondria to fully oxidize lipid substrates. As we considered other factors that may be important in regulating mitochondrial function, the transcriptional co-activators PGC1 and PGC1 emerged as attractive candidates. Both PGC1 isoforms have been implicated as master regulators of mitochondrial biogenesis and important modulators of PPAR activity. In liver, heart, and brown adipose tissue, PGC1 and PGC1 are transcriptionally regulated by various nutritional and/or energy stresses (31). We therefore proceeded to investigate skeletal muscle mRNA expression of both PGC1 isoforms under a variety of physiological conditions (Fig. 2). Expression of PGC1 was 4.5-fold greater in soleus (red) than in gastrocnemius (mixed) or extensor digitorum longus (white) muscle, whereas conversely, PGC1 mRNA was most abundant in the extensor digitorum longus muscle (Fig. 2A). An acute bout of exercise increased PGC1 expression 2.2-fold in the gastrocnemius muscle (Fig. 2B), whereas denervation (a model of inactivity) provoked a 50% decrease in expression levels (Fig. 2C). In contrast, neither exercise nor denervation altered expression of PGC1 (Fig. 2, B and C). Chronic high fat feeding resulted in a marked decrease (80%) in soleus muscle PGC1 but no significant change in PGC1 (Fig. 2D). Starvation also tended to decrease PGC1. Thus, transcriptional regulation of PGC1, but not PGC1, correlated with changes in mitochondrial fatty acid oxidation capacity (Fig. 1). This expression pattern indicates that PGC1 and PGC1 are differentially responsive to energy stress and suggests that the isoform plays a primary role in mediating adaptive changes in muscle mitochondrial function.
Overexpression of PGC1 in L6 Muscle Cells-Data from our animal models implied that PGC1 may permit more efficient adaptation to a high lipid environment. To test this hypothesis, we used recombinant adenovirus to increase the normally low levels of PGC1 expressed in rat L6 myotubes. Subsequent experiments then evaluated the transcriptional and metabolic consequences of a 24-h fatty acid challenge in cells expressing high compared with low PGC1 activity. Preliminary studies using recombinant adenoviruses encoding either -galactosidase or green fluorescent protein confirmed high efficiency transduction of differentiated myotubes (Fig. 3A). Likewise, both PGC1 mRNA and protein levels increased dose-dependently with an increasing titer of rAd-CMV-PGC1 (Fig. 3, B and C). Similar to previous reports, PGC1 overexpression caused a marked induction of the COXIV protein (Fig. 3D), consistent with an increase in mitochondrial biogenesis.
Microarray Analysis of PGC1-mediated Reprogramming of Muscle Cells-We used DNA microarray analysis to gain a more comprehensive perspective of how PGC1 modulates transcriptional responses to lipid exposure (TABLE ONE). Treatment of myocytes with high fatty acid resulted in predictable increases in the expression of several classic PPAR-targeted genes involved in fatty acid catabolism (PDHK4, 2.7-fold; UCP2, 2.2-fold; AAC2, 1.7-fold; and ECH1, 1.7-fold). Fatty acid exposure alone produced little or no effects on transcripts associated with the tricarboxylic acid cycle or the electron transport chain. In contrast, PGC1 overexpression not only activated several -oxidative and lipid trafficking genes but also triggered the global induction of genes involved in the tricarboxylic acid cycle, the electron transport chain, oxidative phosphorylation, and protection against oxidative stress. In general, overexpression of PGC1 potentiated lipid-mediated activation of classic PPAR target genes.
TABLE ONE
Pathway analysis of fatty acid-induced gene activation in L6 myocytes expressing high or low PGC1a
Rat L6 myocytes were treated with recombinant adenoviruses encoding -galactosidase (-gal) or PGC1 on differentiation day 2. Twenty four h after addition of virus, cells were incubated an additional 24 h in control differentiation medium with bovine serum albumin or with the addition of 500 µM oleate (FA). Pooled RNAs from replicate experiments were analyzed by the Duke University DNA Microarray Center using the Rat Genome Oligo Set version 1.1 custom-printed microarray. Pathway analysis was performed using GeneSpring 5.1 software. Gene expression data are presented as fold change relative to -galactosidase control cells grown in the absence of fatty acid.
