Satellite cells from dystrophic (Mdx) mice display accelerated differentiation in primary cultures and in isolated myofibers


  • Zipora Yablonka-Reuveni,

    1. Department of Biological Structure, School of Medicine, University of Washington, Seattle, Washington
    Search for more papers by this author
  • Judy E. Anderson

    Corresponding author
    1. Department of Human Anatomy and Cell Science, Faculty of Medicine, University of Manitoba, Winnipeg, Canada
    • Department of Human Anatomy and Cell Science, Faculty of Medicine, University of Manitoba, 730 William Avenue, Winnipeg, MB, R3E 0W3, Canada
    Search for more papers by this author


In the dystrophic (mdx) mouse, skeletal muscle undergoes cycles of degeneration and regeneration, and myogenic progenitors (satellite cells) show ongoing proliferation and differentiation at a time when counterpart cells in normal healthy muscle enter quiescence. However, it remains unclear whether this enhanced satellite cell activity is triggered solely by the muscle environment or is also governed by factors inherent in satellite cells. To obtain a better picture of myogenesis in dystrophic muscle, a direct cell-by-cell analysis was performed to compare satellite cell dynamics from mdx and normal (C57Bl/10) mice in two cell culture models. In one model, the kinetics of satellite cell differentiation was quantified in primary cell cultures from diaphragm and limb muscles by immunodetection of MyoD, myogenin, and MEF2. In mdx cell cultures, myogenin protein was expressed earlier than normal and was followed more rapidly by dual myogenin/MEF2A expression and myotube formation. In the second model, the dynamics of satellite cell myogenesis were investigated in cultured myofibers isolated from flexor digitorum brevis (FDB) muscle, which retain satellite cells in the native position. Consistent with primary cultures, satellite cells in mdx myofibers displayed earlier myogenin expression, as well as an enhanced number of myogenin-expressing satellite cells per myofiber compared to normal. The addition of fibroblast growth factor 2 (FGF2) led to an increase in the number of satellite cells expressing myogenin in normal and mdx myofibers. However, the extent of the FGF effect was more robust in mdx myofibers. Notably, many myonuclei in mdx myofibers were centralized, evidence of segmental regeneration; all central nuclei and many peripheral nuclei in mdx myofibers were positive for MEF2A. Results indicated that myogenic cells in dystrophic muscle display accelerated differentiation. Furthermore, the study demonstrated that FDB myofibers are an excellent model of the in vivo state of muscle, as they accurately represented the dystrophic phenotype. Developmental Dynamics 235:203–212, 2006. © 2005 Wiley-Liss, Inc.


The interplay between proliferation and differentiation during myogenesis is tightly regulated by the helix-loop-helix family of muscle-specific transcription factors. The four members of this family (Myf5, MyoD, myogenin, and MRF4) are often referred to as myogenic regulatory factors (MRFs). MyoD is expressed by proliferating myoblasts, while the onset of myogenin expression marks the entry of myoblasts into differentiation. Myogenin expression is rapidly followed by terminal withdrawal from the cell cycle and expression of muscle-specific structural proteins such as sarcomeric myosin (Yablonka-Reuveni and Rivera 1994, 1997a; Ludolph and Konieczny, 1995; Andres and Walsh, 1996).

The myocyte enhancer factor 2 (MEF2) family of transcription factors also plays a pivotal regulatory role in myogenesis. This family includes at least 4 proteins, MEF2 A–D. Proliferating myoblasts express MEF2 at low levels. As myogenic differentiation begins, MEF2 expression is up-regulated, coinciding with or shortly after the onset of myogenin expression (Cserjesi and Olson, 1991; Naya and Olson, 1999; McKinsey et al., 2000; Yablonka-Reuveni and Rivera, 1997a, b). While the expression of MEF2 is not limited to skeletal muscle, members of the MEF2 family interact in skeletal muscle with MRFs to regulate various critical phases in the progression of myogenesis. These interactions lead to increased expression of muscle structural genes during myoblast differentiation, muscle hypertrophy, electrical stimulation, and other stressful stimuli (Naya and Olson, 1999; McKinsey et al., 2002a, b). Multiple pathways converge on MEF2A to regulate the function of that gene, and interaction of MEF2 and MRF proteins can be uncoupled from activation of transcription (Puri and Sartorelli, 2000). MEF2 involvement in various cell lineages complicates the interpretation of changes in intramuscular MEF2-expression level during muscle regeneration in vivo, such as reported using RT-PCR (Sakuma et al., 2005). In a disease state like muscular dystrophy, the outcome of adaptations to changes in activity level, inflammation, and growth signals are collectively represented in muscle tissue. The same muscle histology section will demonstrate necrotic and apoptotic cell death, and myogenic cell proliferation and differentiation during repair (Garrett and Anderson, 1995; Anderson et al. 1998a, b; McIntosh et al., 1998). Therefore, the effect of pathology on MEF2 expression is particularly difficult to anticipate without cell-by-cell analysis of MEF2 in the context of MRF gene expression. The activity and location of myofiber nuclei also change during disease, as exemplified by the central nucleation that characterizes longitudinal segments of myofibers that have regenerated following focal damage in rat, mouse, and human muscle (Karpati et al., 1988). In such regenerated segments, myonuclei that fused with the myofiber during regeneration are centrally located, while satellite cell nuclei will be visible at the myofiber periphery within the layer of basal lamina. However, there is little information to distinguish gene expression specifically in nuclei of regenerated segments.

