Dose-Dependent Restoration of Dystrophin Expression in Cardiac Muscle of Dystrophic Mice by Systemically Delivered Morpholino

Home , Dystrophin

Gene Therapy (2010) 17, 132–140 & 2010 Macmillan Publishers Limited All rights reserved 0969-7128/10 $32.00 www.nature.com/gt ORIGINAL ARTICLE Dose-dependent restoration of dystrophin expression in cardiac muscle of dystrophic mice by systemically delivered morpholino

B Wu, P Lu, E Benrashid, S Malik, J Ashar, TJ Doran and QL Lu McColl-Lockwood Laboratory for Muscular Dystrophy Laboratory, Neuromuscular/ALS Center, Department of Neurology, Carolinas Medical Center, Charlotte, NC, USA

We have earlier shown that antisense morpholino oligomers rescues dystrophin expression dose dependently in both are able to restore dystrophin expression by systemic skeletal and cardiac muscles. Therapeutic levels of delivery in body-wide skeletal muscles of dystrophic mdx dystrophin were achieved in cardiac muscle albeit at higher mice. However, the levels of dystrophin expression vary doses than in skeletal muscles. Up to 50 and 30% normal considerably and, more importantly, no dystrophin levels of dystrophin were induced by single systemic delivery expression has been achieved in cardiac muscle. In this of3gkg–1 of morpholino in skeletal and cardiac muscles, study, we investigate the efﬁciency of morpholino-induced respectively. High doses of morpholino treatment reduced exon skipping in cardiomyoblasts and myocytes in vitro, and the serum levels of creatine kinase without clear toxicity. in cardiac muscle in vivo by dose escalation. We showed These ﬁndings suggest that effective rescue of dystrophin in that morpholino induces targeted exon skipping equally cardiac muscles can be achieved by morpholino for the effectively in both skeletal muscle myoblasts and treatment of Duchenne muscular dystrophy. cardiomyoblasts. Effective exon skipping was achieved in Gene Therapy (2010) 17, 132–140; doi:10.1038/gt.2009.120; cardiomyocytes in culture. In the mdx mice, morpholino published online 17 September 2009

