DEVELOPMENTAL DYNAMICS 211:141–152 (1998) Alteration in Myosatellite Cell Commitment With Muscle Maturation JIWEI YANG,1 ROBERT KELLY,2 MOLLY DAOOD,3 MARTIN ONTELL,1 JON WATCHKO,3 and MARCIA ONTELL1* 1Department of Cell Biology and Physiology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 2Department of Molecular Biology, Pasteur Institute, Paris, France 3Department of Pediatrics and Magee Womens Research Institute, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania ABSTRACT Myosatellite cells are myoblasts found between the basal lamina and sarcolemma of myofibers of postnatal mice. The extent to which these cells are programmed, upon differentiation, to express isoforms of contractile protein genes specific to the type of fiber with which they are associated has been evaluated in vitro using myosatellite cells derived from the soleus and the extensor digitorum longus muscles (EDL) of 4-day-old and adult transgenic mice, which express nuclear localizing b-galactosidase (nlsb-gal) under the control of the promoter and 38 enhancer of the gene encoding fast myosin light chain 3F (MLC3F) (Kelly et al.  J. Cell Biol. 129:383–396). Cultures were allowed to differentiate either as myocytes (mononucleated cells), to prevent possible modification of the myosatellite phenotype by other myonuclei in mosaic myotubes, or as myotubes. Transgene expression was age related, with 90% and 70% of the myocytes derived from the neonatal EDL and soleus muscles (muscles that had not yet achieved their mature phenotype), respectively, having nuclei encoding b-gal; 61% and 32% of the myocyte nuclei derived from myosatellite cells of the adult EDL (a fast muscle) and the adult soleus muscle (a mixed muscle containing many slow myofibers), respectively, expressed this transgene. Because myosatellite cells found in adult muscles are the progeny of those found in the neonate, an alteration of myosatellite cell commitment to express this transgene occurs with muscle maturation. Because expression of the transgene in neonatal and adult muscle in vivo reflects the expression of the endogenous MLC3F gene (Kelly et al.  J. Cell Biol. 129:383–396), it is likely that expression of the transgene by differentiated myosatellite cells reflects the extent of commitment of these cells to produce MLC3F. A hypothesis is presented that MLC3F is widely expressed in developing muscles but eliminated in myofibers that undergo maturation toward a slower phenotype. Dev. Dyn. 1998;211:141–152. r 1998 Wiley-Liss, Inc. Key words: myosatellite cell; commitment; MLC3F transgene; myogenesis; in vitro r 1998 WILEY-LISS, INC. INTRODUCTION The formation of skeletal muscle fibers during in utero development, during muscle regeneration, or in culture involves the fusion of mononucleated myoblasts to form multinucleated myofibers. In developing avian muscle, in which diversity among myotubes is present from the time myotubes first form (Miller et al., 1985), it has been demonstrated that there are discrete populations of embryonic and fetal myoblasts (Bonner and Hauschka, 1974; Miller, 1992; Stockdale, 1992; DeCusella Angelis et al., 1994). Moreover, three types of embryonic myoblasts can be identified based on the myosin heavy chain (MHC) isoforms produced by myotubes formed by clonal cultures of these myoblasts (Miller et al., 1985). In developing mammals, in which the primary myotubes initially display homogeneity of their MHCs (Narusawa et al., 1987), differences in cultured embryonic and fetal myoblasts in terms of their MHC accumulations (Vivarelli et al., 1988; Smith and Miller, 1992), reactions to tumor promoters (Cossu et al., 1988), and reactions to TGF-b (De-Cusella et al., 1994) have also been reported. The myoblasts found in postnatal muscle are termed myosatellite cells (Mauro, 1961). It has been suggested that these cells are of different lineage from myoblasts found in developing muscles (c.f., Cossu and Molinaro, 1987). These relatively quiescent, mononucleated cells found between the sarcolemma and basal lamina of postnatal muscle fibers are the source of additional myonuclei for growing muscle fibers. When muscle is injured, myosatellite cells are stimulated to proliferate; depending on the extent of the trauma, they may participate in myofiber repair or may fuse with each other to form new muscle fibers. In the chick, it has been suggested that populations of myosatellite cells differ depending on the type of fiber (i.e., fast versus slow muscle fibers) with which they are associated (Feldman and Stockdale, 1991; Hartley et Grant sponsor: National Institutes of Health; Grant number: AR36294; Grant sponsor: Wellcome Traveling Fellowship; Grant sponsor: EC Biotechnology; Grant number: PL 950228. *Correspondence to: Marcia Ontell, Department of Cell Biology and Physiology, University of Pittsburgh School of Medicine, South BST— Room 313, Pittsburgh, PA 15261. E-mail: email@example.com Received 22 July 1997; Accepted 24 October 1997 142 YANG ET AL. al., 1991; Bourke et al., 1995; McFarland et al., 1997). Recently, there has been interest in determining whether the myosatellite cells found in mature mammalian muscles also comprise a heterogeneous population. One way to study this in vivo is to evaluate regenerated myofibers that form de novo, by the fusion of myosatellite cells. If, upon differentiation, myosatellite cells are predetermined to accumulate fiber type specific isoforms of contractile proteins, it can be expected that regenerated myofibers will express the same contractile proteins as the original fibers with which the myosatellite cells were associated, even in the absence of innervation. When fast or slow muscles regenerate in situ, in the presence of their respective nerves, they accumulate transcripts for contractile protein mRNAs similar to those found in the original myofibers. However, in the absence of innervation, regenerated slow muscles accumulate similar transcripts (Esser et al., 1993) and proteins (Whalen et al., 1990) to those found in fast muscles. These regeneration studies suggest that myosatellite cells are not predetermined, at least in terms of contractile protein gene expression. In contrast, when head muscles of the cat, which express a superfast MHC, are minced and allowed to regenerate in the bed of a limb muscle that normally contains fibers with type IIb MHC, the regenerated muscle expresses only superfast MHC (Hoh and Hughes, 1991), suggesting that head muscle myosatellite cells are committed. Using retroviral lineage marked myosatellite cells injected into muscles of growing rats, Hughes and Blau (1992) observed that single clones of myoblasts can