Báo cáo khoa học: Cardiac ankyrin repeat protein is a marker of skeletal muscle pathological remodelling pot

Cardiac ankyrin repeat protein is a marker of skeletal
muscle pathological remodelling
Lydie Laure
1
, Laurence Suel
1
, Carinne Roudaut
1
, Nathalie Bourg
1
, Ahmed Ouali
2
, Marc Bartoli
1
,
Isabelle Richard
1
and Nathalie Danie
`
le
1
1Ge
´
ne
´
thon-CNRS FRE3087, Evry, France
2 INRA de Theix, Saint Gene
`
s Champanelle, France
Muscle atrophy can result from disuse of the organ or
be associated with ageing or severe systemic conditions

such as diabetes, AIDS and cancer. It is also a feature
common to many hereditary muscle diseases, including
muscular dystrophies (MDs). Duchenne MD (DMD),
caused by mutation in the dystrophin gene, is the most
common form of the disease and is particularly severe:
skeletal and cardiac muscles are affected, and the life-
span of the patients is seriously impaired [1]. Limb
girdle MDs (LGMDs) represent another important
subgroup of MD, grouped together on the basis of
common clinical features: they all primarily and
Keywords
CARP; FoxO1; muscle; p21
WAF1/CIP1
;
remodelling
Correspondence
I. Richard, Ge
´
ne
´
thon, CNRS FRE3087, 1 bis
rue de l’Internationale, 91000 Evry, France
Fax: +33 0 1 60 77 86 98
Tel: +33 0 1 69 47 29 38
E-mail:
(Received 31 July 2008, revised 20 October
2008, accepted 24 November 2008)
doi:10.1111/j.1742-4658.2008.06814.x
In an attempt to identify potential therapeutic targets for the correction of
muscle wasting, the gene expression of several pivotal proteins involved in

protein metabolism was investigated in experimental atrophy induced by
transient or definitive denervation, as well as in four animal models of
muscular dystrophies (deficient for calpain 3, dysferlin, a-sarcoglycan and
dystrophin, respectively). The results showed that: (a) the components of
the ubiquitin–proteasome pathway are upregulated during the very early
phases of atrophy but do not greatly increase in the muscular dystrophy
models; (b) forkhead box protein O1 mRNA expression is augmented in
the muscles of a limb girdle muscular dystrophy 2A murine model; and (c)
the expression of cardiac ankyrin repeat protein (CARP), a regulator of
transcription factors, appears to be persistently upregulated in every condi-
tion, suggesting that CARP could be a hub protein participating in com-
mon pathological molecular pathway(s). Interestingly, the mRNA level of
a cell cycle inhibitor known to be upregulated by CARP in other tissues,
p21
WAF1/CIP1
, is consistently increased whenever CARP is upregulated.
CARP overexpression in muscle fibres fails to affect their calibre, indicating
that CARP per se cannot initiate atrophy. However, a switch towards fast-
twitch fibres is observed, suggesting that CARP plays a role in skeletal
muscle plasticity. The observation that p21
WAF1/CIP1
is upregulated, put in
perspective with the effects of CARP on the fibre type, fits well with the
idea that the mechanisms at stake might be required to oppose muscle
remodelling in skeletal muscle.
Abbreviations
AAV2/1, adeno-associated virus 2/1; Ankrd2, ankyrin repeat domain-containing protein 2; CARP, cardiac ankyrin repeat protein; DAPI,
4¢,6-diamidino-2-phenylindole; DMD, Duchenne muscular dystrophy; EDL, extensorum digitorum longus; FoxO, forkhead box protein O;
FP, fluorescent protein; LGMD, limb girdle muscular dystrophy; MAFbx, muscle atrophy F-box protein; MD, muscular dystrophy; MLC-2v,
myosin light chain 2v; MLC-f, myosin light chain, fast; MuRF1, muscle RING finger protein 1; NF, neurofilament protein; NF-jB, nuclear

factor-jB; qRT-PCR, quantitative RT-PCR; TA, tibialis anterior; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labelling; Ub,
ubiquitin; UPS, ubiquitin–proteasome system; YFP, yellow fluorescent protein.
FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS 669
predominantly affect proximal muscles around the
scapular and the pelvic girdles [2]. About 20 different
forms of LGMD are currently recognized; among the
most frequent are LGMD2A, LGMD2B and the
sarcoglycanopathies (LGMD2C–F), caused by muta-
tion in the calpain 3, dysferlin and sarcoglycan genes,
respectively [2].
Disuse-induced atrophy and MDs might share some
molecular mechanisms that are possibly involved in
skeletal muscle wasting. Muscle atrophy results from
the negative balance in the ratio between protein syn-
thesis and protein degradation, hence leading towards
protein wasting. One of the key players in the degrada-
tion of myofibrillar proteins is the ubiquitin–protea-
some system (UPS) [3]. The elimination process is
initiated by labelling of the targeted proteins with mul-
tiple ubiquitin molecules, and requires the coordinated
action of three classes of enzymes known as E1 (ubiqu-
itin-activating enzymes), E2 (ubiquitin-conjugating
enzymes) and E3 (ubiquitin ligases) [4]. The ubiquitin–
proteasome cascade is stimulated at many levels in
several conditions leading to muscle wasting: the
expression of proteasome subunits, the hydrolytic
activity, and the general substrates ubiquitination [5].
In particular, the 14 kDa ubiquitin carrier protein E2
(E2-14 kDa) and two recently identified E3s, muscle
atrophy F-box protein (MAFbx; also commonly called

atrogin-1) and muscle RING finger protein 1 (MuRF1;
also named TRIM63), are upregulated in many skele-
tal muscle-wasting conditions [5]. During atrophy,
expression of MAFbx and MurF1 is stimulated by the
forkhead box protein O (FoxO) family of transcription
factors, through inhibition of the Akt pathway [6,7].
In addition, it was also shown that transcriptional
stimulation of MuRF1 is under the control of the
nuclear factor-jB (NF-jB) pathway [8].
Even though the literature largely explores the con-
vergent role of the UPS components in atrophy, mus-
cle wasting is a complex mechanism in which specific,
although poorly understood, pathways could play a
role. In particular, cardiac ankyrin repeat protein
(CARP) was suggested to be involved in these pro-
cesses. CARP, together with ankyrin repeat domain-
containing protein 2 (Ankrd2) and diabetes-related
ankyrin repeat protein, forms a family of transcription
regulators known as muscle ankyrin repeat proteins.
These three isoforms share in their C-terminal region a
minimal structure composed of several ankryrin-like
domains possibly involved in protein–protein inter-
action, PEST motifs characteristics of rapidly degraded
protein, and a putative nuclear localization signal.
CARP is expressed in both cardiac and skeletal mus-
cles, and was reported to be either upregulated [9] or
downregulated [10,11], depending on the atrophic situ-
ation considered, and upregulated in hypertrophic con-
ditions in heart [12–17] and in skeletal muscle [18–21].
From the functional point of view, in heart cells,

CARP overexpression suppresses troponin C and atrial
natriuretic factor expression [22], and its interaction
with the transcription factor YB1 inhibits the synthesis
of the ventricular-specific myosin light chain 2v (MLC-
2v) [23]. In vascular smooth muscle cells, increased
CARP expression has been demonstrated to be associ-
ated with upregulation of the protein p21
WAF1/CIP1
,an
inhibitor of the cell cycle [24]. Taken as a whole, these
findings suggest that CARP coordinates the expression
of genes involved in cell structure and proliferation,
and could play a role during muscle mass variation.
In an attempt to identify hub proteins that may be
potential diagnostic markers or even therapeutic tar-
gets for the correction of muscle wasting, the expres-
sion of pivotal proteins involved in all the mechanisms
discussed previously was investigated in denervation-
induced atrophy, as well as in three animal models of
LGMD and in the mdx mouse, a DMD model. Our
study demonstrates that: (a) the UPS is transiently
upregulated after denervation, consistent with its
known role in atrophy, but it does not seem to be
greatly activated in MD; (b) FoxO1 is a biological
marker specific for the LGMD2A murine model;
and (c) among all the genes considered, the expression
of CARP, together with its downstream target,
p21
WAF1/CIP1
, appears to be the only one that system-

atically increases. CARP overexpression in muscle
fibres fails to induce an atrophic phenotype, indicating
that CARP per se cannot initiate the phenomenon.
Nonetheless, the switch towards fast-twitch fibres
observed in this situation, together with the observa-
tion that the p21
WAF1/CIP1
expression pattern seems to
reflect CARP level, suggests that CARP might play a
role in muscle plasticity.
Results
The proteasome pathway components are
only transiently upregulated, whereas increased
CARP expression is maintained throughout
denervation-induced-atrophy
The expression of several factors possibly involved in
atrophy was investigated by the evaluation of their
mRNA level by quantitative RT-PCR (qRT-PCR) in
conditions leading to transitory or definitive atrophy.
The genes studied were those encoding: (a) two tran-
scription factors involved in the control of muscle
mass: NF-jB-p65 and FoxO1; (b) several components
Cardiac ankyrin repeat protein in muscle plasticity L. Laure et al.
670 FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS
of the UPS – ubiquitin (Ub), E2-14 kDa, two E3
ubiquitin ligases, MuRF1 and MAFbx, and the C2,
C8 and C9 subunits of the proteasome; and (c) CARP,
a transcriptional regulator associated with perturbation
of muscle mass. Transient or chronic denervation of
the posterior limb was induced and the mRNA levels

were measured in tibialis anterior (TA) muscles at four
different times following the initiation of the treat-
ments (days 3, 9, 14 and 21). Atrophy was efficiently
triggered by the treatment, as 40% of the TA weight
was lost after 21 days of chronic denervation
(Fig. 1A). When denervation was only transient, the
TA weight also initially decreased, but slowly increased
again from day 14 while innervation occurred [25] (sta-
tistically higher than chronic denervation from day 14
to day 21, P < 0.05).
0
50
100
150
200
T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
C9
0
50
100
150
200
0
50
100
150
200
T0 T3 T9 T14 T21 D0 D3 D9 D14D21T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control