Keywords: morpholino; exon skipping; DMD; heart

Introduction appears not to be critical for its functions. For instance, deletion of exons from 17 to 49 was associated with only Duchenne muscular dystrophy (DMD) affects one in a mild clinical phenotype.6 Similarly, an artificially approximately 3500 male births and is characterized by constructed microdystrophin with deletion from exon rapid progression of muscle degeneration. The disease is 18 to 58 remains largely functional.7 caused by nonsense and frame-shift mutations in the Antisense oligonucleotide (AON)-mediated exon skip- DMD gene at Xp21.1, resulting in lack of dystrophin ping is able to restore the reading frame of dystrophin production.1 In-frame mutations in the dystrophin gene gene disrupted by DMD mutations. This can lead to the also cause a milder form of Becker muscular dystrophy production of truncated but functional dystrophin, and with expression of truncated but partially functional can reverse a DMD to a milder Becker muscular dystrophin proteins.2–4 The amount of dystrophin in dystrophy or near normal phenotypes.8–11 The thera- Becker muscular dystrophy varies, but can reach 30% or peutic potential of antisense therapy for DMD was higher of normal levels, in contrast to only a few initially showed in the dystrophic mdx mouse that revertant fibers in the majority of DMD.5 The DMD gene harbors a nonsense point mutation in the exon 23, consists of 79 exons spanning 42.3 million base pairs. leading to general absence of dystrophin in the muscles. The muscle form of dystrophin protein has 3685 amino Specifically designed 20O methyl phosphorothioate acids (427 kDa) that can be divided into four structural AONs delivered by intramuscular injections were able domains: amino terminal, rod, cysteine-rich and carboxy to skip the exon 23 effectively in tibialis anterior (TA) terminal domains. The majority (approximately 60%) of muscles and produce functional amount of proteins.8 the DMD mutations occur within the rod domain of The same 20O methyl phosphorothioate AON was shown dystrophin, which spans about half the length of the to induce dystrophin expression in body-wide muscles protein. However, the rod domain of the dystrophin gene when delivered systemically.9 More recently, regular systemic administration of the phosphorodiamidate Correspondence: Dr QL Lu, McColl-Lockwood Laboratory for morpholino oligomers targeting the same mouse dystro- Muscular Dystrophy Laboratory, Neuromuscular/ALS Center, phin exon 23 induced effective exon skipping and Department of Neurology, Carolinas Medical Center, 1000 Blythe produced functional levels of dystrophin in skeletal Boulevard, Charlotte, NC 28231, USA. 10 E-mail: [email protected] muscles. Systemic effect of morpholino oligomers for 11 Received 23 April 2009; revised 23 July 2009; accepted 5 August exon skipping has now been shown in dystrophic dog. 2009; published online 17 September 2009 The dog harbors a splice site mutation in intron 6, Dose-dependent restoration of dystrophin expression BWuet al 133 leading to exclusion of exon 7 from the mRNA transcript. toxicity. Considerable reduction in the serum levels of Treating the dogs weekly or biweekly with a cocktail of creatine kinase was detected in mice treated with three morpholino oligomers (120–200 mg kg–1) systemi- morpholinoE23 at 0.03 g kg–1 and above. Improved cally resulted in further removal of exon 6 and 8 with muscle pathology was observed in the diaphragm after extensive dystrophin expression and significant func- treatment with 1.5 g kg–1 or higher dose of morpholi- tional stabilization by several criteria.11 Furthermore, noE23. These findings suggest that effective rescue of effective restoration of reading frame and production of dystrophin in cardiac muscles can be achieved by dystrophin have now been achieved in Phase I clinical unmodified morpholino with a dose probably tolerable trials targeting human dystrophin exon 51 in muscles of for long-term treatment. DMD patients by local injections with both 20O methyl phosphorothioate AONs and morpholino oligomers (personal communication).12,13 Results DMD affects body-wide muscles including the cardiac muscle. As DMD patients live longer because of MorpholinoE23 induces skipping of exon 23 efficiently improved multidisciplinary patient care, rescuing dys- in cardiac myoblast and myocytes trophin expression in cardiac muscle becomes more To examine whether cardiac myoblasts and myocytes critical for their longevity and quality of life.14–20 More could be effectively induced to skip dystrophin exon 23, importantly, a recent study suggests that restoration of we established cardiac myoblast culture from young mdx dystrophin in skeletal muscles only may exacerbate the mice (2 weeks old) for morpholinoE23 transfection failure of heart function if dystrophin expression cannot (Figure 1a). The nature of cardiac myoblasts was be effectively restored in cardiac muscle.16 However, confirmed by immunocytochemistry for Troponin I both unmodified 20O methyl phosphorothioate AONs (Figure 1b). The cells and C2C12 myoblasts as a control and morpholino oligomers have been unable to induce were first incubated with 10 mgml–1 fluorescein isothio- effective exon skipping and dystrophin expression in cyanate-labeled morpholino and the delivery was mon- cardiac muscle. This could severely limit the therapeutic itored directly under fluorescence microscope. Efficient values of the treatment to DMD patients with the current delivery of the labeled morpholino was clearly demon- chemistries on clinic trials.12,13 The mechanism(s) for the strated, as almost all cells showed fluorescence with lower antisense effect in cardiac muscle is not under- stronger signal within the nuclear areas (Figure 2a). The stood, but one likely contributing factor is the lower cells were then incubated with morpholinoE23 specific delivery efficiency of antisense oligomers to cardio- for targeting exon 23 of the mouse dystrophin. Exon myocytes than to myofibers in skeletal muscles.10 skipping was examined 1, 2 and 4 days after treatment. In this study, we investigated the efficiency of RT-PCR showed targeted skipping of exon 23 in both cell morpholino-induced exon skipping in cardiomyoblasts types. Approximately 30 and 50% of exon 23 skipping and myocytes in vitro, and in cardiac muscles in vivo by were observed with 10 and 50 mgml–1 morpholinoE23 dose escalation. Morpholino oligomer (morpholinoE23) respectively, in both skeletal and cardiac myoblasts was able to induce targeted exon 23 skipping in both (Figure 2b). Maximum efficiency of exon skipping was skeletal and cardiac myoblasts with similar efficiency. observed 4 days after 50 mgml–1 morpholinoE23 treat- Effective exon skipping was also achieved in freshly ment in both types of myoblasts (Figure 2c). isolated cardiomyocytes. In the mdx mice, morpholi- We also examined exon 23 skipping in freshly noE23 rescued dystrophin expression dose dependently prepared cardiomyocytes. Exon 23 skipping was clearly in both skeletal and cardiac muscles. However, potential detected 2 days after treatment with 50 mgml–1 morpho- therapeutic levels of dystrophin were achieved in cardiac linoE23. As expected, the isolated cardiomyocytes only muscles at considerable higher dosage than in skeletal survived for a limited time in culture and no exon muscles. MorpholinoE23 treatment, up to 3 g kg–1 deliv- skipping was detected by day 5 after the isolation (Figure ered systemically, produced approximately 50 and 30% 2d). These results, therefore, suggest that specific exon normal levels of dystrophin in skeletal and cardiac skipping can be achieved in cardiac myocytes when muscles, respectively, without acute and short-term effective delivery of morpholino is achieved.

Figure 1 Cardiomyoblasts (a) and immunocytochemistry showing signals (red) for cardiac troponin 1 (b).