FIGURE 4.
Microarray validation by real time PCR. Rat L6 myocytes were treated with recombinant adenoviruses encoding -galactosidase (-gal) or PGC1 on differentiation day 2. Twenty four h after addition of virus, cells were incubated an additional 24 h in differentiation medium without (control) or with the addition of 500 µM oleate (FA). Microarray results were validated by real time quantitative PCR performed on individual samples. A, classic PPAR target genes, uncoupling proteins (UCP2 and UCP3), and pyruvate dehydrogenase kinase 4 (PDK4); B, tricarboxylic acid cycle enzymes; aconitase and citrate synthase (CS); C, enzymes of -oxidation; ACAA2, acetyl-CoA acyltransferase 2; CAT, carnitine acylcarnitine translocase; CPT1, carnitine palmitoyltransferase 1; MCAD, medium chain acyl-CoA dehydrogenase; VLCAD, very long chain acyl-CoA dehydrogenase. D, cytosolic enzymes, glutamate-oxaloacetate transaminase (GOT) and MDH. Gene expression was normalized to 18 S as an endogenous control, and data are presented as fold change (mean ± S.E.) relative to -galactosidase-treated control cells. Experiments were performed in triplicate, and differences between -galactosidase and PGC1 groups were analyzed by Student's t test.
Expression levels of a select set of metabolic genes were then validated by real time quantitative PCR (Fig. 4). These assays demonstrated strong agreement with the microarray data and confirmed PGC1-mediated induction of genes encoding mitochondrial enzymes such as aconitase and citrate synthase, as well as cytosolic enzymes such as malate dehydrogenase (MDH1) and glutamate oxaloacetate transaminase (GOT1). Consistent with data presented in TABLE ONE, real time quantitative PCR revealed subsets of lipid-responsive genes that could be classified into three distinct patterns of regulation: 1) genes that were similarly responsive to fatty acid in the -galactosidase compared with PGC1 overexpressing cells (UCP2 and UCP3); 2) genes that responded to fatty acid only when PGC1 was overexpressed (medium chain acyl-CoA dehydrogenase, very long chain acyl-CoA dehydrogenase, and carnitine acylcarnitine translocase); and 3) genes for which a response to fatty acid was present in control cells but then further potentiated by high expression of PGC1 (carnitine palmitoyltransferase 1, acetyl-CoA acyltransferase 2, and PDK4).
PGC1 Overexpression Enhances Complete Fatty Acid Oxidation-We next investigated the impact of increased PGC1 activity on metabolic responses to a high lipid load. Similar to muscle mitochondria from starved or high fat fed rats (Fig. 1), L6 myocytes exposed to increasing fatty acid concentrations exhibited disproportionate increases in the rates of incomplete (ASM) relative to complete (CO2) -oxidation. In both sets of control cells, [14C]oleate oxidation to CO2 increased only 1.6-fold when the fatty acid concentration was raised from 100 to 500 µM (Fig. 5A), whereas label incorporation into ASM increased 2.2-fold (Fig. 5B). In comparison, myocytes treated with rAd-CMV-PGC1 displayed 2.2-2.6-fold increases in 14CO2 production, but only 1.8- to 1.6-fold increases in labeling of ASM. The relationship between incomplete and complete fatty acid oxidation was quantified by calculating the ratio of radiolabeled ASM relative to CO2 (Fig. 5C). In control cells treated with 100 µM oleate, label incorporation into ASM was 1.3-fold greater than into CO2. This ratio increased to 1.8 when the oleate concentration was raised to 500 µM. Notably, however, under conditions of high PGC1 activity, production of 14CO2 and 14C-labeled ASM was equal (ratio of 1.0) at both low and high fatty acid concentrations. Thus, PGC1 modified mitochondrial function in a manner that favored tighter coupling between -oxidation and tricarboxylic acid cycle flux. Oximetry was used as an alternative measure of tricarboxylic acid cycle activity. In the presence of both lipid (palmitate) and non-lipid (glucose) substrates, oxygen consumption was 2.5-fold greater in myotubes transduced with rAd-CMV-PGC1 compared with rAd-CMV--gal (Fig. 5D). Likewise, in the presence of FCCP, an uncoupling agent that permits assessment of maximal respiratory capacity, oxygen consumption was 80% greater in myotubes that overexpressed PGC1.
