DEVELOPMENTAL

BIOLOGY

143,320-334

(1991)

Skeletal Muscle Satellite Cell Diversity: Satellite Cells Form Fibers of Different Types in Cell Culture JEFFREY L. FELDMANANDFRANK

Accepted

October

E. STOCKDALE'

~24~19.90

Following skeletal muscle injury, new fibers form from resident satellite cells which reestablish the fiber composition of the original muscle. We have used a cell culture system to analyze satellite cells isolated from adult chicken and quail pectoralis major (PM; a fast muscle) and anterior latissimus dorsi (ALD; a slow muscle) to determine if satellite cells isolated from fast or slow muscles produce one or several types of fibers when they form new fibers in vitro in the absence of innervation or a specific extracellular milieu. The types of fibers formed in satellite cell cultures were determined using immunoblotting and immunocytochemistry with monoclonal antibodies specific for avian fast and slow myosin heavy chain (MHC) isoforms. We found that satellite cells were of different types and that fast and slow muscles differed in the percentage of each type they contained. Primary satellite cells isolated from the PM formed only fast fibers, while up to 25% of those isolated from ALD formed fibers that were both fast and slow (fast/slow fibers), the remainder being fast only. Fast/slow fibers formed from chicken satellite cells expressed slow MHCl, while slow MHCZ predominated in fast/slow fibers formed from quail satellite cells. Prolonged primary culture did not alter the relative proportions of fast to fast/slow fibers in high density cultures of either chicken or quail satellite cells. No change in commitment was observed in fibers formed from chicken satellite cell progeny repeatedly subcultured at high density, while fibers formed from subcultured quail satellite cell progeny demonstrated increasing commitment to fast/slow fiber type formation. Quail satellite cells cloned from high density cultures formed colonies that demonstrated a similar change in commitment from fast to fast/slow, as did serially subcloned individual satellite cell progeny, indicating that the observed change from fast to fast/slow differentiation resulted from intrinsic changes within a satellite cell. Thus satellite cells freshly isolated from adult chicken and quail are committed to form fibers of at least two types, satellite cells of these two types are found in different proportions in fast and slow muscles, and repeated cell proliferation of quail satellite cell progeny may alter satellite cell progeny to increasingly form fibers of a single type. 0 1991 Academic Press. Inc.

lite cell (Mauro, 1961; White et al., 1975; Miller and Stockdale, 1987). Satellite cells normally are quiescent cells that reside between the basal lamina and sarcolemma of the muscle fiber (for a review of satellite cells see Campion (1984)), becoming developmentally active, perhaps in response to mitogens released when fibers are damaged (Bischoff, 1986; DiMario and Strohman, 1988; Bischoff, 1990). During avian muscle repair, satellite myoblasts proliferate, fuse, and differentiate, with or without innervation, into muscle fibers of specific types (Bandman et al, 1989). It has not been previously demonstrated whether satellite cells are a uniform or a subdivided population of myoblasts, which differ in their commitment to the types of muscle fibers they form. Factors both intrinsic and extrinsic to developing and maturing muscle fibers are responsible for initiating and sustaining expressed differences among muscle fiber types. It has been proposed that intrinsic factors establish a state of myoblast commitment in the early embryo that serves to define the developmental fates of myoblasts as indicated by the type(s) of myosin heavy

INTRODUCTION

Skeletal muscles differ from each other on the bases of contraction rates, expression of different isoforms of muscle-specific contractile proteins, metabolism, and fiber type composition (Buller et al., 1960; Salmons and Vrbova, 1969; Arndt and Pepe, 1975). Each of these properties is established during development (Crow and Stockdale, 1986a) and persists when maturity is reached. When muscle fibers are damaged, they regenerate and the properties of the original fibers are reestablished (Cerny and Bandman, 1987; Saad et al., 1987; Abe et ah, 198’7; d’Albis et ah, 1988; Bandman et al., 1989). However, the cellular basis for reestablishing fiber types in damaged muscle of the adult is poorly understood. While it is recognized that formation and growth of developing avian muscle require at least two types of myogenic precursor cells, embryonic (early) and fetal (late) myoblasts, muscle repair and regeneration require at least one additional type of myoblast, the satel’ To whom

0012-1606191 Copyright All rights

correspondence

should

$3.00

6 1991 by Academic Press, Inc. of reproduction in any form reserved.

be addressed.

320

chain (MHC)’ expressed when they form a muscle fiber, and that MHC expression can then be modulated within limits by innervation, contractile activity, hormones, and perhaps other factors (Buller et al., 1960; Salmons and Vrbova, 1969; Gambke et al., 1983; Silberstein et al., 1986; Silberstein and Blau, 1986; Miller et cd., 1985; Miller and Stockdale, 1986a,b; Cerny and Bandman, 1986; Schafer et ah, 1987; Stockdale et cd., 1989). It has not been established that similar cellular processes are operative in the satellite cells that form regenerative muscle fibers in adult vertebrates. It is known from in lli~o studies, however, that the fiber types found in a muscle following regeneration are the same as those of the original muscle (Cerny and Bandman, 1987; Abe et al., 1987; d’Albis et al., 1988; Saad et al., 1986; 1987). This observation suggests that either differences in the external milieu or intrinsic differences among satellite cells within a given muscle, or a combination of the two, could be operative in the generation of diversity in newly formed fibers of regenerating muscle. To determine if regeneration of muscles of different fiber type may in part be rooted in diversity among the cells committed to be satellite cells, we examined the types of muscle fibers formed in vitro from satellite cells isolated from adult muscles of diverse types. The isolation and culture of satellite cells allowed us to assay the intrinsic property of satellite cells to form fibers of specific types in the absence of environmental cues that may modulate fiber type in z~iz~o.Our observations indicate that satellite cells isolated from muscles of adult animals are of differing types, unequally distributed among muscles composed of differing fiber type. MATERIALS

Sutellite

AND

METHODS

Cell Culture

Satellite cell myoblasts were isolated from the fasttwitch Pectoralis (PM) and slow-tonic Anterior latissimus dorsi (ALD) muscles of 2.5 to 6-month-old adult White Leghorn chickens or Japanese quail. Isolation of the satellite cells from the dissected muscles was based on the procedures of Yablonka-Reuveni et al. (1987), with modifications. The birds were sacrificed by CO, asphyxiation and muscles dissected aseptically and placed in sterile culture dishes with Hank’s balanced salt solution (HBSS). All solutions used during the prep-

’ Abbreviations used: ALD, anterior latissimus dorsi; Ara-C, cytosine arabinoside; DAB, diaminobenzidine; EDL, extensor digitorum longus; HBSS, Hank’s balanced salt solution; HRP, horseradish peroxidase; MHC, myosin heavy chain; MLC, myosin light chain; PM, pectoralis muscle; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; SMHCl, slow myosin heavy chain 1; SMHCB, slow myosin heavy chain 2.