Gene Therapy Dose-dependent restoration of dystrophin expression BWuet al 134 systemic injections of 2 mg per adult mdx mouse (approximately 60–80 mg kg–1 bodyweight) morpholi- noE23, although considerable variation in dystrophin levels persists. We, therefore, predict that higher dosage might be able to achieve a more homogeneous expression of dystrophin in skeletal muscle. A series of doses, ranging from 0.015 (data not shown), 0.03, 0.06, 0.15, 0.3, 0.6, 1.5 to 3 g kg–1, were used for single intravenous injections into the mdx mice, and body-wide muscles were examined 2 weeks later. MorpholinoE23 at the doses of 0.06 g kg–1 or below did not induce detectable increase in the number of dystrophin positive fibers and no increase in the dystrophin protein was observed in nearly all muscles examined. Consistently, RT-PCR showed no detectable transcript with targeted exon 23 skipped (data no shown). However, 0.15 g kg–1 morpho- linoE23 induced dystrophin expression in 415% fibers of nearly all skeletal muscles, although most of the fibers were only weakly stained by immunohistochemistry (Figures 3 and 4a). The number of dystrophin positive fibers reached to 30 and 40% with the doses of 0.3 g and 0.6 g kg–1, respectively. The signal intensity also increased with the increase in doses, and the amount of dystrophin measured by western blots reached 10–15% of normal levels. However, dystrophin expression remained highly variable within all skeletal muscles examined, and areas of fibers completely lacking dystrophin expression persisted (Figures 3 and 4). Levels of dystrophin continued to improve as the dose of morpholinoE23 increased. Nearly all fibers became Figure 2 MorpholinoE23 induces skipping of exon 23 in cardiac dystrophin positive in all skeletal muscles examined in myoblasts and myocytes and C2C12 myoblasts. (a) C2C12 myo- the mice treated with 3 g kg–1, although variation in blasts (C2C12, left panel) and mouse cardiomyoblasts (CMB, right panel) were incubated with 10 mgml–1 fluorescein isothiocyanate- signal intensity remained. At this dose, the amount of labeled morpholino (upper panel). No fluorescence was detected in dystrophin protein reached 50% of the normal levels in the untransfected cells (lower panel). (b) MorpholinoE23 dose- all skeletal muscles including TA, intercostals, digital dependently induced exon 23 skipping in C2C12 myoblast (left muscles, back trapezius, thoracic and lumbar muscles panel) and cardiomyoblast cells (right panel). E23+, normal (Figures 3–5), and the proportion of truncated dystro- dystrophin mRNA containing exon 23; E23À, mRNA with exon 23 phin mRNA accounted for approximately 40% when skipped. (c)50mgml–1 morpholinoE23 time-dependently increased exon 23 skipping in C2C12 myoblasts (left panel) and cardio- compared with the un-truncated form. Appreciably, myoblasts (right panel). (d) Detection of exon 23 skipping in dystrophin expression became rather uniform with cultured TA muscle (1–6), heart tissues (7–11) and cadiomyocytes approximately 85% normal levels in the diaphragm. (12–14). 1, sample from control TA muscle; 12, freshly isolated cardiomyocytes without treatment; 2, 3, 7, 8 and 13, 2 days after 50 mgml–1 morpholinoE23 treatment; 4, 5, 9, 10 and 14, 5 days after Higher doses of morpholinoE23 induced dystrophin the treatment; 6 and 11, 7days after the treatment. expression in the cardiac muscle More importantly, expression of dystrophin in the To investigate if exon skipping could be more cardiac muscle was also dose dependent. Dystrophin effectively tested using ex vivo tissues from cardiac expression was not clearly detectable in the cardiac muscles, freshly isolated heart muscles and TA muscles muscle after treatment with morpholinoE23 at the doses –1 were incubated with 50 mgml morpholinoE23 and exon of 0.15 g kg–1 or below by either immunohistochemistry, skipping was examined at different time points. No exon western blots or RT-PCR (Figures 3 and 4). However, skipping was detected by day 2 and 5 after transfection. dystrophin expressing fibers, approximately 10%, were Only o5% normal levels of RNA were isolated com- clearly shown when the dose reached 0.3 g kg–1, pared with that isolated from the same amount of freshly although only a small proportion of the cardiomyocytes dissected tissues of both types by day 5 after transfection, exhibited strong signal for dystrophin. Cardiomyocytes consistent with significant degradation of the muscle expressing high and low levels of dystrophin were tissues in culture. However, exon skipping was detected distributed relatively evenly throughout the cardiac from the morpholinoE23-treated tissues at day 7 with muscle. This pattern of distribution was different from efficiency similar in both types of muscles (Figure 2d). that in skeletal muscles treated with lower doses of morpholinoE23, in which fibers expressing strong dys- Intravenous injection of morpholinoE23 induced trophin frequently formed clusters of various sizes and dose-dependent expression of dystrophin in skeletal intermingled with areas of fibers that completely lack muscles detectable dystrophin. Similar to that observed in the We have earlier shown that body-wide dystrophin skeletal muscles, dystrophin expression improved induction in skeletal muscles can be enhanced by regular further with the increase in doses. A measure of

Gene Therapy Dose-dependent restoration of dystrophin expression BWuet al 135

Figure 3 Dose-dependent restoration of dystrophin expression by intravenous injection of morpholinoE23 in skeletal and cardiac muscles. A series of doses, ranging from 0.03, 0.15, 0.3, 0.6, 1.5 to 3 g kg–1 were used for single intravenous injections into the mdx mice and body-wide muscles were examined 2 weeks later. Blue nuclear staining with DAPI. Note the relatively uniform appearance of dystrophin positive ﬁbers in the diaphragm treated with 1.5 and 3 g kg–1 of morpholino.

