Resumos que serão apresentados no 6o Encontro Anual da Sociedade Americana de Terapia Gênica que será realizado de 4 a 8 de junho de 2003 em Washington.

1) Correction of the Dystrophin Gene Mutation in the mdx5cv Mouse Model of Duchenne Muscular Dystrophy Mediated by Chimeric and DNA Oligonucleotides In Vitro and In Vivo

Carmen Bertoni,1 Thomas A. Rando.1

1Department of Neurology, Stanford University Medical Center,Stanford, CA, United States.

Gene correction represents an appealing option for the treatment of genetic disorders due to the prospect of permanent restoration of gene expression. We have investigated the possibility of inducing single base pair alterations at the genomic level to restore the expression of dystrophin in mouse models of Duchenne muscular dystrophy, a severe muscle disease caused by mutations in the dystrophin gene. We have shown the ability of chimeric RNA/DNA oligonucleotides (chimeraplasts) to correct a point mutation in the dystrophin gene in the mdx mouse. We have also shown the feasibility of using chimeraplasts to mutate a base in an intron/exon boundary of the dystrophin gene to alter splicing, an approach that may be applicable to a variety of dystrophin gene defects. We have now compared the correction ability of chimeraplasts to that of DNA oligonucleotides in muscle cells of the mdx5cv mouse. This model has a point mutation in exon 10 of the dystrophin gene that creates a cryptic splice site. Exon 10 is thus aberrantly spliced resulting in alteration of the dystrophin coding sequence leading to a lack of dystrophin expression. We have designed a targeting chimeraplast (MDX7¹) and targeting DNA oligonucleotides (MDX7² and MDX7³) to specifically correct the mdx5cv mutation. Each oligonucleotide is perfectly homologous to the region of exon 10 of the mdx5cv dystrophin gene containing the mutation, except for a mismatch at the mutated base. The chimeric oligonucleotide is designed to pair with both strands of the DNA; the DNA oligonucleotides are designed to pair with either the coding (MDX7³) or the non-coding strand (MDX7²). As controls, we have used a chimeraplast (MDX8¹) and DNA oligonucleotides (MDX8² and MDX8³) identical to the targeting oligonucleotides but lacking the mismatch with the mdx5cv mutation. Fluorescently labeled oligonucleotides are efficiently taken up in muscle precursor cells in vitro using all 3 types of oligonucleotides. Fluorescence persists longer in cells transfected with MDX7² or MDX7³, suggesting that DNA oligonucleotides have increased stability compared to chimeraplasts. Restoration of dystrophin expression was assessed at the mRNA and protein level. All targeting oligonucleotides were capable of restoring dystrophin expression, while control oligonucleotides had no effect. Gene correction was demonstrated at the genomic level in cells transfected with targeting oligonucleotides. Quantitative RT-PCR indicated that the level of gene correction varied between 0.2 to 5%. The most efficient oligonucleotides were the chimeraplast (MDX7¹) and the DNA oligonucleotide that was designed to anneal with the coding strand (MDX7³). The chimeric and DNA oligonucleotides also corrected the mdx5cv mutation in vivo as determined by the restoration of dystrophin expression. The expression of dystrophin was assessed as early as 2 weeks after injection and was stable for at least 3 months after injection. Our studies provide evidence that oligonucleotide-mediated gene correction is a feasible approach to the treatment of certain genetic disorders in which long-term gene expression is required. Thus this technology has the potential to be a viable, non-viral approach to stable restoration of gene expression.

2)  Long-Term Transgene Expression from High-Capacity Adenoviral Vectors Delivered to Fetal Muscle In Utero

Roberto Bilbao,1 Daniel Reay,1 Volker Biermann,2 Christoph Volpers,2 Zhilong Jiang,1 Stefan Kochanek,2 Paula R. Clemens.1,3

1Neurology, University of Pittsburgh, Pittsburgh, PA; 2Center for Molecular Medicine, University of Cologne, Cologne, Germany;3Department of Veterans Affairs Medical Center, Pittsburgh, PA.

In utero gene delivery holds promise for the treatment of hereditary diseases such as Duchenne muscular dystrophy (DMD). To date, efficient transduction has been achieved using firstgeneration adenoviral vectors. Due to the large size of the dystrophin gene cDNA (14 kD), gene transfer of the full-length cDNA will require a vector with a larger insert capacity such as the high-capacity adenoviral (HC-Ad) vector. In this study, we analyzed the longevity of transgene expression achieved by direct HC-Ad vector-mediated gene delivery to muscle in utero. We also studied the efficiency of muscle gene delivery by HC-Ad vectors with intravascular delivery. We first evaluated the transduction levels in muscle after intramuscular delivery of an HC-Ad vector carrying the lacZ gene (AdGS46) to fetal C57BL/6 mice 16 days after conception (E-16).Hind limb muscles were collected 1 and 5 months after infection and analyzed for â-galactosidase (â-Gal) expression by the onitrophenyl-â-D-galactopyranoside (ONPG) assay. High levels of transgene expression in muscle were found. We also investigated intravascular delivery of HC-Ad vector to C57BL/6 E-16 fetuses and observed high transduction efficiency in limb muscles. In addition,higher survival rates were observed in those mice transduced with an HC-Ad vector as compared to a first-generation Ad vector. To assess the potential of HC-Ad vector-mediated gene transfer to fetal muscle in a therapeutic model, we performed intramuscular injections of an HC-Ad vector carrying the dystrophin gene (AdDYS) to E-16 mdx mice, the animal model for DMD. Immunohistochemical staining showed dystrophin expression in muscle of mdx mice transduced in utero with AdDYS. Our results demonstrate that 1) Long-term transgene expression can be achieved by HC-Ad vector-mediated gene delivery to fetal muscle; 2) The HC-Ad vector can deliver full-length dystrophin to fetal muscle in utero.

3) A Novel Approach To Identify Patients with Duchenne Muscular Dystrophy Caused by Stop Codon Mutations Using Aminoglycoside Antibiotics

Shigemi Kimura,1 Tishihiko Miyagi,1 Takashi Hiranuma,1 Kowashi Yoshioka,1 Shiro Ozasa,1 Kaori Ito,1 Makoto Matsukura,1 Makoto Ikezawa,1 Masafumi Matsuo,2 Yasuhiro Takeshima,3 Teruhisa Miike.1

1Department of Child Development, Kumamoto University School of Medicine, Kumamoto, Kumamoto, Japan; 2Division of Molecular Medicine, Kobe University Graduate School of Kobe, Kobe, Hyogo, Japan; 3Department of Pediatrics, Kobe University Graduate School of Kobe, Kobe, Hyogo, Japan.

Intro: Aminoglycoside antibiotics have been found to suppress premature stop codons located in the defective dystophin gene in mdx mice, suggesting a possible treatment for Duchenne muscular dystrophy (DMD) . However, it is very difficult to find patients that are applicable for this therapy, because: 1) only 5 to 10% of DMD patients have nonsense mutations in the dystrophin gene, 2) it is challenging to find nonsense mutations in the gene because dystrophin cDNA is very long (14kb), and 3) the efficiency of aminoglycoside-induced read-through is dependent on the type of nonsense mutation. Recently, our research has focused on MyoD, a transcriptional factor that has the ability to differentiate fibroblasts into myotubes in vitro. Adenoviral vectors encoding MyoD, regulated by CAG Promoter (AdMyoD), can efficiently transduce fibroblasts to express MyoD. In this study, we introduce an easy system to identify patients for this therapy and report for the first time, that dystrophin expression was detected in myotubes of DMD patients using gentamicin. Methods:Fibroblasts were isolated from six DMD patients. In patient 1, a deletion of exons 48~50 in the dystrophin gene resulted in an out of frame pattern of the gene. Patients 2-6 had nonsense mutations in the dystrophin gene; the stop codon is TGA for patients 2-4 and TAA for patients 5 and 6. Control fibroblasts were isolated from a non-DMD patient. The fibroblasts were infected in vitro with AdMyoD using a multiplicity of infection (MOI) of 100. Following infection, the cells were cultured in DMEM supplemented with 2% FBS and 300 mg/ml of gentamicin. At 2 weeks postinfection, the dystrophin expression was analyzed by dystrophin staining and Western blotting analysis. Results:The in vitro immunofluorescence staining and Western blot analysis for dystrophin showed that dystrophin expression was not detected in the myotubes of patient 1 (deletion of dystrophin gene) cultured with and without gentamicin . In contrast, dystrophin expression was detected in myotubes of patients 2, 3 and 4 (stop codon mutation TGA) cultured with gentamicin, but not detected in myotubes cultured without genatamicin. Interestingly, dystrophin expression was not observed in myotubes from patients 5 and 6 with the stop codon mutation TAA, in spite of being cultured with gentamicin. Dystrophin expression was detected in control myotubes from fibroblasts of a non-DMD patient after culturing them with and without gentamicin. Discussion:We have developed a system to identify DMD patients caused by stop codon mutations in the dystrophin gene that are eligible for gentamicin treatment. By monitoring dystrophin expression of myotubes differentiated from fibroblasts infected with AdMyoD and cultured in gentamicin, we are able to determine which patients will benefit from such treatments. In addition, our results show that this system for the aminoglycoside treatment is far more effective for DMD patients that have nonsense mutation TGA than for patients that have nonsense mutation TAA.