aration procedure included 1% penicillin, streptomycin, and Fungizone (GIBCO). Tissue was digested with 2 vol of fresh 0.2% collagenase (Worthington) in HBSS for 45 min at 37”C, with gentle agitation. The muscle digest was triturated with a lo-ml serological pipet, washed with HBSS, centrifuged, and the supernatant discarded. The pellet was digested with 2 vol of 0.1% trypsin (Difco) for 45 min at 37°C with gentle agitation to release the satellite cells from the basement membranes. Digestion was halted by addition of an equal volume of HBSS, 10% horse serum followed by centrifugation. The pellet was resuspended in 5 ml of HBSS with serum, triturated several times first with 5-ml serological pipets followed by Pasteur pipets, and passed through a 20-pm Nitex filter to remove undigested tissue masses and muscle fiber fragments. The cells were collected by centrifugation, resuspended in 2 ml of culture medium (defined below), and counted. Typical yields were -1.5 x lo6 cells/g of ALD and -3 X lo4 cells/g of PM. Cells were plated at 3-7 cells/cm2 for clonal culture and lo”lo4 cells/cm’ for high density cultures. Cells were plated on either collagen- or diluted Matrigel (Collaborative Research)-coated dishes (Matrigel was diluted 10X with MEM (Allen and Boxhorn, 1987)) for mass cultures or on collagen-coated dishes for clonal density cultures. The media used to maintain long-term cultures of satellite cells was 83% MEM, 10% horse serum (previously screened to support clonal culture), 5% chicken embryo extract, glutamine, penicillin, streptomycin, Fungizone, while that used for passaged cultures consisted of 77% F-10, 15% horse serum, 5% chicken embryo extract, 0.132 mg/ml CaCl,, glutamine, penicillin, streptomycin, Fungizone and was used at 0.1 ml/cm2 (used for the passaged cultures). Media was replaced every other day, except in the clonal cultures which were fed every 3 days. Satellite cell cultures at high density were divided into several groups. One group was maintained on the same culture dishes throughout the course of the experiment. Another group was periodically lifted off the culture dish with 0.02% trypsin, passed through a 20-pm Nitex filter to remove fibers, counted, and replated as mononucleated cells on new dishes at both clonal and mass densities (see above). Some muscle colonies on clonal dishes were subcloned by the methods of Rutz and Hauschka (1982). Primary muscle colonies from 8-day cultures were isolated using 3- or 5-mm (inner diameter) glass cloning rings, washed twice with HBSS, and lifted from the dish with 0.02% trypsin. The satellite cells were then plated onto collagen-coated 35-mm dishes for 1 week and then expanded to three or four 60-mm dishes. Satellite cell cultures at high density were set up for 4-week-long experiments. Dishes that were to be sub-

322

DEVELOPMENTALBIOLOGY

cultured were passaged at S-day intervals. Dishes to be used for fiber type assays (either by immunostaining or myosin extraction for immunoblotting) were harvested at either 8- or g-day intervals. Typically, four or five dishes would be harvested from each group of mass culture dishes; two would be fixed and two or three extracted. Twelve to 15 dishes were set out at clonal density with each passage, and all of these would be fixed. Serial Subcloning of Satellite Myoblasts

Satellite cells were isolated and plated at clonal density on five 60-mm dishes, as previously described. After 8 days in culture, the dishes were rinsed with HBSS and 10 colonies each from PM and ALD cultures (two from each of five dishes) were isolated with cloning rings. The dish was refilled with medium (to prevent any trypsin that might leak out under the cloning ring from disturbing unselected colonies) and the cloning rings were filled with 0.02% trypsin to lift the colonies. The cells from each colony that was lifted were plated on three 60-mm dishes and allowed to incubate for 8 days. The cloning rings were removed from the original dishes and the dishes were washed, fixed, and stained with either F59, S58, or S46, followed by a horseradish peroxidase (HRP) label (see below). The fibers within the remaining colonies were scored for reactivity with the antibodies. Two or three colonies were isolated and lifted from the clonal dishes at the end of the g-day incubation, and these subclones were in turn plated on two or three 60-mm dishes, while the colonies that were not isolated were fixed and stained as before. The process was repeated through four subclonings. Immunohistochemistry

and Immunocytochemistry

PM and ALD were excised from adult chicken and rapidly frozen in melting isopentane followed by immersion in liquid nitrogen. Ten-micrometer cross sections of the frozen muscle were cut serially with an A/O Histostat and mounted on slides coated with a gelatin substrate. Sections were stained with either F59, S58, or S46 antibodies followed by a 500~ dilution of HRPtagged secondary antibody (electrophoresis grade; BioRad) and a 100x dilution of peroxidase-anti-peroxidase complex (PAP; Boehringer-Mannheim). Labeled myofibers were visualized with diaminobenzidine (DAB; Sigma Chemical Co.; 20 mg in 40 ml of 50 mM Tris-HCl, pH 7.6; 20 /.J of 30% H,O,). The MHC content of individual fibers in culture was determined as previously described (Miller et al., 1985; Miller and Stockdale, 1986a,b). Cultures were washed with HBSS, fixed with 100% ethanol, washed with phosphate-buffered saline (PBS), blocked for 30 min with 2% horse serum, 2% BSA in PBS, and incubated in

V0~~~~143,1993

a 10X dilution of hybridoma supernatant for 1 hr at room temperature. Clonal cultures were incubated in the HRP-tagged secondary antibody and PAP described above and visualized with DAB. The dishes were counterstained with 0.1% methylene blue in 25% ethanol. Myogenic colonies were identified by the presence of fibers and scored according to their reactivity with the primary antibody. Mass cultures were double immunofluorescently stained. The first antibody was always the S58 hybridoma supernatant. This was visualized with 0.01 mg/ml rhodamine (TRITC)-labeled rabbit antimouse IgA (a-chain-specific; EY Laboratories), followed by either F59 or S46 hybridoma supernatants, visualized with 0.01 mg/ml fluorescein (FITC)-labeled rabbit anti-mouse IgG (y-chain-specific; Zymed Laboratories). After several washes with PBS, the cells were mounted with 2.5% 1,4-diazabicyclo[2,2,2]octane (to reduce the rate of fluorescence quenching) in 90% glycerol, 10% PBS under a 22 x 40-mm coverslip. Fibers were observed with a Zeiss Photomicroscope fluorescence microscope in rhodamine and fluorescein channels and in phase-contrast, and were scored as positive for either antibody alone, both together, or neither. Random fields were scored under phase-contrast and fluorescence. All fibers that intersected the ocular cross hairs as the dish was moved in a straight line across its entire width were scored. This randomized sampling minimized the effects of branching and swirling of fibers because both positive and negative fibers have an equal chance of being encountered.

SDS-Gel Electrophoresis and Immunoblotting

Myosin was extracted from satellite cell cultures as previously described (Miller et al., 1985; Miller and Stockdale, 1986a,b). Protein concentrations of the extracts were determined by dye binding assay (Bradford, 1976) or optical density. SDS-PAGE samples were made using a 3~ sample buffer (Langer et al., (1982) modified from Laemmli (1970). These samples were run on 5% SDS-PAGE (Rushbrook and Stracher, 1979) mini-gels and stained with Coomassie blue to determine the amount of protein to load that would yield myosin bands of the same staining intensity. Samples equalized for myosin heavy chain were then electrophoresed on triplicate 5% SDS-PAGE mini-gels and transferred to nitrocellulose paper by the method of Towbin et al. (1979). The blots were immunostained with 10x dilutions of either F59, S58, or S46 hybridoma supernatants, followed by a 103X dilution of HRP-tagged secondary antibody (electrophoresis grade; Bio-Rad); and a 200X dilution of PAP complex, visualized with DAB alone or in the presence of 0.0075% nickel ammonium sulfate and cobalt