FIGURE 5.
PGC1 enhances complete oxidation of fatty acids. Rat L6 myocytes were treated with recombinant adenoviruses encoding -galactosidase (-gal) or PGC1 on differentiation day 2 and were compared against a no virus control (NVC) group. Forty eight h after addition of virus, cells were incubated 3 h with 100-500 µM [14C]oleate. Fatty acid oxidation was determined by measuring 14C label incorporation into CO2 (A) and ASM (B), representing complete and incomplete oxidation, respectively. C, the relationship between incomplete and complete fatty acid oxidation was expressed as a ratio of label incorporated into ASM divided by labeling of CO2. D, oximetry experiments were performed 48 h after addition of virus in L6 cells that were trypsinized and transferred to an oximetry chamber. Oxygen consumption was measured in the presence of 400 µM palmitate, 25 mM glucose, or the uncoupling agent FCCP. Data represent the mean ± S.E. from three to four experiments. Differences between groups were analyzed by ANOVA and Student's t test, * indicates p to NVC and -galactosidase treatments; indicates p
Metabolic Analysis of Exercise Training by Acylcarnitine Profiling-Results from our cell-based experiments suggested that exercise-induced activation of PGC1 might alleviate mitochondrial inefficiencies caused by HF feeding. To test this possibility, we performed a chronic (14-week) HF feeding study in which half of the mice in each group were exercise-trained during the final 2 weeks of the diet. Consistent with our studies in rats (Fig. 2), the HF diet decreased PGC1 expression 50% in mouse gastrocnemius muscle. In this experiment muscles were harvested from overnight-fasted animals that had been rested for 48 h, thereby minimizing confounding effects of acute activity. Although an acute exercise bout increased muscle PGC1 mRNA abundance (Fig. 2), we did not observe a training effect on transcript levels in the mice fed the SC diet (Fig. 6A). This was not surprising because several exercise-induced transcripts, including PGC1, are transiently increased with each exercise bout (32, 33). More interestingly however, the exercise regimen reversed the HF diet-mediated suppression of PGC1 and increased mRNA expression by 2.8-fold over sedentary mice that were fed the HF diet. Exercise tended to increase COXIV protein abundance, although these changes did not reach statistical significance (Fig. 6B).
Similar to the results in Fig. 1, preliminary studies in mice confirmed a robust fasting-induced rise in circulating NEFA and a coincident increase in muscle acylcarnitine levels (Fig. 7, A and B). Accordingly, we chose the overnight-fasted condition, representing a state of high mitochondrial fatty acid influx, to evaluate the impact of endurance training on muscle acylcarnitine profiles. Resembling the data in Fig. 1G, when muscles were harvested from starved mice most short and medium acylcarnitines were similarly abundant regardless of the diet (Fig. 7C). However, the HF diet did increase the C18:0 and C18:1 acylcarnitine species. The exercise training regimen lowered most lipid-derived acylcarnitines but did not affect the leucine metabolite (C5), suggesting lipid-specific adaptations in mitochondrial fuel metabolism. Comparable with the interplay that we observed between PGC1, fatty acids, and gene regulation in cultured muscle cells (Fig. 4), the most robust adjustments in fatty acylcarnitine metabolites occurred in exercise-trained mice fed the HF regimen. Thus, like PGC1 overexpression in L6 myocytes, exercise training appeared to optimize mitochondrial processing of excess lipid substrate.
