An adult tissue-specific stem cell in its niche: A gene
profiling analysis of in vivo quiescent and activated
muscle satellite cells
Giorgia Pallafacchinaa,1, Stéphanie Françoisa,2, Béatrice Regnaultb,
Bertrand Czarnyc, Vincent Divec, Ana Cumanod,
Didier Montarrasa,⁎, Margaret Buckinghama
aMolecular Genetics of Development Unit, Department of Developmental Biology, URA CNRS 2578, Institut Pasteur,
25 rue du Dr. Roux, 75015 Paris, France
bGenopole, Institut Pasteur, 28 rue du Dr. Roux, 75015 Paris, France
cService of Molecular Engineering of Proteins, CEA, 91191 Gif/Yvette, Saclay, France
dLymphocyte Development Unit, Department of Immunology, INSERM U668, Institut Pasteur, 25 rue du Dr. Roux,
75015 Paris, France
Received 9 October 2009; received in revised form 21 October 2009; accepted 21 October 2009
carried out transcriptome analyses on satellite cells purified by flow cytometry from Pax3GFP/+mice. We compared samples
from adult skeletal muscles where satellite cells are mainly quiescent, with samples from growing muscles or regenerating
(mdx) muscles, where they are activated. Analysis of regulation that is shared by both activated states avoids other effects due
to immature or pathological conditions. This in vivo profile differs from that of previously analyzed satellite cells activated
after cell culture. It reveals how the satellite cell protects itself from damage and maintains quiescence, while being primed for
activation on receipt of the appropriate signal. This is illustrated by manipulation of the corepressor Dach1, and by the
demonstration that quiescent satellite cells are better protected from oxidative stress than those from mdx or 1-week-old
muscles. The quiescent versus in vivo activated comparison also gives new insights into how the satellite cell controls its niche
on the muscle fiber through cell adhesion and matrix remodeling. The latter also potentiates growth factor activity through
proteoglycan modification. Dismantling the extracellular matrix is important for satellite cell activation when the expression of
proteinases is up-regulated, whereas transcripts for their inhibitors are high in quiescent cells. In keeping with this, we
demonstrate that metalloproteinase function is required for efficient regeneration in vivo.
© 2009 Elsevier B.V. All rights reserved.
The satellite cell of skeletal muscle provides a paradigm forquiescent and activated tissue stem cell states. We have
⁎ Corresponding author. Fax: +33 1 40613452.
E-mail address: email@example.com (D. Montarras).
1Present address: Department of Biomedical Sciences, University of Padova, Viale G. Colombo 3, 35121 Padova, Italy.
2Permanent address: Department of Experimental Medicine, University of Milano-Bicocca, 20052 Monza, Italy.
1873-5061/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
available at www.sciencedirect.com
Stem Cell Research (2010) 4, 77–91
Stem cells in many adult tissues are in a quiescent state.
Their contribution to tissue repair depends on activation,
which leads to proliferation and subsequent differentiation
or return to quiescence. Skeletal muscle provides a model
system for studying this pattern of cell behavior. This tissue
regenerates after injury, and the muscle satellite cell is key
to this process (Collins et al., 2005; Montarras et al., 2005).
These quiescent cells lie under the basal lamina of the
muscle fiber until activated in response to injury, when they
leave this niche and proliferate before differentiating to
form new muscle fibers or reverting tothe quiescent satellite
cell state on the newly formed fibers (Montarras and
Buckingham, 2008). The satellite cell is characterized by
the expression of Pax7 (Seale et al., 2000) and also Pax3 in
many muscles (Relaix et al., 2006).
With the advent of transcriptome analyses, it has become
possible to perform gene profiling on satellite cells and
hence to further investigate their regulation. Previous
transcriptome analyses on skeletal muscle were carried
out either with whole muscle tissue in the course of
regeneration (Zhao and Hoffman, 2004; Porter et al.,
2004; van Lunteren and Leahy, 2007) or on cultured muscle
78G. Pallafacchina et al.
cells (Seale et al., 2004; McKinnell et al., 2008; Kumar et
al., 2009). The former includes several cell types in addition
to differentiating muscle fibers, while with the latter the
question of quiescence cannot be addressed since culture
leads to immediate activation.
Strategies have now been developed which permit
characterization of satellite cells from adult muscle. They
rely on the direct isolation of cells by flow cytometry on the
basis of the expression of surface markers or targeted
reporter expression (Montarras et al., 2005; Fukada et al.,
2007; Day et al., 2007; Cerletti et al., 2008; Bosnakovski et
al., 2008; Tanaka et al., 2009). Fukada et al. (Fukada et al.,
2007) isolated quiescent satellite cells in vivo and compared
their transcriptome to that of activated satellite cells under
in vitro culture conditions. We now take this analysis an
important step further by comparing the transcriptomes of
quiescent with activated states in vivo. Our approach
depends on the purification of satellite cells by flow
cytometry from Pax3GFP/+mice (Montarras et al., 2005).
Such cells from adult skeletal muscle, which are mainly
quiescent, were compared to satellite cells, from 1-week
postnatal muscle and from adult dystrophic muscle of mdx
mice (Ikemoto et al., 2007), that are undergoing activation.
Pax3GFP-positive cells from adult muscle were also cultured,
providing a population of activated satellite cells, but with
potential modifications due to in vitro culture conditions.
Our analysis provides novel insight into the nature of the
quiescent state and the response of the satellite cell to
activation in vivo.
Results and discussion
We selected diaphragm, pectoralis, and abdominal mus-
cles to purify satellite cells from Pax3GFP/+mice by flow
cytometry (Montarras et al., 2005). In order to compare
quiescent versus activated satellite cell states, we
isolated these cells from adult (6-week-old) mice (Ad),
in which most satellite cells are quiescent, confirmed by
Ki67 labeling for which 97% of the GFP-positive satellite
cells were negative (Supplementary Fig. 1). These adult
cells were compared to cells isolated from Pax3GFP/+:mdx/
mdx(Ad.mdx) mice at the same stage, where the lack of
dystrophin has engendered major skeletal muscle degen-
eration/regeneration, with activation of satellite cells
(30% Ki67 positive, Supplementary Fig. 1), (Ikemoto et
al., 2007). The mdx mouse represents a pathological
situation and we therefore also used postnatal (1-week-
old) muscles of Pax3GFP/+
satellite cells are activated (80% Ki67-positive cells,
Supplementary Fig. 1), to ensure muscle growth (Relaix
et al., 2006). The same FACS criteria of GFP fluorescence,
size, and granularity, excluding other cell types and
muscle fibers, were applied in all cases. The GFP reporter
is relatively stable and continues to be detectable at a
low level in differentiating cells when Pax3 is down-
regulated. We took advantage of this to separate the
satellite cells into high and low GFP-positive fractions
(Fig. 1A). Antibody staining of cells isolated by cytospin
after flow cytometry revealed expression of some differ-
entiated muscle markers in the low GFP fraction (results
not shown). In the comparative analysis reported here, we
have concentrated on the transcriptome of the high GFP
fraction. Also included in this analysis is the transcriptome
of cultured satellite cells, for comparison with previously
published work. These ex vivo activated cells correspond
to the progeny of satellite cells from the high GFP fraction
of normal adult muscles, after 3 days in culture under
conditions that promote their proliferation.
Analyses of the microarray data are presented in
Supplementary Tables 1 to 6 and are summarized in
Figs. 1B and C. In each comparison a minimum of 1.5-fold
change was retained. We used both Benjamini-Hochberg
(BH) and Bonferroni adjustments (see Materials and
Methods), except for the comparison with ex vivo
activated cells (Supplementary Table 3) where only the
latter was applied because of the much greater number of
differences. The highly stringent Bonferroni adjustment
reduces the number of genes identified in the compar-
isons, but with a risk of eliminating transcripts of interest.
