RESEARCH THAT WILL BE PRESENTED IN ANNUAL MEETING OF AMERICAN SOCIETY OF GENE THERAPY 2012, MAY 2012

 

Demonstration of Systemic Exon 45-55 Multiple Skipping in Dystrophic mdx52 Mice

Yoshitsugu Aoki, Tetsuya Nagata, Akinori Nakamura, Takashi Saito, Jun Tanihata, Stephanie Duguez, Kanneboyina Nagaraju, Eric Hoffman, Terence Partridge, Toshifumi Yokota, Shin'ichi Takeda. Department of Molecular Therapy, National Institute of Neuroscience, National Center of Neurology and Psychiatry (NCNP), Kodaira, Tokyo, Japan; Department of Medicine (Neurology and Rheumatology), Shinshu University School of Medicine, Matsumoto, Nagano, Japan; Research Center for Genetic Medicine, Children's National Medical Center, Washington, DC; Department of Medical Genetics, School of Human Development, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Canada

Background: Duchenne muscular dystrophy (DMD), the most common form of muscular dystrophy, is caused by a lack of dystrophin. A most promising therapeutic approach is antisense oligo-mediated elimination of frame disrupting mutations by exon skipping. Two major hurdles are, limited applicability, and uncertain function of the resulting quasi-dystrophin. Skipping of exon 45-55 at the mutation hot-spot would address both issues. Here, we tested the feasibility of antisense oligo-mediated exon 45-55 multiple skipping in mdx52 mice which harbor a deletion mutation in exon 52 of the Dmd gene. Methods: We newly designed 87 antisense oligo sequences against the ten exons between exons 45 and 55 except exon 52. We screened a series of phosphorodiamidate morpholino oligomers with cell-penetrating moiety (vPMOs) singly and in ten-oligo cocktail using H-2Kb-tsA58 (H-2K)-mdx52 myogenic cells. Then we tested the molecular efficacy by intramuscular injection and intravenously injected an optimized vPMOs cocktail (0.6 mg/kg each of the ten-oligo, biweekly for 18 weeks) into 5-week-old mdx52 mice. The efficiency and efficacy of exon 45-55 multiple skipping were tested at the mRNA, protein, histological and functional levels. Results: We demonstrated skipping of all 10 exons in H-2K-mdx52 myotubes. Better still, in dystrophic mdx52 mice, systemic injections of such ten-oligo cocktail induced extensive dystrophin expression at the subsarcolemma in skeletal muscles throughout the body with 15% dystrophin protein levels on average. The expression of β-dystroglycan and α1-syntrophin were fully restored, but neuronal nitric oxide synthase (nNOS) was not detected by western blotting. Dystrophic pathology and skeletal muscle function were improved without any detectable toxicity after the systemic vPMO cocktail treatment. Conclusion: We demonstrated an effective rescue by systemic exon 45-55 multiple skipping in a DMD animal model. Interpretation: This represents a major advance on the simple use of exon skipping to change the phenotype from DMD to its milder form Becker muscular dystrophy (BMD), because, on the basis of clinical reports, it could rescue more than 63% of patients with deletion mutations, while spontaneous deletions of this region are associated with asymptomatic or exceptionally mild phenotypes with almost normal life expectancy.

RNAi Therapy for Dominant Limb Girdle Muscular Dystrophy Type 1A

Jian Liu, Lindsay Wallace, Sara E. Garwick-Coppens, Darcy Nelson, Carol Davis, Susan V. Brooks, Michael E. Hauser, Jerry R. Mendell, Scott Q. Harper. The Research Institute at Nationwide Children's Hospital, Columbus, OH; Duke University Medical Center, Durham, NC; University of Michigan School of Medicine, Ann Arbor, MI

Limb Girdle Muscular Dystrophy (LGMD) refers to a group of 22 disorders characterized by progressive wasting and weakness of pelvic and shoulder girdle muscles. Commonly, patients require wheelchair assistance, and individuals with some forms of LGMD may have cardiac and respiratory muscle involvement. Modest improvements in a limited set of muscles may dramatically improve patients' quality of life, but there is currently no effective treatment. Over the last two decades, the feasibility of using gene therapy to treat muscular dystrophy has improved, but previous strategies have almost exclusively focused on replacing defective or missing genes underlying recessive disorders. These strategies are not feasible for treating dominant muscular dystrophies, including the 7 dominant forms of LGMD (LGMD1). Instead, patients with dominant LGMD1 would benefit from reduction of their pathogenic alleles. In this study, we developed the first RNAi-based pre-clinical treatment for LGMD1A, which is caused by dominant mutations in one allele of the myotilin (MYOT) gene. Our strategy involved delivering MYOT targeted artificial microRNAs (miMYOT) to muscles of the T57I transgenic mouse model of LGMD1A using AAV6 vectors. We found that miMYOT vectors, but not controls, significantly reduced soluble mutant MYOT protein and MYOT aggregates that characterize LGMD1A were either absent or very small in treated muscles. This reduction was accompanied by significantly recovery in function, as indicated by improvements in muscle mass, treadmill running performance, and whole muscle force in treated T57I mice. We are currently quantifying these obvious histopathological improvements, and are determining whether our miMYOT vectors will improve overall muscle weakness in T57I mice. These studies represent important first steps toward translating targeted RNAi gene therapy approaches for LGMD1A.