4) Prolonged Dystrophin Expression and Functional Correction of mdx Mouse Muscle Following Gene Transfer with a Helper- Dependent (Gutted) Adenovirus Encoding Murine Dystrophin

Renald Gilbert,1 Roy W. R. Dudley,2 An-Bang Liu,3 Basil J.Petrof,2 Josephine Nalbantoglu,4 George Karpati.4

1Genomics and Gene Therapy Vector Group, Biotechnology Research Institute, NRC, Montreal, QC, Canada; 2Respiratory Division, McGill University, Montreal, QC, Canada; 3Department of Neurology, Tzu Chi Medical Center, Hualien, Taiwan;4Neuromuscular Research Group, Montreal Neurological Institute, Montreal, QC, Canada.

Dystrophin gene transfer using helper-dependent adenoviruses (HDAd), which are deleted of all viral genes, is a promising option to treat muscles in Duchenne muscular dystrophy. We investigated the benefits of this approach by injecting the tibialis anterior (TA) muscle of neonatal and adult dystrophin-deficient (mdx) mice with a fully deleted HDAd (HDCBDysM). This vector encoded two full-length murine dystrophin cDNAs regulated by the powerful cytomegalovirus enhancer/â-actin promoter. At 10 days postinjection of neonatal muscles, 712 fibers (42% of the total number of TA fibers) were dystrophin positive (dys+), a value that did not decrease for 6 months (the study duration). In treated adults, maximal transduction occurred at 30 days post-injection (414 dys+ fibers,24% of the total number of TA fibers), but decreased by 51% after 6 months. All studied aspects of the pathology were improved in neonatally-treated muscles: the percentage of dys+ fibers with centrally localized myonuclei remained low, localization of the dystrophin associated protein complex was restored at the plasma membrane, muscle hypertrophy was reduced, maximal force generating-capacity and resistance to contraction-induced injuries were increased. The same pathological aspects were improved in the treated adults, except for reduction of muscle hypertrophy and maximal force generating capacity. We demonstrated a strong humoral response against murine dystrophin in both animal groups, but mild inflammatory response occurred only in the treated adults. Our data indicate that HDCBDysM is one of the most promising and efficient vectors for treating DMD by gene therapy, and that early muscle treatment using this vector would mitigate the DMD pathology more efficiently.

5) Gel-Based Delivery of Recombinant AAV Vectors to Adult Murine Diaphragm

Thomas J. Fraites, Jr.,1 Cathryn Mah,1 Irene Zolotukhin,1 Barry J. Byrne.1

The diaphragm and respiratory muscles are important targets for gene transfer and therapy for a host of dystrophies and myopathies.Transgenic and knockout mice provide the most widely available and genetically homogenous models in which to test potential therapies. The murine diaphragm, which is six to eight cells thick in most normal mice, presents unique delivery challenges. We have sought to develop methods for efficient and uniform delivery of recombinant adeno-associated virus (rAAV) vectors to mouse diaphragm. We constructed rAAV vectors based on rAAV serotypes 1, 2, and 5, and evaluated their utility for diaphragmatic gene delivery with and without a gel-based delivery vehicle. Recombinant AAV2 vector plasmids encoding the cytomegalovirus immediate-early promoterdriven beta galactosidase reporter gene were cross-packaged into AAV1, 2, and 5 capsids as previously described (Zolotukhin, et al.,2002. Methods. 28(2):158-67). Vectors were mixed at room temperature with a water-soluble, bacteriostatic, gelatin-based gel and directly applied to the abdominal surface of the diaphragm. Free virus, without vehicle, was also directly applied to control diaphragms. All vectors were administered at a dose of 5x109 particles. Four weeks after gene delivery, diaphragm tissues were harvested and assessed for enzyme activity by X-gal staining.

6) Widespread Gene Expression of dsarcoglycan in the Bio14.6 Dystrophic Hamster Hanidleg Muscles by Pressurized Delivery of a Double-Stranded AAV Vector

Tong Zhu,1 Liqiao Zhou,1 Bing Wang,1 Juan Li,1 Xiao Xiao.1

Introduction: Gene transfer of the missing d-sarcoglycan in the limb girdle muscular dystrophy hamster Bio14.6 by AAV vectors is an effective treatment. But systemic gene delivery through blood vessel and transduction efficiency of AAV vectors need further improvement. Here we tested the pressurized intra-arterial injection method to transfer a novel double-stranded AAV vector carrying the d-sarcoglycan gene into the hindlimb muscle of Bio14.6 hamsters. Method: To deliver the AAV vector through artery into a large group of the muscles, the femoral vessels were carefully dissected under a surgical microscope. Two overlapping rubber tourniquets were transmuscularly placed at the level of the proximal thigh. A microvascular clamp was placed to temporarily occlude the femoral vessels. A 32G intracranial catheter was canulated distally into the femoral artery. After the tourniquets were tightened , 4x1012 particles of dsAAV-CMV-d-sarcoglycan diluted in 1 ml PBS was injected into the artery as fast as possible (normally in 8-10s). Local Intramuscular injection was also performed as a positive control with 2x1012 AAV vector particles into the gastrocnemius and tibialis anterior muscles, respectively. All animals were sacrificed 1 or 2 months after injection. Cryosections of Quadriceps, GAS and TA muscles were obtained for anit-d-sarcoglycan immunostaining. Results: The untreated muscle cells showed a degeneration morphology without sarcoglycan expression; In contrast, in the intravascularly treated hindlimb, sustained and uniform expression of sarcoglycan was observed on the cell membranes of all muscle groups downstream of the vasculature. However, in local intramuscular injection group, the sarcoglycan only expressed locally on the membrane of muscle cells in the injected muscle. Uneven expression of the d-sarcoglycan was observed. In general, no toxicity was observed for both gene delivery methods. Conclusion: The pressurized intravascular injection method can systemically deliver the dsAAV-CMV-d-sarcoglycan to the hindlimb muscles, which offers potential clinical significance for future gene therapy trials.

7) AAV Mediated Co-Delivery of Igf-I and Dystrophin to mdx Mouse Muscle

Simone Abmayr,1 Tara J. McNair,1 Rob W. Crawford,1 Jeffrey S. Chamberlain.1

1Department of Neurology, University of Washington School of Medicine, Seattle, WA.

Gene replacement therapy for muscular dystrophy has been shown to improve morphological features and functional properties of dystrophic mdx mouse muscle. However, viral vector mediated delivery of dystrophin to all muscle fibers remains a challenging goal. Insulin-like growth factor-I (Igf-I) was found to enhance muscle regeneration and to maintain muscle mass and function in old and dystrophic animals. Igf-I is a secreted polypeptide and can therefore target virus infected and non-infected cells by binding to its receptor and triggering proliferative and differentiation responses and antiapoptotic pathways. Our goal is to co-deliver dystrophin and Igf- I to dystrophic muscle to determine if the protective effect of Igf-I is synergistic with the beneficial effects of dystrophin in ameliorating the mdx phenotype. We have cloned and characterized the isoforms of Igf-I that are expressed in normal and dystrophic mouse muscle. Based on this information, we have generated several AAV vectors that express the major muscle isoform of Igf-I. Our AAV vectors express Igf-I under the control of the CMV and the muscle specific promoters, CK6 and desmin. We have developed a real time PCR assay to quantify RNA expression and to characterize the relative strength of the promoters. Intramuscular injection of these AAV Igf-I vectors into C57/BL10 mice leads to levels of Igf-I mRNA expression up to 500-fold above normal. Currently, we are investigating the effects of this overexpression on the muscle function and morphology. Additionally, we have co-injected AAV carrying micro-dystrophin with AAV carrying Igf-I into dystrophic muscle and are studying their relative and combined potential for reversing the dystrophic pathology of the mdx mouse.