When the two in vivo states are compared, this results in
217 genes which can be regarded as potential markers of
quiescent satellite cells in adult muscles (Fig. 1B,
Supplementary Table 4) and 398 which are up-regulated
on in vivo activation (Fig. 1C, Supplementary Table 5).
We then employed the more commonly used BH
mice (1wk), where many
regulated in quiescent satellite cells. (A) Flow cytometry diagram showing high and low GFP fluorescent fractions of satellite cells
from adult skeletal muscle. (B) Comparison of the transcriptome of Pax3GFP/+cells from adult muscle (Ad), in which most satellite cells
are quiescent, with those from regenerating Pax3GFP/+:mdx/mdx (Ad.mdx, blue), growing 1-week-old muscle (1wk, yellow), or adult
Pax3GFP/+cells after 3 days in culture (Ad.cult, pink). The number of genes with significant differential expression was identified using
two algorithms for multiple comparisons: Benjamini-Hochberg (BH), results within parentheses, and the more stringent Bonferroni
algorithm, results shown in bold. The numbers of genes which are up-regulated in Ad satellite cells compared to Ad.mdx and 1wk
activated states are shown hatched in red. (C) A similar comparison showing the numbers of genes up-regulated in activated Ad.mdx
and 1wk compared to Ad satellite cells (hatched in blue). (D) An intensity map, showing differences in transcript levels in satellite cells
from Ad, Ad.mdx, 1wk, and Ad.cult experimental groups. Triplicate samples are shown for each group. Blue color indicates low and
red indicates high relative transcript levels. Genes shown were selected from the list presented in Supplementary Tables 4 and 5 and
include those involved in cell cycle progression and cell division (cyclins/Ccn, Cdc, Cdk), DNA replication (Rfc, Orc, Pole, Dna2l), or
transcriptional activity (E2f, Rbl1/p107). (E) Transcript levels measured by qPCR for cyclin E1, calcitonin receptor (CalcR),
dermatopontin (Dpt), and Nap1l5. Samples are as in D. (F) Immunofluorescence on a single fiber preparation at time zero (t0) and after
3 days in culture (t3) showing satellite cells labeled with a Pax7 or a Nap1l1 antibody. Hoechst staining of Pax7-positive satellite cell
nuclei (arrow heads) and myonuclei is shown. Bar=100 μm.
Scheme of the experimental strategy and validation of the in vivo models: genes associated with proliferation are down-
79Gene profiling of muscle satellite cells in vivo
adjustment to analyze the microarray data, in order to
include more genes of potential interest in our compar-
isons. With the BH adjustment, these numbers are
respectively 1186 (Fig. 1B, Supplementary Table 4) and
1870 (Fig. 1C, Supplementary Table 5). This common
fraction obviates variations due to the pathological state
of adult mdx mice, or to physiological conditions
associated with growth of 1-week postnatal mice, as
well as the many changes which are introduced as a result
of culture. In subsequent analyses, we focused our
attention on genes showing changes in both Ad-Ad.mdx
and Ad-1wk comparisons after the BH adjustment and at
least in one of the two after Bonferroni adjustment.
Genes associated with proliferation
When quiescent cells are activated in response to a
requirement for muscle growth or repair, they proliferate.
Examination of transcripts of genes associated with prolif-
eration shows that they are up-regulated in cells from
postnatal or regenerating muscle, as well as in culture, thus
involvement of Dach1 in the inhibition of satellite cell proliferation and differentiation. (A) An intensity map, showing differences in
transcript levels, presented as in Fig. 1D. ⁎ did not show differences in the in vivo Bonferroni comparisons. (B) Transcript levels
measured by qPCR for Pax7, Pax3, Dach1, Sox17, and Klf9. Samples are as in Fig. 1E. (C–E) Transduction of freshly isolated satellite
cells from adult Pax3GFP/+muscles by a lentiviral vector expressing human (h) Dach1 or control RFP. (C) Western blot analysis showing
overexpression of h Dach1 and RFP proteins. β-Tubulin is shown as a control for protein loading. (D) Immunofluorescence staining of
satellite cells using anti-RFP or anti-hDach1 antibodies (red) or anti-myogenin (Myog) antibodies (green) after 4 days of culture. Bar =
100 µm. (E) Proliferation (left panel), after 3 days of culture, is expressed as the percentage of colonies formed by Dach1-transduced
cells compared to RFP-transduced cells. Differentiation (right panel), measured by the level of myogenin mRNA determined by qPCR
(black bars) and the number of myogenin expressing cells (green bars) detected by immunofluorescence, is similarly expressed as a
percentage. Two independent experiments with triplicate samples were carried out for each measurement. Error bars represent
Changes in the expression of genes encoding transcription factors associated with Pax-positive progenitor cell behavior:
80G. Pallafacchina et al.
validating the in vivo models and demonstrating that normal
adult satellite cells are mainly quiescent. This is the case for
the Gene Ontology classes referring to cell cycle progression.
Supplementary Table 6 shows, as an example, the list of
genes found in the DNA replication class. A selection of genes
involved in cell cycle progression is shown in the intensity
map (Fig. 1D) with Cyclin E1 (Ccne1) shown as an example of
qPCR analysis (Fig. 1E). Transcripts of the calcitonin receptor
(Calcr), which was identified as a quiescent satellite cell
marker based on ex vivo comparison (Fukada et al., 2007),
are also lower in activated cells in vivo, but the comparison
shows a much more striking (160-fold) down-regulation in
cultured cells (Fig. 1E). Dermatopontin (Dpt) also referred to
as “Early quiescence 1” (Supplementary Reference 1, SR1)
provides another example of a gene that is more highly
expressed in adult quiescent satellite cells (Fig. 1E) and
strikingly down-regulated in culture.
Transcripts for major chromatin remodeling complexes
associated with cell proliferation/transcription are up-
regulated in Ad.mdx and 1wk satellite cells. This is
illustrated by Smarca4/Brg1 and Smarcc1, components of
the SWI/SNF complex (Fig. 1D), also directly implicated in
the promotion of myogenesis (Ohkawa et al., 2007). An
interesting example is provided by members of the Nap
family of nucleosome assembly factors (SR2). Nap1l1
accumulates after satellite cell activation in culture (Fig.
1F), as expected for the role of this protein in chromatin
assembly after cell division. However, Nap1l5 transcripts are
high in quiescent satellite cells (Fig. 1E), suggesting a
Transcription factors associated with Pax-positive
myogenic progenitor cells
In Fig. 2A, the expression of genes encoding transcription
factors potentially involved in maintenance of the muscle
progenitor cell state, as illustrated by the presence of Pax3/
7, is compared. Pax7, and to a lesser extent Pax3, transcripts
are relatively higher in Ad satellite cells (Figs. 2A and B). Six
homeodomain proteins, Six1 and Six4, are important
upstream regulators of myogenesis intervening with Pax3
and Pax7 in the embryo (Buckingham and Relaix, 2007). Six1
transcripts and those for the Six coactivators Eya1, 3, and 4,
are present, but show little differences, except in cultured
cells (Supplementary Tables 1–3). Six4, also transcribed in
Ad cells, shows up-regulation in Ad.mdx and 1wk samples
(Fig. 2A). Transcripts for Dach1 are notably up-regulated in
quiescent adult satellite cells (Fig. 2B). Dach1 is potentially
involved in repressing Six activity (Li et al., 2003), thus
avoiding myogenic activation by Six in quiescent satellite
cells. Dach1 also inhibits cyclin D1 and cell cycle progression
(Wu et al., 2006), which may reflect a role in maintenance of
quiescence. In order to address these issues, freshly isolated
adult satellite cells were transduced with a lentiviral vector
carrying human (h) Dach1 cDNA and their proliferation and
differentiation potential was assayed in culture (Figs. 2C–E).