*
0
50
100
150
200
250
300
T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
FoxO1
0
50
100
150
200
250
300
0
50
100
150
200
250
300
50
100
150
200
250

300
T0 T3 T9 T14 T21 D0 D3 D9 D14D21T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
*
*
**
0
100
200
300
400
T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
NF-kB
0
100
200
300
400
0
100
200
300
400
T0 T3 T9 T14 T21 D0 D3 D9 D14D21T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
**
0
50
100

150
T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
E2
0
50
100
150
0
50
100
150
T0 T3 T9 T14 T21 D0 D3 D9 D14D21T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
**
1
10
100
1000
T0 T3 T9 T14T21 D0 D3 D9 D14D21
% of control
MuRF1
1
10
100
1000
1
10
100
1000

T0 T3 T9 T14T21 D0 D3 D9 D14D21T0 T3 T9 T14T21 D0 D3 D9 D14D21
% of control
*
**
**
**
T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
MAFbx
1
10
100
1000
10 000
T0 T3 T9 T14 T21 D0 D3 D9 D14D21T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
*
**
**
**
15
20
25
30
35
40
45
50
Day 3 Day 9 Day 14 Day 21
TA weight (mg)

15
20
25
30
35
40
45
50
Day 3 Day 9 Day 14 Day 21
15
20
25
30
35
40
45
50
15
20
25
30
35
40
45
50
Day 3 Day 9 Day 14 Day 21Day 3 Day 9 Day 14 Day 21
Control
Transient
Definitive
TA weight (mg)

*
*
*
*
*
0
50
100
150
T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
Ubiquitin
0
50
100
150
0
50
100
150
T0 T3 T9 T14 T21 D0 D3 D9 D14D21T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
*
*
*
*
*
*
**
**

T0 T3 T9 T14T21 D0 D3 D9 D14D21
% of control
CARP
1
10
100
1000
10 000
100 000
T0 T3 T9 T14T21 D0 D3 D9 D14D21T0 T3 T9 T14T21 D0 D3 D9 D14D21
% of control
*
**
**
**
**
**
**
0
50
100
150
200
T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
C2
0
50
100
150

200
0
50
100
150
200
T0 T3 T9 T14 T21 D0 D3 D9 D14D21T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
*
*
0
50
100
150
200
T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control
C8
0
50
100
150
200
0
50
100
150
200
T0 T3 T9 T14 T21 D0 D3 D9 D14D21T0 T3 T9 T14 T21 D0 D3 D9 D14D21
% of control

*
**
A
B
Fig. 1. Effect of transient or definitive
denervation on muscle weight and gene
expression profiles. Male mice of the
129SvPasIco strain were treated transiently
(T) by crushing or definitively (D) by section
of the sciatic nerve. Samples were taken
from six animals on each date (control, 3, 7,
9, 14 and 21 days after nerve injury). (A)
Weight of TA muscles from control, crushed
and sectioned limbs (n = 6 per time point).
Other muscles of the lower limb, such as
EDL and soleus muscles, present similar
proportional loss of weight. P-values are
shown as *P < 0.05 for significance
between control and each time point, and
as
h
P < 0.05 for significance between tran-
sient and definitive denervation. (B) Each
graph demonstrates the expression level for
a gene of interest (FoxO1, NF-jB-p65, Ub,
E2-14 kDa, C2, C8, C9, MuRF1, MAFbx and
CARP ) as assessed by qRT-PCR in TA
muscles of treated animals (n = 2–6 for
each time point). Results are expressed as
percentage of expression level measured in

the respective sham-operated muscles.
*P < 0.05 and **P < 0.01 for significance
between control and each time point;
h
P < 0.05 for significance between
transient and definitive denervation.
L. Laure et al. Cardiac ankyrin repeat protein in muscle plasticity
FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS 671
The results showed that, with both transient and
definitive denervation, FoxO1, NF-jB-p65 and several
components of the UPS (subunits C2, C8 and C9, and
the two E3s MuRF1 and MAFbx) were immediately
and transiently upregulated, with higher variations in
the case of the two E3s (note the logarithmic scale)
Fig. 1B). After this initial increase, their expression
returned rapidly to normal levels, even displaying a
slight reduction for every proteasome subunit consid-
ered (C2, C8 and C9). Ubiquitin mRNA levels
decreased very early during the time course of atrophy,
remaining very low when denervation was definitive,
but progressively increasing again from the start of
reinnervation when the sciatic nerve was only crushed
(Fig. 1B). E2-14 kDa expression, which remained sta-
ble when atrophy was only transient, was reduced at
late stages (from day 14) of definitive denervation-
induced atrophy (Fig. 1B). CARP expression increased
with atrophy (Fig. 1B). Whereas CARP expression
slowly decreased back to control level with the reduc-
tion of atrophy in transient denervation, it stayed high
when sciatic nerve regeneration was prevented. CARP

upregulation was particularly important, as reflected
by the logarithmic scale.
CARP is robustly upregulated in murine MDs,
whereas FoxO1 expression is increased
specifically in C3-null animals
The expression levels of the mRNAs measured in
denervation conditions were also compared by qRT-
PCR in several models of MD: a natural model of
dysferlin deficiency [26], which we backcrossed on a
C57BL/6 background and renamed B6.A/J-dysf
prmd
(model for LGMD2B), and three engineered models
deficient in either dystrophin (mdx
4Cv
[27]), calpain 3
(C3-null; unpublished), or a-sarcoglycan (Sgca-null
[28]), models of DMD, LGMD2A and LGMD2D,
respectively. Every strain was used at an age where the
symptoms of the disease are detectable (4 months of
age for all models except C3-null mice, which were
evaluated at 7 months of age) and was compared to its
respective control breed. The levels of mRNA expres-
sion were measured in five muscles [quadriceps,
extensorum digitorum longus (EDL), TA, soleus and
psoas], chosen in order to reflect the muscle impair-
ment specificity – which varies between models – and
the type of fibres composing the muscle (see Experi-
mental procedures).
The results of qRT-PCR showed that the level of
NF-jB-p65 was slightly increased in specific muscles of

every murine model, especially in the two most inflam-
matory models, mdx
4Cv
and Sgca-null (Fig. 2). FoxO1
was upregulated to very similar levels in every muscle
of the C3-null strain (about two-fold over control, with
P < 0.05 for quadriceps, EDL and psoas), whereas
its expression was slightly decreased in all the other
models (Fig. 2).
The expression of Ub was not affected in any of
the four pathologies considered, whereas that of
E2-14 kDa showed a tendency to decrease in several
muscles (Fig. 2). In the mdx
4Cv
model, the levels of the
three proteasome subunits (C2, C8 and C9) were
affected, C2 and C8 being downregulated and C9
upregulated. Unexpectedly, considering their role in
atrophy, neither MuRF1 nor MAFbx expression
increased in these animal models, their levels being
even significantly reduced in some cases (Fig. 2).
The most remarkable effect observed herein was
robust upregulation of CARP mRNA (note the loga-
rithmic scale) in most muscles of all models of MDs
(Fig. 2). Interestingly, this increase seemed to be higher
in the muscles strongly affected by the pathologies.
The increase was far more important in the Sgca-null
and in the mdx
4Cv
models, two dystrophies character-

ized by a similar pathogenesis and caused by a defect
in one of the components of the dystrophin-associated
glycoprotein complex.
CARP is expressed at the protein level in
myofibres of denervation-induced atrophy
models and in mononucleated cells of highly
regenerative MD animals
Among all the genes whose expression was investi-
gated in the different models of muscle disorder, we
demonstrated that the CARP gene was the only one
whose expression systematically increased, which is
consistent with CARP’s role as a hub protein partici-
pating in common pathological molecular pathway(s).
CARP protein expression was hence measured by
western blot in conditions of denervation-induced
atrophy and in murine models of MD (Fig. 3). Inter-
estingly, we observed that the protein was detected
by western blot provided that the mRNA level
reached 60-fold over the basal condition. This ele-
ment probably accounts for the inability to detect
CARP in many conditions in which its mRNA
upregulation is indeed important, although not
important enough. The protein was therefore
detected from day 3 in both denervation conditions,
remaining high until day 21 when the sciatic nerve
was sectioned, but dropping to undetectable levels
when reinnervation occurred during transient dener-
vation (Fig. 3A, upper left panel). As regards the
murine models of dystrophies, CARP protein was
Cardiac ankyrin repeat protein in muscle plasticity L. Laure et al.