 

Genetic Correction of Dystrophin by Engineered Nucleases

David G. Ousterout, Pablo Perez-Pinera, Matthew T. Brown, Charles A. Gersbach. Biomedical Engineering, Duke University, Durham, NC

Duchenne Muscular Dystrophy (DMD) is the most common hereditary monogenic disease, occurring in about 1 in 3500 male births. DMD is caused by a genetic defect in dystrophin, an essential musculoskeletal protein. The absence of dystrophin leads to muscle weakness and wasting, resulting in fatal respiratory and cardiac disease. At present, there are no treatments that can effectively address the poor life expectancy and quality of life of these patients. Recent studies have demonstrated the potential of a new class therapeutics based on targeted gene editing by designer nucleases. These nucleases take advantage of natural DNA repair mechanisms to create desired changes to a genetic sequence. Importantly, zinc-finger nucleases (ZFNs) against the HIV-1 co-receptor CCR5 are in an ongoing Phase 1/2 and two Phase 1 clinical trials, demonstrating the feasibility of this type of therapeutic approach. Other studies are utilizing ZFNs to correct genetic mutations associated with sickle cell anemia, X-linked severe combined immunodeficiency, hemophilia B, and alpha-1 anti-trypsin deficiency. This project utilizes synthetic nucleases to edit and correct the dystrophin gene as a novel potential therapy for DMD. The advantage of this method is that the native dystrophin gene is restored, presumably including all of the major isoforms and functions of dystrophin. We designed ZFNs and TALE nucleases (TALENs) targeted to exon 51 of the human dystrophin gene that potentially address greater than 13% of all DMD mutations. ZFNs were generated using modular assembly (MA) and Context-Dependent Assembly (CoDA). TALENs were generated using the Golden Gate Assembly method. Five MA-ZFNs, seven CoDA ZFNs, and twelve TALENs were tested for gene editing activity at the endogenous dystrophin gene in human K562 cells. Two MA-ZFNs, three CoDA ZFNs, and four TALE nucleases were highly active, modifiying 6-13% of alleles as measured by the Surveyor nuclease assay. One ZFN of interest binds directly in exon 51 and mediated homology-directed repair in 6% of alleles using plasmid- and single-strand oligonucleotide-based donor templates. One TALEN mediated high-efficiency gene editing in 13% of alleles. Assuming that one third of these NHEJ events results in reading frame restoration, this approach has a potential gene correction rate of up to 4.3%. These nucleases were also active in immortalized myoblasts from DMD patients with an NHEJ rate of 1-12% as measured by Surveyor. We also detected successful homology-directed repair in these DMD cells after co-delivering the designer nucleases with a donor template designed to correct the dystrophin gene. We predict that this will lead to restored dystrophin function and expression in the corrected cells. Ongoing studies are assessing dystrophin gene correction and protein restoration in DMD patient-derived myoblasts in vitro and by in vivo muscle transplantation into immunodeficient mice. The development of these nucleases therefore presents a transformative approach to treating DMD.
 

Muscle Force Improvement by Long-Term Systemic Expression of an AAV9 Minidystrophin after Delivery in Young Adult GRMD Dogs without Immune Suppression

Juan Li, Chunping Qiao, Dan Bogan, Ruhan Tang, Janet Bogan, Tao Bian, Jianbin Li, Jennifer Dow, Zhenhua Yuan, Joshua C. Grieger, Richard Jude Samulski, Joe Kornegay, Xiao Xiao. Division of Molecular Pharmaceutics, Eshelman School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, NC; Gene Therapy Center, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC

Recently we reported efficient and long-term transgene expression without immune suppression in a young adult golden retriever muscular dystrophy (GRMD) dog, by a single intravenous (i.v.) injection of an AAV9 human minidystrophin vector containing a muscle-specific (MS) promoter (ASGCT 2011 abstract). However, we did not examine muscle functions in that dog. To repeat the gene expression results and also to examine immunology and muscle force improvement, we carried out an experiment using 8 young GRMD dogs from the same litter randomly divided into 2 groups, 2 males and 2 females per group for the gene therapy study. In this study,we first pre-screened anti-AAV9 antibodies by an in vitro infection assay and found no detectable neutralizing antibodies in all the dogs. At 12 weeks of age, four dogs (weight 5.8kg, 7.1kg, 7.5kg, 7.9 kg respectively) were injected with the AAV9-MS-human Mini-dystrophin vector by a leg vein at the dose of 3.0x1013/kg. The other 4 dogs were used as untreated controls. Initial muscle biopsies at 4 weeks post injection showed detectable but low levels of minidys expression by immunofluorescent (IF) staining and western blot. The second muscle biopsies at four month post injection, however, showed significantly increased gene expression levels. The results of quantitative PCR of vector DNA were also consistent with the IF staining and western blot data. ELISPOT assays against minidys peptide library at 7 weeks post AAV injection revealed no differences in all 8 GRMD dogs and a normal control dog. None of them had readings above background level (50 spots/106 PBM cells). We next investigated if muscle forces could be improved after AAV9-MS-human Mini-dystrophin vector gene therapy. Several phenotypic markers were evaluated in dogs at both 2 (pretreatment) and 6 months. The principal outcome parameter was tetanic isometric tibiotarsal joint extension force, which showed consistent decline in GRMD dogs in our natural history studies. Values for the gene therapy treated dogs increased from 1.75 N/kg ± 0.345 at 2 months to 2.05 N/kg ± 0.356 at 6 months, while those of untreated dogs decreased from 1.32 N/kg ± 0.11 to 1.09 ± 0.270 (p < 0.01). The muscle force data are in general agreement with long-term expression of minidystrophin gene. In addition, other therapeutic parameters such as muscle MRI and joint angle measurement are being investigated. In conclusion, the above results further demonstrated that long-term minidystrophin gene expression could be achieved by a single i.v. injection without immune suppression. More importantly, muscle functions could be significantly improved as a result of AAV9-MS-minidystrophin gene therapy.
 