contribute to fibers of different types in vivo. They concluded that de novo myotube formation results in the expression of the MHC isoform of the muscle fibers from which the myosatellite cells were derived; when myosatellite cells fuse with existing myofibers, any intrinsic commitment of myosatellite cell nuclei to particular programs of gene expression is overridden by extrinsic signals (e.g., signals emanating from the endocrine and nervous systems or from other nuclei within the mosaic myofiber). By establishing cultures of myosatellite cells and determining the phenotype of the myotubes that they form upon differentiation, one can study the commitment of myosatellite cells in the absence of innervation and without the influence of endocrine factors. In differentiated cultures derived from cat jaw muscle, myotubes fail to expresses superfast MHC (Hill et al., 1989). Similarly, differentiated cultures of human muscle do not show diversity of myosatellite cells based on MHC isoform accumulation (Cantini et al., 1980; Cho et al., 1993; Mouly et al., 1993; Edom et al., 1994). However, Dusterhoft and Pette (1993) have demonstrated by immunocytochemistry that myotubes formed in vitro by rat myosatellite cells derived from the fast tibialis anterior muscle almost exclusively contain fast MHCs; 10% of the myotubes formed by myosatellite cells of the slow soleus muscle coexpress fast and slow MHC (with the remainder expressing only the fast isoform). In those myotubes coexpressing fast and slow isoforms, the slow myosin often accumulates around specific myonuclei, whereas the fast isoform is more widely distributed. This suggests that these fibers are formed from a heterogeneous population of myosatellite cells. Rosenblatt et al. (1996) evaluated myotubes formed in cultures derived from explants of single, adult mouse muscle fibers of known fiber type. They reported that MHCb/slow positive myotubes are more frequently found in cultures established from myosatellite cells associated with type I myofibers; however, a substantial number of myotubes established from myosatellite cells associated with type IIa and a lesser number of myotubes established from type IIb myofibers also contain this isoform. Thus, although there appears to be a tendency toward commitment of the myosatellite cells of rodent muscle related to the type of myofiber with which it is associated, all myosatellite cells associated with a given fiber type do not have the same commitment to the expression of particular MHCs. A more effective way of evaluating the diversity of the myosatellite cells in vitro is to allow these cells to differentiate as mononucleated myocytes. This will prevent possible alterations in the programmed phenotype of individual myosatellite cells by other myonuclei in multinucleated, mosaic myotubes. In the present study, myosatellite cell diversity has been evaluated using cultured primary myoblasts derived from the extensor digitorum longus muscle (EDL), a 99% fast twitch muscle in the adult mouse (Thomas et al., 1984), and the soleus muscle, a 40–60% slow twitch muscle of the adult mouse (Wirtz et al., 1983; Wigston and English, 1992). The cultures have been derived from both newborn and adult muscles of transgenic mice which express nuclear localizing b-galactosidase (nlsbgal) under the control of the myosin light chain 3F (MLC3F) promoter and 38 enhancer (Kelly et al., 1995). MLC3F is one of four alkali MLCs commonly found in rodent skeletal muscle and is preferentially associated with fast fibers in the adult mouse. The relative concentration of MLC3F is highest in type IIb, intermediate in type IIx (IId), and lowest in type IIa myofibers (Wada and Pette, 1993; Zardini and Parry, 1994). During in utero development, the transgene is activated in myotubes of hindlimb muscles from 11.5 days of embryonic development, 3.5 days before high level expression of the endogenous gene. However, during the fetal period and in the adult, transgene expression in striated muscle reflects the expression of the endogenous MLC3F gene (e.g., it is downregulated in adult muscles, such as the soleus muscle and the diaphragm, that have few type IIb fibers) (Kelly et al., 1995). We have determined the frequency of finding mononucleated myocytes with b-gal–positive myonuclei and the frequency of finding myotubes with b-gal–positive myonuclei in differentiated cultures of myosatellite cells derived from newborn and adult transgenic mice. These frequencies have been compared with the frequency with which myofibers with b-gal–positive nuclei are found in intact EDL MYOSATELLITE CELL COMMITMENT 143 2C,D). Quantitative analyses revealed that in the newborn EDL and soleus muscles and in the adult EDL, ,90% of the myofiber profiles that contained sublaminal nuclei had nuclei that contained b-gal (i.e., reacted with X-gal staining solution). The frequency of finding b-gal–positive myonuclei in the adult soleus muscle was less than one half (t-test; P , 0.001) of that observed in the adult EDL (Fig. 3). MLC3F lacZ Transgene Expression in Myocytes Derived From Newborn and Adult Mice Fig. 1. Whole mount, b-gal–stained EDL and soleus muscles of neonatal and adult MLC3F-nlslacZ-E transgenic mice. All muscles were exposed to the X-gal staining solution for 1.5 hr, the minimal time for reaction product to appear in the adult soleus muscle. The staining in the newborn soleus (Neo Sol) appeared to be slightly less intense than that of the newborn EDL (Neo EDL). The Neo EDL stained with the same intensity as the Adult EDL. Little reaction product is seen in the adult soleus (Adult Sol) muscle, even after prolonged staining. Scale bar 5 1 mm. and soleus muscles of the transgenic mouse. This has enabled us to determine the extent to which myosatellite cells are committed to a particular muscle phenotype as a result of the type of fiber with which they are associated. It has also permitted us to attempt to evaluate whether the commitment is present at birth (i.e., before the emergence of the muscle fiber’s mature phenotype) or whether it develops during the postnatal period. RESULTS MLC3F lacZ Transgene Expression in Newborn and Adult Whole Muscles Whole newborn and adult muscles of transgenic mice (MLC3F-nlacZ-E), which express nlsb-gal under the control of the MLC3F promoter and 38 enhancer (Kelly et al., 1995), were incubated with X-gal staining solution. The newborn and adult EDL and the newborn soleus muscle demonstrated reaction product ,20 min after exposure to X-gal staining solution; in contrast, the adult soleus muscle required 1.5 hr of exposure before any reaction product was detectable. The intensity