1.5 g kg–1 morpholinoE23 induced dystrophin expression improvement was observed in the diaphragm of the mice in 450% of the cardiomyocytes by immunohistochem- treated with morpholinoE23 at the doses of 1.5 g kg–1 and istry, although most of the cells showed only weak higher. Severe muscle degeneration with areas of heavy signals for dystrophin with non-continuous membrane infiltrates, frequently observed in the control mdx mice, staining (Figures 3 and 4). At the dose of 3 g kg–1, was considerably reduced. The total number of CD3, morpholinoE23 induced dystrophin in nearly all cardi- CD4 and CD8 positive cells in the diaphragms of the omyocytes, although variation in levels of expression mice treated with 3 g kg–1 morpholino was only 15 (±6) remained clear. RT-PCR showed that mRNA with exon compared with 154 (±36) in the control mice (per five 23 skipped was approximately 30% of normal levels. microscopic fields under Â 200 magnification). The This was supported by the levels of dystrophin protein muscle fibers were more densely packed with less restoration with western blot (Figure 4c). These results variation in size in the diaphragm of treated mice suggest that effective exon skipping and restoration of compared with the untreated control mice (Figure 6a). dystrophin can be achieved in cardiac muscle, but The other evidence suggesting improvement of muscle requires higher doses of AON than in skeletal muscles. pathology after morpholinoE23 treatment was the Correct skipping of the targeted exon 23 was confirmed progressive reduction in serum creatine kinase levels as by sequencing (data not shown). Dystrophin expression the doses of morpholinoE23 increased. Creatine kinase was also detected in smooth muscles in intestine, levels decreased after 0.15 g kg–1 morpholinoE23 treat- stomach and blood vessels in the lung (Figure 5). ment, although the difference was not significant. Substantial decrease of creatine kinase levels was observed in the mice treated with morpholinoE23 of High doses of morpholinoE23 improved muscle 0.3 g kg–1 or higher when compared with those of the pathology and serum creatine kinase level untreated mdx mice. Significant reduction was observed without toxicity in the mice receiving 1.5 and 3 g kg–1 of morpholinoE23, Immediately after injection of all doses of morpholi- although the levels were still higher than the normal noE23, the mice woke up and started to move around, ranges (Figure 6b). Both liver and kidney were normal in and the control mice was injected with saline only. All histology without obvious inflammatory infiltrate and experimental mice ate and drank normally and the the levels of serum enzymes indicating the organ’s weight of treated mice remained similar to that of control functions remained within the normal ranges (Figures mice (data not shown). There was no death in the mice 6a and b). treated with all doses of morpholinoE23. No increase in mononucleate infiltrates was observed and, similar to untreated controls, degenerating fibers in small focal Discussion areas were seen occasionally in all limb muscles of mice treated with any dose of morpholinoE23 (data not Recently, studies in our laboratory and other places have shown). As expected within the period of 2 weeks shown the great potential of using cationic peptides or treatment, percentage of centrally nucleated fibers polymers for enhanced delivery of morpholino and exon remained similar in all muscles of morpholinoE23- skipping effect.21,22 Systemic delivery of a morpholino treated and control mdx mice. However, a considerable conjugated with the peptide of (RXRRBR)2XB sequence

