SIGNALING AND CELL PHYSIOLOGY

Uncoupling protein 3 expression levels influence insulinsensitivity, fatty acid oxidation, and relatedsignaling pathways

Rosalba Senese & Vivien Valli & Maria Moreno & Assunta Lombardi &Rosa Anna Busiello & Federica Cioffi & Elena Silvestri & Fernando Goglia &

Antonia Lanni & Pieter de Lange

Received: 7 July 2010 /Revised: 10 September 2010 /Accepted: 4 October 2010 /Published online: 7 November 2010# Springer-Verlag 2010

Abstract Controversy exists on whether uncoupling pro-tein 3 (UCP3) positively or negatively influences insulinsensitivity in vivo, and the underlying signaling pathwayshave been scarcely studied. We studied how a progressivereduction in UCP3 expression (using UCP3 +/+, UCP3 +/−,and UCP3 −/− mice) modulates insulin sensitivity andrelated metabolic parameters. In order to further validateour observations, we also studied animals in which insulinresistance was induced by administration of a high-fat diet(HFD). In UCP3 +/− and UCP3 −/− mice, gastrocnemiusmuscle Akt/protein kinase B (Akt/PKB) (serine 473) andAMP-activated protein kinase (AMPK) (threonine 171)phosphorylation, and glucose transporter 4 (GLUT4)membrane levels were reduced compared to UCP3 +/+

mice. The HOMA-IR index (insulin resistance parameter)was increased both in the UCP3 +/− and UCP3 −/− mice.In these mice, insulin administration normalized Akt/PKBphosphorylation between genotypes while AMPK phos-phorylation was further reduced, and sarcolemmal GLUT4levels were induced but did not reach control levels.Furthermore, non-insulin-stimulated muscle fatty acid oxi-dation and the expression of several involved genes both inmuscle and in liver were reduced. HFD administrationinduced insulin resistance in UCP3 +/+ mice and theaforementioned parameters resulted similar to those ofchow-fed UCP3 +/− and UCP3 −/− mice. In conclusion,high-fat-diet-induced insulin resistance in wild-type micemimics that of chow-fed UCP3 +/− and UCP3 −/− miceshowing that progressive reduction of UCP3 levels resultsin insulin resistance. This is accompanied by decreasedfatty acid oxidation and a less intense Akt/PKB and AMPKsignaling.

Keywords Uncoupling protein 3 . Insulin . Akt/PKB .

AMPK .GLUT4

Introduction

UCP3 is predominantly expressed in skeletal muscle, and toa lesser extent in heart and brown adipose tissue (BAT) [1],all of which are metabolically highly active tissues. Due toits homology to the classic uncoupling protein, uncouplingprotein 1 (UCP1), which is expressed in the thermogenicBAT [2], UCP3 was originally thought to perform afunction similar to that of UCP1 (viz. a dissipation of theproton gradient across the mitochondrial inner membraneleading to an uncoupling of oxidative phosphorylation and

Rosalba Senese and Vivien Valli contributed equally to the work.

R. Senese :V. Valli : F. Cioffi :A. Lanni (*) : P. de Lange (*)Dipartimento di Scienze della Vita,Seconda Università degli Studi di Napoli,Via Vivaldi 43,81100 Caserta, Italye-mail: antonia.lanni@unina2.it

P. de Langee-mail: pieter.delange@unina2.it

M. Moreno : E. Silvestri : F. GogliaDipartimento di Scienze Biologiche ed Ambientali,Università degli Studi del Sannio,Via Port’Arsa 11,82100 Benevento, Italy

A. Lombardi :R. A. BusielloDipartimento delle Scienze Biologiche, Sez. Fisiologia ed Igiene,Università degli Studi di Napoli “Federico II”,Via Mezzocannone 8,80134 Naples, Italy

Pflugers Arch - Eur J Physiol (2011) 461:153–164DOI 10.1007/s00424-010-0892-3

the release of energy as heat) in skeletal muscle. However,UCP3 mRNA and protein increase not only in situations inwhich energy expenditure is significantly increased (hyper-thyroidism [3–6], high levels of leptin [7], cold exposure[8]) but also in a situation in which it would be expected tobe depressed, being starvation [9–11]. As yet, an increasedexpression of UCP analogues has not been shown to bealways associated with increased mitochondrial uncoupling[12, 13]. As for mitochondrial respiration control rate(RCR) or basal proton conductance, UCP3-knockout micehave been found not to differ from wild-type mice [9].However, validating the ongoing hypothesis proposed byGoglia and Skulachev on a functional role for UCP3 as atransporter of lipid hydroperoxides outside of the mito-chondrial matrix [14], we have recently shown that thereindeed exists a function for UCP3 as a translocator of lipidhydroperoxide and a mediator of lipid hydroperoxide-dependent mitochondrial uncoupling [15]. Even if a directrole of UCP3 in the regulation of fatty acid oxidation inmuscle has recently been questioned [16], several studiesshow a correlation between increased UCP3 expression andincreased oxidation of fatty acids [17–20].

Besides being a major consumer of fatty acids as fuel,skeletal muscle is known to account for approximately 80%of insulin-stimulated glucose uptake. In this context, recentevidence suggests a role for UCP3 in the balance betweenglucose and lipid metabolism. When overexpressed in vivo,UCP3 induces reductions in the fasting blood glucose andinsulin concentrations, indicating an elevated insulin sensi-tivity [21, 22], while in L6 myotubes UCP3 improvesglucose uptake through activation of phosphoinositol 3-kinase (PI3K) [23]. However, depletion of UCP3 in micehas also been associated with increased insulin sensitivity[24], also when the mice were subjected to a high-fat diet[25]. Thus, up to now the relationship between UCP3expression and glucose tolerance/insulin sensitivity iscontradictory.

Clinical data seem in favor of an insulin-sensitizing role forUCP3: one study has reported that a 50% reduction of UCP3protein in muscle of human subjects could be correlated withthe incidence of type 2 diabetes mellitus (T2DM) [26], and ithas recently been shown that T2DM patients show strikingincreases in the concentration of total plasma fatty acylcar-nitines reflecting incomplete LCFA beta oxidation [27].Since polymorphisms in the UCP3 gene are increasinglyassociated with fat metabolism, obesity, and diabetes inhumans [28], it thus seems that despite the contrasting basicresearch data thus far published in the literature, reducedUCP3 levels would negatively interfere with insulin sensi-tivity and fatty acid oxidation, and this encouraged us tolook for an association between reduced UCP3 levels andthese parameters in a model of heterozygous and homozy-gous UCP3-depleted mice.

