QUARTO ENCONTRO ANUAL DA SOCIEDADE AMERICANA DE TERAPIA GÊNICA - RESUMOS RELACIONADOS COM DISTROFIA MUSCULAR

Abstract No. 68

Antisense-mediated exon skipping and restoration of dystrophin synthesis in muscle cells from DMD patients

Dept. of Human and Clinical Genetics, Leiden University Medical Center, Leiden, The Netherlands

Abstract
Duchenne muscular dystrophy (DMD) is a lethal muscle disease typically caused by frame-shifting mutations in the DMD gene that impede the synthesis of the dystrophin protein. In the milder allelic Becker muscular dystrophy (BMD), the mutations do not disturb the reading frame and a still partly functional protein is expressed. In this study, we explored a therapeutic strategy aimed at restoration of the reading frame in DMD patients through targeted modulation of dystrophin pre-mRNA splicing. Considering that exon 45 is the single most frequently deleted exon in DMD, whereas exon 45+46 deletions only cause a milder form of BMD, we designed a system to induce exon 46 skipping. Following transfection of antisense oligoribonucleotides (AONs) complementary to a putative splicing regulatory sequence within exon 46, we induced the skipping of this exon from the transcript in cultured myotubes of both mouse and human origin. Immunohistochemical and RT-PCR analysis of transfected myotubes from two unrelated DMD patients carrying an exon 45 deletion, showed that the skipping of exon 46 to a level of 20% allowed the synthesis of normal amounts of properly localized dystrophin. Our results demonstrate the therapeutic potential of antisense oligonucleotides to restore dystrophin expression from the endogenous gene in DMD patients. This gene therapy strategy should be applicable to a variety of DMD mutations, as well as other genetic diseases.

Abstract No. 178

A Novel Electrode Design for Efficient Electroporative Gene Delivery

Center for Pharmacogenetics, School of Pharmacy, University of Pittsburgh

Abstract
Recently, gene delivery in vivo using electroporation has became an area of greater research interest. The prospect of utilizing electroporation for gene transfer in vivo is evident based on the success of delivering genes in vitro and effective delivery of drugs in electrochemotherapy studies. Although the detailed molecular mechanism of electroporation is still not completely understood, there is agreement in the literature that the field strength induces some sort of metastable structure defect in the membrane which serves as a pathway from the extracellular space to the cell interior. Electroporation is a threshold phenomenon. The field strength necessary for gene delivery must exceed a threshold value. Moderate increases in the applied field strength resulting delivery; however, large increases of the applied field from this threshold result in tissue damage. Here, we report a novel design of an injectable electrode, which utilizes comparatively much lower electric field strength to achieve high transfection efficiency. The hypothesis of this design was based on two electroporation parameters we defined. One is the Effective Threshold Value (ETV), which is required for the electrotransfer and is defined by the cell and its environment. Another one is the Apparent Threshold Value (ATV), which is defined by the electroporater. ATV is usually much greater than ETV. Theoretically, the field strength is reversely proportional to the distance from its source. Therefore, when the field strength is delivered from the conventional electrodes to the injection site of DNA, the field strength is dissipated due to the distance between the injection site and the electrode. To compensate for the loss in the field strength, higher ATV has to be applied which results in toxicity to the tissue. A logical approach to decrease ATV is to directly deliver the field strength to the injection site. In this case, ATV is approximately equal to ETV such that gene transfer can be achieved at lower field strength. This hypothesis was supported by our data. The maximal level of gene transfer was achieved at as low as 50 V/cm of field using an injectable electrode, whereas 200 V/cm was required by using a conventional electrode to achieve the similar level of gene transfer. Histochmical analysis showed that no muscle damage was observed using the new electrode. However, severe damage was induced by using the conventional electrode. We conclude that the new injectable electrode represents a significant improvement in elctroporative gene transfer to tissues.
Supported by NIH grants CA74918, DK54225, AR45925, DK44935 and by Muscular Dystrophy Association of America.

Abstract No. 202

Enhancement of Electroporation-Mediated Gene Expression In Muscle by Enzymatic Modulation of Extracellular Matrix

Center for Pharmacogenetics
School of Pharmacy
University of Pittsburgh

Abstract
Gene delivery to skeletal muscle is a promising method for the treatment of muscular diseases as well as for the production of therapeutic proteins. Electroporation has been proved to be an efficient method for gene delivery to muscle. In this study, we report a significant enhancement of electroporation-mediated gene expression in skeletal muscle by injecting plasmid DNA mixed with hyaluronidase before the application of electroporation. Luciferase gene expression was about 10 fold higher in mice injected with DNA and hyaluronidase, compared with mice injected with DNA alone. Hyaluronic acid showed no effect on promoting gene expression. Moreover, the enhancement of gene expression required enzymatic activity, because co-injection of DNA with denatured hyaluronidase or bovine serum albumin had no effect on gene expression. More importantly, the area of the muscle that expressed transgene product was increased very significantly in mice treated with hyaluronidase. More than 40% of muscle fibers expressed b-galactosidase when mice were co-injected with a plasmid containing LacZ gene and hyaluronidase. On the other hand, less than 5% of muscle fibers expressed b-galactosidase in mice injected with plasmid alone. Supported by NIH grant AR45925 and by Muscular Dystrophy Association of America.

