Cell therapy for degenerative muscle diseases such as the muscular dystrophies requires a source of cells with the capacity to participate in the formation of new muscle fibers. We investigated the myogenic potential of human fetal mesenchymal stem cells (hfMSCs) using a variety of stimuli. The use of 5-azacytidine or steroids did not produce skeletal muscle differentiation, whereas myoblast-conditioned medium resulted in only 1%-2% of hfMSCs undergoing muscle differentiation. However, in the presence of galectin-1, 66.1% +/- 5.7% of hfMSCs, but not adult bone marrow-derived mesenchymal stem cells, assumed a muscle phenotype, forming long, multinucleated fibers expressing both desmin and sarcomeric myosin via activation of muscle regulatory factors. Continuous exposure to galectin-1 resulted in more efficient muscle differentiation than pulsed exposure (62.3% vs. 39.1%; p < .001). When transplanted into regenerating murine muscle, galectin-1-exposed hfMSCs formed fourfold more human muscle fibers than nonstimulated hfMSCs (p = .008), with similar results obtained in a scid/mdx dystrophic mouse model. These data suggest that hfMSCs readily undergo muscle differentiation in response to galectin-1 through a stepwise progression similar to that which occurs during embryonic myogenesis. The high degree of myogenic conversion achieved by this method has relevance for the development of therapies for muscular dystrophies.

Previous studies have shown that transfer of human myoblasts carrying a mitochondrial DNA mutation into muscles of the severe combined immunodeficient mouse may provide an important animal model for mitochondrial myopathy. However, a major drawback of this mouse is its extreme sensitivity to ionising radiation, a pre-treatment which enhances the efficiency of myoblast transfer success. We implanted human myoblasts into the tibialis anterior muscles of another immunodeficient mouse, mutated in the recombinase activating gene-1 (RAG-1), to determine if this mouse could be an alternative to the severe combined immunodeficient for our mitochondrial myoblast transfer model. We also examined several different methods of muscle degeneration prior to myoblast transfer to determine which method resulted in the greatest amount of human tissue in implanted muscles. Our results show that the RAG-1 mouse displayed no sensitivity to the irradiation process compared to the high sensitivity in the severe combined immunodeficient mouse which resulted in early termination of the study. We also show that degeneration of host muscles by the myotoxin barium chloride (BaCl2) resulted in the greatest amount of regenerating human muscle fibres in both the severe combined immunodeficient and RAG-1 mice. In addition, the maximum amount of human fibres observed in transplanted muscles was similar in each mouse strain. The average number of fibres throughout muscles was significantly greater in severe combined immunodeficient mice injured by BaCl2, but was similar between all other muscle groups. This study suggests that the RAG-1 mouse is a suitable host for the mitochondrial myoblast transfer model and may also prove valuable for other myoblast transfer models such as muscular dystrophy.

The dogma that a cell is rigidly committed to one tissue type has been heavily challenged over the past few years with numerous reports of transdifferentiation of cells between different lineages. Cells capable of entering lineages other than that of their tissue of origin have been identified in several diverse tissues. Recently we have focussed on a non-committed myogenic cell within the dermis that is capable, under certain conditions, of expressing muscle specific markers and even fusing to the terminally differentiated stage of muscle cell development. We have identified galectin-1 as being a potent factor implicated in this process. In this review we discuss our findings and consider the involvement of galectin-1 in muscle determination, differentiation and regeneration. Published in 2004.

Continuous cycles of muscle fiber necrosis and regeneration are characteristic of the muscular dystrophies, and in some cases this leads to premature replicative senescence of myoblasts in vitro. The molecular mechanism of senescence in human myoblasts is poorly understood but there is evidence to suggest that telomeric attrition may be one of the ways by which this is achieved. We report here, for the first time, the extension of normal human skeletal muscle cell replicative life span by the reconstitution of telomerase activity. The telomerase-expressing cells show no features of transformation in vitro and have stable genomes with diploid karyotypes, do not express exceptionally high levels of c-myc and have wild-type, unmethylated CDKN2A genes. In vivo, they regenerate to repair muscle injury in immunosuppressed RAG-1 mice. This work suggests that telomerase expression to repair short telomeres may aid the expansion of diploid human muscle cells and consequently attempts at gene therapy for muscle diseases.