672 FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS
Fig. 2. Gene expression profiles in murine models of MD. Each graph demonstrates the expression level for a gene of interest (FoxO1,
NF-jB-p65, Ub, E2-14 kDa, C2, C8, C9, MuRF1, MAFbx and CARP ) as assessed by qRT-PCR in quadriceps, EDL, TA, soleus and psoas
muscles of C3-null, B6.A/J-dysf
prmd
, Sgca-null and mdx
4Cv
animals (n = 3–4 for each point). Results are expressed as percentage of expres-
sion level measured in the respective control muscles (129svPasIco and C57BL/6). P-values for significance between wild-type and deficient
animals: *P < 0.05 and **P < 0.01.
L. Laure et al. Cardiac ankyrin repeat protein in muscle plasticity
FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS 673
detected in Sgca-null animals only (Fig. 3A, lower
left panel).
In order to clarify CARP cellular distribution within
the muscle, immunodetection of the protein was
performed on sections of muscles from denervated
(3 days after denervation), a-sarcoglycan-deficient and
dystrophin-deficient animals, and their appropriate
control strains. Specificity of the CARP antibody was
T0 T3 T9 T14
T3
C3-null 129SvPasIco T3
C57BL/6 B6.A/J-Dysf T3
CARP
Ponceau red
CARP
Ponceau red
Control
Denervated

Control
Sgca-null
Day3 after denervation
Sgca-nullControl
CARP Pax7 Merge
255.0
0.0
1440.0 pixels
1560.0 pixels
1440.0 pixels
1560.0 pixels
2550
0.0
A
B
C
D
E
T21 D3 D9 D14 D21
Sgca-null
C57BL/6
mdx
4cn
mdx
4cn
mdx
4cn
50 µm 50 µm
50 µm
Cardiac ankyrin repeat protein in muscle plasticity L. Laure et al.

674 FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS
first confirmed by the very specific staining observed in
cultured HER911 cells transfected with a plasmid
encoding CARP (data not shown). In all sections (con-
trol, denervated and MD animals), intense staining
was seen within scattered clusters of small myofibres
(Fig. 3B). No difference in the number of these clusters
was observed between conditions, indicating that the
increase in CARP expression did not originate from
these cells. In denervation-induced atrophy, additional
diffuse checked-pattern staining of higher-calibre fibres
was also detected, with a higher intensity in denervated
muscles than in control sections (Fig. 3C). Considering
the dystrophic process present in Sgca-null and mdx
4Cv
animals, it is difficult to evaluate whether such upregu-
lation also occurred in these models. In any case, very
intense foci corresponding to the cytoplasm of small
round cells flanking the muscle fibres were observed in
Sgca-null and mdx
4Cv
animals (Fig. 3D). These cells
expressed Pax7, the first transcription factor activated
during myogenesis (Fig. 3E). Immunostaining of the
neurofilament protein (NF) failed to reveal any colo-
calization with CARP (data not shown).
The p21
WAF1/CIP1
gene expression profile parallels
CARP in both MD and denervation-induced

atrophy models
In an attempt to dissect the molecular mechanisms
activated downstream of the CARP gene, the gene
expression of three relevant target genes chosen on
account of CARP targets in cardiac and vascular tis-
sues was measured by qRT-PCR in both denervation
and MD models: the slow isoform of myosin light
chain MLC-2v [23], its paralogous gene in skeletal
muscle fast fibres, myosin light chain, fast (MLC-f),
and the cell cycle inhibitor p21
WAF1/CIP1
[24]. Although
it was previously reported to be expressed at low levels
in skeletal muscle [29], MLC-2v gene expression
remained undetectable in our conditions (data not
shown). Whereas MLC-f expression was inversely
correlated with CARP level in denervated animals, its
level was generally unaffected, or even slightly
increased, in muscles of MD models (data not shown).
As neither MLC-2v nor MLC-f expression were corre-
lated consistently with CARP level, neither of these
proteins seems to be involved in the CARP signalling
pathway in skeletal muscle. In contrast, in both dener-
vation and MD models, p21
WAF1/CIP1
gene expression
paralleled the CARP profile, i.e. increased when muscle
degeneration occurred, and progressively decreased
back to control level during the reinnervation phase of
transient denervation (Fig. 4). It is worth mentioning

that p21
WAF1/CIP1
upregulation was of the same order
of magnitude as CARP upregulation, as reflected by
the logarithmic scale.
CARP overexpression in wild-type mouse TA
muscle does not induce atrophy, but alters fibre
type composition
Considering that the upregulation of CARP persisted
in definitive denervation and was consistent in MD
models, we tried to understand its contribution to
these conditions and therefore investigated its func-
tion(s) in skeletal muscle. A pseudotyped adeno-associ-
ated virus 2/1 (AAV2/1) vector in which the CARP
coding sequence is fused with the yellow fluorescent
protein (YFP) sequence was injected into the TA mus-
cle of normal mice. One month after injection, direct
observation of the skinned injected muscle using con-
focal fluorescence microscopy allowed the visualization
of a high level of YFP fluorescence. Measurement of
the level of CARP mRNA by qRT-PCR confirmed
strong expression of the transgene (more than 60 times
the level of mRNA in the control experiment,
P < 0.01; Fig. 5A,B). This was indeed reflected by the
appearance of a band corresponding to CARP expres-
sion in western blots (Fig. 5C).
Fig. 3. Analysis of CARP protein level and cellular localization in denervation-induced atrophy and in murine models of MD. (A) In conditions
of both transient (T) and definitive (D) denervation, the level of expression of CARP protein was assessed on equivalent amounts of lysate
proteins resolved by western blot. The standardization of the loading was verified by Ponceau red staining. The expression of CARP protein
correlates perfectly with the mRNA profile. The expression of CARP protein was estimated by the same method in the psoas muscle (all

models but C3-null) or the deltoid muscle (C3-null mice) of the different MD models (comparison made in each case with the adequate wild-
type strain). We previously verified that the upregulation of the level of CARP transcripts is similar in both deltoid and psoas in the C3-null
strain (five-fold over wild-type control, data not shown). The results show that the upregulation of the expression of CARP protein can be
visualized in the Sgca-null model only. (B) CARP was detected by specific immunostaining (in green) on transverse sections of control
(129svPasIco), denervated (129svPasIco), Sgca-null and mdx
4Cv
muscles. Staining with dystrophin (in red) was used to delimit the fibres. A
view of each muscle taken with a 40· objective is presented, showing the very specific staining observed within clusters of small myofibres.
Scale bars: 30 lm. (C) Surface plots representing the density of pixels from whole muscle sections after immunostaining show that CARP
expression increases after denervation. Original images were processed using
IMAGEJ software (8-bit images, Fire look-up table; http://rsb.
info.nih.gov/ij/). (D) Very intense foci of mononucleated cells are observed in Sgca-null and mdx
4Cv
muscles, but not in control muscles. Scale
bars: 50 lm. (E) Costaining for CARP and Pax7 shows that the cells identified in (D) are positive for Pax7.
L. Laure et al. Cardiac ankyrin repeat protein in muscle plasticity
FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS 675
We next investigated whether any phenotype was
apparent following CARP expression. In these condi-
tions, the TA muscle weight was not affected
(Fig. 5D). The histological appearance of the muscles
was normal (Fig. 5E). Morphometric analyses per-
formed on sliced muscles (Fig. 5F) revealed no differ-
ences in terms of number or mean diameter of fibres
in comparison with the untreated control, although
the slight switch of the curve detected in the presence
of CARP might reflect a tendency to generate bigger
fibres. Muscle sections were negative for terminal
deoxynucleotidyl transferase dUTP nick end labelling
(TUNEL) staining, a marker of apoptosis (data not

