Differentiation 84 (2012) 322–329

Contents lists available at SciVerse ScienceDirect

Differentiation

0301-46

Join the

http://d

n Corr

Genetic

France.

E-m

journal homepage: www.elsevier.com/locate/diff

In vitro characterization of proliferation and differentiationof pig satellite cells

Marie-Hel�ene Perruchot a,b,n, Patrick Ecolan a,b, Inge Lise Sorensen c, Niels Oksbjerg c, Louis Lefaucheur a,b

a INRA, UMR1348 Physiology, Environnement and Genetics for the Animal and Livestock Systems (PEGASE), F-35590 Saint-Gilles, Franceb AgroCampus-Ouest, UMR1348 Physiology, Environnement and Genetics for the Animal and Livestock Systems (PEGASE), 65 rue de St Brieuc, F-35000 Rennes, Francec Department of Food Science, Faculty of Science and Technology, Aarhus University, DK-8830 Tjele, Denmark

a r t i c l e i n f o

Article history:

Received 16 February 2012

Received in revised form

31 May 2012

Accepted 9 August 2012Available online 27 September 2012

Keywords:

Satellite cell culture

Myosin heavy chain

Fiber type

Pig

81/$ - see front matter & 2012 International

International Society for Differentiation (ww

x.doi.org/10.1016/j.diff.2012.08.001

esponding author at: INRA, UMR1348 Phy

s for the Animal and Livestock Systems (PE

Tel.: þ33 2 23 48 70 54; fax: þ33 2 23 48 50

ail address: marie-helene.perruchot@rennes.i

a b s t r a c t

Skeletal muscle contains various muscle fiber types exhibiting different contractile properties based on

the myosin heavy chain (MyHC) isoform profile. Muscle fiber type composition is highly variable and

influences growth performance and meat quality, but underlying mechanisms regulating fiber type

composition remain poorly understood. The aim of the present work was to develop a model based on

muscle satellite cell culture to further investigate the regulation of adult MyHC isoforms expression in

pig skeletal muscle. Satellite cells were harvested from the mostly fast-twitch glycolytic longissimus

(LM) and predominantly slow-twitch oxidative rhomboideus (RM) muscles of 6-week-old piglets.

Satellite cells were allowed to proliferate up to 80% confluence, reached after 7 day of proliferation (D7),

and then induced to differentiate. Kinetics of proliferation and differentiation were similar between

muscles and more than 95% of the cells were myogenic (desmin positive) at D7 with a fusion index

reaching 6579% after 4 day of differentiation. One-dimensional SDS polyacrylamide gel electrophor-

esis revealed that satellite cells from both muscles only expressed the embryonic and fetal MyHC

isoforms in culture, without any of the adult MyHC isoforms that were expressed in vivo. Interestingly,

triiodothyronine (T3) induced de novo expression of adult fast and a-cardiac MyHC in vitro making our

culture system a valuable tool to study de novo expression of adult MyHC isoforms and its regulation by

intrinsic and/or extrinsic factors.

& 2012 International Society of Differentiation. Published by Elsevier B.V. All rights reserved.

1. Introduction

Skeletal muscle is a heterogeneous tissue mostly composed ofmyofibers exhibiting different contractile and metabolic properties.In mature skeletal muscles, the myofiber contractile type is gen-erally based on adult myosin heavy chain (MyHC) isoforms that arepresent within each fiber. During myogenesis and postnatal musclegrowth, pig skeletal muscles exhibit eight skeletal MyHC isoforms,i.e. four adult (I, IIa, IIx and IIb), three developmental (embryonic,fetal and a-cardiac) and one extraocular isoform (Lefaucheuret al., 1997, 1998; Davoli et al., 2002). Fiber type composition ofmature skeletal muscles is highly variable depending on muscletype, species, genotype and breeding conditions (Lefaucheur, 2010),and the mechanisms underlying the regulation of the differentMyHC isoform expression are complex and involve several factorssuch as innervation, physical activity, hormones, growth factors, and

Society of Differentiation. Publish

w.isdifferentiation.org)

siology, Environnement and

GASE), F-35590 Saint-Gilles,

80.

nra.fr (M.-H. Perruchot).

substrate availability (Schiaffino and Reggiani, 1996; Pette andStaron, 2001). In addition, as presented below, available data alsosuggest that fiber type could be dependent on the intrinsic proper-ties of satellite cells which proliferate and fuse with existingmyofibers during growth. These cells are located between the basallamina and sarcolemma of each myofiber, and based on in vitro

studies, the proportion of myotubes expressing the adult slow type IMyHC has been reported to be greater in satellite cell culturesderived from slow muscles, whereas this isoform was absent incultures from fast muscles in rat (Cantini et al., 1993; Dusterhoftand Pette, 1993; Huang et al., 2006), mouse (Rosenblatt et al., 1996;Bryla and Karasinski, 2001), rabbit (Barjot et al., 1995), cat (Kanget al., 2010) and chicken (Stockdale and Miller, 1987; Feldman andStockdale, 1991). However, other studies did not find any intrinsicdifference between satellite cells originating from different muscles(Whalen et al., 1990; Dusterhoft et al., 1990; Mouly et al., 1993;Edom et al., 1994; Wehrle et al., 1994; Bourke et al., 1995; Midrioet al., 1998; Bonavaud et al., 2001; LaFramboise et al., 2003; Kanget al., 2010). Therefore, the mechanisms involved in fiber typedetermination are still not fully understood and deservefurther study.

ed by Elsevier B.V. All rights reserved.