Satellite cells are the myogenic progenitors in adult muscle, situated on the myofiber surface. In healthy adult muscle, these cells are mostly quiescent but can enter a proliferative state upon muscle trauma (Hawke and Garry, 2001). In the dystrophic (mdx) mouse model of human Duchenne muscular dystrophy, satellite cells are thought to be active and participate in ongoing myofiber regeneration (Carnwath and Shotton, 1987; Sabourin and Rudnicki, 2000; Anderson, 1998; Anderson and Wozniak, 2004). However, it has remained unclear if the performance of satellite cells from mdx mouse muscle is inherently different from those that originate from normal wild type mouse muscle. Experiments presented in the current study were designed to characterize differentiation kinetics of satellite cells from mdx mice as the cells undergo myogenesis in two culture models. In one set of studies, cells were isolated from limb and diaphragm muscles of normal and mdx dystrophic mice and examined at various intervals after plating, for the timing and co-expression of molecular (MyoD, myogenin, MEF2A) and phenotypic (myotube formation) markers of myogenic differentiation. Quantification of cells expressing these markers was previously demonstrated to provide an excellent means for analyzing the myogenic potential of cultured skeletal muscle cells (Yablonka-Reuveni and Rivera, 1997a; Kastner et al., 2000; Yablonka-Reuveni and Paterson, 2001). In a second study, single muscle myofiber cultures were prepared from normal and mdx mice to analyze the kinetics of satellite cell differentiation in their response to FGF2, a mitogen known to affect myogenesis. Single myofiber cultures from wild-type and mutant rodent muscle were used previously to model gene expression patterns and response to growth factors during myogenesis of satellite cells as characterized by the satellite cell nuclear expression of distinct protein markers associated with proliferation and differentiation (Yablonka-Reuveni and Rivera, 1997b; Yablonka-Reuveni et al. 1999a, b).

Both cell culture systems investigated in the present study demonstrated an accelerated differentiation of satellite cells from mdx mice, suggesting that some features of myogenic differentiation are cell-autonomous. The myofiber culture model additionally demonstrated the novel observation of strong MEF2 expression in central myofiber nuclei contained within mdx regenerated myofiber segments. This distinctive characteristic of the mdx myofibers, not previously investigated in culture, likely reflects ongoing differentiation events in vivo, related to segmental dystrophy and repair. We conclude that the mouse myofiber culture model permits insight into cellular and molecular events associated with muscular dystrophy.


Primary Cell Cultures

Figure 1 shows representative micrographs of cultures of dissociated myogenic cells from the diaphragm of mdx mice at 3 and 6 days following initial culturing. Figure 1 consists of 4 columns, each of 3 panels, representing cultures that were immunostained with antibodies against MyoD and MEF2A (columns A and C; culture days 3 and 6, respectively) or myogenin and MEF2A (columns B and D; culture days 3 and 6, respectively), and counterstained with DAPI. Similar immunostaining patterns were observed for cultures from mdx limb muscles and normal diaphragm and limb counterparts (data not shown). Overall, primary cell cultures from normal and dystrophic muscles showed the typical progression of robust proliferation to differentiation and fusion into thick, branching myotubes, despite being maintained in serum-rich growth medium. The progression was noted by the appearance of MyoD+ cells first, followed in time by the development of myogenin+ cells and the formation of myotubes. However, as detailed below, the kinetics of progression into the myogenin+ state (and subsequent fusion into myotubes) differed for mdx versus normal primary cell cultures.

Figure 1.

Myogenic cultures isolated from diaphragm muscle of mdx mice at day 3 (columns A and B) and day 6 (columns C and D) following cell culturing. Cultures were maintained in high-serum growth medium and allowed to undergo spontaneous differentiation. Cultures were immunostained for the expression of MyoD and MEF2A (A,C) or myogenin and MEF2A (B,D) and counterstained with DAPI to show nuclei. Mdx cultures were typically more confluent after plating at equivalent density and contained myotubes 2–3 days earlier than normal cell cultures. In dispersed cell cultures, the expression of myogenin and MEF2A was observed in both individual cells and in myotube nuclei. Magnification = 400×.