Abstract No. 252

In Vivo and In Vitro Expression of Transgenes in Mammalian Vestibular Hair Cells Infected with [E1-], and [E1-, E2b-] Adenoviral Vectors

Harvard Medical School, Boston, MA 02114
Duke University Medical Center, Durham, NC 27710
University of Miami School of Medicine, Miami, FL 33136

Abstract
To understand the functions of specific proteins in vestibular sensory hair cells, it is necessary to add or inactivate those proteins in a system where their physiological effects can be studied both in vitro and in vivo.
The loss of vestibular hair cells, or the loss of their capacity to transduce vestibular stimuli, is the cause of many balance disorders. In order to develop viral vectors that could allow an understanding of specific protein functions, and for the potential treatment of balance disorders, recombinant adenovirus vectors were used to infect vestibular hair cells of the guinea pig inner ear in vivo, and cultured mouse vestibular endorgans. Recombinant adenovirus vectors [E1-], [E1-, E2b-] containing either green fluorescent protein (GFP) or lacZ as a reporter gene were used in this study. [E1-] adenovirus vectors (105-108 pfu/ml) were found to be toxic to cultured vestibular hair cells, such that by 48 hrs after infection, there were no intact stereocilia bundles. On the other hand, [E1-, E2b-] adneovirus vectors (108 pfu/ml) were not toxic to cultured vestibular hair cell stereocilia up to 2 weeks post-infection. The [E1-, E2b-] adenovirus vector (5 x 108 pfu/infused ear) was also capable of driving transgene expression to vestibular hair cells in vivo.
The development of a viral vector that infects hair cells of the vestibular system in vivo and in vitro without ototoxicity, is crucial for gene replacement therapy to be effectively used to treat balance disorders.
This work was supported by grants from the Public Health Service (DC03086, DK52925, DC00304), Muscular Dystrophy Association, USA, and funds from Glaser Foundation and the University of Miami's Chandler Chair.

Abstract No. 264

Using transgenic CMT1A mice as preclinical model.

INSERM U491, Fac de Medecine, 13385 Marseille Cedex5 France

Abstract
We have constructed 4 mouse lines by introducing a human YAC containing the human gene PMP22 into the murine oocyte (Huxley et al, 1996; 1998). This gene has been involved in the Charcot-Marie-Tooth type 1A disorder, the most frequent inherited peripheral neuropathy (1/3000). Patients suffering from this disease present a duplication (1.5 Mb) of the proximal short arm of chromosome 17, which includes the PMP22 gene. Two transgenic lines exhibit a peripheral neuropathy transmitted as an autosomal dominant trait, and clinical signs of transgenic mice are very similar to the human disorder. In a further step, we have explore the model, and these data clearly demonstrate the following points: - The neuropathic phenotype linked to PMP22 overexpression, appears by a non linear mechanism. - The peripheral neuropathy is not a demyelinating neuropathy, but a dysmyelinating. – Disorders linked to PMP22 overexpression are in fact pathologies of the differentiation of Schwann cells.
In a last step, we used these CMT mice, as a preclinical model, in order to test therapeutical approaches (we may note that it is a human YAC we have introduced in the mouse). In a first attempt, we have tested a battery of molecules susceptible to act on peripheral nerve and myelination. We tested transgenic mice from the same litter either with an active molecule or a placebo. We choose, as inclusion criteria, only males (more severely affected than females) of an age of two month (clinical signs appear about one month after birth in this line). Muscular capacity has been evaluated using a rotarod test, and at the end of the test, by measuring the percentage of nerve fibers with myelin on sciatic histological slices. Our two first set of experiments was very promising, as one molecule is reverting the phenotype with a great efficiency, allowing us to envisage the first clinical trials on CMT1A patients.

Abstract No. 367

Lentiviral Vectors in the Fetus: Distribution and Efficiency is Dependent on the Route of Administration and Vector Pseudotype

Children’s Institute for Surgical Science, Children’s Hospital of Philadelphia, Philadelphia, PA 19104
Institute for Human Gene Therapy, Philadelphia, PA 19104
The Salk Institute, La Jolla, CA 92037

Abstract
In utero gene therapy is a promising technique to allow long-term correction of congenital genetic diseases. The early gestational fetal environment may provide unique biological and practical advantages for gene therapy in that actively proliferating stem cell compartments are accessible. Given the potential of lentiviral vectors to transduce stem cells, we sought to define the biology of these vectors following in utero administration in pre-immune mouse fetuses.
Balb/c fetuses at 14-15 days’ gestation were injected with lentiviral vectors carrying the transgene lacZ under the control of the human CMV promoter in one of three routes: intraperitoneal (IP), intrahepatic (IH), or intramuscular (IM). The lentiviral vectors were pseudotyped with either VSV-G, Mokola, or Ebola virus proteins at a titer of 1e9 or 1e10 infectious units (IU)/ml (VSV-G), 2e8 IU/ml (Mokola) or 3e8 IU/ml (Ebola). 5 ul of virus was introduced directly into each fetus using a pulled glass micropipette. The pups were harvested at time points between 5 days to 19 weeks following injection. Detailed histologic assessment was performed by analyzing 1-2 um plastic sections of X-gal stained tissues. Mortality following IH injection was compared to control pups injected with PBS alone.
Survival to birth following direct intrahepatic injection was 53% with PBS, 42% with VSV-G, 56% with Ebola, and 50% with Mokola pseudotyped virus. IH injected animals showed gene expression in liver hepatocytes and peri-ductular cells, with the comparative efficiency of each pseudotype being Mokola>VSV-G>Ebola. IM injection resulted in transduction of myocytes, satellite cells, neuromuscular spindles, and perivascular cells, with greatest efficiency from the Mokola-pseudotyped vector. Systemic spread of the virus was documented by transduction of cardiomyocytes after IP, IH, and IM injection with the VSV-G-pseudotyped vector and after IH and IM injection with the Mokola-pseudotyped vector. Expression has persisted in liver, heart, and muscle up to the time of last analysis at 19 weeks following injection without evidence of an inflammatory infiltrate in the transduced tissues.
Long-term gene expression in multiple tissues can be achieved after in utero administration of lentiviral vectors with minimal apparent toxicity. The efficiency and distribution of transduction after in utero administration was highly dependent upon the route of administration and the pseudotype of vector used. Our findings of long term transduction of skeletal myocytes, satellite cells, cardiomyocytes, and hepatocytes after in utero administration suggests a novel strategy for the treatment of congenital muscular dystrophies, coagulopathies, and inborn errors of metabolism.

