Total Skeletal Muscle PGC-1 Deficiency Uncouples
Mitochondrial Derangements from Fiber Type
Determination and Insulin Sensitivity
Christoph Zechner,1,3,6Ling Lai,1,3Juliet F. Zechner,1,3,6Tuoyu Geng,4,5Zhen Yan,4,5John W. Rumsey,3Deanna Collia,3
Zhouji Chen,1David F. Wozniak,2Teresa C. Leone,1,3and Daniel P. Kelly1,3,*
1Department of Medicine
2Department of Psychiatry
Washington University School of Medicine, St. Louis, MO 63110, USA
3Sanford-Burnham Medical Research Institute, Orlando, FL 32827, USA
4Department of Medicine
5Center for Skeletal Muscle Research at Robert M. Berne Cardiovascular Research Center
University of Virginia School of Medicine, Charlottesville, VA, 22908, USA
6Present address: Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, TX 75390, USA
Evidence is emerging that the PGC-1 coactivators
serve a critical role in skeletal muscle metabolism,
function, and disease. Mice with total PGC-1 defi-
ciency in skeletal muscle (PGC-1a?/?bf/f/MLC-Cre
mice) were generated and characterized. PGC-
1a?/?bf/f/MLC-Cremice exhibit a dramatic reduction in
exercise performance compared to single PGC-1a-
or PGC-1b-deficient mice and wild-type controls.
The exercise phenotype of the PGC-1a?/?bf/f/MLC-Cre
mice was associated with a marked diminution in
muscle oxidative capacity, together with rapid deple-
tion of muscle glycogen stores. In addition, the
PGC-1a/b-deficient muscle exhibited mitochondrial
structural derangements consistent with fusion/
fission and biogenic defects. Surprisingly, the propor-
tion of oxidative muscle fiber types (I, IIa) was not
reduced in the PGC-1a?/?bf/f/MLC-Cremice. Moreover,
insulin sensitivity and glucose tolerance were not
altered in the PGC-1a?/?bf/f/MLC-Cremice. Taken
together, we conclude that PGC-1 coactivators are
necessary for the oxidative
programs of skeletal muscle but are dispensable for
fundamental fiber type determination and insulin
cellular energy metabolism (Kelly and Scarpulla, 2004; Lin et al.,
2005). Originally discovered in brown adipose tissue (Puigserver
et al., 1998), PGC-1a enhances transcription by interacting
directly with target transcription factors, including PPARg and
other members of the nuclear receptor superfamily (reviewed in
Finck and Kelly, 2006; Handschin and Spiegelman, 2006). The
PGC-1 coactivators are enriched in tissues with high capacity
for mitochondrial respiratory function, such as brown adipose
tissue, heart, and skeletal muscle, and are highly inducible in
response to physiological stimuli, including exercise, fasting,
and cold exposure (Wu et al., 1999; Goto et al., 2000; Lehman
et al., 2000; Baar et al., 2002; Pilegaard et al., 2003; Finck and
Evidence is emerging that PGC-1 signaling plays an important
role in skeletal muscle structure and function (Lin et al., 2002a;
Arany et al., 2005, 2007; Leone et al., 2005; Lelliott et al., 2006;
Handschin et al., 2007). Tissue-specific transgenic approaches
in mice have shown that forced overexpression of PGC-1 coac-
tivators in skeletal muscle increases muscle oxidative capacity
and the proportions of type I and IIa fibers (Lin et al., 2002a).
PGC-1b has been shown to drive formation of type IIx fibers
(Arany et al., 2007). Using a skeletal-muscle-specific, doxycy-
cline-regulated system, we found that acute induction of
PGC-1a expression in mice increased muscle glycogen stores,
a signature of training (Wende et al., 2007). Collectively, the
results of the forced expression studies suggest that induction
of PGC-1a, which is known to occur following exercise, drives
a trained skeletal muscle phenotype. However, gain-of-function
mental skeletal muscle development, fiber type specification, or
mitochondrial biogenesis. Insights gained from PGC-1 coactiva-
tor mouse ‘‘knockout’’ studies to date have been limited due to
the functionally redundant actions of PGC-1a and PGC-1b;
single PGC-1 gene deletion lines exhibit relatively minimal
phenotypes (Lin et al., 2004; Leone et al., 2005; Lelliott et al.,
2006; Vianna et al., 2006; Sonoda et al., 2007). Mice with gener-
(PGC-1ab?/?mice) die of severe heart failure shortly after birth,
precluding evaluation of the skeletal muscle phenotype in this
model (Lai et al., 2008).
cated in the development of skeletal muscle insulin resistance
(Mootha et al., 2003; Patti et al., 2003), although this notion is
controversial. Single PGC-1a or PGC-1b ‘‘knockout’’ mouse
Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc. 633
models have generally not demonstrated insulin-resistant or
glucose-intolerant phenotypes (Lin et al., 2004; Leone et al.,
2005; Lelliott et al., 2006; Vianna et al., 2006; Sonoda et al.,
2007). Again, the difficulty in interpreting these studies relates
to the functional overlap between the PGC-1 coactivators,
setting the stage for compensatory or even supercompensatory
responses by the remaining coactivator.
and their potential role in the metabolic basis of disease, we em-
barked on a study to define the necessary functions of these
regulatory factors in skeletal muscle using a conditional loss-
of-function strategy in mice. To this end, a gene-targeting
approach was used to establish mice with skeletal-muscle-
specific PGC-1b deficiency on a generalized PGC-1a-deficient
background. We found that combined deficiency of PGC-1a
and PGC-1b in skeletal muscle results in a dramatic exercise
performance phenotype related to mitochondrial structural and
functional abnormalities. Surprisingly, PGC-1a?/?bf/f/MLC-Cre
mice did not exhibit a defect in fundamental muscle fiber type
formation or abnormalities in glucose tolerance or insulin
Dramatic Exercise Deficit in Mice with Muscle
The PGC-1b gene was disrupted via loxP-mediated excision in
skeletal muscle by Cre recombinase (Cre) expressed under the
control of the myosin light chain 1f (MLC1f) promoter (PGC-
1bf/f/MLC-Cremice) (Figure S1A). Consistent with the type II fiber
specificity of the MLC-1f promoter (Donoghue et al., 1991;
Parsons et al., 2004), PGC-1b gene expression was markedly
reduced in gastrocnemius (GC) (residual levels 10.9% ± 1.3%
of PGC-1bf/f) and white vastus (9.0% ± 1.5% of PGC-1bf/f)
muscles of PGC-1bf/f/MLC-Cremice, but to a lesser extent in
soleus muscle (44% ± 8.5% of PGC-1bf/f) (Figure S1B), which
contains a greater proportion of type I fibers. Immunoblotting
studies confirmed that PGC-1b protein levels were undetectable
mice (Figure S1C). PGC-1b gene expression was not reduced in
heart or liver in the PGC-1bf/f/MLC-Cremice (Figure S1B). PGC-1a
mRNA levels were unchanged in the muscle of the PGC-
1bf/f/MLC-Cremice (data not shown). PGC-1bf/f/MLC-Cremice
survived, appeared normal compared to WT littermates, and
did not exhibit overt abnormalities in locomotor activity. To
generate lines with skeletal muscle deficiency of both PGC-1a
and PGC-1b, the PGC-1bf/f/MLC-Cremice were crossed into
a generalized PGC-1a?/?background (Leone et al., 2005),
a strategy that was used previously to generate mice with
combined deficiency of PGC-1a and PGC-1b in heart (Lai
et al., 2008). Mice with combined PGC-1a and PGC-1b defi-
ciency in skeletal muscle (PGC-1a?/?bf/f/MLC-Cremice) were
viable, appeared normal on inspection, and did not exhibit any
overt abnormalities in locomotor activity. Fasting circulating
triglyceride and glucose levels were not abnormal in the PGC-
1a?/?bf/f/MLC-Cremice (data not shown).