Cultures were examined for the transition from proliferative, MyoD+ myoblasts to differentiated MyoD+/MEF2A+ myogenic cells. A progressive decrease in the proportion of mononucleated cells expressing MyoD alone was displayed in both C57 and mdx cultures. In C57 diaphragm cell cultures (Fig. 2A), about 20% of “differentiated” myogenic cells (defined in the Experimental Procedures section as the sum of cells that were identified in the myogenic program by their expressing a MRF protein plus the number of cell nuclei contained in myotubes) were MyoD+/MEF2A+ by 5–6 days; myotubes contained about 2.5% of nuclei. In contrast, in mdx cultures, the transition from MyoD+ to MyoD+/MEF2A+ expression was 1–2 days earlier; by day 6 in culture, there were approximately 30% of nuclei in myotubes and more than double the proportion of MyoD+/MEF2A+ cells compared to primary cultures from control C57 mice. There were very few cells (<1%) that expressed MEF2A protein without MyoD protein in either C57 or mdx primary cultures. These data indicate that the decline in the number of cells that were positive only for MyoD expression, the transition to MEF2A expression, and myotube formation all were initiated earlier in mdx than C57 primary cultures.

Figure 2.

Stacked histograms of the distribution of cells expressing MyoD and MEF2A (A) and myogenin and MEF2A (B) in primary cell cultures isolated from normal and mdx dystrophic diaphragm, and cultured for up to 10 days. The numbers of cells expressing MyoD and MEF2A (A) and myogenin and MEF2A (B), and the number of myotube nuclei were counted and expressed as proportions (percent) of the total number of “differentiated” myogenic cells in culture. Cells stained positive for MyoD or myogenin without MEF2A (black bars) or with MEF2A (white bars) and the proportion of nuclei located in myotubes (dotted bars) are represented. The x-axes are labeled to indicate the mouse strain (C57 or mdx), days in culture (top number), and the total number of “differentiated” myogenic cells (bottom number, averaged for 10 microscopic fields). The total number of “differentiated” myogenic cells was defined as a term to include all cells that could be identified in the myogenic program, based on the expression of a MRF protein and nuclei in myotubes. The number of “differentiated” cells was calculated as the sum of mononuclear cells with nuclei stained positive for the respective muscle regulatory factor (MRF+/MEF2A− plus MRF+/ MEF2A+) plus the number of nuclei in myotubes (see Experimental Procedures section for further details).

Immunolocalization studies of myogenin and MEF2A were used to follow the further progress of mononuclear cell differentiation and myotube formation. As detailed in the Introduction section, as differentiation begins, MEF2 expression is up-regulated, coincident with or shortly after the onset of myogenin expression. Cultures from C57 muscles showed nuclei that expressed myogenin together with MEF2A, starting at day 6 in the diaphragm cultures with intense staining for myogenin (images not shown). Similar timing of MEF2A expression was observed in soleus, TA, and quadriceps primary cultures (days 6, 8, and 8 respectively, data not shown). Primary cultures from mdx muscle showed myogenic nuclei with dual myogenin+ and MEF2A+ expression starting much earlier than in C57 primary cultures, at day 3 (diaphragm) or day 3 and 4 (soleus and TA, not shown), and increasing over time in culture. In mdx cultures, there was earlier expression of myogenin beginning at 3 days in culture and more nuclei within myotubes. There was also notably more robust proliferation in mdx muscle cell cultures than normal, making mdx cell cultures more confluent than C57 cell cultures.

Collectively, these data show that progressive expression of two differentiation markers, myogenin and MEF2A, and, ultimately, myotube formation, were more rapid in primary cultures isolated from the diaphragm muscle of mdx compared to C57 mice. Comparison of the frequency distributions of the three types of myogenic cells (myogenin+/MEF2A−, myogenin+/MEF2A+, myogenin−/MEF2A+) between primary cell cultures of diaphragm muscle cells from normal and mdx mice showed significantly higher proportions of myogenin+/MEF2A+ cells in mdx cultures at day 6 (P < 0.001, Chi-squared statistics).

Myofiber Cultures

Figure 3 shows representative FDB myofibers from normal C57 and mdx dystrophic mice. Characteristic of normal myofibers (Fig. 3D), myonuclei were located at the periphery of the long myofiber cylinder, seen in profile (at the projected edge of a myofiber) or in multiple focal planes through the myofibers. This appearance of myofiber nuclei means, practically, that in normal myofibers, satellite cell nuclei and myonuclei are not distinguishable using DAPI alone, before activation and differentiation of satellite cells. After activation, satellite cells become positive for MyoD, then expressed myogenin and MEF2A.

