INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

THE MYOGENIC REGULATORY FACTORS (MRFs) MyoD (2), myogenin (7), Myf-5, and MRF-4 (17), which is also known as Herculin or Myf-6, are basic helix-loop-helix (bHLH) proteins that are known to have vital roles in the regulation of muscle gene transcription during the differentiation process of developing skeletal muscle (5, 20).

During muscle development, each of the MRF family has a distinct temporal pattern of expression (3). The phenotype of mice carrying mutations in either one MRF or combinations of genes have revealed redundant and essential functions of the MRFs (2).

Postnatally, MyoD, myogenin, and Myf-5 continue to be expressed, although at low levels (8, 11, 19), whereas MRF-4 is present at relatively high levels in adult muscle (17, 19). The persistent expression of these transcription factors in the adult animal suggests that they may play a functionally significant role postnatally. Hughes et al. (11) observed differential patterns of expression of two of these factors, MyoD and myogenin, between rat muscles of different fiber type compositions. The slow oxidative soleus muscle was observed to express high levels of myogenin, whereas MyoD transcripts were found to predominate in fast-twitch muscles. These workers also noted that the distribution of MyoD and myogenin transcripts differed within a single muscle, with myogenin preferentially accumulating in slow regions and MyoD accumulation appearing highest in the faster regions that contain IIx and/or IIb fibers. Furthermore, it was also observed that an alteration of the fast/slow fiber type distribution, by cross-reinnervation or thyroid hormone treatment, resulted in a corresponding alteration in the pattern of expression of the MyoD and myogenin transcripts.

Subsequently, studies have suggested that the relationship between the expression of these MRFs in skeletal muscle and the phenotype of their component fiber populations may not be so straightforward (12). Further studies by Hughes and co-workers (10) showed that MyoD expression is required for the maintenance of normal fiber type balance in muscles; the absence of this factor in mice in which the MyoD gene has been disrupted leads to a shift in fiber type composition of muscles. In fast muscles, this shift is toward a slower phenotype but, surprisingly, to a faster phenotype in the slow soleus muscle.

Most studies to date that have examined the potential role of MRFs in postnatal skeletal muscle have examined MyoD and myogenin. Despite its considerably higher level of expression in the adult, MRF-4 has been examined in only a few studies. It has been suggested that whereas myogenin and MyoD may play a regulatory role in muscle fiber phenotype, MRF-4 probably only has a noninstructive role in the maintenance of the differentiated state (10). This passive role has been proposed, because, unlike MyoD and myogenin, MRF-4 transcript levels, when examined at the whole muscle level, have been shown not to differ considerably between muscles. It has, therefore, been assumed that there is little difference between fiber types, with respect to MRF-4 expression, within muscles. In this study, we have examined expression levels at the cellular level and have shown this assumption to be incorrect. We demonstrated that not only does MRF-4 exhibit a highly fiber type-dependent pattern of expression but that this pattern itself depends on the muscle in which it is expressed.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Six-week-old rats were killed by cervical dislocation, and the plantar group of muscles (gastrocnemius, plantaris, and soleus) were dissected, mounted on cork, covered in OCT compound (BDH), coated with talcum powder (to facilitate rapid freezing), and plunged into liquid nitrogen. Serial sections (10 µm) were cut on a cryostat and mounted on 3-aminopropyltriethoxy-silane-treated slides. The sections were treated with proteinase K, fixed in paraformaldehyde, and dehydrated through an ascending series of alcohols.

MRF-4 nucleotide probes were synthesized using the MRF-4 (17) and myogenin (7) cDNAs as templates for T3 and T7 RNA polymerase enzymes (Boehringer Mannheim RNA labeling kit) to generate sense and anti-sense digoxigenin (Dig)-labeled riboprobes. Serial sections were incubated with a prehybridization buffer composed of 4× sodium chloride-sodium citrate (SSC) and 50% formamide for 30 min at 37°C. Five to ten nanograms of either the sense or anti-sense Dig-labeled probe in hybridization buffer (40% formamide, 10% dextran sulfate, 1× Denhardt's solution, 4× SSC, 10 mM DTT, 1 mg/ml yeast tRNA, 1 mg/ml denatured salmon sperm) was applied to the sections. Hybridization took place overnight at 42°C in a sealed box.

After several stringency washes (2× SSC, 1× SSC, 0.1× SSC) at 37°C, unbound RNA probe was removed by incubating the sections with a Tris-EDTA (pH 8.0) buffer containing 20 µg/ml RNase A. After washing the sections with Tris-buffered saline (TBS), the sections were incubated with an alkaline phosphatase conjugated anti-Dig secondary antibody (Boehringer Mannheim) for 2 h, washed in TBS (pH 9.5), and incubated overnight with a nitroblue tetrazolium substrate (Sigma). The sections were mounted with an aqueous mounting medium and visualized.

Adjacent sections were incubated overnight at 4°C with a monoclonal anti-slow myosin heavy chain (MHC) antibody, generously given to us by Dr. Dhoot (6). The antibody was diluted 1:100 in 1% BSA in PBS. After washing the sections with PBS, the sections were incubated with an anti-mouse biotinylated secondary antibody diluted 1:200 in BSA for 2-3 h at room temperature. After several washes in PBS, the sections were incubated with an avidin-horseradish peroxidase conjugate diluted 1:200 for 30 min. A diaminobenzidine substrate was then applied, and the localization of slow MHC was visualized.