FELDMANAND~TOCKDALE

chloride (De Blas and Cherwinski, 1983) to increase sensitivity. Monoclonal Antibodies and Nomenclature Myosin heavy chain expression was probed using a panel of monoclonal antibodies (mAb) produced and characterized in our laboratory (Crow and Stockdale, 1986a; Miller et al., 1985; Miller and Stockdale, 1986a,b; Schafer et al., 1987; Crow and Stockdale, 1984, 1986b; Stockdale and Miller, 1987). Briefly, F59 is an IgG specific for all fast isoforms of myosin heavy chain, S58 is an IgA specific for a slow MHC isoform that migrates with slow MHC2 in SDS-PAGE (with a slight reactivity to slow MHCl), and S46 is an IgG specific for slow isoforms that migrate with slow MHCs 1 and 2 in SDSPAGE. These antibodies were in hybridoma supernatant solution and were used at a 10x dilution for all procedures. We defined muscle fibers in culture (myotubes) as being fast type fibers if they stained with F59 only, as slow fibers if they stained with S58 and/or S46 alone, and as mixed fast/slow fibers if they stained with F59 and S58 and/or S46 (Miller and Stockdale, 1986a,b; Miller and Stockdale, 1987; Schafer et al., 1987; Stockdale and Miller, 1987; Stockdale et al., 1989; Miller and Stockdale, 1989; Stockdale, 1989; Stockdale, 1990a,b,c; Stockdale and Feldman, 1990). RESULTS

Fiber Types, and Myosin Heavy Chains in Adult Avian Pectoralis and Anterior Latissimus Dorsi Muscles The adult avian Pectoralis and Anterior Latissimus Dorsi muscles are composed of different fiber types. A panel of mAb specific for avian fast or slow isoforms of MHC were used to determine the MHCs expressed in these muscles and to assign fiber type (Crow and Stockdale, 1986a; Miller et al., 1985; Miller and Stockdale, 1986a,b). Sections of PM and ALD from adult chicken were stained with antibody mAbs F59, S58, or S46 (Fig. 1). All fibers in the adult PM were fast type. Fibers in the adult ALD were of three types, most were slow type with occasional fibers that contained either a mixture of both fast and slow or only fast MHC isoforms. Immunoblots of MHC extracted from these adult muscles confirmed the immunohistochemistry findings (e.g., Fig. 3, lanes 1 and 2). Quail PM and ALD are composed of the same fiber types as described for the chicken (data not shown). Muscle Fiber Types Formed in High Density Primary Satellite Cell Cultures Because adult avian PM and ALD muscles differ in their expression of MHC, we wished to determine if sat-

Satellite

Ceil

Divrrsity

323

ellite cells isolated from these muscles would form fibers of single or multiple types in cell culture. Fibers formed, in cell culture, from satellite cells isolated from PM and ALD were found to differ. Satellite cells from adult chicken or quail PM muscles were isolated and plated in high density culture, and, using immunocytochemistry, the types of fibers formed were determined at Days 8-9, 16-17, 24-25, and 32-33 after their initial plating. Fibers formed from chicken and quail adult PM satellite cells were of the fast type, because they only stained with mAb F59 at all time periods (Fig. 2 and Fig. 4a and 4~). On rare occasions a mononucleated myocyte was observed that, in addition, stained with mAb S46 in the chicken, or mAbs S46 and S58 in the quail. The preponderance of fast MHC isoforms in fibers of PM cultures was confirmed by immunoblotting extracts of these fibers using the same three monoclonal antibodies. Only fast MHC(s) were found (Fig. 3), when extracts were blotted from primary cultures of chicken PM (lanes 3-5) or quail PM (lanes 11-13) after 8 and 32 days, and after 32 days in the presence of cytosine arabinoside (Ara-C), respectively. Fibers formed in primary culture of ALD satellite cells isolated from adult chickens were of two types: fast and fast/slow (SMHCl). All of the fibers formed from ALD satellite myoblasts stained with mAb F59, and thus contained fast isoforms (Fig. 2). Approximately 25% of these fibers also expressed SMHCl-like isoform. This was deduced by comparing cocultures stained with mAbs F59 and S58 with those stained with mAbs S46 and S58. When cultures were stained with mAb S46, 25% of the fibers stained, while none of the fibers reacted with S58 alone (an occasional mononucleated myocyte stained with mAbs F59 and S58). Therefore about 25% of chicken ALD fibers contained an isoform like SMHCl in addition to a fast isoform of MHC (Fig. 4b) and 75% contained only fast isoforms. The presence of slow MHCl was confirmed by immunoblotting of extracts from ALD satellite cultures (Fig. 3). These blots showed that both fast MHC(s) and slow MHCl were present in extracts made from chicken ALD primary cultures after 8 and 32 days (lanes 7 and 8), and after 32 days in the presence of Ara-C (lane 9). The percentage of fibers that expressed both fast and slow MHC isoforms in chicken ALD satellite cell high density primary cultures decreased with longer durations of incubation (Fig. 4b). To determine if this decrease in fast/slow fibers was because fast/slow fibers became fast fibers, or because new fast fibers continued to form from remaining satellite cells, Ara-C was added to long-term cultures 8 days after initial plating to prevent formation of new satellite cell progeny and new fibers. The percentage of fibers that expressed SMHCl did not decrease under these conditions (Fig. 4b), sug-

324

DEVELOPMENTALBIOLOGY

VOLUME 143.1991

FIG. 1. Frozen sections of adult chicken PM and ALD muscles stained with monoclonal antibodies specific to fast and slow isoforms of MHC. Serial sections of frozen muscle were stained with either mAbs F59, S46, or S58, followed by a HRP-labeled secondary antibody specific for mouse immunoglobulins. Staining was visualized with DAB and hydrogen peroxide. Antibody F59 is specific for fast MHC isoforms, mAb S58 is specific for a slow isoform that migrates with slow MHCZ (SMHC2) in SDS-PAGE, and mAb S46 is specific for slow isoforms that migrate with slow MHCl (SMHCl) and SMHC2 in SDS-PAGE. Fibers of the fast PM stained exclusively with F59. Most fibers in the ALD stained with mAbs S46 and S58, but not mAb F59 and were, therefore, slow. However, the ALD contained some fibers that stained with mAb F59 alone: these were fast type (arrow), while others stained with all three and were, therefore, fast/slow type (fat arrowheads). Bar, 100 pm.

gesting that proliferation of principally fast type satellite cells was responsible for the decrease in fast/slow fibers in long-term ALD cultures. No change in fiber types was noted in chicken PM satellite cell cultures, whether they were exposed to Ara-C or not, for the same duration of time (Fig. 4a). These results from chicken PM and ALD satellite culture suggested that there was more than one type of satellite cell in adult muscles, those that formed fibers that expressed only fast MHC(s) and those that expressed both fast MHC(s) and a slow MHC, and that distribution of the two types differed between the muscles. Fibers formed in primary culture of adult quail ALD satellite cells were also of two types. A small percentage of fibers (up to 10%) contained both fast and slow MHC isoforms but the slow isoforms expressed were different than those expressed in fibers from chicken ALD satellite cells (Fig. 2; Fig. 4d). All fibers formed from quail ALD satellite cells reacted with mAb F59 and some of these also reacted with both mAbs S46 and S58 (Fig. 4d and first data point in Fig. 7d). The presence of slow MHCs was demonstrated by immunoblots of extracts of quail satellite cultures, as illustrated in Fig. 3, where lanes 15 and 16 were extracts from 8 and 32 days, and

lane 17 was 32 days with the addition of Ara-C. The relative percentages of fast and fast/slow fibers did not change in long-term cultures of either quail PM or ALD satellite cells in the presence or absence of Ara-C. Thus satellite cells from both chicken and quail produced at least two different types of muscle fibers in cell culture, fast and fast/slow, but the fast/slow type fibers formed from quail ALD satellite cells differed from those of the chicken, because they expressed SMHCl- and SMHCBlike isoforms while those of the chicken expressed SMHCl-like isoforms alone. Analysis of Satellite Cell Composition Avian Muscles