In relation to this, we compared insulin sensitivity andsignaling events and related metabolic parameters in-volved in insulin action in wild-type, heterozygous, orhomozygous UCP3-knockout mice. Several metabolicparameters of the same mouse strain have already beenpublished [29]. We studied basal and insulin-inducedphosphorylation of Akt/protein kinase B (Akt/PKB,predominantly influencing insulin-stimulated glucose up-take), AMP-activated protein kinase (AMPK, involved inglucose uptake and in fatty acid oxidation), and membranelevels of glucose transporter 4 (GLUT4). These parameterswere correlated to insulin sensitivity. In addition, musclefatty acid oxidation was measured as well as theexpression of genes involved in fatty acid oxidation andgluconeogenesis in gastrocnemius muscle and liver. Inorder to gain further insight into the role of UCP3 ininsulin sensitivity, we decided to measure the sameparameters under conditions in which insulin resistanceis known to be induced, namely, after exposure to a high-fat diet (HFD, [21, 25, 30]); an induction in UCP3 +/+mice of a biochemical/expression profile similar to thatobserved in UCP3 +/− and −/− mice would furthervalidate the conclusions.

Materials and methods

Hormones

The insulin used in this study (humalog rapid acting, insterile saline) was from Eli Lilly (Indianapolis, IN,USA).

Determination of UCP3 genotype

Mice (Swiss Black; Taconic, USA) harboring a truncatedUCP3 gene were originally constructed by Gong andcoworkers [29], and were a kind gift from Dr. Michael N.Sack, Translational Medicine Branch, National Institute ofhealth, Bethesda, MD, USA. Mice were first backcrossedwith fresh WT Swiss Black, then intercrossed to obtainthree different lines: UCP3 −/−, UCP3 +/−, and UCP3 +/+.For genotype analysis, tail DNA was isolated according tostandard protocols and submitted to multiplex PCR analy-sis. Oligos were used as described in [29] to distinguish thewild-type UCP3 gene from the transgene, containing aneomycin replacing exon 2 of the UCP3 gene [29]. Theoligos had the following sequence: UCP3-S exon 2: 5′-CATAGGCAGCAAAGGAAC-3′ UCP3-AS intron 2: 5′-CTATATGGTTTACACAGC-3 ′ , and NEO-S: 5 ′-GCAAAACCAAATTAAGGG-3′. Multiplex PCR thus pro-vided amplification fragments of 490 bp (derived from theNEO-S and UCP3-AS oligo, identifying the transgene) and

154 Pflugers Arch - Eur J Physiol (2011) 461:153–164

of 400 bp (derived from the UCP3-S and UCP3-AS oligo,identifying the wild-type gene).

Animal treatments

UCP3 +/+, +/−, and −/− mice were separated by gender,and littermates were housed in cages at 24°C. Water wasavailable ad libitum, as was a standard chow diet [fatty acidcontent (mg/kg): palmitate (16:0) 4,387, palmitoleate (16:1)202, stearic (18:0) 675, oleic (18:1) 5,046, linoleic (18:2)12,335, linoenic (18:3) 1,169; total metabolizable percent-age of energy: 60.4 carbohydrates, 29 proteins, 10.6 fat J/J;15.88 kJ gross energy/g, (Muscedola s.r.l., Milan, Italy)]. Atthe age of 6 weeks, male mice of each genotype wereallocated to one of two groups. For the next 5 weeks, thefirst group (termed N) continued to receive the chow diet,whereas the second group (termed HFD) received a high-fatdiet [consisting of 280 g chow diet supplemented with395 g lyophilized lamb meat (Liomellin, Milan, Italy),120 g cellulose (Sigma), 20 g mineral mix (ICN Biomed-ical, Solon, OH, USA), 7 g vitamin mix (ICN), and 200 glow-salt butter (Lurpak, Denmark) (total metabolizablepercentage of energy=21 carbohydrates, 29 proteins, 50fat J/J; 19.85 kJ gross energy/g)]. After these treatments, themice were weighed. The mice were subjected to fasting for16 h and subgroups were administered insulin (1 mU/g).From these, mice were killed 15 min after insulin injectionfor Western blot analysis or were subjected to insulintolerance test. Mice were killed by cervical dislocation,after which the gastrocnemius muscle and liver wereexcised and either processed for the isolation of mitochon-dria and membranes or immediately frozen in liquidnitrogen.

Preparation of mitochondria and membrane fractions

Gastrocnemius muscle mitochondria were isolated afterhomogenization of the tissue in an isolation mediumconsisting of 220 mM mannitol, 70 mM sucrose, 20 mMTris–HCl, 1 mM EDTA, 5 mM EGTA, and 5 mM MgCl2,pH 7.4 (all from Sigma-Aldrich Corp., St Louis, MO,USA). After this homogenization, samples were centrifugedat 700×g, and supernatants were collected and transferredinto new tubes for subsequent centrifugation at 4,000×g.The obtained mitochondrial pellet was washed twice andwas then resuspended in a minimal volume of isolationmedium and kept on ice. The freshly prepared mitochondriawere used immediately either for assessments of respirationrate (fatty acid oxidation) or for enzymatic assays.Mitochondria for Western blot analysis were isolated aspreviously described [6]. Isolation of the membrane fractionwas performed using a protocol slightly modified from onepreviously published [23]. The supernatants separated from

the mitochondrial pellets were recentrifuged at 10,000×gfor 10 min and the resulting supernatant was thencentrifuged at 190,000×g for 1 h at 4°C (Beckman OptimaTLX; Beckman Coulter S.p.A.). The pellet containing theenriched plasma membrane fraction was suspended in lysisbuffer containing 20 mM Tris–HCl pH 7.5, 150 mM NaCl,1 mM EDTA, 1 mM EGTA, 2.5 mM Na2H2P2O7, 1 mM b-CH3H7O6PNa2, 1 mM Na3VO4, 1 mM PMSF, 1 mg/mlleupeptin, and 1% Triton X-100 (all from Sigma-Aldrich).The protein content of each fraction was determined usingBio Rad’s DC method (Bio Rad Laboratories, Hercules,CA, USA).

Preparation of total lysates

For Western blotting analysis, gastrocnemius muscle washomogenized in the previously described lysis buffer usingan Ultraturrax homogenizer and ultracentrifuged at86,000×g for 10 min at 4°C (Beckman Optima TLX). Theprotein concentrations of the supernatants of the ultra-centrifuged cleared lysates were determined using BioRad’s DC method (Bio Rad Laboratories).

RNA isolation

Total gastrocnemius muscle and liver RNAwas isolated usingTRIZOL standard protocol (Invitrogen Life Technologies,Milano, Italy). Tissue/TRIZOL mixtures were homogenizedusing a polytron, keeping the viscosity of the solution to aminimum to ensure effective inactivation of endogenousRNAse activity.