Abstract No. 394

Functional Correction of Muscular Dystrophy by Four-Repeat Micro-Dystrophins

Department of Neurology, University of Washington, School of Medicine, Seattle, WA 98115-7720
Cellular and Molecular Biology Program, University of Michigan Medical School, Ann Arbor, MI 48109
Department of Physiology, University of Michigan Medical School, Ann Arbor, MI 48109

Abstract
The goal of this study was to evaluate the correction of muscular dystrophy in a mouse model (the mdx mouse) using human dystrophin transgenes (micro-dystrophins) that contain a series of large deletions sufficient enough to allow packaging into the small (~5 kb cloning capacity) gene therapy vector, adeno-associated virus (AAV). AAV is the most efficient skeletal muscle delivery vector that does not generate a significant immune response, yet it is incapable of accommodating a full-length dystrophin cDNA (14 kb). The creation of functional micro-dystrophin expression vectors of ~4.6 kb in size is therefore essential to develop a successful AAV-mediated gene therapy strategy for Duchenne muscular dystrophy (DMD).
DMD is an X-linked, recessive disorder caused by defects in the dystrophin gene. It is the most common form of muscular dystrophy and is characterized by progressive muscular degeneration and weakness, leading to death in the early twenties. The full-length dystrophin gene product is a 427 kDa, multi-domain protein that is localized to the subsarcolemmal membrane and is believed to provide a critical link between the actin cytoskeleton and the extracellular matrix via interactions with the dystrophin associated protein complex, or DAP. The binding motifs of the dystrophin molecule that are necessary for assembly of the DAP complex and subsequent stabilization of the muscle membrane are located near the N and C termini, while the majority of the protein lies in the central rod domain, which is comprised of 24 spectrin-like, triple helical repeats spaced by four non-repeat hinge regions. Natural in-frame deletions in this region often result in a less severe form of muscular dystrophy, called Becker muscular dystrophy.
A recent paper in PNAS vol. 97 (Wang et al., 2000) demonstrated that dystrophin molecules with large C terminal and rod domain deletions (maintaining 5-6 repeats) can be delivered by AAV and act to prevent the development of muscular dystrophy in young mdx mice. Although this work suggested that protection can be achieved in limb muscle by a micro-dystrophin molecule, it left unanswered whether these dystrophins possessed the functional capacity to protect against the harsh forces associated with damaging eccentric contractions, and whether this protection would extend to the mdx diaphragm, which is the most severely affected muscle in mice and humans.
We generated transgenic mice expressing several different micro-dystrophin clones containing only four repeats, and analyzed their ability to correct muscular dystrophy by both morphological and physiological assays. Two of three transgenes significantly reduced the amount of dystrophy observed in the mdx limb and diaphragm muscles. The best transgene produced muscle with wild-type levels of central nuclei, Evans Blue dye exclusion, resistance to contraction-induced injury, and the ability to run distances on a treadmill. In addition, no areas of fibrosis nor monocyte infiltration were observed. We are currently evaluating these micro-dystrophin clones in adult animals using AAV vectors.

Abstract No. 393

Sequences permitting episomal plasmid replication in primary human myoblasts

Human Genetics Laboratory, Laval University Hospital Center, Sainte-Foy, Canada.
Institut fur Zellbiologie, Universitat Witten/Herdecke, Stockumer Strasse 10, D-58448 Witten, Germany.

Abstract
Many human gene therapies require the use of non-immunogenic vectors for stable gene expression. Stable gene expression can be mediated by an integrated or episomal DNA sequence. Episomal vectors offer the advantage of not modifying the chromosomes, but most episomal vectors require the expression of viral genes. A plasmid with a Scaffold/Matrix-Attached Region from the human interferon- gene and a SV40 replication origin has previously been described as mitotically stable in chinese hamster ovary cells. We introduced the EGFP gene into this plasmid and transfected it into primary human myoblasts. This plasmid replicated episomally for more than two months under selective pressure. Intact episomal plasmid DNA could be recovered and used to transform E. coli cells. This plasmid is thus very interesting because it permits efficient episomal replication in primary human cells without requiring the expression of any viral gene. This plasmid would be particularly usefull in our field of autologous myoblast transplantation for Duchenne muscular dystrophy since human myoblasts are difficult to transfect with large plasmids (containing the dystrophin cDNA) and the stability of the DNA introduced into a few cells is desired. Research is underway to characterize the segregation efficiency of this plasmid, experiment alternate replication origins and introduce the dystrophin cDNA.