RNA Interference Inhibits DUX4-Induced Muscle Toxicity In Vivo: Implications for a Targeted FSHD Therapy

Lindsay M. Wallace, Sara E. Garwick-Coppens, Jian Liu, Jacqueline Domire, Scott Q. Harper. Molecular, Cellular, and Developmental Biology Graduate Program, The Ohio State University, Columbus, OH; Center for Gene Therapy, The Research Institute at Nationwide Children's Hospital, Columbus, OH; Integrated Biomedical Sciences Graduate Program, The Ohio State University, Columbus, OH; Department of Pediatrics, The Ohio State University, Columbus, OH

Facioscapulohumeral muscular dystrophy (FSHD), one of the most common inherited muscle disorders, is characterized by asymmetric wasting and weakness in muscles of the face, scapula, and limbs. The clinical features of FSHD were described over a century ago, and a chromosomal deletion that reduces the copy number of D4Z4 repeats on chromosome 4q35 was linked to the disease in 1992. Unfortunately, because these D4Z4 contractions do not completely remove any obvious genes, the pathogenic mechanisms underlying FSHD remained unresolved. As a result, FSHD research for the last two decades has been primarily focused on understanding pathogenesis, while translational research for the disease has been largely lacking. Today no targeted treatment for FSHD exists, but recent findings suggest the barrier to translation may be lowering. Several studies, including our work, support an FSHD pathogenesis model involving over-expression of the pro–myopathic gene, DUX4. Thus, DUX4 inhibition may be a promising therapeutic strategy for FSHD. In this study, we tested a pre-clinical RNAi-based DUX4 gene silencing approach as a prospective treatment for FSHD. We used adeno-associated viral (AAV) vector-delivered therapeutic microRNAs to correct DUX4-associated myopathy in vivo. We found that our DUX4-targeted microRNAs (miDUX4) significantly reduced DUX4 RNA and protein and protected mouse muscles from DUX4-induced damage. Specifically, miDUX4-treated muscles showed minimal to no myofiber degeneration, regeneration or activation of apoptotic pathways, while control-treated muscles had widespread damage associated with elevated DUX4 levels. These histological improvements were also associated with a functional improvement measured by grip strength. Our results provide proof–of–principle for RNAi therapy of FSHD through DUX4 inhibition, and importantly, this strategy can be modified to target other genes that may be mis-expressed in FSHD.
 

A Novel Rationally Designed AAV Micro-Utrophin Vector Recruits nNOS to the Sarcolemma

Yi Lai, Junling Zhao, Yongping Yue, Dongsheng Duan. Molecular Microbiology and Immunology, University of Missouri, Columbia, MO

Sarcolemmal neuronal nitric oxide synthase (nNOS) is crucial for normal muscle function. Absence of sarcolemmal nNOS plays a critical role in the pathogenesis of muscular dystrophy. Duchenne muscular dystrophy (DMD) is caused by mutations in the dystrophin gene. Dystrophin anchors nNOS to the sarcolemma. Utrophin is a homolog of dystrophin and has been extensively investigated as a candidate gene for DMD gene therapy. Unfortunately, unlike dystrophin, utrophin cannot recruit nNOS to the sarcolemma. Inability of utrophin to restore sarcolemmal nNOS may significantly attenuate its therapeutic effect. To overcome this drawback of utrophin, we set to engineer novel utrophin genes that can fix nNOS to the muscle cell membrane. We recently found that dystrophin spectrin-like repeats 16 and 17 (R16/17) mediate sarcolemmal nNOS localization. Utrophin R15/16 shares great homology with utrophin R16/17. In this study, we examined whether replacing utrophin R15, or R16 or both with respective dystrophin R16 and R17 can result in novel utrophin genes capable of nNOS anchoring. A series of four repeats micro-utrophin AAV vectors were generated. All micro-utrophin genes contained a flag tag, utrophin N-terminus, utrophin repeats 1, and 22 and utrophin cysteine-rich domain. The two repeats between utrophin R1 and R22 were either original utrophin R15/16 or engineered with dystrophin R16/17 substitution. The micro-utrophin gene was packaged into AAV-9 and delivered AAV viruses to the tibialis anterior muscle of utrophin and dystrophin double knockout mice. One month after injection, we examined microgene expression and sarcolemmal nNOS localization. Robust expression was observed in all AAV infected muscle. As expected, the microgene construct carrying original utrophin R15/16 did not recover sarcolemmal nNOS. Constructs with either dystrophin R16 or R17 failed to restore nNOS to the sarcolemma. However when both repeats were replaced, sarcolemmal nNOS expression was restored. Our results suggest that both dystrophin R16 and R17 are required for nNOS localization. Further, dystrophin R16/17 is sufficient for sarcolemmal nNOS expression in the context of utrophin. The novel micro-utrophin gene described here may hold great promise for DMD gene therapy.
 

