Background
Like many muscle diseases, Duchenne muscular dystrophy (DMD) is characterised by a gradual loss of muscle function with age. Patients are initially ambulatory and have mild muscle pathology, despite ongoing degeneration and repair. In later stages of DMD, patients experience progressively more severe muscular changes, accompanied by loss of function, physical dependency, and ultimately, death [
1].
DMD patients generally lack the cytoskeletal protein dystrophin, a member of the spectrin-like superfamily of actin binding proteins. Functional dystrophin is localised to the inner face of the sarcolemma and binds to cytoskeletal F-actin and transmembrane beta-dystroglycan as part of multiprotein complex that mediates signalling between the cytoskeleton and the extracellular matrix. The consequences of lack of dystrophin appear to be an enhanced susceptibility to fibre damage and possibly poor signalling between fibres and their environment. A milder form of dystrophin deficiency in humans is the Becker-Kiener type of muscular dystrophy (BMD). Here, dystrophin is not completely absent, but mutations lead to a quantitatively and/or qualitatively reduced gene product which does not accomplish its full function. The onset of BMD is generally later than DMD. In childhood, symptoms are usually very mild and muscle weakness becomes more evident only in the teens or twenties. BMD is non-lethal and patients often achieve normal life span, although the disease can progress in later life [
2].
To date, it is not clear what controls disease progression in either DMD or BMD and no consensus has been reached in the literature. The progressive loss of muscle in DMD and other muscle disorders could be due to a sustained or increasing rate of degeneration above the rate of repair, or a progressive decline in the ability to regenerate the muscle. Some pathological changes appear to be similar to those observed in healthily ageing people, yet premature and exacerbated. A popular, yet challenged view is that in DMD, disease progression is attributable to accumulating deficiencies in the ability of satellite cells resident within the muscle to mediate regeneration and/or their own replacement [
3]. Satellite cell deficiencies could arise because of the excessive demands on repair mechanisms necessitated by the continuous degeneration of unstable muscle that does not express dystrophin. Indeed, symptoms of rapid early turnover of muscle fibre material and cells are apparent before birth in DMD patients [
4], yet serious functional deficits arise only late in the first decade.
A commonly used model for studies of DMD is the
mdx mouse. These animals, like human patients, show delayed onset of debilitating muscle degeneration. Although a transient burst of frank degeneration occurs in
mdx mice during the period of muscle growth around the third postnatal week, degeneration leading to debility only occurs late in life, primarily in specific muscles, such as the diaphragm [
5]. In
mdx mice, as in human DMD patients, disease is caused by the absence of functional dystrophin, owing to a nonsense mutation in exon 23. The relatively mild phenotype of
mdx mice can, in part, be attributed to the compensatory function of the dystrophin-related protein utrophin, which is highly upregulated in regenerating muscle fibres in adult
mdx mutants [
6]. Over time, enhanced muscle turnover and satellite cell numbers are also seen in
mdx mice [
7].
Other muscle diseases also show variable clinical progression. Human motor and sensory neuropathy type 1A (HMSN1A, also known as Charcot-Marie-Tooth disease type 1A, CMT1A) is a dominantly inherited demyelinating disorder of the peripheral nervous system. It is most frequently caused by over-expression of the
PMP22 gene due to duplication of a 1.5-Mb region on chromosome 17, but it can also result from point mutations in the
PMP22 gene [
8‐
12]. The affected individuals typically have distal muscle weakness and atrophy often associated with mild to moderate sensory loss, depressed tendon reflexes, and high-arched feet. Individuals with HMSN1A experience slowly progressive weakness and atrophy of distal muscles in the feet and/or hands. Disease progression is variable for unknown reasons.
PMP22 C22 transgenic mice which were modified to harbour seven copies of the human
PMP22 gene demonstrate developmental delays in myelination, decreased numbers of myelinated fibres, and abnormally thin myelin similar to HMSN1A [
13]. Being a neuropathy, PMP22 C22 mice can be used as reference animals that display a muscle phenotype without harbouring intrinsic muscle defects.