M.-H. Perruchot et al. / Differentiation 84 (2012) 322–329 323

Because fiber type composition has been reported to influencegrowth performance and meat quality traits, the control of adultMyHC expression is a major challenge in meat producing animals.To better understand the underlying mechanisms leading to fibertype diversity, a satellite cell culture system expressing no adultMyHC isoforms in basal conditions would be a valuable tool tostudy the regulation of de novo expression of the adult MyHC, inresponse to different factors. To best of our knowledge, such aculture system is not available because different adult MyHCisoforms are usually expressed in vitro in addition to the devel-opmental embryonic (MyHCemb) and fetal (MyHCfetal) isoforms,making it difficult the identification of factors that regulate theexpression of adult MyHC. Therefore, the aim of the present studywas to find an in vitro culture system where adult MyHC wouldnot be expressed in basal conditions, and could be induced bya physiological factor. Among all hormones, thyroid hormonesappear to have the greatest effects on muscle fiber contractilephenotype (D’Albis and Butler-Browne, 1993; Pette and Staron,2001). In particular, triiodothyronine (T3) has been shown tostimulate the disappearance of embryonic and fetal MyHC infavor of adult fast isoforms during the early postnatal period, andactivate the expression of adult fast isoforms in adult muscles.Thus, the present study used T3 as the physiological factor toinfluence MyHC expression in our satellite cell culture system.

D11D0 D7D2 D5 D9 D10D8

2. Materials and methods

2.1. Animals and muscle sampling

The experiment was conducted on five 6-week-old Pietrain X(Large White X Landrace) crossbred female piglets (Sus scrofa)from the INRA experimental herd (Saint-Gilles, France). Allanimals were free of the mutated RYR-1 halothane gene. Animalswere raised and slaughtered in accordance with the Frenchlegislation on animal experimentation and ethics, and in com-pliance with the current national French regulations applied incommercial slaughterhouses. Just after slaughter by electricalstunning and exsanguination, muscle samples were taken fromtwo distinct skeletal muscles, i.e. the longissimus (LM), a predo-minantly fast-twitch glycolytic muscle involved in voluntarymovements of the back, and the tubular portion of rhomboideus

(RM), a mostly slow-twitch oxidative postural muscle involved insupport of the head. For histological analyses, a muscle sample of0.5�0.5�1 cm parallel to the myofiber direction was taken inthe middle of each muscle, fixed with pins on flat sticks, frozen in2-methylbutane (isopentane) cooled by liquid nitrogen, andstored at �70 1C. A contiguous muscle sample was removedunder aseptic conditions for isolation of satellite cells from bothmuscles. Briefly, 10 g of 1-2 mm thick slices of fresh LM and RMwere excised and placed on ice in 20 ml Dulbecco’s PhosphateBuffer Saline (DPBS, Dutscher Brumath, France) containing 1%glucose (Sigma, Saint-Quentin Fallavier, France).

ProliferationDMEM

10%FCS10%HS

DifferentiationDMEM5%FCSInsulin1µM

TransitionDMEM

10%FCS

CA 1µM

Fig. 1. Experimental protocol for satellite cell cultures. Cells are left to proliferate

from D0 to 80% confluence (D7 in present case). Then, cells are placed in a

transition medium for approximately 16 h, and finally switched to a differentia-

tion medium for 4 days (D8–D11). DMEM¼Dulbecco’s Modified Eagle’s Medium,

FCS¼Fetal Calf Serum, HS¼Horse Serum, CA¼Cytosine Arabinoside.

2.2. Myofiber histological typing

Transverse 10 mm thick serial cross sections were cut on acryostat (2800 frigocut N, Reichert-Jung, Heidelberg, Germany) at�20 1C, and processed for the conventional acto-myosin-ATPasestaining after preincubation at pH 4.35 to identify types I, IIAand IIB fibers (Brooke and Kaiser, 1970). A serial section wasreacted for succino-dehydrogenase (SDH) to determine the myo-fiber metabolic type defined as oxidative (R, red) or glycolytic(W, white) (Nachlas et al., 1957).

2.3. Satellite cell isolation

Satellite cells were isolated from LM and RM as previouslydescribed (Theil et al., 2006; Mau et al., 2008). Briefly, musclesamples were trimmed of visible connective tissue, finelychopped with scissors on a Petri dish placed on ice, and digestedfor 3�20 min under shaking in a water-bath at 37 1C in a Ca2þ-free medium containing 0.25% trypsin (Invitrogen, Cergy Pontoise,France), 1.5 mg/ml type II Collagenase (PAA, Les Mureaux, France)and 0.1% DNAse (Sigma, Saint-Quentin Fallavier, France). Cellsuspensions from the three successive digestions were pooledand centrifuged at 800 g for 10 min at 4 1C. The pellets weresuspended in ice-cold PGM (Proliferation Growth Medium) con-taining Dulbecco’s modified Eagle’s medium (DMEM) supplemen-ted with 10% fetal calf serum (FCS), 10% horse serum, 3 mg/mlamphotericin B and 20 mg/ml gentamycin, and successively fil-tered through 200 and 50 mm nylon membranes (Nytex, DutscherBrumath, France). Then, cells were centrifuged on top of a 20%Percoll discontinuous gradient at 15,000 g for 8 min at 4 1C toenrich the population in satellite cells (Mau et al., 2008). Thelower part in the centrifugation tube was collected in ice-coldPGM, centrifuged (800 g, 10 min, 4 1C) and resuspended in ice-cold PGM. The number of satellite cells was counted using aMalassez hemocytometer. The use of a Percoll gradient centrifu-gation yielded clean refringent mononucleated cells free fromdebris and myofibrillar fragments. The number of satellite cellsisolated per gram of fresh muscle reached 1.570.2�106 cells inboth LM and RM muscles.

2.4. Satellite cell culture conditions

Cells were seeded at a density of 7�104 cells/cm2 in six-wellplates (9.6 cm2/well) coated with Matrigel (1/50 v/v, BD, Le Pontde Claix, France) and grown at 37 1C in PGM containing antibioticsunder 95% air and 5% CO2. Cells were allowed to proliferate up to80% confluence in PGM (up to D7 in present case), then placed in atransition medium containing 10% FCS for approximately 16 h,and finally switched to a differentiation medium (DMEM contain-ing 5% FCS, 1 mM porcine insulin and 1 mM cytosine arabinosideto fully stop cell proliferation (Sigma, Saint-Quentin Fallavier,France)) up to D11 (Fig. 1).