Abstract No. 395

Long-term gene therapy and contractile force recovery in a mouse model of Duchenne muscular dystrophy (DMD) after AAV-mini-dystrophin vector treatment

Department of Molecular Genetics and Biochemistry, Department of Pediatrics, University of Pittsburgh School of Medicine, Pittsburgh, PA15261

Abstract
Duchenne Muscular dystrophy (DMD) is one of the most common, disabling and lethal genetic diseases afflicting one of every 3500 boys. The patients suffer from progressive muscle degeneration and weakness, become wheelchair-bound in their early teens and die prematurely in their late teens and early twenties. No effective treatment is currently available for DMD. The etiological cause of the disease is the mutation in the dystrophin gene. The lack of a functional dystrophin protein in the striatal muscle causes instability of the muscle cell membrane and results in persistent damage by muscle contraction. Because DMD is X-linked and genetically recessive, it has been considered as a good candidate for gene replacement therapy. Adeno-associated viral (AAV) vector has proven to be the most promising vector system, which renders efficient and stable gene delivery in muscle tissues without triggering cellular immune responses to the transgene products. Previously, a major hurdle preventing the use of AAV vectors for DMD gene therapy is the small capacity of AAV (5 kb) and the immense size of the dystrophin gene, which spans over 3 million basepairs on the X-chromosome and has a cDNA of 14 kb. Recently, we have overcome this hurdle by selectively truncating the non-essential regions of the human dystrophin and created novel minidystrophin genes that can be packaged into AAV vectors. Injection of AAV-mini-dystrophin vectors into the hindleg muscles of mdx mice, a DMD animal model, effectively restored the missing dystrophin protein along with the dystrophin associated protein complex onto the muscle cell membrane. Moreover, AAV vector treatment alleviated the muscle histopathological conditions and protected the muscle cell membrane integrity for over a year.
In this report, we show that AAV-mini-dystrophin vector treatment in the mdx mouse muscle is also able to improve the contractile forces. To test the effect of muscle contractile force recovery, adult male mdx mice at the age of 1.5 to 2.5 months were injected with AAV-mini-dystrophin on the tibialis anterior (TA) muscle of the left legs, while the right legs remained untreated as the internal control. At the time of vector treatment, the muscle tissues already went through massive waves of degeneration and regeneration, and exhibited overt signs of pathology. At 7 months after vector injection, the AAV treated and untreated muscles were isolated for in vitro contractile force tests. The experimental results showed that 1) the specific peak force was slightly higher after vector treatment (P=4.08±0.53N/cm2, n=8) when compared to the untreated control muscle (P=3.75±0.65N/cm2, n=8); 2) the specific tetanic force was 30% higher in the vector treated muscles (P=13.05±2.00N/cm2, n=8) than the untreated control muscles (P=10.30±1.56N/cm2, n=8). 3) during the eccentric lengthening tests, the vector treated muscles retained significantly higher contractile forces than the untreated muscles. Immunofluorescent staining of minidystrophin in the AAV vector treated muscles after in vitro force measurement showed 20% to 50% of the myofibers positive for dystrophin expression. Measurement of the age-matched normal male mouse TA muscle contractile forces is in progress. In summary, this study demonstrated that the mini-dystrophin can recover the contractile force deficits in the adult dystrophic muscles of the mdx mice, therefore, further supporting the therapeutic functionality of the minigenes delivered by AAV vectors.

Abstract No. 531

Production of Adeno-associated Virus Vectors by Cre-loxP Recombination System

Department of Neurology and Neurosurgery, Montreal Neurological Institute, McGill University, Montreal, Quebec, Canada

Abstract
Production of Adeno-associated Virus Vectors by Cre-loxP Recombination System*

Qingshan Teng, Josephine Nalbantoglu, George Karpati, Paul C. Holland
Department of Neurology and Neurosurgery, Montreal Neurological Institute, McGill University, Montreal, Quebec, Canada

Adeno-associated viruses are attractive vectors for the delivery of foreign genes into mammalian cells for gene therapy. The production of AAV vectors usually involves transfection of 293A cells with an AAV plasmid which contains a foreign gene expression cassette flanked by AAV terminal repeat sequences, the AAV helper functions rep and cap, and adenovirus genes E1A, E1B, E2A, E4 and VA RNA. We describe a novel approach to produce AAV vectors by using the prokaryotic Cre-loxP recombination system. Cells that stably express the Cre recombinase (293Cre4) were cotransfected with AAV vector plasmids and AAV helper plasmids containing rep and cap, followed by infection with AdLC8cluc helper viruses. The AdLC8cluc helper viruses have their packaging signals flanked by loxP sequences, rendering them unpackageable in 293Cre4 cells. However, they retain the ability to replicate and provide all the functions necessary for packaging of AAV vectors. We tested this strategy to produce AAV-LacZ, a recombinant AAV containing the reporter gene E.coli LacZ. 293Cre4 cells were cotransfected with the AAV vector AAV-LacZ and AAV helper plasmid ACG2, followed by infection with AdLC8cluc helper viruses. The cells were harvested 72 hours post-infection. AAV-LacZ virus was titrated on 293A cells by staining for the LacZ product, beta-galactosidase. AAV vectors produced by this scheme yielded 6.0 x 10⁷ blue cell forming units/1.2 x 10⁶ cells. Contamination with adenovirus was estimated less than one in 10⁴ AAV functional particles. The use of such packaging deficient helper adenovirus may circumvent some of the problems associated with the large-scale production of recombinant AAV.

*Supported by Muscular Dystrophy Association (USA).

Abstract No. 785

FUNCTIONAL CORRECTION OF ADULT MDX MOUSE MUSCLE USING GUTTED ADENOVIRAL VECTORS EXPRESSING FULL-LENGTH DYSTROPHIN.