Quantitative real-time PCR

PCR primers were designed using Primer Express version 2.0(Invitrogen). Expression levels of Cyclophilin F (CycF) wereused for normalization. Primers used were: CycF sense: 5′-AGGCAGATGTCGTGCCAAAGAC-3′, antisense: 5′-ACATGAAGGCGGGATGACC-3′; PEPCK sense: 5′-CAACTGTTGGCTGGCTCTCA-3 ′ , antisense: 5 ′-CTGGCGTTGAATGCTTTCTC-3′; G6Pase sense: 5′-CAGTGGTTGGAGACTGGTTTC-3′, antisense: 5′-TTTCCACGAAAGATAGCGAGA-3′; LPK sense, 5′-GCATTGAAAGTGGAAAGCTC-3 ′ , antisense: 5 ′-GGGGCTAGATGGCAGATGTA-3′; AOX sense: 5′-TTGGAAACCACTGCCACATA-3 ′ , antisense: 5 ′-GGTCCCATACGTCAGCTTGT-3′; PPARδ sense: 5′-AACATCCCCAACTTCAGCAG-3′, antisense: 5′-GGAA-GAGGTACTGGCTGTCG-3′; ′CPT1a sense: 5′-ATGTG-G A C C T G CAT T C C T T C - 3 ′ , a n t i s e n s e : 5 ′ -CCTTCAGGAAAGGTGAGTCG-3′; and CPT1b sense: 5′-CTCAGCCTCTACGGCAAATC-3 ′ , antisense: 5 ′-CTTCTTGATCAGGCCTTTGC-3′. A total of 1 μg total

Pflugers Arch - Eur J Physiol (2011) 461:153–164 155

RNA was used to generate cDNA strands in a 20-μlreaction volume using the Cloned AMV First StrandSynthesis Kit (Invitrogen). An equivalent of 25 ng totalRNA was subsequently used in the amplification with50 nM gene-specific primers and 4 ml SYBR green mix(Applied Biosystems) in a total volume of 8 μl, usingstandard cycle parameters on an Applied Biosystemsmodel 7500.

Basal serum fatty acid measurement

Subgroups of N and HFD mice were subjected to fastingfor 16 h, and basal serum fatty acid levels were measuredusing a WAKO NEFA C kit (WAKO Chemicals GmbH,Neuss, Germany).

Basal serum insulin measurement

Subgroups of N and HFD mice were subjected to fastingfor 16 h, and basal serum insulin concentrations (mU/l)were measured using an ELISA kit from B-bridgeInternational (Mountain View, CA, USA).

Insulin tolerance test

Subgroups of N and HFD mice were subjected to fastingfor 16 h. Whole blood glucose (mM) was measured before(basal level) and 5, 10, 15, 30, and 60 min afterintraperitoneal injection of insulin (1 mU/g). To collectblood, a light incision was made in the tail, and completetime courses were obtained from each animal. Glucoseconcentrations were measured using the One Touch Ultraglucometer from LifeScan Inc. (Milpitas, CA, USA). Aftermeasurement, the animals were refed and successfullyrecovered.

Western immunoblot analysis

UCP3 protein levels were measured in skeletal musclemitochondria using a polyclonal antibody from ChemiconInternational (Temecula, CA, USA). The protein levels ofAkt and phosphorylated Akt (Ser 473) were determined inthe supernatants of ultracentrifuged lysates. The GLUT4protein level was measured in the membrane fractions. Allthe above measurements were made using polyclonalantibodies. The total protein levels of AMPK weredetermined in the same lysates by employing specificantibodies against AMPK (Cell Signaling). Phosphorylationof AMPK (the α-subunit) was examined using an antibodyagainst Thr 172 (AMPK). All Akt, AMPK, and GLUTantibodies were from Cell Signaling Technology Inc.(Beverly, MA, USA). The β-actin antibody used here wasfrom Sigma-Aldrich Corp. Proteins were detected by a

chemiluminescence protein-detection method based on theprotocol supplied with a commercially available kit (NEN,Boston, MA, USA) and using the indicated antibodies andsecondary antibodies. The protein concentrations of thesamples were determined using Bio Rad’s DC method (BioRad Laboratories). Signals were quantified by means of aBio-Rad Molecular Imager FX using the supplied software(Bio-Rad Laboratories).

Measurement of fatty acid oxidation rate

Mitochondrial fatty acid oxidation was measured usingthe protocol described previously [31]. The rate ofmitochondrial fatty acid oxidation was assessed polaro-graphically using a Clark-type electrode at 30°C in a finalvolume of 0.5 ml of 80 mM KCl, 50 mM HEPES(pH 7.0), 1 mM EGTA, 5 mM K2HPO4, 1% BSA (w/v),and 2.5 mM malate in the presence of ADP (120 μg/ml).The reaction was started by the addition of palmitoyl-L-carnitine (40 μM). All the chemicals were purchasedfrom Sigma-Aldrich Corp.

Measurement of CPT activity

Total CPT (CPT1b plus CPT2) activity was measuredspectrophotometrically by following (at 412 nm) thekinetics of CoASH production in the presence of 5,5′-dithio-bis(2-nitrobenzoic acid) (DTNB) and carnitine, usingpalmitoyl-CoA as substrate [32]. A Σ412 value of13.6 mM−1cm−1 was applied for the calculation of totalCPT activity.

Statistical analysis

Results are expressed as means±SEM. The statisticalsignificance of differences between groups was deter-mined using a one-way ANOVA followed by a Student–Newman–Keuls test. Differences were considered signif-icant when P <0.05.

Results

Mitochondrial UCP3 protein levels in gastrocnemiusmuscle of UCP3 +/− mice

Mice were genotyped as described in the Materials andmethods section. To determine whether UCP3 mitochon-drial protein actually is halved in the UCP3 +/− mice, weperformed Western blot analysis on protein isolated fromfreshly prepared mitochondria from gastrocnemius muscle(see Materials and methods). We observed a 50% reductionof the mitochondrial UCP3 protein level in gastrocnemius