In a recent study, we found that satellite cells of MSV
ski transgenic mice display a differentiation defect compared to wildtype control animals and that this defect is exacerbated in ageing animals [
14]. Like
mdx mice, hypertrophic MSV
ski transgenic mouse muscles have muscle degeneration that is initially efficiently repaired, but which eventually shows defective regeneration and frank muscle defects. In the present paper, we investigate the differentiation potential of satellite cells of single muscle fibres from the hypertrophic
mdx and
PMP22 mouse models and corresponding wildtype control animals in order to clarify whether ageing-related change in differentiation potential of satellite cells might influence disease progression.
Discussion
Satellite cells are normally quiescent cells that reside under the basal lamina of muscle fibres. Under conditions of growth and repair, satellite cells become activated and begin a coordinated myogenic program: they initially proliferate and express a range of myogenic genes, including desmin, before aligning and fusing to form terminally differentiated multinucleate syncitia that organise and express the contractile apparatus, which includes the MyHC proteins. We have shown recently that, in
MSVski transgenic mice, satellite cell differentiation
in vitro was influenced by the age of the study animals and showed a progressive decline with age [
14]. This effect was much less obvious in wildtype control animals. Therefore, we concluded that a satellite cell differentiation defect developed in
MSVski mice. We speculated that the defect might be caused by continual degeneration/repair apparent in hypertrophic
MSVski mice and that the changes in satellite cells might underlie the worsening of the muscle pathological profile with age. The
mdx mouse model of human DMD displays a hypertrophy phenotype in the skeletal musculature reminiscent of
MSVski mice. Similarly, DMD patients initially show a mild phenotype that gradually progresses throughout the lifespan, manifesting as muscle loss and fibrosis which culminates in death from respiratory or cardiac failure. As the progression of DMD is not understood and because of similarities between DMD/
mdx pathology and the MSV
ski animal phenotype we hypothesised that progressive defect in the differentiation potential of satellite cells might contribute to the pathologic mechanism of the debilitating human disease.
We find that satellite-derived cells from ageing
mdx mice are, in general, capable of differentiating to the same degree as satellite-derived cells from control animals. We did not assess satellite-derived cells that remained physically juxtaposed to the explanted fibre because such cells could display different behaviour(s) due to variations in the fibre integrity or its surrounding matrix. We chose to measure differentiation as the fraction of desmin-expressing cells co-expressing MyHC rather than fusion for several reasons. Firstly, without using assays that detect syncitia [
19], there is no way of clearly ascertaining whether two cells are truly fused or closely apposed. Secondly, mononucleate cells are capable of terminally differentiating and expressing MyHC; such cells would be missed if assessing fusion. Although we can not eliminate the possibility that the
mdx condition leads to an alteration in the cells that express desmin, our study provides evidence against this view. First, yields of desmin
+ cells are similar between
mdx and control. Second, the proliferation rate of desmin
+ cells appears similar between
mdx and control. Third, and as discussed further below, although numbers of desmin
- cells are increased in
mdx cultures the increase is similar both before and after the differentiation phase suggesting that interconversion of desmin
- and desmin
+ cells is not a significant factor in our experiments. Overall, it is unlikely that a defect in differentiation of satellite-derived cells is a major contributor to disease progression in
mdx mice.
Despite this lack of significant change in overall differentiation capacity, satellite-derived cells from some individual older mdx animals displayed lower differentiation efficiencies than those from other mdx animals of the same age. Age- and background genotype-matched control mice, young mdx mice, and mice with severe muscle weakness due to the PMP22 transgene did not show this variation. We were unable to correlate this effect to any factor analysed. All animals were held under similar conditions. We consciously used both genders and analysed data in combination of both sexes and separately. We conducted additional experiments using increased or decreased collagenase type I incubation times as well as a more severe collagenase type II digestion to determine if variations in satellite cell activation or yield, basal lamina digestion, or fibre damage could affect differentiation potential, but found no differences (data not shown). Similarly, we performed dilution cloning of satellite cells to assess the effects of proliferation rate on satellite cells and whereas we observed heterogeneity of proliferation rates amongst satellite cells grown from single cells, we again found no ultimate difference in differentiation efficiency (data not shown). For those reasons we can not explain the inter-animal variation of mdx results by differences in the experimental design or genetic background of the animal. We speculate that uncontrolled environmental effects or epigenetic factors affecting other genes in the mdx background explain the variation. It is striking that fibres yielding poorly-differentiating cells are numerous in affected individuals, but nevertheless, some fibres yield cells differentiating as well in controls. This emphasises that relatively heritable heterogeneity in myogenic cells must exist in mdx mice and demands elucidation. Moreover, we can not eliminate the possibility that the mdx individuals showing poor differentiation in our assay would have undergone a worse progression of disease in later life. Additionally, we cannot exclude the possibility that very subtle differences in differentiation behaviour were not detected in our assay system as we have utilized matrigel, a matrix in which growth factors are abundant. Thus, small variations might have been masked that only would be detectable at the application of collagen or gelatine matrices.