We observed a 46% reduction in the capacity of the cells to
proliferate as measured by the number of colonies formed in
culture (Fig. 2E). An effect on differentiation was detected
at the onset of terminal differentiation, marked by the
expression of myogenin. A 2.2-fold reduction in myogenin
expression was found at both the transcript and the protein
level in Dach1-transduced cells, compared to control cells
transduced with a lentiviral vector encoding RFP (Figs. 2D
and E). Taken together, these results argue in favor of a role
of Dach1 in the control of proliferation and differentiation in
muscle satellite cells.
Meox2, expressed in myogenic progenitor cells in the
embryo and implicated in limb myogenesis (Mankoo et al.,
1999), is transcribed at a high level in Ad satellite cells and,
like Dach1, is down-regulated in Ad.mdx and 1wk samples
and notably in cultured satellite cells. Pitx3, which encodes
another homeobox transcription factor active in later fetal
myogenesis, is expressed at a lower level in adult quiescent
satellite cells compared to activated cells, in keepingwith its
role in myogenesis (L'Honore et al., 2007).
Transcripts of a number of Sox genes are present in
satellite cells (Supplementary Tables 1–3). Sox7, Sox17, and
Sox18 transcripts are higher in quiescent adult satellite cells
(Figs. 2A and B). The Sox family of transcription factors acts
with Pax factors in the specification of a number of tissues
(Buckingham and Relaix, 2007). This might suggest that these
members of the Sox family are involved in the maintenance
of myogenic identity and/or quiescence of adult satellite
cells. Our analysis also revealed increased levels of tran-
scripts encoding Prdm16 (Fig. 2A), which is a cotranscrip-
tional activator of the nuclear receptor Pparγ, a master
regulator of adipocyte differentiation. Pparγ is also tran-
scribed at higher levels in Ad cells (Supplementary Tables 1,
2, and 4). Prdm16 has been implicated in switching myogenic
precursor cells to the brown fat lineage (Kajimura et al.,
2009) and the presence of these transcripts probably reflects
some degree of intrinsic plasticity of satellite cells, which
can undergo adipogenesis under certain conditions (Asakura
et al., 2001; Shefer et al., 2004).
This may also be suggested by the high levels of Klf7 and
Klf9 transcripts in these cells (Figs. 2A and B). Klf4, another
member of this family of Kruppel-like transcription factors,
is associated with the acquisition of pluripotency (Takahashi
and Yamanaka, 2006). Klf9, like Klf4, also inhibits prolifer-
ation (Good and Tangye, 2007), thus pointing to a potential
role in quiescence.
Canonical Wnt and Notch signaling pahways have been
implicated in satellite cell activation (Brack et al., 2008).
Consistent with this, we see higher levels of transcripts for
Wisp1 (Wnt induced signaling pathway protein 1) (Figs. 3A
and B) and Hes6, a readout of Notch signaling (Figs. 3C and D)
in Ad.mdx and 1wk samples. In contrast, transcripts for
Axud1, also referred to as Csrpn1, are higher in Ad cells
(Figs. 3A and B). This gene is up-regulated by Axin 1, an
inhibitor of Wnt signaling. Noncanonical Ca2+/PKC-depen-
dent Wnt signaling, onthe other hand, maybe involved in the
maintenance of quiescence (Otto et al., 2008), as indicated
by transcript levels of PKCη (Prkch) (Fig. 3A). Small changes
(detected after the BH adjustment) were observed in
transcripts for Frizzled7 and Prickle3, components of the
noncanonical Wnt planar cell polarity pathway (Supplemen-
tary Tables 1 and 2). Interestingly Lnx1, which encodes a
Numb interacting scaffold protein, is highly expressed in
81Gene profiling of muscle satellite cells in vivo
quiescent satellite cells. Transcripts for Numb, which
antagonizes Notch activity, are higher in Ad cells (BH
adjustment only). Lnx1 also interacts with cell adhesion
molecules involved in apical tight junction formation (SR3)
and may therefore play a role in asymmetric cell division of
satellite cells on the muscle fiber, observed at the onset of
activation (Kuang et al., 2007).
Changes in the expression of the Igf (Figs. 3D and E) and
Pdgf (Supplementary Figs. 2A and B) signaling pathways
reflect their involvement in satellite cell activation. This is
illustrated by the up-regulation of the gene for Igf binding
protein, Igfbp3, that transports Igf and potentiates its sig-
naling (SR4), whereas transcripts for Igfbp6, that sequesters
Igfs (SR4) and thus prevents activation, are high in Ad cells
(Figs. 3D and E).
Tgfβ/Bmp signaling tends to antagonize myogenesis
(Kollias and McDermott, 2008) and transcripts for compo-
nents of this pathway are highly expressed in Ad cells
(Supplementary Figs. 2C and D), in keeping with a role in
preventing myogenic progression in quiescent satellite cells.
Signaling through sphingolipids is also a feature of Ad
satellite cells (Fig. 3F), which express sphingomyelin in their
plasma membrane (Nagata et al., 2006). Transcripts for
sphingomyelin synthetase (Sgms2), that converts ceramide
to sphingomyelin, are high in quiescent cells, whereas the
gene encoding sphingomyelin phospodiesterases (Smpdl3b)is
down-regulated. Accumulation of ceramide triggers apopto-
sis and is also avoided by high level expression of genes for
ceramidases (Acer2 and 3) that convert it to sphingosine
(SR5). Transcripts for sphingosine receptors (Sr1pr1 and 3)
(SR5) are high in quiescent satellite cells. Signaling through
these receptors is likely to promote survival and activation
when spingosine-1-phosphate becomes available (Nagata et
Resistance to xenobiotics and oxidative stress
Quiescent satellite cells have developed strategies that
confer resistance to xenobiotics, genotoxics, and oxidative
stress. This is illustrated in Fig. 4 (see also Supplementary
Tables 4 and 5). They express high levels of genes such as
Abcb1a, Abca5, and Abcc9 encoding efflux channels of the
multidrug resistance family that pump toxic substances
out of the cell. Transcripts for proteins involved in
the solubilization of toxins by hydroxylation, such as many
cytochrome mono-oxygenases of the p450 family (examples,
Cyp4b1, Cyp26b1) or the Flavin mono-oxygenase (Fmo2), are
also high. This is also the case for other modifying enzymes
that increase solubilization of xenobiotics, such as UDP-
glucuronosyl transferase (Ugt1), glutathione transferase
(Gstm1), or sulfotransferase (Sult1a1) (SR6). Transcriptional
activation of further detoxification enzymes depends on the
gene encoding the Aryl hydrocarbon receptor (Ahr)
expressed at a higher level in quiescent, compared with
activated, satellite cells. Interestingly, in screening for
Pax3/7 target genes in the C2 muscle cell line, Ahr and
Abcb1a were identified (Kumar et al., 2009; McKinnell et al.,
2008). The Ahr receptor translocates to the nucleus on
binding toxic molecules such as dioxin derivatives or
polycyclic aromatic hydrocarbons, where it interacts with
transcription factors of the Aryl hydrocarbon nuclear
translocator (Arnt) family (SR7). Arnt2 transcripts are higher
in activated satellite cells, indicating that this form is more
important for cell activation. The Arnt2 factor also interacts
with hypoxia inducible factor, which is expressed after
muscle injury and during growth, when the vasculature is
damaged or still developing and tissue oxygenation is
compromised. In this context, transcripts for angiotensin II
during in vivo activation. (A, C, E, F) Intensity maps, showing
examples of differences in the level of transcripts for compo-
nents of Wnt (A), Notch (D), Igf (E), and sphingolipid (F) signaling
pathways, presented as in Fig. 1D. (B, D) Levels of transcripts
measured by qPCR, for genes implicated in Wnt (B, Wisp1, Wnt1
inducible signaling pathway protein 1, and Axud1, Axin-1 up-
regulated gene 1), Notch and Igf (D, Hes6, Hairy and enhancer of
split 6 and Igfbp6, Igf binding protein 6) signaling.