AAV-Mediated Gene Therapy in Limb-Girdle Muscular Dystrophy 2I (LGMD2I) Mouse Model

Chi-Hsien Wang, Bin Xiao, Chungping Qiao, Yiumo M. Chan, Ru-Hang Tang, Juan Li, Elizabeth Keramaris-Vrantsis, Peijuan Lu, Qilong Lu, Xiao Xiao. Department of Molecular Pharmaceutics, Eshelman School of Pharmacy, University of North Carolina-Chapel Hill, Chapel Hill, NC; McColl-Lockwood Laboratory for Muscular Dystrophy Research, Neuromuscular/ALS Center, Carolinas Medical Center, Charlotte, NC

Fukutin-related protein (FKRP) is a putative glycosyltransferase. The genetic mutations of the FKRP cause a wide spectrum of disease phenotypes ranging from late onset and milder limb-girdle muscular dystrophy 2I (LGMD2I) without central nervous system defect to severe allelic diseases such as congenital muscular dystrophy 1C (CMD1C), Walker-Warburg Syndrome (WWS) and muscle-eye-brain disease (MEB), both with defects in the central nervous system. Currently, neither effective treatment is available for FKRP-related muscular dystrophies. Recently, we have created a novel and viable FKRP knockin mouse model L276Ineo+ with disease phenotypes recapitulating the mild LGMD2I. The availability of this unique animal model enabled us to test postnatal AAV-mediated gene therapy for LGMD2I in this study. The optimized FKRP cDNA was cloned into AAV vector containing ubiquitous CB promoter (CMV enhancer and beta-actin promoter), and different dosages of AAV9-CB-opti-FKRP-myc vectors were delivered into neonatal homozygous L276Ineo+ mice via intraperitoneal injection. Untreated or AAV9-CB-GFP-treated homozygotes were used as controls. The mice have been treated for over a year. Overexpression of FKRP in homozygous L276Ineo+ mice significantly improved muscle morphology, pathology, and muscle membrane leakage. Specifically, the presence of pronounced central nucleation is one of the characteristics of the muscle of homozygous L276Ineo+ mice. However, the central nucleation was barely found in the AAV FKRP vector-treated muscle. The serum creatine kinase level, an indication of muscle leakage, was also significantly decreased after the treatment (816.77±546.80 in homozygous L276Ineo+ mice vs 151±57.11 in treated mice). More importantly, the motor functions of the treated mice has been greatly improved as evidenced by treadmill running test (homozygous L276Ineo+ mice: 353.8±172.9 meters vs AAV-treated mice: 697.4±97.9 meters) and grip force measurement (homozygous L276Ineo+ mice: 0.1713±0.0429 kg vs AAV-treated mice: 0.2296±0.0127kg). In summary, our preliminary data indicated that AAV-mediated gene therapy offers great therapeutic effects in the LGMD2I mouse model. Overexpression of FKRP in treated mice ameliorated muscle pathology, improved muscle physiology, and normalized serum creatine kinase level.
 

rAAV9-Mediated Microdystrophin Gene Transfer with Immune Tolerance Induction Improves Dystrophic Phenotype of Canine X-Linked Muscular Dystrophy

Hiromi Hayashita-Kinoh, Naoko Yugeta, Hironori Okada, Yuko Nitahara-Kasahara, Tomoko Chiyo, Takashi Okada, Shin'ichi Takeda. Department of Molecular Therapy, National Institute of Neuroscience, NCNP, Kodaira, Tokyo, Japan

Background: Duchenne muscular dystrophy (DMD) is a congenital disease causing progressive deterioration of skeletal and cardiac muscles because of mutations in the dystrophin gene. We reported that local injection of rAAV2 or rAAV8 into canine skeletal muscles without immunosuppression resulted in insufficient transgene expression with immune responses. Here we transduced fetuses of the canine X-linked muscular dystrophy in Japan (CXMDJ) to investigate the strategy of inducing immune tolerance to the rAAV as well as the therapeutic effects of rAAV9-microdystrophin. Methods: For fetal transduction, pregnant CXMDJ heterozygote was anesthetized and fetuses at post-coital day 35 were injected with 1x1012 g. c. of rAAV9-CMV-microdystrophin along with 1x1011 g. c. of rAAV9-CAG-luciferase into amniotic fluid under ultrasonic guidance. To examine the immune tolerance to the rAAV, purified canine peripheral leukocytes were exposed to rAAV9-microdystrophin for 4 hours, and then IFN-γ induction was evaluated using qRT-PCR. For systemic microdystrophin expression, we additionally injected rAAV9-CMV-microdystrophin into the jugular vein of 6 weeks old CXMDJ. Transduced littermates were periodically compared each other to assess gait function. Moreover, we analyzed the respiratory and cardiac functions of the rAAV-injected dogs by using whole body plethysmography, echocardiography and electrocardiography. Skeletal muscles of the rAAV-injected animals were sampled to examine rAAV-derived microdystrophin expression at 28 months of age. Results: IFN-γ expression in the purified peripheral blood leukocytes after the rAAV exposure was not induced in one of the rAAV-injected dogs, suggesting successful induction of immune tolerance against rAAV. The animal was additionally transduced with the rAAV9-CMV-microdystrophin at 6 weeks old. rAAV-derived microdystrophin expression in the tibialis anterior muscles and cardiac muscles was confirmed by immunohistochemistry in the transduced dog. Monthly analysis of the transduced dystrophic dogs demonstrated improved gait function compared to the non-injected littermate CXMDJ. At 1 year old, respiratory function and cardiac function of the transduced affected dog was better than those of the untransduced affected dog. Macroscopic findings of heart and diaphragm of the transduced CXMDJ dog maintained nearly normal, suggesting that the combination of the immune tolerance induction and single additional intravenous injection of the rAAV-microdystrophin achieved successful long-term transgene expression with improved dystrophic phenotype. Conclusion: Our results demonstrate that induction of oral immune tolerance against rAAV with long-term transgene expression can be achieved by direct delivery of rAAV into amniotic fluid. We also showed that rAAV-mediated microdystrophin transduction of the dystrophic dog improved their dystrophic phenotype. These findings support the therapeutic benefits of microdystrophin and future feasibilities of rAAV-mediated gene therapy strategies.
 