shown). As members of the CARP family were
recently suggested to play a role in fibre typing [30],
immunohistochemistry of sections was performed
using an antibody against slow myosin. A shift
towards a reduction of slow-twitch fibre type was
observed in the presence of CARP (P < 0.05;
Fig. 5G).
Discussion
In this study, in an attempt to identify proteins
involved in the physiopathology of muscle wasting, we
examined the variation in the expression levels of sev-
eral atrophy-associated genes during transient and
definitive denervation and in four models of MD. The
main results gained from these studies are: (a) that the
levels of essential components of the UPS are aug-
mented rapidly and transiently during denervation-
induced atrophy, but are not elevated in most MD
muscles; (b) that FoxO1 mRNA expression is signifi-
cantly increased in an LGMD2A model; and (c) that
CARP is robustly upregulated in numerous murine
MD models and in denervation-induced atrophy.
First, in line with their documented role in atrophy
[31–33], we demonstrated that the expression levels of
most investigated components of the UPS increase
transitorily during transient and definitive denervation.
However, the mRNA expression levels of both Ub and
E2-14 kDa, previously reported to be upregulated in
atrophic conditions [5], do not increase, suggesting that
neither protein is rate-limiting in this atrophic situa-
tion. Consistent with this result, the role of E2-14 kDa

has lately been reconsidered, as the inactivation of the
corresponding genes does not seem to induce atrophy
resistance, at least in the conditions tested [34]. In con-
trast to the denervation situation, the mRNA expres-
sion levels of the UPS elements were almost never
increased in the four MDs tested, suggesting that the
UPS is not overly activated in these diseases. Whether
this reflects the slow progression of the diseases with
respect to the atrophy phenomenon or weak involve-
ment of the UPS in the pathogenesis remains to be
determined.
Second, FoxO1 was demonstrated to be specifically
upregulated in every muscle of the C3-null strain.
Besides raising the interesting possibility that FoxO1
could be used as a diagnostic marker for LGMD2A,
our results indicate that FoxO1 expression increases as
a consequence of the absence of calpain 3, either
because of a functional relationship between the two
proteins, or by a specific pathophysiological mecha-
nism unique to calpain 3 deficiency. Regardless of its
cause, this upregulation of FoxO1 is very likely to play
an important role in the atrophy observed in this dis-
ease, as its in vivo overexpression was previously dem-
onstrated to induce reduction of muscle mass [6,35].
However, this phenomenon does not seem to proceed
through MuRF1 and MAFbx, as their expression
levels did not increase in our C3-null strain, but might
p21
WAF1/CIP1
p21

WAF1/CIP1
Fig. 4. Gene expression profiles of p21
WAF1/CIP1
after transient or definitive denervation and in murine models of MD. The gene expression of
p21
WAF1/CIP1
was measured by qRT-PCR in TA muscles subjected to denervation-induced atrophy (n = 2–6 for each time point), and in quadri-
ceps, EDL, TA, soleus and psoas muscles of the four MD models (n = 3–4 for each point). T, transient denervation; D, definitive denervation.
Results are expressed as percentage of expression level measured in the respective control muscles for MDs (129svPasIco and C57BL/6)
or in the sham-operated muscles for denervation models. In every situation, p21
WAF1/CIP1
gene expression reflects CARP level.
Cardiac ankyrin repeat protein in muscle plasticity L. Laure et al.
676 FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS
instead involve other FoxO1-dependent signalling
cascades, such as the autophagy pathway [36–38] and/
or the control of satellite cell proliferation [39], two
mechanisms important for muscle mass regulation [40].
Provided that upregulation of FoxO1 is found also in
LGMD2A patients, it seems highly likely that imped-
ing FoxO1 increase and/or inhibiting its activity might
improve the phenotype.
0
20
40
60
Control +CARP
TA weight (mg)
Control + CARP
0

20
40
60
80
100
Control + CARP
**
*
0
5000
10 000
15 000
20 000
25 000
Control CARP
CARP mRNA level
(% of control)
WB CARP
Ponceau red
+ CARPControl
0
100
200
300
400
500
600
700
800
900

0–10
10–20
20–30
30–40
40–50
50–60
60–70
70–80
80–90
90–100
100–110
Control
+ CARP
Fibre number/TA
Fibre diameter (µm)
Mean number of slow-twitch
fibres/section
**
A
D
F
G
B
C
E
Fig. 5. Effect of CARP overexpression in muscle. (A) One month after intramuscular injection of AAV–CARP–FP into the TA muscle, trans-
duction efficiency was visualized by fluorescence microscopy. Note that most observable fibres expressed the construct, apart from a few
negative fibres, which reflected the fluorescence background level. Scale bar: 50 lm. (B) The level of CARP transcript overexpression was
evaluated by qRT-PCR. Results are expressed as percentage of expression level measured in untransduced control muscles. n = 5–7;
**P < 0.01 for significance between AAV–CARP–FP-injected TA muscle and contralateral control. (C) Expression of CARP protein was evalu-

ated by western blot. Equivalent amounts of proteins were resolved, and Ponceau red staining was also used to confirm the standardization
of the loading. (D) Weights of injected TA muscles (n = 13) were compared to those of control samples. No significant difference was
observed. (E) Histological analyses of muscles. Frozen sections of injected TA muscles (right panel) stained with haematoxylin–phloxin–sa-
fran show features identical to normal sections (left panel). Scale bars: 20 lm. (F) Morphometric analysis of muscles overexpressing CARP.
The number of fibres and their minimum diameter in injected muscles are not significantly different as compared to the control (n = 4). (G)
Slow fibres were detected using slow myosin immunostaining, and their numbers were determined on three slices of the TA muscle mid-
section. The number of slow fibres is reduced significantly (*P < 0.05) in CARP-expressing muscles as compared to noninjected muscles,
indicating that CARP can influence the fibre type (n = 6).
L. Laure et al. Cardiac ankyrin repeat protein in muscle plasticity
FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS 677
Third, the most striking evidence obtained from our
investigation is that CARP expression appears to be
persistently upregulated in denervation-induced atro-
phy and is also elevated in all the MD models investi-
gated. This last observation adds to the panel of muscle
pathologies already reported to be associated with an
increase in CARP expression: DMD, spinal muscular
atrophy, facio-scapulo-humeral muscular dystrophy,
amyotrophic lateral sclerosis, and peroxisome prolifera-
tor-activated receptor-induced myopathy [41], as well
as the mdx, Swiss Jim Lambert (SJL) and muscular
dystrophy with myositis (MDM) animal models, defi-
cient respectively in dystrophin, dysferlin and titin [42–
48]. Overall, CARP seems to be a general marker of
muscle damage. The reason(s) for CARP upregulation
remain(s) obscure, and whether CARP expression par-
ticipates in or represents an attempt to resist the unre-
lenting muscle degeneration is an important issue.
It is of interest that CARP is the only protein show-
ing a variation of profile between transient and defini-

tive denervation, with persistence of upregulation in
the latter condition. The CARP profile precisely
reflects muscle atrophy, which could be consistent with
the idea that CARP is an important factor in this
mechanism. However, several facts support the idea
that CARP probably has no active part to play in
muscle atrophy per se. First, there is no consistent
positive correlation between CARP expression and
atrophic situations [9,11], and it can even be upregulated
when skeletal muscle mass increases [18–21]. Second,
in our hands, CARP overexpression in a normal
muscle background failed to induce significant changes
in the number and calibre of fibres.
Interestingly, although CARP is upregulated to very
similar levels in both denervation and MD models,
two different CARP expression sites are observed, in
Pax7-positive mononucleated cells and within the cyto-
plasm of large myofibres, suggesting that CARP plays
a role in myogenic activation, as well as in mature
fibres. It is possible that a common molecular signal-
ling pathway encompassing CARP and p21
WAF1/CIP1
occurs at these two locations. Indeed, among the three
potential target genes tested herein, the p21
WAF1/CIP1
gene is the only one whose expression matches strictly
with CARP level. In addition, p21
WAF1/CIP1
expression
was observed at the same locations (proliferating myo-

blasts [49] and terminally differentiated myotubes [50])
as CARP overexpression. First, in the skeletal myo-
genic lineage, p21
WAF1/CIP1
upregulation leads to the
irreversible withdrawal of myoblasts from the cell
cycle, stimulates differentiation, and confers protection
against apoptosis [49]. However, intense regeneration
is still ongoing in both the Sgca-null and mdx
4Cv
mod-
els, which suggests that either p21
WAF1/CIP1
is not
inhibiting the cell cycle or else that the inhibition pro-
cess is not entirely efficient. Second, p21
WAF1/CIP1
has
previously been reported to be upregulated within the
myonuclei of denervated muscles, a location where it
might be required to protect fibres against denerva-
tion-induced apoptosis [50]. Taken as a whole, the
findings in the MD and denervation models studied
herein suggest that the systematic upregulation of
p21
WAF1/CIP1
whenever CARP expression increases
might oppose cell proliferation and/or inhibit apop-
tosis, thus preventing muscle remodelling.
It should also be noted that muscle ankyrin repeat