156 Pflugers Arch - Eur J Physiol (2011) 461:153–164

Fig. 1 Basal metabolic parameters of wild-type (+/+), UCP3-heterozygous (+/−), and UCP3-knockout (−/−) mice. a Genotypeanalysis and mitochondrial UCP3 protein levels in gastrocnemiusmuscle of standard chow-fed UCP3 +/+, UCP3 +/−, and UCP3 −/−mice. Upper panel: on the ethidium bromide-stained gel, the RT–PCRfragment of 400 base pairs (bp) represents amplification of the UCP3endogenous gene, whereas that of 490 bp represents amplification ofthe disrupted UCP3-NEO transgene (for further explanation, seeMaterials and methods). Lower panel: the Western blot, using protein(20 μg) isolated from gastrocnemius muscle mitochondria, shows the35-kDa UCP3 protein. Equal loading was verified by Ponceau Sstaining (not shown). b Histogram representing the HOMA-IR valuesin the different genotypes, fed either a chow or a high-fat diet (seeTable 2 for more details). The UCP3 genotypes are indicated under the

histograms. Asterisks in the histograms indicate difference (P<0.05)between the different genotypes. c Fatty acid oxidation rates in thedifferent genotypes, fed either a chow or a high-fat diet. The UCP3genotypes are indicated under the histograms. Asterisks in the histo-grams indicate difference (P<0.05) between the different genotypes. dCPT activity in the different genotypes, fed either a chow or a high-fatdiet. The UCP3 genotypes are indicated under the histograms. Asterisksin the histograms indicate difference (P<0.05) between the differentgenotypes. e Basal phosphorylation of Akt/PKB, AMPK (total lysates),and protein levels of GLUT4 (membranes) in gastrocnemius muscle ofmice of the different genotypes, fed either a chow or a high-fat diet. Alldata are representative or the mean±SE of five independent experi-ments. For quantification, see the histograms under Figs. 3 and 4. TheUCP3 genotypes are indicated under the blots

Pflugers Arch - Eur J Physiol (2011) 461:153–164 157

muscle from the UCP3 +/− mice (versus wild-type mice)(Fig. 1a). UCP3 −/− mice did not display a UCP3-specificsignal.

Comparison of phenotypic features and body weightsbetween the different genotypes

Six-week-old mice were weighed and then subjected toeither a chow or a high-fat diet for 5 weeks. At the end themice were weighed again. There were no phenotypicdifferences, nor were there differences in body weightbetween the different genotypes, before and after thefeeding period. The HFD-fed mice all gained relativelymore weight with respect to the chow-fed mice (seeTable 1).

UCP3 genotype and basal insulin sensitivity

Basal fasting serum glucose levels did not significantlyvary between the genotypes (either on chow or HFD), butwith respect to the UCP3 +/+ mice, serum insulin levelswere significantly higher in the UCP3 +/− and UCP3 −/−mice (see Table 2). As a consequence, the homeostaticmodel assessment of insulin resistance (HOMA-IR) index,indicative for insulin resistance, was also significantlyhigher in these mice. This was true for both diets, albeitthat the high-fat diet induced an increase in insulin levels inall genotypes (Table 2), leading to relatively higherHOMA-IR indexes with respect to the chow-fed animals(for comparison with the data discussed in the followingparagraphs, the HOMA-IR data are also shown as ahistogram in Fig. 1b).

UCP3 genotype, mitochondrial fatty acid oxidation,and CPT activity

Blood fatty acid levels in animals under both dietregimens were increased in both UCP3 +/− and −/−

mice with respect to UCP3 +/+ mice (Table 2). Westudied if this was due to decreased fatty acid oxidation inmuscle mitochondria. Indeed, muscle mitochondrial fattyacid oxidation decreased in both the UCP3 +/− and −/−mice with respect to the UCP3 +/+ mice in both dietregimens. These data indicate that reducing UCP3 levelsin any given metabolic context leads to a reduction inmuscle’s capacity to oxidize lipids. To further investigatewhether a reduction of UCP3 may also influence fatty aciduptake, we next measured the activity of CPT. CPTactivities in the UCP3 +/− and UCP3 −/− mice werereduced by around 20% with respect to the UCP3 +/+mice. In the HFD-fed mice, CPT activities in theUCP3 +/− and UCP3 −/− mice were lowered by around33% with respect to the UCP3 +/+ mice (Fig. 1d). Thevariations in CPT activity are associated to similarvariations in CPT expression (see Fig. 2).

UCP3 genotype, basal (non-insulin stimulated) Akt/PKBand AMPK phosphorylation, and GLUT4 membrane levels

In agreement with the parameters measured above, usingboth diets, we observed a gradual decrease in basalphosphorylation levels of Akt/PKB (indicative for de-creased insulin sensitivity) and AMPK (indicative fordecreased fatty acid oxidation) in the UCP3 +/− and −/−mice with respect to the UCP3 +/+ mice (Fig. 1e). TotalAkt/PKB and AMPK protein levels did not differ betweenthe genotypes. In gastrocnemius muscle of the chow-fed,wild-type mice, basal GLUT4 levels were relatively highcompared to those in the chow-fed UCP3 +/− and −/− mice(Fig. 1e, CHOW). HFD-associated GLUT4 membranelevels in the UCP3 +/+ mice were downregulated to levelscomparable to those in the chow-fed UCP3 −/− mice(Fig. 1e, HFD).

Table 2 Basal blood lipid, glucose and insulin levels, and HOMAindex of UCP3 +/+, +/−, and −/− mice at the end of the feeding period

UCP3 +/+ UCP3 +/− UCP3 −/−

FA chow (nM) 0.4±0.02 0.53±0.03* 0.63±0.02*

Glucose chow (mM) 5.6±0.06 6.64±0.14* 7.0±0.07*

Insulin chow (mU/l) 6.5±0.4 10.63±0.5* 12.5±0.8*

HOMA-IR 1.52±0.05 3.13±0.01* 3.87±0.15*

FA HFD (nM) 0.45±0.02 0.7±0.02* 0.93±0.06*

Glucose HFD (mM) 5.9±0.06 6.8±0.17 7.5±0.14

Insulin HFD (mU/l) 19.9±1.3 31.1±1.2* 36.1±2.4*

HOMA-IR 5.23±0.5 9.44±0.5* 12.04±0.55*

All measurements were performed in 11-week-old male mice. All dataare the mean±SE of six independent experiments

*P<0.05 vs. UCP3 +/+ animals

Table 1 Body weights and their increase after the feeding period inUCP3 +/+, +/−, and −/− mice

UCP3 +/+ UCP3 +/− UCP3 −/−

BW chow (g) initial 22.43±1.33 22.27±1.28 23.47±0.46

BW chow (g) final 23.60±1.12 23.45±0.28 24.22±0.28

ΔBW 1.17±0.29 1.17±0.40 0.75±0.39

BW HFD (g) initial 24.50±0.5 23.33±0.33 22.37±1.03

BW HFD (g) final 31.85±0.15 31.00±0.53 29.67±1.20

ΔBW 7.35±0.35 7.67±0.24 7.3±0.21

All measurements were performed starting from 6-week-old malemice. All data are the mean±SE of six independent experiments

158 Pflugers Arch - Eur J Physiol (2011) 461:153–164

UCP3 genotype and expression of genes involvedin muscle and liver energy metabolism

In gastrocnemius muscle, the expression of PPARδ wasdownregulated in both the UCP3 +/− and UCP3 −/− micewith respect to UCP3 +/+ mice under both dietary conditions,and so were two key PPAR target genes involved in fatty acidoxidation: CPT1b, involved in mitochondrial fatty aciduptake; and AOX, a key enzyme in peroxisomal fatty acidoxidation, significantly so in the UCP3 −/− mice (Fig. 2a).