Department of Physiology, University of Michigan, Ann Arbor
Department of Neurology, University of Washington, Seattle.
Cellular and Molecular Biology Program, University of Michigan, Ann Arbor
Department of Human Genetics, University of Michigan, Ann Arbor

Abstract
Duchenne muscular dystrophy (DMD), a degenerative lethal muscle disorder that affects one in 3500 newborn males, is caused by mutations in the dystrophin gene. Adenoviral vectors are promising tools that may be used to deliver the dystrophin gene to affected muscle to alleviate the symptoms of DMD. First generation adenoviral vectors, however, elicit a potent immune response that prevents long term transgene expression, and are limited in cloning capacity. Therefore, we have constructed "gutted" adenoviral vectors that are devoid of all viral genes and contain a full-length human (HDys) or mouse (MDys) dystrophin cDNA driven by a muscle specific promoter. Preliminary five day experiments in six week old mdxmouse tibilais anterior (TA) muscles injected with either viral construct showed high levels of dystrophin protein expression and exclusion of the vital dye Evans Blue, demonstrating improved muscle sarcolemma integrity in the presence of dystrophin. Injections into adult (11-13 mo.) mdxmouse TA muscles were performed and a novel eccentric (lengthening) contraction protocol was used to stringently test for restored muscle function. This protocol demonstrates marked differences between mdxand wild type TA muscle susceptibility to contraction-induced injury; after one injury-inducing lengthening contraction, wild type TA muscle force is reduced by 10%, while mdxmuscles demonstrate a 72% force loss. After five days, HDys injected muscles were significantly protected from injury and showed high levels of dystrophin expression in contrast to the sham injected contralateral control muscles. However, 25 days after injection, significant decreases in force generating capacity were detected. This loss of force generation was similar to that observed after injection of a first generation virus containing a ß-galactosidase transgene (CNß), suggesting that an immune response against human dystrophin protein was elicited. MDys injected muscles demonstrated high levels of dystrophin protein expression and no functional defects after 25 days. In addition, MDys injected muscles were able to produce 62% of wild type force levels after one lengthening contraction that, in contrast, reduced mdxand sham injected control muscles to 38 and 29% of wild type force, respectively. FACS sorting of injected muscles revealed CD4+ and CD8+ cells present in all virus-injected muscles. MDys injected muscles showed the least amount of infiltration, HDys injected muscles showed approximately 1/3 more immune cells than MDys injected muscles, and CNß injected muscles displayed the highest amount of immune cell infiltration. Collectively, these data demonstrate 1) efficient gutted adenoviral vectors are successful in transducing dystrophin in TA muscles of one year old mdxmice, 2) evidence for an immune response against the human dystrophin protein that causes atrophy and loss of function, and 3) a partial functional correction of adult mdxmouse muscle after delivery of full-length mouse dystrophin with a gutted adenoviral vector.

Abstract No. 786

High-efficiency and Long-term Gene Transfer in Cardiomyopathic Hamster Hearts by AAV Vectors

Abstract
1 Department of Molecular genetics and Biochemistry
2 Department of Orthopedic Surgery, and 3 Department of Surgery
University of Pittsburgh, School of Medicine, Pittsburgh, PA 15261

Limb girdle muscular dystrophy (LGMD) 2F is caused by mutations in the d-sacroglycan (SG) gene. Previously, by directly injecting an adeno-associated virus (AAV) vector carrying human d-sacroglycan gene into the skeletal muscle, we have shown successful genetic and biochemical rescue and muscle force recovery in the Bio.14 hamster, a homologous animal model for LGMD 2F(J. Li et al, Gene Therapy 6:74; X. Xiao et al, J.Virology, 74:1438). Like the human patients, the Bio14.6 hamsters also suffer from cardiac myopathy and die prematurely. Hence, the animal has also been widely used as a model of congestive heart failure. In this report, using a heterotopic heart transplant model, we show high-efficiency and long-term intra-cardiac gene transfer by AAV vectors carrying either the reporter Lac-Z gene or the therapeutic d-sarcoglycan gene.
In this study, the 2-month old dystrophic Bio14.6 hamsters and healthy control F1B hamsters were subjected to the heterotopic heart transplantation with the syngeneic donor hearts, which had a single perfusion ex vivo with the AAV vectors via the aorta. In detail, the vectors were delivered into hamster coronary circulation by perfusion through aorta after the heart was bathed in cold Ringer’s buffer for 30 minutes before transplanted into recipient hamsters. For each donor heart, 0.5 ml (5x1012 viral particles/ml) of AAV-LacZ or AAV-SG was used for the ex vivo perfusion. To examine the gene transfer efficiency, the recipient hamsters were sacrificed at four-month and one-year after heart transplantation. Cryo-thin-sections (10 mm) of the heart tissue were subjected to immunofluorescent staining for d-SG expression and X-gal staining for Lac-Z expression. Our experiments demonstrated 1) highly-efficient AAV-LacZ vector transduction in over 90% of the cardiac muscle tissues for one year in the F1B normal hamsters; 2) highly efficient AAV-SG vector transduction in 80% to 90% of the cardiac muscle in the Bio 14.6 hamster heart for four month (the duration of the experiment). 3) no immune rejection of the transplanted hearts that expressed either E. coli LacZ or the human SG proteins. Our studies suggest that the Bio 14 hamster is an excellent LGMD animal model for gene therapy, and that AAV is the vector of choice for the long-term intra-cardiac gene therapy strategy. Further studies are needed to evaluate the physiology and histopathology improvement after AAV vector-mediated gene therapy in the cardiomyopathic hamster hearts.