Figure 3.

Single myofibers prepared from FDB muscles of mdx dystrophic mice (columns A–C) and normal C57 mice (D) that were maintained in culture for 2 days (A–C) or 1 day (D). Myogenin and MEF2A proteins were detected by immunostaining. Myofibers were then counterstained in DAPI for nuclei. Magnification = 400×. Straight arrowheads point to the position of MyoD+ or myogenin+ cells that are negative for MEF2A; concave arrowheads point to the position of myogenin+ cells that are also positive for MEF2A; arrows point to central nuclei, either single nuclei or nuclei placed in variable-length rows that mark previously regenerated segments of a myofiber, all of which are positive for MEF2A. Additional immuno-positive cells on myofibers and nuclei within myofibers are not visible, since they lie outside the focal plane of the micrographs.

In beautiful contrast, many mdx myofibers demonstrated long, segmentally-displayed chains or rows of DAPI-stained myonuclei that were displaced to the central region, aligned as a core of the myofiber (Fig. 3A–C). In those segments with long rows of central nuclei, the peripheral, “side-lined” or “flanking” nuclei that stained with DAPI were identified as satellite cells, based on previous characterization of muscle regeneration (Grounds and Yablonka-Reuveni, 1993; Karpati et al., 1988; Anderson, 1998). Satellite cells at the periphery of dystrophic myofiber segments also progressed to express MyoD, myogenin, and MEF2A over time in culture. In addition, mdx myofibers showed various combinations of regions with and without central nuclei, reflecting the focal nature of segmental damage and regeneration processes that characterize the progression of dystrophy. Both the central myonuclei and the nuclei of flanking satellite cells on the myofiber segments displaying central nuclei were consistently observed to stain positive for MEF2A in mdx myofibers between 1 and 5 days in culture. Peripheral myonuclei in myofiber segments without central nuclei were variably observed to be MEF2A+ or MEF2A−. No other epitope was similarly identified in both the central myonuclei of mdx myofibers and the nuclei of satellite cells flanking the mdx myofibers. MyoD and myogenin proteins were not detected in central myonuclei. MEF2A staining was much more intense and more homogeneous in mdx satellite cell nuclei compared to the punctate staining pattern in satellite cell nuclei on normal myofibers.

The distributions of numbers of myogenin+ satellite cells per myofiber from C57 and mdx myofiber cultures are shown in Figure 4A and B, respectively. Higher numbers of myogenin+ satellite cells per myofiber were consistently observed on mdx myofibers than for normal myofibers, and mdx myofibers showed an earlier appearance of myogenin+ satellite cells than C57 myofibers. The number of myofibers with zero myogenin+ satellite cells dropped rapidly in mdx myofibers over time in culture compared to a more gradual decline in C57 myofibers. With added FGF2, C57 and mdx myofibers showed a shift to higher numbers of myogenin+ satellite cells per myofiber from days 1 to 5 in culture; this was especially apparent from culture day 3 onward and more robust in mdx myofibers. Notably, nearly 60% of mdx myofibers showed more than 4 myogenin+ satellite cells at day 4, compared to 30% in C57 myofiber cultures. Overall, in both C57 and mdx myofibers, the average numbers of myogenin+ satellite cells per myofiber typically increased until day 3 and then plateaued to a mean value (for days 3,4, and 5) as follows: C57 without FGF, 1.5 ± 0.2; C57 with FGF, 2.6 ± 0.2; mdx without FGF 2.5 ± 0.2; and mdx with FGF, 3.6 ± 0.1.

Figure 4.

The distributions of satellite cells stained positive for myogenin expression during myogenesis on isolated normal (C57) myofibers (A) and dystrophic (mdx) myofibers (B) in FDB cultures. Distributions were compiled from counting stained cells on a total of 58–60 myofibers per time point (i.e., 28–30 per time point in each experiment from at least 2 experiments). Cultures were maintained in media without FGF2 (front row of bars) or with 2 ng/ml FGF2 (second row of bars) for 1–5 days.