Proliferation was assessed by determining cell number andviability at each step of proliferation (D2–D7) with a viabilityanalyzer using the trypan-blue exclusion test (Vi-CellTM XR,Beckman Coulter, Paris, France). Immunocytochemical detectionof desmin and MyHC was made during proliferation at D2, D5, D7and differentiation at D9, D10 and D11. Cells were fixed inmethanol at �20 1C for 10 min and rinsed with PBS. To reduce

M.-H. Perruchot et al. / Differentiation 84 (2012) 322–329324

nonspecific binding, cells were incubated in a blocking solutioncontaining 2% Bovine Serum Albumin (Sigma, Saint-QuentinFallavier, France). Anti-desmin (1:100, D1033, Sigma) and anti-MyHC (MF-20, 1:10, Developmental Studies Hybridoma Bank(University of Iowa)) mouse monoclonal antibodies were usedas primary antibodies. MF-20 has been reported to react with allskeletal MyHC isoforms, including embryonic and fetal MyHC(LaFramboise et al., 2003). Cells were incubated with primarymonoclonal antibodies in PBS/0.1% BSA at room temperature for2 h, and the primary antibodies visualized using a fluoresceine-conjugated goat anti-mouse IgG (Jackson, Interchim Montluc-on,France). Cells were mounted in fluorescent mounting mediumcontaining 4,6-diami-dino-2-phenylindole (DAPI) (Dako, Trappes,France) to stain nuclei in blue. A Nikon DS-Ri1 epifluorescencemicroscope (X10 magnification lens) was used to acquire digitalimages using an Eclipse E400 digital camera and NIS–Elementssoftware version 3.0 (Nikon). The fusion percentage was esti-mated from 3 randomly chosen fields containing a total ofapproximately 900 nuclei per well, 3 wells per animal, andexpressed as the number of DAPI stained nuclei inside myotubesdivided by the total number of nuclei in the same field.

For the study of T3 influence, cells were grown in basalconditions as described above (C, control), in a T3-depletedmedium (–T3), and the T3-depleted medium supplemented withincreasing amount of T3 to final concentrations of 1, 10, and100 nM 3,5,30–L-T3 sodium salt from Sigma (T6397). For informa-tion, plasma level of total T3 in 8 week-old piglets is around 3 nM(Herpin and Lefaucheur, 1992), but the actual concentration inthe direct vicinity of satellite cells in vivo could be different fromthe plasma level due to local deiodination of thyroxine (T4) to T3.Serum was T3-depleted using BioRad AG 1�8 resin according toSamuels et al., 1979. Total T3 concentrations in culture mediawere assessed using a radioimmunoassay kit (M.P. Biomedicals,Illkirch, France). The detection limit of the assay was 0.07 nM, andthe intra- and inter-assay CV were 5.6 and 7.5%, respectively. T3concentrations in the control proliferation and differentiationmedia were 0.40 and 0.14 nM, respectively, whereas it was belowthe detection limit of the assay in the T3 depleted proliferationand differentiation media.

2.5. MyHC electrophoresis

Laemmli 1X (Sigma, S-3401) was added to 20 mm thickcryostat muscle sections (10 ml/mg fresh muscle) or satellite cellcultures following two washes with PBS. Cell cultures werescrapped using Laemmli 1X (50–300 ml per well). Proteins wereextracted by heating the mixture at 100 1C for 6 min. Proteinconcentration was determined using Quick Start Bradford DyeReagent 1X (BioRad, Marnes-la-Coquette, France), adjusted to0.5 mg/ml in 1X Laemmli and stored at �70 1C until electrophor-esis. MyHC were separated using a standard one dimensionalsodium dodecyl sulfate-polyacrylamide gel electrophoresis (stan-dard SDS-PAGE) as described by Talmadge and Roy (1993) afterslight modifications (Lefaucheur et al., 2001). The electrophoresiswas carried out in the Bio-Rad-Mini-Protean II Dual Slab Cellelectrophoretic system (Bio-Rad, Richmond, CA, USA) by loading5–10 mg extracted proteins per well. Migration was performed at5 1C for 30 h at 72 V. Then, gels were either silver (Blum et al.,1987) or blue silver (Candiano et al., 2004) stained. Bandscorresponding to the different MyHC isoforms were identified asMyHCemb, MyHCfetal, adult fast MyHC (IIaþ IIxþ IIb) and adultslow MyHC type I and a-cardiac MyHC as previously described bycomparing their mobility with that of bands from different pigskeletal and cardiac muscles sampled during the fetal and post-natal periods (Lefaucheur et al., 2001). To further study theinfluence of T3 treatment on expression of the a-cardiac MyHC

in satellite cell cultures, additional gradient SDS-PAGE wasperformed as previously described (Lefaucheur et al., 2001). Gelswere scanned with ImageQuant LAS 4000 (GE Healthcare, Amer-sham Biosciences, Sweden) and bands were quantified using theImageQuantTL program. The relative proportions of the differentMyHC isoforms were expressed as percentage of the sum of allMyHC within each lane.

2.6. Statistical analysis

Data were analyzed by a two-way analysis of variance usingthe GLM procedure of SAS (SAS Inst. Inc., Cary, NC). The modelincluded the main effects of muscle type, day of cell culture, andtheir interaction, as well as the animal within muscle effect. Theeffect of muscle type was tested against the animal within muscleresidual error. When the muscle type�day of cell culture inter-action was significant, muscle type effects were analyzed withineach day of cell culture with a model including the main effects ofmuscle type and animal. Differences were considered significantat Po0.05. Data are presented as means7SE.

3. Results

3.1. Fiber type composition of selected muscles

Using conventional acto-myosin-ATPase staining after prein-cubation at pH 4.35 (Fig. 2a and c), the percentages of type I, IIAand IIB fibers were approximately 10%, 15% and 75% in LM, and70%, 15% and 15% in RM, respectively. Based on SDH staining(Fig. 2 band d), all types I and IIA fibers were highly oxidative inboth muscles, whereas type IIB fibers exhibited a heterogeneousstaining (R or W). Indeed, 80% of IIB fibers exhibited no SDHstaining (IIBW, fast glycolytic fibers) in LM, whereas nearly all IIBfibers showed a strong SDH staining (IIBR, fast oxidative fibers)in RM. Thus, at 6 weeks of age, LM and RM muscles can beconsidered as mostly fast-twitch glycolytic and slow-twitchoxidative, respectively.