a mild exercise performance phenotype (Leone et al., 2005;
Handschin et al., 2007; Lai et al., 2008). To define the impact
of muscle-specific PGC-1b deficiency and combined PGC-
1a/PGC-1b deficiency on exercise performance, the PGC-
1bf/f/MLC-Creand PGC-1a?/?bf/f/MLC-Cremice were subjected to
a low-intensity run-to-exhaustion protocol on a motorized tread-
mill. Compared to the performance of ‘‘floxed’’ controls (170.2 ±
7.1 min), PGC-1a?/?bf/fand PGC-1bf/f/MLC-Cremice demon-
strated a mild but significant performance deficit (mean running
times of 114 ± 7.1 min and 140 ± 8.8 min, respectively) (Figure 1).
In striking contrast, PGC-1a?/?bf/f/MLC-Cre
a dramatic exercise phenotype, showing exhaustion after an
average of only 6.6 ± 1.6 min (Figure 1).
Histological assessment of the GC muscle of PGC-1a?/?
bf/f/MLC-Cremice did not reveal any overt cellular derangements
or fibrosis (Figure S1D and data not shown). However, a very
slight increase in the proportion of myocytes with centralized
nuclei was noted in PGC-1a?/?bf/f/MLC-Cremuscle fibers, indica-
tive of increased muscle fiber regeneration (Figure S1E). As
a measure of muscle damage, circulating creatine kinase (CK)
was assayed in PGC-1a?/?bf/f/MLC-Cre
30 min following a bout of exercise (protocol described in
Supplemental Experimental Procedures). Mean plasma CK
levels were mildly but significantly elevated in PGC-1a?/?
bf/f/MLC-Cremice compared to the other groups (bf/f, 24.0 ±
3.5 U/l; a?/?bf/f, 23.5 ± 2.3 U/l; bf/f/MLC-Cre, 31.1 ± 6.3 U/l;
a?/?bf/f/MLC-Cre, 103 ± 37.4 U/l; p < 0.004), possibly consistent
with mild skeletal muscle injury. Taken together, these results
are unlikely to account for the profound exercise performance
deficit of the PGC-1a?/?bf/f/MLC-Creline.
The exercise phenotype of the PGC-1a?/?bf/f/MLC-Cremice
prompted us to assess general ambulatory activity and other
parameters that may influence ambulation and exercise perfor-
mance. Using a 1 hr locomotor activity test, both PGC-1a?/?bf/f
and PGC-1a?/?bf/f/MLC-Crelines exhibited a similar, modest
compared to the PGC-1bf/fgroup (Figure S1F, left). Time spent in
the central and peripheral zones of the test area, as a measure of
exploratory behavior, was also recorded and analyzed. PGC-
1a?/?bf/fand PGC-1a?/?bf/f/MLC-Cremice traveled significantly
shorter distances within the peripheral zone (Figure S1F, middle)
and entered the center less frequently (Figure S1F, right) when
mice and controls
Running Time (min)
Figure 1. Global Loss of PGC-1a Combined with Muscle-Specific
Loss of PGC-1b Results in a Dramatic Exercise Performance Deficit
Two- to three-month-old male PGC-1bf/f(ab+/+), PGC-1a?/?bf/f(a?/?),
PGC-1bf/f/MLC-Cre(b?/?), and PGC-1a?/?bf/f/MLC-Cre(ab?/?) mice were sub-
jected to a run-to-exhaustion protocol on a motorized treadmill (n = 6–7) as
described in Experimental Procedures. Bars represent mean running time
(± SEM) in minutes. *p < 0.05 versus ab+/+; zp < 0.05 versus a?/?; #p < 0.05
versus b?/?. See also Figure S1.
PGC-1 Actions in Skeletal Muscle
634 Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc.
compared to the PGC-1bf/fgroup. These collective results are
consistent with our earlier findings that PGC-1a deficiency results
in a mild decrease in general ambulatory activity and may induce
type was not more severe in the PGC-1a?/?bf/f/MLC-Cremice.
Interestingly, muscle strength, as assessed by forelimb grip
studies, did not reveal differences between groups (Figure S1F,
left). In addition, PGC-1a?/?bf/f/MLC-Cremice did not differ from
controls on ledge, platform (data not shown), or inverted screen
tests (Figure S1G, right). Thus, PGC-1a?/?bf/f/MLC-Cremice do
not exhibit abnormalities in coordination, balance, or muscle
strength beyond the phenotype that can be attributed to gener-
alized PGC-1a deficiency.
PGC-1a/b Deficiency Uncouples Muscle Oxidative
Capacity from Fiber Type Determination
To further evaluate the basis for the exercise phenotype in PGC-
1a?/?bf/f/MLC-Cremice, formal skeletal muscle fiber typing was
Fiber Type %
Fiber Type %
αβ αβ+/+ +/+
Fibers without a Change in Type II Fiber
(A) Cross-sections of gastrocnemius (Gastroc)
muscle of 3- to 6-month-old male PGC-1bf/f
(b?/?), and PGC-1a?/?bf/f/MLC-Cre(ab?/?) mice
(n = 5–6) stained for myosin I ATPase activity.
Type I (MHC1-positive) fibers are stained dark.
See also Figure S2.
(B) Top: Sections of plantaris muscle of 3- to
4-month-old male mice (n = 5–7) were immuno-
stained for MHC1 (red), MHC2a (blue), and
MHC2b (green). Representative
shown. Scale bars denote 200 mm. Bottom: Quan-
tification results of the plantaris muscle MHC im-
munostaining studies expressed as mean percent
(± SEM) of total muscle fibers. Unstained muscle
fibers were counted as MHC2x positive. Insert
shows magnification of MHC1 results. *p < 0.05
versus ab+/+; zp < 0.05 versus a?/?; #p < 0.05
(C) Succinate dehydrogenase (SDH) activity stain-
ing was performed on sections of gastrocnemius
muscle (top row; n = 5–6) and soleus muscle
(bottom row; n = 2) of 3- to 6-month-old male
mice of the indicated genotypes.
conducted. First, fiber type distribution
was examined in GC muscle using meta-
chromatic ATPase staining under acidic
conditions, which are permissive for
myosin heavy chain 1 (MHC1) ATPase
ingly, analysis of the oxidative regions of
GC muscle demonstrated a mild increase
in numbers of MHC1-positive muscle
fibers in PGC-1a?/?bf/f/MLC-Cremuscles
compared to the other groups (Figure 2A).