Finally, Fig. 7D shows the results of an intraperitoneal glucose tolerance test that was performed 48 h after the last exercise bout. Similar to previous reports (26), animals fed the HF diet increased their body weight and displayed impaired glucose tolerance. Although the short term exercise intervention did not affect body weight, food intake, or circulating NEFA (not shown), it was sufficient to normalize the glucose tolerance test, even with continued feeding of the HF diet. Thus, notably, exercise-induced lowering of lipid-derived muscle acylcarnitines was accompanied by the reversal of HF diet-induced glucose intolerance.
DISCUSSION
PPARs,, and comprise a family of lipid-activated nuclear hormone receptors that regulate fatty acid homeostasis. Numerous reports indicate that PPAR-targeted fatty acid oxidative genes are induced by physiological manipulations that raise circulating NEFA, such as exercise, starvation, and high fat feeding (18, 28, 29). Still, data that address the direct impact of these transcriptional adaptations on muscle mitochondrial performance are lacking. In our study both overnight starvation and chronic HF feeding of rats produced the anticipated induction of known PPAR target genes, but without comparable changes in the potential of muscle mitochondria to completely oxidize fatty acid substrate. In contrast, both maneuvers caused a disproportionate increase in mitochondrial rates of incomplete -oxidation. A similar phenomenon occurred when muscle cells were exposed to increasing doses of fatty acid. In both the animal and cell culture models, the inability of mitochondria to increase complete fatty acid oxidation in response to lipid exposure was associated with low expression of PGC1. Conversely, raising PGC1 activity, in vivo by exercise training or in cultured myocytes by rAd-mediated gene delivery, favored complete fatty acid oxidation to CO2. Although previous studies have established a clear role for PGC1 in regulating mitochondrial biogenesis (15, 16, 34), our report now demonstrates a strong functional link between PGC1 and efficient mitochondrial oxidation of fatty acid.
FIGURE 6.
Effects of high fat diet and exercise on muscle PGC1 mRNA levels in mice. Mice were fed on SC or HF diets for 14 weeks. During the final 2 weeks of the diet, half of the mice in each group were kept sedentary (Sed) or exercise-trained (Ex) by voluntary running wheel. Gastrocnemius muscles from five to six mice per group were harvested 48 h after the last exercise bout following a 24-h fast. Training-induced changes in PGC1 mRNA levels normalized to 18 S (A) and COX IV protein expression normalized to -tubulin (B). Values are mean ± S.E., and statistical differences between groups were analyzed by ANOVA. * indicates p indicates p
FIGURE 7.
Metabolic analysis of exercise training in mice fed a standard or high fat diet. The effects of 24 h of starvation (Starved) compared with ad libitum feeding (Fed) on serum NEFA (A) and gastrocnemius acylcarnitine profile (B) (nmol/mg protein) were tested in mice fed on SC. Subsequent analyses were performed in mice fed on SC or HF diets for 14 weeks. During the final 2 weeks of the diet, half of the mice in each group were kept sedentary (Sed) or exercise-trained (Ex) by running wheel. Intraperitoneal glucose tolerance tests and harvest of gastrocnemius muscles were executed 48 h after the last exercise bout, following a 24-h starvation. Results show training-induced changes in muscle acylcarnitine profile, expressed as a percent of SC-fed controls (C) and systemic glucose tolerance tests (D). Values are mean ± S.E. for 5-6 mice per group. Statistical differences between groups were analyzed by two-way ANOVA. * indicates p ** indicates p p p
Our results also show that isolated mitochondria from red versus white and trained versus untrained muscle exhibited greater potential to fully oxidize lipid substrate. These observations imply that fiber type and exercise-related adjustments in fuel homeostasis are mediated not only by an increase in mitochondrial number but also by functional changes in the resident mitochondrial population. Consistent with previous reports (16, 35), we found that exercise training and red fiber type are associated with increased mRNA expression of PGC1, but no changes in PGC1. We therefore surmised that the isoform might be primarily responsible for maintaining a mitochondrial population that can accommodate a high influx of lipid fuel. Supporting this notion, treatment of myocytes with rAd-PGC1 triggered a broad program of genes involved in -oxidative production of acetyl-CoA as well as its subsequent flux into the tricarboxylic acid cycle. PGC1-mediated transcriptional reprogramming coincided with a shift from incomplete to complete oxidation of fatty acids. Together, these results suggest that efficient substrate switching from glucose to fatty acid not only requires PPAR-mediated induction of -oxidative enzymes but also the coordinated regulation of downstream metabolic pathways that are not directly targeted by the PPAR system.