As shown by others [
20] and in this report, non-myogenic cells, probably fibroblasts, can be obtained from single fibre cultures and are more abundant in
mdx samples compared with C57BL/10 controls. These cells probably reside on the fibre surface and migrate away from the fibre onto the substrate as do satellite-derived cells.
In vivo, these cells may mediate the fibrotic response to fibre degeneration and could potentially secrete factors such as TGF-β that have been shown to interfere with satellite cell differentiation [
21]. We analysed the proportion of non-myogenic cells in the cultures and whether they influenced the efficiency of differentiation of myogenic cells. We were unable to find a correlation between the contamination of the satellite cell culture with desmin
- non-myogenic cells and the differentiation efficiency of the myogenic cells in the same culture well. This confirms what we have observed in wild type,
MSVski,
PMP22, and
myoD null situations [
14]. There was also no difference in desmin
- cell levels between mice showing poor or normal differentiation. Thus, at least in
mdx animals up to one year of age, no correlation of fibrosis with poor myoblast differentiation is apparent.
Differentiation is delayed, not inhibited in some mdx mice
We found that satellite-derived cells from
mdx mice showing poor differentiation after two days differentiation, recovered and differentiated as well as controls after three further days in differentiation conditions. No morphological differences in the nature of the differentiated cells were detected at this stage. Thus, the reduction in differentiation observed in some
mdx animals is most simply explained as a reduced rate of differentiation. If such a decrease in differentiation rate occurs
in vivo, it could have serious consequences for muscle repair, which may require rapid satellite cell mobilisation and can occur within a few days [
22].
IGF-1 treatment has been shown to enhance the efficiency of differentiation of satellite-derived myoblasts and this has been suggested to mediate an autocrine loop triggered by myogenin expression [
23]. However, IGF-1 treatment of poorly-differentiating
mdx satellite-derived cells did not enhance their differentiation rate. Moreover, although it has been reported that IGF-treatment of wildtype myogenic cells leads to enhanced proliferation [
24,
25], in our experimental conditions using recombinant R
3 IGF-I we were not able to confirm those findings.
Functional impairment in HMSN1A model mice is not accompanied by satellite cell changes
A rodent model for the progressive human neuropathy HMSN1A is the
PMP22 C22 transgenic mouse that harbours seven copies of the human
PMP22 gene in its genome. These animals manifest symptoms similar to human HMSN1A patients including demyelination of Schwann cells and, with later age, progressive skeletal muscle weakness caused by poor innervation [
13]. Examination of postural soleus muscle from such mice revealed a heterogeneous progression of muscle pathology. Some animals showed severe signs of atrophic fibres and changes in fibre type proportion, probably caused by altered electrical activity consequent to slowed nerve conduction. Other individuals, which also showed poor movement, had relatively healthy muscle histology. A cohort of affected
PMP22 C22 mice with altered gait were analysed by single fibre culture and revealed no changes in number, proliferation, or differentiation of desmin
+ cells compared to age- and genetic background-matched controls. Thus, the HMSN1A model mice show a strong phenotype which seems to involve muscle fibre atrophy after demyelination of its innervating motorneuron followed by satellite cell recruitment in the regrowth phase after myelination is restored. However, these processes have no observable effect on mononucleate cells in single fibre cultures.