We next performed quantitative immu-
nohistochemical analyses of fiber type
distribution in the muscle of the four
genotypes. These studies were initially conducted on plantaris
muscle, which exhibits less regional heterogeneity in oxidative
fiber type compared with GC. Again, higher proportions of
MHC1-positive fibers were noted in PGC-1a?/?bf/f/MLC-Cre
muscle when compared with the other genotypes (Figure 2B).
among the groups (Figure 2B). Quantitative analysis of MHC
mRNA transcripts in GC muscle of PGC-1a?/?bf/f/MLC-Cremice
revealed an increase in MHC1 mRNA levels in PGC-1a?/?bf/f
and PGC-1a?/?bf/f/MLC-Cremice and a modest but significant
decrease in MHC2a and 2x mRNA levels in PGC-1a?/?
bf/f/MLC-Cremuscle (Figure S2). These latter results suggest that
posttranscriptional compensatory mechanisms may contribute
to the maintenance of normal type II fiber distribution in the
muscle of PGC-1a?/?bf/f/MLC-Cremice.
We next assessed the oxidative capacity of the PGC-
1a?/?bf/f/MLC-Cremuscle. For these studies, activity of the mito-
chondrial enzyme succinate dehydrogenase (SDH) was used
PGC-1 Actions in Skeletal Muscle
Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc. 635
as an endpoint. SDH activity staining was only minimally
reduced in PGC-1a?/?bf/fand PGC-1bf/f/MLC-CreGC and soleus
muscle compared to PGC-1bf/fcontrols (Figure 2C). In striking
contrast, SDHstaining was
PGC-1a?/?bf/f/MLC-Cremuscle (Figure 2C). Thus, PGC-1 coacti-
vators are required to maintain skeletal muscle oxidative
capacity but not fiber type determination.
The fiber typing studies described above only allowed us to
explore theeffectsof PGC-1a/PGC-1bdeficiencyintypeIIfibers
given the fiber specificity of MLC1f-Cre. To assess the effects of
combined PGC-1a/PGC-1b deficiency on type I fibers, PGC-
1a?/?bf/fmice were bred with mice expressing a muscle creatine
kinase (MCK) promoter-driven Cre (MCK-Cre) to establish the
PGC-1a?/?bf/f/MCK-Creline, in which the PGC-1b gene is disrup-
ted in both fast- and slow-twitch muscle fiber types in a PGC-
1a-deficient background. As expected, expression of the
PGC-1bgene was significantly reduced in fast and slowmuscles
in PGC-1a?/?bf/f/MCK-Cremice (Figure S3). As we found for the
PGC-1bf/f/MLC-Cremice, PGC-1a mRNA levels were unchanged
in the muscle of the PGC-1bf/f/MCK-Cremice (data not shown).
The PGC-1a?/?bf/f/MCK-Creline also had significant reduction in
cardiac PGC-1b expression, which resulted in cardiomyopathy
by 1 month of age (data not shown), a finding that was predicted,
postnatal heart failure (Lai et al., 2008). As observed in
PGC-1a?/?bf/f/MLC-Cremuscle, type I fiber type proportion was
slightly increased without changes in the relative proportions of
the other fiber types in PGC-1a?/?bf/f/MCK-Cremice, based on
metachromatic staining and MHC immunohistochemistry per-
with the results obtained from the PGC-1a?/?bf/f/MLC-Cremice,
these data strongly suggest that PGC-1a and PGC-1b are
dispensable for skeletal muscle development and fundamental
fiber type determination.
Mitochondrial Structural and Functional Derangements
Transcriptional profiling studies were conductedusing GCto gain
We were particularly interested in the subset of genes that were
downregulated only in the doubly deficient mice. The number of
genes (array probes) that were downregulated uniquely or shared
by the three genotypes is displayed in the Venn diagram (Fig-
ure 4A). Of the 5526 genes that were downregulated in the PGC-
1a?/?bf/f/MLC-Cre, 2839 were uniquely downregulated compared
to the other two genotypes. In addition, 1086 genes were upregu-
lated in PGC-1a?/?bf/f/MLC-Cremuscle (data not shown).
Gene ontology (GO) analysis demonstrated that mitochondrial
pathways were rarely downregulated in singly PGC-1a- or PGC-
1b-deficient muscle (4/229 or 1.7% of total pathways). In
contrast, 47/202 (or 23%) of total pathways downregulated
only in the double knockout muscle were involved in mitochon-
drial processes, including the TCA cycle, electron transport,
OXPHOS, and mitochondrial membrane organization and
biogenesis (Table S1, shaded pathways). Only two pathways
were significantly upregulated in the double knockout, neither
of which was mitochondrial. Quantitative RT-PCR validation
studies were conducted on a subset of genes representative of
the latter pathways (identified as uniquely downregulated in the
enzymes involved in the TCA cycle (citrate synthase [Cs]), elec-
tron transport chain/OXPHOS (cytochrome oxidase subunit IV
isoform 1 [Cox4i1] and ATP synthase F1 complex beta subunit
[ATP5b]), mitochondrial membrane transport (adenine nucleo-
tide translocator 1 [Ant1]), and reactive oxygen species
scavenging (mitochondrial superoxide dismutase 2 [Sod2])
were downregulated in PGC-1a?/?bf/fmuscle (and some in
PGC-1bf/f/MLC-Cre) and further reduced in PGC-1a?/?bf/f/MLC-Cre
muscle when compared to PGC-1bf/fcontrols (Figure 4B).
However, levels of mRNA encoding other enzymes, such as
those involved in fatty acid oxidation (carnitine palmitoyltransfer-
ase 1b [Cpt1b] and medium-chain acyl-coenzyme A dehydroge-
nase [Acadm]), was reduced in PGC-1a?/?bf/fmuscle, but not
further downregulated in the double knockout muscle. Interest-
ingly, expression of the gene encoding nuclear respiratory factor
1 (NRF-1), a PGC-1a target (Wu et al., 1999) involved in regu-
lating OXPHOS genes and coordinating nuclear and mitochon-
drial genomes, was similar among the groups. This latter result
is consistent with the findings of a previous study showing no
effect of loss of PGC-1a on NRF-1 expression in the heart (Arany
et al., 2005) or skeletal muscle (Geng et al., 2010). Taken
mitochondrial energy transduction pathways, are shared by the
PGC-1 coactivators. It should also be noted that within the
shared target gene data set, loss of PGC-1a alone often had
a greater impact than PGC-1b deficiency (Figure 4B).
Further review of the gene expression array data revealed
α αβ β+/++/+
Fiber Type %
Figure 3. PGC-1a–/–bf/f/MCK-CreMuscle Exhibits a Mild Increase
in MHC1-Positive Fibers without a Change in Type II Fiber Formation
mice (see also Figure S3) stained for myosin ATPase activity.
(B) Quantification of IHC immunostaining studies expressed as mean percent
(± SEM) of total muscle fibers (n = 3–5/group). *p < 0.05 versus ab+/+; zp <
0.05 versus a?/?; #p < 0.05 versus b?/?.
PGC-1 Actions in Skeletal Muscle
636 Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc.
including fusion and fission, were also downregulated in
PGC-1a?/?bf/f/MLC-Cremuscle (GO ID 7006, 5741, 5743, 31966)
(Table S1). Validation studies confirmed that the expression of
mitofusin (Mfn) 1 and 2 genes and the gene encoding dyna-
min-related protein 1 (Drp1, dnm1l), a key mitochondrial fission
protein, was significantly downregulated in the muscle of the
PGC-1a?/?bf/f/MLC-Cremice compared to the other groups (Fig-
ure 5A). These results suggested that the PGC-1 coactivators
are necessary for mitochondrial dynamics and quality control
in skeletal muscle. Therefore, mitochondrial volume density,
size, and ultrastructure were assessed by electron microscopy
in the intermyofibrillar (IM) and subsarcolemmal (SS) compart-
revealed striking mitochondrial morphologic abnormalities in IM
the other genotypes, including a reduction in density and hetero-
geneity of size (Figure 5B and data not shown). The muscle mito-
chondria of PGC-1a?/?bf/f/MLC-Cremice contained many small,
fragmented mitochondria juxtaposed to elongated, thin mito-
chondria (arrows, Figure 5B). Mitochondrial DNA levels were
reduced in parallel with the observed ultrastructural derange-
ments among the genotypes (Figure S4). Taken together, these
results implicate the PGC-1 coactivators in muscle mitochon-
drial biogenesis and fusion/fission programs.