A better understanding of the distinctive muscle regeneration capacity in mdx mice can provide important insight regarding the pathology of muscular dystrophy. In the present study, we used single-cell analysis techniques to compare the differentiation cascade through the MyoD-myogenin-MEF2A interval by satellite cells from mdx and wild type mice. Experiments employed muscle cell cultures, and also myofiber cultures where the interactions of the satellite cells with respective “parent” myofibers are maintained. Myofiber cultures from dystrophic muscles showed a shift toward more rapid differentiation, possibly due to their in vivo history of disease and their activation status (Anderson, 1998) that would lead to ongoing rounds of proliferation and differentiation. Furthermore, cultures from mdx mice showed earlier differentiation compared to control C57 cultures, seen as an early decline in expression of MyoD (in cell cultures) and a proportionate early increase in myogenin expression (in satellite cells on cultured myofibers). Exogenous FGF2 had the typical effect to increase numbers of myogenin+ satellite cells on normal C57 and especially on dystrophic myofiber cultures as previously described by us for myofiber cultures from wild type rodents (Yablonka-Reuveni and Rivera, 1994; Yablonka-Reuveni et al., 1999a; Kastner et al., 2000). In myofiber cultures, many of the myogenin+ satellite cells were also MEF2A+. Notably, in mdx myofiber cultures, MEF2A was also localized in myonuclei, especially in central nuclei of regenerated segments. Collectively, results of the present study are interpreted as evidence of distinctive kinetics of proliferation and differentiation in myogenic cells and in post-mitotic myonuclei in regenerated mdx dystrophic mouse muscle.

This study traced the expression of MyoD, myogenin, and MEF2A, genes known to regulate myogenic differentiation. Cell cultures from mdx muscles showed earlier myogenin expression, a phenotype that is generally accepted to indicate the onset of differentiation in the myogenic lineage, and also showed earlier transition to co-expression of myogenin and MEF2A than normal. This transition to the myogenin+/MEF2A+ state and the formation of myotubes were observed 3 days earlier in primary cell cultures from mdx diaphragm muscle (and also from TA, quadriceps, and soleus; not shown) compared to normal primary cell cultures, and notably occurred under conditions of serum-rich medium. This accelerated differentiation in primary cultures from mdx mouse muscle was not simply due to accelerated proliferation, because even those mdx cultures started at 2–4-fold lower cell density showed more rapid differentiation than observed for normal cell cultures. Observations of high proliferation were reported for myogenic cultures from dystrophic chickens (Johnson et al., 1983). The present findings were interpreted to indicate that the transition was regulated independently of the external proliferative signals from a high serum concentration, since both control and mdx cultures were maintained in the same serum-rich medium, while the enhanced differentiation was seen only in the mdx cultures. This finding is consistent with a similar observation of early myogenin expression in satellite cells on mdx myofibers compared to normal myofibers, when myofibers were maintained without added FGF2. While 10% of mdx FDB myofibers showed 1 or more myogenin+ satellite cells by the first day after plating, satellite cells on normal C57 myofibers did not show myogenin expression until day 2. Furthermore, by day 2 there were many more myogenin+ satellite cells on myofibers from mdx mice. Altogether, experiments with both cell and myofiber cultures indicated a rapid differentiation program in mdx muscle preparations compared to normal.

The myofiber culture study offers an insight into a unique interplay between the satellite cells and the myofiber in mdx muscle. The early appearance of myogenin+ satellite cells on mdx myofibers, the prolonged plateau of myofibers with 2 or more myogenin+ satellite cells after day 3, and the high number of myogenin+ satellite cells at day 4 and 5 in cultures receiving FGF all suggest that in mdx myofiber cultures, there is a continuous influx of satellite cells into the myogenin+ cell pool. Recent studies have demonstrated that satellite cells express Pax7 and that some satellite cells in certain muscles may show a differential expression of Pax3 (Seale et al., 2000; Shefer et al., 2004; Relaix et al., 2005; Kassar-Duchossoy et al., 2005; Shefer and Yablonka-Reuveni, unpublished data). Hence, future studies on the distribution of cell expression Pax3 and Pax7 in isolated myofibers may enable us to further investigate the initial number and type of satellite cells in dystrophic and normal muscles as well as the proportion of satellite cells that respond to FGF.

Our novel finding of MEF2A expression in central nuclei of isolated myofibers also prompts future in vivo experiments aimed at the regulation and impact of prolonged MEF2A expression in myofiber nuclei. Central nuclei are stable over the lifetime of a muscle myofiber, enabling their use as an index of muscular dystrophy (Karpati et al., 1988). Routine histology cannot distinguish whether central nuclei within isolated myofibers have been derived from recent or previous regeneration events. Comprehensive studies on tissue sections would be required to determine whether central-nuclear expression of MEF2 varies with the progression of mdx mouse muscular dystrophy. The cellular origin of enhanced MEF2 transcriptional activity in regenerating muscle (Akkila et al., 1997) was not identified. In addition, it remains to be determined whether regeneration itself, in normal or mdx muscle, rather than the absence of dystrophin per se, may induce sustained or new transient ME2A expression in satellite cell nuclei that fuse into myofibers during regeneration.