Clonul

of Adult

If satellite cells differ intrinsically from one another, then it should be possible to show that fibers formed from a single satellite cell differ from those formed from another satellite cell. Clonal analyses of satellite cells from the ALD and PM demonstrated that chicken and quail satellite cells are of at least two different types. Primary satellite cells from chicken and quail PM and ALD were plated at clonal density and the types of myogenic colonies they formed were determined by im-

FELDMANANDSTOCKDALE

F59

S58

FIG. 2. Double immunofluorescent staining of fibers from long-term primary cultures of chicken and quail satellite cells. Primary chicken and quail satellite cells isolated from the PM and ALD wew cultured for 32 days without passage, fixed. and double-immunostained with either mAhs F59 and S58 (first pair of columns) or mAhs S46 and S5X (second pair of columns). Monoclonal antibodies F59 and S46 were visualized with FITC-labeled anti-I& (T-chain specific) and S58 was visualized with TRITE-lahelrd anti-IgA ((u-chain specific). All fibers stained with mAh F59. Some lihers in chicken ALD cultures stained, though not strongly, with mAh S4ti. hut not with mAh S58. Bar. 100 pm.

munocytochemistry. Triplicate dishes of colonies were stained with mAbs F59, S58, or S46 and the percentage of positive myogenic colonies was scored as to fiber type using each of the antibodies (Table 1). All of the fibers in a colony formed from an individual chicken or quail satellite cell were of the same type-all of the fibers within the colony stained with the same antibodies. Colonies formed from chicken or quail PM satellite cells were primarily of the fast MHC phenotype. On the other hand, 71% of colonies formed from chicken ALD were exclusively of the fast fiber type and 29% of the colonies were exclusively of the fast/slow (SMHCl) type. None of the colonies contained fibers that stained with mAb S58 (contained SMHCZ). As with the chicken ALD, colonies from the quail ALD were of two types. About 71%’ of

quail ALD satellite cells formed colonies of the fast fiber type and 29% formed colonies of fast/slow SMHCl and SMHC2 type (stain with both S58 and S46). Because all colonies were derived from individual satellite cells cultured under identical conditions, primary chicken and quail satellite cells are of at least two distinct types, and the PM and ALD muscles of the adult differ in the proportions of the two types of satellite cells they contain. These results are in agreement with the fiber type measurements made in the high density cultures.

To determine the stability of the commitment of satellite cells to form fibers of specific types, high density

326

DEVELOPMENTAL

BIOLOGY

VOLUME

143,1991

Chicken PM -3456

Quail ALD

7 a

PM

ALD

-1 910

11 12

13

14

15 16 17

ia

F59

FIG. 3. Immunoblots of myosin extracted from chicken skeletal muscle and cell cultures of chicken and quail satellite cells. Myosin was extracted from adult chicken PM (lane 1) and ALD (lane 2) and from satellite cell cultures derived from chicken PM (lanes 3-6), ALD (lanes 7-lo), and quail PM (lanes ll-14), ALD (lanes 15-18). Extracts were prepared from primary satellite cell cultures after 8 days (lanes 3, ‘7,11,15), 32 days (lanes 4,8,12,16), 32 days with Ara-C (lanes 5,9,13,1’7), and from cultures serially subcultured at high density three times (4” culture) at 8-day intervals (lanes 6, 10, 14, 18). Extracts were separated on 5% SDS-PAGE, electrophoretically transferred to nitrocellulose, stained with mAbs F59, S46, or S58, followed by a HRP-tagged secondary antibody, and visualized with DAB and hydrogen peroxide. The amounts of myosin loaded in each lane were equalized by running test gels after doing protein determinations of the extracts. However, the loadings from chicken quarternary cultures were somewhat less because very little protein could be recovered from the small number of fibers produced. The apparent reaction of mAb S58 with chicken ALD culture in lane ‘7 may be due to slight reactivity of this antibody with SMHCl.

cultures of satellite cells were repeatedly subcultured and at each passage the fibers formed during the previous incubation period were discarded and only mononucleated cells were replated. The newly formed fibers in these high density cultures were stained, at each subculture, with the three mAbs 8 days after each replating. The progeny of satellite cells from the chicken PM carried through repeated high cell density subcultures formed fibers, virtually all of which were fast (all reacted with F59 and less than 1% also reacted with mAbs S46 or S58) (Fig. 5 and Fig. ‘7a). Serial subculturing of chicken ALD satellite cells led to an increase in the formation of fibers of the fast phenotype and a decrease in those of the fast/slow phenotype as had been seen in long-term culture without subculture (Fig. 6 and Fig. 7b; compare to Fig. 4b). Initially, about 75% of the fibers in high density ALD satellite cultures were of the fast phenotype, whereas by the second (3”) or third (4”) subculture 90-95% of the fibers formed by the progeny of chicken ALD satellite cells were of the fast phenotype (Fig. 7b). The decrease from 35 to 5-10% fast/slow fibers in later passages may have been due to overgrowth of one satellite cell type by another as was

shown in unpassaged cultures using Ara-C. Immunoblotti.ng of MHC extracted from the third subculture (4”) showed that PM satellite cultures only contained fast MHC(s) (Fig. 3, lane 6), while ALD satellite cultures contained increased amounts of fast MHC(s) and decreased amounts of slow MHC that reacted with S46 (Fig. 3, lane 10). Unlike with the chicken, there was a dramatic change in the types of new fibers formed when the progeny of quail satellite cells from either the PM or ALD were repeatedly subcultured. After two subcultures of either the PM or the ALD quail satellite cells there was a progressive increase in newly formed fast/slow fibers and a decrease in those of the fast phenotype (Figs. 5 and 6; and Figs. 7c and 7d). Using double immunofluorescence staining with mAbs F59 and S58, the proportion of fibers staining with both antibodies increased from less than 1% in primary culture (1”) to 90% of all the fibers formed in the second subculture (3”) of PM satellite cells and from 10% in primary culture (1”) to 95% of all the fibers formed in the first subculture (2”) of ALD satellite cells. Virtually all (99%) of the fibers formed from ALD satellite cell progeny were of the fast/slow pheno-

a

Chicken

PM

b

- Continuous

Chicken

Quail

C :

PM

a 3 5 21 r: z '; 8 &

80 70 6050 4030 zolo-

a

a

-

Ob 8

I 24

25

Quail

ALD

I 32

Days

33

:!

Contmuous

-_-----____-_-________

__I 16

I

17

d

-

- Continuous

in Culture

- Continuous

100 90 -

ALD

1

I 9 Days

327

Satellite Cell Lkrrsity

FELDMANAND~TOCKDALE

8

16

w

iij i I

24

32

in Culture

FIG. 4. Effect of prolonged primary culture on isoforms of MHC expressed in chicken and quail satellite cell fibers. The relative percentages of fibers stained with mAb F59 alone (m, Cl) or with mAbs F59, S58, and/or S46 (A, a) were determined in high density primary cultures of (a) chicken PM, (b) chicken ALD, (c) quail PM, and (d) quail ALD satellite cells both in the presence of Ara-C (0, a) and in the absence of Ara-C (w, A). Cultures were fixed with 1007% ethanol at S-day intervals, immunofluorescently double-stained with either mAbs F59 and S58, or mAbs S46 and S58, and then individual fibers were scored as positive or negative for staining by observation under fluorescence and phase-contrast microscopy (see Methods). At least 300 fibers were counted on each dish.

type by the third (4”) subculture (Figs. 6 and 7d). Immunoblots of extracts from sequential subcultures of quail PM and ALD satellite cells contained increased amounts of slow MHCs that reacted with both S46 and S58 (Fig. 3, lanes 14 and 18). The progressive appearance of a different type of newly formed fiber after sequential subcultures of adult quail satellite cells resulted from changes in the percentages of satellite cell progeny committed to particular