8) Local Delivery of VEGF165 by AAV Vectors Protects Skeletal Muscle from Injury and Promotes Muscle Regeneration

Nikola Arsic,1 Serena Zacchigna,1 Lorena Zentilin,1 Genaro Ramirez-Correa,1 Sabrina Tafuro,1 Lucia Pattarini,1 Alessandro Salvi,2 Gianfranco Sinagra,2 Mauro Giacca.1

1Molecular Medicine Laboratory, International Centre for Genetic Engineering and Biotechnology, Trieste, Italy; 2Cardiology Unit, Ospedale Maggiore, Trieste, Italy.

Vascular endothelial growth factor (VEGF) is a main regulator of blood vessel formation during embryogenesis and a potent inducer of neovascularization during adult life. Recent evidence suggests that VEGF activity is not strictly specific for endothelial cells, but is also exerted on other cell types. Here, we report on the role of VEGF165 in promoting myogenic precursor cell differentiation to form multinucleated myotubes in vitro, as well as in the enhancement of muscle regeneration in vivo. By immunofluorescence on cultured satellite cell-derived primary myoblasts, we showed that VEGFR2 is strongly upregulated after 2 days of culture under differentiating conditions, and remains highly expressed until the last stages of the differentiation process. Similar results were obtained using the C2C12 myogenic cell line. In both primary myoblasts and C2C12 cells VEGF determined cell cycle arrest and protected cells from apoptotic death. Moreover, the administration of recombinant VEGF during C2C12 differentiation resulted in a significant increase in the number and in the length of the newly formed myotubes, resulting from myoblast fusion. The effects of VEGF on muscle cell survival and regeneration in vivo were assessed by the injection of a high titer preparation of an AAV vector expressing VEGF in two established murine models of muscle damage. Muscle fiber injury was obtained by injection of either 50% glycerol or 1 mM cardiotoxin, which both induce rapid destruction of muscle fibers and strong inflammatory reaction. The former treatment also recapitulates some of the hallmarks that define the physiopathology of Duchenne muscular dystrophy. By immunohistochemistry we observed that injury with both glycerol and cardiotoxin induced expression of VEGFR2 in muscle fibers. Delivery of AAV-VEGF resulted in a remarkable improvement in the preservation of viable fibers and in the induction of fiber regeneration at day 20 after damage with both agents. Preservation of tissue architecture was almost complete after injury with glycerol. This effect involved a marked reduction in fiber apoptosis (as detected by reactivity to anti-caspase-3 antibody) and an increase in the number of regenerating myofibers. Activity of AAV-VEGF strictly correlated with the dose of vector administered. These results demonstrate that VEGF exerts a powerful and specific effect on muscle cell survival and myogenic differentiation. This conclusion implicates that gene delivery of VEGF, besides induction of therapeutic angiogenesis, might be considered for the induction of muscle regeneration for the treatment of a variety of muscular disorders.

9) Rapid Identification of Novel Canine Models of Duchenne Muscular Dystrophy

Bruce F. Smith,1 Glenn E. Morris,2 Joe N. Kornegay,3 Richard J.Bartlett.4

1Scott-Ritchey Research Center, Auburn University, Auburn, AL,United States; 2NEWI, University of Wales, Wrexham, United Kingdom; 3College of Veterinary Medicine, University of Missouri, Columbia, MO, United States; 4NIAMS, National Institutes of Health, Bethesda, MD, United States.

Duchenne muscular dystrophy (DMD) is an X-linked, progressive muscle wasting disease with fatal consequences, which is caused by mutations in the human dystrophin gene. DMD presents unique challenges to gene therapy, due to the size of the gene and resulting cDNA and the wide variety and complexity of the mutations involved. As an X-linked recessive disease, new mutations present themselves at higher rates than in autosomally inherited diseases. In addition, mutations in DMD differ from those classically seen in inherited disease, with an emphasis on deletions. Animal models for DMD have been described in mice, cats and dogs with mutations in the respective dystrophin genes, located on the xchromosome. Both cats and mice have less severe forms of the disease. In the murine model, additional mutations such as a utrophin gene knockout must be bred into the mdx background to reproduce the pathology, symptomatology and fatal consequences found in DMD patients. Only in dogs does the disease carry the full spectrum of clinical phenotypes found in patients with DMD, including the fatal consequences of muscle wasting. Thus, only canine dystrophinopathies recapitulate the human disease with a single gene mutation. In dogs and humans the structure, size and exon junction location of the dystrophin genes are remarkably similar. Thus, it is not surprising that the mutations in the animal models described thus far prove to be different mutations. Any therapy developed for treatment of this spectrum of mutations must be capable of overcoming the consequential spectrum of clinical conditions. To maximize the potential for pre-clinical evaluation of potential therapies, a spectrum of mutations in canine breeds with the consequential continuum of clinical phenotypes would prove invaluable. To date, canine models have been described in the Golden Retriever, Rottweiler, and German Short Haired Pointer breeds. We have developed a program to rapidly screen possible new canine models of dystrophin deficiency using a combination of specific antibodies and rapid, directed sequencing. In the initial screen, antibody binding data is used to identify the likely portion of the molecule affected by the mutation. This region is then amplified by RT-PCR in one kilobase sections and sequenced. Data from these PCR reactions also provides an indication of small deletions or insertions, either due to the failure of the PCR reaction to amplify, or due to alterations in amplicon size. Analysis of the cDNA sequence data allows the specific mutation to be identified or it may indicate possible genomic alterations that affect cDNA structure. This approach allows rapid identification of the mutation in the model, allowing the suitability of a model to be determined without significant investment or delay. By determining the mutation in additional dogs that have presented with clinical muscular dystrophy, a spectrum of single gene mutations in canine models can be described that will allow more appropriate screening of therapeutic approaches than is possible in inbred mice. Data from the analysis of new canine models of Duchenne Muscular Dystrophy will be presented.

Bing Wang,1 Mengnan Tian,1 Chunping Qiao,1 Tong Zhu,1 Juan Li,1 Xiao Xiao.1,2

1Dept.of Molecular Genetics and Biochemistry, University of Pittsburgh, Pittsburgh, PA, United States; 2Dept. of Orthopaedic Surgery, University of Pittsburgh, Pittsburgh, PA, United States.

Duchenne muscular dystrophy (DMD) is the most common disabling and lethal genetic muscle disorder, affecting one of every 3,500 males. No effective treatment is currently available for DMD. The mdx mouse has been the most widely used animal model for DMD, although mdx lacks major clinical deterioration seen in human patients. In contrast, the golden retriever muscular dystrophy (GRMD) dog, as a large animal model, displays remarkable clinical and pathological similarities to the human DMD patients. Therefore, the GRMD is well conceived as a clinically more relevant DMD model. Previously, we have attempted AAV-mediated human mini-dystrophin gene delivery to treat the GRMD, but the result was not striking. We believe that the human dystrophin gene expression in dog may have elicited immune-response in the dystrophic dog. In this report, we have cloned the dog dystrophin cDNA via RTPCR from normal dog muscle and generated a dog version minidystrophin gene (3.8 kb, including 5 rods) that can be readily package into AAV vector along with CMV promoter. We show that the dog mini-dystrophin can be expressed in mdx mice at high levels at one month after vector injection. Immunostaining of the consecutive sections, using antibody against N-terminus of human dystrophin, revealed the restoration of the missing dystrophin onto the plasma membrane. H & E staining also displayed normal histology and the lack of fibrosis and infiltration in the vector transduced area. The results prove the feasibility of using AAV dog mini-dystrophin vector in the large GRMD canine model.