Differentiation defect in satellite cells lacking myoD
There has been controversy surrounding the role of the myogenic transcription factor MyoD in satellite cell differentiation. Early reports suggested the
myoD null muscle regenerated poorly and that myoblasts from young
myoD null mice differentiated poorly
in vitro [
17,
26]. However, a recent study showed the
myoD null satellite cells can differentiate efficiently under some circumstances [
27]. In our single EDL muscle fibre cultures, satellite-derived myoblasts from wild type or heterozygous
myoD
+/- mice differentiate efficiently. In contrast, few desmin-expressing satellite-derived cells from
myoD deficient mice were able express MyHC within two days. Given the rapid repair of entire muscles within several days of toxin-induced injury [
22], the differentiation delay in
myoD null would have severe consequences in wild populations if a similar delay occurred in any
in vivo setting.
An issue not addressed directly by our study is the relationship of the assayed population of desmin-expressing cells from wild type to that from myoD null animals. For example, lack of myoD might cause a change in the numbers of cells expressing desmin. We think this unlikely because the yield of desmin+ cells, and the ratio of desmin+ to desmin- cells, were unaltered in our single fibre cultures, irrespective of myoD genotype. Whatever the case, our experiment shows that lack of myoD leads to substantial reduction in the capacity of muscle fibre-associated migratory proliferative cells to undergo terminal differentiation into myotubes.
Conclusion
As there appears to be no distinct differentiation defect in the satellite cells of
mdx mice, at least by our assay, the cause of progressive muscle loss in the diseased state remains unclear. The differentiation potential of satellite cells has generated contradictory results in several early studies, where differentiation was measured as myoblast fusion in primary cultures (reviewed in [
28,
29]). One study concluded no differentiation defect [
4], whereas Jasmin and colleagues found a reduced differentiative capacity in myoblasts derived from DMD patients [
30]. A very recent publication by Schaefer et al. also revealed heterogeneity in the number of satellite cells between individual mdx and C57 animals. The authors observed that this heterogeneity did not correlate with age, gender, or degree of degeneration, but possibly reflected additional genetic factors that influence the maintenance of the satellite cell pool [
31].
Altered myoblast number appears not to explain disease progression. Some studies find an increase in the number of satellite cells associated with DMD/
mdx muscle fibres ([
32‐
35] and our unpublished observations), although others observe little change [
7]. Thus loss of available cells for repair is unlikely to cause disease progression. Despite reports of increased cell death in DMD/
mdx myoblasts [
36‐
38], we found no evidence of differential apoptosis between control and
mdx. It has also been suggested that decreased proliferative potential and early senescence of satellite cells is the primary cause of disease progression [
3,
39]. One group measured this limitation as an accelerated age-related shortening of satellite cell telomere length [
40], a result contradicted by another study using a similar assay [
41]. We observed no change in proliferative capacity of
mdx satellite-derived cells. These findings suggest that the proliferative potential of satellite cells or their ability to self renew is not compromised; the number of satellite cells remains higher than controls throughout the lifespan of the animals ([
32]; our observations).
In summary, our study is the first to provide data on the differentiation efficiency of satellite cells in older
mdx and
PMP22 C22 mice. We found that, generally, the age of the individual animal had little impact on the differentiation of satellite cells. The pathological processes in muscle of mice with
mdx or
PMP22 C22 genotype are not necessarily accompanied by defects in satellite cell differentiation. However, in some
mdx individuals, satellite cell differentiation is impaired after the age of two months, in a manner akin to that we observed in
MSVski hypertrophic muscle pathology [
14]. It seems clear that there is heterogeneity in the pathology of muscle in various animal models of human disease, probably due to epigenetic/environmental contributions. It remains to be further investigated what causes cases of poor differentiation and whether this impaired differentiation of satellite cells enhances deterioration of the physical condition of the individual.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
MMS and CJM have contributed equally and have carried out the cellular experiments and drafted the manuscript. HB helped with the animals. CH provided PMP22 C22 animals. SMH conceived the study, participated in its design and coordination, and contributed to the writing. All authors read and approved the final manuscript.