TABLE

1

PERCENTAGESOFPRIMARYSATELLITECELLCOLONIESEXPRESSING ONLYFASTORBOTHFASTAND SLOW MYOSINHEAVYCHAINS

Muscle

Fast colonies ‘5 (&SEM)

Fast/slow colonies %(?SEM)

Chicken PM Chicken ALD Quail PM* Quail ALD*

100 71(B) 90( k6) 71(k13)

0 29(B) 10(?6) 29(k13)

* The individual experiments are shown in Table 2.

from

which

these

Total colonies 224 303 433 1398

data

were

derived

phenotypes. Clonal density cultures of satellite cell progeny were established after each high density subculture and the myogenic colonies they formed were assayed for MHC expression to determine the composition of the satellite cell population before each subculture (Table 2). By the third high density subculture (4”) nearly 90% of quail PM satellite cell progeny formed colonies of the fast/slow fiber type, while 89-98% of colonies formed by the second to third subculture (3” to 4“) of ALD satellite cell progeny were of the fast/slow type. Thus, serial subculture at high or low density resulted in a change within the population from fast to principally fast/slow satellite cell type. The change in composition of the quail satellite population that occurred with expansion through repeated subculture at high density was investigated further to prove that the nature of the effect was an intrinsic change within individual satellite cell progeny rather than being solely an overgrowth of the population by one type of satellite cell. Several colonies, all derived from one quail PM or ALD satellite cell, were serially recloned up to four consecutive times. All of the fibers in all colonies at each sequential recloning reacted with

328

DEVELOPMENTALBIOLOGY

F59

FIG. 5. Double immunofluorescent tures). Primary satellite cultures satellite cell cultures from adult centlg double-stained with either

S58

VOLUME 143,lW

S46

S58

staining of chicken and quail PM satellite fibers after primary and quarternarg culture (three subculwere fixed after 8 days of incubation. Quarternary cultures were derived from three passages of high density satellite cell progeny which were subcultured at R-day intervals. The fibers were fixed and immunofluoresmAbs F59 and S58 (first pair of columns) or mAbs S46 and S58 (second pair of columns). Bar, 100 Wm.

F59 and thus expressed fast isoform(s) of MHC. Three patterns of behavior were observed with regard to slow MHC expression in fibers of serially cloned satellite cells: (1) those in which no fibers reacted with the slow antibodies; (2) those in which all of the fibers reacted with the slow antibodies; and (3) those in which some of the fibers within a colony reacted while others did not react with the slow antibodies (Fig. 8). The most common finding was that satellite cells derived from either the ALD or the PM formed colonies of the fast/slow type with serial recloning (Fig. 9). In some clonal lines of satellite cells, however, repeated recloning did not result in progeny that formed colonies containing fibers of a single type. Such clonal lines continued to form colonies of several types-exclusively fast, mixed fast and fast/slow, or exclusively fast/slow fiber types (as de-

scribed in Fig. 8). These results suggest that the changes in myosin isoform expression reported in passaged high density cultures were the result of intrinsic changes in the satellite myoblasts themselves and not of the overgrowth of one population of satellite myoblasts by another. DISCIJSSION

These data demonstrate that skeletal muscle satellite cells are of more than a single type. At least two types of satellite cells were discerned, one that formed only fast fibers, and another that formed fast/slow fibers. In addition, these two types of satellite cells had differences in distribution between muscles. The PM contained satellite cells committed to forming fibers that expressed

FELDMANANDSTOCKDALE

F59

S58

FIG. 6. L~ouble immunofluorescent staining of chicken tures). Primary satellite cultures were fixed after 8 days satellite cell cultures which were subcultured at R-day mAhs F’S9 and S58 (first pair of columns) or mAhs S46

and quail ALD satellite fibers after primary and quarternary of incubation. Quarternary cultures were derived from three interyals. The fibers were fixed and double immunofluorescently and S58 (second pair of columns). Bar, 100 pnl.

only fast MHC(s), whereas the ALD contained satellite cells committed to forming fibers that either exclusively expressed fast MHC(s) or expressed both fast MHC(s) and slow MHC(s). The slow isoforms expressed in fast/ slow type fibers differed between chicken and quail. Repeated cell division at high or clonal density was shown to result in an intrinsic change in the progeny of quail satellite cells, resulting in a change in the type of fiber they formed. Other investigators of satellite cell differentiation in cell culture did not find differences in the type of MHC that were expressed when the source of satellite cells was either a fast or a slow muscle. Matsuda and coworkers (1983), using biochemical methods, showed that fibers formed in cell culture from satellite cells derived from either fast or slow type muscles expressed only

cultures (three subculpassages of high density lahrled with (sither

fast MHC isoforms, regardless of the origin of the satellite cells. On the other hand these and others showed that muscles regenerating in vim from the resident satellite cells, following cold injury under innervated or denervated conditions, expressed only fast myosin isoform(s) in fast muscle regenerates, and fast MHC isoforms, followed by slow MHC isoforms in the slow muscle regenerates (Cerny and Bandman, 1987; Bandman et trl., 1989; Saad et (xl., 1986, 1987; Matsuda ef al., 1983). These observations were consistent with the direct observations reported here on intrinsic differences among satellite cells. The results of studies of mammalian regeneration in zji~o are somewhat in conflict with the proposition that there is diversity among satellite cells. Fast extensor digitorum longus muscle (EDL) of the rat or cat regener-

330

DEVELOPMENTALBIOLOGY a 3

lOOl# 90 -

I2 3 5 x I

so 70 6050 -

z z g

4030zo-

Chicken

PM

VOLUME 143,1991

b

Chlcken

d

Quail

Subculture

ALD

- Subculture

i !I

Passage

C

Quail

PM

Subculture

ALD

Subculture

Passage FIG. 7. Effect of serial subculture on the percentages of fast and fast/slow fibers formed in high density chicken and quail satellite cell cultures. (a) Chicken PM, (b) chicken ALD, (c) quail PM, and (d) quail ALD derived satellite cell cultures were subcultured at &day intervals three times and immunofluorescently stained for fast and slow MHC after each subculture. The proportions of fibers stained with mAb F59 alone (m) or with mAbs F59, S58, and/or S46 (A) were determined for satellite cells from each muscle.

ates as fast muscle fibers and innervated slow soleus as slow muscle fibers. However, in the absence of innervation, regenerated fibers in either the EDL or the soleus only expressed fast MHC isoforms (Kelly and Rubinstein, 1980; Rubinstein and Kelly, 1980; Hoh and Hughes, 1988). These experiments suggest that the limb musculature of the rat and cat possessonly a single type of satellite cell which in the presence of innervation can form different types of fibers. But, Hoh and co-workers (1988, 1989) showed that regenerated fibers of the cat superfast muscle (posterior temporalis) either expressed superfast or slow MHC when they regenerated under any state of innervation, even in the bed of a muscle that normally expressed only adult fast MHC. Regenerating fast (EDL) or slow (soleus) muscles produced fibers expressing fast and slow MHC, but not superfast MHC, even when transplanted into the bed of the posterior temporalis muscle. The observations on regeneration of the mammalian posterior temporalis are consistent with the direct demonstrations reported here that individual satellite cells can form fibers of different types and that satellite cell myoblast types are not uniformly distributed throughout the musculature, but are segregated to specific muscles.