11) Lentiviral Vector Mediated Gene Transfer to Mouse Skeletal Muscle Cells: PotentialApplications for Duchenne Muscular Dystrophy

Sheng Li,1 En Kimura,1 R. W. Crawford,1 B. Fall,1 J. M. Scott,1 J. C. Angello,2 R. Welikson,2 S. D. Hauschka,2 J. S. Chamberlain.1

1Department of Neurology, University of Washington School of Medicine, Seattle, WA; 2Department of Biochemistry, University of Washington School of Medicine, Seattle, WA.

Mutations in the dystrophin gene cause Duchenne muscular dystrophy (DMD). We have shown previously that delivery of mini- or full-length dystrophin genes to muscles of mdx mice, a model of DMD, can prevent and partially reverse the dystrophic pathology. However, gene therapy for DMD will require systemic delivery, and sustained expression, of therapeutic dystrophins in widely distributed skeletal muscles. Lentiviral vectors have a relatively large transgene carrying capacity and are able to integrate into non-dividing cells. We explored the use of lentiviral vectors for transferring genes into mouse skeletal muscle cells in vitro and in vivo. The lentiviral vectors efficiently transduced both proliferating and terminally differentiated mdx muscle cells in vitro, and transgeneexpressing myoblasts were able to differentiate normally without any obvious toxicity. We demonstrated that even a small version of the murine creatine kinase regulatory cassette maintained musclespecific activity in lentivirally-transduced cells. We were also able to transduce with high efficiency the minidystrophin-lentivirus cassettes into a variety of other cell types in vitro, including myoblasts derived from dystrophic dogs and hematopoietic stem cells. Although we were able to obtain moderate levels of skeletal muscle transduction in vivo by direct intramuscular injection, the relatively low titer of lentiviral preparations combined with physical barriers to virus diffusion will limit the direct application of lentiviral vectors for transferring therapeutic genes into muscles. Nonetheless, the ability to successfully transduce both muscle and bone marrow cells in vitro with mini-dystrophin expressing lentiviral vectors suggests that this system may have great potential for developing ex vivo cell therapies for DMD.

12) Functional Analysis of Dystrophin in Vascular Smooth Muscle Cells in Duchenne Muscular Dsytrophy

Kaori Ito,1 Shigemi Kimura,1 Shiro Ozasa,1 Makoto Ikezawa,1 Misao Suzuki,2 Kowashi Yoshioka,1 Makoto Matsukura,1 Takashi Hiranuma,1 Takeshi Miwa,3 Teruhisa Miike.1

1Department of Child Development, Kumamoto University Medical School, Kumamoto, Kumamoto, Japan; 2Division of Transgenic Technology Center for Animal Resources and Development (CARD), Kumamoto University, Kumamoto, Kumamoto, Japan; 3Department of Oncogene Research, Research Insutitute for Microbial Deseases, Osaka University, Suita, Osaka, Japan.

Duchenne muscular dystrophy (DMD) is an X-linked fatal disease caused by mutations of the gene encoding the cytoskeletal protein dystrophin. Dystrophin is a membrane-associated protein that provides a link in a chain of proteins between the actin cytoskeleton to extracellular matrix and comprises a dystrophin-glycoprotein complex (DGC). Dispite the abundance of new information on these molecules, there currently is no effective treatment for DMD because the mechanism by which dystrophin deficiency produces the clinical phenotype is poorly understood. Two principal theories have been proposed to explain the pathogenesis of DMD. The first is that dystrophin deficiency destabilizes the sarcolemmal integrity, thereby rendering the muscle fibers susceptible to damage during contractions. The second theory is that disruption of DGC gives rise to the reduction of the scaffolding function that recruits signaling proteins such as neuronal nitric oxyde synthase (nNOS) to the membrane. Recent studies indicate that nNOS in skeletal muscle plays a key role in the regulation of the blood flow within exercising skeletal muscle by blunting the vasoconstrictor response to alphaadrenergic receptor activation. This protective mechanism is defective in both, DMD patients and mdx mice, an animal model of DMD. We hypothesized that dystrophin deficiency also causes the reduction of nNOS in vascular smooth muscle cells (VSMCs), leads to vascular dysfunction and exacerbates muscle pathology. We therefore generated transgenic mice expressing 14Kb full-length human dystrophin cDNA under the transcriptional control of the  mooth muscle alpha-actin promoter. These mice were then crossed with mdx mice, resulting in three independent SMTg/mdx lines which harbor the dystrophin gene only in SMCs. PCR and southern blot analysis were performed to verify founders and stable transgenic lines. The expression pattern was detectable by semi-quantitative RT-PCR analysis and immunohistochemical staining, which showed the specific expression of transgene in SMCs. We also report the histological characteristics of SMTg/mdx mice such as central nucleation, fiber size variability, and CK concentrations as compared to C57BL/10 control mice and mdx mice. We believe that our SMTg/ mdx mouse model is worth exploring to gain a better understanding of the functon of dystrophin in VSMCs and the pathophysiology of DMD patients. We also believe that the introduction of dystrophin gene into VSMCs is necessary for the effective treatment for DMD.

13)  An AAV Vector-Mediated Micro-Dystrophin Expression Ameliorates Dystrophic Phenotypes of mdxMuscles

Miki Sakamoto,1 Madoka Yoshimura,1 Katsutoshi Yuasa,1 Toshifumi Yokota,1 Takaaki Ikemoto,2 Xiao Xiao,3 Yuko Miyagoe-Suzuki,1 Shin’ichi Takeda.1

Duchenne muscular dystrophy (DMD) is an X-linked, lethal muscle disorder caused by a defect in the dystrophin gene, and characterized by progressive muscle weakness, cardiomyopathy and early death. An adeno-associated virus (AAV) vector-mediated gene transfer is one of attractive approaches for the treatment of DMD, but it has a limitation in insertion size up to 4.9 kb. Therefore, a full-length dystrophin cDNA (14 kb) cannot be incorporated into an AAV vector. To find a short but functional dystrophin cDNA, we generated a series of rod-truncated micro-dystrophin (M3, 3.7 kb; AX11, 4.4 kb; CS1, 4.9 kb), and generated transgenic (Tg) dystrophin-deficient mdx mice expressing each of microdystrophins. CS1 Tg mdx mice showed lowest levels of serum creatine kinase, complete amelioration of muscle pathology, and nearly full restoration of contractile force (Biochem Biophys Res Commun. 293:1265, 2002). To test whether AAV vector-mediated CS1 micro-dystrophin expression can ameliorate the dystrophic phenotypes of mdx muscle, we constructed an AAV vector expressing micro-dystrophin CS1. We used skeletal muscle-specific MCK promoter to drive the CS1 gene, since the MCK promoter in AAV vector drives longer expression of the LacZ gene than the CMV promoter in skeletal muscle (Gene Ther. 23:1576, 2002). To reduce the length of CS1 cDNA, we deleted 5’- and 3’-untranslated regions and the coding region corresponding to exons 71- 78 (.CS1, 3.8 kb). To evaluate the therapeutic effects of AAV-MCK.CS1, we first injected the AAV vector into anterior tibial (TA) muscles of 10-dayold mdx mice. At this age,mdx mice show no signs of muscle degeneration. Therefore, it is easy to evaluate the therapeutic effects of the vector by counting the ratio of centrally nucleated myofibers. We then injected the vector into 5-week-old mdx mice whose muscles show active cycles of muscle degeneration/regeneration. When the AAV vector was introduced into 10-day-old mdx mice, the expression of micro-dystrophin continued for a long time, but dystrophin positive fibers scattered; 10 to 23% at 24 weeks after the AAV injection. H&E staining of muscle tissues showed nearly normal m orphology of dystrophin-positive fibers. In contrast, extensive expression of micro-dystrophin was achieved when 5-week-old mdx muscles were treated. At 8 weeks after the AAV vector injection, a large percentage of fibers were dystrophin-positive (10 to 50%). Even 24 weeks after the injection, 15 to 75 % of myofibers expressed micro-dystrophin. Dystrophin-positive fibers often had centrally located nuclei, however, ratio of these fibers was significantly reduced compared with that of dystrophin-negative fibers. We then isolated AAV-MCK.CS1-treated and non-treated mdx TA muscle and measured tetanic force. Non-treated mdx muscle showed reduced specific tetanic force, but AAV-injected mdx TA muscles showed moderate improvement of the specific force. Thus, our study demonstrated that .CS1 micro-dystrophin introduced by an AAV vector both before the onset of dystrophic changes and during ongoing muscle degeneration successfully protected mdx muscle.