FIGURE 8.
Schematic of PGC1-mediated remodeling of muscle lipid metabolism. In the presence of low PGC1 activity, fatty acid influx and PPAR-mediated activation of target genes (shown in blue) promote incomplete -oxidation. Exercise and/or PGC1 coordinates ligand-induced PPAR activity with remodeling of downstream metabolic pathways (shown in red), thereby favoring complete fatty acid oxidation (FAO) to CO2. ACO, aconitase; AMPK, AMP-activated kinase; ANT, adenine nucleotide transporter; ATPS, ATP synthase; CPT-CAT, carnitine palmitoyltransferase system; CS, citrate synthase; ETC, electron transport chain; FABP, fatty acid-binding proteins; GOT, glutamate-oxaloacetate transaminase; ICD, isocitrate dehydrogenase; PC, pyruvate carboxylase.
PDK4, which encodes an enzyme that phosphorylates and inhibits the PDH complex, ranked highest among the metabolic transcripts that were induced by PGC1. We and others have shown that this gene is highly responsive to both exercise and starvation (28). Inactivation of PDH by PDK4-mediated phosphorylation not only promotes lipid oxidation but also serves to spare pyruvate for use as an anaplerotic substrate. Similarly, increased PGC1 activity yielded higher mRNA levels of both the mitochondrial and cytosolic isoforms of GOT and MDH, enzymes that play important roles in shuttling tricarboxylic acid cycle intermediates and reducing equivalents in and out of the mitochondria (Fig. 8). Thus, PGC1 appears to facilitate fatty acid catabolism by activating anaplerotic reactions that provide the carbon skeletons necessary to sustain high tricarboxylic acid cycle activity.
Consistent with other reports showing that GOT1 is increased by exercise training and expressed more abundantly in red compared with white muscle (36-38), we found that mRNA levels of this enzyme in L6 myocytes were increased over 20-fold by PGC1 overexpression. GOT1 catalyzes the cytosolic conversion of aspartate to oxaloacetate, which represents the rate-limiting step of the malate-aspartate shuttle (36). The conversion of oxaloacetate to malate is then catalyzed by MDH1 in an NADH-consuming reaction. Malate can be transported into the mitochondria where the opposite reaction then delivers NADH to the respiratory chain. This system plays a principal role in permitting oxidation of cytosolic reducing equivalents that are produced by glycolysis and/or via the oxidation of lactate, both of which increase in response to muscle contraction. Most importantly, however, the malate-aspartate carrier system operates as a reversible shuttle that also allows mitochondrial export of reducing equivalents. This process is likely to occur during transitions from exercise to rest, when fatty acid supply and -oxidative production of NADH exceeds ATP demand. A high NADH/NAD ratio slows oxidative phosphorylation at several steps. Thus, by dissipating the rapid intramitochondrial accumulation of reducing equivalents that occurs upon cessation of contractile activity, the malate shuttle might play a key role in preventing the complete shutdown of oxidative metabolism.
