Isabelle Asselin

Very efficient myoblast allotransplantation in mice under FK506 immunosuppression

Transgenic CD1 mice expressing β‐galactosidase were used as myoblast donors. The myoblasts were injected in normal or mdx muscles previously irradiated and injected with notexin. Twenty‐eight days after myoblast transplantation, the percentage of muscle fibers β‐galactosidase‐positive was low in mice not immunosuppressed but was high (80%) in those treated with FK506. In mdx mice, muscle fibers expressing β‐galactosidase were also dystrophin positive. Most of the mice not treated with FK506 produced antibodies against the donor myoblasts. These results indicate that FK506 is a very useful immunosuppressive drug for myoblast transplantation in mice. Irradiation and notexin injection used in our experiments are, however, not feasible in humans. Other manipulations capable of increasing the participation of donor myoblasts to regeneration will therefore have to be identified before new clinical trials are attempted.

Evidence of mdx mouse skeletal muscle fragility in vivo by eccentric running exercise

Duchenne muscular dystrophy is an X‐linked devastating disease due to the lack of expression of a functional dystrophin. Unfortunately, the dystrophin‐deficient mdx mouse model does not present clinical signs of dystrophy before the age of 18 months, and the role of dystrophin in fiber integrity is not fully understood. The fragility of the skeletal muscle fibers was investigated in transgenic mice expressing β‐galactosidase under the control of a muscle specific promoter. Adult mdx/β‐galactosidase (dystrophin‐negative) and normal/β‐galactosidase (dystrophin‐positive) mice were submitted to one short session of eccentric, downhill running exercise. The leakage of muscle enzymes creatine kinase and β‐galactosidase was investigated before, 1 h after, and 3 days after the running session. A significant and transient rise in the level of these enzymes was noted in the serum of mdx mice following the exercise session. Thus, the lack of dystrophin in the mdx model led to local microdamages to the exercised muscle allowing leakage of proteins from the fibers. The peak leakage was transient, suggesting that muscle fiber lesions were rapidly repaired following this short, noninvasive eccentric running session.

Myoblast transplantation in monkeys: control of immune response by FK506.

Myoblasts were grown from monkey muscle biopsies and infected in vitro with a defective retroviral vector containing a cytoplasmic β-galactosidase (β-gal) gene. These myoblasts were then transplanted to 14 different monkeys, 6 of which were immunosuppressed with FK506. Without immunosuppression, only a few myoblasts and myotubes expressing β-gal were observed 1 week after the transplantation, but no cells expressing β-gal were observed after 4 weeks. This result was attributed to immune responses since infiltration by CD4+ or CD8+ lymphocytes was abundant 1 week after transplantation but not after 4 weeks. The expression of interleukin 6 (IL-6), interleukin 2 (IL-2), granulocyte/macrophage colony stimulating factor (GM-CSF), transforming growth factor-beta (TGF-β) and granzyme B mRNAs was increased in the myoblast-injected muscle indicating that the infiltrating lymphocytes were activated. Moreover, antibodies against the donor myoblasts were detected in 3 out of 6 cases. When the monkeys were immunosuppressed with FK506, muscle fibers expressing beta-galactosidase (β-gal) were present 1, 4 and 12 weeks after the transplantation. There was neither significant infiltration by CD4 or CD8 lymphocytes, nor antibodies detected. The mRNA expression of most cytokines was significantly reduced as compared to the nonimmunosuppressed monkeys. These results indicate that FK506 is effective in controlling short-term immune reactions following myoblast transplantation in monkeys and suggest that it may prove useful for myoblast transplantation in Duchenne Muscular Dystrophy patients.

Successful myoblast allotransplantation in mdx mice using rapamycin.