We reasoned that the reduction of fatty acid oxidation in themuscle due to reduced UCP3 levels cause adaptive responsesin other metabolically active organs such as the liver. Indeed,in liver, the expression of PPARδ was downregulated in theUCP3 +/− and UCP3 −/− mice (significantly so) with respectto UCP3 +/+ mice under both dietary conditions, and so wereCPT1a and AOX (Fig. 2b, left panel). This indicates thatreduced UCP3 levels diminish the use of fatty acids as fuelboth in muscle and in liver. As the liver is an importantregulator of blood glucose levels, it may compensate for thereduced use of fatty acids as fuel by enhancing glucoseproduction. This would in turn lead to the need to produce

more insulin to normalize basal serum glucose levels inanimals with reduced UCP3 levels, which indeed weobserved (Table 2). With this in mind, we measured theconsequence of reduced UCP3 levels on gluconeogenic geneexpression. In mice fed the chow diet, expression of thegluconeogenic enzyme PEPCK remained constant betweenthe genotypes, whereas the high-fat diet significantly reducedthe expression of this gene only in the UCP3 +/+ mice(compared to chow-fed UCP3 +/+ mice; Fig. 2b, right panel).Under both dietary conditions, the expression of G6Pase,crucial for hepatic export of glucose, tended to bedownregulated in the UCP3 +/− and UCP3 −/− micewith respect to UCP3 +/+ mice, and the HFD relativelylowered the expression of this gene in all genotypes(reaching a significant difference compared to theUCP3 +/+ mice; Fig. 2b, right panel).

Insulin-induced signaling response and insulin tolerance,comparison between UCP3 genotype in chow-fed mice

Under chow-fed conditions, whereas basal Ser 473phosphorylation of Akt/PKB in gastrocnemius muscle

Fig. 2 Real-time PCR analysisof genes involved in fatty acidoxidation and gluconeogenesisin gastrocnemius muscle andliver. a Gastrocnemius muscle, bliver. Asterisks in the histogramsindicate difference (P<0.05) be-tween the different genotypes.Hashes indicate difference (P<0.05) between diets. All data arethe mean±SE of five indepen-dent experiments. For abbrevia-tions, see text

Pflugers Arch - Eur J Physiol (2011) 461:153–164 159

gradually decreased in the UCP3 +/− and UCP3 −/−with respect to the UCP3 +/+ mice, after injection ofinsulin, all genotypes had similar levels of phosphory-lated Akt/PKB in gastrocnemius muscle (Fig. 3a).AMPK phosphorylation was inversely correlated withthe induction of Akt phosphorylation in response toinsulin (Fig. 3b). In response to insulin, AMPK phos-phorylation fell significantly (by more than 2-fold) in theUCP3 +/+ mice and even more (by 4- and 5-fold,respectively) in the UCP3 +/− and UCP3 −/− mice.Whereas insulin did not stimulate GLUT4 membranelevels in gastrocnemius muscle of the UCP3 +/+ mice, theUCP3 +/− and UCP3 −/− mice did show an increase inGLUT4 membrane protein in response to insulin (by 2-and 3-fold, respectively), but absolute GLUT 4 membrane

levels did not reach those observed in the UCP3 +/+ mice(Fig. 3c). Insulin tolerance tests revealed that within the20-min time interval after insulin injection, the UCP3+/+mice were significantly more insulin sensitive comparedto the UCP3 +/− and −/− mice (Fig. 3d).

Insulin-induced signaling response and insulin tolerance,comparison between UCP3 genotype in HFD-fed mice

Under HFD conditions, basal Ser 473 phosphorylation ofAkt/PKB in gastrocnemius muscle gradually decreased inthe UCP3 +/− and UCP3 −/− with respect to the UCP3 +/+mice. After injection of insulin, the UCP3 +/+ mice did notshow increased Akt/PKB phosphorylation in gastrocnemiusmuscle, whereas in the UCP3 +/− and UCP3 −/− mice, Akt/

Fig. 3 Response to insulin of standard chow-fed UCP3 +/+, +/−, and−/− mice. a Effects of insulin on levels of phosphorylation of Akt(Ser473) in gastrocnemius muscle. b Effects of insulin on levels ofphosphorylation of AMPK (Thr172) in gastrocnemius muscle. cEffects of insulin on membrane levels of GLUT4 in gastrocnemiusmuscle. a, b, c analysis was performed on tissue isolated 15 min afterinsulin injection. Each lane contained 30 μg of protein from a singlemouse. Even protein loading was reassessed after blotting by probingthe blot with a beta-actin antibody. Quantified data are the mean±SE,ratios being calculated with respect to non-insulin-treated controls.

Asterisks in the histograms indicate difference between insulin-stimulated (P<0.05) and non-insulin-treated controls. Hash marksindicate difference between non-insulin-treated UCP3 +/− and UCP3−/− animals and non-insulin-treated UCP3+/+ animals. N.S. nosignificant difference between the insulin-treated animals. d Insulintolerance test (ITT) in the chow-fed mice of all three genotypes. Afterinjection with insulin, tail blood glucose was measured over time.Asterisks in the curves indicate difference between insulin-stimulated(P<0.05) and non-insulin-treated controls. All data are representativeor the mean±SE of five independent experiments

160 Pflugers Arch - Eur J Physiol (2011) 461:153–164

PKB phosphorylation levels increased. However, theselevels did not reach a significant difference compared tothose of the insulin-stimulated UCP3 +/+ mice (Fig. 4a). Asalso observed in the chow-fed mice, AMPK phosphoryla-tion was inversely correlated with the induction of Aktphosphorylation in response to insulin (Fig. 3b). Inresponse to insulin, AMPK phosphorylation fell signifi-cantly (by more than 2-fold) in the UCP3 +/+ mice andeven more (by 4- and 5-fold, respectively) in the UCP3 +/−and UCP3 −/− mice. Gastrocnemius muscle GLUT4membrane levels (insulin stimulated and not) did not varybetween the genotypes (Fig. 4c). Indeed, under theseconditions, insulin tolerance of the UCP3 +/+ mice werealmost comparable to that of UCP3 +/− and −/− mice

(Fig. 4d), although within the first 10 min UCP3 +/+ micewere more insulin sensitive.