Figure 4. PGC-1a and b Drive an Overlap-
ping Subset of Target Genes Involved in
(A) Venn diagram displaying the results of gene
expression microarray analysis conducted on
RNA isolated from PGC-1a?/?bf/f(a?/?), PGC-
(ab?/?) GC musclecompared
controls. The numbers denote number of downre-
gulated gene probes (%0.7) in the corresponding
(B) Levels of mRNAs based on quantitative real-
time RT-PCR performed on RNA isolated from
GC muscle of 3- to 4-month-old male mice from
the genotypes indicated (n = 8–10 per group).
Citrate synthase, Cs; cytochrome c oxidase
subunit 4 isoform 1, Cox4i1; beta polypeptide of
the H+transporting mitochondrial F1 complex
ATP synthase, Atp5b; adenine nucleotide translo-
cator 1, Ant1 = Slc25a4; mitochondrial superoxide
dismutase 2, Sod2; the muscle isoform of carni-
tine-palmitoyl transferase 1, Cpt1b; medium chain
acyl-CoA dehydrogenase, Acadm; and nuclear
respiratory factor 1, Nrf1. Bars represent mean
values (± SEM) normalized to 36B4 mRNA levels
and expressed relative (= 1.0) to PGC-1bf/f(ab+/+)
muscle. *p < 0.05 versus ab+/+; zp < 0.05 versus
a?/?; #p < 0.05 versus b?/?.
Respiration rates were determined on
mitochondria isolated from the hindlimb
of all four genotypes using pyruvate
or palmitoylcarnitine as substrates. Con-
derangements, state 3 respiration rates
were markedly reduced in PGC-1a?/?
bf/f/MLC-Cremuscle compared to the WT
controls (Figure 5C). State 3 rates in mitochondria from single
PGC-1 gene-deficient muscle were reduced to a level midway
between the WT controls and that of PGC-1a?/?bf/f/MLC-Cremice
Muscle PGC-1a/b Deficiency Does Not Affect Glucose
Tolerance or Insulin Resistance
We next sought to assess glucose utilization and insulin respon-
siveness in the PGC-1a?/?bf/f/MLC-Cremice. Mice of all four
chow, we compared results obtained with PGC-1bf/fand
PGC-1bf/f/MLC-Creor PGC-1a?/?bf/fand PGC-1a?/?bf/f/MLC-Cre
so that littermate comparisons were possible. Neither the
nor the double knockout (PGC-1a?/?bf/f/MLC-Cre) (Figure 6B) line
exhibited glucose intolerance or insulin resistance when
compared with their respective controls. The studies were then
repeated after an 8 week high-fat-diet regimen to determine
dard diet, there was no difference in ITT or GTT among the geno-
types on the high-fat diet (Figure 6C). Inaddition, fasting levels of
glucose, insulin, free fatty acids, and triglycerides were not
different among the groups (Figure S5). Skeletal muscle
PGC-1 Actions in Skeletal Muscle
Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc. 637
triglyceride levels were also similar among the four genotypes
To further evaluate muscle glucose utilization in the PGC-
1a?/?bf/f/MLC-Cremice, glycogen levels were assessed pre- and
postexercise. For these experiments, the four genotype groups
were run for a duration matched to the time required for the
glycogen levels in the GC did not differ between groups. The
bout of exercise rapidly depleted stores in PGC-1a?/?bf/f/MLC-Cre
mice but did not significantly change muscle glycogen levels in
the PGC-1bf/f, PGC-1a?/?bf/f, and PGC-1bf/f/MLC-Cregroups (Fig-
ure 7A). These results suggested that the PGC-1a?/?bf/f/MLC-Cre
muscle relies largely on anaerobic glycolysis for ATP production
observation that blood lactate levels increased markedly in the
compared to that of WT and PGC-1bf/f/MLC-Creafter exhaustive
exercise (Figure 7B). Blood lactate levels also increased in
the PGC-1a?/?bf/fmice, but to a lesser extent than the PGC-
1a?/?bf/f/MLC-Cregroup. These results are consistent with the
observation of normal glucose tolerance and help explain the
severe exercise deficit, given that glucose derived from glycogen
serves as an obligate fuel source during exercise.
is supportedby the
We devised a strategy to disrupt the PGC-1b gene in a skeletal-
muscle-selective manner in a generalized PGC-1a-deficient
mouse background. As described herein, we found that PGC-
1a and PGC-1b serve overlapping roles in skeletal muscle and
together are necessary for exercise performance (even minimal
exertion) and maintenance of mitochondrial structure and func-
tion, but are dispensable for fundamental fiber type determina-
tion and normal muscle insulin sensitivity and glucose disposal.
Our findings demonstrated that the cooperative and overlap-
ping actions of PGC-1a and PGC-1b are required for muscle
mitochondrial function and structure. Transcriptional profiling
studies revealed that the expression of a broad array of genes
involved in multiple mitochondrial energy transduction and
OXPHOS pathways is dependent on having at least one func-
tional PGC-1 gene. PGC-1a/b deficiency resulted in dramatic
derangements in mitochondrial structure and respiratory func-
tion, indicating that the PGC-1 coactivators are required for
maintaining a healthy population of normal mitochondria. This
latter function has been ascribed to the ongoing dynamics of
fusion and fission, which serve to maintain mitochondrial quality
control (Chen et al., 2005; Waterham et al., 2007; Zhang and
Chan, 2007). Consistent with this conclusion, we found that
the expression of genes involved in fusion (Mfn1, Mfn2)
and fission (Drp1) was downregulated in the muscle of
PGC-1a?/?bf/f/MLC-Cremice compared to the other genotypes.
Others have recently shown that PGC-1a and PGC-1b are
capable of activating transcription of the Mfn2 gene by coacti-
vating the nuclear receptor ERRa (Soriano et al., 2006; Liesa
et al., 2008). Taken together, these results suggest that the
PGC-1 coactivators are necessary for maintaining high-level
coordinated expression of genes involved in mitochondrial
energy transduction and ATP production pathways, as well as
maintenance of a healthy population of muscle mitochondria.
Mfn1 Mfn2Drp1 Fis-1
Norm. Arbitrary Units
nmol O2/min/mg protein
‡ # *
‡ # *
Figure 5. PGC-1a–/–bf/f/MLC-CreMuscle Exhibits Mitochondrial Struc-
tural and Functional Abnormalities and Dysregulation of Genes
Involved in Mitochondrial Dynamics
(A) Levels of mRNAs encoding mitochondrial fission and fusion genes,
including mitofusin 1 (Mfn1), mitofusin 2 (Mfn2), dynamin-related protein 1
(Drp1 or dmn1l), and mitochondrial outer membrane fission 1 homolog (yeast)
(Fis-1), determined by quantitative real-time RT-PCR (n = 7–8 per group). Bars
represent mean values (± SEM) normalized to 36B4 mRNA levels and
expressed relative (= 1.0) to PGC-1bf/f(ab+/+) muscle.