Abstract No. 787

Markedly enhanced dystrophin expression without inflammation in mdx muscle after transduction with a fully deleted (gutted) adenovirus encoding two dystrophin cDNAs regulated by a strong promoter

Montreal Neurological Institute, McGill University, Montreal, QC, Canada

Abstract
Duchenne muscular dystrophy (DMD) is a relatively common inherited neuromuscular disease caused by deficiency of dystrophin. A potential and useful treatment of DMD by gene therapy will require the efficient and prolonged expression of dystrophin in most muscles of the body. Helper-dependent adenovirus (HDAd), also known as gutted adenovirus (Ad), are deleted of most viral genes and have permitted the safe an long term expression of various transgenes in several tissues. We have previously constructed a HDAd encoding the full-length dystrophin regulated by the cytomegalovirus (CMV) promoter (HDAdCMVDysFl), and have observed a weak dystrophin expression after gene transfer with this vector in muscle. In the present study, in an attempt to increase dystrophin expression, we have constructed a HDAd encoding a single copy of dystrophin regulated by the strong hybrid CMV enhancer/-actin promoter (CB) (HDAdCBDys) and a HDAd encoding two copies of dystrophin regulated by the CB promoter (HDAdCBDysDys). To evaluate dystrophin expression from these vectors, the tibialis anterior muscle of neonatal mdx mice (the mouse model for DMD) was injected with HDAdCMVDysFl, HDAdCBDys or HDAdCBDysDys at a titer ranging from 1.0 to 2.0 X 10¹² particles/ml, and the number of dystrophin positive fibers was determined by immunohistochemistry. At the earliest time point investigated (10 days post-injection), the average number of dystrophin positive fibers was higher after gene transfer with HDAdCBDysDys (mean = 240) compared to HDAdCBDys (mean = 70) and to HDAdCMVDysFl (mean = 20). At 30 days post-injection, the number of dystrophin positive fibers remained the same after gene transfer with HDAdCBDys and HDAdCBDysDys, but in some muscles, nearly 100% of fibers were dystrophin positive. In conclusion, in the backbone of a HDAd the CB promoter confers a much stronger dystrophin expression than the CMV promoter, which persists unabated for at least 30 days without any evidence of inflammation. The use of the double dystrophin expression cassette in some muscles dramatically increased the amount of dystrophin produced by the HDAd.

Abstract No. 789

Early region 4 of adenovirus is sufficient to increase dystrophin expression from a fully deleted (gutted) adenovirus in cell culture but not in dystrophic muscle in vivo

Montreal Neurological Institute, McGill University, Montreal, QC, Canada
Department of Genetics, Duke University Medical Center, Durham, NC, USA
Respiratory Division, McGill University Health Center and Meakins-Christie Laboratories, McGill University, Montreal, QC, Canada
Biotechnology Research Institute, the National Reasearch Council of Canada, Montreal, QC, Canada

Abstract
Helper-dependent adenovirus (HDAd) vectors lack all or most of adenoviral (Ad) genes. They hold great promise for gene therapy of diseases such as Duchenne muscular dystrophy (DMD), because they are less immunogenic than E1/E3-deleted Ad (first generation Ad or FGAd) and can carry the full-length dystrophin cDNA (12 kb). We have previously observed that the transgene expression was less efficient in cell culture and in the muscle in vivo after gene transfer with a HDAd encoding the full-length dystrophin cDNA (HDAdCMVDysFl) compared to a FGAd encoding the mini-dystrophin cDNA (FGAdCMV-dys). The two dystrophin transgenes were regulated by the same cytomegalovirus (CMV) promoter. We have also demonstrated that Ad gene products provided in trans by a FGAd could increase the amount of dystrophin produced by HDAdCMVDysFl in cell culture as well as in muscle. In the present study, we have tried to characterize the Ad gene products responsible for this enhancement. Cultures of HeLa cells and C2C12 myotubes were infected with HDAdCMVDysFl alone, or mixed with an E1/E4-deleted Ad (Ad5/E1(-gal)E4), or with a HDAd encoding the complete E4 region (HDAdMCKDysE4), or with an E1/E3-deleted Ad lacking the expression of the terminal protein and polymerase (AdLacZpp). The tibialis anterior muscle of neonatal mdx mice (the mouse model for DMD) was also injected with the same vector preparations. Dystrophin expression was monitored by western blot analysis at 4 days post-infection (cell culture) and by immunohistochemistry at 10 days post-injection (mdx muscle). No dystrophin enhancement occurred in the presence of (Ad5/E1(-gal)E4) indicating that E4 was required to increase dystrophin expression from HDAdCMVDysFl in vitro and in vivo. Enhanced dystrophin expression was observed in cell culture but not in mdx muscle in the presence of HDAdMCKDysE4 and AdLacZpp. Thus, E4 is sufficient to increase dystrophin expression of HDAdCMVDysFl in vitro but not in the muscle in vivo where additional gene products encoded by E2 are needed. The inclusion of E4 and E2 in the backbone of a HDAd may be required to produce sufficient dystrophin to mitigate the pathology of DMD after gene transfer with a HDAd.