Our preliminary observations on myofibers stained for PCNA suggest that satellite cell proliferation on mdx myofibers is not accelerated; the initial number of PCNA+ cells on myofibers from mdx and control mice is similar on day 1. Rather, proliferation on mdx myofibers varies or “pulses” over time and persists, while in C57 myofibers satellite cells proliferate all at once, followed by a rapid exit from the PCNA+ state, and enter the myogenin+ state. This “all at once” proliferation was previously described by us for satellite cells in various studies of isolated myofibers from growing and adult rodents (Yablonka-Reuveni and Rivera, 1994, 1999a, b; Yablonka-Reuveni and Rivera, 1997a; Kastner et al., 2000). The unique dynamics of myogenic differentiation by satellite cells in cultures of isolated mdx myofibers may be governed by the parent myofiber and could be related to structural changes in dystrophic mdx myofibers that slow the transition to a more mature state. Indeed, the immaturity of the entire myofiber unit was demonstrated in the present study by enhanced expression of MEF2 and the presence of central nuclei in mdx myofibers. The activation of satellite cells in vivo was previously demonstrated to change in dystrophin-deficient mdx muscle, as a result of a loss of nitric oxide signaling from the sub-sarcolemmal cytoskeleton of dystrophic myofibers underlying the satellite cells (Anderson, 2000). In contrast to the unique proliferative dynamics of satellite cells on mdx compared to control myofibers, dissociation of satellite cells from the muscle tissue and their subsequent maintenance in primary cultures would be anticipated to remove the myofiber-dependent features regulating satellite cell proliferation. Indeed, in this study, separation of satellite cells from myofibers and their maintenance in primary culture conditions abolished the differential myofiber-dependent regulation of proliferation; primary cultures derived from mdx muscle proliferated more extensively than normal C57 cultures, as seen by the higher density of mdx dispersed cell cultures. These studies suggest that progeny of satellite cells from mdx muscle are more active with respect to cell proliferation, whereas the myofiber niche exerts a more precise control on satellite cell proliferation, possibly due to regulators of cell cycle progression or cell surface proteins that are expressed by the myofibers. Together, current and previous reports show that satellite cells on mdx dystrophic myofibers are activated, albeit differently from normal, appear to proliferate after the same interval as satellite cells on normal myofibers, and then differentiate rapidly.

Central myonuclei and the MEF2 expression by such central myonuclei appear to be correlated with the distinctive myofiber-dependent regulation of satellite cell differentiation on dystrophic myofibers. MEF2A expression by central nuclei may also contribute to establishing the nature of mdx muscle plasticity during regeneration, given the regulatory role of MEF2 in myogenesis, hypertrophy, activity, or stress responses (McKinsey et al., 2002a, b). It is interesting that atypical proliferation and defects in differentiation have been reported in cultures of muscle cells isolated from patients with Duchenne or congenital muscular dystrophy (Miike, 1983; Jasmin et al., 1984; Lucas-Heron et al., 1989, 1994; Oexle and Kohlschutter, 2001).

It has been established already that development of the dystrophin-associated cytoskeleton during myogenic regeneration has differential effects on cell-cell signaling between myofibers and satellite cells in dystrophic muscle (Kong and Anderson, 2001; Rando, 2001; Anderson and Vargas, 2003; Anderson and Wozniak, 2004; Wozniak et al., 2005). Here, the single myofiber system displayed a microcosm of in vivo events as reported for mdx mouse muscular dystrophy (Zacharias and Anderson, 1991; McIntosh et al., 1994; Pernitsky et al., 1996; Anderson et al., 1998a; Anderson, 2000): accelerated differentiation represented by earlier myogenin+ cells; more proliferation represented by the marked persistence of responsiveness to FGF2 in later myofiber cultures; and evidence for marked central nucleation in myofibers, distributed in longitudinal myofiber segments or along whole myofibers, depending on prior history of damage and regeneration in the particular myofiber under examination. FDB cultures, therefore, provide an excellent model to study the regulation of satellite cell dynamics, myogenesis, and the activity of myofiber nuclei in response to various stimuli and disease processes.

In conclusion, we report the acceleration of differentiation in satellite cells from mdx mice, and the ongoing expression of MEF2 in mdx myofiber nuclei that was especially prominent in the central myonuclei that are the hallmark signs of previous regeneration in dystrophic myofibers. This study demonstrates that satellite cells retain in culture systems their inherent differentiation behavior in vivo during the rounds of degeneration and regeneration. Results suggest the possibility of a cell-autonomous, myofiber-independent control of the differentiation sequence of proliferating satellite cells, in addition to the myofiber-dependent processes that regulate satellite cell activation and proliferation.