Transcriptome differences in signaling pathways
82G. Pallafacchina et al.
type 2 receptor (Agtr2) are markedly up-regulated in in vivo
activated cells (Figs. 4A and B) where it may protect them
from DNA damage and senescence, as shown for vascular
smooth muscle cells (Min et al., 2008).
Genes for enzymes implicated in the response to oxidative
stress (SR8) are also highly expressed in quiescent cells, as
exemplified by Srxn (Sulfiredoxin) (Jonsson and Lowther,
2007) which controls reactive oxygen species-mediated
cytotoxicity, glutathione peroxidase 3 (Gpx3), a secreted
form, which scavenges oxygen radicals as part of the reactive
oxygen species-redox system, or the intracellular thiore-
doxin reductase 1 (Txnrd1). Transcripts for ceruloplasmin
(Cp), which chelates metals such as copper or iron required
for redox reactions, are also high in quiescent satellite cells.
Administration of hydrogen peroxide to satellite cells shows
that adult quiescent cells are better protected from
oxidative stress, as measured by the number of cells that
form colonies (Fig. 4C). The role of Gstm1, Gpx3, and Srxn is
demonstrated by increased sensitivity of these cells on
addition of the reduced glutathione depleting agent,
diethylmaleate (DEM) (Plummer et al., 1981) (Fig. 4C). This
is in keeping with observations on Gpx1 mutants (Lee et al.,
Cell adhesion and the extracellular matrix
Genes for cell–cell adhesion molecules (Figs. 5A and B), such
as claudins (Cldn1,5), intercellular adhesion molecule 2
(Icam2), epithelial V-like antigen 1 (Eva1/Mpzl2), and
endothelial specific adhesion molecule 1 (Esam1), are
notably expressed in Ad satellite cells (Figs. 5A and B).
Esam1 and claudin 5 were thought to be confined to blood
vessels where they mark tight junctions between endothelial
cells (SR9). Their expression, together with CD38 that is the
receptor for Pecam1, on satellite cells may be important in
promoting contiguity with blood vessels (Christov et al.,
2007); see also (Abou-Khalil et al., 2009). Esam1 has also
recently emerged as a marker of adult hematopoietic stem
cells (SR10). Integrins on the cell surface promote adhesion
and signaling through interaction with fibronectin, col-
lagens, and laminins (SR11, SR12). Integrin β1 (Itgb1)
transcripts are high in quiescent cells, as are those for
syndecan 4 (Sdc4), a satellite cell marker which promotes
integrin-mediated cell adhesion and is required for a rapid
satellite cell response to activation (Cornelison et al., 2004).
In contrast, genes for the laminins, such as Lamb1 and the
integrin β3 binding protein (Itgb3 bp) that promotes Integrin
reactions with substrates and integrin signaling (Integrin β3
is also transcribed in satellite cells), are up-regulated in Ad.
mdx and 1wk cells. The Collagen genes, Col1α1 and Col1α2,
are also more highly expressed in these cells, as is the gene
for Syndecan 1 (Sdc1) which enhances integrin α2β1
interactions with collagens, leading to up-regulation of ma-
trix metalloproteinase genes (SR13, Fig. 6).
The genes for N-cadherin (Cdh2) and its intracellular
interactor, α-catenin (Ctnna2) are up-regulated in Ad.mdx
and 1wk satellite cells. N-cadherin in the embryo marks cells
that will enter myogenesis (Cinnamon et al., 2006) and is also
associated with their migration (Brand-Saberi et al., 1996)
(Fig. 5A). Transcripts for the surface glycoprotein CD24 that
prevents proliferation of progenitor cells in self-renewing
ification, protection from xenobiotics, and resistance to oxidative
the BH adjustment in Ad versus Ad.mdx cells. (B) Transcript levels,
measured by qPCR, for a member of the cytochrome P450 family
(Cyp4b1) and for the angiotensin type 2 receptor (Agtr2). Samples
formed after 3 days in culture in cells after treatment with 50 μM
H2O2(red bars) compared to untreated cells (black bars). Right
panel, treatment with the reduced glutathione depleting agent,
diethylmaleate (DEM), increases the sensitivity of adult quiescent
cells to H2O2(green bar), compared to controls: nontreated cells
(black), cells treated with DEM alone (gray), and H2O2alone (red).
minimum of three independent experiments, performed in tripli-
cate. The error bars represent standard errors.
Quiescent satellite cells up-regulate genes for detox-
83 Gene profiling of muscle satellite cells in vivo
tissues (SR14) are higher in Ad cells (Fig. 5A), providing
another potentially interesting surface marker.
Many genes encoding extracellular matrix proteins (Fig. 5A)
Fgl2, encoding fibrinogen-like 2, involved in cell adhesion
(SR15) and not previously associated with muscle, or Smoc2
(Figs. 5A and B), encoding a calcium binding protein which
regulates cell/matrix interactions (SR16). As previously noted,
transcripts for the small secreted proteoglycan, Decorin (Dcn),
2007) are also very high in Ad cells (Supplementary Fig. 2D).
modulating their availability (SR17), such as Versican (Vcan),
expressed in Ad.mdx and 1wk cells.
Expression of genes encoding enzymes involved in
proteoglycan sulfation also undergoes major changes
(Figs. 5B and C). Ad.mdx and 1wk satellite cells express
high levels of heparan sulfate 6-O sulfotransferase (Hs6st1
and 2) transcripts (Figs. 5A and B). Hs6st transfers sulfate
groups to heparan, allowing it to bind signaling molecules,
promoting their activity and presentation to cellular recep-
tors. Genes encoding components of the Fgf signaling
pathway, such as Fgf receptors and Sproutys, implicated in
(A, C, and D) Intensity maps, showing differences in transcript levels of genes for adhesion molecules and extracellular matrix
components (A), proteoglycan modifying enzymes (C), and proteins implicated in cell mobility (D). Samples are as in Fig. 1D. (B)
Transcript levels, measured by qPCR, for cell adhesion (Eva1, Cldn5), matrix component (Smoc2), and heparan sulfotransferase
(Hs6st2) genes. (E) Immunofluorescence staining with Pax7 and doublecortin (Dcx) antibodies of satellite cells, marked by Pax7
(arrowheads), on single fibers from adult EDL muscle immediately after isolation (Ad t0), and after 2 days in culture (t2). A fiber from
an adult mdx EDL muscle at t0 (Ad.mdx t0) shows that satellite cells already express detectable levels of Dcx. Hoechst staining marks
nuclei. Bar=50 μm.
Genes implicated in cell adhesion, extracellular matrix, and cell mobility are modulated during satellite cell activation.
84 G. Pallafacchina et al.
myogenesis (Buckingham and Relaix, 2007; Lagha et al.,
2008), did not show significant changes in expression. We
conclude that in the postnatal in vivo context, Fgf signaling
is mainly regulated by heparan sulfation orchestrated by the
satellite cell. Indeed, sulfatase deficiency negatively affects
muscle regeneration (Langsdorf et al., 2007). The up-
regulation of the chondroitin sulfate synthase 1 gene
(Chsy1) in Ad.mdx and 1wk cells is also indicative of matrix
remodeling (Izumikawa et al., 2008). On the other hand,
transcripts for iduronate-2 sulfatase (Ids), an enzyme
responsible for the degradation of glycosaminoglycan (Moro
et al., 2010), are high in Ad satellite cells (Fig. 5C).
On activation, as the niche environment breaks down,
satellite cells move from the fiber. Transcripts for factors
associated with motility are up-regulated. Examples are
shown in Fig. 5D. Doublecortin (Dcx) is 14- to 16-fold up-
Fig. 5A, not reproduced in culture. Doublecortin is a
microtubule stabilizing protein, indispensable for cortical
migration of neurons (SR18). Its unexpected presence on
5E). Transcripts for a protein very similar to doublecortin,
Dclk1, also show up-regulation in activated satellite cells,
where it may also be involved in migration, as well as mitosis
motility receptor (Hmmr), required for migration of fibro-
blasts in injured skin (SR19) and for thymosin β10 (Tmsb10),
that regulates actin dynamics critical for motility (SR20), is
also high in Ad.mdx and 1wk cells and provides another
example of the potentiation of satellite cell migration.