A seminal observation presented here is that HF diet-induced insulin resistance in rodents occurred in association with decreased muscle PGC1 expression and persistent elevations in intramuscular acylcarnitines, metabolic by-products of incomplete fatty acid oxidation. Conversely, exercise training, which is thought to increase PGC1 activity via both transcriptional and post-translational mechanisms (43, 44), reversed lipid-induced glucose intolerance in conjunction with the lowering muscle acylcarnitine levels. Perhaps these results move us closer to solving a long standing paradox in exercise and diabetes research, i.e. both red and exercise-trained muscles are more insulin-sensitive than their white or untrained counterparts, despite higher lipid influx and storage (6, 27, 39). Our findings suggest that increased PGC1 activity and/or enhanced mitochondrial efficiency may protect against lipid-induced insulin resistance. Similarly, emerging evidence from human studies links decreased PGC1 activity, and accompanying reductions in the expression of genes involved in oxidative phosphorylation, to the etiology of type 2 diabetes (9, 11, 40). Moreover, both inherited and acquired insulin resistance has been associated with perturbations in the metabolic and morphologic properties of skeletal muscle mitochondria (11, 14, 41, 42).
Still unanswered, however, is whether or not PGC1 activation alone, in the absence of exercise, would be sufficient to confer protection against diabetes. Although muscle-specific overexpression of PGC1 in transgenic mice results in mitochondrial proliferation and increased expression of genes involved in oxidative phosphorylation (16), the impact of this manipulation on whole body glucose homeostasis has not been reported. Most surprisingly, PGC1 null mice are paradoxically lean and resistant to diet-induced obesity (45). However, the phenotype of these animals was complicated by dramatic abnormalities in the brain and centrally mediated hyperactivity; thus, a clearer understanding of the metabolic ramifications imparted by low muscle PGC1 levels will depend on a tissue-specific knock-out.
In summary, our conceptual model predicts that conditions of low PGC1 activity, when combined with excess lipid and tonic PPAR activation, provoke a mitochondrial disconnect between -oxidation and tricarboxylic acid cycle flux, thereby triggering the production and/or accumulation of -oxidative intermediates. Although the pathophysiological implications of incomplete fat oxidation are yet unexplored, it is tempting to speculate that this event might contribute to mitochondrial malfunction and whole body metabolic dysregulation. This possibility is supported by the finding that muscle-specific overexpression of PPAR in transgenic mice caused both local and systemic glucose intolerance, in association with marked induction of several lipid-oxidative genes (46). The diabetic phenotype of these animals was reversed by administration of the carnitine palmitoyltransferase 1 inhibitor, oxfenicine, further implicating a role for excessive -oxidation. Moreover, several reports have shown that PPAR null mice are protected against lipid-induced diabetes despite greater susceptibility to obesity (47, 48). Likewise, we have shown that genetic reversal of HF diet-induced insulin resistance in rats correlates with decreased muscle production of -oxidative by-products (26). Studies to investigate potential links between discordant regulation of PPAR and PGC1, incomplete -oxidation, and glucose intolerance now await further investigation.