3.2. Proliferation and differentiation of satellite cells in culture

Cell counting after 2, 5 and 7 day of culture (D2, D5, D7)indicated similar proliferation rates between muscles (Fig. 3b),and 80% confluence was achieved after 7 day of culture in bothmuscles. Immunostaining for desmin, a muscle specific protein,was used to evaluate the purity of satellite cell preparations in LMand RM derived primary cultures (Fig. 3a). The percentage ofdesmin positive cells was 92.472.0% in LM and 95.470.6% in RMat D5, and 96.671.7% and 97.670.9% at D7, showing that mostcells were myogenic cells, without any significant difference inthe proportion of myogenic cells between muscles (Fig. 3c).

The presence of MyHC in cell cultures was assessed byimmunohistochemistry during proliferation and differentiation(Fig. 4a) using MF20, a monoclonal antibody reacting with allskeletal MyHC isoforms. No labeling was observed at D2. At D5,only mononucleated cells were present among which approxi-mately 20% reacted with MF20 in both muscles. At D7 (80%confluence), about 10% of mononucleated cells reacted with thisantibody, and few myotubes were present and strongly reactedwith MF20. During differentiation, the number of mononucleatedcells stained with MF20 decreased, whereas the number and sizeof myotubes dramatically increased up to D9, and then moreslowly up to D11, they all strongly reacted with MF20. Thecombined staining of nuclei with DAPI and MyHC with fluores-ceine was used to measure the fusion index (Fig. 4b). This indexincreased during differentiation and reached 65.378.8% after

W

RI

IIAIIB

Fig. 2. Histochemical identification of fiber types on serial transverse frozen sections of longissimus (a and b) and rhomboideus (c and d) muscles of 6 week-old pigs. (a),

(c) Acto-myosin ATPase after preincubation at pH 4.35 to identify types I, IIA and IIB fibers. (c), (d) Succino-dehydrogenase staining to identify red oxidative (R) and white

glycolytic (W) fibers. Scale bar¼100 mm.

M.-H. Perruchot et al. / Differentiation 84 (2012) 322–329 325

4 day of differentiation (D11), without any significant differencebetween LM and RM. Thus, kinetics of satellite cell proliferationand differentiation did not differ between LM and RM in our basalcell culture conditions.

3.3. Myosin heavy chain analysis

The standard SDS-PAGE separation of MyHC in pig LM andRM muscles and corresponding satellite cell cultures is shownin Fig. 5. A clear separation of the embryonic (band a) andfetal (band b) MyHC isoforms from the adult fast (band c¼ IIaþIIxþ IIb) and slow (band d¼ I) MyHC isoforms was achieved aspreviously described (Lefaucheur et al., 2001). Band c was a wideband composed of a mixture of the adult IIa, IIx and IIb MyHCisoforms which could not be clearly separated from each other.Embryonic and fetal MyHC were not detected in vivo in LM andRM of the 6 week-old piglets. Thus, in vivo LM and RM containedon average 85% adult fast and 65% adult slow MyHC at this stage,respectively. In satellite cell cultures, MyHC was absent at D2 butcould be observed as soon as at D5 in both muscles, in accordancewith the immunostaining of cultures with the MF20 antibody(Fig. 4a). The standard SDS-PAGE showed that only the embryonicand fetal MyHC isoforms were present in culture, whereas noadult fast or slow MyHC could be detected, whatever the muscleorigin of the cells. This shows that the staining with the MF-20antibody in the immunocytochemical study of satellite cellcultures corresponded to the embryonic and fetal MyHC isoforms.Surprisingly, the embryonic/fetal MyHC ratio was highly variable(2.370.9) and did not significantly change from D5 to D11 ineither muscle (data not shown).

3.4. Influence of T3 on myosin heavy chain expression

T3-depleted medium was suitable for porcine satellite cellssurviving, proliferation and differentiation as no obvious difference

was observed between the control and T3-depleted media for thekinetic of proliferation, differentiation and basal expression ofMyHC (Fig. 6). Therefore, the T3-depleted medium was next usedas the reference for the study of T3 influence. T3 treatments did notvisibly modify the kinetics of satellite cell proliferation anddifferentiation (data not shown). In contrast, the 10 and 100 nMT3 doses induced de novo expression of two bands correspondingto adult fast MyHC (Fig. 6, band c). A faint d band was inducedby 100 nM T3 (white arrowhead) at the level of the a-cardiacMyHC known to be expressed in the atrium (Fig. 6, lane A), justabove the slow type I MyHC (Fig. 6, lane V) as previously describedby Lefaucheur et al. (2001). In order to confirm this de novo

expression of the a-cardiac MyHC, a complementary study usinggradient SDS-PAGE allowing a clear separation of a-cardiac fromthe slow type I MyHC was performed as previously reported(Lefaucheur et al., 2001).A band migrating at the level of thestrongest band present in the atrium (Fig. 7, lane A) was de novo

induced by the 10 and 100 nM T3 doses (arrowhead), whereas noadditional band could be detected at the level of the slow type IMyHC. Therefore, present data strongly suggest that the additionalfaint band induced by T3 in standard SDS-PAGE correspond to thea-cardiac MyHC isoform.

4. Discussion

We present for the first time a system of satellite cell culturethat only expresses the embryonic and fetal MyHC isoforms at theprotein level in basal conditions, without any expression of theadult fast or slow isoforms that are present in vivo in the LM andRM muscles at 6 weeks of age. This stage was chosen becausemost post-natal maturation of contractile and metabolic proper-ties of skeletal muscle fibers is established by this stage (Suzukiand Cassens, 1980; Lefaucheur and Vigneron, 1986), and satellitecells are still abundant in relation to a still very high relative

LM RM

D2

D5

D7

10

15

85

90

95

100

0

5

D2 D5 D7

Sate

llite

cel

l cou

ntin

g (X

106 )

70

75

80

D5 D7

Des

min

pos

itive

cel

ls (%

)

Fig. 3. Satellite cells from longissimus (LM) and rhomboideus (RM) muscles of 6 week-old pigs after 2 (D2), 5 (D5) and 7 (D7) days of proliferation. (a) Cell nuclei were

stained in blue by 4,6-diami-dino-2-phenylindole (DAPI), and desmin in green using a mouse monoclonal anti-desmin antibody visualized by a fluoresceine-conjugated

goat anti-mouse IgG (Scale bar¼10 mm). (b) Growth curve of longissimus (empty rings) and rhomboideus (filled rings) satellite cell cultures from D2 to D7. Results are

means7SE (n¼6 animals). No significant difference between muscles was observed. (c) Percentage of desmin positive cells after 5 (D5) and 7 (D7) days of proliferation in

satellite cell cultures of pig LM (open bars) and RM (solid bars) muscles. Results are means7SE (n¼6 animals).