Proteinases and inhibitors involved in the regulation
of the niche
Strikingly, a number of genes encoding inhibitors of
proteinases, such as Tfpi2 or Serpin1, that prevent extra-
cellular matrix remodeling, are up-regulated in Ad satellite
cells, whereas transcripts for proteinases are up-regulated in
Ad.mdx and 1wk cells (Figs. 6A and C). Proteinases of the
matrix metalloproteinase (MMP) and Adam families and their
inhibitors illustrate this aspect. Transcripts for the former
(ex. Adam12 and 19, Adamts2 and 7, Mmp11) are low in
quiescent cells and increase on activation. This is illustrated
for Adam19 at the protein and transcript level (Figs 6A, B and
C). Exceptions are Mmp2 and Adamts1; however, the former
encodes an enzyme that requires proteolytic cleavage to be
active (SR21) and the latter an enzyme involved in the
release of Tgfβ (SR22), implicated in quiescence. Many genes
for the inhibitors of matrix degrading enzymes (Timp2, 3, 4)
or proteins involved in their association with the extracel-
lular matrix (Efemp1) are up-regulated in Ad cells, also
shown by qPCR for Timp4 and Tfpi2 (Fig. 6C). The
requirement for MMPs in satellite cell activation and muscle
regeneration is substantiated by the result of systemic
delivery of the MMP inhibitor, AM409 (Defamie et al.,
2008), in mdx mice (Figs. 6D–G). We found a 3.5-fold
decrease in the percentage of regenerating fibers in the
diaphragm of AM409 treated mdx mice, together with a 2.5-
fold decrease in the percentage of activated MyoD-positive
cells (Fig. 6F). This suggests that MMPs are involved in the
activation of satellite cells required for regeneration. We
also found that muscle regeneration was delayed in
cardiotoxin-injured tibialis anterior muscle of AM409 treated
wild-type mice (Supplementary Fig. 3). MMPs, implicated in
regeneration (Oh et al., 2004; Li et al., 2009), are also
activated in other cell types at the site of injury, such as
macrophages. In order to show that the effect on regener-
ation also reflects a direct role of MMPs in satellite cell
activation, we treated cultured single fibers from normal
adult muscle for 2 days with 10 μM AM409. This led to a
marked reduction in the expansion of MyoD-positive cells
along the fibers, measured as the number of clustered MyoD-
positive cells that have progressed beyond the first cell
division. Thenumber of clusters with more than two cells was
reduced from 59 to 34% on treatment with AM409. A
significant reduction in the average number of MyoD-positive
cells per fiber was also observed in AM409-treated cultures
(Fig. 6G, left panel). These results suggest a cell autonomous
requirement for MMP activity in satellite cell expansion.
We report here the first transcriptome analysis of quiescent
versus in vivo activated satellite cells. As shown by Ki67
labeling and expression of genes associated with prolifera-
tion, most adult satellite cells are quiescent. Satellite cells
from mdx and growing muscle, which show different degrees
of activation, are subject to particular environmental
influences; however, the comparison of transcripts that
show changes between adult and both adult mdx and 1week
satellite cells selects those involved in quiescence versus
activation. Proliferating cultured satellite cells display
accentuated differences in the expression of many genes, as
already noted (Fukada et al., 2007), but this comparison can
be misleading in terms of the member of a gene family that
changes (e.g., Notch3 instead of Notch1, or Igfbp7 instead of
Igfbp6) or of genes that show no or different changes in
expression on in vivo versus in vitro activation (e.g., Dcx,
involved in satellite cell mobilization, or Gcp3 implicated in
extracellular matrix modifications that affect signaling).
Our analysis reveals expression of genes for novel cell
surface markers (e.g., CD24) and regulatory factors (e.g.,
Prdm16) in quiescent satellite cells. Accumulation of
transcripts for family members of factors associated with
pluripotency suggests that satellite cells, like neural stem
cells (Kim et al., 2009), may be prone to reprogramming.
High levels of transcripts for Sox17, and the closely related
Sox18, in quiescent cells, may also be important for
protection from apoptosis, as shown for hematopoietic
stem cells (Kim et al., 2007). This is a function that Pax7
performs before and at birth (Relaix et al., 2006). However,
Pax7 is not essential later (Lepper et al., 2009), when Sox17/
18 may ensure survival, together with other protective
mechanisms, such as that of sphingolipid signaling.
Protection of stem cells from aggressive agents is critical,
particularly in a tissue like muscle where there is little
turnover and satellite cells remain quiescent for long
periods. The transcriptome analysis shows that adult
satellite cells are well armed in this respect, expressing a
battery of genes that confer resistance to toxic molecules.
Adult satellite cells are primed to respond rapidly to
activation signals, but are held back by expression of
85Gene profiling of muscle satellite cells in vivo
86 G. Pallafacchina et al.
inhibitors. This is illustrated by high levels of transcripts for
the insulin binding protein, Igfbp6, or by Dach1 that can act
as a co-repressor of Six transcription factors.
Another characteristic feature of stem cells in vivo is that
they are sequestered in a particular environment or “niche.”
Our comparison shows down-regulation of genes encoding
adhesion molecules on activation and changes in transcripts
for extracellular matrix components that will also affect the
response to growth factors, such as Fgfs. Notably, proteinase
activity, required for breakdown of the niche and efficient
regeneration, is inhibited in quiescent satellite cells.
Our analysis underlines the importance of in vivo
experiments for understanding the muscle stem cell. We
provide new insight into how these cells orchestrate their
own maintenance and protection as quiescent satellite cells
within the niche where they are poised for activation and
also how they respond in order to leave the niche to become
activated, proliferating myogenic cells. The skeletal muscle
paradigm has its special features—a single quiescent stem
cell within its niche, for example. However the control of
quiescence versus activation is a central issue in stem cell
biology. Characterization of the muscle stem cell in this
respect reveals underlying general principles.
Materials and methods
The Pax3GFP/+mouse line (Montarras et al., 2005) and
Pax3GFP/+:mdx/mdx mice obtained by crossing mdx/mdx
(Bulfield et al., 1984) with Pax3GFP/+mice were used to
prepare satellite cells. All animal experiments were carried
out according to the regulations of the French Ministry of
Agriculture and Fisheries practiced by the Institut Pasteur
Flow cytometry, satellite cell isolation, and
Satellite cells were isolated from a pool of diaphragm,
pectoralis, and abdominal muscles of Pax3GFP/+mice, by flow
cytometry as previously described (Montarras et al., 2005),
window of GFP positivity was subdivided into the same high
and low GFP fractions for all samples (Fig. 1A). Muscle
satellite cells were isolated from three experimental groups:
(1) adult 6-week-old Pax3GFP/+(Ad), (2) adult 6-week-old
Pax3GFP/+:mdx/mdx (Ad.mdx), and (3) postnatal 1-week-old
Pax3GFP/+(1wk) mice. Cells were directly collected in RLT
buffer (Qiagen) for RNA extraction or plated on gelatin-
coated dishes in growth medium (Relaix et al., 2006) for 3
days before RNA extraction, to obtain the fourth experimen-
tal group of ex vivo activated satellite cells (Ad.cult).
For each experimental group three independent samples
were prepared for analysis. For each sample, cells were
isolated from at least six mice (males and females in equal
The percentage of GFP-positive cells in each sample was
100%, as expected since this was the basis on which cells
were purified by flow cytometry. The GFP-positive cells that
we analyzed corresponded to 10–15% of the small, nongran-
As a control of viability and purity, cells from each sample
were plated at low density. All the colonies formed after 3 to
5 days were myogenic on the basis of MyoD, myogenin, or
troponin T expression (data not shown), indicating that we
had isolated pure samples of muscle progenitor cells.