Discussion

In the experimental model utilized here, UCP3 wild-typemice are compared with those in which UCP3 is partially orcompletely ablated, focusing on insulin sensitivity, skeletalmuscle fatty acid oxidation, phosphorylation of Akt/PKBand AMPK, GLUT4 membrane levels, and gene expressioninvolved in fat and glucose handling in muscle and liver.Here, we show that reduced UCP3 levels correlate withreduced basal Akt/PKB phosphorylation and with basal

Fig. 4 Response to insulin of HFD-fed UCP3 +/+, +/−, and −/− mice.a Effects of insulin on levels of phosphorylation of Akt (Ser473) ingastrocnemius muscle. b Effects of insulin on levels of phosphoryla-tion of AMPK (Thr172) in gastrocnemius muscle. c Effects of insulinon membrane levels of GLUT4 in gastrocnemius muscle. a, b, canalysis was performed on tissue isolated 15 min after insulininjection. Each lane contained 30 μg of protein from a single mouse.Even protein loading was reassessed after blotting by probing the blotwith a beta-actin antibody. Quantified data are the mean±SE, ratiosbeing calculated with respect to non-insulin-treated controls. Asterisks

in the histograms indicate difference between insulin-stimulated (P<0.05) and non-insulin-treated controls. Hash marks indicate differencebetween non-insulin-treated animal and non-insulin-treated UCP3+/+animal. N.S. no significant difference between the insulin-treatedanimals. d Insulin tolerance test (ITT) in the HFD-fed mice of all threegenotypes. After injection with insulin, tail blood glucose wasmeasured over time. Asterisks in the curves indicate differencebetween insulin-stimulated (P<0.05) and non-insulin-treated controls.All data are representative or the mean±SE of five independentexperiments

Pflugers Arch - Eur J Physiol (2011) 461:153–164 161

insulin resistance reflected by increased HOMA-IR values.In addition, reduced UCP3 levels and AMPK activitiesinter-relate, at least in part, through reduced activity ofCPT, a known target of AMPK [20]. This also means thatreduced fatty acid oxidation in the UCP3 +/− and UCP3 −/− mice, on both diet regimens tested in this study, is partlydue to decreased fatty acid uptake into the mitochondria.Detailed overviews of basal metabolic parameters of thesame mice, housed and fed under the same conditions asused here, have already been presented [29], and of adifferent mouse strain in response to a high-fat diet [33],and it was concluded that there were no significantphenotypic differences between the three genotypes [29,33]. However, HOMA-IR indexes for basal insulin resis-tance were not reported in these studies. Interestingly, fromthe data of the previously mentioned papers [29, 33], it canbe extrapolated that basal insulin sensitivity was reduced intheir male UCP3 KO mice, due to increased insulin levels,which is in line with our observations. From these reports,it can be deduced that it is important to note that the age ofthe animals considered for such kind of studies maydrastically influence the data obtained on the degree ofinsulin resistance in the UCP3-depleted mice with respectto the UCP3 +/+ mice. Costford et al. [24, 25] haveproposed that in their model both UCP3-overexpressing(UCP3tg) mice as well as UCP3 −/− mice tended to bemore insulin sensitive compared to UCP3 +/+ mice, all ofthe male gender. However, within the first 3 months ofhigh-fat feeding, they observed similar responses to insulinin all genotypes [24]. In that study, differences were onlyobserved when high-fat-diet feeding periods were increasedto 4 months, but then after 8 months the differencesdisappeared again. In a later publication, the authorsrepeated the experiment and now found differences insensitivity after 3 months. However, the gender of theanimals was not reported [25]. In addition, from the workof Choi and colleagues [21], one can extrapolate thatinsulin sensitivity of UCPtg mice after a high-fat diet isclearly elevated with respect to UCP3 +/+ mice [21]. Theresults of this paper, together with the data obtained byVidal-Puig and colleagues [33], are in line with ourobservations. In light of the previous observations, in orderto avoid aging-related variability, we decided to study thisparameter in mice with an age of 11 weeks. That, as judgedfrom the previous studies, seems to be an age which isneither too precocious nor overlapping with aging-relatedmechanisms. In the UCP3 +/+ mice, basal GLUT4 levelsdo not change after injection with insulin. This result isperhaps different from what one may expect, but it confirmsearlier observations on skeletal muscle in vivo wherefasting and streptozotocin treatment (blocking pancreaticactivity) both increased plasma membrane GLUT4 levels[34]. Interestingly, the lowered basal Akt phosphorylation

levels seen in mice with reduced UCP3 levels arenormalized to wild-type Akt phosphorylation levels afterinjection with insulin; however, GLUT4 membrane levelsdo not reach wild-type levels, at least under chow-fedconditions. In the HFD-fed mice, the GLUT4 membranelevels in the WT mice were lowered to reach the values ofthe UCP3 +/− and −/− mice. This implies that theendogenous insulin levels in the UCP3 +/− and UCP3 −/−mice, although elevated with respect to the levels in thewild-type mice (Table 2), are not sufficient to compensatefor reduced Akt response caused by the lowered UCP3levels. Despite a recovery of Akt phosphorylation byexogenous insulin administration, UCP3 +/− and −/− miceremain insulin resistant with respect to the wild-type mice.It is well known that Akt/PKB and AMPK are determinantsof glucose and lipid use, respectively [20, 35], and we haverecently shown that specific inhibition in vivo of Akt/PKBsuppressed activity of the glycolytic enzyme phosphofruc-tokinase, whereas in vivo inhibition of AMPK activitysuppressed fatty acid oxidation in gastrocnemius muscle[36]. Here, we show that reduction of UCP3 levels ingastrocnemius muscle reduces both AMPK phosphoryla-tion and fatty acid oxidation. AMPK is a majordeterminant of the use of fatty acids as fuel in skeletalmuscle under conditions in which insulin signaling isdepressed [20]. These data are in line with the hypothesisthat UCP3 may be one of the determinants involved in theswitch from glucose to lipid utilization by skeletal muscle[37], as its reduction leads to an increase in Akt/PKB—and decrease in AMPK—phosphorylation in response toinsulin. Since inhibition of AMPK phosphorylation causesa less efficient oxidation of fatty acids in gastrocnemiusmuscle [36], a decline in AMPK activity in response toinsulin should reduce insulin sensitivity, and this is inagreement with previous data from other groups showingthat the factors responsible for mediating fat-inducedinsulin resistance may be accumulated intramusculartriglyceride (IMTG) breakdown products including fattyacyl-CoAs, diacylglycerol (DAG), and ceramides (forreview, see [35]). It has recently been shown that T2DMpatients show striking increases in the concentration oftotal plasma fatty acylcarnitines reflecting incompleteLCFA beta oxidation [27] which may be explained byrepression of AMPK signaling and fatty acid oxidationrelated to the low UCP3 levels often found in thesepatients [26]. The HFD increased HOMA-IR values in theUCP3 +/+ mice to values similar to those of chow-fed +/−and −/− mice, thus definitely establishing that UCP3 +/−and −/− mice are insulin resistant. But what happenedwhen HFD was administrated to UCP3 +/− and −/−animals? In these animals, further validating the efficacyof HFD as inducer of insulin resistance, there was anincrease in HOMA index values also in the UCP3 +/−