of staining after 1.5 hr was similar for the neonatal and adult EDL. The neonatal soleus muscle was slightly less intensely stained than the EDL, and only minimal staining of the adult soleus muscle was detected (Fig. 1). Observations of transverse sections stained with X-gal suggested that there was a similar frequency of myofiber profiles (i.e., cross-sections of muscle fibers) containing b-gal–positive nuclei relative to the total population of fiber profiles containing sublaminal nuclei (myonuclei or myosatellite cell nuclei) in the neonatal EDL and soleus muscles (Fig. 2A,B); this was less in the adult soleus (Fig. 2E,F) than in the adult EDL (Fig. In primary cultures established from the adult (8 to 12 weeks old) and neonatal EDL and soleus muscles of the transgenic mice, ,1% of the myoblasts on plates exposed to proliferation media expressed the transgene. This percentage was similar to the percentage of cells on these plates that reacted with antibody to striated muscle MHC (not shown). For studies in which the myoblasts were allowed to differentiate as myocytes (mononucleated cells), replicate plates were seeded at low density. After 6 days in differentiation media, one plate was reacted with an antibody specific for skeletal muscle myosin (MF20; Bader et al., 1982) and with Hoechst 33258 (Fig. 4C,D) while the others were reacted with X-gal staining solution and Hoechst 33258 (Figs. 4A,B,E,F, 5). Myocytes with b-gal–positive myonuclei and myocytes whose nuclei failed to stain with X-gal were found in all differentiated cultures (Figs. 4A,E, 5A,C). The nuclei of b-gal–negative myocytes were readily visible when the cells were viewed with fluorescence optics that included a Hoechst cube (Figs. 4B,F, 5B,D). To correct for the possibility that myoblasts had not differentiated into myocytes after 6 days in differentiation media and to account for the few fibroblasts that were present in the culture, the myosin antibody stained plates were used to determine the percentage of mononucleated cells that had differentiated into myocytes (Fig. 4C,D). By determining the frequency of myocytes with b-gal– positive myonuclei versus the total number of mononucleated cells (b-gal–positive and b-gal–negative cells) and by correcting for the percentage of mononucleated cells that had not differentiated into myocytes, it was possible to calculate the ratio of myocytes that expressed, or failed to express, the transgene. Ninety percent of the myocytes cultured from the neonatal EDL were b-gal–positive, whereas only 70% of those from the newborn soleus were positive (t-test, P , 0.001) (Fig. 3). In contrast, in cultures of adult muscle, 61% of the myocytes found in EDL-derived cultures and 32% of the myocytes in soleus-derived cultures exhibited reaction product after X-gal staining (t-test, P , 0.001; Fig. 3). MLC3F lacZ Transgene Expression in Cultured Myotubes Derived From Newborn and Adult Muscles For studies in which the myosatellite cells were allowed to differentiate as myotubes (defined as MHC- 144 YANG ET AL. Fig. 2. Sections of neonatal (A,B) and adult (C–F) EDL (A,C,D) and soleus (B,E,F) muscles seen in Figure 1. There is no difference in the frequency of finding fibers with b-gal–positive nuclei (black nuclei) in the neonatal EDL (A) and soleus (B) muscles. Sections of the adult EDL (C) and soleus (E) muscle stained with X-gal are shown. Only the nuclei containing b-gal are visible with brightfield optics. There are significantly more fibers with b-gal–positive nuclei in the adult EDL (C) than in the adult soleus (E) muscles. Figures D and F are of the same sections seen in C and E, respectively, viewed with fluorescence optics. The white dots seen in D and F are nuclei stained with propidium iodide. The white arrows in C and E indicate myofibers that have nuclei that are not b-gal–positive. White arrows in D and F indicate the same fibers seen in C and E (arrows), respectively. Nuclei in these fibers are stained with propidium iodide (arrows, D and F), but these nuclei do not stain with X-gal (C and E). By comparing brightfield and fluorescence images and by focusing with phase optics (not shown), it is possible to determine the number of myofiber profiles containing b-gal–positive nuclei relative to the total populations of myofiber profiles that contain sublaminal nuclei. Scale bars 5 50 µm. MYOSATELLITE CELL COMMITMENT Fig. 3. Percentage of myofibers, myocytes, and myotubes with b-gal– positive nuclei. The percentage of fiber profiles with nuclei that contained b-gal in intact EDL (open bar) and soleus (filled bar) muscles of neonatal and adult MLC3F-nlslacZ mice, percentage of myocytes in differentiated cultures derived from myosatellite cells of the EDL and soleus muscles of neonatal and adult transgenic mice, and percentage of myotubes with b-gal-positive myonuclei in differentiated cultures derived from the myosatellite cells of the EDL and soleus muscles of newborn and adult transgenic mice are shown. Asterisks indicate significant differences at P , 0.001 between the EDL and soleus muscles of a given group. N, neonatal; A, adult. positive cells with two or more myonuclei), replicate plates were seeded at high density. After 6 days in differentiation media, in cultures derived from the neonatal EDL and soleus muscles (Fig. 6A,6B), 90% of the myotubes were b-gal–positive (Fig. 3). A similar percentage of the myotubes in adult EDL cultures (Fig. 6C) contained b-gal–positive myonuclei; only 55% of the myotubes derived from the adult soleus muscle (Fig. 6D) reacted with X-gal staining solution (Fig. 3). Endogenous MLC3F Expression in Cultured Myotubes Derived From Transgenic Mice Because there is some disagreement as to whether MLC3F is routinely produced in differentiated cultures of mammalian muscle (Rubinstein and Holtzer, 1979; Whalen et al., 1979; Caravatti et al., 1982; Cox et al., 1990; Edom et al., 1994) and some concern that the accumulation of muscle-specific proteins in differentiated cultures depends on culture conditions (Dusterhoft and Pette, 1993), the ability of differentiated myotubes in our system to accumulate MLC3F was evaluated with polyacrylamide gel electrophoresis (PAGE). MLC3F was found in differentiated cultures of the adult and neonatal EDL and soleus muscles grown under the same conditions as those used for the b-gal assay (Fig. 7). DISCUSSION This study establishes for the first time that there is diversity among adult mouse myosatellite cells in terms of determination of MLC gene expression, based on the 145 type of myofiber with which the myosatellite cells are associated. The present study evaluates the expression of MLC3F