M.-H. Perruchot et al. / Differentiation 84 (2012) 322–329326

muscle growth. Conventional histochemical fiber typing andstandard SDS-PAGE showed that LM and RM were mostly fast-twitch glycolytic and slow-twitch oxidative at 6 weeks of age,respectively, as observed in older pigs (Lefaucheur et al., 2002).

In both muscles, the pattern of satellite cell proliferation anddifferentiation were very similar to those previously reported insemimembranosus muscle from 6 week-old and newborn pigletsusing the same methodology (Theil et al., 2006; Mau et al., 2008).The yield of satellite cell isolation (1.570.2�106 cells/g muscletissue) was lower than the 4.1�106 value reported by Mau et al.(2008) in newborn piglets, in accordance with data reporting adecrease in the abundance of satellite cells with increasing age(Campion et al., 1981). Surprisingly, we found no difference in theyields of satellite cells isolated from LM and RM, in contrast toother findings reporting higher yields in slow than fast muscles inrats (Gibson and Schultz, 1983; Dusterhoft et al., 1990; Le Moigneet al., 1990; Lagord et al., 1998) and rabbits (Barjot et al., 1995).The discrepancy may be due to the much younger physiologicalage of our animals compared with that of laboratory animalsusually close to the adult stage. More than 90% of the cellscontained desmin at D5 and D7 indicating that myogenic cells

predominated in our cultures. Moreover, the proportions ofdesmin-positive cells were similar in LM and RM derived cultures,making our conditions suitable to compare in vitro proliferationand differentiation between LM and RM.

In vitro data in rats and rabbits usually report that cells fromslow muscles proliferate more actively than those isolated fromfast muscles (Le Moigne et al., 1990; Barjot et al., 1995; Lagordet al., 1998; Martelly et al., 2000; Ono et al., 2010), and fuse intomyotubes either more (Le Moigne et al., 1990; Barjot et al., 1995)or less efficiently (Lagord et al., 1998; Ono et al., 2010). In ourbasal conditions, kinetics of in vitro proliferation and differentia-tion were similar in LM and RM, and the difference with studiescarried out in laboratory animals could still be related to theyounger physiological age of our animals.

It is generally stated that MyHC are synthesized upon terminaldifferentiation and constitute a key marker of the terminal stageof myogenesis. However, the present study clearly shows thatMyHC was already present in some mononucleated cells duringthe proliferation phase, in accordance with data recently obtainedin rat (Balan et al., 2009). The pattern of in vitro MyHC expressiondid not differ between LM and RM during the proliferation and

70

80

90

100

LM

RM

30

40

50

60

D9 D10 D11

Fusi

on (%

)

LM

D2

RM

D5

D7

D9

D10

D11

Fig. 4. Myosin heavy chain (MyHC) expression during cell culture. (a) Immunolabelling in satellite cell cultures during proliferation at D2, D5 and D7 and

differentiation at D9, D10 and D11. Cells were isolated from longissimus (LM) and rhomboideus (RM) muscles. Nuclei were stained in blue by 4,6-diami-dino-2-

phenylindole (DAPI), and MyHC in green using a mouse monoclonal anti-MyHC antibody (MF-20) visualized by a fluoresceine-conjugated goat anti-mouse IgG. Arrows

indicate mononucleated cells. Scale bar¼10 mm. (b) Changes in the fusion index during satellite cell differentiation at D9, D10 and D11 in pig LM and RM muscles.

Fusion index is expressed as the percentage of total nuclei that are located in myotubes (at least 3 nuclei) stained in green using a mouse monoclonal anti-MyHC

antibody (MF-20) visualized by a fluoresceine-conjugated goat anti-mouse IgG. Results are means7SE (n¼6 animals). No significant difference between LM and RM

were observed.

M.-H. Perruchot et al. / Differentiation 84 (2012) 322–329 327

differentiation phases. Interestingly, only the embryonic and fetalMyHC were present in satellite cell cultures in basal conditions,without any expression of the adult fast or slow MyHC isoformsthat were currently present in vivo in LM and RM muscles. Thisindicates that in vitro and in vivo MyHC profiles are not correlated,and that our satellite cell culture system exhibits an immaturephenotype with the exclusive expression of the embryonic and fetalMyHC, whatever the muscle origin. A prominent expression of theembryonic and fetal MyHC isoforms has also been previouslyreported in satellite cell cultures from different adult rat and mouseskeletal muscles (Matsuda et al., 1984; Dusterhoft et al., 1990;

Wehrle et al., 1994; Rosenblatt et al., 1996), but our culture systemis the first one to be devoid of adult MyHC, which makes it avaluable tool to study the de novo expression of the adult MyHCisoforms and their regulation in response to different intrinsicand/or extrinsic factors.

Whether the different in vivo expression of MyHC between LMand RM is dependant on intrinsic properties of satellite cells and/or extrinsic factors remains to be clarified. As presented in theintroduction, numerous studies support the existence of differentintrinsic properties of satellite cells between fiber types, butseveral data presented below suggest that these differences could

LM45 dpc S A C -T3 1 10 100

α-cardiac

I

Fig. 7. Gradient SDS-PAGE separation of myosin heavy chains (MyHC) in fetal

longissimus (LM 45 dpc), adult semispinalis (S) and atrium (A), and LM satellite cell

cultures after 4 days of differentiation in media containing increasing amount of

T3 (C, control medium; –T3, T3-depleted medium; 1, 10 and 100, T3-depleted

medium supplemented with 1, 10 and 100 nM T3, respectively). 5 mg of proteins

were loaded per lane and gels were silver stained. A band migrating at the level of

the strongest band observed in the atrium (lane A) was induced by the 10 and

100 nM T3 doses (arrowhead), whereas no additional band was induced by T3 at

the level of the slow type I MyHC (arrow).

