YXW and RME conceived and designed the experiments. YXW, CLZ, RTY, MCN, and CRBO performed the experiments. YXW, CLZ, and RME analyzed the data. HKC, JH, and HK contributed reagents/materials/analysis tools. YXW and RME wrote the paper.
The authors have declared that no conflicts of interest exist.
Endurance exercise training can promote an adaptive muscle fiber transformation and an increase of mitochondrial biogenesis by triggering scripted changes in gene expression. However, no transcription factor has yet been identified that can direct this process. We describe the engineering of a mouse capable of continuous running of up to twice the distance of a wild-type littermate. This was achieved by targeted expression of an activated form of peroxisome proliferator-activated receptor δ (PPARδ) in skeletal muscle, which induces a switch to form increased numbers of type I muscle fibers. Treatment of wild-type mice with PPARδ agonist elicits a similar type I fiber gene expression profile in muscle. Moreover, these genetically generated fibers confer resistance to obesity with improved metabolic profiles, even in the absence of exercise. These results demonstrate that complex physiologic properties such as fatigue, endurance, and running capacity can be molecularly analyzed and manipulated.
Engineered expression of the peroxisome proliferator-activated receptor δ in skeletal muscle increases type I muscle fibers allowing the modified mice to run twice the distance of wild-type littermates.
Skeletal muscle fibers are generally classified as type I (oxidative/slow) or type II (glycolytic/fast) fibers. They display marked differences in respect to contraction, metabolism, and susceptibility to fatigue. Type I fibers are mitochondria-rich and mainly use oxidative metabolism for energy production, which provides a stable and long-lasting supply of ATP, and thus are fatigue-resistant. Type II fibers comprise three subtypes, IIa, IIx, and IIb. Type IIb fibers have the lowest levels of mitochondrial content and oxidative enzymes, rely on glycolytic metabolism as a major energy source, and are susceptible to fatigue, while the oxidative and contraction functions of type IIa and IIx lie between type I and IIb (
Muscle fiber specification appears to be associated with obesity and diabetes. For instance, rodents that gain the most weight on high-fat diets possess fewer type I fibers (
We have previously established that peroxisome proliferator-activated receptor (PPAR) δ is a major transcriptional regulator of fat burning in adipose tissue through activation of enzymes associated with long-chain fatty-acid β-oxidation (
A role of PPARδ in muscle fiber was suggested by its enhanced expression—at levels 10-fold and 50-fold greater than PPARα and γ isoforms, respectively (unpublished data). An examination of PPARδ in different muscle fibers reveals a significantly higher level in type I muscle (soleus) relative to type II–rich muscle (extensor digitorum longus) or type I and type II mixed muscle (gastrocnemius) (
(A) Pooled RNA isolated from various muscles of five wild-type male C57B6 mice was hybridized with indicated probes. EDL, extensor digitorum longus; Gastro, gastrocnemius.
(B) Pooled nuclear proteins (15 μg/lane) isolated from muscles of five wild-type male C57B6 were probed with anti-PPARδ antibody. RNA polymerase II (Pol II) is shown as a loading control.
(C) Expression of the VP16-PPARδ transgene in various tissues. 10 μg of total RNA from each tissue was hybridized with a VP16 cDNA probe. Gastrocnemius muscle was used here.
(D) Nuclear proteins (15 μg/lane) isolated from gastrocnemius muscle of the transgenic mice (TG) and the wild-type littermates (WT) were probed with indicated antibodies. The upper, nonspecific band that cross-reacted with the anti-PPARδ antibody serves a loading control.
To directly assess the role of activation of PPARδ in control of muscle fiber plasticity and mitochondrial biogenesis, we generated mice expressing a transgene in which the 78-amino-acid VP16 activation domain was fused to the N-terminus of full-length PPARδ, under control of the 2.2-kb human α-skeletal actin promoter. In agreement with the previous characterization of this promoter (
Type I muscle can be readily distinguished from type II or mixed muscle by its red color, because of its high concentration of myoglobin, a protein typically expressed in oxidative muscle fibers. We found that muscles in the transgenic mice appeared redder (
(A and B) Muscles in transgenic mice (TG) are redder than those in wild-type mice (WT).
(C) Metachromatic staining of the type II plantaris muscle. Type I fibers are stained dark blue.
(A) Total RNA (10 μg/lane) prepared from gastrocnemius muscle of transgenic (TG) and wild-type (WT) littermates was probed with indicated probes. The fold increase of induction of each gene is shown.
(B) Total genomic DNA (10 μg/lane) prepared from gastrocnemius muscle was digested with Nco1 and subjected to Southern analysis with COXII (mitochondrial genome–encoded) and MCIP1 (nuclear genome–encoded) DNA probes.
(C) Equal amounts of gastrocnemius muscle were collected from both transgenic mice and control littermates. Total mitochondrial DNA was isolated and separated on 1% agarose gel. The relative abundance of mitochondrial DNA in transgenic and wild-type mice is presented.
(D) Western blot analysis of muscle fiber markers and mitochondrial components. Each lane was loaded with 80 μg of total gastrocnemius muscle extracts.
(E) Wild-type C57B6 mice were treated with vehicle or PPARδ agonist GW501516 for 10 d. Total RNA (10 μg/lane) prepared from the gastrocnemius muscle was probed with indicated probes.
A number of previous studies have shown that obese individuals have fewer oxidative fibers, implying that the presence of oxidative fibers alone may play a part in obesity resistance. To test this possibility, we fed the transgenic mice and their wild-type littermates with a high-fat diet for 97 d. Although the initial body weights of the two groups were very similar, the transgenic mice had gained less than 50% at day 47, and only one-third at day 97, of the weight gained by the wild-type animals (
(A) Seven-week-old transgenic (TG) and wild-type (WT) littermates (
(B) Histology of inguinal fat pad in the transgenic and wild-type littermates under a high-fat diet for 2 mo.