using primary cultures derived from muscles of transgenic mice which express nlsb-gal under the control of the MLC3F promoter and 38 enhancer (Kelly et al., 1995). (In intact muscles, the expression of this transgene mimics the expression of the endogenous MLC3F gene in adult mice [Kelly et al., 1995].) Heterogeneity of myosatellite cells in vitro is displayed in cultures derived from muscles of postnatal day 4 mice. This is before the time that the fibers of intact muscles express their mature phenotype. However, the neonatal myosatellite cells retain plasticity, so that with muscle maturation, some myosatellite cells (or their progeny) lose the ability to spontaneously express the MLC3F transgene when they undergo differentiation in vitro. The extent to which there is diversity among myosatellite cells is difficult to assess in cultures which differentiate as myotubes, because the phenotype of a mosaic myotube is likely to reflect that of the majority of myoblasts that participated in that myotube’s formation (Dusterhoft and Pette, 1993). By evaluating the expression of the MLC3F transgene in cultures in which myosatellite cells have differentiated as mononucleated myocytes, more accurate quantification of the percentage of myosatellite cells programmed to express the MLC3F transgene has been possible. The frequency with which myotubes that contain nuclei expressing this transgene are found in vitro has also been evaluated. In comparing the frequency with which cultured myocytes and myotubes express the transgene, it must be remembered that in multinucleated myotubes in vitro (as in multinucleated myofibers in the intact muscles), nlsb-gal is capable of migrating for distances .350 µm from the nucleus that encodes it to enter noncoding myonuclei (Yang et al., 1997). This would predict a higher frequency of myotube or myofiber profiles with b-gal–positive myonuclei than myocytes with b-gal–positive myonuclei. This is precisely the observation made in the present study. At 4 days postnatal, there are similar frequencies of myofiber profiles with b-gal–positive myonuclei in the intact EDL and soleus muscles of MLC3F-nlslacZ-2E mice. There are no published studies comparing the frequency of myofibers expressing MLC3F in these two muscles during the neonatal period. However, in a recent study (Washabaugh et al., in press), it has been determined by using competitive polymerase chain reaction (PCR) that the number of molecules of MLC3F per nanogram of total RNA is the same in the soleus and EDL muscles of the mouse at birth. The frequency of myofibers with b-gal–positive myonuclei decreases by ,50% with age in the soleus muscle, but there is no age-related effect on this frequency in the EDL muscle. This is in keeping with the alterations in muscle phenotype that occur as these two muscles mature. At birth, these muscles do not contain any adult fast MHCs, but at 4 days postnatal transcripts for MHCIIb— but not other adult fast MHC mRNAs—can be demon- 146 YANG ET AL. Fig. 4. Differentiated cultures of myosatellite cells derived from neonatal EDL (A–D) and soleus (E,F) muscles that contained myocytes. Plates have been exposed to either X-gal solution and Hoechst dye (A,B,E,F) or MF20 antibody that has been visualized with CY3-labeled second antibody (C) and Hoechst dye (D). A: There are many b-gal–positive myocytes (mononucleated cell) in the EDL derived cultures, and only a few b-gal–negative myocytes (arrow). The double arrowheads indicate myotubes, which have been excluded from the quantification. B: Same culture as seen in A. When a myonucleus contains significant amounts of product after X-gal staining, Hoechst staining of that myonucleus is quenched. The arrow indicates the nucleus of the myocyte seen in A (arrow) that lacked reaction product. By comparing figures such as A and B, it is possible to determine the relative number of myosatellite cells that expressed the transgene upon differentiation. C: All the differentiated myocytes in the culture react with MF20 antibody. D: Same culture as in C. Arrows indicate nuclei of mononucleated cells that failed to differentiate into myocytes (MF20-negative cells). By comparing paired micrographs such as C and D, it is possible to calculate the percentage of mononuclear cells that have differentiated into myocytes. These data have been used to correct the calculation of the frequency of myocytes having b-gal–positive nuclei. E: In cultures derived from the neonatal soleus muscle, the frequency of b-gal–negative myocytes (arrows) is greater than for similar EDL-derived cultures (A). F: Same culture as in E. Arrows indicate the same myocytes, with b-gal–negative myonuclei, as seen in E (arrows). Scale bar 5 50 µm. strated by in situ hybridization (Ontell, unpublished data). During the perinatal period, the EDL and soleus muscles contain a considerable amount of the develop- mental isoform MHCperinatal (MHCpn). It has been suggested that heterodimers consisting of MLC1F and MLC3F as well as homodimers of MLC3F are associ- MYOSATELLITE CELL COMMITMENT 147 Fig. 5. Differentiated cultures of myosatellite cells from the adult EDL (A,B) and soleus (C,D) that contain myocytes, stained with X-gal (A,C) and with Hoechst dye (B,D). Arrows in A and C indicate mononucleated cells whose nuclei fail to express b-gal. B and D are of the same cultures seen in A and C, respectively. The arrows (B and D) indicate nuclei of myocytes that were a-gal–negative in A and C (arrows). The double arrowheads in C and D indicate myotubes that were excluded from the quantification. The frequency of b-gal–positive myocytes in cultures established from adult muscles is less than for cultures of neonatal muscles (compare 5A and 4A, and 5C and 4E). Cultures of the adult soleus muscle (C) had a lower frequency of b-gal–positive myocytes than cultures of the adult EDL (A). Scale bar 5 50 µm. ated with this isoform (c.f., d’Albis and Butler-Browne, 1993). This would account for relatively high levels of expression of the transgene at 4 days postnatal. Although the MHCpn is lost in both muscles with age, the EDL undergoes a selective decrease in the percentage of fibers that contain MHCb/slow, and fast MHCs become the predominant isoforms (Whalen et al., 1984). In contrast, the decrease in MHCpn in the soleus muscle is accompanied by an increase in MHCb/slow and the appearance of MHCIIa. In adult muscle, MLC3F is preferentially associated with fibers containing MHCIIb and to a lesser extent with fibers containing MHCIIx (MHCIId) and MHCIIa (Wada and Pette, 1993; Zardini and Parry, 1994). Thus, the decreased frequency of myofiber profiles with b-gal–positive