(B)Representative electron micrographs of gastrocnemiusshowingsubsarco-
lemmal mitochondria in sections from PGC-1a?/?bf/fmice (a?/?; top row) and
PGC-1a?/?bf/f/MLC-Cremice (middle row and magnified in bottom row). Note
small, fragmented (white arrowhead), and elongated (black arrowhead) mito-
chondria. Scale bars represent 500 nm.
(C) Bars represent mean (± SEM) mitochondrial respiration rates determined
shown using pyruvate or palmitoylcarnitine as a substrate (4–6 male animals
per group). Rates were measured under the following conditions: basal, state
3 (ADP-stimulated), and after oligomycin treatment (oligo-induced State 4).
*p < 0.05 versus ab+/+; zp < 0.05 versus a?/?; #p < 0.05 versus b?/?. See
also Figure S4.
PGC-1 Actions in Skeletal Muscle
638 Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc.
The impact of combined PGC-1a and PGC-1b deficiency on
exercise performance was dramatic. Our results strongly
suggest that the exercise phenotype is due to severe mitochon-
drialdysfunction forcing reliance onanaerobic glycolysis for ATP
production, leading to rapid depletion of glycogen and prema-
ture fatigue. The marked elevation in circulating lactate levels
in the PGC-1a/b-deficient mice postexercise is consistent with
this conclusion. Interestingly, lactate levels were also increased
postexercise in the PGC-1a?/?mice, albeit to a lesser extent
than the PGC-1a/b-deficient animals. This latter observation
could reflect the fact that hepatic PGC-1a deficiency results in
an altered Cori cycle response. It is also possible that PGC-1 co-
activators play a primary role in glycogen metabolism. However,
the protein levels of the enzymes involved in glycogen synthesis
and degradation were not significantly altered in the PGC-
1a?/?bf/f/MLC-Cremice (data not shown).
able uncoupling of muscle fiber type and oxidative programs.
Specifically, phenotypic analysis of the PGC-1a?/?bf/f/MLC-Cre
AUC x 1000
0 3060 90120
Blood Glucose [mg/dl]
AUC x 1000
Blood Glucose [mg/dl]
0 30 60 90120
0 3060 90120
Blood Glucose (% t0)
Blood Glucose (% t0)
0 306090 120
0 1530 456090
Blood Glucose [mg/dl]
Blood Glucose (% t0)
High Fat Chow
Time (min)Time (min)
αβ αβ+/+ +/+
AUC x 1000
Figure 6. Global Loss of PGC-1a Combined
with Muscle-Specific Loss of PGC-1b Does
Not Result in Glucose Intolerance or Insulin
(A) Mean blood glucose (± SEM) levels during GTT
(left) and ITT (right) using 2- to 3-month-old male
littermate PGC-1bf/f(ab+/+) and PGC-1bf/f/MLC-Cre
(b?/?) mice on standard chow as described in the
Experimental Procedures (n = 9–10). Total area
under the glucose excursion curve (± SEM) is dis-
played in the inset for the GTT.
(B) GTT and ITT results (mean blood glucose ±
SEM) for 2- to 3-month-old male littermate PGC-
1a?/?bf/f(a?/?) and PGC-1a?/?bf/f/MLC-Cre(ab?/?)
mice on standard chow (n = 11–13 per group).
(C) Results of GTT (left) and ITT (right), expressed
as mean blood glucose ± SEM, conducted on
13-week-old male mice after 8 weeks of high-fat
diet (n = 3–6/group). See also Figure S5.
mice did not reveal a shift toward a
‘‘detrained’’ muscle fiber type profile, as
would be predicted by PGC-1 overexpres-
sion studies. Rather, we found a modest
increase in type I fibers in the muscles of
increase in type I fibers in the mutant mice
suggests that an independent, currently
unidentified pathway involved in funda-
mental fiber type is compensatorily acti-
vated. PGC-1 overexpression in mice has
been shown to inhibit the protein degrada-
tion known to occur in disuse atrophy
(Wenz et al., 2009; Brault et al., 2010).
However, we did not find an overt abnor-
mality in muscle mass in the PGC-
PGC-1 coactivators are capable of driving
a shift toward oxidative fibers and muscle
or fiber type specification. One caveat to our conclusions is that
a small amount of a mutant form of a naturally occurring, alterna-
tively spliced form of PGC-1a, referred to as NT-PGC-1a (Zhang
et al., 2009), could be theoretically expressed in the PGC-1a-defi-
cient line used in this study. However, RT-PCR analysis demon-
aswe have shown inthe generalized PGC-1a-deficient line (Leone
phenotype shown here, it would seem unlikely that this could
account for a completely normal fiber phenotype.
Significant evidence suggests a link between skeletal muscle
mitochondrial dysfunction and the development of insulin resis-
tance (Lowell and Shulman, 2005; Morino et al., 2005; Petersen
et al., 2005). In addition, several studies in humans have shown
reduced expression of PGC-1a (Mootha et al., 2003; Patti
et al., 2003) in skeletal muscle of insulin-resistant and diabetic
dysfunction and altered PGC-1 signaling in the pathogenesis
PGC-1 Actions in Skeletal Muscle
Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc. 639
of insulin resistance. The studies shown herein demonstrate that
total skeletal muscle PGC-1 deficiency in mice does not lead to
insulin resistance or glucose intolerance on standard chow or
high-fat diet. Moreover, given the mitochondrial derangements
present in PGC-1a?/?bf/f/MLC-Cremuscle, our results suggest
that mitochondrial dysfunction is unlikely to contribute to insulin
resistance. It is certainly possible, however, that insulin resis-
tance drives muscle mitochondrial dysfunction.
Generation of PGC-1bf/fmice has been described elsewhere (Lai et al., 2008).
PGC-1bf/fmice were crossed with mice expressing Cre under control of
a MLC1f promoter (Bothe et al., 2000) or the MCK promoter (Bru ¨ning et al.,
1998) to achieve muscle-specific deletion of PGC-1b. These mice in turn
were crossed to obtain mice homozygous for the loss of PGC-1a (Leone
et al., 2005) and the PGC-1b flox allele (PGC-1a?/?bf/fmice), with and without
MLC1f-Cre. (See Supplemental Experimental Procedures for more detail
regarding the breeding strategy.)
Mice were maintained in the hybrid background, C57BL6/J x sv129, and
littermate controls were used. PGC-1bf/fand PGC-1bf/f/MLC-Cremice were
generated fromthe same breedingpairs,while PGC-1a?/?bf/f
PGC-1a?/?bf/f/MLC-Cremice were generated from a different set of breeding
pairs. Male mice were used for all experiments unless otherwise stated.
Mice were allowed ad libitum access to standard laboratory rodent chow
(diet 5053, Purina Mills Inc.) or high-fat diet (TD97268, Harlan Teklad) as indi-
cated. All animal experiments and euthanasia protocols were conducted in
strict accordance with the National Institutes of Health guidelines for humane
treatment of animals and were reviewed and approved by the Institutional
Animal Care and Use Committees of Washington University School of Medi-
cine and the Sanford-Burnham Medical Research Institute.