RNA collection and microarray analysis
Total RNA was extracted and purified after DNase treat-
ment (Promega) using the RNAeasy Micro kit (Qiagen). RNA
and cRNA quality was monitored on Agilent RNA Pico
LabChips (Agilent Technology). cRNAs obtained from
100 ng of RNA were amplified by using the GeneChip
Expression Two-Cycle 3′ amplification system (Affymetrix).
Fragmented biotin-labeled cRNA samples were hybridized
on Mouse 430_2.0 GeneChip Genome arrays, according to
the manufacturer's protocol. Three independent isolates
were analyzed for each experimental group. The generation
of cell intensity files and the quality control of hybridiza-
tions were performed with GeneChip Operating Software
is absent in quiescent cells (t0). Hoechst staining marks nuclei. Bar=100 μm. (C) Transcript levels, measured by qPCR, for the proteinase
Adam19, the MMP inhibitor Timp4, and the proteinase inhibitor Tfpi2. (D and E) In vivo treatment of mdx mice with the AM409 MMP
inhibitor (see Materials and Methods for details) affects regeneration as shown by the altered morphology of regenerating fibers in
diaphragm muscle after treatment with the inhibitor for 14 days (AM409), compared to control untreated muscle (veh), labeled with
antibodies to embryonic myosin heavy chain (MyHCemb, red) that marks newly regenerating fibers and to laminin (green) that marks the
fiber outline (D). (E) Sections stained with MyHCemb (red) and MyoD (green) antibodies from diaphragm muscles as in D. Bar in D and
E=50 μm. (F) Quantification of the number of regenerating fibers after AM409 treatment, expressed as the percentage of MyHCemb-
positive fibers in the total number of fibers in the mdx diaphragm muscle (red bars). MyoD quantification is expressed as the number of
percentage of cells in groups of 1–2 cells or of more than 2 cells along the fibers (left panel), as well as the number of satellite cells per
fiber found in AM409-treated compared to nontreated preparations, after 2 days in culture. Three sets of single fiber preparations, 35
fibers, and a total of 300 satellite cells were scored for each sample.
Expression analysis of genes encoding extracellular proteinases and their inhibitors: matrix metalloproteinases (MMP) are
87 Gene profiling of muscle satellite cells in vivo
Microarray data processing
Differentially expressed transcripts between samples were
analyzed using SPlus ArrayAnalyser software (Insightfull
Corporation) and processed by the Robust Multichip Analysis
(RMA) algorithm in order to correct the background, to adjust
intensity summarization into a unique probe set signal. These
normalized data sets are expressed on a logarithmic scale and
(dChip) software (available at http://www.hsph.harvard.
edu/∼cli/complab/dchip/ (Li and Wong, 2003)) for genes
selected from Supplementary Tables 4 and 5, according to our
stringency criteria (see below). Rows in intensity maps
represent genes and columns represent samples (analyzed in
triplicate). Two-way clustering was performed using a
correlation-based metric and the unweighted pair group
method with the arithmetic mean for gene color assignment.
Redundant probe sets were omitted in the analysis.
Differential expression between samples was calculated
using the local pool error test for pairwise sample compar-
isons. To determine statistical significance, the Benjamini-
Hochberg (BH) algorithm for multiple testing was implemen-
ted to adjust the P value. We also used the more stringent
Bonferroni algorithm for P value adjustment, indicated in
As expected, fewer genes were selected by this method. The
threshold P value was set at 0.05 for both BH and Bonferroni
adjustments, which assigns a 95% confidence interval to the
microarray analysis. For the Ad-Ad.cult comparison, only
genes significant after the Bonferroni adjustment were
retained, since the number of genes after the BH method
was very large. Only transcripts showing at least 1.5-fold
the BH stringency criteria and that are in common in the two
in vivo comparisons (Ad-Ad.mdx and Ad-1wk, Supplementary
Tables 1 and 2) are presented in Supplementary Tables 4 and
5. Gene function enrichment using dChip software was
performed on genes form Supplementary Tables 4 and 5 to
obtain Gene Ontology classification of genes associated with
DNA replication presented in Supplementary Table 6.
The complete microarray data set, including the RMA data
used to produce intensity maps, have been uploaded onto
the GEO (Gene Expression Omnibus, http://www.ncbi.nlm.
nih.gov/geo) website under Accession Number GSE15155.
Single fiber preparation and culture
Single fibers were prepared from EDL muscles as described
(Collins et al., 2005) and immediately fixed or transferred
into growth medium (see cell culture section).
accession number, forward and reverse primer sequence from 5’ to 3’
Primer sets used in the qPCR experiments. Table columns from left to right correspond to: gene symbol, transcript
Gene SymbolAcc Num Forward primerReverse primer
purchased from Applied Biosystem TaqMan® Gene Expression Assays
purchased from SuperArray Bioscience Corporation
88G. Pallafacchina et al.
Immunodetection was as described on single fibers (Collins et
al., 2005), on cultured cells, and on sections (Relaix et al.,
2006). Antibodies and procedures are in Supplementary
Quantitative real-time PCR
Total RNA was isolated as described for the microarray
analysis experiments from freshly prepared satellite cells.
The list of primers used for quantitative real-time PCR
(qPCR) is shown in Table 1; procedures are in Supplementary
Lentiviral vector production and satellite cell
The human Dach1 full-length cDNA was cloned into a
lentiviral vector using the Gateway system (Invitrogen) and
the strategy designed by Charneau and colleagues (Arhel et
al., 2007). A lentiviral vector coding for the red fluorescent
protein, RFP, was used to monitor cell transduction. This
type of vector permits efficient transduction of freshly
isolated satellite cells. For detailed procedures see Supple-
Western blot analysis
Protein extracts were obtained from lentivirus transduced
satellite cell cultures, separated on SDS-PAGE gels, and
transferred to nitrocellulose membranes as described (Crist
et al., 2009) and membranes were probed with antibodies
(see Supplementary Information).
In vivo muscle regeneration and matrix
metalloproteinase (MMP) inhibitor treatment
Cardiotoxin injury of tibialis anterior (TA) muscles was as
described (Cooper et al., 2001). In vivo MMP inhibitor
treatment of injured and mdx mice was as in (Defamie et
al., 2008). See Supplementary Information.
Antioxidant capacity of satellite cells
Treatment of freshly isolated satellite cells in culture with
50 μm H202 provoked oxidative stress, estimated as a
percentage of colonies formed compared to untreated
cells. Prior treatment with 50 μm diethylmaleate (DEM)
was used to deplete reduced glutathione; see also Supple-
M.B.'s laboratory is supported by the Pasteur Institute, the
CNRS, and by grants from the AFM and the E.U. through the
MYORES network of excellence (6th F.R.P.) and the
EuroSyStem and Optistem projects (7th F.R.P.). G.P. was
the recipient of an FRM postdoctoral fellowship and later,
was supported by EuroSyStem. The funding sources had no
role in the design or execution of this work. We are grateful
to the Pasteur Institute staff of the DNA microarray platform
and Imagopole and also to Pierre Charneau, Yves Jacob, and
Philippe Souque for help with lentivirus production. We also
thank Catherine Bodin for technical assistance with the
preparation of sections.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.scr.2009.
Abou-Khalil, R., Le Grand, F., Pallafacchina, G., Valable, S.,
Authier, F.J., Rudnicki, M.A., Gherardi, R.K., Germain, S.,
Chretien, F., Lafuste, P., Montarras, D., Chazaud, B., 2009.
Autocrine and paracrine angiopoietin 1/Tie-2 signalling promotes
muscle satellite cell self-renewal. Cell Stem Cell 5, 298–309.