nuclei in the soleus muscle with age reflects the decreased accumulation of MLC3F in that muscle. In contrast, the higher percentage of fiber profiles containing b-gal– positive myonuclei in the adult EDL muscle is consistent with that muscle’s fast phenotype and its accumulation of MLC3F. The present study provides the first evaluation of the commitment of myosatellite cells associated with myofi- bers that have not achieved their adult phenotype. Evaluation of myocyte cultures derived from myosatellite cells of neonatal muscle clearly demonstrates that a significant proportion of the myosatellite cells in both the EDL and soleus muscles express the MLC3F transgene upon differentiation. However, the frequency is significantly higher in the neonatal EDL. How does this finding fit with the observation that the frequency of myofiber profiles with myonuclei which contain b-gal in the intact neonatal muscles is similar? One hypothesis is that the neonatal cultures are derived from a single source of myoblasts (myosatellite cells), whereas the intact muscle fibers contain nuclei derived from two or possibly three different populations of myoblasts (embryonic or fetal myoblasts 1 myosatellite cells, or a combination of all three types of myoblasts) with different commitments. An alternate hypothesis would be that a larger percentage of the myonuclei that contain b-gal in the intact neonatal EDL are nuclei that express this fusion protein, while many of the nuclei in the neonatal soleus muscle that are b-gal–positive are nonexpressing nuclei that have received nlsb-gal by translocation of the fusion protein or its message from the region of 148 YANG ET AL. Fig. 6. Differentiated cultures of myosatellite cells from neonatal EDL (A) and soleus (B) muscles and adult EDL (C) and soleus (D) muscles that contained myotubes. The majority of the myotubes in both of the cultures derived from neonatal muscles have b-gal–positive myonuclei (A,B). Fig. 7. PAGE of myosins extracted from differentiated cultures of the adult and neonatal EDL and soleus muscles (lanes 2–5). Myosin extracted from the intact EDL (lane 1) served as a standard for identification of the various fast MLC isoforms. Samples were of differentiated cultures (myotubes) derived from the myosatellite cells of the adult soleus muscle (lane 2), adult EDL (lane 3), neonatal soleus muscle (lane 4), and neonatal EDL (lane 5). All cultures accumulated MLC3F protein (arrows). the expressing nuclei (Yang et al., 1997). In the present study, data derived from cultures of myosatellite cells that differentiated as myotubes fail to support the first Arrows (A,B) indicate the few myotubes that do not have b-gal–positive myonuclei. The frequency of finding myotubes with b-gal–negative myonuclei (arrows in C and D) is higher in cultures derived from the adult soleus muscle (D) than from the adult EDL (C). Scale bar 5 50 µm. hypothesis, in that the frequency of finding myotubes with b-gal–positive myonuclei is similar in differentiated cultures of neonatal myosatellite cells and in muscle fibers of intact neonatal muscles, despite the differences in the populations of myoblasts that formed them. Supporting the second hypothesis is the observation that although both neonatal cultures that have differentiated as myocytes and those that have differentiated as myotubes have been established by the same population of myosatellite cells, the frequency of myotubes with b-gal–positive myonuclei is similar to the frequency of myocytes with b-gal–positive nuclei for cultures derived from the EDL but greater than the frequency of myocytes with b-gal–positive nuclei for cultures derived from the soleus muscle. Again, a major difference between the myocytes and the myotubes is the ability of b-gal or its mRNA to migrate from expressing to nonexpressing myonuclei in mosaic myotubes (Yang et al., 1997). Evaluation of myocyte cultures of adult muscles demonstrates a more striking diversity in the commitment of myosatellite cells derived from the soleus and EDL muscles to express the MLC3F transgene than is present in corresponding neonatal myocyte cultures. In cultures derived from neonatal muscles, only 10% of the myosatellite cells of the EDL and 30% of the myosatel- MYOSATELLITE CELL COMMITMENT lite cells of the soleus muscle fail to express the transgene upon differentiation. In contrast, 39% and 68% of the myosatellite cells of the adult EDL and soleus muscles, respectively, fail to express the MLC3F transgene upon differentiation. For the soleus muscle, the change in the frequency of myoblasts committed to express the MLC3F transgene which occurs with age roughly parallels the changes in frequencies of myofiber profiles with b-gal–positive myonuclei in the intact muscles (Fig. 3). For the EDL, the age-related decrease in the frequency of myosatellite cells committed to express the transgene occurs in the absence of any significant age-related change in the frequency of finding myofiber profiles with b-gal–positive myonuclei. Here, too, it could be postulated that fewer of the b-gal–positive myonuclei present in the adult EDL (compared with the neonatal EDL) are nuclei expressing the fusion protein, and that more of them receive b-gal by translocation of the fusion protein or its mRNA from expressing myonuclei in the same myofiber. Because myosatellite cells found in adult muscle are derived from the population of myosatellite cells found in the neonate, it is clear that any commitment that the myosatellite cells of the newborn may have to express MLC3F is plastic, and with muscle maturation that commitment is altered. The results of the cell culture experiments with adult muscles clearly indicate a marked heterogeneity of the myosatellite cells of adult muscles in terms of transgene expression. Given that the EDL muscle of the mouse accumulates significantly more MLC3F (a fast isoform) than does the soleus muscle and that the frequency of finding b-gal–positive myocytes in differentiated cultures of the EDL is twice that of the soleus muscle, the heterogeneity in the expression of the MLC3F transgene (which is expressed in adult muscle in a pattern similar to the endogenous gene [Kelly et al., 1995]) appears to be related to fiber type. To further evaluate the extent to which the adult myofiber of this transgenic mouse may imprint or change the commitment of its population of myosatellite cells, it will be necessary to establish cultures from single transgenic myofibers of known fiber type and to evaluate the myocytes that form when its myosatellite cells undergo differentiation (c.f., Rosenblatt et al., 1996). To understand the