162 Pflugers Arch - Eur J Physiol (2011) 461:153–164

and −/− mice, again causing a relative increase in theUCP3 +/− and UCP3 −/− mice with respect to their wild-type HFD littermates. The observation that HFD feedingcaused a decrease in PEPCK expression in the UCP3 +/+ mice but not in the UCP3 +/− and UCP3 −/− mice(indicating that liver glucose production is increasedwhen muscle UCP3 levels are reduced) allows us tosuggest that under HFD conditions, also the livercontributes to the insulin resistance we observed in theUCP3 +/− and UCP3 −/− mice through an interplaywith muscle. A similar example of an interplay betweendifferent metabolic tissues upon manipulation by trans-genes is shown in the work of Izumiya and coworkers[38] who inserted a muscle-specific inducible Akt trans-gene and observed a drastic adaptive response in liver.Another interesting observation coming from our study isthat reduced UCP3 levels also resulted in a decrease inPPARδ expression in both skeletal muscle and liver.These results underline the recently published observa-tion that in liver of PPARδ-depleted mice, pathwaysrelated to glucose utilization are downregulated, whichcorrelated with elevated plasma glucose and triglycerides[39]. In conclusion, through various metabolic altera-tions, reduction of UCP3 levels in mice fed a chow dietresults in reduced insulin sensitivity, which can bemimicked when wild-type mice are exposed to a high-fat diet. In UCP3 +/− and UCP3 −/− mice, on the otherhand, a HFD resulted in a further worsening of theinsulin-resistance profile.

Acknowledgment This study was supported by a Grant fromMinistero dell’Istruzione dell’Università, e della Ricerca COFIN—Protocol Number 20089SRS2X.

Ethical standards The experiments comply with the current Italianlaws.

Conflict of interest The authors declare that they have no conflict ofinterest.

References

1. Vidal-Puig A, Solanes G, Grujic D, Flier JS, Lowell BB (1997)UCP-3, an uncoupling protein homolog expressed preferentiallyand abundantly in skeletal muscle and brown adipose tissue.Biochem Biophys Res Commun 235:79–82

2. Lowell BB, Susulic VS, Hamann A, Lawitts JA, Himms-Hagen J,Boyer BB, Kozak LP, Flier JS (1993) Development of obesity intransgenic mice after genetic ablation of brown adipose tissue.Nature 366:740–742

3. Silvestri E, Moreno M, Lombardi A, Ragni M, de Lange P,Alexson SE, Lanni A, Goglia F (2005) Thyroid hormoneeffects on putative biochemical pathways involved in UCP3activation in rat skeletal muscle mitochondria. FEBS Lett579:1639–1645

4. Gong DW, He Y, Karas M, Reitman M (1997) Uncoupling protein-3is a mediator of thermogenesis regulated by thyroid hormone, beta3-adrenergic agonists, and leptin. J Biol Chem 272:24129–24132

5. Lanni A, Beneduce L, Lombardi A, Moreno M, Boss O, MuzzinP, Giacobino JP, Goglia F (1999) Expression of uncouplingprotein-3 and mitochondrial activity in the transition fromhypothyroid to hyperthyroid state in rat skeletal muscle. FEBSLett 444:250–254

6. De Lange P, Lanni A, Beneduce L, Moreno M, Lombardi A,Silvestri E, Goglia F (2001) Uncoupling protein-3 is a moleculardeterminant for the regulation of resting metabolic rate by thyroidhormone. Endocrinology 142:3414–3420

7. Cusin I, Zakrzewska KE, Boss O, Muzzin P, Giacobino JP,Ricquier D, Jeanrenaud B, Rohner-Jeanrenaud F (1998) Chroniccentral leptin infusion enhances insulin-stimulated glucose metab-olism and favors the expression of uncoupling proteins. Diabetes47:1014–1019

8. Simonyan RA, Jimenez M, Ceddia RB, Giacobino JP, Muzzin P,Skulachev VP (2001) Cold-induced changes in the energycoupling and the UCP3 level in rodent skeletal muscles. BiochimBiophys Acta 1505:271–279

9. Cadenas S, Echtay KS, Harper JA, Jekabsons MB, BuckinghamJA, Grau E, Abuin A, Chapman H, Clapham JC, Brand MD(2002) The basal proton conductance of skeletal muscle mito-chondria from transgenic mice overexpressing or lacking uncou-pling protein-3. J Biol Chem 277:2773–2778

10. De Lange P, Ragni M, Silvestri E, Moreno M, Schiavo L,Lombardi A, Farina P, Feola A, Goglia F, Lanni A (2004)Combined cDNA/RT–PCR analysis of gene expression profile inrat gastrocnemius muscle: relation to its adaptative function inenergy metabolism during fasting. FASEB J 18:350–352

11. De Lange P, Farina P, Moreno M, Ragni M, Lombardi A, SilvestriE, Burrone L, Lanni A, Goglia F (2006) Sequential changes in thesignal transduction responses of skeletal muscle following fooddeprivation. FASEB J 20:2579–2581

12. Cadenas S, Buckingham JA, Samec S, Seydoux J, Din N, DullooAG, Brand MD (1999) UCP2 and UCP3 rise in starved ratskeletal muscle but mitochondrial proton conductance is un-changed. FEBS Lett 462:257–260

13. Hesselink MK, Greenha PL, Constantin-Teodosiu D, Hultman E,Saris WH, Nieuwlaat R, Schaart G, Kornips E, Schrauwen P(2003) Increased uncoupling protein-3 content does not affectmitochondrial function in human skeletal muscle in vivo. J ClinInvest 111:479–486

14. Goglia F, Skulachev VP (2003) A function for novel uncouplingproteins: antioxidant defense of mitochondrial matrix by trans-locating fatty acid peroxides from the inner to the outer membraneleaflet. FASEB J 17:1585–1591

15. Lombardi A, Busiello RA, Napolitano L, Cioffi F, Moreno M, deLange P, Silvestri E, Lanni A, Goglia F (2010) Uncouplingprotein-3 (UCP3) translocates lipid hydroperoxide and mediateslipid hydroperoxide-dependent mitochondrial uncoupling. J BiolChem 285:16599–16605

16. Seifert EL, Bezaire V, Estay C, Harper ME (2008) Essential rolefor uncoupling protein-3 in mitochondrial adaptation to fasting butnot in fatty acid oxidation or fatty acid anion export. J Biol Chem283:25134–25131

17. Bezaire V, Seifert EL, Harper ME (2007) Uncoupling protein-3:clues in an ongoing mitochondrial mystery. FASEB J 21:312–324