Duchenne muscular dystrophy (DMD)* is due to a mutation occurring on a gene located on the X chromosome that normally encodes for the expression of a large, 420-kDa protein called dystrophin (Dys) (1). This protein, which is normally located along the inner part of the plasma membrane of skeletal muscle cells, is absent from DMD patient muscles (2). The lack of Dys is responsible for the absence of Dys-associated proteins and glycoproteins. The absence of this complex would be responsible for the progressive degeneration of skeletal muscles in DMD patients (3). A point mutation in the Dys gene leads to the expression of a truncated protein in the mdx mouse, considered as an animal model of human DMD (4). Normal myoblast transplantation restores Dys expression in the grafted muscles of mdx mice. Indeed, normal myoblasts contain a normal Dys gene, and when they fuse together or with the host myoblasts, nuclei from these normal myoblasts restore Dys expression (5, 6). Normal myoblast transplantation had only limited success in human. Some groups demonstrated, a short time after grafting, the presence of Dys+ fibers in some patients (7–10). Moreover, transient strength increases have been measured in a few patients (10). In some cases, hosts produced antibodies (Ab) against the myoblasts and/or myotubes of the donor, even when donor and recipient were fully compatible for class I and class II DR MHC (7, 10). Some Abs were able to induce a complement-dependent lysis of myotubes, and some were also directed against the donor Dys (10). Such immune reactions have also been demonstrated in mdx mice transplanted either with human or rat (xenografts) or mouse (allografts) myoblasts (11, 12). High doses of CsA apparently diminish or suppress in some animals this humoral reaction (12, 13). Thus, controlling the immune system seems to be a crucial step to performing successful myoblast transplantations. CsA is not very efficient for myoblast transplantation in mice. It is toxic and may not be used for long-term treatments. Therefore, the new immunosuppressant, rapamycin (RAPA), has been tested to evaluate its potential to reduce immune reactions and allow myoblast graft success. RAPA is a fungal, FK506 structurally related macrolid (14). Rapa inhibits cytokine and immunoglobulin synthesis, blocks lymphocyte proliferation, and leads to T lymphocyte unresponsiveness upon cytokine stimulation (15,16). RAPA allows, at low doses, long-term graft survival in many models and many animal species (17–19). Short-term graft success under RAPA treatment has been evaluated in this study in 4 different muscle-grafting models. Graft success has been assessed mainly by the characterization of donor fibers in the host muscles, by the examination of Dys expression, and by the analysis of production of anti-donor cell Ab. Finally, the infiltration of cells involved in inflammation and immune reactions has been studied in grafted muscles.

Anti-inflammatory effect of transforming growth factor-beta1 in myoblast transplantation.

BACKGROUND The inflammatory reaction that occurs during the 5 days after transplantation led at 3 days to the death of 70% of injected myoblasts. Use of anti-inflammatory agents appeared to be a possible way to increase myoblast survival. The application of gene transfer techniques to cell transplantation offers the potential for the prevention of inflammatory reaction. METHODS In this study, transforming growth factor-beta1 (TGF-beta1) gene was introduced in myoblasts with a retroviral vector to permit the secretion of this anti-inflammatory cytokine. Survival of (1) infected myoblasts expressing TGF-beta1 or (2) normal myoblasts transplanted with genetically modified cloned myoblasts was compared with survival of normal myoblasts. RESULTS Expression of TGF-beta1 by myoblasts or by cotransplanted cells decreased myoblast mortality after 3 days by roughly 20% (66.0+/-3.0% in control vs. 46.3+/-4.2% and 46.2+/-5.9%). The increase of myoblast survival by TGF-beta1 expression was correlated with a lower polymorphonuclear cell and macrophage infiltration in muscles compared with control. In addition, cytotoxicity of neutrophils against myoblasts was assayed in vitro. The oxidation of myoblasts by activated neutrophils was decreased after infection of the myoblasts with the TGF-beta1 retroviral vector. CONCLUSIONS These data demonstrate that the insertion of TGF-beta1 decreases inflammatory reaction observed after myoblast transplantation and thus prolongs their survival.