significance of heterogeneity among the myosatellite cells of the neonate in terms of the expression of the MLC3F transgene, it is important to understand the pattern of expression of the endogenous MLC3F during development. High levels of accumulation of MLC3F mRNA are detected with in situ hybridization in the developing crural muscles of the mouse before 14 days gestation (E14) (Ontell et al., 1993), at a time when primary myofibers (but not secondary myofibers) are present in crural muscles (Ontell and Kozeka, 1984; Ontell et al., 1988). At this stage, there is no preferential accumulation of this transcript among or within any of the crural muscles (Ontell et al., 1993). At E17, there is enhanced accumulation of the endogenous transcript in the more mature anterior compartment 149 muscles (including the EDL) than the less mature posterior compartment muscles (including the soleus muscle). However, at birth (with the maturation of the posterior compartment muscles), the intensities of signal for MLC3F mRNA in the two compartments are similar with in situ hybridization, and the numbers of molecules of MLC3F mRNA per nanogram of total RNA isolated from the EDL and soleus muscles are also similar (Washabaugh et al., in press). This indicates that expression of the endogenous gene before birth is not related to fiber type and that it is only during the postnatal period that the expression of this gene becomes fiber-type–specific. If this is true, then MLC3F can be regarded as a developmental isoform whose gene is actively transcribed in most (if not all) muscle fibers during in vivo development and whose transcription is downregulated postnatally in fibers that develop a ‘‘slower’’ phenotype. The diversity of myofibers in adult muscle, based on the accumulation of different isoforms of musclespecific proteins, adenosinetriphosphatase activity, and muscle contractile characteristics, is well established. In adult muscle it has been clearly demonstrated that the phenotype of a muscle fiber is plastic, in that it may be altered by changing the pattern of muscle activity (c.f., Pette and Vrbova, 1985). It would be of interest to determine the effect of changing a mature muscle’s fiber type (by altering muscle activity) on the population of myosatellite cells associated with that muscle. The results of the present study are consistent with the hypothesis that muscle phenotype correlates with myosatellite cell’s commitment to express specific isoforms of contractile proteins upon differentiation. As neonatal muscles alter their phenotypes toward the adult fiber type, the commitment of the myosatellite cells to express the MLC3F is altered, resulting in a selective loss of commitment of the myosatellite cells of slow muscle fibers to express the transgene. The mechanism by which muscle phenotype directly or indirectly influences commitment of myosatellite cells remains to be determined. EXPERIMENTAL PROCEDURES The EDL and soleus muscles were removed from newborn (2–4 days postnatal) and adult transgenic mice (MLC3F-nlacZ-E; killed by an anesthetic overdose) that express nlsb-gal under the control of a 2-kb MLC3F promoter and 38 enhancer (Kelly et al., 1995). Some muscles were used to determine the frequency of finding b-gal–positive nuclei in intact muscles; others served as the source of myosatellite cells for in vitro studies. Intact Muscles Newborn and adult EDL and soleus muscles (five muscles for each group) were fixed in 4% paraformaldehyde (1 hr at 4°C), washed twice in phosphate-buffered saline (PBS), incubated in a staining solution [2 mM MgCl2, 0.1% Tween-20, 5 mM K3Fe(CN)6, 5 mM 150 YANG ET AL. K4Fe(CN)6, and 0.1% 5-bromo-4-chloro-3-indolyl-bgalactopyranoside (X-gal) (which was dissolved in dimethyl sulfoxide)] at 37°C for 1.5 hr (Sanes et al., 1986), cryoprotected in 15% sucrose, and frozen in liquid-N2– cooled isopentane. Serial, transverse cryosections (10-µm thick) were restained (with X-gal solution) for 3 hr (to ensure that nuclei in the center of the section had sufficient exposure to the staining solution) and then stained with propidium iodide (to permit visualization of nuclei that failed to exhibit blue reaction product after exposure to X-gal staining solution). Sections was viewed with a Nikon FXA microscope (Tokyo, Japan) equipped with brightfield, phase, and fluorescence optics. One thousand muscle fiber profiles (i.e., crosssections of muscle fibers) observed in five different transverse sections per muscle were evaluated. Fiber profiles containing sublaminal nuclei (myonuclei or myosatellite cells) were scored according to whether the nuclei they contained were b-gal–positive or b-gal– negative, and the ratio of b-gal–positive myofibers to the total fiber population that contained sublaminal nuclei was determined. Evaluation of Primary Myosatellite Cell Cultures Adult and neonatal muscles were excised and transferred into dishes containing Ham’s F10 serum free medium. For cultures of adult muscles, the same muscles from both legs were pooled and minced. For cultures of neonatal muscles, all of the soleus muscles and all of the EDL muscles obtained from pups in a single litter (litter size, 6–12 pups) were separately pooled. Six cultures were evaluated for each group. Isolation of myosatellite cells and culture conditions were as described in Ontell et al. (1992). Briefly, muscles were incubated at 37°C for 20 min in 0.025% collagenase in Puck’s saline G and centrifuged (1000 rpm 3 5 min). The pellet was resuspended in 10 ml of complete medium (Ham’s F10 supplemented with 15% v/v donor horse serum) and seeded into 100-mm gelatincoated dishes. Bovine basic fibroblast growth factor (b-FGF) was added (4 ng/ml, twice daily), and the dishes were incubated at 37°C in a water-saturated atmosphere containing 5% CO2. After 48 hr, the medium containing the suspended muscle debris was removed and centrifuged, and the pellet was resuspended in 10 ml of complete medium. After preplating the suspended fragments for 20 min, they were transferred into a 100-mm gelatin-coated dish. (Preplating enhances the ratio of myoblasts to fibroblasts in primary culture [Richler and Yaffe, 1970].) Basic-FGF (4 ng/ml) was added to the cultures twice a day. Fortyeight hours after transfer, the cultures were refed with complete medium plus b-FGF. On day 3, cultures were briefly (,10–20 sec) treated with 0.025% collagenase at 37°C, a procedure that favors the release of myoblasts while leaving fibroblasts still attached to the plate. Cells were centrifuged and resuspended in complete medium. Three sets of plates were established from the suspension, with the following purposes. 