14) Chronic Inflammation-Induced Extrasynaptic Utrophin Upregulation in Muscle Fibers of Immune Competent mdx Mice Is Related to Reduced Calpain Activity of Muscle

Ishrat Waheed,1 Renald Gilbert,1 Basil Petrof,2 Josephine Nalbantoglu,1 George Karpati.1

Chronic inflammation induced in the anterior tibialis muscles of immune competent mdx mice by intramuscular injection of a first generation adenovirus (FGAV) with strong beta galactosidase expression produces, by 30 days, appreciable amounts of extrasynaptic utrophin (utr). Utr is close structural and functional analogue of dystrophin (dys), and the amount of the extrasynaptic utr produced by chronic inflammation is sufficient to mitigate the deleterious effects of dys deficiency (reduced muscle fiber necrosis, restoration of dys-associated proteins, force generation impairment, etc.). Certain proinflammatory cytokines have been suspected to have a role in this process which, among other things, is supported by our finding that in certain types of immune incompetent mice (i.e. TNF-alpha gene-ablated), the above-described extrasynaptic utr upregulation does not occur. By contrast, in the IL-6 gene-ablated animals, utr levels remain unaffected, possibly because of vigorous compensation by IL-6 analogues. Here we also report that activity of calpain was significantly reduced in the inflammatory mdx muscles compared to non-inflammatory mdx controls (p<.0047). We also determined that in myotube cultures, utr is subject to calpainmediated proteolysis in a calcium-dependent manner. Furthermore, certain proinflammatory cytokines (IL-1 beta and IL-6, but not TNF-alpha) inhibited calcium-induced calpain activity. We suggest that inhibition of calcium-dependent muscle calpain activity by proinflammatory cytokines is related to the increase of the extrasynaptic utr, which is normally produced in small quantities and is subject to degradation by calpain. Inhibition of calpain would thus increase the extrasynaptic utr level to a new, higher-than-normal, steady state. Thus, inhibition of muscle calpain activity by other means may turn out to have a therapeutic role for dys deficiency.

15) Transfection of mdx and Normal Murine Muscle Fibers by Electrotransfer of Plasmid Encoding GalNac Transferase

Margaret Durko,1 Renald Gilbert,1 Josephine Nalbantoglu,1George Karpati.1

Recent work by P. Martin’s group demonstrated that in skeletal muscle fibers of mdx mice that are transgenic for CT GalNac transferase (GNT), abundant GNT was present in the extrasynaptic sarcolemma. (PNAS, 99:5616-5621, 2002). Coextensive with the overexpression of the extrasynaptic GNT, there was expression of extrasynaptic sarcolemmal utrophin, which was sufficient to mitigate some of the dystrophic features of the mdx muscles. We wished to determine if overexpression of extrasynaptic GNT starting in adult mdx muscle is also associated with extrasynaptic utrophin expression and mitigation of the dystrophic phenotype. Anterior tibialis (AT) muscles of adult mdx and normal CD1 mice were injected with 30 microliters (30-90 micrograms) of pyrogen-free plasmid containing an expression cassette of CMV-murine GNT cDNA (kind gift of Dr. P. Martin) and hyaluronidase (30 ul at 0.4 mg/ul). Subsequently, a train of eight 200 V/cm electrical pulses, each of 20 millisecond duration, was delivered by needle electrodes using an ECMA 30 electroporator to the AT muscles. Control muscles were not injected with plasmid but otherwise treated like the experimental ones. Ten-12 days later, transverse cryostat sections of the treated and control AT muscles were stained with biotinylated Wisteria Floribunda (WFA) or biotynilated Vicia Villosa (VVA) lectins with and without neuraminidase pretreatment, as well as with an antibody to GNT (gift of Dr. P. Martin) and utrophin. In several of the treated mdx and CD1 muscles, large groups of large diameter muscle fibers showed strong extrajunctional sarcolemmal WFA and VVA binding. Neuraminidase pretreatment increased peripheral staining of myofibers with WFA and VVA. However, the same fibers showed no extrasynaptic utrophin immunoreactivity even though synaptic utrophin expression was normal. We suggest that lack of extrasynaptic utrophin expression in the GNT muscle fibers may be explained by either that, unlike transgenic fibers, GNT-transfected mature mdx muscle fibers do not express utrophin, or the expression would take longer than 12 days. These points have practical importance if overexpression of extrasynaptic GNT is contemplated for the therapy of dystrophin deficiency.

16) Gene Transfer of vIL-10 To Improve Myoblast Survival in MDX Mice

Yong Li,1 William Forster,1 Levent Balkir,2 Paul Robbins,2 Johnny Huard.1,2

1Orthopeadic Surgery, University of Pittsburgh, Pittsburgh, PA, United States; 2Molecular Genetics and Biochemistry, University of Pittsburgh, Pittsburgh, PA, United States.

Myoblast transplantation is a potential therapy for Duchenne muscular dystrophy (DMD). However, the poor survival of myoblasts following transplantation has hindered the overall application of this technology. Immuno-rejection has been considered to be a major limitation following myoblast transplantation. Decreased immune rejection is known to increase the number of myoblasts that survive post-transplantation. Viral interleukin-10 (vIL-10) can suppresses autoimmune response after transplantation, and has already been found to prolong survival rates for different cells and tissues post transplantation in several animal models. We hypothesis that vIL-10 may improve myoblasts survival after transplantation due to its immunosuppressive function. A plasmid, encoding vIL-10 gene and the neomycin resistance gene, was transfected into myoblasts (C2C12) by lipofectin. The clone cells were selected by 500m g/ml G418 for 14 days after gene transfection. ELISA and western blot were performed to ensure that the clone cells were expressing of vIL-10. Both the selected clone cells (C2C12vIL-10) and normal myoblasts (C2C12) were retrovirally transduced to express a LacZ gene and then directly injected into gastrocnemius muscles (C2C12vIL-10 clone cells in left legs and C2C12 cells in right legs) of MDX mice (C57BL6J, age 6week). All mice were sacrificed for histological analysis at different time points (3days and 1,2,3,5,7 weeks) post cell transplantation. We also performed physiological testing at 7 weeks post transplantation. From counting the LacZ positive myofibers on both sides, we found that the surviving numbers of C2C12vIL-10 cells were greater than those of the control C2C12 cells. The immunohistochemical results show that the numbers of dystrophin positive myofibers in the C2C12vIL-10 injected muscle were significantly high than those of the control C2C12 injected muscle at different times point post transplantation. This increased survival of myoblasts resulted in improved MDX mouse muscle strength as assessed by fast twitch and tetanus strength tests 7 weeks post transplantation. Our results demonstrated that vIL-10 gene transfected myoblasts could survive longer than control non-gene transfected myoblast in MDX mice. This prolonged survival of transplanted myoblasts subsequently led to an increase in muscle strength of DMD muscle.

17) The Differentiation of Embryonic Stem Cells into Muscle Cells with Tet-Off System for Duchenne Muscular Dystrophy Therapy

Shiro Ozasa,1 Shigemi Kimura,1 Kaoru Ito,1 Makoto Ikezasa,1Takashi Hiranuma,1 Makoto Matsukura,1 Kowashi Yoshioka,1 Teruhisa Miike,1 Kimi Araki,2 Kuniya Abe,2 Kenichi Yamamura,2 Hitoshi Niwa.3

1Child Development, Kumamoto University School of Medicine, Kumamoto, Kumamoto, Japan; 2Department of DevelopmentalGenetics, Institute of Molecular Embryology and Genetics, Kumamoto University, Kumamoto, Kumamoto, Japan;3Laboratory for Pluripotent Cell Studies, RIKEN Center forDevelopmental Biology, Kobe, Hyogo, Japan.