proteins, which include CARP, have recently been sug-
gested to be important for sarcomere length stability
and muscle stiffness and to have an inhibitory role in
the regenerative response of muscle tissue [30]. Here,
we showed that CARP overexpression induces a switch
towards fast-twitch fibres. All of these elements add to
the previous observations related to the effects of
p21
WAF1/CIP1
, and support the idea that CARP plays a
global role in muscle plasticity. Accordingly, the main-
tenance of CARP expression during chronic denerva-
tion suggests that this protein plays an active part in
this static condition and might contribute actively to
the prevention of remodelling through blockade of
adaptive pathways during deleterious muscle processes.
Interestingly, besides MDs, CARP has been reported
to be upregulated in many other pathological tissues
(hypertrophic hearts [12–17], nephropathic kidneys
[51], and wounded epidermis [52]), which suggests that
CARP is a widely spread marker of tissue alterations.
If the consequences of CARP overexpression prove
to be detrimental for the skeletal muscle, impeding
CARP expression would seem to be especially interest-
ing, as CARP expression is increased in many different
muscle diseases. Considering that transforming growth
factor-b, tumour necrosis factor-a and interleukin-1a
are known stimuli of CARP expression, pharmacologi-
cally targeting these pathways might be an option.
Indeed, as it has already been demonstrated that drug-

mediated inhibition of tumour necrosis factor-a [53] or
transforming growth factor-b [54] in the mdx mouse
model greatly improves the muscle histology, it would
be interesting to investigate the role of CARP in these
signalling pathways.
Experimental procedures
Animals
All mice were handled in accordance with the European
guidelines for the humane care and use of experimental
Cardiac ankyrin repeat protein in muscle plasticity L. Laure et al.
678 FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS
animals. The C3-null model corresponds to complete inac-
tivation of calpain 3 expression. Exon 1 of the calpain 3
gene has been targeted by an IRES–LacZ–Lox-PGK–
hygro–lox cassette to allow the expression of the LacZ
transgene and prevent the expression of the calpain 3 gene
from the muscle-specific promoter upstream of exon 1.
The a-sarcoglycan-deficient (Sgca-null) model has been
previously described [28]. The mdx
4Cv
model is an engi-
neered model carrying a missense mutation in exon 53 of
the dystrophin gene [55,56] and was obtained from the
Jackson Laboratory (Ann Harbor, USA). A/J mice, which
have a retrotransposon insertion in intron 4 of the dysfer-
lin gene [26], were backcrossed with C57BL/6 for four
generations and renamed B6.A/J-dysf
prmd
. Control mice
from the 129SvPasIco and C57BL/6 strains were pur-

chased from Charles River Laboratories (Les Oncins,
France).
In vivo experiments
To obtain denervation of the muscle of the lower limb,
the dorsal skin of the thigh of 2-month-old male
129SvPasIco mice (n = 6 for each time point) was cut,
and the posterior muscles were split apart to reveal the
sciatic nerve. Chronic denervation was obtained by resec-
tion of a 5 mm nerve segment. To avoid regeneration, the
proximal end of the nerve was ligatured. In contrast, for
transient denervation, the sciatic nerve was crushed for
10 s at midthigh with a no.°5 Dumont forceps. This con-
dition allows nerve regeneration. In each group, addi-
tional animals were sham operated, and their muscles
were used as controls. In these animals, the skin was cut,
and muscles or sciatic nerves were visualized, but no
treatment was applied. TA muscles were sampled 3, 9, 14
and 21 days after denervation. Chronic denervation was
checked at the time of muscle excision by visualization of
abnormal gait of the limb and by verifying the disconti-
nuity of the sciatic nerve at the thigh. The TA, EDL and
soleus muscles were sampled at the appropriate times
(after 3, 9, 14, 21 days) after transient or definitive dener-
vation.
To obtain CARP overexpression in muscle, 2 · 10
11
viral
genomes suspended in 25 lL of recombinant AAV2/1 viral
preparation were injected into the left TA muscle of anaes-
thetized 2-month-old 129SvPasIco male mice (n = 13).

TA muscles were sampled 1 month after injection, and
directly observed using confocal fluorescence microscopy
(emission wavelength used for data collection: 514 nm) to
allow visualization of YFP fluorescence. Muscles were then
quickly frozen in liquid nitrogen.
Quantitative RT-PCR
Quadriceps, EDL, TA, soleus and psoas muscles were
sampled from four animals aged 4 months for the mdx
4Cv
,
B6.A/J-dysf
prmd
and Sgca-null models and their controls,
and aged 7–8 months for the C3-null model and its con-
trol. Muscles were chosen in order to reflect the muscle
impairment specificity – which varies between models –
and the type of fibres composing the muscle. In the
C3-null model, the quadriceps muscle is unaffected by the
disease, the EDL and TA muscles are weakly affected,
and the soleus and psoas muscles are the most strongly
affected (C. Roudaut & I. Richard, unpublished data). In
the A/J mouse, the first dystrophic signs are seen at
2 months of age, and progress as a function of age. The
quadriceps femoris and triceps brachii muscles are the
most severely affected, whereas the gastrocnemius, soleus
and tibialis anterior muscles show mild histopathology,
even at late stages of the disease [26]. Placing the A/J
mutation on a C57BL/6 background has no effect on the
disease presentation (N. Bourg & I. Richard, unpublished
data). In the Sgca-null and mdx

4Cv
models, every muscle
is strongly affected by the disease (even though no con-
clusions can be drawn concerning the psoas muscle in the
mdx
4Cv
model, as no data are currently available) [28,56].
The type of fibres forming the muscle differs between
muscles: the quadriceps muscle is composed roughly of
50% type 1 fibres and 50% type 2 fibres, the TA and
EDL muscles contain about 5–10% type 1 fibres, the
soleus muscle is composed exclusively of type 1 fibres,
and the psoas muscle is composed exclusively of type 2
fibres.
Total RNA was isolated from mouse muscle using
Trizol reagent (Gibco, BRL). cDNA was synthesized
from 1 lg of total RNA using the SuperScript first
strand synthesis system for the RT-PCR kit (Invitrogen,
Cergy Pontoise, France) and random oligonucleotides.
Expression of genes encoding NF-jB-p65, FoxO1,
E2-14 kDa, Ub, the C2, C8 and C9 subunits of the pro-
teasome, MuRF1, MAFbx, CARP, p21
WAF1/CIP1
, MLC-
2v and MLC-f was monitored by a real-time qRT-PCR
method using TaqMan probes (Perkin Elmer, Courta-
boeuf, France). The ubiquitous acidic ribosomal
phosphoprotein was used to normalize the data across
samples. Its expression was monitored by SYBRGreen
incorporation. The primer pairs and TaqMan probes used

for amplification are given in Table 1. Each experiment
was performed in duplicate and repeated at least twice.
Construction and production of viral vector
The coding sequence of CARP was obtained by PCR
amplification using mouse cDNA extracted from muscle
of a 129SvTer mouse as a template. The primers used
were 5¢-CACCATGATGGTACTGAGAG-3¢ and 5¢-GAA
TGTAGCTATGCGAGAGTTC-3¢. The PCR product was
first cloned into pcDNA3.1D/V5–His–Topo (Invitrogen),
and then transferred, after enzymatic restriction, into an
AAV-based pSMD2-derived vector [57], where the CARP
L. Laure et al. Cardiac ankyrin repeat protein in muscle plasticity
FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS 679
sequence is placed under the control of a cytomegalovirus
promoter and fused at its 5¢-end with YFP and at its
3¢-end with cyan fluorescent protein. This last construct
is named pAAV.CMV.CARP-FP. The integrity of all
constructs was confirmed by automated sequencing.
AAV2/1 viral preparations were generated by packaging
AAV2-inverted terminal repeat recombinant genomes in
AAV1 capsids using a three-plasmid transfection protocol
as previously described [58]. Recombinant vectors were
purified by using double caesium chloride gradient
ultracentrifugation followed by dialysis against sterile
NaCl/P
i
. After DNA extraction by successive treatments
with DNase I and proteinase K, viral genomes were
quantified by a TaqMan real-time PCR assay using prim-
ers and probes complementary to the inverted terminal

repeat region. The primer pairs and TaqMan MGB probes
used for inverted terminal repeat amplification were:
1AAV65/Fwd, 5¢-CTCCATCACTAGGGGTTCCTTGT
A-3¢; 64AAV65/rev, 5¢-TGGCTACGTAGATAAGTAGC
ATGGC-3¢; and AAV65MGB/taq, 5¢-GTTAATGATT
Table 1. Primers and probe sets used for qRT-PCR.
Acronym Name Accession no.
Upper primer (5¢-to3¢)
Probe (5¢-to3¢)
Lower primer (5¢-to3¢)
C2 Proteasome subunit C2 (Psma1) AF060088 289mC2.F: ATGCAACTTTATGCGCCAGG
313mC2.P: TTTGGATTCCAGATTTGTGTTTGACAGACCA
392mC2.R: GGATCTGGGTTTTGCTTCCA
C8 Proteasome subunit C8 (Psma3) AF060089 343mC8.F: TCTTGCAGACAGAGTGGCCA
391mC8.P: CGCTGTTAGACCTTTTGGCTGCAGTTTC
439mC8.R: CGCACTGTAAGACCCCAACA
C9 Proteasome subunit C9 (Psma4) AF060093 6mC9.F: TCTGCACCCTCACCGTCTTC
58mC9.P: TCTCGAAGATATGACTCCAGGACCACAATATTTTCT
135mC9.R: GGCTTCCATGGCATACTCCA
CARP Cardiac ankyrin repeat protein NM_013468 616mCARP.F: CTTGAATCCACAGCCATCCA
641mCARP.P: CATGTCGTGGAGGAAACGCAGATGTC
706mCARP.R: TGGCACTGATTTTGGCTCCT
E2-14 kDa Ubiquitin-conjugating enzyme
E2B (Ube2b)
NM_009458 83E2_14.F: GGGATTTCAAGCGATTGCAA
129E2_14.P: CGCCCCATCTGAAAACAACATCATGC
191E2_14.R: GGTGTCCCTTCTGGTCCAAA
FoxO1 Forkhead box protein O1 (FoxO1) NM_019739 1297mFoxO1.F: CTAAGTGGCCTGCGAGTCCT
1369mFoxO1.P: CCAGCTCAAATGCTAGTACCATCAGTGGGAG
1445mFoxO1.R: GTCCCCATCTCCCAGGTCAT