We have previously shown that galectin-1 is a factor capable of converting mouse dermal fibroblasts to the myogenic lineage [Cell Transplant 2000;9:519]. Here, we report that human dermal fibroblasts are also capable of expressing the myogenic marker, desmin, when grown in muscle-cell-conditioned media. Furthermore, the human foetal skin cells also express this marker when grown in the presence of galectin-1. These results highlight the importance of galectin-1 in the conversion of both human and murine skin cells to a myogenic lineage. Thus galectin-1 could be an important tool for use in autologous cell therapies for the treatment of human muscular dystrophies.

Since its discovery four decades ago, the satellite cell of skeletal muscle has been implicated as the major source of myogenic cells involved in growth and repair of muscle fibres. This review not only looks at the role of the satellite cell in these processes but discusses how cells derived from other sources and tissues have recently been implicated in muscle formation and regeneration. Muscle itself also yields cells that contribute to other cell lineages although it is currently debated as to whether these cells originate within muscle or have migrated there from other tissues. The reality of using cells from muscle or other tissues to repair diseased muscle fibres is also addressed.

Normal murine dermal fibroblasts implanted into the muscles of the mdx mouse, a model for Duchenne muscular dystrophy, not only participate in new myofibre formation but also direct the expression of the protein dystrophin which is deficient in these mice. We have reported that the lectin galectin-1 is implicated in the conversion of dermal fibroblasts to muscle. In the current work we confirm the presence of galectin-1 in the medium used for conversion. Furthermore we report that exposure of clones of dermal fibroblasts to this lectin results in 100% conversion of the cells. Conversion was assessed by the expression within the cells of the muscle-specific cytoskeletal protein desmin. We also investigate the effects of galectin-1 on cells of the C2C12 mouse myogenic cell line and on primary mouse myoblasts. Exposing both transformed and primary myoblasts to the lectin resulted in an increase in fusion of cells to the terminally differentiated state in both types of cultures. Galectin-1 does not cause the myogenic conversion of murine muscle-derived fibroblasts.

Both in vitro and in vivo studies have described the conversion of fibroblasts to myogenesis when in the presence of dysfunctional myogenic cells. Myogenic conversion of fibroblasts subjected to a normal, as opposed to a diseased muscle environment has only been reported in vitro. The primary aim of this work was to determine if fibroblasts can convert to a myogenic lineage and contribute to new fibre formation when implanted into the regenerating muscle of a normal mouse. Dermal fibroblasts were prepared from neonatal mouse skin and labelled prior to implantation with the fluorescent nuclear marker 4′,6‐diamidino‐2‐phenylindole (DAPI). Cells were implanted into muscles of host mice that had been subjected to either cold/crush or minced muscle injury. Some host muscles were x‐irradiated to deplete the muscle of endogenous muscle precursor cells. Muscles were removed at 3 wk postimplantation and analysed both histologically and for the presence of DAPI labelled nuclei. Fibres containing DAPI labelled central nuclei indicated that the implanted cells had participated in the regenerative process. Mouse dermal fibroblasts therefore do contribute to muscle fibre formation in regenerating normal mouse muscle but the extent of their contribution is dependent on the nature of the trauma induced in the host muscle. The study also showed that regeneration was more successful in muscles which had not been irradiated, which is contrary to the previous studies where dermal fibroblasts were introduced into myopathic mouse muscle.

Myotonic dystrophy is caused by a CTGn expansion in the 3′-untranslated region of a serine/threonine protein kinase gene (DMPK), which is flanked by two other genes, DMWD and SIX5. One hypothesis to explain the wide-ranging effects of this expansion is that, as the mutation expands, it alters the expression of one or more of these genes. The effects may vary in different tissues and developmental stages, but it has been difficult to develop these hypotheses as the normal postnatal developmental expression patterns of these genes have not been adequately investigated. We have developed accurate transcript quantification based on fluorescent real-time reverse transcription-polymerase chain reaction (TaqMan) to develop gene expression profiles during postnatal development in C57Bl/10 mice. Our results show extensive independent postnatal regulation of the myotonic dystrophy-locus genes in selected tissues and demonstrate which are the most highly expressed of the genes in each tissue. All three genes at the locus are expressed in the adult lens, questioning a previous model of cataractogenesis mediated solely by effects onSix5 expression. Additionally, using an in vivomodel, we have shown that Dmpk levels decrease during the early stages of muscle regeneration. Our data provide a framework for investigation of tissue-specific pathological mechanisms in this disorder.