ACKNOWLEDGMENTS
We are grateful to Dr. Bruce Spiegelman for generously providing us with the recombinant adenovirus encoding PGC1.
【参考文献】
Zimmet, P., Alberti, K. G., and Shaw, J. (2001) Nature 414, 782-787
Lichtenstein, A. H., and Schwab, U. S. (2000) Atherosclerosis 150, 227-243
Shulman, G. I. (2000) J. Clin. Investig. 106, 171-176
McGarry, J. D. (2002) Diabetes 51, 7-18
Perseghin, G., Petersen, K., and Shulman, G. I. (2003) Int. J. Obes. Relat. Metab. Disord. 27, Suppl. 3, S6-S11
Goodpaster, B. H., He, J., Watkins, S., and Kelley, D. E. (2001) J. Clin. Endocrinol. Metab. 86, 5755-5761
Astrup, A. (2001) Public Health Nutr. 4, 499-515
Adhihetty, P. J., Irrcher, I., Joseph, A. M., Ljubicic, V., and Hood, D. A. (2003) Exp. Physiol. 88, 99-107
Patti, M. E., Butte, A. J., Crunkhorn, S., Cusi, K., Berria, R., Kashyap, S., Miyazaki, Y., Kohane, I., Costello, M., Saccone, R., Landaker, E. J., Goldfine, A. B., Mun, E., Defronzo, R., Finlayson, J., Kahn, C. R., and Mandarino, L. J. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 8466-8471
Patti, M. E. (2004) Curr. Opin. Clin. Nutr. Metab. Care 7, 383-390
Lowell, B. B., and Shulman, G. I. (2005) Science 307, 384-387
Iossa, S., Mollica, M. P., Lionetti, L., Crescenzo, R., Tasso, R., and Liverini, G. (2004) Diabetes 53, 2861-2866
Schrauwen, P., and Hesselink, M. K. (2004) Diabetes 53, 1412-1417
Petersen, K. F., Dufour, S., Befroy, D., Garcia, R., and Shulman, G. I. (2004) N. Engl. J. Med. 350, 664-671
Puigserver, P., and Spiegelman, B. M. (2003) Endocr. Rev. 24, 78-90
Lin, J., Wu, H., Tarr, P. T., Zhang, C. Y., Wu, Z. D., Boss, O., Michael, L. F., Puigserver, P., Isotani, E., Olson, E. N., Lowell, B. B., Bassel-Duby, R., and Spiegelman, B. M. (2002) Nature 418, 797-801
Mootha, V. K., Handschin, C., Arlow, D., Xie, X., St. Pierre, J., Sihag, S., Yang, W., Altshuler, D., Puigserver, P., Patterson, N., Willy, P. J., Schulman, I. G., Heyman, R. A., Lander, E. S., and Spiegelman, B. M. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 6570-6575
Gilde, A. J., and Van Bilsen, M. (2003) Acta Physiol. Scand. 178, 425-434
Koves, T. R., Noland, R. C., Bates, A. L., Henes, S. T., Muoio, D. M., and Cortright, R. N. (2005) Am. J. Physiol. 288, C1074-C1082
Jones, J. P., Tapscott, E. B., Olson, A. L., Pessin, J. E., and Dohm, G. L. (1998) J. Appl. Physiol. 84, 1661-1666
Waters, R. E., Rotevatn, S., Li, P., Annex, B. H., and Yan, Z. (2004) Am. J. Physiol. 287, C1342-C1348
Almind, K., and Kahn, C. R. (2004) Diabetes 53, 3274-3285
Kim, J. Y., Koves, T. R., Yu, G. S., Gulick, T., Cortright, R. N., Dohm, G. L., and Muoio, D. M. (2002) Am. J. Physiol. 282, E1014-E1022
Muoio, D. M., Way, J. M., Tanner, C. J., Winegar, D. A., Kliewer, S. A., Houmard, J. A., Kraus, W. E., and Dohm, G. L. (2002) Diabetes 51, 901-909
Becker, T. C., BeltrandelRio, H., Noel, R. J., Johnson, J. H., and Newgard, C. B. (1994) J. Biol. Chem. 269, 21234-21238
An, J., Muoio, D. M., Shiota, M., Fujimoto, Y., Cline, G. W., Shulman, G. I., Koves, T. R., Stevens, R., Millington, D., and Newgard, C. B. (2004) Nat. Med. 10, 268-274
