CTS-IPITA-IXA 2007 Joint Conference – Abstracts - Minneapolis – September 15–20, 2007

1) Revisiting the problem of the limited migration of myoblasts transplanted in a muscle

Jacques P. Tremblay

Centre Hospitalier Universit Laval, Centre de Recherche, Unit de Gntique Humaine, Qubec QC, Canada

The transplantation of normal myoblasts in the muscles of Duchenne Muscular Dystrophy (DMD) patients is a possible treatment to restore the expression of dystrophin in their muscle fibers. One of the main problems of this potential therapeutic approach is that the myoblasts injected in a muscle fuse only with the muscle fibers, which are close to the injection trajectories. Experiments made in mouse muscles indicated that the myoblasts indeed migrated only 150 microns. Thus close injection trajectories (i.e., at 1 mm interval) are necessary to obtain the expression of dystrophin in a high percentage of the muscle fibers. Our research group has thus made experiments to pretreat myoblasts with various growth factors (including IGF-1, bFGF, MGF and IL-4) to increase their migration. We found that IGF-1 and bFGF increased gelatinase (MMP9) activity in human myoblasts. These growth factors improved the overall migration of monkey myoblasts in vitro in a serum free medium and in the muscles of SCID mice. The potentiating effect ofIGF-1 on myoblast migration was blocked by Amiloride (uPA inhibitor) and marimastat (BB94 MMP inhibitor). Increased myoblast migration was also obtained with a synthetic peptide (MGF-Ct24E) based on the E domain of IGF-1. This peptide increased the expression of uPA, uPAR and MMP7 and reduced PAI-1. It also promoted the in vitro migration of human myoblasts. The motogenic activity of this peptide does not involve the IGF-1R since it was not blocked by a mAb against the IGF-1 receptor. MGF-Ct24E peptide also promoted the in vivo migration of human myoblasts. The mechanismby which the MGF-Ct24E peptide enhances the in vivo migration involves both u-PA and MMP proteolytic systems. Our group also found that IL-4 is secreted by differentiating myoblasts and that these cells express the IL-4 and IL-13 receptors. IL-4 only slightly increased MMP2 and MMP-9 but significantly increased active uPA and uPAR and reduced PAI-1. IL-4 did not increase a5 integrin but increases b1 and b3 integrins. A mAb against IL-4 prevented formation of large myotubes. Co-injection of IL-4 with human myoblasts improved their migration in vivo. Thus IL-4 is involved in the migration of myoblasts towards myotubes to increase their fusion with these myotubes. After this series of experiments in vitro and in mouse muscles, we were however disappointed to find that bFGF and IGF-1 did not improve the long-term graft success following transplantation of monkey myoblasts coinjected with these growth factors. In fact, a series of experiments led us to conclude that increased migration alone is not sufficient to increase the number of dystrophin+ fibers. Our experiments indicated that muscle fiber repair or induction of muscle hypertrophy is required. Indeed short and long term observations following transplantation through a polyethylene microtube of monkey myoblasts in SCID mice indicated that although myoblasts labeled with PKH67 migrated 700 mm in at 60 h, 14 days later dystrophin+ fibers were observed only near the microtube. This is probably because damaged muscles fibers were present at 5 h only near the myotube and not at 700 mm from it. New experiments in monkeys indicated that although muscle fibers expressing ab-Gal reporter gene present at 30 days only along the injection trajectories, the b-gal labeled myoblasts were present in extrafascicular pockets 3 days after their injection in a monkey muscle. Moreover, myoblasts injected subcutaneous migrated in a monkey muscle and fused with previously damaged muscle fibers. Several treatments to increase the damage to the muscle fibers, such as notexin, bupivicaine, a large needle, electroporation and exercise increased the numbers of dystrophin+ fibers in a muscle because the myoblasts were induced to fuse with these damaged muscle fibers. Following the transplantation of normal myoblasts, induction of muscle hypertrophy by blocking the myostatin signal, either with follistatin, a dominant myostatin receptor or a siRNA for the myostatin receptor also increased the frequency of muscle fibers expressing dystrophin. Thus to increase the success of myoblast transplantation is not sufficient to improve the migration of the myoblasts, the fusion of these cells with the muscle fibers must be induced by either damaging the fibers or by inducing their hypertrophy.

2) Myoblast transplantation in Duchenne Muscular Dystrophy: lessons from human and nonhuman primate studies

Daniel Skuk

Human Genetics Unit, CHUL, Laval University, Quebec, Canada.