D2 D5 D7 D9 D10 D11

Proliferation Differentiation

In vivo

LM

Stdabc

d

RM

abc

d

Fig. 5. Standard SDS-PAGE separation of myosin heavy chains (MyHC) in pig

longissimus (LM) and rhomboideus (RM) muscles from a 6 week-old pig (in vivo)

and corresponding satellite cell cultures during proliferation (D2, D5 and D7) and

differentiation (D9, D10 and D11). Standard (Std) is a mixture of 6 week-old pig

LM and RM with 70 dpc LM to localize the developmental embryonic and fetal

MyHC isoforms as previously described (Lefaucheur et al., 2001). Bands a and b

represent embryonic and fetal MyHC, respectively, they are clearly separated from

adult fast type II (band c¼IIaþIIxþ IIb) and slow type I (band d) MyHC isoforms.

5 mg of proteins were loaded per lane and gels were silver stained.

a b

C -T3 1 10 100 std LM+RM V Aab

d

cα-cardiac

d

c

90

50607080

C

10203040

-T3T3 1nMT3 10nMT3 100nM

MyH

C p

ropo

rtion

(%)

0Emb Fetal Fast α-cardiac

Fig. 6. Standard SDS-PAGE separation of myosin heavy chains (MyHC) in long-

issimus satellite cell cultures after 4 days of differentiation in media containing

increasing amount of T3 (C, control medium; –T3, T3-depleted medium; 1, 10 and

100, T3-depleted medium supplemented with 1, 10 and 100 nM T3, respectively)

(a). Standard (Std) is similar to std in Fig. 5. Lanes 7, 8 and 9 correspond to a

mixture of 6 week-old pig longissimus and rhomboideus (LMþRM), ventricle (V)

and atrium (A), respectively. Identification of bands a–d are as described in Fig. 5.

The faint d band induced by 100 nM T3 (white arrow) was at the level of the

a-cardiac MyHC known to be expressed in the atrium (A), and shown to migrate

just above the slow type I MyHC (Lefaucheur et al., 2001). 10 mg of proteins were

loaded per lane and gels were silver blue stained. (b) Influence of T3 on MyHC

proportions in longissimus satellite cell cultures after 4 days of differentiation. Data

are means7SE expressed as the percentage of total MyHC within each lane. The

experiment was performed twice using two independent cultures and MyHC

proportions were obtained from three replicates within each culture.

M.-H. Perruchot et al. / Differentiation 84 (2012) 322–329328

only be observed under the application of permissive extrinsicfactors. Thus, in vitro chronic electrical stimulation of myotubecultures has been shown to induce slow MyHC I expression toa much greater extend in slow- than fast-muscle derived cul-tures, thus inducing differences between cultures that were notobserved in basal conditions (Wehrle et al., 1994). Amongpotential effectors, calcium ions (Ca2þ), the second messengerin excitation-contraction coupling, likely play a key role inmodulating MyHC expression. Indeed, addition of a Ca2þ iono-phore to rabbit hind limb primary satellite cell cultures has beenshown to reversibly shift the initially fast MyHC profile to aslower more oxidative type (Kubis et al., 1997). Specific factors

originating from myofibers or surrounding tissues could alsoinduce differences in satellite cell properties. Indeed, supplemen-tation of myoblast cultures from 7-day-old chick embryos orC2C12 cells with a slow or fast muscle extract has been reportedto increase the expression of slow or fast MyHC, respectively(Matsuoka and Inoue, 2008). Similarly, a specific expression of theslow MyHC in cultures derived from slow chick fetal musclescould only be observed in the presence of the neural tube(DiMario and Stockdale, 1997). Therefore, expression of adultMyHC is likely dependant on factors which can be both intrinsicand extrinsic to the satellite cells, and our culture system, with noexpression of adult MyHC in basal conditions, is potentially apowerful tool to identify these factors.

To validate our culture system, we have challenged cells withT3 known to induce a shift to a faster phenotype in adult skeletalmuscle, the expression of the a-cardiac MyHC in the ventricularmyocardium (Lompre et al., 1984) and to stimulate the disap-pearance of embryonic and fetal MyHC in favor of adult fastisoforms during the early postnatal period (D’Albis and Butler-Browne, 1993). The a-cardiac MyHC has also been shown to beexpressed in limb skeletal muscle after chronic low frequencystimulation in rabbit (Peuker et al., 1998), and in vivo in earlypostnatal skeletal muscle in pig (Lefaucheur et al., 1997). More-over, the expression of the a-cardiac MyHC has been reported tobe increased in pig skeletal muscle after early postnatal coldexposure, likely in relation with the increased plasma level of T3induced by cold exposure (Lefaucheur et al., 2001). In accordancewith these data reported in vivo, present results provide evidencefor de novo expression of adult fast and a-cardiac MyHC inresponse to T3 in our satellite cell culture system. Therefore,our in vitro culture system can respond to extrinsic factors in aphysiological way, and the absence of expression of adult MyHCin basal condition make our system a powerful tool to study theregulation adult MyHC expression in the pig.

In conclusion, we have characterized a system of pig satellitecell culture that exhibits an immature MyHC profile, i.e. anexpression of only the embryonic and fetal MyHC isoforms inbasal conditions, without any expression of the adult fast or slowisoforms, and whatever the muscle type of origin. The de novo

induction of the adult fast and a-cardiac MyHC expression inresponse to T3 makes our culture system a physiologicallyrelevant tool to study the regulation of adult MyHC expression,and thus identify intrinsic and/or extrinsic factors leading to fibertype diversity in pig skeletal muscle.

Acknowledgments

This work was financially supported by INRA. We acknowledgeall the staff involved in animal care and slaughtering.