Duchenne muscular dystrophy (DMD) is an X-linked recessive muscle disease caused by the deficiency of dystrophin. Various approaches, including cell transplantation and gene therapy, have been carried out in order to restore the missing dystrophin gene in DMD patients. Although gene therapy for DMD has shown some promise, viral vectors are limited to the injected area and we believe that DMD is a systemic disease. Conversely, muscle and/or bone marrow derived stem cell transplantation to dystrophic muscle bysystemic, intravenous injections has succeeded for the delivery of dystrophin (1, 2). These cells cannot retain their stem cell characteristics if cultured for a long time. In contrast, embryonic stem (ES) cells can be cultured for a long time, keeping their original condition and having the capacity to differentiate into all tissues and cell types. ES cells induced to differentiate in vitro gave rise to many cell types including hematopoietic precursors, cardiac and skeletal muscle, endothelium, and neural cells. Previously, it was impossible to limit differentiation of the ES cells to the muscle lineage. We have shown that ES cells can be induced to differentiate into myotubes by the establishment of genetically engineered ES cells harboring a tetracycline-regulated expression (Tet-Off System)(4) vector encoding the myogenic transcriptional factor, MyoD (3).

Supertargeting is a good strategy to make Tet-Off System-regulated ES cells. ZHTc6 is a mouse-derived feeder-free ES cell line, which carries Oct-3/4 cDNA regulated by the Tet-Off System. In the Tet-Off System, a cDNA artificially drives expression by the removal of tetracycline. The supertargeting vector, which was designed to replace the Oct-3/4 with MyoD cDNA and have neomycin resistant gene expression, was constructed for the supertargeting. Electroporation of the supertargeting vector to ZHTc6 and subsequent G418 selection were performed. Subsequently, 6 colonies were isolated, picked up, and screened by Southern blot analysis.

Results

The Southern blot analysis showed that the Oct-3/4 cDNA in 3  lones was replaced by MyoD cDNA by homologous recombination. The clones were then induced to differentiate into myotubes following the removal of tetracycline.

In this system, we can maintain ES cells undifferentiated in the presence of tetracycline and induce them to differentiate exclusively into myogenic pathways by the removal of tetracycline in vitro. Further in vivo and in vitro analysis is needed to delineate the possibilities of the clinical application of this system.

18) Matrylisin over Expression Enhance the In Vivo Migration of Myoblasts
Jean-Francois Lafreniere,1 Philippe Mills,1 E. Mostafa El Fahime,1 Jacques P. Tremblay.1
1Genetique Humaine, Centre de Recherche du CHUL, Ste-Foy,
QC, Canada.
The success of myoblast transplantation as a potential treatment for Duchenne muscular dystrophy (DMD) has been limited in part by the low dispersion of transplanted myoblasts. Injection sites at every millimeters is required to provide a good transplantation success. We previously showed that in vivo myoblast migration throughout the muscle extracellular matrix requires matrix metalloproteinase activity. We now demonstrate that over-expression
of matrilysin (MMP7) could helps to reduce the number of injection sites for the treatment of DMD patients. We constructed a retroviral vector including the matrilysin cDNA. RT-PCR and Western blots were used to verify that the retrovirus is effective and that the protein is expressed and secreted by the infected cells. After the evaluation of the infection rate by immunocytochemistry, the infected mouse myoblasts were transplanted and the in vivo migration was quantified after 60 hours by the micro-tube technique [ El Fahime E. et al : Exp Cell Res, 2000. ] Our results show that the over-expression of MMP7 improves the in vivo migration of mouse myoblasts. After 60 hours, the migration distance of normal mouse myoblasts is about 340 µm. After infection of only 25% of the cells with the retrovirus, the migration distance reach 961µm. These results suggest that matrilysin over-expression is a effective way to improve intramuscular migration and reduce the number of injection sites required for a successful myoblast transplantation.

19) Toward Optimization of Foamy Viral Infection of Myoblasts
Simon P. Quenneville,1 Joel Rousseau,1 Jaques P. Tremblay.1 1Genetique Humaine, Centre de Recherche du CHUL, Sainte-Foy, QC, Canada.
Myoblast transplantation is a possible therapy for several muscle affections such as Duchenne muscular dystrophy In this therapy, muscle precursor cells (myoblasts) are cultivated from a unaffected donor. These cells are then transplanted to the patient. However,  this method is limited by the immune system of the  recipient. It is possible to avoid rejection using immunosuppressive drugs like FK- 506, but these drugs are usually toxic. Ex-vivo gene therapy may be able to avoid these problems. The patient own myoblasts would be
genetically modified before their retransplantation. Several vectors are in principle able to do this task, but each of them have their problems and limitations such. We have investigated the possibility of using a new vector, i.e., the foamy virus, to infect myoblasts. We have produced a eGFP containing virus with a transient transfection system. The production was made in 293T cells. Virus were produced at titers around 105 particles per ml. The foamyvirus was found able to infect NIH 3T3, C2C12 cell line and normal human
myoblasts in culture. The viral infection was then optimized using centrifugation and several charged molecules like polybrene and chondroitin sulfate sulfate. Both of these methods improved the infection rate. We are thus currently trying to develop a foamy vector containing a mini-dystrophin gene and we will present the results at the meeting.

20) Microdystrophin Gene Therapy of Cardiomyopathy in Mdx Mice

Yongping Yue,1 Zhenbo Li,2 Scott Q. Harper,3 Robin L. Davisson,2 Jeffrey S. Chamberlain,3 Dongsheng Duan.1

Majority Duchenne muscular dystrophy (DMD) patients develop cardiomyopathy and many also die from cardiac failure. Encouraging gene therapy results have been achieved in the skeletal muscle of a mouse DMD model (mdx), yet few gene therapy attempts have been made in treating dystrophic cardiomyopathy. Unlike limb muscles, the anatomic location and the physiologic function of the heart present a formidable challenge to gene transfer. To overcome  his hurdle, we have developed a novel heart gene transfer method in  ewborn mdx mice. Anesthesia was achieved in 46 s (n=114) by placing newborn mice in a custom-designed ice-chamber. Recombinant adeno-associated virus (AAV) was delivered to the cardiac cavity with a glass microinjector. Injection was completed in 93 s (n=112). To optimize the method, we studied the effect of cold-shock time on recovery. Without injection, it took 528 s (n=7) and 453 s (n=16) for mice to recover from a 2 and a 5 min cold-shock respectively. In AAV injected mice, the overall recovery time was not significantly different from controls (2 min, 484 s, n=77; 5 min, 492 s, n=86). However, there was a slightly higher survival rate in  the 2 min group (96.43%) compared to the 5 min group (94.64%). The ubiquitous CMV and the muscle specific CK6 promoters have been used in mdx limb muscle gene therapy. To determine the most suitable promoter for cardiac muscle, we compared expression of AV.CMV.LacZ and AV.CK6.LacZ. Despite a similar level of expression in the limb muscle, the CMV promoter led to substantially higher expression in the cardiac muscle (2.96x106RU/mg, n=10) than the CK6 promoter (7.23x104RU/mg, n=10) at 1 m post-infection (pi). This result highlighted the importance of selecting appropriate expression cassettes for DMD gene therapy. DMD therapy needs longterm expression. To evaluate if neonatal gene transfer lead stableexpression in the heart, we compared AV.RSV.AP expression at 1and 6 m pi. Quantitative study revealed a 3-fold increase in AP+ heart cells at 6 m pi (238±38/section, n=6) compared to that at 1 m pi (82±24/section, n=6). It is possible that gene delivery to neonatal mice may have targeted some progenitor cells. Microdystrophin genes are highly truncated versions of the dystrophin gene. They hold tremendous potentials in DMD therapy. Since cardiomyopathyis an independent process, it is critical to evaluate these microgenesin the heart. One of the most important consequences of losing dystrophin is the disappearance of the dystrophin glycoprotein complex (DGC). The DGC mediates mechanical interactions between the cytoskeleton and the extra-cellular matrix. Since the C-terminal domain is completely deleted in microgenes, we investigated whether AAV-mediated microdystrophin expression could restore the DGC in the heart. Consistent with reporter gene transfer, AAV-mediated .R4-R23/.71-78 microgene transfer in neonatal mdx hearts resulted in profound gene expression in outer- and inner-layers of mdx hearts at 10 m pi (n=9). Importantly, immunofluorescent staining confirmed the successful restoration of other DGC components such as β-sarcoglycan and β-dystroglycan in heart cells. Our data have significant implications in the development of DMD heart genetherapy.