Normal control mice (C57Bl/10, here referred to as C57) and mdx mice were purchased from Jackson Laboratory (Bar Harbor, ME). Cultures were prepared from muscles of young adult (8–11-week-old) mice. Animal care and experimental procedures were approved by the Institutional Animal Care and Use Committees at the University of Manitoba and the University of Washington.

Primary Cell Cultures

Myogenic cultures were prepared in parallel from mdx and C57 mice, using 3–6 mice per isolation. Cells were isolated from diaphragm, soleus, tibialis anterior (TA), and quadriceps muscles after careful dissection of the muscles to minimize connective tissue contribution. Collected muscles, processed separately by source muscle, were enzymatically digested, and released single cells were cultured in parallel, as reported previously (Yablonka-Reuveni et al., 1999a). Isolated cells were re-suspended from a pellet into serum-rich growth medium consisting of Dulbecco's minimum essential medium (DMEM) supplemented with 25% fetal bovine serum (Hyclone, Logan, UT), 10% horse serum (Hyclone), 1% chicken embryo extract, and antibiotics. Sera were pre-selected based on their capacity to support growth of low-density cultures and chicken embryo extract was prepared from 10–11-day-old embryos as previously described (Shefer and Yablonka-Reuveni, 2005). Cells were plated at a density of 105 cells per plate using 35-mm plates, pre-coated with 2% gelatin as previously described (Yablonka-Reuveni, 2004). Cultures were maintained in a standard tissue culture, with fresh growth medium replaced following the first 3 days in culture and every 2 days thereafter, and were harvested at different time points, as specified in Results. Cultures are typically prepared with this protocol to enrich the myogenic cell population, as reported (Yablonka-Reuveni et al., 1999a).

Cultures of Isolated Muscle Myofibers

Single muscle myofibers were isolated from the flexor digitorum brevis muscle (FDB) following our previously reported protocol for young adult mice (Yablonka-Reuveni et al., 1999a). Muscles from both hind feet of 3–6 mice were used for each experiment. FDBs were dissected cleanly from the foot to remove connective tissues and neurovascular bundles, and incubated at 37°C for 2.5 hr in 0.2% collagenase type 1 (Sigma-Aldrich). Single myofibers were released by gentle trituration and purified by three rounds of settling through columns of 10 cc DMEM containing 10% horse serum. Following the final settling, myofibers were dispensed as 50-μl aliquots into 35-mm plates coated with 120 μl of Vitrogen 100 (Collagen Corp., Freemont, CA) in DMEM. Plated myofibers attached to the Vitrogen as it gelled in the tissue culture incubator (for 15–20 min at 37°C) and then received 1 ml of basal medium consisting of 20% controlled-process serum replacement (CPSR2, Sigma, St. Louis, MO; bovine plasma stripped by charcoal and dialyzed to yield a serum-rich, lipid-poor, mitogen-poor product; further details in Shefer and Yablonka-Reuveni, 2005), 1% horse serum, and antibiotics. Parallel dishes were supplemented with fibroblast growth factor 2 (FGF2, 2 ng/ml, kindly provided by Dr. S. Hauschka, University of Washington) or left untreated for controls. Cultures were maintained for 1–5 days with fresh medium (± FGF) replaced every 24 hr.


Single and double immunofluorescence were carried out on methanol-fixed cultures of primary cultures or isolated myofibers as previously described (Yablonka-Reuveni et al., 1999a; Graves and Yablonka-Reuveni, 2000). The following previously characterized primary antibodies were used: mouse anti-MyoD (Mab 5.8A, IgG fraction diluted 1:100, gift from Drs. P. Dias and P. Houghton, St. Jude Children's Research Hospital, Memphis TN; Yablonka-Reuveni et al., 1999a; Yablonka-Reuveni, 2004); mouse anti-myogenin (hybridoma supernatant diluted 1:10, clone F5D, a gift of Dr. W. Wright, University of Texas; Yablonka-Reuveni and Rivera, 1994; Yablonka-Reuveni et al., 1999a; Yablonka-Reuveni, 2004); and rabbit anti-MEF2A (1:100 dilution, Santa Cruz Biotechnology, clone C-21; Molkentin et al., 1996; Yablonka-Reuveni and Rivera, 1997a; Kastner et al., 2000; Yablonka-Reuveni and Paterson, 2001). Secondary antibodies were fluorescein-conjugated goat anti-mouse IgG and rhodamine-conjugated goat anti-rabbit IgG (each diluted 1:1,000, from Organon-Technika Cappel, Downington PA). DAPI (4,6-diamidino-2-phenylindole, 1 μg/ml) was used to counterstain nuclei in cell and myofiber cultures.