Gene Therapy Dose-dependent restoration of dystrophin expression BWuet al 136 in toxicity is expected when the cationic polymers are delivered systemically, although no clear toxicity is observed at the reported doses for a relatively short time period. Thus, cationic polymer-modified morpholi- nos can be used for clinical trials only when their acute and chronic toxicity, such as LD50 (dosage leading to 50% death) and long-term effect to vital organs, are determined. Therefore, alternatives for improving antisense effect with higher safety margin need to be explored. In this study, we investigate the potential of unmodified morpholino for exon skipping in cardiomyoblasts and cardiomyocytes, and dose effect of morpholino on dystrophin expression in the dystrophic model of mdx mice. Our results clearly show that morpholino induces specific exon skipping effectively in both cardiomyoblasts and cardiomyocytes in culture. Exon skipping and dystrophin expression improve in all muscles as the dose of morpholino increases in vivo. Single dose of morpholino at 3 g kg–1 produces approximately 50 or 30% normal levels of dystrophin protein in skeletal and cardiac muscle, respectively. No toxicity is detected at the highest dosage. Thus, unmodified morpholino could be used to achieve therapeutic effect in dystrophic heart as well as in skeletal muscles if long-term safety is established. Restoration of dystrophin expression in cardiac muscle becomes increasingly important for the life quality and longevity of DMD patients, as improved multidisciplinary patient care reduces their mortality from other causes. However, earlier studies have shown the difficulty to achieve effective exon skipping and dystrophin induction in the cardiac muscle with unmodified 9,10 Figure 4 Intravenous injection of morpholinoE23 dose-depen- AONs including morpholino. The mechanism for a dently induced exon 23 skipping and expression of dystrophin in lower efficiency in exon skipping in cardiac muscle skeletal and cardiac muscles. (a) The % of dystrophin positive fibers compared with skeletal muscles is not understood. One induced by morpholinoE23 with the doses ranging from 0.015 to likely explanation is the less efficiency in delivery of –1 3gkg in muscles of mdx mice. (b) Detection of exon 23 skipping unmodified oligomers systemically to cardiac muscle in muscles by RT-PCR. Lane 1 size marker. (c) Western blot showed dystrophin expression levels in tibialis anterior (TA) muscle, (20 mg than to skeletal muscles in vivo, as delivery of morpho- total protein loaded), diaphragm (20 mg total protein loaded) and lino is considered through a passive diffusion process heart (40 mg total protein loaded) of mdx mice. (d) Western blot for rather than an active transportation.23,24 Thus, dystrophic a-actin as protein loading control. Letter C and M under Panels c muscle fibers are likely to be more permissible for the and d represent the sample from TA muscle of normal C57BL6 oligomers than those fibers with little damage or mouse (10 mg total protein loaded) and untreated mdx mice (20 mg membrane permeabilization such as cardiomyocytes in total protein loaded), respectively. (f) The average % of dystrophin expression (signal intensity with normalized protein loading) in the the dystrophic mdx mice. This is consistent with the three muscles using normal muscle as 100%. results from this study that efficiency in exon skipping is still lower in the cardiac muscle than in skeletal muscles even when high doses of the morpholino were used. (R:arginine, X:6-aminohexanoic acid, and B:alanine) is Nevertheless, the results show that the charge-neutral able to restore dystrophin to almost normal levels in the morphoplino can enter the cardiac myocytes and induce cardiac and skeletal muscles in dystrophic mdx mouse. effective and specific exon skipping, suggesting that The treatment also increases muscle strength and degeneration process is not a prerequisite for effective prevents cardiac pump failure induced by dobutamine rescue of dystrophin by unmodified morpholino.21 stress in vivo. However, the applicability of peptide- Determination of effective dosage is a prerequisite for conjugated morpholino for life-long treatment of mus- clinic applications of any potential drug. A clinic trial cular dystrophy remains to be determined. Specifically, with morpholino for dystrophin exon 51 skipping is the potential of immune response to the delivery being conducted in UK for the treatment of DMD enabling peptides needs to be assessed with long-term patients (personal communication). However, effective administration in animal models and ultimately by clinic dose to be used for systemic administration has to be trials.21 More recently, we reported that a non-peptide determined for individual antisense oligomer. Our ear- cationic polymer, Octa-guanidine, also significantly lier study suggests that therapeutic effect could be improves morpholino-mediated exon skipping and achieved in some skeletal muscles with dose of morpho- dystrophin expression in both cardiac and skeletal lino at approximately 60–80 mg kg–1 by regular injections muscles, resulting in amelioration in pathology of in the mdx mice.10 However, the same dose and treatment dystrophic mdx mice.22 However, considerable increase regiment fail to induce clearly detectable dystrophin