1. To confirm that proliferating myoblasts derived from MLC3F-nlacZ-E mice did not express the transgene in our culture system. To achieve this goal, myosatellite cells were plated at a density of 1 3 105 cells per 60-mm dish and grown for 24–48 hr in complete medium in the presence of b-FGF. Plates were then fixed in either 2% paraformaldehyde or AFA fix (63% ethanol, 3.2% formaldehyde, and 750 mM acetic acid). Paraformaldehyde fixed plates were stained with X-gal for 3 hr at 37°C. Cells on AFA fixed plates were permeabilized with 0.1% Triton, nonspecifically blocked by 1% bovine serum albumin (BSA) in PBS at 37°C for 30 min, and incubated at room temperature with MF20 (an antibody to skeletal muscle myosin; Bader et al., 1982) for 1 hr, followed by an incubation with goat anti-mouse-IgG CY3 conjugate. By staining with MF20, it was possible to determine the percentage (if any) of myosatellite cells that differentiated into myocytes under our ‘‘proliferating’’ conditions. If the expression of the transgene required differentiation, only those cells that expressed MHC would be expected to be b-gal– positive. 2. To evaluate myosatellite cell commitment to a specific phenotype in cultures of differentiated, mononucleated myocytes. To favor the formation of differentiated mononucleated cells (myocytes), the plates were seeded at low density (1 3 104 cells per 60-mm dish) and grown overnight in complete medium. Cultures were then switched to differentiation medium (Ham’s F10/5% horse serum with 1.9 mg/ml of insulin) and fed once every other day for 6 days. Plates were fixed, treated as described above, and reacted with Hoechst 33258 to facilitate counting of all mononucleated cell on the plate. Plates were observed with a Nikon FXA microscope equipped for brightfield and fluorescence microscopy. For data collection, 20 different fields for each plate were randomly chosen. From the MF20 reacted plates, the percentage of mononucleated cells that had not become differentiated myocytes (under the differentiation conditions described above) relative to the total number of mononucleated cell on the plates was determined. From the plates stained with X-gal, the percentage of myocytes with b-gal–positive nuclei relative to the total number of mononucleated cells on the plate was evaluated. (The few myotubes present on these plate were not included in any evaluations.) Because not all myoblasts had undergone differentiation (as judged from MHC immunostaining) and because a few fibroblasts were found in each plate, it was necessary to correct for the percentage of undifferentiated cells to determine what percentage of myosatellite cells expressed the transgene upon differentiation. 151 MYOSATELLITE CELL COMMITMENT 3. To evaluate myosatellite cell commitment to a specific phenotype in cultures of differentiated, multinucleated myotubes. Cells were seeded at a density of 1 3 105 cells per 60-mm dish and grown for 48 hr in complete medium. They were then switched to differentiation medium for 6 days, fixed, and stained as described above. For data collection, 20 different fields on each plate were randomly chosen. The percentage of myotubes (defined as cells with two or more nuclei) containing ‘‘blue’’ myonuclei relative to the total myotube population was determined. Representative micrographs of sectioned and cultured material were obtained with a Nikon FXA microscope equipped with an Optronics CCD camera (Goleta, CA). This camera was linked to a computer that had Optimas (Seattle, WA) image analysis software. Statistical Analysis Analysis of variance (ANOVA) and t-test were used for statistical analysis. Differences were considered significant at P , 0.05. Accumulation of MLC3F in Cultured Myotubes Polyacrylamide gel electrophoresis (PAGE) was used to evaluate the accumulation of the MLC3F protein in cultures that differentiated as myotubes. Cultured cells were rinsed with PBS and scraped into a myosin extraction buffer (300 mM NaCl, 100 mM NaH2PO4, 50 mM Na2HPO4, 1 mM MgCl2, 10 mM Na4P2O7, 10 mM EDTA, and 0.1% b-mercaptoethanol, pH 6.5) and frozen at 280°C. Cells were thawed, homogenized with a polytron homogenizer on ice, and allowed to incubate on ice in a centrifuge tube for 40 min to extract the myosin. Tubes were centrifuged for 30 min at 4°C at 14,000 3 g. The supernatant was diluted 1:10 in a low salt buffer (1 mM EDTA and 0.1% b-mercaptoethanol), and the myosin filaments were allowed to precipitate overnight at 4°C. Filaments were pelleted at 14,000 3 g for 30 min, resuspend in myosin sample buffer (500 mM NaCl and 10 mM NaH2PO4, pH 7.0), and allowed to incubate overnight at 4°C. The extract was diluted with an equal volume of sodium dodecyl sulfate (SDS) sample buffer (62.5 mM Tris-HCl, 2% SDS, 30% glycerol, 0.001% bromophenol blue, and 5% b-mercaptoethanol, pH 6.8), boiled for 2 min, and stored at 280°C. A separating gel consisted of 12.5% acrylamide, and a stacking gel of 5% acrylamide was used. Approximately 15 µg of protein was loaded in each lane of the gel. Electrophoresis was performed at constant current of 20 mA through the stacking gel and at 25 mA through separating gel. The gel was then silver-stained (Oakley et al., 1980), scanned, and processed with an Adobe Photoshop program. ACKNOWLEDGMENTS This study was supported by NIH grant AR36294 (to M.O.). R.K. was supported initially by a Wellcome Traveling Fellowship and subsequently by EC Biotech- nology Grant PL 950228. The authors thank Dr. Simon Watkins, Structural Biology Imaging Center, for his help. REFERENCES Bader D, Masaki T, Fischman D. Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro. J. Cell Biol. 1982;95:763–770. Bonner PH, Hauschka SD. Clonal analysis of vertebrate myogenesis. I. Early developmental events in the chick limb. Dev. Biol. 1974;37: 317–328. Bourke DL, Wylie SR, Theon A, Bandman E. Myosin heavy chain expression following myoblast transfer into regenerating chick muscle. Basic Appl. Myol. 1995;5:43–56. Cantini M, Sartore S, Schiaffino S. Myosin types in cultured muscle cells. J. Cell Biol. 1980;85:903–909. Caravatti M, Minty A, Robert B, Montarras D, Weydart A, Cohen A, Daubas P, Buckingham M. Regulation of muscle gene expression: The accumulation of messenger RNAs coding for muscle-specific proteins during myogenesis in a mouse cell line. J. Mol. Biol. 1982;160:59–76. Cho M, Webster SG, Blau HM. Evidence for myoblast-extrinsic regulation of slow myosin heavy chain expression during muscle fiber formation in embryonic development. J. Cell Biol. 1993;121:795– 810. Cossu G, Molinaro M. Cell heterogeneity in the myogenic lineage. Curr. Top. Dev. Biol. 1987;23:185–208. Cossu