Cortright, R. N., Muoio, D. M., and Dohm, G. L. (1997) Nutr. Biochem. 8, 228-245
Muoio, D. M., MacLean, P. S., Lang, D. B., Li, S., Houmard, J. A., Way, J. M., Winegar, D. A., Corton, J. C., Dohm, G. L., and Kraus, W. E. (2002) J. Biol. Chem. 277, 26089-26097
de Fourmestraux, V., Neubauer, H., Poussin, C., Farmer, P., Falquet, L., Burcelin, R., Delorenzi, M., and Thorens, B. (2004) J. Biol. Chem. 279, 50743-50753
Van Hove, J. L., Zhang, W., Kahler, S. G., Roe, C. R., Chen, Y. T., Terada, N., Chace, D. H., Iafolla, A. K., Ding, J. H., and Millington, D. S. (1993) Am. J. Hum. Genet. 52, 958-966
Spiegelman, B. M., and Heinrich, R. (2004) Cell 119, 157-167
Pilegaard, H., Saltin, B., and Neufer, P. D. (2003) J. Physiol. (Lond.) 546, 851-858
Pilegaard, H., Ordway, G. A., Saltin, B., and Neufer, P. D. (2000) Am. J. Physiol. 279, E806-E814
St. Pierre, J., Lin, J., Krauss, S., Tarr, P. T., Yang, R., Newgard, C. B., and Spiegelman, B. M. (2003) J. Biol. Chem. 278, 26597-26603
Akimoto, T., Pohnert, S. C., Li, P., Zhang, M., Gumbs, C., Rosenberg, P. B., Williams, R. S., and Yan, Z. (2005) J. Biol. Chem. 280, 19587-19593
Barron, J. T., Gu, L., and Parrillo, J. E. (1998) J. Mol. Cell. Cardiol. 30, 1571-1579
Barron, J. T., Gu, L., and Parrillo, J. E. (1999) Mol. Cell. Biochem. 194, 283-290
Hittel, D. S., Kraus, W. E., Tanner, C. J., Houmard, J. A., and Hoffman, E. P. (2005) J. Appl. Physiol. 98, 168-179
Muoio, D. M., Dohm, G. L., Tapscott, E. B., and Coleman, R. A. (1999) Am. J. Physiol. 276, E913-E921
Mootha, V. K., Lindgren, C. M., Eriksson, K. F., Subramanian, A., Sihag, S., Lehar, J., Puigserver, P., Carlsson, E., Ridderstrale, M., Laurila, E., Houstis, N., Daly, M. J., Patterson, N., Mesirov, J. P., Golub, T. R., Tamayo, P., Spiegelman, B., Lander, E. S., Hirschhorn, J. N., Altshuler, D., and Groop, L. C. (2003) Nat. Genet. 34, 267-273
Ritov, V. B., Menshikova, E. V., He, J., Ferrell, R. E., Goodpaster, B. H., and Kelley, D. E. (2005) Diabetes 54, 8-14
Petersen, K. F., Befroy, D., Dufour, S., Dziura, J., Ariyan, C., Rothman, D. L., DiPietro, L., Cline, G. W., and Shulman, G. I. (2003) Science 300, 1140-1142
Puigserver, P., Rhee, J., Lin, J. D., Wu, Z. D., Yoon, J. C., Zhang, C. Y., Krauss, S., Mootha, V. K., Lowell, B. B., and Spiegelman, B. M. (2001) Mol. Cell 8, 971-982
Baar, K., Wende, A. R., Jones, T. E., Marison, M., Nolte, L. A., Chen, M., Kelly, D. P., and Holloszy, J. O. (2002) FASEB J. 16, 1879-1886
Lin, J., Wu, P. H., Tarr, P. T., Lindenberg, K. S., St. Pierre, J., Zhang, C. Y., Mootha, V. K., Jager, S., Vianna, C. R., Reznick, R. M., Cui, L., Manieri, M., Donovan, M. X., Wu, Z., Cooper, M. P., Fan, M. C., Rohas, L. M., Zavacki, A. M., Cinti, S., Shulman, G. I., Lowell, B. B., Krainc, D., and Spiegelman, B. M. (2004) Cell 119, 121-135
Finck, B. N., Bernal-Mizrachi, C., Han, D. H., Coleman, T., Sambandam, N., LaRiviere, L., Holloszy, J. O., Semenkovich, C. F., and Kelly, D. P. (2005) Cell Metab. 1, 133-144
Tordjman, K., Bernal-Mizrachi, C., Zemany, L., Weng, S., Feng, C., Zhang, F., Leone, T. C., Coleman, T., Kelly, D. P., and Semenkovich, C. F. (2001) J. Clin. Invest. 107, 1025-1034
Guerre-Millo, M., Rouault, C., Poulain, P., Andre, J., Poitout, V., Peters, J. M., Gonzalez, F. J., Fruchart, J. C., Reach, G., and Staels, B. (2001) Diabetes 50, 2809-2814
订阅登记：
请您在下面输入常用的Email地址、职业以便我们定期通过邮箱发送给您最新的相关医学信息，感谢您浏览首席医学网！