18. Samec S, Seydoux J, Dulloo AG (1998) Role of UCP homologuesin skeletal muscles and brown adipose tissue: mediators ofthermogenesis or regulators of lipids as fuel substrate? FASEB J12:715–724

19. Samec S, Seydoux J, Dulloo AG (2000) Downregulation ofskeletal muscle UCP-3 gene expression during refeeding isprevented by cold exposure. Pflugers Arch 439:723–729

Pflugers Arch - Eur J Physiol (2011) 461:153–164 163

20. De Lange P, Moreno M, Silvestri E, Lombardi A, Goglia F, LanniA (2007) Fuel economy in food-deprived skeletal muscle:signalling pathways and regulatory mechanisms [review]. FASEBJ 21:3431–3441

21. Choi CS, Fillmore JJ, Kim JK, Liu ZK, Kim S, Collier EF, KulkarniA, Distefano A, Hwang YJ, Kahn M, Chen Y, Yu C, Moore IK,Reznick RM, Higashimori T, Shulman GI (2007) Overexpression ofuncoupling protein 3 in skeletal muscle protects against fat-inducedinsulin resistance. J Clin Invest 117:1995–2003

22. Clapham JC, Arch JR, Chapman H, Haynes A, Lister C, MooreGB, Piercy V, Carter SA, Lehner I, Smith SA, Beeley LJ, GoddenRJ, Herrity N, Skehel M, Changani KK, Hockings PD, Reid DG,Squires SM, Hatcher J, Trail B, Latcham J, Rastan S, Harper AJ,Cadenas S, Buckingham JA, Brand MD, Abuin A (2000) Miceoverexpressing human uncoupling protein-3 in skeletal muscle arehyperphagic and lean. Nature 406:415–418

23. Huppertz C, Fischer BM, Kim YB, Kotani K, Vidal-Puig A,Slieker LJ, Sloop KW, Lowell BB, Kahn BB (2001) Uncouplingprotein 3 (UCP3) stimulates glucose uptake in muscle cellsthrough a phosphoinositide 3-kinase-dependent mechanism. J BiolChem 276:12520–12529

24. Costford SR, Chaudhry SN, Salkhordeh M, Harper ME (2006)Effects of the presence, absence, and overexpression of uncou-pling protein-3 on adiposity and fuel metabolism in congenicmice. Am J Physiol Endocrinol Metab 290:E1304–E1312

25. Costford SR, Chaudry SN, Crawford SA, SalkhordehM, Harper ME(2008) Long-term high-fat feeding induces greater fat storage inmicelacking UCP3. Am J Physiol Endocrinol Metab 295:E1018–E1021

26. Schrauwen P, Hesselink MKC, Blaak EE, Borghouts LB, SchaartG, Saris WHM, Keizer HA (2001) Uncoupling protein 3 contentis decreased in skeletal muscle of patients with type 2 diabetes.Diabetes 50:2870–2873

27. Adams SH, Hoppel CL, Lok KH, Zhao L, Wong SW, Minkler PE,Hwang DH, Newman JW, Garvey WT (2009) Plasma acylcarni-tine profiles suggest incomplete long-chain fatty acid (beta)oxidation and altered tricarboxylic acid cycle activity in type 2diabetic African-American women. J Nutr 139:1073–1081

28. Jia J-J, Zhang X, Ge CR, Jois M (2009) The polymorphisms ofUCP2 and UCP3 genes associated with fat metabolism, obesityand diabetes. Obes Rev 10:519–526

29. Gong DW, Monemdjou S, Gavrilova O, Leon LR, Marcus-Samuels B, Chou CJ, Everett C, Kozak LP, Li C, Deng C, Harper

ME, Reitman ML (2000) Lack of obesity and normal response tofasting and thyroid hormone in mice lacking uncoupling protein-3. J Biol Chem 275:16251–16257

30. Tiraby C, Tavernier G, Capel F, Mairal A, Crampes F, Rami J,Pujol C, Boutin LD (2007) Resistance to high-fat-diet-inducedobesity and sexual dimorphism in the metabolic responses oftransgenic mice with moderate uncoupling protein 3 overexpres-sion in glycolytic skeletal muscles. Diabetologia 50:2190–2199

31. Kerner J, Turkaly PJ, Minkler PE, Hoppel CL (2001) Agingskeletal muscle mitochondria in the rat: decreased uncouplingprotein-3 content. Am J Physiol Metab 281:E1054–E1062

32. Alexson SEH, Nedergaard J (1988) A novel type of short- andmedium-chain acyl-CoA hydrolases in brown adipose tissuemitochondria. J Biol Chem 263:13564–13571

33. Vidal-Puig AJ, Grujic D, Zhang CY, Hagen T, Boss O, Ido Y,Szczepanik A, Wade J, Mootha V, Cortright R, Muoio DM,Lowell BB (2000) Energy metabolism in uncoupling protein 3gene knockout mice. J Biol Chem 275:16258–16266

34. Camps M, Castello A, Munoz P, Monfar M, Testar X, Palacin M,Zorzano A (1992) Effect of diabetes and fasting on GLUT-4(muscle/fat) glucose transporter expression in insulin-sensitivetissues. Biochem J 282:765–772

35. Kiens B (2006) Skeletal muscle lipid metabolism in exercise andinsulin resistance. Physiol Rev 86:205–243

36. De Lange P, Senese R, Cioffi F, Moreno M, Lombardi A, SilvestriE, Goglia F, Lanni A (2008) Rapid activation by 3, 5, 3′-L-triiodothyronine of AMPK/ACC and Akt/PKB signaling path-ways: relation to changes in fuel metabolism and myosin heavychain protein content in rat gastrocnemius muscle in vivo.Endocrinology 149:6462–6470

37. Criscuolo F, Mozo J, Hurtaud C, Nubel T, Bouillaud F (2006)UCP2, UCP3, avUCP, what do they do when proton transport isnot stimulated? Possible relevance to pyruvate and glutaminemetabolism. Biochim Biophys Acta 175:1284–1291

38. Isumiya Y, Hopkins T, Morris C, Sato K, Zeng L, Viereck J,Hamilton J, Ouchi N, LeBrasseur NK, Walsh K (2008) Fast/glycolytic muscle fiber growth reduces fat mass and improvesmetabolic parameters in obese mice. Cell Metab 7:159–172

39. Sanderson LM, Boekschoten MV, Desvergne B, Muller M,Kersten S (2009) Transcriptional profiling reveals divergent rolesof PPAR alpha and PPAR beta/delta in regulation of geneexpression in mouse liver. Physiol Genomics 41:42–52

164 Pflugers Arch - Eur J Physiol (2011) 461:153–164