Increased Overexpression of Human Alpha7 Integrin as a Potential Therapy for Duchenne Muscular Dystrophy

Kristin N. Heller, Chrystal L. Montgomery, Vinod Malik, Kimberly M. Shontz, Paul M. L. Janssen, K. Reed Clark, Louise R. Rodino-Klapac, Jerry R. Mendell. The Research Institute at Nationwide Childrens Hospital, Columbus, OH; Department of Pediatrics, The Ohio State University, Columbus, OH

Duchenne Muscular Dystrophy (DMD) is a severe muscle disease caused by mutations in the dystrophin gene. Dystrophin helps link integral membrane proteins to the actin cytoskeleton and stabilizes the sarcolemma during muscle activity. We have focused on a unique approach utilizing adeno-associated virus (AAV) to develop a treatment for DMD using the mdx mouse model. Our strategy involves the upregulation of α7 integrin, a laminin receptor in skeletal and cardiac muscle that links the extracellular matrix to the actin skeleton, similar to that of the dystrophin-glycoprotein complex. There is an increase in α7 in patients with DMD and the mdx mouse and α7 has been proposed to be an important modifier of dystrophic symptoms. Burkin et al. showed that transgenic expression of the rat isoform in the mdx/utr -/- promoted satellite cell proliferation and activation, maintenance of muscle integrity, promoted hypertrophy and reduced cardiomyopathy. We have generated rAAV8.MCK.human α7 integrin and delivered it to the lower limb of mdx mice through isolated limb perfusion. As mdx mice have endogenous α7 expression, we measured overexpression human α7 as a biomarker of efficacy. Outcome measures include reversal of pathology and improvement in muscle strength and resistance to eccentric contraction induced injury in the EDL muscle. We have demonstrated approximately fifty percent of fibers in the TA and EDL overexpressing human α7 integrin after delivery through the femoral artery. The increase of human α7 integrin in skeletal muscle significantly protected against loss of force compared with untreated (contralateral) muscles, although measurement of normalized specific force in rAAV8.MCK.huα7 treated and untreated contralateral mdx EDL muscles show no significant change. α7 integrin's protection against eccentric contraction gives hope that this treatment will preserve the muscle phenotype over time. In addition, as an endogenous protein it has the benefit of being a non-immunogenic. We are currently treating mdx/utrn-/- mice to determine if there is a similar improvement in muscle physiology and investigating the mechanism that is involved in the protection from eccentric contraction induced injury.
 

Systemic Delivery of nNOS-Recruiting Mini-Dystrophin AAV Vectors Ameliorated Muscular Dystrophy in a Mouse DMD Model

Yadong Zhang, Dongsheng Duan. Molecular Microbiology and Immunology, University of Missouri School of Medicine, Columbia, MO

The nNOS-recruiting mini-dystrophin gene is a promising candidate gene for Duchenne muscular dystrophy (DMD) gene therapy. We recently developed a set of dual AAV vectors (YZ27/YZ22) to express the nNOS-recruiting minigene. Local injection yielded robust transduction and sarcolemmal nNOS localization (Zhang & Duan Hum Gene Ther 23:98-103, 2012). Here, we examined the therapeutic efficacy of this dual vector set following systemic gene transfer. We packaged YZ27/YZ22 in AAV-9 and administered to 3-week-old dystrophin-deficient mdx mice via tail vein injection. Four months later we examined minigene expression, histopathology and muscle force. Immunohistochemistry and Western blot revealed broad mini-dystrophin expression in skeletal muscle. Inflammation and central nucleation were reduced in treated mice. Physiology assay showed significant improvement of specific muscle force in both anterior tibialis and extensor digitorum longus muscles. These data have further demonstrated therapeutic efficacy of the nNOS recruiting mini-dystrophin dual AAV vectors. Further evaluation of this vector set in large animal models may set the foundation for future clinical trial (Supported by NIH and MDA).
 

Adeno Associated Virus (AAV) Mediated Follistatin Gene Transfer Toxicology Studies in Preparation of Phase I/II Clinical Trial

Janaiah Kota, Christopher Shilling, Chrystal Montgomery, Sarah Lewis, Adam Bevan, Kim Shontz, Yuuki Kaminoh, Kristin Heller, Xiomara Rosales, Laurence Viollet, Kevin Flanigan, K. Reed Clark, Brian Kaspar, Zarife Sahenk, Jerry Mendell. Center for Gene Therapy, The Research Institute at Nationwide Childrens Hospital, Columbus, OH