21) Highly Efficient and Comprehensive Gene Transfer to All Muscle Groups Necessary for the Potential Treatment of Duchenne/Becker Muscular Dystrophies Following In Utero Delivery of a Lentivirus Vector in Immuno-Competent Mice

Michael Themis,1 Simon Waddington,1 Maxine Holder,1 Susan Buckley,1 Kyriokos Mitrophanos,2 Brian Bigger,1 Laurane Lawrence,1 Terry Cook,1 Faisal Allaf,1 Susan Kingsman,2 CharlesCoutelle,1 Lisa Gregory.1 1Cell and Molecular Biology, Imperial College School of Medicine,London, United Kingdom; 2OxfordBiomedica, Oxford, United Kingdom.

Gene therapy for Duchenne and Becker muscular dystrophy (DMD/BMD) has so far not been successful because of the difficulty in achieving efficient and permanent gene transfer to a large number of affected muscles. In this study, we have investigated multiple routes of in utero vector administration applicable to DMD gene therapy. Using an Equine infectious anaemia lentivirus vector (EIAV) carrying the bgalactosidase reporter gene and immunocompetent mice we describe substantial gene delivery to the musculature involved in breathing and locomotion. To achieve widespread muscle transduction, vector was administered systemically and/or by direct injection into skeletal muscles and the peritoneal and thoracic cavities of fetal MF1 mice on day 15 of gestation (DG). Significant gene transfer to cardiac myocytes and low level skeletal muscle and diaphragm gene expression was observed following intravascular injection. High transduction levels of skeletal myofibres were achieved by direct limb muscle injection and of the diaphragm and intercostals muscles by intra-peritoneal, intra-thoracic and supracostal injections, respectively. Combined intravascular, intra-peritoneal and intramuscular injections were tolerated without increased mortality. Furthermore, germ-line analysis of mouse sperm using sensitive Taqman assay showed no germ-line gene transfer in fetally treated adult mice. Because we demonstrate high level and long term gene expression and have previously described tolerance against a therapeutic transgene in immuno-competent mice we suggest that the EIAV lentivirus vector carrying the dystrophin gene may be used for future rescue experiments in dystrophin deficient mdx mice. These findings support the concept of in utero gene delivery for the future prevention/correction of DMD in the clinic.

22) Ribozyme and Antisense RNA-Based Gene Therapies for Myotonic Dystrophy

Marc-Andre Langlois,1 Gilles Doucet,1 Nan Sook Lee,2 John J. Rossi,2 Jack Puymirat.1

1Department of Human Genetics, Laval University Medical Research Center, Quebec, QC, Canada; 2Division of Molecular Biology, Beckman Research Institute of the City of Hope, Duarte, CA, United States.

Myotonic Dystrophy type 1 (DM1) is an autosomal dominant neuromuscular disease characterized by myotonia, skeletal muscle weakness and wasting, cardiac conduction problems, cataracts and other multi-system dysfunctions. The molecular defect associated with DM1 is a large CTG expansion in 3’UTR of the DM protein kinase (DMPK) gene. Gaining evidence now suggests that specific targeting of mutant DMPK transcripts may prove successful in alleviating many of the cellular and molecular phenotypes seen in DM1. Until now, no method was shown capable of targeting preferentially the mutant transcripts. Here we have devised two approaches for targeting mutant DMPK mRNAs, one using antisense RNAs the other using nuclear-retained ribozymes. We have devised a retroviral system capable of expressing high levels of an antisense RNA directed to a site in 3’UTR of the DMPK mRNA encompassing the CUG repeats. We also developed a ribozyme expressed under the control of a modified tRNA-meti promoter. The modifications in this expression cassette causes nuclear retention of it’s expression products. Accessible ribozyme cleavage sites were identified within the 3’UTR of DMPK mRNA by semirandom oligonucleotide scanning and a hammerhead ribozyme was designed to cut the most accessible sequence tested. In vitro assays show that the antisense RNA produced reduces levels of mutant and normal DMPK mRNA by 50% and 80% respectively. This effect is also associated with restoration of glucose uptake, myoblast differentiation and reduction of CUGBP1 sequestration in the nucleus. Transient transfection of DM1 myoblasts with the wild type ribozyme expression cassette showsa 62% and a 50% reduction in mutant and normal transcript expression. In-situ hybridization using a Cy3-labeled DMPK probe showed a significant reduction in the number of nuclear speckles in ribozyme-transfected myoblasts. Destruction of mutant DMPK mRNA by the wild type ribozyme also coincided with insulin receptor splicing restoration by nearly 32%. Our results show for the first time that molecular targeting of the nuclear-retained mutant DMPK mRNA by antisense RNAs and ribozymes is effective in restoring some of the cellular and molecular phenotypes associated with DM1. Thus, these strategies offer a promise of possible therapeutics for the near future, once an effective and safe delivery system is developed.

23) Bone Marrow Derived Stem Cell Transplantation in the Canine Model of Duchenne Muscular Dystrophy (GRMD)

Luis Garcia,1 Catherine Escriou,2 Aurélie Avril,1 Stéphane Blot,2 Richard Mulligan,3 Olivier Danos.1

1Genethon CNRS UMR 8115, Evry, France; 2Ecole Nationale Vétérinaire de Maisons Alfort, Maisons Alfort, France; 3HHMI,Harvard Medical School, Boston, MA.

In mice, hematopoietic stem cells (HSC) have been reported to give rise to differentiated muscle cells following bone marrow transplantation. We have asked whether such a phenomenon can take place in the context of a large animal affected with muscular dystrophy. In this aim, we have performed allogeneic bone marrow transplantations in GRMD dogs, in which a mutation in the dystrophin gene is associated with typical symptoms of Duchenne Muscular Dystrophy. Dystrophin positive revertant fibers (0.01%) are occasionnaly observed in the muscles of GRMD dogs, probably due to exon-skipping (exons 4-9 and exons 5-12). Eleven GRMD animals received full body irradiation and weretransplanted with total bone marrow from normal, DLA-matched littermates. After hematopoietic reconstitution, blood chimerism was found to be over 95%. Although clinical investigations showed no obvious improvement of the muscular dystrophy condition, dystrophin positive fibers (0.1%) were detected in muscle sections prepared  from biopsies taken from the biceps and quadriceps muscles nine months after transplantation. Muscle injury using successive intramuscular injections of notexin did not significantly improve the number of positive fibers. These data suggests that the dystrophin positive fibers which appear following bone marrow transplantation originate from the circulating compartment and that the pool of resident muscle progenitor is not affected. The poor efficiency of the process could be due either to a low mobilization of the cells of interest from the circulating compartment to the muscle or to their limited ability to contribute to muscle . In order to test the myogenic potential of bone marrow cells, we performed intramuscular injections of either unfractionated bone marrow cells or stem cell-enriched (SP) bone marrow fraction in GRMD dogs. Only the injection of the SP fraction (104 cells) resulted in numerous dystrophin positive fibers spread over one centimeter near the site of injection (equivalent to the injection of 40x106 myoblasts). This confirmed the high plasticity of immature bone marrow cells when placed into the environment of the skeletal muscle. It suggested that the limiting factor for efficiently contributing to muscle reconstruction is the low amount of cells available in the circulating compartment or their inability to reach the muscle tissue. To address this issue, we performed local intramuscular injections of an adenoviral vector encoding the Stromal Derived Factor 1 (SDF-1), a chemokine known to attract HSC. Preliminary results indicate a significant enhancement of the number of dystrophin positive fibers after three weeks at the site of SDF-1 production.

24) Transplantation of Bone Marrow Stem Cells Expressing a Mini-Dystrophin/eGFP Fusion Protein into mdx Mice

Sheng Li,1 En Kimura,1 Robert W. Crawford,1 Paul Gregorevic,1 Leonard Meuse,1 Xin Ye,2 Brent Fall,1 J. M. Scott,1 Robert Welikson,3 J. C. Angello,3 Qiliang Li,2 G. Stamatoyannopoulos,2 S. D. Hauschka,3 J. S. Chamberlain.1

1Department of Neurology, University of Washington School of Medicine, Seattle, WA, United States; 2Department of Medicine,University of Washington School of Medicine, Seattle, WA, United States; 3Department of Biochemistry, University of Washington School of Medicine, Seattle, WA, United States.