Arhel, N.J., Souquere-Besse, S., Munier, S., Souque, P., Guadagnini,
S., Rutherford, S., Prevost, M.C., Allen, T.D., Charneau, P.,
2007. HIV-1 DNA Flap formation promotes uncoating of the pre-
integration complex at the nuclear pore. EMBO J. 26 (12),
multipotential stem cells that exhibit myogenic, osteogenic, and
adipogenic differentiation. Differentiation 68 (4-5), 245–253.
Bosnakovski, D., Xu, Z., Li, W., Thet, S., Cleaver, O., Perlingeiro,
R.C., Kyba, M., 2008. Prospective isolation of skeletal muscle
stem cells with a Pax7 reporter. Stem Cells 26 (12), 3194–3204.
Brack, A.S., Conboy, I.M., Conboy, M.J., Shen, J., Rando, T.A., 2008.
A temporal switch from notch to Wnt signaling in muscle stem
cells is necessary for normal adult myogenesis. Cell Stem Cell 2
Brand-Saberi, B., Gamel, A.J., Krenn, V., Muller, T.S., Wilting, J.,
Christ, B., 1996. N-cadherin is involved in myoblast migration and
muscle differentiation in the avian limb bud. Dev. Biol. 178 (1),
Buckingham, M., Relaix, F., 2007. The role of Pax genes in the
development of tissues and organs: Pax3 and Pax7 regulate
muscle progenitor cell functions. Annu. Rev. Cell Dev. Biol. 23,
Bulfield, G., Siller, W.G., Wight, P.A., Moore, K.J., 1984. X
chromosome-linked muscular dystrophy (mdx) in the mouse.
Proc. Natl. Acad. Sci. USA 81 (4), 1189–1192.
Cabello-Verrugio, C., Brandan, E., 2007. A novel modulatory
mechanism of transforming growth factor-beta signaling through
decorin and LRP-1. J. Biol. Chem. 282 (26), 18842–18850.
Cerletti, M., Jurga, S., Witczak, C.A., Hirshman, M.F., Shadrach,
J.L., Goodyear, L.J., Wagers, A.J., 2008. Highly efficient,
functional engraftment of skeletal muscle stem cells in dystro-
phic muscles. Cell 134 (1), 37–47.
Christov, C., Chretien, F., Abou-Khalil, R., Bassez, G., Vallet, G.,
Authier, F.J., Bassaglia, Y., Shinin, V., Tajbakhsh, S., Chazaud,
B., Gherardi, R.K., 2007. Muscle satellite cells and endothelial
cells: close neighbors and privileged partners. Mol. Biol. Cell 18
Cinnamon, Y., Ben-Yair, R., Kalcheim, C., 2006. Differential effects
of N-cadherin-mediated adhesion on the development of
myotomal waves. Development 133 (6), 1101–1112.
Collins, C.A., Olsen, I., Zammit, P.S., Heslop, L., Petrie, A.,
Partridge, T.A., Morgan, J.E., 2005. Stem cell function, self-
89 Gene profiling of muscle satellite cells in vivo
renewal, and behavioral heterogeneity of cells from the adult
muscle satellite cell niche. Cell 122 (2), 289–301.
Cooper, R.N., Irintchev, A., Di Santo, J.P., Zweyer, M., Morgan, J.E.,
Partridge, T.A., Butler-Browne, G.S., Mouly, V., Wernig, A.,
2001. A new immunodeficient mouse model for human myoblast
transplantation. Hum. Gene Ther. 12 (7), 823–831.
Cornelison, D.D., Wilcox-Adelman, S.A., Goetinck, P.F., Rauvala,
H., Rapraeger, A.C., Olwin, B.B., 2004. Essential and separable
roles for Syndecan-3 and Syndecan-4 in skeletal muscle devel-
opment and regeneration. Genes Dev. 18 (18), 2231–2236.
Crist, C.G., Montarras, D., Pallafacchina, G., Rocancourt, D.,
Cumano, A., Conway, S.J., Buckingham, M., 2009. Muscle stem
cell behavior is modified by microRNA-27 regulation of Pax3
expression. Proc. Natl. Acad. Sci. USA 106 (32), 13383–13387.
Day, K., Shefer, G., Richardson, J.B., Enikolopov, G., Yablonka-
Reuveni, Z., 2007. Nestin-GFP reporter expression defines the
quiescent state of skeletal muscle satellite cells. Dev. Biol. 304
Defamie, V., Laurens, M., Patrono, D., Devel, L., Brault, A., Saint-
Paul, M.C., Yiotakis, A., Barbry, P., Gugenheim, J., Crenesse, D.,
Dive, V., Huet, P.M., Mari, B., 2008. Matrix metalloproteinase
inhibition protects rat livers from prolonged cold ischemia-warm
reperfusion injury. Hepatology 47 (1), 177–185.
Fukada, S., Uezumi, A., Ikemoto, M., Masuda, S., Segawa, M.,
Tanimura, N., Yamamoto, H., Miyagoe-Suzuki, Y., Takeda, S.,
2007. Molecular signature of quiescent satellite cells in adult
skeletal muscle. Stem Cells 25 (10), 2448–2459.
Good, K.L., Tangye, S.G., 2007. Decreased expression of Kruppel-
like factors in memory B cells induces the rapid response typical
of secondary antibody responses. Proc. Natl. Acad. Sci. USA 104
Ikemoto, M., Fukada, S., Uezumi, A., Masuda, S., Miyoshi, H.,
Yamamoto, H., Wada, M.R., Masubuchi, N., Miyagoe-Suzuki, Y.,
Takeda, S., 2007. Autologous transplantation of SM/C-2.6(+)
satellite cells transduced with micro-dystrophin CS1 cDNA by
lentiviral vector into mdx mice. Mol. Ther. 15 (12), 2178–2185.
Izumikawa, T., Koike, T., Shiozawa, S., Sugahara, K., Tamura, J.,
Kitagawa, H., 2008. Identification of chondroitin sulfate glucur-
onyltransferase as chondroitin synthase-3 involved in chondroitin
polymerization: chondroitin polymerization is achieved by
multiple enzyme complexes consisting of chondroitin synthase
family members. J. Biol. Chem. 283 (17), 11396–11406.
Jonsson, T.J., Lowther, W.T., 2007. The peroxiredoxin repair
proteins. Subcell. Biochem. 44, 115–141.
Kajimura, S., Seale, P., Kubota, K., Lunsford, E., Frangioni, J.V.,
Gygi, S.P., Spiegelman, B.M., 2009. Initiation of myoblast to
brown fat switch by a PRDM16-C/EBP-beta transcriptional
complex. Nature 460 (7259), 1154–1158.
Kim, I., Saunders, T.L., Morrison, S.J., 2007. Sox17 dependence
distinguishes the transcriptional regulation of fetal from adult
hematopoietic stem cells. Cell 130 (3), 470–483.
Kim, J.B., Sebastiano, V., Wu, G., Arauzo-Bravo, M.J., Sasse, P.,
J., Hubner, K., Bernemann, C., Ortmeier, C., Zenke, M.,
pluripotency in adult neural stem cells. Cell 136 (3), 411–419.
Kollias, H.D., McDermott, J.C., 2008. Transforming growth factor-
beta and myostatin signaling in skeletal muscle. J. Appl. Physiol.
104 (3), 579–587.
Kuang, S., Kuroda, K., Le Grand, F., Rudnicki, M.A., 2007.
Asymmetric self-renewal and commitment of satellite stem
cells in muscle. Cell 129 (5), 999–1010.
Kumar, D., Shadrach, J.L., Wagers, A.J., Lassar, A.B., 2009. Id3 is a
direct transcriptional target of Pax7 in quiescent satellite cells.
Mol. Biol. Cell 20, 3170–3177.