Facioscapulohumeral muscular dystrophy (FSHD) is an autosomal dominant disease linked to a deletion within tandem repeats termed D4Z4 located on chromosome 4q35, and is characterized by a typical regional distribution, featuring composed patterns of clinically affected and unaffected muscles. No treatment is available for this condition, in which the pathophysiological mechanism is still unknown, especially regarding its selectivity. The most popular hypothesis is that the reduced D4Z4 array would allow chromatin opening and expression of more centromeric genes. Autologous transfer of myoblasts from unaffected to affected territories could be considered as a potential strategy to delay or stop muscle degeneration. To validate this concept, we produced, characterized and compared myoblasts prepared from clinically unaffected muscles of FSHD patients (n=5) and from healthy donors (n=10), in view of establishing if FSHD myoblasts grown from unaffected muscle, but still carrying the FSHD molecular defect, would manifest or not its deleterious effect when submitted to a large scale expansion or in vitro differentiation. According to a clinically-approved procedure, 109 cells were obtained within 16–23 d. More than 80% of these cells were myoblasts (CD56 and desmin-positive). FSHD myoblasts presented a doubling time equivalent to that of control cells; they kept high proliferation ability and did not show early telomere shortening. In vitro, these cells were able to differentiate and to express muscle-specific antigens. In vivo, they participated to muscle structures when injected into immunodeficient mice. Then, we analysed the characteristics of myogenic cells prepared from affected FSHD muscles (n=6). Myoblasts from dystrophic FSHD muscles displayed important alterations in morphology, proliferation and in vitro differentiation ability when compared to myoblasts from non dystrophic muscles and matched controls. These data suggested that myoblasts expanded from unaffected FSHD muscles could be suitable tools in view of autologous cell transplantation clinical trials, and mandated the set-up of a phase I. The primary objectives of this trial are to evaluate the feasibility, safety, biological outcomes of injecting high numbers of autologous cells (prepared from vastus lateralis) in a large portion of a clinically affected muscle (tibialis anterior). Three groups of 3 patients are scheduled, allowing an escalation of the number of injection sites between groups. The trial was authorized by the French regulatory agency and Ethics committee, and started in October 2006. The first three patients received cell injections without complications. Upon analysis of the first group results, the next group will be included.

Several different recessive muscular dystrophies, including Duchenne Muscular Dystrophy (DMD), can be treated by transplanting normal allogeneic myoblasts in the muscles. These myoblasts containing a normal copy of the gene, which is mutated in the patient, will fuse with the existing muscle fibers and lead to the expression of this normal gene. Our research group has recently demonstrated the feasibility of this approach in a Phase 1 clinical trial inDMD patients. However, such allogeneic myoblast transplantations currently require a sustained immunosuppression. We have developed in mice a safe protocol to develop immunological tolerance for this cellular allo-transplantation. This tolerigenic treatment is called TTCB (for Transfusion of donor blood, Treosulfan, Cyclophosphamide and Bone marrow transplantation). We are currently trying to validate this protocol in monkeys and dogs. An alternative to avoid the requirement for a sustained immunosuppression is to transplant to the patients their own genetically corrected myoblasts. We have indeed been able to introduce not only a micro-dystrophin gene with a lentiviral vector but also the full length dystrophin cDNA using a hybrid adeno/AAV vector. We have also been able to induce the skipping of a dystrophin exon using a lentivirus coding for a U7RNA. This approach permitted to restore the reading frame of the dystrophin mRNA in DMD myoblasts and led to the expression of quasi-dystrophin in SCID mice transplanted with these genetically corrected myoblasts. Another problem of the clinical application of myoblast transplantation is these cells fused only with the muscle fibers located near the injection trajectories. We initially hypothesized that this was because the myoblasts did not migrate away from the injection trajectories. However, more recent experiments indicated than even when these cells migrated more than 700 microns away from their injection site they did not fuse the neighboring muscle fibers. However, damage induced to the fibers with notexin, bupivicaine, exercice and electroporation increased the number of muscle fibers expressing dystrophin due to increased fusion of the donor myoblasts with the host muscle fibers. Blocking the myostatin signal with follistatin or with a dominant negative Activin IIB receptor also increased these hybrid fibers expressing dystrophin. Thus for the myoblasts to be able to fuse with the host muscle fibers, the fibers have either to be undergoing repair or undergoing hypertrophy. Finally, roughly 75% of the myoblasts transplanted in a muscle died in the first 3 to 5 days following their transplantation. This death is due to many mechanisms, which induce their apoptosis or their necrosis. Anoikis, a form of apoptosis due to the lack of matrix attachment of the myoblasts, is one of the mechanisms involved. Moreover, the myoblasts located in the center of a cell pocket died due to necrosis. Most of the surviving cells are affected by oxidative stress that may be due in part to the inflammatory reaction. The recent years have thus permitted to better identify the problems limiting the success of myoblast transplantation to dystrophic patients and solutions to improve the success of this therapy are progressively being developed.