Satellite cells isolated from chicken sustained a commitment to expression of specific classes of MHC within the fibers they formed through many more cell generations than did quail satellite cells. Repeated subculture of the progeny of chicken PM satellite myoblasts resulted in no apparent change in MHC expression in the fibers they formed. While repeated passage of chicken ALD satellite cells resulted in an increase in the proportion of fast type versus fast/slow type fibers. This appeared to reflect overgrowth of one satellite cell type by another. Quail satellite cells, on the other hand, produced progeny that underwent a unidirectional change that resulted in the formation of a different type of fiber. The rapidity of change in the quail ALD satellite cells from those that formed only fast to those that formed only fast/slow fibers, after serial recloning of single satellite cells and their progeny, suggested that this phenomenon resulted from an intrinsic change in the satellite cells rather than from overgrowth alone. Serially subcultured quail fetal myoblasts have also been shown to change such that they form fibers that express different MHCs (Schafer et al., 1987), different myosin light chains, and/or different tropomyosins (Montarras and Fiszman, 1983). Such a change has been

Sutellife tkll Diwrsit g TABLE2 PERCENTAGESOF SATELLITE CELLPROGENYCOMMITTEDTOFORMINGFASTANDFAST/SLOWMYOGENICCOLONIESATEACHSUBCULTURE OFHIGHDENSITYCULTURES Fast Muscle Quail

PM

Passage” 1”

2” 3” 4” Quail

ALD

Experiment codeb

Total colonies

78

22

46

: d d” d

100 92 62 19 9 11

80 38 81 91 89

313 74 110 47 35 88

E %

99 42 84 58

1 58 42 16

290 278 212 618

d” a d d

62 16 17 2 8

38 84 83 98 92

64 76 29 68 86

2”

4”

Fast/slow colonies (‘5)

b

1”

3”

colonies (a)

’ Primary (1”) cultures were established from satellite cells isolated directly from the muscles and subcultured at high density ever 8-9 days. Cultures designated 2”-4” were derived from satellite cell progeny lifted from high density culture dishes after 8 days of incubation and plated clonal cell density (approx. 3-7 cells/cm’ dish). b All experiments and time points in which at least 10 colonies grew in each dish were included in the data.

designated as a change in commitment (Stockdale et a,l., 1989; Stockdale, 1989). The phenomenon of change in commitment of individual satellite cells was best illustrated in clones of individual satellite cells. The change appeared to be uniformly rapid when ALD satellite cells were cloned but less so when PM satellite cells were cloned (Table 2 and

331

Fig. 9). No change occurred in long-term primary cultures of either ALD or PM, where continued new fiber formation was uncommon. The change in quail satellite cells from those that formed fast fibers to those that formed fast/slow fibers only occurred subsequent to the removal of the first generation of fibers from the cultures and the formation of new fibers by satellite cell progeny. Of the colonies derived from clones picked from low density primary cultures of ALD satellite cells (where 80% or more of the colonies expressed only fast MHC), over 80% of the colonies formed by the first subcloning contained fibers that expressed both fast and slow MHCs. However, clones isolated from primary quail PM cultures varied more widely in the fiber types formed in subclones. Many of the individual colonies derived from a cloned satellite cell contained fibers of different types. Often, only a section of a colony would stain with the S58 antibody (Fig. 8), suggesting that a change in commitment occurred as progeny were produced in the colony and that whatever event or mechanism was responsible for the change was initiated after a small number of divisions. The mechanism of change in commitment of fetal and satellite myoblasts is unknown, but because it is associated with the production of new cells it could be related to a change in methylation within the genome. Recent observations on the regulation of MLC lf/3f gene expression have shown that hypomethylation within the promoter region of this gene was associated with expression of light chain 3f (Lamson and Stockdale, 1989). Likewise, hypomethylation of DNA has been shown to be a mechanism associated with commitment of mesenchymal cells to a myogenie fate (Konieczny and Emerson, 1984; Chiu and Blau, 1985; Davis et al., 1987). Satellite cells join other myoblasts in demonstrating differences in differentiated fate. Analyses of avian and mammalian myoblasts isolated at different times dur-

FIG. 8. Three types of fibers found in colonies in clonal density cultures of serially subcloned quail satellite cells. Low density cultures were grown for 8 days after subcloning, fixed with 100% ethanol, stained with either mAbs F59, S58, or S46, and an HRP-tagged secondary antibody. After visualizing the staining with DAB and hydrogen peroxide, the dishes were counter-stained with 0.01% methylene blue in 254 ethanol. Shown are three quail PM satellite cell colonies formed from a single clonal subculture found on the same dish after four serial subclonings and stained with mAb S58. Fibers in one colony (left) did not stain with S58, in another colony some of the fibers, but not others, stained with mAb S58 (center), and all fibers in another colony stained with mAb S58 (right). Bar, 100 p,

332

DEVELOPMENTAL

Quail 100%

BIOLOGY

VOLUME

b

PM

143, 1991

Quail

PM

+

1'

2'

3~

4'

5~

1"

2"

Passage

C

d

Quail ALD

1’

2'

3'

4'

3'

4'

Passage

3'

4'

Passage

bail

1'

ALD

2'

PaSSage

FIG. 9. Changes in myosin isoform expression in fibers formed by serially recloned single yuail PM and ALD satellite cells. Primary (I”) quail satellite cells from the PM (a, b) and ALD (c, d) were plated at clonal density; a colony was isolated, lifted from the dish, and replated into three dishes (secondary (2”) colonies)-this procedure was repeated through tertiary (3”) and quarternary (4”) reclonings. The dishes were fixed and stained with mAbs F59, S46, or S58 and scored for the different colony types (see Fig. 8). The same procedure was followed with all of the reclonings. Each graph represents one experiment (clones isolated from one primary culture) in which the data were derived from the progeny of individual parental satellite cells. Graph a represents the average values from all colonies from six clones and associated subclones, b and c represent the average values from all colonies from three clones, and d represents the average values from all colonies from five clones. Shown are the average percentages (k standard deviations) of colonies of the fast/slow type in primary culture and in each subsequent recloning. The clones followed in each experiment were those that survived and produced at least 10 colonies per dish at each recloning. To simplify the results, colonies that contained fibers, some or all of which reacted with antibodies to slow MHC (see Fig. 8). were scored as positive for slow type fibers.

ing embryonic and fetal development in vivo show that different categories of myoblasts exist. Distinct populations of early (embryonic) and late (fetal) myoblasts have been demonstrated to exist in chick, quail, mouse, and human embryos (White et al., 1975; Rutz and Hauschka, 1982; Stockdale and Miller, 1987; Hauschka, 1974; Bonner and Hauschka, 1974; Mouly et al., 1987) and do not appear to be lineal descendents of one another (White et ab, 1975; Miller et al, 1985; Miller and Stockdale, 1986a,b; Schafer et al., 1987; Rutz and Hauschka, 1982; Bonner and Hauschka, 1974; Mouly et al., 1987; Seed and Hauschka, 1984). The myoblasts that appear in embryonic and fetal development form fibers that differ in fiber and colony morphology and media requirements for fiber formation in cell culture. They also form fibers that exhibit differences in the MHCs expressed (Crow and Stockdale, 1986a; Miller and Stockdale, 1986a,b; Schafer et al., 1987; Rubinstein and Holtzer, 1979; Vivar-

elli et ah, 1988; Hoffman et al., 1989). Differences between fetal myoblasts and satellite cells also exist. Cultured satellite cells from both mouse and human are capable of differentiating in the presence of the tumor promoter TPA, while fetal myoblasts are not (Cossu et al., 1983,1985). Embryonic myoblasts, like satellite cells, are also resistant to the effects of TPA (Cossu et ab, 1988). Mouse satellite cells express acetylcholine receptors while fetal myoblasts do not (Cossu et al., 1987) and chicken satellite cells have also been reported to express different surface antigens than fetal myoblasts (Yablonka-Reuveni, 1988). As shown here for chicken satellite cells and previously for chicken embryonic and fetal myoblasts, the constellation of MHCs expressed in cell culture by the fibers they each form differs as well (Miller et al., 1985; Miller and Stockdale, 1986a,b; Miller and Stockdale, 1989). Thus satellite cells have properties that permit them to be distinguished from one another

FELDMAN

AND STOCKDALE

and from myoblasts that precede them in development. It remains to be determined whether earlier myoblasts exhibit differential segregation within putative fast, mixed, and slow muscles; how myoblasts become restricted in the types of fibers they form; how this restriction is altered by cell proliferation; and what the lineal relationship is between satellite cells and earlier forms of myoblasts. We thank Barbara Hill and Sandra Conlon for their excellent technical assistance and Gloria Garcia for her invaluable secretarial assistance. This work was supported by MDA and NIH grants to F.E.S. J.L.F. was initially supported by the Program in Cancer Biology at Stanford and was a recipient of an MDA fellowship. This paper is dedicated to the memory of Dr. Alexander Mauro, who first recognized and defined the satellite cell.