M.-H. Perruchot et al. / Differentiation 84 (2012) 322–329 329

References

Balan, O.V., Miuge, N.S., Ozerniuk, N.D., 2009. Analysis of expression of heavymyosin chains during in vitro differentiation of satellite cells and myoblastsderived from rat skeletal muscles. Biology Bulletin 36, 213–219.

Barjot, C., Cotten, M.L., Goblet, C., Whalen, R.G., Bacou, F., 1995. Expression ofmyosin heavy chain and of myogenic regulatory factor genes in fast or slowrabbit muscle satellite cell cultures. Journal of Muscle Research and CellMotility 16, 619–628.

Blum, H., Beier, H., Gross, H.J., 1987. Improved silver staining of plant proteins,RNA and DNA in polyacrylamide gels. Electrophoresis 8, 93–99.

Bonavaud, S., Agbulut, O., Nizard, R., D’Honneur, G., Mouly, V., Butler-Browne, G.,2001. A discrepancy resolved: human satellite cells are not preprogrammed tofast and slow lineages. Neuromuscular Disorders 11, 747–752.

Bourke, D.L., Wylie, S.R., Theon, A., Bandman, E., 1995. Myosin heavy chainexpression following myoblast transfer into regenerating chicken muscle.Basic and Applied Myology 5, 43–56.

Brooke, M.H., Kaiser, K.K., 1970. Muscle fiber types: how many and what kind?Archives of Neurology 23, 369–379.

Bryla, K., Karasinski, J., 2001. Diversity of myosin heavy chain expression insatellite cells from mouse soleus and EDL muscles. Folia Histochemica etCytobiologica 39, 295–300.

Campion, D.R., Richardson, R.L., Reagan, J.O., Kraeling, R.R., 1981. Changes in thesatellite cell population during postnatal growth of pig skeletal muscle.Journal of Animal Science 52, 1014–1018.

Cantini, M., Fiorini, E., Catani, C., Carraro, U., 1993. Differential expression of adulttype MHC in satellite cell cultures from regenerating fast and slow rat muscles.Cell biology International 17, 979–983.

D’Albis, A., Butler-Browne, G.S., 1993. The hormonal control of myosin isoformexpression in skeletal muscle of mammals: a review. Basic and AppliedMyology 3, 7–16.

Davoli, R., Fontanesi, L., Zambonelli, P., Bigi, D., Gellin, J., Yerle, M., Milc, J., Braglia,S., Cenci, V., Cagnazzo, M., Russo, V., 2002. Isolation of porcine expressedsequence tags for the construction of a first genomic transcript map of theskeletal muscle in pig. Animal Genetics 33, 3–18.

Candiano, G., Bruschi, M., Musante, L., Santucci, L., Ghiggeri, G.M., Carnemolla, B.,Orecchia, P., Zardi, L., Righetti, P.G., 2004. Blue silver: a very sensitive colloidalCoomassie G-250 staining for proteome analysis. Electrophoresis 25,1327–1333.

DiMario, J.X., Stockdale, F.E., 1997. Both myoblast lineage and innervationdetermine fiber type and are required for expression of the slow myosinheavy chain 2 gene. Developmental Biology 188, 167–180.

Dusterhoft, S., Yablonka-Reuveni, Z., Pette, D., 1990. Characterization of myosinisoforms in satellite cell cultures from adult rat diaphragm, soleus and tibialisanterior muscles. Differentiation 45, 185–191.

Dusterhoft, S., Pette, D., 1993. Satellite cells from slow rat muscle express slowmyosin under appropriate culture conditions. Differentiation 53, 25–33.

Edom, F., Mouly, V., Barbet, J.P., Fiszman, M.Y., Butler-Browne, G.S., 1994. Clones ofhuman satellite cells can express in vitro both fast and slow myosin heavychains. Developmental Biology 164, 219–229.

Feldman, J.L., Stockdale, F.E., 1991. Skeletal muscle satellite cell diversity: satellitecells form fibers of different types in cell culture. Developmental Biology 143,320–334.

Gibson, M.C., Schultz, E., 1983. Age-related differences in absolute numbers ofskeletal muscle satellite cells. Muscle & Nerve 6, 574–580.

Herpin, P., Lefaucheur, L., 1992. Adaptive changes in oxidative metabolism inskeletal muscle of cold-acclimated piglets. Journal of Thermal Biology 17,277–285.

Huang, Y.C., Dennis, R.G., Baar, K., 2006. Cultured slow vs. fast skeletal muscle cellsdiffer in physiology and responsiveness to stimulation. American Journal ofPhysiology Cell Physiology 291, 11–17.

Kang, L.H.D., Rughani, A., Walker, M.L., Bestak, R., Hoh, J.F.Y., 2010. Expression ofmasticatory-specific isoforms of myosin heavy-chain, myosin-binding Protein-C and tropomyosin in muscle fibers and satellite cell cultures of cat mastica-tory muscle. Journal of Histochemistry and Cytochemistry 58, 623–634.

Kubis, H.P., Haller, E.A., Wetzel, P., Gros, G., 1997. Adult fast myosin pattern andCa2þ-induced slow myosin pattern in primary skeletal muscle culture.Proceedings of the National Academy of Sciences of the United States ofAmerica 94, 4205–4210.

LaFramboise, W.A., Guthrie, R.D., Scalise, D., Elborne, V., Bombach, K.L., Armanious,C.S., Magovern, J.A., 2003. Effect of muscle origin and phenotype on satellitecell muscle-specific gene expression. Journal of Molecular and CellularCardiology 35, 1307–1318.

Lagord, C., Soulet, L., Bonavaud, S., Bassaglia, Y., Rey, C., Barlovatz-Meimon, G.,Gautron, J., Martelly, I., 1998. Differential myogenicity of satellite cells isolatedfrom extensor digitorum longus (EDL) and soleus rat muscles revealed in vitro.Cell and Tissue Research 291, 455–468.

Lefaucheur, L., Vigneron, P., 1986. Post-natal changes in some histochemical andenzymatic characteristics of three pig muscles. Meat Science 16, 199–219.