G, Ranaldi G, Senni MI, Molinaro M, Vivarelli E. ‘Early’ mammalian myoblasts are resistant to phorbol ester-induced block of differentiation. Development 1988;102:65–69. Cox RD, Garner I, Buckingham ME. Transcriptional regulation of actin and myosin genes during differentiation of a mouse muscle cell line. Differentiation 1990;43:183–191. d’Albis A, Butler-Browne G. The hormonal control of myosin isoform expression in skeletal muscle of mammals: A review. Basic Appl. Myol. 1993;3:7–16. De-Cusella Angelis MG, Molinari S, Le DA, Coletta M, Vivarelli E, Bouche M, Molinaro M, Ferrari S, Cossu G. Differential response of embryonic and fetal myoblasts to TGFb: A possible regulatory mechanism of skeletal muscle histogenesis. Development 1994;120: 925–933. Dusterhoft S, Pette D. Satellite cells from slow rat muscle express slow myosin under appropriate culture conditions. Differentiation 1993; 53:25–33. Edom F, Mouly V, Barbet JP, Fiszman MY, Butler-Browne GS. Clones of human satellite cells can express in vitro both fast and slow myosin heavy chains. Dev. Biol. 1994;164:219–229. Esser K, Gunning P, Hardeman E. Nerve-dependent and -independent patterns of mRNA expression in regenerating skeletal muscle. Dev. Biol. 1993;159:173–183. Feldman JL, Stockdale FE. Skeletal muscle satellite cell diversity: Satellite cells form fibers of different types in cell culture. Dev. Biol. 1991;143:320–334. Hartley RS, Bandman E, Yablonka-Reuveni Z. Myoblasts from fetal and adult skeletal muscle regulate myosin expression differently. Dev. Biol. 1991;148:249–260. Hill MA, Ecob-Prince MS, Hoh JF. Regeneration of cat posterior temporalis muscle in culture. Cell Diff. Dev. 1989;28:145–151. Hoh JF, Hughes S. Expression of superfast myosin in aneural regenerates of cat jaw muscle. Muscle Nerve 1991;14:316–325. Hughes SM, Blau H.M. Muscle fiber pattern is independent of cell lineage in postnatal rodent development. Cell 1992;68:659–671. Kelly R, Alonso S, Tajbakhsh S, Cossu G, Buckingham M. Myosin light chain 3F regulatory sequences confer regionalized cardiac and skeletal muscle expression in transgenic mice. J. Cell Biol. 1995;129: 383–396. Mauro A. Satellite cells of skeletal muscle fibers. J. Biophys. Biochem. Cytol. 1961;9:493–498. McFarland DC, Gilkerson KK, Pesall JE, Wellenreiter RH, Ferrin NH, Ye WV, Yun Y, Vander Wall LS. Comparison of growth factor receptors and metabolic characteristics of satellite cells derived 152 YANG ET AL. from the biceps femoris and pectoralis major muscles of the turkey. Gen. Comp. Endocrinol. 1997;105:114–120. Miller JB. Myoblast diversity in skeletal myogenesis: How much and to what end? Cell 1992;69:1–3. Miller JB, Crow MT, Stockdale FE. Slow and fast myosin heavy chain content defines three types of myotubes in early muscle cell cultures. J. Cell Biol. 1985;101:1643–1650. Mouly V, Edom F, Barbet JP, Butler-Browne GS. Plasticity of human satellite cells. Neuromuscul. Disord. 1993;3:371–377. Narusawa M, Fitzsimons RB, Izumo S, Nadal-Ginard B, Rubinstein NA, Kelly AM. Slow myosin in developing rat skeletal muscle. J. Cell Biol. 1987;104:447–459. Oakley BR, Kirsch DR, Morris NR. A simplified ultrasensitive silver stain for detecting proteins in polyacrylamide gels. Anal. Biochem. 1980;105:361–363. Ontell M, Kozeka K. Organogenesis of the mouse extensor digitorum longus muscle: A quantitative study. Am. J. Anat. 1984;171:149–161. Ontell M, Hughes D, Bourke D. Morphometric analysis of the developing mouse soleus muscle. Am. J. Anat. 1988;181:279–288. Ontell MP, Hughes D, Hauschka S, Ontell M. Transient neonatal denervation alters the proliferative capacity of myosatellite cells in dystrophic (129 ReJ dydy) muscle. J. Neurobiol. 1992;23:407–419. Ontell M, Ontell MP, Sopper MM, Mallonga R, Lyons G, Buckingham M. Contractile protein gene expression in primary myotubes of embryonic mouse hindlimb muscles. Development 1993;117:1435– 1444. Pette D, Vrbova G. Neural control of phenotypic expression in mammalian muscle fibers. Muscle Nerve 1985;8:676–689. Richler C, Yaffe D. The in vitro cultivation and differentiation capacities of myogenic cell lines. Dev. Biol. 1970;23:1–22. Rosenblatt JD, Parry DJ, Partridge TA. Phenotype of adult mouse muscle myoblasts reflects their fiber type of origin. Differentiation 1996;60:39–45. Rubinstein NA, Holtzer H. Fast and slow muscles in tissue culture synthesis only fast myosin. Nature 1979;280:323–325. Sanes JR, Rubenstein JL, Nicolas JF. Use of a recombinant retrovirus to study post-implantation cell lineage in mouse embryos. EMBO J. 1986;5:3133–3142. Smith T, Miller J. Distinct myogenic programs of embryonic an fetal mouse muscle cells: Expression of the perinatal myosin heavy chain isoform in vitro. Dev. Biol. 1992;149:16–26. Stockdale FE. Myogenic cell lineages. Dev. Biol. 1992;154:284–298. Thomas D, Kleuber K, Bourke D, Ontell M. The size of the myofibers in mature grafts of the mouse extensor digitorum longus muscle. Muscle Nerve 1984;7:226–231. Vivarelli E, Brown WE, Whalen R, Cossu G. The expression of slow myosin during mammalian somitogenesis and limb bud differentiation. J. Cell. Biol. 1988;107:2191–2197. Wada M, Pette D. Relationships between alkali light-chain complement and myosin heavy-chain isoforms in single fast-twitch fibers of rat and rabbit. Eur. J. Biochem. 1993;214:157–161. Washabaugh CH, Ontell MP, Shau Z, Hoffman EP, Ontell M. The role of the nerve in determining fetal skeletal muscle phenotype. Dev. Dyn. 1998;211:177–190. Whalen RG, Butler-Browne GS, Sell S, Gros F. Transitions in contractile protein isozymes during muscle cell differentiation. Biochimie 1979;61:625–632. Whalen RG, Johnstone D, Bryers PS, Butler-Browne GS, Ecob MS, Jaros E. A developmentally regulated disappearance of slow myosin in fast-type muscles of the mouse. FEBS Lett. 1984;177:51–56. Whalen RG, Harris JB, Butler-Browne GS, Sesodia S. Expression of myosin isoforms during notexin-induced regeneration of rat soleus muscles. Dev. Biol. 1990;141:24–40. Wigston DJ, English AW. Fiber-type proportions in mammalian soleus muscle during postnatal development. J. Neurobiol. 1992;23:61–70. Wirtz P, Loermans HMT, Peer PGM, Reinjes AGM. Postnatal growth and differentiation of muscle fibers in the mouse. I. A histochemical and morphological investigation of normal muscle. J. Anat. 1983;137: 109–126. Yang J, Ontell MP, Kelly R, Watkins SC, Ontell M. Limitations of nls beta-galactosidase as a marker for studying myogenic lineage or the efficacy of myoblast transfer. Anat. Rec. 1997;248:40–50. Zardini DM, Parry DJ. Identification, distribution, and myosin subunit composition of type IIX fibers in mouse muscles. Muscle Nerve 1994;17:1308–1316.