Inhibition of myostatin signaling is a therapeutic strategy to improve muscle size and strength for muscle disease. Follistatin has been demonstrated as a potent myostatin antagonist. In our previous pre-clinical work in both rodents and non-human primates, intramuscular (IM) administration of an alternatively spliced follistatin isoform, FS344, delivered using AAV serotype1 effectively improved muscle mass and strength with no off target effects. To establish the safety of rAAV1.CMV.follistatin344 for clinical trial use, a dose-ranging toxicology study was carried out in mice including a dose 10-fold greater than that planned for a phase I/II gene therapy clinical trial. C57BL/6 mice were randomized to three experimental arms delivering rAAV1.CMV.follistatin344 bilaterally to quadriceps muscle by IM injection. Dose ranging corresponded to the highest clinical dose (2x1012 vg/kg) or 10-fold higher (2x1013 vg/kg) compared to vector diluent. Animals were necropsied at 6, 12, 24 and 36 weeks (10 animals/timepoint) post injection (PI). Outcomes included: Body weight, hematology, chemistry panels, reproductive hormones, immune studies, viral vector biodistribution and histopathology of 24 organs from each animal. No treatment related adverse clinical signs were observed throughout the study. None of the animals showed abnormalities in hematology, clinical chemistry parameters and reproductive hormones and viral DNA dissipated in non-injected tissues by 24 weeks. At 6 weeks PI, in high dose animals, scattered focal infiltrates of inflammatory cells were observed in the injected muscle consistent with T-cell immunity to AAV1 as identified by IFN-g ELISpot, but no other pathology was observed in any of the organs. Increase in quadriceps muscle mass was observed at all-time points that correlated with myofiber hypertrophy and muscle follistatin expression. Mean fiber diameter increased and remained persistent from 6 weeks (63.3 ±2.1 µm) to 36 weeks (65.9 ± 2.1 µm) PI compared to controls (46.6 ± 5.1µm). The density of PaX7 positive nuclei increased rapidly at 6 weeks, prior to the occurrence of hypertrophic-multinucleated muscle fibers at 12 weeks. No circulating antibody to follistatin was observed. In conclusion, this study shows rAAV1.CMV.follistatin344 is safe and well tolerated at the proposed clinical doses and causes satellite cell proliferation and significant muscle hypertrophy by fusion with differentiated satellite cells. Early transient inflammation in muscle is consistent with an immune response to AAV1. These results support this approach for clinical trial use

Abbreviated Dystrophins Restore the Passive Properties of the Extensor Digitorum Longus Muscle in Dystrophin-Null Mice

Chady H. Hakim, Dongsheng Duan. Molecular Microbiology and Immunology, University of Missouri, Columbia, MO

Muscle and joint stiffness is a major clinical feature in Duchenne muscular dystrophy (DMD) patients. DMD is one of the most common lethal inherited muscle wasting diseases in childhood and it is caused by the loss of the dystrophin protein. We recently showed that the extensor digitorum longus (EDL) muscle of mdx mice (a DMD mouse model) exhibits disease-associated muscle stiffness. Abbreviated forms of the dystrophin gene (micro and mini-dystrophin) are the leading candidate genes for DMD gene therapy. Unfortunately, it has never been clear whether these abbreviated genes can mitigate muscle stiffness. To address this critical question, we examined the passive properties (elastic and viscous properties) of the EDL muscle in transgenic mdx mice that express either a representative minigene (ΔH2-R15) or one of the two most promising microgenes (ΔR2-15/ΔR18-23/ΔC or ΔR4-23/ ΔC). The passive properties were measured in 6, 14, and 20-month-old transgenic mice and the results were compared to those of age and sex-matched normal and mdx mice. Surprisingly, despite significant truncation of the gene, the elastic and viscous properties were completely normalized in every strain of transgenic mice we examined. Our results demonstrate for the first time that abbreviated dystrophin genes can effectively treat muscle stiffness in a DMD model. Our findings provide important support to further develop mini-/micro-dystrophin gene therapy.
 

A Novel Cell Therapy for Muscular Dystrophy by Bone Marrow Stromal Cell: Mesenchymal Stem Cell Derived from Bone Marrow Can Affect Skeletal Muscle Regeneration

Yasushi Maeda, Asuka Koga, Masatoshi Ishizaki, Tomohiro Suga. Neurology, Kumamoto University Faculty of Life Sciences, Kumamoto, Japan; Neurology, Kumamoto Saishunso National Hospital, Koshi, Kumamoto, Japan

Bone marrow stromal cells are stem cells differentiating in various mesenchymal tissues: bone, fat, skeletal muscle and so on. These cells are called MSC and widely noticed as an ideal cell for regenerative medicine. Its immunomodulatory competence becomes evident as well. MSC having these multi biological properties is very interesting in therapeutic application for various diseases. We hypothesized that MSC may affect the regeneration of mesenchymal tissue like skeletal muscle. In order to prove this hypothesis, we repetitively transplanted MSC (dko-MSC) from a muscular disease model mouse, dystrophin / utrophin double knockout mouse (D/U dko), into D/U dko mouse's peritoneal cavity. Nine times transplantations, two millions of dko-MSC in each transplantation, were carried out by 8-week-old. The transplantation obviously improved the body build, skeletal structure and locomotor activity.

Histologically, the transplantation made myofiber hypertrophic comparing to the wild-type mouse's myofiber. Surprisingly, transplantation drastically prolonged life span.

We can prove that MSC modifes the skeletal muscle regeneration. Although the mechanism is still unknown, autologous MSC transplantation to patient suffering from muscular dystrophy will become a novel therapy. More importantly, this cell therapy may be applied not only to muscular dystrophy but also common myopathy.
 