In some studies of isolated myofiber cultures, to verify that nuclei that stained positively for the nuclear antigens (i.e., MyoD or myogenin) were in cells associated with the myofiber (i.e., satellite cells) and were not myonuclei, a few parallel cultures were dually reacted with the antibody against the nuclear antigen and a rabbit antibody against the mitogen-activated protein kinases (MAPK) ERK1/ERK2 (serum, 1:1,000 dilution; a gift from Drs. R. Seger and E. Krebs, University of Washington). As we previously reported (Shefer and Yablonka-Reuveni, 2005; Yablonka-Reuveni et al., 1999a, b), the strong immunolabeling of satellite cell cytoplasm with the anti-MAPK facilitates a clear distinction between nuclei within satellite cells (well-demarcated from myofibers, by a distinctively-labeled MAPK+ cytoplasm) and myofiber nuclei (which are not). Notably, MAPK-immunolabeling was only used in the present study for demarcating satellite cells to determine which nuclear entity was labeled by myogenin, and not for enumerating satellite cells per individual myofibers. We previously established that only differentiating satellite cells, and not myofiber nuclei, are positive for myogenin in cultured FDB myofibers. Nevertheless, it was important to revisit this topic in the present study since it was possible that myonuclei in dystrophic myofibers may express myogenin. Indeed, dual immunolabeling for myogenin and MAPK indicated that only nuclei in myofiber-associated cells (i.e., satellite cells) and not myofiber nuclei were labeled for myogenin in both mdx and C57 myofibers.

Quantification of Positively-Stained Cells in Primary Cultures and Isolated Myofiber Cultures

Cultures were observed and photographed under epifluorescence and phase contrast optics on a Nikon Optiphot 2 fluorescence microscope.

Each culture plate of dispersed cells was analyzed by separate counts of: (1) the total number of cells (i.e., all the DAPI-stained nuclei); (2) the number of cells positive for MyoD or myogenin; (3) the number of cells double-positive for MEF2A and a MRF protein (MyoD or myogenin); and (4) the number of nuclei contained in myotubes. The total number of “differentiated” myogenic cells was defined as a term that included any cell that could be identified in the myogenic program, based on the specific markers used for the immunostaining, plus nuclei in myotubes. This method of analysis was selected for the highly myogenic cultures to avoid inclusion of possible non-myogenic cells; the presence of such cells in primary myogenic cultures from adult muscle cannot completely be avoided and these non-myogenic cells could not be distinguished from cells that had not yet become identifiable with the markers used for enumeration of myogenic progeny (Yablonka-Reuveni, 2004). The number of “differentiated” cells was calculated as the sum of mononuclear cells with nuclei stained positive for a MRF gene (MRF+/MEF2A− cells plus MRF+/MEF2A+ cells) plus the number of nuclei in myotubes. Cells positive only for MEF2A were rare and were not included in the analysis as their myogenic identity is unclear. Typically, the density of cultures was such that, depending on culture day, 20–40 arbitrarily-chosen microscopic fields were counted using a 40× objective; data were then averaged for each phenotypic subpopulation based on expression pattern, and standardized to a common number of fields per each plate (i.e., per 10 fields), as reported (Yablonka-Reuveni and Rivera, 1997a; Kastner et al., 2000; Yablonka-Reuveni, 2004). Experiments were repeated at least twice.

Myofiber cultures were quantified for the number of myogenin+ cells. At least two plates were used for each time point within the same experiments, and experiments were repeated at least twice. Labeled satellite cells were counted as the number of positively-stained cells on each of at least 28–30 myofibers, randomly selected per plate. The proportions of myofibers demonstrating 0, 1, 2, 3, 4, or more than 4 (>4) satellite cells stained positive for myogenin were graphed as a frequency distribution for cultures with and without exogenous FGF2, as a function of the total number of myofibers analyzed per time point, for each day in culture.

In both culture models, results report analysis of frequency distributions of single- or double-positive cell staining as a function of time (days) in culture using Chi-squared tests, and a probability of P < 0.05 to determine statistical significance.


The authors thank Antony Rivera and Priscilla Natanson (University of Washington) for expert technical assistance and Dr. Gabi Shefer (University of Washington) for helpful input to the manuscript. The work was supported by grants from National Institutes of Health (AG 13798, AG 21566; Z.Y.R.), USDA Cooperative State Research, Education, and Extension Service (NRI, 99-35206-7934; Z.Y.R.), the Muscular Dystrophy Association (J.E.A.), and the Manitoba Institute of Child Health–Children's Hospital Foundation (J.E.A.).