Using the mdx mouse model for human Duchenne muscular dystrophy we have shown that a cell population residing in the dermis of C57B1/10ScSn mouse skin is capable of converting to a myogenic lineage when implanted into the mdx muscle environment. It was important to determine the characteristics of the converting cell. A previous in vitro study indicated that 10% of cells underwent conversion but only when the cells were grown in medium previously harvested from a myogenic culture. In the present study we cloned cells derived from the dermis to identify the converting cells. Clones grown in normal growth medium showed no conversion, but when grown in medium conditioned by muscle cells around 40% conversion was achieved in several individual clones. We investigated whether the protein β-galactoside binding protein (βGBP), which is secreted by myoblasts and acts as a cell growth regulator of fibroblasts, could be a candidate factor responsible for conversion. Medium harvested from COS-1 cells infected with a construct containing βGBP has been used for this investigation. Growth of dermal fibroblasts in medium enriched with this factor showed a high rate of conversion to cells expressing muscle-specific factors.

A fundamental keystone of developmental biology has been the growing use of reporter genes in model transgenic systems. Their use has greatly facilitated investigations of cell lineage and cell fate in addition to aiding experiments aimed at determining patterns of gene expression, gene interaction and gene regulation. Through construction of transgenic mice, ubiquitously expressing human placental alkaline phosphatase (PLAP), we demonstrate the suitability of PLAP as a reporter gene for use in conjunction with, or as an alternative to, beta-galactosidase (lacZ). Our findings demonstrate that over-expression of PLAP has no adverse effects on mouse development or viability, despite a widespread pattern of expression. This technology provides a simple yet effective mechanism based on eukaryotic reporter gene technology to facilitate the identification of transgenic cells within complex in vivo systems.

Defects of the mitochondrial genome are important causes of disease. Despite major advances in our investigation of patients, there is no effective therapy. Progress in this area is limited by the absence of any animal models in which we can evaluate treatment. To develop such a model we have injected human myoblasts into the tibialis anterior of SCID mice after inducing necrosis. After injection of normal human myoblasts, regenerating fibers expressed human beta-spectrin, confirming they were derived from fusion of human myoblasts. The stability of the muscle fibers was inferred by demonstrating the formation of motor end plates on the regenerating fibers. In addition, we show the presence of human cytochrome c oxidase subunit II, which is encoded by the mitochondrial genome, in the regenerated fibers. After injection of human myoblasts containing either the A8344G or the T8993C heteroplasmic mitochondrial DNA mutations, human beta-spectrin positive fibers were found to contain the mutation at a similar level to the injected myoblasts. These studies highlight the potential value of this model for the study of mitochondrial DNA defects.

Laminin alpha 2, a sub-unit of the basement membrane component laminin-2, is deficient in the dy/dy and allelic dy2 J/dy2 J mouse. It is also the defective protein in a proportion of children with congenital muscular dystrophy. Linkage and mutational analysis have established that this is a primary effect caused by defects in the LAMA2 gene. Laminin alpha 2 has previously been shown to be deficient in dy/dy skeletal muscle, peripheral nervous system and brain. We report here preliminary observations on differences in detection of laminin alpha 2 in muscle, peripheral nerves and brain of dy/dy mice using three, well characterized antibodies. In normal muscle laminin alpha 2 is localized to the basement membrane of the myofibres and the Schwann cells of peripheral nerves, whilst in adult brain it is only detected on blood vessels. Our results show that there is appreciable, but slightly reduced, expression of laminin alpha 2 in skeletal muscle of dy/dy mice but almost no detectable protein in the brain, peripheral nerve and spinal nerve roots. Our observations are at present unexplained but they raise the possibility for the first time that there may be different tissue specific isoforms of laminin alpha 2. Molecular characterization of possible differences responsible for our observations may aid the identification of the mutation in the dy/dy mouse and lead to a better understanding of the role and expression of laminin alpha 2 in pathological conditions.