An Approach for Systemic Delivery of Embryonic Stem Cells into Mouse Skeletal Muscle

Shigemi Kimura, Kowashi Yoshioka, Shiro Ozasa. Child Development, Kumamoto University Graduate School, Kumamoto, Japan

We established genetically engineered ES (ZHTc6-MyoD) cells that harbor a tetracycline-regulated expression vector encoding myogenic transcriptional factor MyoD, for the therapy of muscle diseases, especially Duchenne muscular dystrophy (DMD). Almost all the ZHTc6-MyoD cells were induced into muscle lineage after removal of tetracycline. The undifferentiated ZHTc6-MyoD cells are Sca-1+ and c-kit+, but CD34-, all well-known markers for mouse hematopoietic stem cells. In addition, they are able to maintain themselves in the undifferentiated state, even after one month of culture. Therefore, it is possible to obtain a large quantity of ZHTc6-MyoD cells in the undifferentiated state that maintain the potential to differentiate only into muscle lineage. Additionally, the cells were infected with lentiviral vector carrying GFP gene as maker. We injected the infected cells into vein of mdx mice, a model mouse of DMD, for systemic delivery. A few GFP positive myofibers were detected in the muscle of the mice. In addition, GFP positive cells were isolated from skeletal muscle of the injected mice, and differentiated to muscle lineage. Finally the myotubes beat. Therefore, our ES cells have considerable therapeutic potential for treating muscle diseases.
 

Strategy for rAAV-Mediated Transduction of Common Marmoset Skeletal Muscle To Generate NHP DMD Model

Hironori Okada, Hidetoshi Ishibashi, Hiromi Hayashita-Kinoh, Tomoko Chiyo, Yuko Nitahara-Kasahara, Takashi Okada, Shin'ichi Takeda. Department of Molecular Therapy, National Institute of Neuroscience, NCNP, Kodaira, Tokyo, Japan; Department of Neurophysiology, National Institute of Neuroscience, NCNP, Kodaira, Tokyo, Japan

[Background] Duchenne muscular dystrophy (DMD) caused by dystrophin gene mutation is the most frequent and severe form of muscular dystrophy. Therapeutic methods developed for DMD have been assessed using mouse and dog model (e.g. mdx, GRMD, and CXMDJ) to reveal the interspecies differences of immune response. Bisides, motor, emotional, and cognitive functions affected by lack of dystrophin in central nervous system are left as issues unable to be assessed by mouse and dog model, even if therapies were achieved to prolong patient's life. Therefore, animal models with more advanced immune system and higher brain function are required to comprehensively assess DMD therapeutics. Common marmoset has various advantages as non-human primate (NHP) model, such as small body, short gestation, multiple birth, and short sexual maturation. Although generation of GFP transgenic common marmosets with gene-transfer to pre-implantation embryo by lentiviral vector (rLV) was reported, this method needs many donor and recipient animals for pre-implantation embryo. On the other hand, adeno-associated virus vector (rAAV) is capable of safe gene transfer and long-term transgene expression with a wide range of tissue specificity along with transduction of non-dividing cells. In order to realize high-throughput generation of DMD model marmoset using rAAV, we investigated in vivo transgene expression patterns with rAAV transduction. [Methods] rAAV9 carrying CAG promoter-driven EGFP (rAAV9-CAG-EW) was injected into adult marmoset TA and ECR muscles. rAAV1-CAG-EW was i.p. injected into new born marmosets. Marmoset dystrophin gene was estimated from their genomic sequences obtained from the UCSC Genome Browser website. For induction of frame-shift mutation by exon skipping, h51AON1 anti-sense morpholino or U7 promoter-driven snRNA which carries h51AON1 sequence, optimized Sm binding sequence, and hnRNP A1 binding site (U7cjE51) was constructed. rAAV9-U7cjE51 was injected into adult marmoset TA and ECR muscles. [Results] After rAAV9-CAG-EW injection into adult marmoset muscles, stringent expression of EGFP was observed at 2 weeks. However, diverse diminution of EGFP expression, fibrosis, and lymphocyte invasion were observed by 10 weeks. Interestingly, proficient EGFP expression in TA muscle was successfully observed at 14 months after rAAV1-CAG-EW i.p. injection into new born marmoset. Both h51AON1 morpholino and h51AON1 snRNA induced exon 51 skipping in a primary myoblasts of marmoset. Five weeks after rAAV9-U7cjE51 injection into adult marmoset muscles, exon 51 skipping was observed at the site of transduction. [Conclusion] Long-term effective muscle transduction after systemic injection into new born marmoset was confirmed. Also, induction of frame-shift mutation in marmoset dystrophin was favorably induced in vivo. This strategy with rAAV transduction would be useful for generating DMD and various models in marmoset.
 