(C and D) Intramuscular glycogen content (C) and triglyceride content (D) of mice in (A) after high-fat feeding (
(E) Glucose tolerance test. Mice in (A) after high-fat feeding were fasted for 6 h and then injected with glucose at a concentration of 1g/kg body weight. Then blood glucose levels were measured periodically over 2 h (
(A) Eleven-week-old wild-type C57B6 mice were fed a high-fat diet in combination with vehicle or GW501516 for 57 d. Total RNA (10 μg/lane) prepared from the gastrocnemius muscle was probed with indicated probes.
(B) Net body weight gain for mice in (A) after treatment was calculated for individual mice and averaged. Initial body weights were 28.54 ± 1.04 g for vehicle group (
(C) Various tissue weights for mice in (A) after treatment. ifat, inguinal fat; rdfat, reproductive fat; retrofat, retroperitoneal fat.
(D) Glucose tolerance test. Mice in (A) after treatment were fasted for 6 h and then injected with glucose at a concentration of 1g/kg body weight. Blood glucose levels were then measured periodically over 2 h.
Muscle oxidative capacity is a crucial factor for determining endurance and fatigue. Indeed, type I fibers adaptively generated through exercise training are considered to be fatigue resistant. However, whether the type I fibers generated molecularly via PPARδ expression can contribute to enhanced performance in the absence of previous training is unclear. In fact, the consequence of genetically induced fiber switch on running capacity has to our knowledge never been evaluated. We thus compared exercise performance between untrained, body-weight-matched transgenic and wild-type littermates. Mice were run on oxygen-infused, enclosed treadmills until exhaustion. Strikingly, the running time and distance the transgenic mice were able to sustain were increased by 67% and 92%, respectively (
(A) Enhanced exercise performance in the transgenic mice. Fourteen-week-old male transgenic and wild-type littermates with similar body weights (
(B) Compromised exercise performance in PPARδ-null mice. Two-month-old male PPARδ-null mice and wild-type controls with similar body weights (
(C) Functions of PPARδ in skeletal muscle.
Our data reveal that a PPARδ-mediated transcriptional pathway can regulate muscle fiber specification, enabling the generation of a strain of mice with a “long-distance running” phenotype. We show that targeted expression of an activated form of PPARδ produces profound and coordinated increases in oxidation enzymes, mitochondrial biogenesis, and production of specialized type I fiber contractile proteins—the three hallmarks for muscle fiber type switching (
The muscle phenotypes described here are remarkably similar to those of transgenic mice expressing either calcineurin, calmodulin-dependent kinase, or PGC-1α (
Skeletal muscle is a major site to regulate whole-body fatty-acid and glucose metabolism. We show that mice with increased oxidative fibers are resistant to high-fat-induced obesity and glucose intolerance. Moreover, ligand studies provide compelling evidence that activation of endogenous PPARδ can produce similar effects. Might PPARδ have any beneficial effects on glucose metabolism in the lean condition? This has not been explored; however, insulin resistance in the elderly is confined mostly to skeletal muscle and may be due to reduction of mitochondrial number and/or function (
The transactivation domain (78 amino acid residues, corresponding to residues 413–490) of VP16 was fused in frame with the N-terminus of mouse PPARδ. The VP16-PPARδ fusion cDNA was placed downstream of the human α-skeletal actin promoter (
Mouse EST clones were obtained from ATCC (Manassas, Virginia, United States), verified by sequencing, and used as Northern probes. Antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, California, United States). Total muscle protein extracts (
Prior to the exercise performance test, the mice were accustomed to the treadmill (Columbus Instruments, Columbus, Ohio, United States) with a 5-min run at 7 m/min once per day for 2 d. The exercise test regimen was 10 m/min for the first 60 min, followed by 1 m/min increment increases at 15-min intervals. Exhaustion was defined when mice were unable to avoid repetitive electrical shocks.
Muscle fiber typing was essentially performed using metachromatic dye–ATPase methods as described (
Number of mice for each group used in experiments is indicated in figure legends. Values are presented as mean ± SEM. A two-tailed Student's t test was used to calculate
This video shows the exercise performance of a representative of the transgenic mice (right chamber) and a representative of wild-type control littermates (left chamber) on the treadmill 15 min into the exercise challenge.
(52.4 MB MOV).
This video shows the exercise performance of a representative of the transgenic mice (right chamber) and a representative of wild-type control littermates (left chamber) on the treadmill 90 min into the exercise challenge.
(41.7 MB MOV).
We thank Dr. M. Downes for providing the α-skeletal actin promoter; Dr. G. D. Barish for comments on the manuscript; M.A. Lawrence for histology; Dr. S. Pfaff for the use of microscopes and photographic equipment; M. Lieberman and K.L. Schnoeker for photography; J.M. Shelton for advice on fiber staining; and E. Stevens and E. Ong for administrative assistance. YXW was supported by a postdoctoral fellowship from California Tobacco-Related Disease Research Program. JH was supported by the BK21 program, Ministry of Education, Korea. HK was supported by grant M1–0311-00–0145 from the Molecular and Cellular BioDiscovery Research Program, Ministry of Science and Technology, Korea. RME is an Investigator of the Howard Hughes Medical Institute at the Salk Institute for Biological Studies and March of Dimes Chair in Molecular and Developmental Biology. This work was supported by the Howard Hughes Medical Institute and the Hillblom Foundation.
cytochrome c oxidase
muscle carnitine palmitoyltransferase-1
Peroxisome proliferator-activated receptor-gamma coactivator 1α
peroxisome proliferator-activated receptor
uncoupling protein