Gene Therapy Dose-dependent restoration of dystrophin expression BWuet al 137

Figure 5 Restoration of dystrophin in skeletal and smooth muscles after 3 g kg–1 morpholinoE23 injection. Back thoracic and lumbar muscle (a), right common extensor muscle of forelimb (b). Arteries and other vessels in the lung (c) and muscles of esophagus (d). expression in the cardiac muscle.10 Similar dose effect of MD, USA) and Liberase Blendzymes (Roche Applied morpholino for dystrophin exon skipping has also been Sciences, Indianapolis, IN, USA).25–28 The nature of observed in the dystrophic dogs.11 Thus, regular treat- cardiac myoblasts was confirmed by immunocytochem- ment of DMD with 80 mg kg–1 or below could rescue the istry for Troponin I (Abacam, Cambridge, MA, USA). dystrophic skeletal muscles, but is most likely insuffi- Cardiomyoblasts, cardiomyocytes and C2C12 myoblasts cient for rescue of cardiac dysfunctions. This study were grown in DMEM supplemented with 20% FBS shows that at the dose of 0.3 g kg–1, morpholinoE23 can (Gibco, Carlsbad, CA, USA), 2 mML-glutamine (Gibco) achieve approximately 5% dystrophin induction in heart and 100 units per ml of Penicillin/Streptomycin (Gibco). muscles. Considering the likely variation in efficacy of Cells were incubated at 37 1C with 10% CO2. Myoblasts antisense oligomers selected for targeting different were plated into six-well culture plates at 2 Â 105 cells per dystrophin exons, this high dose of morpholino may well and allowed to reach 70–80% confluence. The well be required if the aim of exon skipping includes the appropriate amount of morpholinoE23 was then added rescue of cardiac muscle through a regiment of biweekly into the media (without transfection reagent). For administration. However, one would argue that effective delivery of morpholinoE23 to cardiomyocytes, the doses in human might be different from that in mice. oligomers were mixed with the cells immediately after Therefore, dose escalating clinic trials are critical. the isolation. For ex vivo culture, TA muscles and hearts In summary, our results showed that effective rescue were quickly removed from the mice after euthanasia. of dystrophin in cardiac muscle as well as skeletal The pieces of the ventricular wall or TA (B6mm2, muscles can be achieved by unmodified morpholino. No 3–3.5 mm thick) were then placed on the Millicell culture toxicity is observed at any dosage. The high safety plate inserts (0.4 mM Pore Size, Millipore, Billerica, MA, window together with the relatively long half-life USA) and cultured in the same media as described above (between 1 and 2 months) of the induced dystrophin at 37 1C for 24 h. MorhplinoE23 was then added into the provides prospect that unmodified morpholino can be media (without transfection reagent). used for long-term rescue of dystrophin for the treatment of DMD patients. Animals, oligonucleotides and delivery methods Five mdx mice aged 4–5 weeks were used in each group. Materials and methods The antisense phosphorodiamidate morpholino oligomer (morpholinoE23) (50-GGCCAAACCTCG GCTTACCTGA Isolation, charactererization and culture of cells AAT-30) against the boundary sequences of exon and and tissues intron 23 of dystrophin gene is used. MorpholinoE23 Cardiomyoblasts and cardiomyocytes were isolated from tagged with fluorescein isothiocyanate was also used. the mdx mice aged 2 weeks. A portion of the left ventricle MorpholinoE23 was prepared in 100 ml saline and of mdx mice heart was used for cell isolation by administrated by retraorbital injections. collagenase digestion with Cellutron cardiomyocyte Mice were killed at desired time points, and muscles isolation kit (Cellutron Life Technology, Baltimore, were snap frozen in liquid nitrogen-cooled isopentane

Gene Therapy Dose-dependent restoration of dystrophin expression BWuet al 138

Figure 6 Examination of pathology and serum testing after morpholinoE23 treatment. (a) Histology of TA, diaphragm and kidney (hematoxylin and eosin staining) from the normal C57BL6 mice (C57), untreated mdx mice (Untreated), 3 g kg–1 morpholinoE23-treated mdx mice (3 g kg–1 treated). (b) Serum testing. CK, creatine kinase (kU l–1), creatinine (mg l–1), urea nitrogen (mg ml–1), total bilirubin (mg l–1), alanine transaminase (ALT) (U per 100 ml), alkaline phosphatase (ALP) (U per 100 ml) and gamma-glutamyltransferase (GGT) (U l–1). Signiﬁcant reduction in creatine kinase levels was observed in the 1.5, 3 g kg–1 morpholinoE23-treated mdx mice compared with untreated mdx mice (n ¼ 5, two-tailed t-test, *Po0.05). No signiﬁcant differences were detected for other serum components between the two groups of mdx mice except for ALT, which showed elevated levels in the untreated mice. Mdx mice were intravenously treated with a series of doses of 0.03 g kg–1 up to 3 g kg–1 of morpholinoE23.

and stored at À80 1C. Experiments were approved by goat-anti-rabbit IgGs Alexa 594 (Invitrogen, Eugene, IACUC Carolinas Medical Center. OR, USA). Sections were also stained with hematoxylin and eosin for histological assessment. Fluorescein iso- Antibodies and immunohistochemistry thiocyanate-labeled monoclonal antibodies to mouse Sections of 6 mm were cut from at least two-thirds of CD3, CD4 and CD8 (EBioscience, CA, USA) were used muscle length of TA, quadriceps, biceps and gastro- in combination for detection of lymphocyte infiltration in cnemius at 100-mm intervals and at least 10 levels from the diaphragm. The number of cells was counted in five all other muscles including heart, diaphragm, inter- microscopic fields under 200 times of magnification. costals and abdominal muscles at 100-mm intervals. The intervening muscle sections were collected either for Protein extraction and western blot western blot or RT-PCR analysis. The serial sections The collected sections were ground into powder and were stained with rabbit polyclonal antibody P7 against lysed with 200 ml protein extraction buffer as described dystrophin. The primary antibody was detected by earlier.21,22 The protein concentration was quantified by