MAFbx F-box only protein 32 (Fbox32) NM_026346 1235mMafBx.F,
CTGGAAGGGCACTGACCATC
1265mMafBx.P, CAACAACCCAGAGAGCTGCTCCGTCTC
1353mMafBx.R, TGTTGTCGTGTGCTGGGATT
MLC-f Myosin light chain, fast BC055869 396mMLCfast.F: TGGAGGAGCTGCTTACCACG
423mMLCfast.P: ACCGATTTTCCCAGGAGGAGATCAAGAA
500mMLCfast.R: TCTTGTAGTCCACGTTGCCG
MLC-2v Myosin light chain, slow NM_010861 381mMLC-2V.F: GAAGGCTGACTATGTCCGGG
403mMLC-2V.P: ATGCTGACCACACAAGCAGAGAGGTTCTC
461mMLC-2V.R: GCTGCGAACATCTGGTCGAT
MuRF1 Tripartite motif-containing 63
(Trim63)
NM_001039048 958mMurf1.F AGGGCCATTGACTTTGGGAC
995mMurf1.P AGGAGGAGTTTACAGAAGAGGAGGCTGATGAG
1047mMurf1.R CTCTGTGGTCACGCCCTCTT
NF-jB-p65 v-Rel reticuloendotheliosis viral
oncogene homolog A (Rela)
NM_009045 M1833p65.F: GGCGGCACGTTTTACTCTTT
M1857p65.P: CGCTTTCGGAGGTGCTTTCGCAG
M1941p65.R: TCAGAGTTCCCTACCGAAGCAG
P0 Acidic ribosomal phosphoprotein XR_004667 MH181PO.F: CTCCAAGCAGATGCAGCAGA
M225PO.P: CCGTGGTGCTGATGGGCAAGAA
M267PO.R: ACCATGATGCGCAAGGCCAT
p21
WAF1/CIP1
Cyclin-dependent kinase
inhibitor 1A (p21
WAF1/CIP1
)
NM_007669 1584p21.F: GTACAAGGAGCCAGGCCAAG

1629p21.P: TCACAGGACACTGAGCAATGGCTGATC
1691p21.R: GTGCTTTGACACCCACGGTA
Ub Ubiquitin X51703 22mUbiq.F: TCGGCGGTCTTTCTGTGAG
51mUbiq.P: TGTTTCGACGCGCTGGGCG
96mUbiq.R: GTTAACAAATGTGATGAAAGCACAAA
Cardiac ankyrin repeat protein in muscle plasticity L. Laure et al.
680 FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS
AACCC-3¢. Titre is expressed as viral genome per mL
(vgÆmL
)1
).
Immunofluorescence, histological and
morphometric analysis of muscle
Cryosections (8 or 10 lm in thickness) were prepared from
frozen muscles. Transverse sections were processed for hae-
matoxylin–phloxin–safran histological staining. For deter-
mination of the number and minimal diameter of fibres,
histochemical immunostaining with a rabbit polyclonal
antibody against laminin (Dako, Trappes, France; Z0097)
was performed to delimit each fibre as follows. Unfixed
transverse cryosections were incubated for 30 min with
NaCl/P
i
/10% fetal bovine serum in order to block unspe-
cific sites, and then overnight with a 1 : 1000 dilution of
primary antibodies at room temperature. After washing
with NaCl/P
i
, sections were incubated with a secondary
antibody, goat anti-(rabbit Ig) conjugated with horseradish

peroxidase (Kit En Vision Rabbit HRP, Dako; K4002) for
30 min at room temperature. Sections were mounted with
Eukitt (Sigma-Aldrich Chimie, Lyon, France) after 4¢,
6-diamidino-2-phenylindole (DAPI) staining and visualized
on a Nikon Eclipse E60 microscope. Digital images of a
slice corresponding to the muscle midsection were acquired
with a 4 · objective, a CCD camera (Sony, Clichy, France)
and a motorized stage. Images were then analysed with
ellix software (Microvision, Evry, France).
Immunostainings were performed on transverse muscle
sections. Unfixed sections were incubated for 60 min at
room temperature in Mouse-On-Mouse blocking reagent
(Vector; MKB-2213) and then overnight at 4 °C with pri-
mary anti-CARP IgG (PTC; 11427-1-AP; dilution 1 : 50).
After extensive NaCl/P
i
washes, sections were incubated
first with biotinylated goat anti-(rabbit IgG) (Vector;
BA-1000; dilution 1 : 500) for 60 min at room temperature
and, after additional washes, with streptavidin conjugated
with Alexa Fluor 488 (Invitrogen; S11223; dilution 1 : 500).
Costainings were performed using various primary anti-
bodies [Pax7 (DSHB; Pax7-a-1ea; dilution 1 : 100) and NF
(Millipore, Saint-Quentin-en-Yvelines, France; AB5539;
dilution 1 : 50)], and detected using antibodies conjugated
to Alexa Fluor 594 (Invitrogen; goat anti-mouse A11020 or
goat anti-chicken A11042; dilutions 1 : 1000). DAPI
nuclear staining was also performed (Invitrogen; D21490).
Sections were mounted with fluoromount-G (CliniSciences,
Montrouge, France; 0100-01) and visualized on a Leica

confocal microscope. Digital images of a slice correspond-
ing to the muscle midsection were acquired with a 40·
objective, a CCD camera (Sony) and a motorized stage.
Slow-fibre staining was performed using the monoclonal
antibody anti-skeletal slow myosin, heavy chain (dilution
1 : 1000; Sigma; M-8421) on transverse sections correspond-
ing to the muscle midsection according to the protocol of
the ARK peroxidase kit (Dako). Sections were mounted
with Eukitt (fluka) and visualized on a Nikon Eclipse E60
microscope. Digital images were acquired with a 4· objec-
tive, a CCD camera (Sony) and a motorized stage. The
number of positive muscle fibres was determined on three
levels of sections per muscle, with cartograph software
(Microvision, Evry, France) (n = 6 for each condition).
DNA fragmentation was visualized on muscle transverse
sections fixed with 3.7% formalin using the TUNEL tech-
nique according to the manufacturer’s protocol (Roche
Applied Science, Meylan, France). Muscle slices were
mounted in Vectashield containing DAPI for analysis by
fluorescence microscopy on a Leica confocal microscope.
Positive controls were performed on some sections using
0.5 mgÆmL
)1
DNase I (Sigma) in NaCl/Tris at room
temperature for 10 min prior to TUNEL staining.
DNase1-treated sections incubated with fluorescein isothio-
cyanate-labelled nucleotide mixture, without addition of
terminal deoxynucleotidyl transferase, were used as negative
controls.
Western blot

CARP-specific polyclonal antibodies were obtained by injec-
tion of the KTLPANSVKQGEEQRK peptide into rabbits.
Horseradish peroxidase-linked donkey anti-(rabbit IgG) and
sheep anti-(mouse IgG) antibodies were purchased from
GE Healthcare Europe (Saclay, France).
Muscles from control and denervated mice were
homogenized using an ultra-turrax T8 in a buffer contain-
ing 20 mm Tris (pH 7.5), 150 mm NaCl, 2 mm EGTA,
1% Triton X-100, 2 lm E64 and protease inhibitors (com-
plete mini protease inhibitor cocktail; Roche Applied
Science). After centrifugation at 10 000 g for 10 min at
4 °C, the supernatants were recovered for western blot
analysis. The samples were denatured before SDS/PAGE
using LDS NuPage buffer (Invitrogen) supplemented with
100 mm dithiothreitol. Protein concentrations were deter-
mined by the bicinchoninic acid methodology (Thermo
Fisher Scientific, Brebie
`
res, France). Fifty micrograms of
protein samples were subjected to SDS/PAGE in precast
4–12% acrylamide gradient gels (NuPage system; Invitro-
gen) and transferred onto poly(vinylidene difluoride) mem-
branes (Millipore) by the application of an electric field
(100 V, 1 h). The transfer efficiency was evaluated by
Ponceau red protein staining (0.2% w/v, in 5% acetic
acid). The membranes were probed with antibodies
against CARP (dilution 1 : 500). Detection was performed
with secondary antibodies (dilution 1 : 10 000) coupled to
horseradish peroxidase. Visualization was performed using
the SuperSignal West Pico substrate (Pierce).