Glucose and Insulin Tolerance Testing
Prior to studies, 2- to 3-month-old male mice on standard chow were fasted
overnight (GTT) or for 6 hr (ITT). For GTT studies, mice were injected with
1 g/kg of D-glucose. For ITT, mice received an intraperitoneal injection of
human regular insulin (Eli Lilly and Co.) at a dose of 0.75 U/kg body weight
for ITT. The tip of the tail (?1 mm) was cut 60 min prior to the glucose or insulin
challenge forblood sampling.Blood glucoselevels weredeterminedatseveral
different time points after challenge using an Accu-Chek Advantage glucome-
ter (Roche). For GTT, area under the curve (AUC) was defined as the difference
between baseline glucose levels and the deflection caused by the glucose
challenge. Total AUC was calculated using the trapezoidal rule. For the high-
fat-diet study, mice were given high-fat diet beginning at 5 weeks of age. At
13 weeks of age, GTT and ITT were performed following a 5 hr fast. A dose
of 2 g/kg D-glucose was used for the GTT.
Blood and Tissue Chemistry
After an 8 week high-fat-diet regimen, mice were fasted overnight (16 hr)
beginning at 5 p.m. Blood samples (?1 ul) were obtained from the tail for
measurement of blood glucose using an Accu-Chek Advantage glucometer
(Roche). Additional blood (?50 ul) was obtained for insulin measurements.
Plasma insulin levels were determined by ELISA (Mercodia, Inc.) following
the manufacturer’s instructions. Mice were euthanized by CO2inhalation after
10 weeks on high-fat diet, and plasma and tissue triglyceride concentrations
were determined by colorimetric assays as previously described (Chen
et al., 2008). Plasma free fatty acids were also determined by a colorimetric
assay (Lin et al., 2002b).
Evaluation of Exercise Performance
Two- to three-month-old mice were run to exhaustion employing a motorized,
speed-controlled, modular treadmill system (Columbus Instruments). The
treadmill was equipped with an electric shock stimulus and an adjustable incli-
nation angle. Running velocity was set to 10 m/min for 1 hr and increased by
2 m/min increments every 15 min. The inclination angle was level. Tail blood
was taken before and immediately following the exhaustion point, and lactate
levels were analyzed using a Lactate Pro blood lactate test meter (ARKRAY,
Inc.). For the assessment of muscle glycogen and serum CK values, male
PGC-1a?/?bf/f/MLC-Cremice were run to exhaustion, and the exercise
time of controls was matched to the running time of the respective PGC-
1a?/?bf/f/MLC-Cremouse. Mice used for glycogen measurements were eutha-
nized immediately after the running bout while mice used for serum CK
measurements were harvested 30 min after the exercise bout. For details on
Evaluation of Muscle Function and Locomotor Activity
Musclestrength,muscle use,andgeneral activitylevels wereassessedin3-to
4-month-old female mice. For the assessment of forelimb grip strength, a grip
strength meter was used as previously described (Wozniak et al., 2007). (See
Supplemental Experimental Procedures for more detail.)
RNA, DNA, and Protein Analysis
Quantitative real-time PCR was performed on total RNA as previously
described (Huss et al., 2004). The mouse-specific primer-probe sets used to
detect specific gene expression can be found in Table S2. 36B4 mRNA was
measured in a separate well (in triplicate) and used to normalize the gene
Genomic/mitochondrial DNA was isolated using the RNAzol method, fol-
lowed by back extraction with 4 M guanidine thiocyanate, 50 mM sodium
(μ μmol glucose/g tissue)
‡ # *
‡ # *
Figure 7. Combined Loss of Skeletal Muscle PGC-1a and b Results
in Rapid Depletion of Muscle Glycogen during Exercise
(A) Meanglycogen levels(±SEM)ofGCmuscle ofsedentary2-to3-month-old
male mice of the indicated genotypes are shown on the left. The results of GC
glycogen levels postexercise are shown on the right.
(B) Mean blood lactate levels (± SEM) at baseline (white bars) and following
exhaustive exercise (black bars) for all four genotypes (n = 5–10/group).
(Run-to-exhaustion protocol is described in text and Experimental Proce-
dures.) *p < 0.05 versus ab+/+; zp < 0.05 versus a?/?; #p < 0.05 versus b?/?.
PGC-1 Actions in Skeletal Muscle
640 Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc.
citrate, and 1 M Tris and an alcohol precipitation. Mitochondrial DNA content
was determined by SYBR Green analysis (Applied Biosystems or Stratagene).
The levels of NADH dehydrogenase subunit 1 (mitochondrial DNA) were
normalized to the levels of lipoprotein lipase (genomic DNA). The primer
sequences are noted in Table S2.
Gene Expression Profiling
The Digestive Diseases Research Core Center (DDRCC) Functional Genomics
Core Facility at Washington University School of Medicine performed hybrid-
ization of total RNA from GC muscle of 12-week-old mice to Agilent G4122F
Whole-Mouse Genome microarrays. Samples were hybridized in four experi-
mental pairs (PGC-1a?/?bf/fversus PGC-1bf/f, PGC-1bf/f/MLC-Creversus PGC-
1bf/f, PGC-1a?/?bf/f/MLC-Creversus PGC-1bf/f, and PGC-1a?/?bf/f/MLC-Cre
imental Procedures for more detail.
Histology and Electron Microscopy
tissue was rapidly fixedwith Karnovsky’s fixative (2% glutaraldehyde, 1%para-
fixedin1% osmium tetroxide, dehydrated ingraded ethanol, embedded inPoly
Bed plastic resin, and sectioned for EM. EM was performed by the Research
Electron Microscopy Core at Washington University School of Medicine.
GC and soleus muscles were analyzed. Muscle tissue was formalin fixed
and embedded in paraffin, sectioned, and stained with hematoxylin and eosin
(H&E) and Masson’s Trichrome. For SDH and MHC ATPase stains, muscle
tissue was frozen down in isopentane that had been cooled in liquid nitrogen.
ATPase stains were performed at pH 4.31. Under these acidic conditions,
MHC2 isoforms are inactivated while MHC1 is still functional, resulting in addi-
tion of black dye to MHC1-positive muscle fibers. Muscles for MHC immuno-
fluorescence (IF) analysis were embedded in OCT Compound (Tissue-Tek),
and IF was performed as described previously (Waters et al., 2004). The fibers
were quantified (MHC1, red; MHC2a, blue; MHC2x, black [unstained];
MHC2b, green), and results were expressed as relative numbers of the
different fiber types.
Mitochondrial respiration was assessed in isolated mitochondria from the hin-
dlimb muscle with pyruvate or palmitoylcarnitine as substrate, as described
previously (Bhattacharya etal.,1991).(See SupplementalExperimentalProce-
dures for more detail.)
Data were analyzed using t tests or ANOVA where appropriate, with Newman-
Keuls tests or pairwise comparisons (Bonferroni adjusted) used for post hoc
analyses. The level of significance was set at p < 0.05 in all cases. Data are re-
ported as mean values ± SEM.
The data discussed in this publication, including tables denoting the genes
significantly up- and downregulated in the PGC-1a?/?bf/f/MLC-Cremuscle,
have been deposited in NCBI’s Gene Expression Omnibus and are accessible
through GEO Series accession number GSE23365 (http://www.ncbi.nlm.nih.