Gene Therapy Dose-dependent restoration of dystrophin expression BWuet al 139 Protein Assay Kit (BioRad, Hercules, CA, USA). Proteins 2 Arahata K, Beggs AH, Honda H, Ito S, Ishiura S, Tsukahara T et from normal C57BL6 mice (positive controls) and from al. Preservation of the C-terminus of dystrophin molecule in the treated mdx mice were loaded onto a 6% polyacrylamide skeletal muscle from Becker muscular dystrophy. J Neurol Sci gel containing 0.2% SDS and 10% glycerol. Samples were 1991; 101: 148–156. electrophoresed overnight at 10 mA at 4 1C and blotted 3 Koenig M, Beggs AH, Moyer M, Scherpf S, Heindrich K, onto nitrocellulose membrane overnight at 300 mA. The Bettecken T et al. The molecular basis for Duchenne versus membrane was then washed and blocked with 5% Becker muscular dystrophy: correlation of severity with type of skimmed milk and probed with monoclonal antibody deletion. Am J Hum Genet 1989; 45: 498–506. NCL-DYS1 against dystrophin rod domain (Vector Labs, 4 Schwartz M, Hertz JM, Sveen ML, Vissing J. LGMD2I presenting Burlingame, CA, USA) overnight. The bound primary with a characteristic Duchenne or Becker muscular dystrophy 64 antibody was detected by horseradish peroxidase-con- phenotype. Neurology 2005; : 1635–1637. 5 Angelini C, Fanin M, Pegoraro E, Freda MP, Cadaldini M, jugated goat anti-mouse IgG and ECL Western Blotting Martinello F. Clinical-molecular correlation in 104 mild X-linked Analysis System (Perkin Elmer, Waltham, MA, USA). muscular dystrophy patients: characterization of sub-clinical The intensity of the bands obtained from the oligomer- phenotypes. Neuromuscul Disord 1994; 4: 349–358. treated mdx muscles was measured and compared with 6 England SB, Nicholson LV, Johnson MA, Forrest SM, Love DR, that from normal muscles of C57BL6 mice (NIH ImageJ Zubrzycka-Gaarn EE et al. Very mild muscular dystrophy 1.42 software). associated with the deletion of 46% of dystrophin. Nature 1990; 343: 180–182. RNA extraction and RT-PCR 7 Wang B, Li J, Xiao X. Adeno-associated virus vector carrying The collected Sections were homogenized in TRIzol human minidystrophin genes effectively ameliorates muscular (Invitrogen) by using an Ultra-Turrax homogenizer dystrophy in mdx mouse model. PNAS 2000; 95: 13714–13719. (Janke and Kunkel, Staufen, Germany). Total RNA 8 Lu QL, Mann CJ, Lou F, Bou-Gharios G, Morris GE, Xue SA et al. was then extracted and 100 ng of RNA template was Functional amounts of dystrophin produced by skipping the used for a 50-ml RT-PCR with RT-PCR Master Mix (USB, mutated exon in the mdx dystrophic mouse. Nat Med 2003; 9: Cleveland, Ohio, USA). The primer sequences for the 1009–1014. RT-PCR were Ex20Fo 50-CAGAATTCTGCCAATTGCT 9 Lu QL, Rabinowitz A, Chen YC, Yokota T, Yin H, Alter J et al. GAG-30 and Ex26Ro 50-TTCTTCAGCTTGTGTCATCC-30 Systemic delivery of antisense oligoribonucleotide restores for amplification of mRNA from exons 20 to 26. A total of dystrophin expression in body-wide skeletal muscles. Proc Natl 40 cycles were carried out for the RT-PCR. Bands with Acad Sci USA 2005; 102: 198–203. the expected size for the transcript with exon 23 deleted 10 Alter J, Lou F, Rabinowitz A, Yin H, Rosenfeld J, Wilton SD et al. Systemic delivery of morpholino oligonucleotide restores dys- were extracted and sequenced. The intensity of the bands trophin expression bodywide and improves dystrophic patho- was measured with the NIH ImageJ 1.42 and percentage logy. Nat Med 2006; 12: 175–177. of exon skipping was calculated with the intensity of the 11 Yokota T, Lu QL, Partridge T, Kobayashi M, Nakamura A, two bands representing both exon 23 unskipped and Takeda S et al. Efficacy of systemic morpholino exon-skipping in skipped as 100%. Duchenne dystrophy dogs. Ann Neurol 2009; 65: 667–676. 12 van Deutekom JC, Janson AA, Ginjaar IB, Frankhuizen WS, Measurement of serum creatine kinase and other Aartsma-Rus A, Bremmer-Bout M et al. Local dystrophin components restoration with antisense oligonucleotide PRO051. N Engl J Mouse blood was taken immediately after cervical Med 2007; 357: 2677–2686. dislocation and centrifuged at 1500 r.p.m. for 3 min. 13 Arechavala-Gomeza V, Graham IR, Popplewell LJ, Adams AM, Serum was separated and stored at À80 1C. The level of Aartsma-Rus A, Kinali M et al. Comparative analysis of antisense serum components was determined by Charles Riverside oligonucleotide sequences for targeted skipping of exon 51 Laboratories, USA. during dystrophin pre-mRNA splicing in human muscle. Hum Gene Ther 2007; 18: 798–810. http://clinicaltrials.gov/ct2/show/ NCT00844597?term ¼ duchenne%2C+AVI&rank ¼ 1. Conflict of interest 14 Foster K, Foster H, Dickson JG. Gene therapy progress and prospects: Duchenne muscular dystrophy. Gene Therapy 2006; 13: The authors declare no conflict of interest. 1677–1685. 15 Wagner KR, Lechtzin N, Judge DP. Current treatment of adult Duchenne muscular dystrophy. Biochim Biophys Acta 2007; 1772: Acknowledgements 229–237. 16 Townsend D, Yasuda S, Li S, Chamberlain JS, Metzger JM. This work was supported by the Carolinas Muscular Emergent dilated cardiomyopathy caused by targeted repair of Dystrophy Research Endowment at the Carolinas dystrophic skeletal muscle. Mol Ther 2008; 16: 832–835. HealthCare Foundation and Carolinas Medical Center, 17 Matsuda R, Nishikawa A, Tanaka H. Visualization of dystrophic Charlotte, NC, and by U.S. Army Medical Research, muscle fibers in mdx mouse by vital staining with Evans blue: Department of Defense (W81XWH-05-1-0616). evidence of apoptosis in dystrophin-deficient muscle. J Biochem (Tokyo) 1995; 118: 959–964. 18 Silver MM, Banerjee D, Hudson AJ. Segmental myofiber necrosis References in myotonic dystrophy—an immunoperoxidase study of immunoglobulins in skeletal muscle. Am J Pathol 1983; 112: 1 Koenig M, Hoffman EP, Bertelson CJ, Monaco AP, Feener C, 294–301. Kunkel LM. Complete cloning of the Duchenne muscular 19 Bostick B, Yue Y, Long C, Duan D. Prevention of dystrophin- dystrophy (DMD) cDNA and preliminary genomic organization deficient cardiomyopathy in twenty-one-month-old carrier mice of the DMD gene in normal and affected individuals. Cell 1987; by mosaic dystrophin expression or complementary dystro- 50: 509–517. phin/utrophin expression. Circ Res 2008; 102: 121–130.