REFERENCES ABE, J., FUJII, Y., KUWAMURA, Y., and HIZAWA, K. (1987). Fiber type differentiation and myosin expression in regenerating rat muscles. Acta Puthol. J(zpun 37, 1537-1547. ALLEN, R. E., and BOXHORN, L. (1987). Inhibition of skeletal muscle satellite cell differentiation by transforming growth factor-0. J, Cell. PhysioL 133, 567-572. ARNDT, I., and PEPE, F. A. (1975). Antigenic specilicity of red and white muscle myosin. J. Histoche~n. C!4toclrenl. 23, 159-168. BANDMAN, E., CERNY, L. C., and BENNETT, T. (1989). Myogenic and non-myogenic factors regulate the expression of myosin heavy chains in developing and regenerating chicken skeletal muscle. 171 “Cellular and Molecular Biology of Muscle Development” (L. H. Kedes and F. E. Stockdale, Eds.), pp. 429-439. A. R. Liss, New York. BISCHOFF, R. (1986). A satellite cell mitogen from crushed adult muscle. Deu. BioL 115, 140-147. BISCHOFF, R. (1990). Cell cycle commitment of rat muscle satellite cells. J. Ccl2 BioL 111, 201-207. BONNER, P. H., and HAUSCHKA, S. D. (1974). Clonal analysis of vertebrate myogenesis. I. Early developmental events in the chick limb. Del,. Bid. 37, 317-328. BRADFORD, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Ad. Bioch,enr. 72,248-254. BULLER, A. J., ECCLES, J. C., and ECCLES, R. M. (1960). Interactions between motoneurones and muscles in respect of the characteristic speeds of their responses. J. Phgsiol. 150, 417-439. CAMPION, D. R. (1984). The muscle satellite cell: A review. Irct. Rrr. Cytol. 87, 225-251, CERNY, L. C., and BANDMAN, E. (1986). Contractile activity is required for the expression of neonatal myosin heavy chain in embryonic chick pectoral muscle cultures. J. Cell Biol. 103, 2153-2161. CERNY, L. C., and BANDMAN, E. (1987). Expression of myosin heavy chain isoforms in regenerating myotubes of innervated and denervated chicken pectoral muscle. Del:. BioL 119, 350-362. CHIU, C.-P., and BLAU, H. M. (1985). 5-Azacytidine permits gene activation in a previously noninducible cell type. Cell 40,417-424. Cossu, G., CICINELLI, P., FIERI, C., COLETTA, M., and MOLINARO, M. (1985). Emergence of TPA-resistant “satellite” cells during muscle histogenesis of human limb. EXJJ. Cell Res. 160, 403-411. Cossu, G., EUSEBI, F., GRASSI, F., and WANKE, E. (1987). Acetylcholine receptor channels are present in undifferentiated satellite cells but not in embryonic myoblasts in culture. Dev. Bid. 123,43-50. Cossu, G., MOLINARO, M., and PACIFICI, M. (1983). Differential re-

SfStrtrllitu

(‘P/I

Diwr.sitj/

333

sponse of satellite cells and embryonic myoblasts to a tumor promoter. L)r,l. Bid. 98, 520-524. Cossu, G., RANALDI, G., SENNI, M. I., MOLINARO, M., and VIVARELLI, E. (1988). “Early” mammalian myoblasts are resistant to phorbol ester-induced block of differentiation. DewZoprrcent 102,65-69. CROW, M. T., and STOCKDALE, F. E. (1984). Myosin isoforms and the cellular basis of skeletal muscle development. EXP. Bid. Med. 9, 165-174. CROW, M. T., and STOCKDALE, F. E. (1986a). Myosin expression and specialization among the earliest muscle fibers of the developing avian limb. &II. Riol. 113, 238-254. CROW, M. T., and STOCKDALE, F. E. (1986b). The developmental program of fast myosin heavy chain expression in avian skeletal muscles. Dm Rid. 118, 333-342. D’ALBIS, A., COUTEAUX, R., JANMOT, C., ROULET, A., and MIRA, J.-C. (1988). Regeneration after cardiotoxin injury of innervated and denervated slow and fast muscles of mammals. Myosin isoform analysis. Errr. J. Biochrtrc. 174, 103-110. DAVIS, R. L., WEINTRAUB, H., and LASSAR, A. B. (1987). Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cdl 51,987-1000. DE BLAS, A. L., and CHERWINSKI, H. M. (1983). Detection of antigens on nitrocellulose paper immunoblots with monoclonal antibodies. Anrzl. Biochen,. 133, 214-219. DIMARIO, J., and STROHMAN, R. C. (1988). Satellite cells from dystrophic (mdx) mouse muscle are stimulated by fibroblast growth factor in vitro. D~f~~reutiution 39, 42-49. GAMBKE, B., LYONS, G., HASELGROVE, J., KELLY, A., and RUBINSTEIN, N. (1983). Thyroidal and neural control of myosin transitions of rat fast and slow muscles. FEBS Left. 156, 335-339. HAUSCHKA, S. D. (1974). Clonal analysis of vertebrate myogenesis. III. Developmental changes in the muscle-colony-forming cells of the human fetal limb. Der Bid. 37, 345-368. HOFFMAN, R. K., RUBINSTEIN, N. A., and KELLY, A. M. (1989). Differential expression of myosin heavy chain genes in primary cultures of rat fast and slow hindlimb muscles. J. &ll Bid 109,260a. [Abstr] HOH, J. F. Y., and HUGHES, S. (1988). Myogenic and neurogenic regulation of myosin gene expression in cat jaw-closing muscles regenerating in fast and slow limb muscle be&. J. Muscle Rex Cell MofiL 9, 59-72. HOH, J. F. Y., HUGHES, S., HUGH, G., and POZGAJ, I. (1989). Three hierarchies in skeletal muscle fibre classification: Allotype, isotype and phenotype. 1~ “Cellular and Molecular Biology of Muscle Development” (L. H. Kedes and F. E. Stockdale, Eds.), UCLA Sgmp. Mol. Cell. Biol., pp. 15-26. A. R. Liss, New York. KELLY, A. M., and RUBINSTEIN, N. A. (1980). Patterns of myosin synthesis in regenerating normal and denervated muscles of the rat. III “Plasticity of Muscle” (D. Pette, Ed.), pp. 161-175. de Gruyter, New York. KONIECZNY, S. F., and EMERSON, C. P. (1984). 5-Azacytidine induction of stable mesodermal stem cell lineages from lOTl/Z cells: Evidence for regulatory genes controlling determination. &/I 38,79-800. LAEMMLI, Ii. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. N(zt/rrc’ (Ltn&~t) 227, 6X0685. LAMSON, G., and STOCKDALE, F. E. (1989). Developmental and musclespecilic changes in methglation of the myosin light chain LClf and LCYf promoters during avian myogenesis. Del,. Bid. 132,62-68. LANGER, B. G., LEUNG,L. L., GONNELLA, P. A., NACHMIAS, V. T., NACHMAN, R. L., and PEPE, F. A. (1982). tu-Actinin and membrane glycoprotein IIIa are different proteins in human blood platelets. Proc. Ntrtl. Amd. Sri. USA 79, 43%5. MATSUDA, R., SPECTOR, D. H., and STROHMAN, R. C. (1983). Regenerating adult chicken skeletal muscle and satellite cell cultures express

334

DEVELOPMENTAL BIOLOGY

embryonic patterns of myosin and tropomyosin isoforms. De?,. Biol. 100,478-488.