Statistics
Data are presented as means ± standard error of the
mean. Individual means and distributions were compared
L. Laure et al. Cardiac ankyrin repeat protein in muscle plasticity
FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS 681
using the Mann–Whitney and the Kolmogorov–Smirnov
nonparametric tests, respectively. Differences were
considered to be statistically significant at P < 0.05 or
P < 0.01.
Acknowledgements
We acknowledge the excellent technical expertise
of I. Adamski, L. Arandel, C. Georger, A. Jollet,
T. Marais, A. Verdot, and L. Van Wittenberghe. We
thank Dr M. Barkats and the Howard Hughes Medi-
cal Institute for providing us with NF specific antibody
and Sgca-null mice (Iowa City, USA), respectively. We
are grateful to S. Lupton for checking the language.
This work was funded by the Association Franc¸ aise
contre les Myopathies.
References
1 Emery AE (2002) The muscular dystrophies. Lancet
359, 687–695.
2 Daniele N, Richard I & Bartoli M (2007) Ins and outs
of therapy in limb girdle muscular dystrophies. Int J
Biochem Cell Biol 39, 1608–1624.
3 Taylor RG, Tassy C, Briand M, Robert N, Briand Y &
Ouali A (1995) Proteolytic activity of proteasome on
myofibrillar structures. Mol Biol Rep 21, 71–73.
4 Scheffner M, Nuber U & Huibregtse JM (1995) Protein
ubiquitination involving an E1–E2–E3 enzyme ubiquitin

thioester cascade. Nature 373, 81–83.
5 Nury D, Doucet C & Coux O (2007) Roles and poten-
tial therapeutic targets of the ubiquitin proteasome sys-
tem in muscle wasting. BMC Biochem 8(Suppl. 1), S7.
6 Sandri M, Sandri C, Gilbert A, Skurk C, Calabria E,
Picard A, Walsh K, Schiaffino S, Lecker SH & Gold-
berg AL (2004) Foxo transcription factors induce the
atrophy-related ubiquitin ligase atrogin-1 and cause
skeletal muscle atrophy. Cell 117, 399–412.
7 Stitt TN, Drujan D, Clarke BA, Panaro F, Timofeyva
Y, Kline WO, Gonzalez M, Yancopoulos GD & Glass
DJ (2004) The IGF-1/PI3K/Akt pathway prevents
expression of muscle atrophy-induced ubiquitin ligases
by inhibiting FOXO transcription factors. Mol Cell 14,
395–403.
8 Cai D, Frantz JD, Tawa NE Jr, Melendez PA, Oh BC,
Lidov HG, Hasselgren PO, Frontera WR, Lee J, Glass
DJ et al. (2004) IKKbeta/NF-kappaB activation causes
severe muscle wasting in mice. Cell 119, 285–298.
9 Baumeister A, Arber S & Caroni P (1997) Accumulation
of muscle ankyrin repeat protein transcript reveals local
activation of primary myotube endcompartments during
muscle morphogenesis. J Cell Biol 139, 1231–1242.
10 Yang W, Zhang Y, Ma G, Zhao X, Chen Y & Zhu D
(2005) Identification of gene expression modifications in
myostatin-stimulated myoblasts. Biochem Biophys Res
Commun 326, 660–666.
11 Stevenson EJ, Giresi PG, Koncarevic A & Kandarian
SC (2003) Global analysis of gene expression patterns
during disuse atrophy in rat skeletal muscle. J Physiol

551, 33–48.
12 Aihara Y, Kurabayashi M, Saito Y, Ohyama Y,
Tanaka T, Takeda S, Tomaru K, Sekiguchi K, Arai M,
Nakamura T et al. (2000) Cardiac ankyrin repeat pro-
tein is a novel marker of cardiac hypertrophy: role of
M-CAT element within the promoter. Hypertension 36,
48–53.
13 Ihara Y, Suzuki YJ, Kitta K, Jones LR & Ikeda T
(2002) Modulation of gene expression in transgenic
mouse hearts overexpressing calsequestrin. Cell Calcium
32, 21–29.
14 Arber S, Hunter JJ, Ross J Jr, Hongo M, Sansig G,
Borg J, Perriard JC, Chien KR & Caroni P (1997)
MLP-deficient mice exhibit a disruption of cardiac
cytoarchitectural organization, dilated cardiomyopathy,
and heart failure. Cell 88, 393–403.
15 Zolk O, Frohme M, Maurer A, Kluxen FW, Hentsch
B, Zubakov D, Hoheisel JD, Zucker IH, Pepe S &
Eschenhagen T (2002) Cardiac ankyrin repeat protein, a
negative regulator of cardiac gene expression, is aug-
mented in human heart failure. Biochem Biophys Res
Commun 293, 1377–1382.
16 Miller MK, Bang ML, Witt CC, Labeit D, Trombitas
C, Watanabe K, Granzier H, McElhinny AS, Gregorio
CC & Labeit S (2003) The muscle ankyrin repeat pro-
teins: CARP, ankrd2/Arpp and DARP as a family of
titin filament-based stress response molecules. J Mol
Biol 333, 951–964.
17 Nagueh SF, Shah G, Wu Y, Torre-Amione G, King
NM, Lahmers S, Witt CC, Becker K, Labeit S &

Granzier HL (2004) Altered titin expression, myocardial
stiffness, and left ventricular function in patients
with dilated cardiomyopathy. Circulation 110, 155–162.
18 Carson JA, Nettleton D & Reecy JM (2002) Differential
gene expression in the rat soleus muscle during early
work overload-induced hypertrophy. FASEB J 16, 207–
209.
19 Hentzen E, Lahey M, Peters D, Mathew L, Barash I,
Friden J & Lieber R (2006) Stress-dependent and
independent expression of the myogenic regulatory
factors and the Marp genes after eccentric contractions
in rats. J Physiol 570, 157–167.
20 Barash IA, Mathew L, Ryan AF, Chen J & Lieber
RL (2004) Rapid muscle-specific gene expression
changes after a single bout of eccentric contractions
in the mouse. Am J Physiol Cell Physiol 286, C355–
C364.
21 Chen YW, Nader GA, Baar KR, Fedele MJ, Hoffman
EP & Esser KA (2002) Response of rat muscle to acute
Cardiac ankyrin repeat protein in muscle plasticity L. Laure et al.
682 FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS
resistance exercise defined by transcriptional and trans-
lational profiling. J Physiol 545, 27–41.
22 Jeyaseelan R, Poizat C, Baker RK, Abdishoo S,
Isterabadi LB, Lyons GE & Kedes L (1997) A novel
cardiac-restricted target for doxorubicin. CARP, a
nuclear modulator of gene expression in cardiac progen-
itor cells and cardiomyocytes. J Biol Chem 272, 22800–
22808.
23 Zou Y, Evans S, Chen J, Kuo HC, Harvey RP & Chien

KR (1997) CARP, a cardiac ankyrin repeat protein, is
downstream in the Nkx2-5 homeobox gene pathway.
Development 124, 793–804.
24 Kanai H, Tanaka T, Aihara Y, Takeda S, Kawabata
M, Miyazono K, Nagai R & Kurabayashi M (2001)
Transforming growth factor-beta/Smads signaling
induces transcription of the cell type-restricted ankyrin
repeat protein CARP gene through CAGA motif in
vascular smooth muscle cells. Circ Res 88, 30–
36.
25 Lefaucheur JP & Sebille A (1995) The cellular events of
injured muscle regeneration depend on the nature of the
injury. Neuromuscul Disord 5, 501–509.
26 Ho M, Post CM, Donahue LR, Lidov HG, Bronson
RT, Goolsby H, Watkins SC, Cox GA, Brown RH Jr
(2004) Disruption of muscle membrane and phenotype
divergence in two novel mouse models of dysferlin defi-
ciency. Hum Mol Genet 13, 1999–2010.
27 Chapman VM, Miller DR, Armstrong D & Caskey CT
(1989) Recovery of induced mutations for X chromo-
some-linked muscular dystrophy in mice. Proc Natl
Acad Sci USA 86, 1292–1296.
28 Duclos F, Straub V, Moore SA, Venzke DP, Hrstka
RF, Crosbie RH, Durbeej M, Lebakken CS, Ettinger
AJ, van der Meulen J et al. (1998) Progressive muscular
dystrophy in alpha-sarcoglycan-deficient mice. J Cell
Biol 142, 1461–1471.
29 Baker PE, Kearney JA, Gong B, Merriam AP, Kuhn
DE, Porter JD & Rafael-Fortney JA (2006) Analysis of
gene expression differences between utrophin/dystro-