Supplemental Information includes Supplemental Experimental Procedures,
five figures, and two tables and can be found with this article online at
This work was supported by NIH grants DK45416 and HL58427 (D.P.K.),
Neuroscience Blueprint Interdisciplinary Core Grant P30 NS057105 (D.F.W.),
an American Diabetes Association Research Award (AR050429, Z.Y.), Deut-
sche Forschungsgemeinschaft Research Fellowship ZE 796/2-1 (C.Z.), and
the Digestive Diseases Research Center (P30 DK052574) at Washington
University School of Medicine (WUSM). Special thanks to Genevieve DeMaria
for assistance with manuscript preparation; Sara Conyers at WUSM for assis-
tance with the behavioral studies; Jochen K. Lennerz at WUSM for helpful
discussion regarding histology; Karen Green and William Kraft for performing
theelectron microscopy(WUSMResearchElectronMicroscopyCore Facility);
Julio Ayala and Emily King of the Sanford-Burnham Cardiometabolic Pheno-
typing Core for assistance with GTT, ITT, and insulin measurements; and
Suellen Greco (WUSM) for performing the CK measurements.
Received: May 28, 2010
Revised: August 19, 2010
Accepted: October 1, 2010
Published: November 30, 2010
Arany, Z., He, H., Lin, J., Hoyer, K., Handschin, C., Toka, O., Ahmad, F.,
Matsui, T., Chin, S., Wu, P.H., et al. (2005). Transcriptional coactivator
PGC-1 alpha controls the energy state and contractile function of cardiac
muscle. Cell Metab. 1, 259–271.
Arany, Z., Lebrasseur, N., Morris, C., Smith, E., Yang, W., Ma, Y., Chin, S., and
Spiegelman, B.M. (2007). The transcriptional coactivator PGC-1beta drives
the formation of oxidative type IIX fibers in skeletal muscle. Cell Metab. 5,
Baar, K., Wende, A.R., Jones, T.E., Marison, M., Nolte, L.A., Chen, M., Kelly,
D.P., and Holloszy, J.O. (2002). Adaptations of skeletal muscle to exercise:
rapid increase in the transcriptional coactivator PGC-1. FASEB J. 16,
Bhattacharya, S.K., Thakar, J.H., Johnson, P.L., and Shanklin, D.R. (1991).
Isolation of skeletal muscle mitochondria from hamsters using an ionic
medium containing ethylenediaminetetraacetic acid and nagarse. Anal.
Biochem. 192, 344–349.
Bothe, G.W., Haspel, J.A., Smith, C.L., Wiener, H.H., and Burden, S.J. (2000).
Selective expressionofCrerecombinase inskeletal musclefibers.Genesis 26,
Brault, J.J., Jespersen, J.G., and Goldberg, A.L. (2010). Peroxisome prolifera-
tor-activated receptor gamma coactivator 1alpha or 1beta overexpression
inhibits muscle protein degradation, induction of ubiquitin ligases, and disuse
atrophy. J. Biol. Chem. 285, 19460–19471.
Bru ¨ning, J.C., Michael, M.D., Winnay, J.N., Hayashi, T., Ho ¨rsch, D., Accili, D.,
Goodyear, L.J., and Kahn, C.R. (1998). A muscle-specific insulin receptor
knockout exhibits features of the metabolic syndrome of NIDDM without
altering glucose tolerance. Mol. Cell 2, 559–569.
Chen, H., Chomyn, A., and Chan, D.C. (2005). Disruption of fusion results in
mitochondrial heterogeneity and dysfunction. J. Biol. Chem. 280, 26185–
Chen, Z., Gropler, M.C., Norris, J., Lawrence, J.C., Jr., Harris, T.E., and Finck,
B.N. (2008). Alterations inhepatic metabolisminfld micereveal arole forlipin 1
in regulating VLDL-triacylglyceride secretion. Arterioscler. Thromb. Vasc. Biol.
Donoghue, M.J., Alvarez, J.D., Merlie, J.P., and Sanes, J.R. (1991). Fiber type-
and position-dependent expression of amyosin light chain-CAT transgenede-
tected with a novel histochemical stain for CAT. J. Cell Biol. 115, 423–434.
Finck, B.N., and Kelly, D.P. (2006). PGC-1 coactivators: inducible regulators of
energy metabolism in health and disease. J. Clin. Invest. 116, 615–622.
Geng, T., Li, P., Okutsu, M., Yin, X., Kwek, J., Zhang, M., and Yan, Z. (2010).
PGC-1alpha plays a functional role in exercise-induced mitochondrial biogen-
esis and angiogenesis but not fiber-type transformation in mouse skeletal
muscle. Am. J. Physiol. Cell Physiol. 298, C572–C579.
Goto, M., Terada, S., Kato, M., Katoh, M., Yokozeki, T., Tabata, I., and
Shimokawa, T. (2000). cDNA Cloning and mRNA analysis of PGC-1 in
PGC-1 Actions in Skeletal Muscle
Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc. 641
epitrochlearis muscle in swimming-exercised rats. Biochem. Biophys. Res. Download full-text
Commun. 274, 350–354.
Handschin, C., and Spiegelman, B.M. (2006). Peroxisome proliferator-acti-
vated receptor gamma coactivator 1 coactivators, energy homeostasis, and
metabolism. Endocr. Rev. 27, 728–735.
and Spiegelman, B.M. (2007). Skeletal muscle fiber-type switching, exercise
intolerance, and myopathy in PGC-1alpha muscle-specific knock-out animals.
J. Biol. Chem. 282, 30014–30021.
Huss,J.M., Torra, I.P., Staels, B.,Gigue `re, V.,and Kelly, D.P. (2004). Estrogen-
related receptor alpha directs peroxisome proliferator-activated receptor
alpha signaling in the transcriptional control of energy metabolism in cardiac
and skeletal muscle. Mol. Cell. Biol. 24, 9079–9091.
Kelly, D.P., and Scarpulla, R.C. (2004). Transcriptional regulatory circuits
controlling mitochondrial biogenesis and function. Genes Dev. 18, 357–368.
Lai, L., Leone, T.C., Zechner, C., Schaeffer, P.J., Kelly, S.M., Flanagan, D.P.,
Medeiros, D.M., Kovacs, A., and Kelly, D.P. (2008). Transcriptional coactiva-
tors PGC-1alpha and PGC-lbeta control overlapping programs required for
perinatal maturation of the heart. Genes Dev. 22, 1948–1961.
Lehman, J.J., Barger, P.M., Kovacs, A., Saffitz, J.E., Medeiros, D.M., and
Kelly, D.P. (2000). Peroxisome proliferator-activated receptor gamma coacti-
vator-1 promotes cardiac mitochondrial biogenesis. J. Clin. Invest. 106,
Lelliott, C.J., Medina-Gomez, G., Petrovic, N., Kis, A., Feldmann, H.M.,
Bjursell, M., Parker, N., Curtis, K., Campbell, M., Hu, P., et al. (2006).
Ablation of PGC-1beta results in defective mitochondrial activity, thermogen-
esis, hepatic function, and cardiac performance. PLoS Biol. 4, e369.
Leone, T.C., Lehman, J.J., Finck, B.N., Schaeffer, P.J., Wende, A.R., Boudina,
S., Courtois, M., Wozniak, D.F., Sambandam, N., Bernal-Mizrachi, C., et al.
(2005). PGC-1alpha deficiency causes multi-system energy metabolic
derangements: muscle dysfunction, abnormal weight control and hepatic
steatosis. PLoS Biol. 3, e101.