Lagha, M., Kormish, J.D., Rocancourt, D., Manceau, M., Epstein,
J.A., Zaret, K.S., Relaix, F., Buckingham, M.E., 2008. Pax3
regulation of FGF signaling affects the progression of embryonic
progenitor cells into the myogenic program. Genes Dev. 22 (13),
Langsdorf, A., Do, A.T., Kusche-Gullberg, M., Emerson Jr., C.P., Ai,
X., 2007. Sulfs are regulators of growth factor signaling for
satellite cell differentiation and muscle regeneration. Dev. Biol.
311 (2), 464–477.
Lee, S., Shin, H.S., Shireman, P.K., Vasilaki, A., Van Remmen, H.,
Csete, M.E., 2006. Glutathione-peroxidase-1 null muscle pro-
genitor cells are globally defective. Free Radic. Biol. Med. 41 (7),
Lepper, C., Conway, S.J., Fan, C.M., 2009. Adult satellite cells and
embryonic muscle progenitors have distinct genetic require-
ments. Nature 460 (7255), 627–631.
L'Honore, A., Coulon, V., Marcil, A., Lebel, M., Lafrance-Vanasse, J.,
Gage, P., Camper, S., Drouin, J., 2007. Sequential expression
and redundancy of Pitx2 and Pitx3 genes during muscle
development. Dev. Biol. 307 (2), 421–433.
Li, C., Wong, W.H., 2003. DNA-Chip Analyzer (dChip). In: Parmigiani,
G., Garrett, E.S., Irizarry, R., Zege, S.L. (Eds.), The Analysis
of Gene Expression Data: Methods and Software. Springer,
New York, pp. 120–141.
Li, H., Mittal, A., Makonchuk, D.Y., Bhatnagar, S., Kumar, A., 2009.
Matrix metalloproteinase-9 inhibition ameliorates pathogenesis
and improves skeletal muscle regeneration in muscular dystro-
phy. Hum. Mol. Genet. 18 (14), 2584–2598.
Li, X., Oghi, K.A., Zhang, J., Krones, A., Bush, K.T., Glass, C.K.,
Nigam, S.K., Aggarwal, A.K., Maas, R., Rose, D.W., Rosenfeld,
M.G., 2003. Eya protein phosphatase activity regulates Six1-
Dach-Eya transcriptional effects in mammalian organogenesis.
Nature 426 (6964), 247–254.
Mankoo, B.S., Collins, N.S., Ashby, P., Grigorieva, E., Pevny, L.H.,
Candia, A., Wright, C.V., Rigby, P.W., Pachnis, V., 1999. Mox2 is
a component of the genetic hierarchy controlling limb muscle
development. Nature 400 (6739), 69–73.
McKinnell, I.W., Ishibashi, J., Le Grand, F., Punch, V.G., Addicks,
G.C., Greenblatt, J.F., Dilworth, F.J., Rudnicki, M.A., 2008.
Pax7 activates myogenic genes by recruitment of a histone
methyltransferase complex. Nat. Cell Biol. 10 (1), 77–84.
Min, L.J., Mogi, M., Iwanami, J., Li, J.M., Sakata, A., Fujita, T.,
Tsukuda, K., Iwai, M., Horiuchi, M., 2008. Angiotensin II type 2
receptor deletion enhances vascular senescence by methyl
methanesulfonate sensitive 2 inhibition. Hypertension 51 (5),
Montarras, D.,Buckingham, M., 2008. Isolation, characterisation and
origin of muscle satellite cells. In: Tsuschida, K., Takeda, S.
(Eds.), Recent Advances in Skeletal Muscle Differentiation,
Research Signpost Series. Trivandrum, Kerala, India, pp. 25–39.
Montarras, D., Morgan, J., Collins, C., Relaix, F., Zaffran, S.,
Cumano, A., Partridge, T., Buckingham, M., 2005. Direct
isolation of satellite cells for skeletal muscle regeneration.
Science 309 (5743), 2064–2067.
Moro, E., Tomanin, R., Friso, A., Modena, N., Tiso, N., Scarpa, M.,
in zebrafish early development. Matrix Biol. 29 (1), 43–50.
Nagata, Y., Partridge, T.A., Matsuda, R., Zammit, P.S., 2006. Entry
of muscle satellite cells into the cell cycle requires sphingolipid
signaling. J. Cell Biol. 174 (2), 245–253.
Oh, J., Takahashi, R., Adachi, E., Kondo, S., Kuratomi, S., Noma, A.,
Alexander, D.B., Motoda, H., Okada, A., Seiki, M., Itoh, T.,
Itohara, S., Takahashi, C., Noda, M., 2004. Mutations in two
matrix metalloproteinase genes, MMP-2 and MT1-MMP, are
synthetic lethal in mice. Oncogene 23 (29), 5041–5048.
Tachibana, T., Imbalzano, A.N., 2007. Myogenin and the SWI/SNF
ATPase Brg1 maintain myogenic gene expression at different
stages of skeletal myogenesis. J. Biol. Chem. 282 (9), 6564–6570.
Otto, A., Schmidt, C., Luke, G., Allen, S., Valasek, P., Muntoni, F.,
Lawrence-Watt, D., Patel, K., 2008. Canonical Wnt signalling
90G. Pallafacchina et al.
induces satellite-cell proliferation during adult skeletal muscle Download full-text
regeneration. J. Cell Sci. 121 (Pt 17), 2939–2950.
Plummer, J.L., Smith, B.R., Sies, H., Bend, J.R., 1981. Chemical
depletion of glutathione in vivo. Methods Enzymol. 77, 50–59.
G., Khanna, S., 2004. Temporal gene expression profiling of
dystrophin-deficient (mdx) mouse diaphragm identifies conserved
and muscle group-specific mechanisms in the pathogenesis of
muscular dystrophy. Hum. Mol. Genet. 13 (3), 257–269.
D., Tajbakhsh, S., Mansouri, A., Cumano, A., Buckingham, M.,
2006. Pax3 and Pax7 have distinct and overlapping functions in
adult muscle progenitor cells. J. Cell Biol. 172 (1), 91–102.
Seale, P., Ishibashi, J., Holterman, C., Rudnicki, M.A., 2004. Muscle
satellite cell-specific genes identified by genetic profiling of
MyoD-deficient myogenic cell. Dev. Biol. 275 (2), 287–300.
Seale, P., Sabourin, L.A., Girgis-Gabardo, A., Mansouri, A., Gruss,
P., Rudnicki, M.A., 2000. Pax7 is required for the specification of
myogenic satellite cells. Cell 102 (6), 777–786.
Shefer, G., Wleklinski-Lee, M., Yablonka-Reuveni, Z., 2004. Skeletal
muscle satellite cells can spontaneously enter an alternative
mesenchymal pathway. J. Cell Sci. 117 (Pt 22), 5393–5404.
Takahashi, K., Yamanaka, S., 2006. Induction of pluripotent stem
cells from mouse embryonic and adult fibroblast cultures by
defined factors. Cell 126 (4), 663–676.
Tanaka, K.K., Hall, J.K., Troy, A.A., Cornelison, D.D., Majka, S.M.,
Olwin, B.B., 2009. Syndecan-4-expressing muscle progenitor
cells in the SP engraft as satellite cells during muscle
regeneration. Cell Stem Cell 4 (3), 217–225.
van Lunteren, E., Leahy, P., 2007. Gene expression microarrays and
respiratory muscles. Respir. Physiol. Neurobiol. 156 (2), 103–115.
Wu, K., Li, A., Rao, M., Liu, M., Dailey, V., Yang, Y., Di Vizio, D.,
Wang, C., Lisanti, M.P., Sauter, G., Russell, R.G., Cvekl, A.,
Pestell, R.G., 2006. DACH1 is a cell fate determination factor
that inhibits cyclin D1 and breast tumor growth. Mol. Cell. Biol.
26 (19), 7116–7129.
Zhao, P., Hoffman, E.P., 2004. Embryonic myogenesis pathways in
muscle regeneration. Dev. Dyn. 229 (2), 380–392.
91 Gene profiling of muscle satellite cells in vivo