Gene Therapy Dose-dependent restoration of dystrophin expression BWuet al 140 20 Quinlan JG, Hahn HS, Wong BL, Lorenz JN, Wenisch AS, Levin 25 Zhou YY, Wang SQ, Zhu WZ, Chruscinski A, Kobilka BK, LS. Evolution of the mdx mouse cardiomyopathy: physiological Ziman B et al. Culture and adenoviral infection of and morphological findings. Neuromuscul Disord 2004; 14: 491–496. adult mouse cardiac myocytes: methods for cellular genetic 21 Wu B, Moulton HM, Iversen PL, Jiang J, Li J, Spurney CF. et al. physiology. Am J Physiol Heart Circ Physiol 2000; 279: Effective rescue of dystrophin improves cardiac function in H429–H436. dystrophin-deficient mice by a modified morpholino oligomer. 26 Kajstura J, Zhang X, Liu Y, Szoke E, Cheng W, Olivetti G et al. Proc Natl Acad Sci USA 2008; 105: 14814–14819. The cellular basis of pacing-induced dilated cardiomyopathy. 22 Wu B, Li Y, Morcos PA, Doran TJ, Lu P, Lu QL. Octa-guanidine Myocyte cell loss and myocyte cellular reactive hypertrophy. morpholino restores dystrophin expression in cardiac and Circulation 1995; 92: 2306–2317. skeletal muscles and ameliorates pathology in dystrophic mdx 27 Beltrami AP, Barlucchi L, Torella D, Baker M, Limana F, mice. Mol Ther 2009; 17: 864–871. Chimenti S et al. Adult cardiac stem cells are multipotent and 23 Kurreck J. Antisense technologies. Improvement through novel support myocardial regeneration. Cell 2003; 114: 763–776. chemical modifications. Eur J Biochem 2003; 270: 1628–1644. 28 Oh H, Bradfute SB, Gallardo TD, Nakamura T, Gaussin V, 24 Summerton J, Weller D. Morpholino antisense oligomers: design, Mishina Y et al. Cardiac progenitor cells from adult myocardium: preparation, and properties. Antisense Nucleic Acid Drug Dev homing, differentiation, and fusion after infarction. Proc Natl 1997; 7: 187–195. Acad Sci USA 2003; 100: 12313–12318.

Gene Therapy