Transplantation of myoblasts from a transgenic mouse overexpressing dystrophin produced only a relatively small increase of dystrophin-positive membrane

Myoblast cultures from normal and Tg‐MDA (transgenic mouse overexpressing dystrophin 50‐fold) mice were transplanted into dystrophin‐deficient mdx mouse muscles. Four weeks after transplantation, dystrophin‐positive fibers were observed four times more frequently in cross sections of muscles injected with Tg‐MDA. Myoblasts from Tg‐MDA mice also expressing the β‐gal transgene (Tg‐MDA/β‐gal) and myoblasts from β‐gal transgenic mice containing one normal dystrophin gene (normal/β‐gal) were also transplanted into mdx mouse muscles. Four weeks after transplantation, the fiber length positive for dystrophin (nuclear domain) was shorter (439 ± 326 μm) than the β‐gal nuclear domain (1466 ± 713 μm) of the same fiber when normal/β‐gal myoblasts were transplanted, but increased (1302 ± 487 μm) when Tg‐MDA/β‐gal myoblasts were used. These experiments show that despite the presence in Tg‐MDA myoblasts of constructions which lead in vivo in transgenic mice to an overexpression of dystrophin 50‐fold, the membrane area over which dystrophin was expressed was increased only threefold. This observation is also expected for vector‐mediated gene therapy.

Increased myogenic potential and fusion of matrilysin-expressing myoblasts transplanted in mice.

The success of myoblast transplantation in clinical trials has been limited in part by the low dispersion of grafted cells outside the injection site. Our research group previously reported that the culture of myoblasts with concanavalin A, a stimulator of metalloproteinase production, increased their migration. Several lines of evidence also suggested that muscle cell fusion involves metalloproteinase-sensitive mechanisms. To determine whether the increased expression of metalloproteinases had an influence on myoblast fusion and dispersion through the muscle following transplantation, we generated a myoblast cell line expressing human matrilysin (MMP-7). The MMP-7-expressing myoblasts were obtained by the stable transfection of a matrilysin expression vector in a TnILacZ immortomouse myoblast clone. Matrilysin-expressing myoblasts showed a highly increased in vitro fusion index, forming seven times (p < 0.001) more myotubes than the control cell line and three times (p < 0.001) more myotubes than the Immortomyoblast parental clone. Single-site transplantation of matrilysin-expressing myoblasts generated more fibers (p < 0.001), over a greater surface (p < 0.001) than the control cell line. The cotransplantation of matrilysin-expressing myoblasts and of normal human myoblasts in SCID mice increased the number of human dystrophin-positive fibers and myotubes by sixfold. Although no significant increased migration of myoblasts outside the injection sites was observed, our results show that the metalloproteinase activity can improve the myogenic potential of myoblasts in vitro and the fusion of myoblasts with host fibers in vivo. MMP-7 expression may be useful in increasing myoblast transplantation success.

Combined immunosuppression of mycophenolate mofetil and FK506 for myoblast transplantation in mdx mice.

Background. Overcoming adverse effects of immunosuppressors can be achieved by combining different drugs, thus allowing a dosage reduction. Myoblast transplantation is a potential therapy for Duchenne muscular dystrophy. Our research group previously established that FK506 (tacrolimus) is an effective immunosuppressive drug for myoblast transplantation in mice and monkeys. Methods. In the present study, a reduced dose of FK506 at 1.0 mg/kg/day was used in combination with mycophenolate mofetil (MMF; 80 mg/kg/day) as an immunosuppressive protocol for myoblast transplantation. Graft success was evaluated by quantifying the number of dystrophin-positive fibers per muscle section that were injected with normal cells. Results. MMF used alone could not prevent immune rejection of the transplanted myoblasts. MMF given in combination with FK506 immediately after transplantation reduced the success of myoblast transplantation by about 50%. A low dose of FK506 combined with MMF after the establishment of the graft (3 weeks) maintained graft success and controlled immune infiltration compared with a low dose of FK506 alone. However, lymphocyte infiltration was observed at longer term using a low dose of FK506 combined with MMF. Conclusions. The diminution of graft success when combining FK506 and MMF by the time of myoblast transplantation could be attributed to the inhibition of myoblast fusion by MMF. The use of MMF and FK506 after the establishment of the graft did not reduce graft success, however, this combination was not effective at controlling long-term immune rejection in comparison with the optimal dose of FK506 alone.