Abstract No. 790

Facilitation of Myoblast Transplantation by Gene Transfer

Departments of Molecular Genetics & Biochemistry, and Human Genetics, University of Pittsburgh, Pittsburgh, PA
Department of Orthopedic Surgery, University of Pittsburgh, Pittsburgh, PA
Department of Molecular Genetics & Biochemistry, University of Pittsburgh, Pittsburgh, PA

Abstract
Duchenne Muscular Dystrophy (DMD) is a genetic muscular disease caused by loss of dystrophin protein, an important component of the cytoskeleton of myofibers. Dystrophin deficiency results in loss of the connection between the sarcolemmal cytoskeleton and the extracellular matrix, leading to progressive muscle weakness and death in young ages in DMD patients. Transplantation of myoblasts has been shown to be effective in restoring normal dystrophin expression in dystrophic muscle. Although primary muscle-derived stem cells isolated from hindlimb muscle of mdx mice have displayed a better survival rate than primary myoblasts after transplantation, all cell populations that we have tested showed significant loss 5 days post-transplantation. However, when we genetically engineered primary muscle-derived cells to express IL-Ra, an anti-inflammatory protein, the early loss of the injected cells was significantly reduced 48 hr after injection. To improve the in vivo survival of transplanted myoblasts, we have compared genes encoding immunosuppressive and anti-inflammatory proteins. For this study, primary muscle-derived stem cells (PP6) isolated from hindlimb muscle of mdx mice were used. Retroviral vectors have been constructed with immunosuppressive genes including TGF-ß, vIL-10, IL-Ira, and soluble IL-1 receptor Type II-Ig (sIL-IRII-Ig). Stable selected populations of primary myoblasts expressing the different immunosuppressive gene products were generated by retroviral infection. The cells were transduced with adenovirus carrying the LacZ reporter gene 24 hr. before the transplantation into the gastrocnemius muscle of mdx mice. The animals were sacrificed on day 5 and 21 post-transplantation and the hindlimb muscles removed, sectioned and stained for LacZ expression. The LacZ reporter-gene expression was detected in all groups including the control group on day 5. However, a significant loss of LacZ expression was observed on day 21. We could not detect any LacZ-positive myofibers in the non-modified control group. In contrast, we have observed varying levels of LacZ expression in the other groups on day 21. In particular, 36 % and 35% of the myoblasts producing sIL-IRII-Ig and vIL-10 respectively survived and continued to express LacZ for 21 days compared to the number of LacZ-positive myofibers on day 5 of the same group. Ratios of 12% and 1.5% were observed for the genetically engineered myoblasts producing IL-IRa and TGF-ß respectively. Comparing the various immunosuppressive genes, sIL-IRII-Ig and vIL-10 are the most promising to facilitate prolonged survival of myoblasts following intra-muscular transplantation. This approach may be help in developing therapeutic strategies in DMD.

Abstract No. 793

Characterization of Myoblasts Derived from Fetal and Post-natal Muscle: Implications for Targeting Adenovirus to Skeletal Muscle

University of Pittsburgh, Dpt. Neurology, 546 S-BST Pittsburgh, PA15213

Abstract
Adenoviruses have a number of features that suit them particularly well as gene therapy vectors for skeletal muscle. In spite of these advantages, adenoviral vector-mediated gene transfer to muscle still faces serious preclinical, technical hurdles. Muscle tissue is widespread, comprising roughly 40% of the body's mass. Therefore, if a therapy is to be effective against an inherited muscle disease such as Duchenne Muscular Dystrophy that affects widely distributed muscles, delivery of the therapy must take advantage of the vascular system. Since the native attachment receptor for adenovirus, Coxsackievirus-Adenovirus receptor, is expressed at a low level in mature muscle and at a high level in other tissues such as liver, systemic vascular delivery of adenovirus to muscle will likely require targeting to a muscle-specific molecule. We therefore seek to identify and characterize muscle-specific ligands that could be used for targeting adenoviral vectors to fetal and adult muscle.
To generate myogenic reagents that will be useful for the development of muscle targeting, we are characterizing myoblasts derived from both fetal and adult murine muscle. We have successfully isolated myoblasts from fetal (embryonic day 16) and post-natal muscle, demonstrating proliferation as mononuclear myoblasts, fusion in vitro to form multinuclear myotubes, and expression of myogenic markers. To characterize adenoviral vector-mediated gene delivery, we infected both fetal and post-natal myoblasts with a first generation adenovirus (E1, E3 deleted) carrying the lacZ reporter gene driven by the CMV promoter over a range of multiplicities of infection. We have found that reporter gene expression is approximately 20-fold higher in fetal myoblasts as compared with post-natal myoblasts.
Successful targeting of adenoviral vectors to either fetal or post-natal muscle will require both an understanding of mechanisms of un-targeted transduction and the identification of candidate muscle-specific molecules. Our studies will compare fetal and post-natal myogenic cells for the expression of cell-surface molecules, including CAR, integrins and muscle-specific molecules, and markers of myogenic differentiation. These myogenic cells can also be used for the investigation of adenoviral vector targeting to known muscle-specific cell surface molecules and for the identification of novel targeting ligands.

Abstract No. 794

The differentiation of embryonic stem cells into muscle cells for Duchenne muscular dystrophy therapy

Department of Child Development, Kumamoto University School of Medicine.

Abstract
Intro: 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 by systemic, intravenous injections have succeeded for the delivery of dystrophin (1, 2). Furthermore, embryonic stem (ES) cells derived from the inner cell mass of blastocyst-stage embryos are totipotent cells and have 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. However, we were capable of inducing differentiation of these ES cells to muscle cells only by transducing them with an Adenovirus encoding for MyoD, a transcription factor used for muscle cell differentiation (3).
Methods: Following the removal of feeder cells (primary embryo fibroblasts), the ES cells were infected in vitro with Adenovirus (AdMyoD) at a multiplicity of infection (MOI) of 50 and cultured in ES medium containing Leukemia inhibitory factor (LIF) and 15% FCS. One day following the AdMyoD infection, LIF was removed from the culture medium and the concentration of FSC was changed to 2% in order to induce differentiation of cells into myotubes. At 7 days post-infection, the differentiated ES cells were assessed for desmin expression by immunofluorescence using a monoclonal mouse anti-desmin antibody (1:200) followed by a FITC-goat anti-mouse IgG (1:50).
Results: Some of the ES cells began to fuse and form myotubes at 3 days post-infection. Many myotube-like cells were detected at 7 days post-infection. The immunofluorescent staining showed that almost all fused (myotubes) and non-fused cells were desmin positive.
Discussion: To achieve success for DMD therapy, differentiation of ES cells must be limited to the muscle lineage. Our results indicated that ES cells are easily transformed to muscle cells upon stimulation with MyoD. This data suggests that further development in ES cell technology could be useful for clinic treatments of DMD.