Lefaucheur, L., Hoffman, R., Okamura, C., Gerrard, D., Leger, J.J., Rubinstein, N.,Kelly, A., 1997. Transitory expression of alpha cardiac myosin heavy chain in asubpopulation of secondary generation muscle fibers in the pig. Develop-mental Dynamics 210, 106–116.

Lefaucheur, L., Hoffman, R.K., Gerrard, D.E., Okamura, C.S., Rubinstein, N., Kelly, A.,1998. Evidence for three adult fast myosin heavy chain isoforms in type IIskeletal muscle fibers in pigs. Journal of Animal Science 76, 1584–1593.

Lefaucheur, L., Ecolan, P., Lossec, G., Gabillard, J.C., Butler-Browne, G.S., Herpin, P.,2001. Influence of early postnatal cold exposure on myofiber maturation in pigskeletal muscle. Journal of Muscle Research and Cell Motility 22, 439–452.

Lefaucheur, L., Ecolan, P., Plantard, L., Gueguen, N., 2002. New insights into musclefiber types in the pig. Journal of Histochemistry and Cytochemistry 50,719–730.

Lefaucheur, L., 2010. A second look into fibre typing - Relation to meat quality.Meat Science 84, 257–270.

Le Moigne, A., Martelly, I., Barlovatz-Meimon, G., Franquinet, R., Aamiri, A., Frisdal,E., Bassaglia, Y., Moraczewski, G., Gautron, J., 1990. Characterization ofmyogenesis from adult satellite cells cultured in vitro. International Journalof Developmental Biology 34, 171–180.

Lompre, A.M., Nadal-Ginard, B., Mahdavi, V., 1984. Expression of the cardiac alpha-and beta-myosin heavy chain genes is developmentally and hormonallyregulated. Journal of Biological Chemistry 259, 6437–6446.

Martelly, I., Soulet, L., Bonnavaud, S., Cebrian, J., Gautron, J., Barritault, D., 2000.Differential expression of FGF receptors and of myogenic regulatory factors inprimary cultures of satellite cells originating from fast (EDL) and slow (Soleus)twitch rat muscles. Cellular and Molecular Biology 46, 1239–1248.

Matsuda, R., Spector, D., Stohman, R.C., 1984. Denerveted skeletal muscle displaysdiscoordinate regulation for the synthesis of several myofibrillar proteins.Proceedings of the National Academy of Sciences of the United States ofAmerica 81, 1122–1125.

Matsuoka, Y., Inoue, A., 2008. Controlled differentiation of myoblast cells into fastand slow muscle fibers. Cell and Tissue Research 332, 123–132.

Mau, M., Oksbjerg, N., Rehfeldt, C., 2008. Establishment and conditions for growthand differentiation of a myoblast cell line derived from the semimembranosus

muscle of newborn piglets. In vitro Cellular and Developmental BiologyAnimal 44, 1–5.

Midrio, M., Danieli-Betto, D., Esposito, A., Megighian, A., Carraro, U., Catani, C.,Rossini, K., 1998. Lack of type 1 and type 2 A myosin heavy chain isoforms inrat slow muscle regenerating during chronic nerve block. Muscle & Nerve 21,226–232.

Mouly, V., Edom, F., Barbet, J.P., Butler-Browne, G.S., 1993. Plasticity of humansatellite cells. Neuromuscular Disorders 3, 371–377.

Nachlas, M.M., Tsou, K.C., De Souza, E., Cheng, C.S., Seligman, A.M., 1957.Cytochemical demonstration of succinic dehydrogenase by the use of a newp-nitrophenyl substituted ditetrazole. The Journal of Histochemistry andCytochemistry 5, 420–436.

Ono, Y., Boldrin, L., Knopp, P., Morgan, E.J., Zammit, P.S., 2010. Muscle satellite cellsare functionally heterogeneous population in both somite-derived and bran-chiomeric muscles. Developmental Biology 337, 29–41.

Pette, D., Staron, R.S., 2001. Transitions of muscle fiber phenotypic profiles.Histochemistry and Cell Biology 115, 359–372.

Peuker, H., Conjard, A., Pette, D., 1998. Alpha-cardiac-like myosin heavy chain asan intermediate between MHCIIa and MHCI beta in transforming rabbitmuscle. American Journal of Physiology—Cell Physiology 43, C595–C602.

Rosenblatt, J.D., Parry, D.J., Partridge, T.A., 1996. Phenotype of adult mouse musclemyoblasts reflects their fiber type of origin. Differentiation 60, 39–45.

Samuels, H.H., Stanley, F., Casanova, J., 1979. Depletion of L-3,5,30-triiodothyronineand L-thyroxine in euthyroid calf serum for use in cell culture studies of theaction of thyroid hormone. Endocrinology 105, 80–85.

Schiaffino, S., Reggiani, C., 1996. Molecular diversity of myofibrillar proteins: generegulation and functional significance. Physiological Reviews 76, 371–423.

Stockdale, F.E., Miller, J.B., 1987. The cellular basis of myosin heavy chain isoformexpression during development of avian skeletal muscles. DevelopmentalBiology 123, 1–9.

Suzuki, A., Cassens, R.G., 1980. A histochemical study of myofiber types in muscleof the growing pig. Journal of Animal Science 51, 1449–1461.

Talmadge, R.J., Roy, R.R., 1993. Electrophoretic separation of rat skeletal musclemyosin heavy chain isoforms. Journal of Applied Physiology 75, 2337–2340.

Theil, P.K., Sorensen, I.L., Nissen, P.M., Oksbjerg, N., 2006. Temporal expression ofgrowth factor genes of primary porcine satellite cells during myogenesis.Animal Science Journal 77, 330–337.

Wehrle, U., Dusterhoft, S., Pette, D., 1994. Effects of chronic electrical stimulationon myosin heavy chain expression in satellite cell cultures derived from ratmuscles of different fiber-type composition. Differentiation 58, 37–46.

Whalen, R.G., Harris, J.B., Butler-Browne, G.S., Sesodia, S., 1990. Expression ofmyosin isoforms during notexin-induced regeneration of rat soleus muscles.Developmental Biology 141, 24–40.