phin-deficient vs mdx skeletal muscles reveals a specific
upregulation of slow muscle genes in limb muscles.
Neurogenetics 7, 81–91.
30 Barash I, Bang ML, Mathew L, Greaser ML, Chen J &
Lieber RL (2007) Structural and regulatory roles of the
muscle ankyrin repeat protein family in skeletal muscle.
Am J Physiol Cell Physiol 293, C218–C227.
31 Bodine SC, Latres E, Baumhueter S, Lai VK, Nunez L,
Clarke BA, Poueymirou WT, Panaro FJ, Na E,
Dharmarajan K et al. (2001) Identification of
ubiquitin ligases required for skeletal muscle atrophy.
Science 294, 1704–1708.
32 Zhao J, Zhang Y, Zhao W, Wu Y, Pan J, Bauman WA
& Cardozo C (2008) Effects of nandrolone on denerva-
tion atrophy depend upon time after nerve transection.
Muscle Nerve 37, 42–49.
33 Sacheck JM, Hyatt JP, Raffaello A, Jagoe RT, Roy
RR, Edgerton VR, Lecker SH & Goldberg AL (2007)
Rapid disuse and denervation atrophy involve transcrip-
tional changes similar to those of muscle wasting during
systemic diseases. FASEB J 21, 140–155.
34 Lecker SH (2003) Ubiquitin–protein ligases in muscle
wasting: multiple parallel pathways? Curr Opin Clin
Nutr Metab Care 6, 271–275.
35 Kamei Y, Miura S, Suzuki M, Kai Y, Mizukami J,
Taniguchi T, Mochida K, Hata T, Matsuda J, Abura-
tani H et al. (2004) Skeletal muscle FOXO1 (FKHR)
transgenic mice have less skeletal muscle mass, down-
regulated type I (slow twitch/red muscle) fiber genes,
and impaired glycemic control. J Biol Chem 279,

41114–41123.
36 Zhao J, Brault JJ, Schild A, Cao P, Sandri M,
Schiaffino S, Lecker SH & Goldberg AL (2007) FoxO3
coordinately activates protein degradation by the
autophagic/lysosomal and proteasomal pathways in
atrophying muscle cells. Cell Metab 6, 472–483.
37 Mammucari C, Milan G, Romanello V, Masiero E,
Rudolf R, Del Piccolo P, Burden SJ, Di Lisi R, Sandri
C, Zhao J et al. (2007) FoxO3 controls autophagy in
skeletal muscle in vivo. Cell Metab 6, 458–471.
38 Mammucari C, Schiaffino S & Sandri M (2008) Down-
stream of Akt: FoxO3 and mTOR in the regulation of
autophagy in skeletal muscle. Autophagy 4, 524–526.
39 Rathbone CR, Booth FW & Lees SJ (2008) FoxO3a
preferentially induces p27Kip1 expression while impair-
ing muscle precursor cell-cycle progression. Muscle
Nerve 37, 84–89.
40 Bechet D, Tassa A, Taillandier D, Combaret L &
Attaix D (2005) Lysosomal proteolysis in skeletal
muscle. Int J Biochem Cell Biol 37, 2098–2114.
41 Casey WM, Brodie T, Yoon L, Ni H, Jordan HL &
Cariello NF (2008) Correlation analysis of gene
expression and clinical chemistry to identify biomarkers
of skeletal myopathy in mice treated with PPAR agonist
GW610742X. Biomarkers 13, 364–376.
42 Bakay M, Zhao P, Chen J & Hoffman EP (2002) A
web-accessible complete transcriptome of normal
human and DMD muscle. Neuromuscul Disord 12(Sup-
pl. 1), S125–S141.
43 Nakamura K, Nakada C, Takeuchi K, Osaki M,

Shomori K, Kato S, Ohama E, Sato K, Fukayama M,
Mori S et al. (2002) Altered expression of cardiac anky-
rin repeat protein and its homologue, ankyrin repeat
protein with PEST and proline-rich region, in atrophic
muscles in amyotrophic lateral sclerosis. Pathobiology
70, 197–203.
44 Nakada C, Tsukamoto Y, Oka A, Nonaka I, Takeda S,
Sato K, Mori S, Ito H & Moriyama M (2003) Cardiac-
restricted ankyrin-repeated protein is differentially
induced in Duchenne and congenital muscular dystro-
phy. Lab Invest 83, 711–719.
L. Laure et al. Cardiac ankyrin repeat protein in muscle plasticity
FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS 683
45 Nakada C, Oka A, Nonaka I, Sato K, Mori S, Ito H &
Moriyama M (2003) Cardiac ankyrin repeat protein is
preferentially induced in atrophic myofibers of congeni-
tal myopathy and spinal muscular atrophy. Pathol Int
53, 653–658.
46 Porter JD, Khanna S, Kaminski HJ, Rao JS, Merriam
AP, Richmonds CR, Leahy P, Li J, Guo W & Andrade
FH (2002) A chronic inflammatory response dominates
the skeletal muscle molecular signature in dystrophin-
deficient mdx mice. Hum Mol Genet 11, 263–272.
47 Suzuki N, Aoki M, Hinuma Y, Takahashi T, Onodera
Y, Ishigaki A, Kato M, Warita H, Tateyama M &
Itoyama Y (2005) Expression profiling with progression
of dystrophic change in dysferlin-deficient mice (SJL).
Neurosci Res 52, 47–60.
48 Witt CC, Ono Y, Puschmann E, McNabb M, Wu Y,
Gotthardt M, Witt SH, Haak M, Labeit D, Gregorio

CC et al. (2004) Induction and myofibrillar targeting
of CARP, and suppression of the Nkx2.5 pathway in
the MDM mouse with impaired titin-based signaling.
J Mol Biol 336, 145–154.
49 Walsh K & Perlman H (1997) Cell cycle exit upon
myogenic differentiation. Curr Opin Genet Dev 7,
597–602.
50 Ishido M, Kami K & Masuhara M (2004) In vivo
expression patterns of MyoD, p21, and Rb proteins in
myonuclei and satellite cells of denervated rat skeletal
muscle. Am J Physiol Cell Physiol 287, C484–C493.
51 Matsuura K, Uesugi N, Hijiya N, Uchida T & Moriy-
ama M (2007) Upregulated expression of cardiac anky-
rin-repeated protein in renal podocytes is associated
with proteinuria severity in lupus nephritis. Hum Pathol
38, 410–419.
52 Shi Y, Reitmaier B, Regenbogen J, Slowey RM, Opale-
nik SR, Wolf E, Goppelt A & Davidson JM (2005)
CARP, a cardiac ankyrin repeat protein, is up-regulated
during wound healing and induces angiogenesis in
experimental granulation tissue. Am J Pathol 166 , 303–
312.
53 Grounds MD & Torrisi J (2004) Anti-TNFalpha (Remi-
cade) therapy protects dystrophic skeletal muscle from
necrosis. FASEB J 18, 676–682.
54 Cohn RD, van Erp C, Habashi JP, Soleimani AA,
Klein EC, Lisi MT, Gamradt M, ap Rhys CM, Holm
TM, Loeys BL et al. (2007) Angiotensin II type 1 recep-
tor blockade attenuates TGF-beta-induced failure
of muscle regeneration in multiple myopathic states.

Nat Med 13, 204–210.
55 Im WB, Phelps SF, Copen EH, Adams EG, Slightom
JL & Chamberlain JS (1996) Differential expression of
dystrophin isoforms in strains of mdx mice with differ-
ent mutations. Hum Mol Genet 5, 1149–1153.
56 Judge LM, Haraguchiln M & Chamberlain JS (2006)
Dissecting the signaling and mechanical functions of the
dystrophin–glycoprotein complex. J Cell Sci 119, 1537–
1546.
57 Snyder RO, Miao CH, Patijn GA, Spratt SK, Danos
O, Nagy D, Gown AM, Winther B, Meuse L, Cohen
LK et al. (1997) Persistent and therapeutic concentra-
tions of human factor IX in mice after hepatic gene
transfer of recombinant AAV vectors. Nat Genet 16,
270–276.
58 Bartoli M, Poupiot J, Goyenvalle A, Perez N, Garcia
L, Danos O & Richard I (2006) Noninvasive monitor-
ing of therapeutic gene transfer in animal models of
muscular dystrophies. Gene Ther 13, 20–28.
Cardiac ankyrin repeat protein in muscle plasticity L. Laure et al.
684 FEBS Journal 276 (2009) 669–684 ª 2008 The Authors Journal compilation ª 2008 FEBS