1.Gussoni, E., Y. Soneoka, C. D. Strickland et al. Dystrophin expression in the mdx mouse restored by stem cell transplantation. Nature. 401 390-394 1999
2.Ferrari, G., G.C.D. Angelis, M. Coletta, et al. Muscle Regeneration by Bone Marrow-Derived Myogenic Progenitors Science 279 1528-1530 1998
3.D. F. Pinney, Pearson-W SH, S. F.Konieczny, et al. Myogenic lineage determination and differentiation: evidence for a regulatory gene pathway. Cell.53 781- 93 1988.

Abstract No. 795

Differential phenotypes of human primary myoblasts subjected to lentivector-mediated reversible immortalization

Faculty of Medicine, University of Geneva, Switzerland

Abstract
Lentiviral vectors can transduce cells irrespective of their proliferating status. We have taken advantage of this property to immortalize conditionally human muscle satellite cells through the use of LoxP-lentivectors expressing growth promoting and antisenescence genes. Primary myoblasts from control individuals and from patients suffering from Duchenne or Steinert muscular dystrophies could be immortalized with gene cocktails that included the SV40 large T antigen (TAg) or the INK4a locus antagonist bmi-1, with or without telomerase (hTERT). The resulting cell lines were not tumorigenic in nude mice, but grew continuously in culture. TAg-expressing clones showed a great degree of genetic instability and exhibited aberrant morphological features. In contrast, myoblasts immortalized with bmi-1 retained a normal karyotype, grew in "proliferation" but not in "differentiation" medium, and conserved more faithfully the phenotypic characteristics of primary myoblasts. However, only rare clones demonstrated a clear fusogenic potential, even once arrested by Cre-mediated excision of the transgenes. These results further validate the use of lentiviral vectors for the conditional immortalization of human primary cells, and also demonstrate that the composition of the immortalization cocktail greatly influences the potential utility of the resulting cells.

Abstract No. 797

Interleukin-6 up-regulates utrophin expression in mdx skeletal muscle

Department of Molecular Therapy, National Institute of Neuroscience, National Center of Neurology and Psychiatry

Abstract
Duchenne muscular dystrophy (DMD) is an X-linked, lethal disorder caused by a defect in the DMD gene. Utrophin is an autosomal homologue of the DMD gene product, dystrophin, and over-expression of utrophin is expected to compensate the defect of dystrophin.
We have previously reported that utrophin was up-regulated at the sarcolemma of dystrophin-deficient mdx mice, when a -galactosidase-expressing adenovirus vector, AxCALacZ was injected into the skeletal muscle. Moreover, up-regulated utrophin mitigated dystrophic phenotypes. Since this over-expression was inhibited by immunosuppression with FK-506 or anti-CD4 antibody treatment, we concluded that the immune response against AxCALacZ resulted in the up-regulation of utrophin (Yamamoto et al., Hum Gene Ther 11, 669-680, 2000).
To apply over-expression of utrophin to treatment of DMD, we need to know the molecular mechanism of the up-regulation. We hypothesize that some cytokines would be produced by infiltrating cells and up-regulate the expression of utrophin in muscle fibers. To clarify which cytokine is involved in the up-regulation of utrophin, we daily injected candidates of cytokines into tibialis anterior muscles of 2-week-old mdx or normal C57Bl/10 mice for five days. Two, five and eight days after the final injection, we evaluated utrophin expression by immunohistochemistry.
The experiment with anti-CD4⁺ treatment suggests that CD4⁺ T cells were mainly involved in the up-regulation of utrophin. Hence, we firstly focused on cytokines involved in differentiation or maturation of CD4⁺ T cells. Interferon- (IFN-) and Interleukin-12 (IL-12) play an important role in differentiation of CD4⁺ cells into Th1 subset. We injected IFN- at a dose of 400 U/day or 1000 U/day, and IL-12 at a dose of 80 ng/day or 800 ng/day, however these cytokines showed no effects on utrophin expression. IL-4 is known to be involved in differentiation of CD4⁺ cells into Th2 subset. Therefore, we next tried IL-4 at a dose of 80 ng/day or 800 ng/day, but it also failed to up-regulate utrophin.
IL-6 is expressed in tissues infected with adenovirus vector. IL-6 is also detected in the muscle of patients with inflammatory myopathies or in regenerating muscle, where utrophin is up-regulated. Therefore we injected IL-6 at a dose of 800 ng/day and found that utrophin was widely expressed at the sarcolemma of mdx muscles. Importantly, the expression of utrophin was observed in mature myofibers as well as regenerating myofibers. In mdx muscles injected with IL-6 at a dose of 80 ng/day utrophin expression was weaker and partial. We also injected IL-6 into the muscles of normal C57Bl/10 mice at a dose of 800 ng/day, but utrophin was not expressed at the sarcolemma.
These results strongly suggest that IL-6 is involved in over-expression of utrophin in the AxCALacZ injected mdx muscle. In contrast, the up-regulation of utrophin did not occur in muscles of normal mice. This implies that other prerequisite factors which particularly express in mdx muscle are also involved in the utrophin up-regulation.

http://www.academicpress.com/www/journal/asgt2001/index.html