Liesa, M., Borda-d’Agua, B., Medina-Go ´mez, G., Lelliott, C.J., Paz, J.C., Rojo,
M., Palacı ´n, M., Vidal-Puig, A., and Zorzano, A. (2008). Mitochondrial fusion is
increased by the nuclear coactivator PGC-1beta. PLoS ONE 3, e3613.
Lin, J., Wu, H., Tarr, P.T., Zhang, C.Y., Wu, Z., Boss, O., Michael, L.F.,
Puigserver, P., Isotani, E., Olson, E.N., et al. (2002a). Transcriptional co-acti-
vator PGC-1 alpha drives the formation of slow-twitch muscle fibres. Nature
Lin, X., Schonfeld, G., Yue, P., and Chen, Z. (2002b). Hepatic fatty acid
synthesis is suppressed in mice with fatty livers due to targeted apolipoprotein
B38.9 mutation. Arterioscler. Thromb. Vasc. Biol. 22, 476–482.
Lin, J., Wu, P.H., Tarr, P.T., Lindenberg, K.S., St-Pierre, J., Zhang, C.Y.,
Mootha, V.K., Ja ¨ger, S., Vianna, C.R., Reznick, R.M., et al. (2004). Defects in
adaptive energy metabolism with CNS-linked hyperactivity in PGC-1alpha
null mice. Cell 119, 121–135.
the PGC-1 family of transcription coactivators. Cell Metab. 1, 361–370.
Lowell, B.B., and Shulman, G.I. (2005). Mitochondrial dysfunction and type 2
diabetes. Science 307, 384–387.
Mootha, V.K., Lindgren, C.M., Eriksson, K.F., Subramanian, A., Sihag, S.,
Lehar, J., Puigserver, P., Carlsson, E., Ridderstra ˚le, M., Laurila, E., et al.
(2003). PGC-1alpha-responsive genes involved in oxidative phosphorylation
are coordinately downregulated in human diabetes. Nat. Genet. 34, 267–273.
Morino, K., Petersen, K.F., Dufour, S., Befroy, D., Frattini, J., Shatzkes, N.,
Neschen, S., White, M.F., Bilz, S., Sono, S., et al. (2005). Reduced mitochon-
drial density and increased IRS-1 serine phosphorylation in muscle of insulin-
resistant offspring of type 2 diabetic parents. J. Clin. Invest. 115, 3587–3593.
Parsons, S.A., Millay, D.P., Wilkins, B.J., Bueno, O.F., Tsika, G.L., Neilson,
J.R., Liberatore, C.M., Yutzey, K.E., Crabtree, G.R., Tsika, R.W., and
Molkentin, J.D. (2004). Genetic loss of calcineurin blocks mechanical over-
load-induced skeletal muscle fiber type switching but not hypertrophy.
J. Biol. Chem. 279, 26192–26200.
Patti, M.E., Butte, A.J., Crunkhorn, S., Cusi, K., Berria, R., Kashyap, S.,
Miyazaki, Y., Kohane, I., Costello, M., Saccone, R., et al. (2003).
resistance and diabetes: Potential role of PGC1 and NRF1. Proc. Natl. Acad.
Sci. USA 100, 8466–8471.
Petersen, K.F., Dufour, S., and Shulman, G.I. (2005). Decreased insulin-stimu-
lated ATP synthesis and phosphate transport in muscle of insulin-resistant
offspring of type 2 diabetic parents. PLoS Med. 2, e233.
Pilegaard, H., Saltin, B., and Neufer, P.D. (2003). Exercise induces transient
transcriptional activation of the PGC-1alpha gene in human skeletal muscle.
J. Physiol. 546, 851–858.
Puigserver, P., Wu, Z., Park, C.W., Graves, R., Wright, M., and Spiegelman,
B.M. (1998). A cold-inducible coactivator of nuclear receptors linked to adap-
tive thermogenesis. Cell 92, 829–839.
Sonoda, J., Mehl, I.R., Chong, L.W., Nofsinger, R.R., and Evans, R.M. (2007).
PGC-1beta controls mitochondrial metabolism to modulate circadian activity,
adaptive thermogenesis, and hepatic steatosis. Proc. Natl. Acad. Sci. USA
Soriano, F.X., Liesa, M., Bach, D., Chan, D.C., Palacı ´n, M., and Zorzano, A.
(2006). Evidence for a mitochondrial regulatory pathway defined by peroxi-
some proliferator-activated receptor-gamma coactivator-1 alpha, estrogen-
related receptor-alpha, and mitofusin 2. Diabetes 55, 1783–1791.
Vianna, C.R., Huntgeburth, M., Coppari, R., Choi, C.S., Lin, J., Krauss, S.,
Barbatelli, G., Tzameli, I., Kim, Y.B., Cinti, S., et al. (2006). Hypomorphic muta-
tion of PGC-1beta causes mitochondrial dysfunction and liver insulin resis-
tance. Cell Metab. 4, 453–464.
Waterham, H.R., Koster, J., van Roermund, C.W., Mooyer, P.A., Wanders,
R.J., and Leonard, J.V. (2007). A lethal defect of mitochondrial and peroxi-
somal fission. N. Engl. J. Med. 356, 1736–1741.
Waters, R.E., Rotevatn, S., Li, P., Annex, B.H., and Yan, Z. (2004). Voluntary
running induces fiber type-specific angiogenesis in mouse skeletal muscle.
Am. J. Physiol. Cell Physiol. 287, C1342–C1348.
Wende, A.R., Schaeffer, P.J., Parker, G.J., Zechner, C., Han, D.H., Chen,
M.M., Hancock, C.R., Lehman, J.J., Huss, J.M., McClain, D.A., et al. (2007).
A role for the transcriptional coactivator PGC-1alpha in muscle refueling.
J. Biol. Chem. 282, 36642–36651.
Wenz, T., Rossi, S.G., Rotundo, R.L., Spiegelman, B.M., and Moraes, C.T.
(2009). Increased muscle PGC-1alpha expression protects from sarcopenia
and metabolic disease during aging. Proc. Natl. Acad. Sci. USA 106, 20405–
Wozniak, D.F., Xiao, M., Xu, L., Yamada, K.A., and Ornitz, D.M. (2007).
fibroblast growth factor 14. Neurobiol. Dis. 26, 14–26.
Wu, Z., Puigserver, P., Andersson, U., Zhang, C., Adelmant, G., Mootha, V.,
Troy, A., Cinti, S., Lowell, B., Scarpulla, R.C., and Spiegelman, B.M. (1999).
Mechanisms controlling mitochondrial biogenesis and respiration through
the thermogenic coactivator PGC-1. Cell 98, 115–124.
Zhang, Y.,and Chan, D.C.(2007).Structural basis for recruitment of mitochon-
drial fission complexes by Fis1. Proc. Natl. Acad. Sci. USA 104, 18526–18530.
Zhang, Y., Huypens, P., Adamson, A.W., Chang, J.S., Henagan, T.M.,
Boudreau, A., Lenard, N.R., Burk, D., Klein, J., Perwitz, N., et al. (2009).
Alternative mRNA splicing produces a novel biologically active short isoform
of PGC-1alpha. J. Biol. Chem. 284, 32813–32826.
PGC-1 Actions in Skeletal Muscle
642 Cell Metabolism 12, 633–642, December 1, 2010 ª2010 Elsevier Inc.