Duchenne muscular dystrophy (DMD) is an X-linked recessive, lethal genetic disease caused by mutations in the dystrophin gene. Truncated dystrophin proteins, including mini-dystrophin (e.g. .H2-R19) and a dystrophin lacking the C-terminal domain, can fully prevent dystrophy in mdx mice, a model for DMD. Gene therapy for DMD will require delivery of therapeutic dystrophins to widely distributed muscles. However, methods do not currently exist for systemic delivery of dystrophin expression vectors, thus limiting their clinical application. The discovery of highly plastic adult stem cells has suggested a potential approach to systemically deliver dystrophin genes into muscles via the blood circulation. To facilitate the appraisal of rare transdifferentiation events, we generate da transgenic mouse strain carrying a MiniDys-eGFP fusion genedriven by the human alpha-skeletal actin promoter (HSA) on themdx background. The fusion gene encodes a fully functional minidystrophin( .H2-R19/ .CT) in which the C-terminal domain is replaced with an enhanced green fluorescence protein (eGFP) codingsequence. These transgenic mdx mice display normal musclemorphology with easily detectable GFP expression that is localizedon the sarcolemma of all skeletal muscles. Furthermore, thedystrophin-glycoprotein complex (DGC) linking cytoskeletal actin and the extracellular matrix was also restored. Functional analysis is underway to evaluate the contractile properties of the mini-dys/ eGFP positive myofibers. Whole bone marrow cells from thesetransgenic mice were transplanted into mdx mice to allow ascertainment of the number and extent of dystrophin and eGFPpositive myofibers in the recipients’ muscles at different time points. These studies should help address the potential for using manipulated bone marrow stem cells to achieve ex vivo gene therapy for DMD.

25) Systemic Delivery of Immunosuppressant Proteins and Dystrophin by High-Capacity Adenoviral Vector Provides Sustained Expression of Dystrophin in Skeletal Muscle of Mdx Mice

Zhilong Jiang,1 Paula R. Clemens.1,2

Adenoviral (Ad) vector-mediated gene delivery of normal, fulllength dystrophin to skeletal muscle provides a promising strategyfor the treatment of Duchenne muscular dystrophy (DMD), an Xlinked recessive, dystrophin-deficient muscle disease. Currentdifficulties preventing successful DMD gene therapy by Ad vectormediated gene transfer include the cellular and humoral immuneresponses induced by vector capsid and transgene proteins, ultimately diminishing expression of dystrophin. Our strategy to overcome this immunity and to prolong the expression of dystrophin in skeletal muscle is combined administration of an immunosuppressant protein high-capacity Ad (HC-Ad) vector expressing mCTLA4Ig or mCD40Ig with a therapeutic HC-Ad vector xpressing dystrophin. In this study, we determined whether systemic administration of immunosuppressant protein HC-Ad vectors expressing mCTLA4Ig and/or mCD40Ig with or without predepletion of liver macrophages with liposome-encapsulated clodronate (LEC) induces long-term expression of dystrophin in muscle of mdx mice intramuscularly injected with an HC-Ad vector expressing dystrophin at the same time. We found that 50-70% of muscle fibers were positive for dystrophin in the dystrophin HCAd vector-injected tibialis anterior muscle at 2 weeks and 9 weeks after administration in all groups (AdFK0 (control), AdmCTLA4Ig, AdmCTLA4Ig/LEC, and AdmCTLA4Ig/AdmCD40Ig). The intensity of dystrophin staining by immunohistochemistry and the number of vector DNA copies by real time PCR was stable between 2 and 9 weeks in groups treated with an immunosuppressant protein HC-Ad vector. In contrast, these parameters decreased with time in the negative control group. In the three CTLA4Ig-expressing groups, the immune response to gene delivery was markedly diminished, as assessed by the infiltration of CD4+ and CD8+ T lymphocytes in the injected muscle, the production of Th1 cytokines (IL-2, IFN-γ) and pro-inflammatory IL-6 from cultured splenocytes, and the activation of vector-specific cytotoxic T lymphocytes (CTL). Neutralizing antibodies against Ad vector were not significantly inhibited in the two groups expressing CTLA4Ig alone, but were significantly inhibited in the group expressing both CTLA4Ig and CD 40Ig, suggesting that co-expression of CTLA4Ig and CD40Ig is required for effective inhibition of a humoral immune response.  These studies demonstrated that systemic expression of CTLA4Ig alone or with CD40Ig effectively inhibited a cellular immune response and provided for stable expression of dystrophin in injected muscle.

26) Congenital Muscular Dysytrophy Gene Therapy with AAV-Mini-Agrin Vectors

Chunping Qiao,1 Tong Zhu,1 Juan Li,1 Jianbin Li,1 Xiao Xiao.1

1Melocular Genetics & Biochemistry, University of Pittsburgh, Pittsburgh, PA, United States.

Congenital muscular dystrophy (CMD) is an autosomal inherited disorder. It is a severe, progressive muscle-wasting disease that frequently leads to death in early childhood. There is, as yet, no way of treating this lethal disease. Most cases of CMD are caused by mutations in LAMA2, the gene encoding the alpha chain of the main laminin isoform expressed by muscle fibers. Using transgenic mouse model, Ruegg and colleagues have shown that the chick miniagrin —which binds to basement membrane and to dystroglycan—rescues dystrophic symptoms in a mouse model (lama2dy) for CMD. This intriguing result encouraged us to incorporate mini-agrin gene to treat the congenital muscular dystrophy disease. To pursue mini-agrin-mediated gene therapy on lama2dy mouse model, we first cloned the mice mini-agrin cDNA by RT-PCR. After careful sequencing examination, the correct mouse mini-agrin DNA fragment was incorporated into AAV vector plasmid containing CMV promoter and SV40 polyA signal sequence. In vitro Western blot showed that pAAV-mini-agrin had robust expression of mini-agrin protein, especially as a secretable form. AAV serotype 2 vectorcarrying mini-agrin gene was delivered to 10-days-old lama2dy young mice by intramuscular injection. Immunofluorescent staining indicated high-level expression of mini-agrin protein in injected muscle. More importantly, H&E staining showed that mini-agrin positive area has a significant muscle morphology improvement compare with agrin negative area. Collagen III staining also demonstrated less fibrosis occurred in agrin positive area. To achieve better therapeutic effect, systemic delivery of mini-agrin protein to the whole body of lama2dy mouse by AAV vector is underway. This study will provide a potential treatment for CMD.

27) Stable Correction of a Genetic Defect by In Vivo Targeting of a Stem Cell Population with a Lentiviral Vector

Gary P. Kobinger,1,2 Jean-Pierre Louboutin,1,2 Elizabeth R. Barton,1 H. Lee Sweeney,1 James M. Wilson.1,2

1University of Pennsylvania School of Medicine, Philadelphia, PA; 2The Wistar Institute, Philadelphia, PA.

Successful gene replacement therapy will require stability of the genetic graft or therapeutic gene in the corrected tissues. Stability of the genetic graft can be assured with integrating or self replicating viruses by targeting post-mitotic cells that have a very long life-time  r stem cells that can replenish defective tissue with corrected cells. Most differentiated cells that are targets of gene transfer turnover with frequencies that could require repeated delivery of vector or vector modified cells at intervals ranging from weeks to months. Analternative strategy is to genetically modify stem cells capable of stably repopulating diseased tissues. This has been achieved in several models in which stem cells are genetically modified ex vivo prior to transplantation. In this study, we explore the possibility of targeting a stem cell population in situ through in vivo administration of vector. For developing this concept, we have selected the mouse model of muscular dystrophy (mdx mice) that undergo rapid turnover of their muscle fibers. Results reveal that in vivo targeting of muscle stem cells, notably satellite cells, with a pseudotyped lentiviral vector encoding for the minidystrophin restores dystrophinexpression and provides functional correction in skeletal muscle of mdx mice. Targeting of satellite cells in mdx mice with Ebola or MuLV pseudotyped HIV vector expressing dystrophin results in the regeneration of functionally and phenotypically corrected mature skeletal muscles. This study shows that progenitor cells can be genetically engineered in vivo and subsequently proliferate into terminally differentiated tissue carrying the genetic graft in a way that stably corrects function.