MAURO, A. (1961). Satellite cell of skeletal muscle fibers. J. Biophys. Biochem.

Cytol.

9, 493-495.

MILLER, J. B., CROW,M. T., and STOCKDALE, F. E. (1985). Slow and fast myosin heavy chain content defines three types of myotubes in early muscle cell cultures. J. Cell Biol. 101, 1643-1650. MILLER, J. B., and STOCKDALE, F. E. (1986a). Developmental origins of skeletal muscle fibers: Clonal analysis of myogenic cell lineages based on expression of fast and slow myosin heavy chains. Proc. N&l.

Acad.

Sci. USA 83, 3860-3864.

MILLER, J. B., and STOCKDALE, F. E. (1986b). Developmental regulation of the multiple myogenic cell lineages of the avian embryo. J. Cell Biol. 103, 2197-2208. MILLER, J. B., and STOCKDALE, F. E. (1987). What muscle cells know that nerves don’t tell them. Trends Neuro Ski. 10,325-329. MILLER, J. B., and STOCKDALE, F. E. (1989). Multiple cellular processes regulate expression of slow myosin heavy chain isoforms during avian myogenesis in vitro. Deu. Biol. 136, 393-404. MONTARRAS, D., and FISZMAN, M. Y. (1983). A new muscle phenotype is expressed by subcultured quail myoblasts isolated from future fast and slow muscles. J. Biol. Chem. 258, 3883-3888. MOULY, V., TOUTANT, M., and FISZMAN, M. Y. (1987). Chick and quail limb bud myoblasts, isolated at different times during muscle development, express stage-specific phenotypes when differentiated in culture. Cell Difleer. 20, 1’7-25. RUBINSTEIN, N. A., and HOLTZER, H. (1979). Fast and slow muscles in tissue culture synthesize only fast myosin. Nuture (London) 280, 323-325.

RUBINSTEIN, N. A., and KELLY, A. M. (1980). The sequential appearance of fast and slow myosins during myogenesis. I?l “Plasticity of Muscle” (D. Pette, Ed.), pp. 147-159. de Gruyter, New York. RUSHBROOK,J. I., and STRACHER, A. (1979). Comparison of adult, embryonic and dystrophic myosin heavy chains from chicken muscle by sodium dodecyl sulfate/polyacrylamide gel electrophoresis and peptide mapping. Proc. Nutl. Acad. Sci. USA 76,4331-4334. RUTZ, R., and HAUSCHKA, S. (1982). Clonal analysis of vertebrate myogenesis. VII. Heritibility of muscle colony type through sequential subclonal passages in n&o. Dew. Biol. 91, 103-110. SAAD, A. D., OBINATA, T., and FISCHMAN, D. A. (1986). Expression of myosin and C-protein isoforms in regenerating muscle. In “Molecular Biology of Muscle Development” (C. Emerson, D. Fischman, B. Nadal-Ginard, and M. A. Q. Siddiqui, Eds.), pp. 309-322. A. R. Liss, New York. SAAD, A. D., OBINATA, T., and FISCHMAN, D. A. (1987). Immunochemical analysis of protein isoforms in thick myofilaments of regenerating skeletal muscle. Dell. Biol. 119, 336-349. SALMONS, S., and VRBOVA, G. (1969). The influence of activity on some contractile characteristics of mammalian fast and slow muscles. J. Physiol.

201, 535-549.

SCHAFER, D. A., MILLER, J. B., and STOCKDALE, F. E. (1987). Cell diversification within the myogenic lineage: In vitro generation of two

VOLUME 143,199l

types of myoblasts from a single myogenic progenitor cell. Cell 48, 659-670. SEED, J., and HAUSCHKA, S. D. (1984). Temporal separation of the migration of distinct myogenic precursor populations into the developing chick wing bud. Dev. BioZ. 106,389-393. SILBERSTEIN, L., and BLAU, H. M. (1986). Two fetal-specific fast myosin isozymes in human muscle. In “Molecular Biology of Muscle Development,” pp. 253-262. A. R. Liss, New York. SILBERSTEIN, L., WEBSTER, S. G., TRAVIS, M., and BLAU, H. M. (1986). Developmental progression of myosin gene expression in cultured muscle cells. Cell 46, 1075-1081. STOCKDALE, F. E. (1989). Skeletal muscle fiber specification during development and the myogenic cell lineage. In “The Assembly of the Nervous System” (L. T. Landmesser, Ed.), pp. 37-50. A. R. Liss, New York. STOCKDALE, F. E. (1990a). The myogenic lineage: Evidence for multiple cellular precursors during avian limb development. Proo. Exp. Biol.

Med.

194, 71-75.

STOCKDALE, F. E. (1990b). Muscle fiber development and origins of fiber diversity. Zn “The Dynamic State of Muscle Fibers” (D. Pette, Ed.), pp. 121-125. de Gruyter, New York. STOCKDALE, F. E. (199Oc). Myoblast diversity and the formation of the early limb musculature. Awn. N. Y Acud. Sci. 599, 111-118. STOCKDALE, F. E., and FELDMAN, J. L. (1990). Satellite cells and the concept of cell lineage in avian myogenesis. In “The Dynamic State of Muscle Fibers” (D. Pette, Ed.), pp. 641-649. de Gruyter, New York. STOCKDALE, F. E., and MILLER, J. 8. (1987). The cellular basis of myosin heavy chain isoform expression during development. Dev. Biol. 123, 1-9.

STOCKDALE, F. E., MILLER, J. B., FELDMAN, J. L., LAMSON, G., and HAGER, J. (1989). Myogenic cell lineages: Commitment and modulation during differentiation of avian muscle. In “Cellular and Molecular Biology of Muscle Development” (L. H. Kedes and F. E. Stockdale, Eds.), UCLA Symp. Mol. Cell. Biol., pp. 3-13. A. R. Liss, New York. TOWBIN, H., STAEHLIN, T., and GORDON, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Nutl. Acnd. Ski. USA 76,4350-4354.

VIVARELLI, E., BROWN, W. E., WHALEN, R. G., and Cossu, G. (1988). The expression of slow myosin during mammalian somitogenesis and limb bud differentiation. J. Cc/l Biol. 107,2191-2197. WHITE, N. K., BONNER, P. H., NELSON, D. R., and HAUSCHKA, S. D. (1975). Clonal analysis of vertebrate myogenesis. IV. Medium-dependent classification of colony-forming cells. Dev. Biol. 44, 346361.

YABLONKA-REUVENI, Z. (1988). Identification of satellite cell-specific antigens. J. Cell. Bioch,em. 12C (suppl.), 334. [Abstr] YABLONKA-REUVENI, Z., QUINN, L. S., and NAMEROFF, M. (1987). Isolation and clonal analysis of satellite cells from chicken pectoralis muscle. Dell. Biol. 119, 252-259.