The relative levels of MRF-4 transcripts in type I fibers of the soleus and gastrocnemius muscles were compared. This was carried out on 20 fibers from each muscle by densitometry using a Kontron image analysis system (KS300).

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

In this study, the cellular expression of the MRF-4 was examined in the slow postural soleus and the mixed gastrocnemius muscles. Armstrong and Phelps (1) described the regional changes in fiber type composition within the rat gastrocnemius muscle, where type I (slow oxidative) fiber composition varied from 0% in the superficial or white region of this muscle to 30% in the deep or red region. In contrast, no regionalization in the smaller soleus muscle was observed, which had a mean composition of 87% of type I fibers with the remainder being of the type IIa (fast oxidative glycolytic) fiber type (1).

The accumulation of MRF-4 transcripts at the individual fiber level, as determined by in situ hybridization, shows a marked difference between the two muscles studied. A coherent pattern was observed in animals studied, and representative sections are shown in Fig. 1. The control sense probe exhibited no differential expression (Fig. 2A) In the gastrocnemius, the MRF-4 transcript was preferentially expressed in fibers expressing slow MHC, present only in the deeper parts of this muscle (Fig. 1, A, B, E, and F). In contrast, in the soleus, which is composed almost exclusively of type I and type IIa fibers, there was no difference in expression between subpopulations of fibers (Fig. 1, C and D). The fiber type-specific pattern of expression observed in the gastrocnemius with the MRF-4 probe was not observed when a myogenin probe was used (Fig. 2B). Furthermore, the mean levels of expression of the MRF-4 transcript in type I fibers in the soleus were 32% lower than in the type I fibers present in the gastrocnemius (P < 0.001, Student's t-test).

View larger version (140K): [in this window] [in a new window]   Fig. 1.   Distribution of slow myosin heavy chain (MHC)-expressing fibers (darkly staining) in gastrocnemius (g) and soleus (s) muscles are shown in A and C and serial sections hybridized with myogenic regulatory factor (MRF)-4 digoxigenin-labeled antisense probe are shown in B and D. E and F show the gastrocnemius under higher power, demonstrating that those fibers that express higher levels of MRF-4 (F) exactly correlate with those expressing slow MHC (E). Same 2 fibers have been marked (x and y) on both serial sections to facilitate orientation. Scale bars on A-D represent 250 µm and in E and F represent 100 µm.

View larger version (60K): [in this window] [in a new window]   Fig. 2.   A: section of gastrocnemius muscle hybridized with MRF-4 digoxigenin-labeled sense probe. Scale bar represents 250 µm. B: section of gastrocnemius muscle hybridized with myogenin digoxigenin-labeled antisense probe. Scale bar represents 100 µm.

It was observed in previous studies that MyoD and myogenin transcripts also demonstrate regional distribution at the cellular level, which correlates with fiber type composition, but the discrete fiber specificity observed in this study has not been described for any MRF. The supposed lack of muscle specificity of MRF-4, elucidated from transcript analysis carried out at the whole muscle level, has been the reason for suggesting that this bHLH protein, unlike MyoD and myogenin, may play little or no part in the regulation of muscle fiber phenotype (10). Our data, in contrast, suggest that in a mixed muscle, such as the gastrocnemius, MRF-4 exhibits a high degree of fiber type specificity at the mRNA level (Fig. 1, E and F). It is also present in the postnatal animal at much higher levels than any other MRF transcript (19). These two factors together suggest that MRF-4 may play a significant role in postnatal muscle, possibly in the regulation of fiber phenotype.

The link between MRF expression level and muscle fiber phenotype does not, however, appear to be simple. It has been shown previously that the absence of MyoD has different and opposite effects on phenotype in fast and slow muscles (10). This may be due to the fact that fibers from different muscles that are categorized as the same "type" due to histochemical or immunohistochemical analysis (on the basis of the predominant MHC composition) have their phenotypes regulated by different pathways. The data in the present study would support this hypothesis. Slow fibers in the gastrocnemius express high levels of MRF-4 compared with other fiber types, whereas the same "fiber type" in the soleus, where it is predominant, shows no such elevation.

The difference between the same fiber type in different muscles may be due to different development pathways or neural activity patterns. Extrinsic factors such as the neurotrophins in the local environment may also play a role (18). The animals used in this study were actively growing, which may have contributed to the pattern of MRF-4 transcript expression observed. At birth, the gastrocnemius is more differentiated than the soleus muscle (21). There is also evidence that different fiber types mature at different rates within the same muscle (21). It is possible that the type I fibers are at different stages of maturation in the two muscles examined in this study and that the different patterns of MRF-4 expression are a reflection of this. Narusawa et al. (16), however, examined the effects of denervation on slow fibers in the soleus and fast extensor digitorum longus (EDL) muscles of the postnatal rat and concluded that "type I fibers in the soleus and EDL are not equivalent." Further work is needed to establish the ontogenic pattern for MRF-4 transcript expression in muscle fibers of the postnatal animal.

Evidence for a neural role in a muscle-specific regulation of fiber phenotype is suggested from previous studies where we have shown that reduced levels of activity in immobilized rat plantaris and soleus muscles produce dramatic changes in MHC gene expression that are markedly different in these two muscles (14). Interestingly, the same model also has a distinctly different effect on the levels of MRF-4 transcripts in these two muscles, which also could suggest a possible role for this MRF in the regulation of muscle phenotype (13).

Perspectives