Intravenous PMO Delivery System with Bubble Liposomes and Ultrasound Exposure into Skeletal Muscles of the mdx Mice

Yoichi Negishi, Yuko Ishii, Takuo Kojima, Hitomi Shiono, Shoko Sekine, Yoko Endo-Takahashi, Ryo Suzuki, Kazuo Maruyama, Yukihiko Aramaki. Department of Drug Delivery and Molecular Biopharmaceutics, School of Pharmacy, Tokyo University of Pharmacy and Life Sciences, Hachioji, Tokyo, Japan; Department of Biopharmaceutics, School of Pharmaceutical Sciences, Teikyo University, Sagamihara, Kanagawa, Japan

Background: Duchenne Muscular Dystrophy (DMD) is genetic disorders caused by mutations in the DMD gene that lead to the absence of essential muscle protein dystrophin. Recently, it is expected that exon skipping, mediated by antisense oligonucleotides (AOs), is one of the most promising methods for restoration of dystrophin expression in DMD treatment. However, efficient delivery method of AOs is still required for the DMD treatment. Ultrasound (US) in combination with microbubbles has recently been acquired much attention in the safe method of gene or antisense oligonucleotides (AOs) delivery. However, microbubbles have problems with size, stability, and targeting function. Liposomes have been known as drug, antigen, and gene delivery carriers. To solve the above-mentioned issues of microbubbles, we have previously developed the polyethyleneglycol (PEG)-modified liposomes entrapping echo-contrast gas, “Bubble liposomes” (BLs), which can function as a novel gene delivery tool by applying them with US exposure. In this study, to achieve a regional PMO delivery into the mdx mice, we tried to deliver a PMO, which is designed to skip the mutated exon 23 from the mRNA of murine dystrophin, into skeletal muscles of the mdx mice by intraveneous injection and subsequently the treatment of BLs and US exposure. Methods: Mdx mice (5 weeks old, male) were anesthetized with pentobarbital throughout each procedure. Prior to each injection, a tourniquet was placed around the upper hind limb to restrict blood flow into and out of the limb. A solution of BLs and PMO for mouse exon 23 was injected into the great saphenous vein of mdx mice. All intravenous injections were performed using a syringe pump. Subsequently, US exposure (1 MHz, 2 W/cm2, duty cycle 50 %) was immediately applied to the hamstring muscle. Two weeks after the treatment, exon 23 skipping levels were analysed by RT-PCR. The number of dystrophin-positive fibers after the treatment was also analysed by immunohistochemical analysis. Results: Mice were euthanized 2 weeks after the single injection, the injected muscles were isolated and analyzed by RT-PCR and by immunohistochemistry. In the combination of BLs and ultrasound exposure, we found that the PMO significantly restored the dystrophin mRNA expression and dystrophin-positive fibers by skipping the mutated exon 23 in the injection site of mdx mice, compared to the PMO injection alone. The dystrophin-positive fibers can be found in the wide area of muscle of mdx mice. Conclusions: By the intravenous delivery of PMO into muscle using BLs and US, the efficient PMO delivery into the muscle of mdx mice can be achieved with mild US application. Taken together, this ultrasound-mediated BL technique using veins may be an effective method for DMD therapy.
 

Efficient Hydrodynamic Limb Perfusion of an AAV9 Mini-Dystrophin in Adult GRMD Dogs by a Clinically Used, Pressure- and Speed-Controlled Pump

Juan Li, Chunping Qiao, Janet Bogan, Jianbin Li, Ruhan Tang, Dan Bogan, Jennifer Dow, Zheng Jane Fan, William J. Powers, Richard Jude Samulski, Joe Kornegay, Xiao Xiao. Division of Molecular Pharmaceutics, Eshelman School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, NC; Gene Therapy Center, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC; Dpatartment of Neurology, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC

Hydrodynamic limb perfusion is a simple gene delivery method to the muscle tissues. Previously we have used a speed-controlled pump (Harvard pump) to deliver AAV vectors into the hind limb muscles of large animals upon occlusion of the blood vessels with a rubber tourniquet. However, it is highly desirable to use clinically proven safe settings with controllable injection pressure and speed and FDA-approved equipment. Recently, Fan et al published safety data in a clinical trial in muscular dystrophy patients with hydrodynamic saline injection in the occluded limbs, using a clinical Belmont high volume pump and a blood pressure cuff. But it was unclear if the same setting could achieve efficient gene vector delivery to muscle tissues. Here we used the clinical equipment and setting to test in a large animal model of muscular dystrophy, the GRMD dogs, for therapeutic gene delivery in limbs. A custom-designed, cone-shaped blood pressure cuff was used to fit the upper hind limb for better occlusion. The initial experiment with an AAV-GFP revealed that the vector was absorbed by the in-path metal heating coils of the Belmont pump. No gene expression was detected. To bypass the heating coils, we added a smaller clinical pump to deliver the undiluted vector in parallel with the Belmont pump that delivers the saline diluent. The vector and saline were mixed in the tubing immediate before the catheter. The two pumped were synchronized for injection speed and total volume. Two adult GRMD dogs (14 months old; weighed 15.5 kg and 15.3 kg) were recruited in the AAV9 minidys vector perfusion study. First, the lower hind limb volume was measured by displacement in a water-filled container. The cuff was placed above the knee at a marked line. A 20-gauge catheter was inserted into the great saphenous vein. The Belmont pump was set at an injection speed of 80 ml/min and pressure limit at 300 mmHg, the same as in the human studies. The total injection volume was set at 40% of the occluded lower limb. After the cuff was inflated to 310 mmHg, the two pumps, one with the vector and the other with the saline, were started simultaneously. Out of the three legs injected, two had good occlusion. Limb muscle biopsies were performed at 3 weeks post injection. Immunofluorescent staining and Western blot of the muscle revealed robust expression of the minidystrophin in all samples, both anterior and posterior compartment of the legs. Additional biopsies will be done at longer time points. The preliminary study validated the effectiveness of the clinical procedure in achieving therapeutic gene delivery in the DMD dog model. It also warrants further investigation and optimization of the hydrodynamic limb perfusion procedure for future clinical trials.