Myoblast transplantation in non-dystrophic dog

Dog myoblasts obtained from muscle biopsies were infected in vitro with a defective retroviral vector containing a cytoplasmic beta-galactosidase (beta-Gal) gene. These myoblasts were initially transplanted in the irradiated muscles of SCID mice and beta-Gal positive muscle fibers were observed. beta-Gal myoblasts were also transplanted back either in the donor dogs (autotransplantation model) or in unrelated recipient dogs (allotransplantation model). Following these myoblast injections, a rapid inflammatory reaction developed within the muscle as indicated by an expression of P-selectin and of pro-inflammatory cytokine mRNAs (interleukin 6 (IL-6) and transforming growth factor beta (TGF-beta), and by a neutrophil infiltration. Following either auto- or allotransplantation in inadequately or non-immunosuppressed dogs, a specific immune reaction also developed within 2 weeks as indicated by the infiltration of CD4+ and of CD8+ lymphocytes, the increased expression of IL-10 and granzyme B mRNAs and the presence of antibodies reacting with the injected cells. Some dogs were immunosuppressed with several combinations of FK506, cyclosporine (CsA) and RS-61443. In dogs immunosuppressed with CsA combined with RS-61443, only a few myoblasts and myotubes expressing beta-Gal were observed 1-2 weeks after the transplantation, but no muscle fibers expressing beta-Gal were observed after 4 weeks, and antibodies against the injected cells were formed. In dogs immunosuppressed with FK506 alone, although no antibodies against the injected cells were produced, there were no small cells and no muscle fibers expressing beta-Gal 1 month after the transplantation. However, FK506 triggered diarrhea and vomiting in dogs. When the dogs were immunosuppressed with FK506 combined with CsA and RS-61443, muscle fibers expressing beta-Gal were present 4 weeks after the transplantation and no antibodies reacting with donor myoblasts were detected. These results indicate that the combination of three immunosuppressive agents (i.e., FK506, CsA and RS-61443) is effective in controlling the specific immune reactions following myoblast transplantation in dogs and they underline that the outcome of myoblast transplantation is dependent in part on an adequate immunosuppression. These results obtained here in normal dogs may justify myoblast transplantation in dystrophic dogs despite the side effects of FK506.

Role of non-major histocompatibility complex antigens in the rejection of transplanted myoblasts

Myoblasts obtained from donors histoincompatible for several non-major histocompatibility complex antigens (i.e., including minor histocompatibility antigens) and from syngeneic donors were transplanted without any immunosuppression into the muscles of male dystrophic C57BL/10J mdx/mdx mice. Myoblasts from syngeneic mice resulted in the formation of a high percentage of dystrophin-positive fibers 16 weeks after the transplantation. There was no evidence of a cellular immune reaction against the donor myoblasts, i.e., no infiltration by CD4 or CD8 lymphocytes and no increased expression of granzyme B and interferon-gamma mRNAs. Transplantation of myoblasts obtained from donors histoincompatible only for non- major histocompatibility complex antigens produced a transient increase of dystrophin-positive fibers at 4 weeks after transplantation for some donor strains but not for others. For donor strains that did produce an increase at 4 weeks, the number of dystrophin-positive fibers was reduced 16 weeks after the transplantation. There was evidence of a cellular immune reaction-infiltration by CD4 and by CD8 lymphocytes and increased expression of granzyme B and interferon-gamma mRNAs. Transplantation of myoblasts obtained from male C57BL/10J +/+ mice into female C57BL/10J mdx/mdx mice also led to the presence of only a few dystrophin-positive fibers with the same signs of cellular immune reaction. In this later case, the cellular immune response was attributed to the H-Y minor antigens. Finally, antibodies against fetal calf serum were detected after both syngeneic and nonsyngeneic transplantations, indicating that the culture medium may also be a source of antigens. In mice, the presence of these antibodies against culture medium did not reduce the success of a first syngeneic transplantation.