| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2003;108:1626.)
© 2003 American Heart Association, Inc.
Basic Science Reports |
From the Department of Molecular Microbiology and Immunology, University of Missouri School of Medicine, Columbia, Mo (Y.Y., D.D.); the Department of Anatomy and Cell Biology, University of Iowa Roy J. and Lucille A. Carver College of Medicine, Iowa City, Iowa (Z.L., R.L.D.); and the Department of Neurology, The University of Washington School of Medicine, Seattle, Wash (S.Q.H., J.S.C.).
Correspondence to Dongsheng Duan, PhD, M610G, MSB, Department of Molecular Microbiology and Immunology, University of Missouri, School of Medicine, 1 Hospital Dr, Columbia, MO 65212. E-mail duand{at}missouri.edu
Received March 26, 2003; revision received June 4, 2003; accepted June 6, 2003.
| Abstract |
|---|
|
|
|---|
Methods and Results AAV was delivered to the newborn mdx mouse cardiac cavity. The procedure was rapid and well tolerated. Efficient expression was achieved in the inner and the outer layers of the myocardium. The ubiquitous cytomegalovirus promoter resulted in substantially higher expression than the muscle-specific CK6 promoter. The therapeutic effects of microdystrophin were evaluated at 10 months after infection. Immunostaining demonstrated extensive microdystrophin expression and successful restoration of the dystrophin-glycoprotein complex. Importantly, AAV-mediated microdystrophin expression improved the sarcolemma integrity in the mdx heart.
Conclusions We established a simple gene transfer method for efficient and persistent transduction of the mdx mouse heart. AAV-mediated microdystrophin expression restored the critical dystrophin-glycoprotein complex and improved sarcolemma integrity of the mdx heart. Our results revealed the promise of AAV-microdystrophin gene therapy for cardiomyopathy in DMD.
Key Words: muscular dystrophy genes viruses gene therapy microdystrophin
| Introduction |
|---|
|
|
|---|
90% of patients with the milder Becker muscular dystrophy (BMD) and female DMD carriers show cardiac involvement.2 For the majority of patients, heart transplantation is the only available treatment for their end-stage cardiomyopathy.3 DMD and BMD are caused by mutations in the dystrophin gene. Replacement of the defective gene with a functional homologue holds great therapeutic promise. In the past several years, DMD gene therapy has focused on correcting skeletal muscle pathological lesions in murine (mdx) and canine (cxmd) models for DMD,4 but very little effort has been made to treat the cardiomyopathy. The only reported study explored fetal cardiomyocyte engraftment in the mdx and the cxmd hearts, but expression was limited to the transplanted cells.5
The paucity of cardiac gene therapy attempts results primarily from 2 causes. First, unlike limb muscles that are easily accessible, the anatomic location and the physiological function of the heart present a formidable challenge to gene transfer. Second, the full-length dystrophin cDNA exceeds the packaging capacity of most currently available viral vectors. To circumvent the size limitation, microdystrophin genes that contained only
30% of the dystrophin cDNA coding sequence were developed (reviewed by Dickson et al6). Some microgenes were capable of providing functional improvement in the mdx mouse skeletal muscles, including the diaphragm.7,7a
Different heart gene transfer techniques have also been explored. Therapeutic genes were delivered to the heart by direct injection or coronary artery perfusion. However, most of the published studies have used larger animals, such as rat, hamster, and rabbit.8,9 The small size of the mouse has made heart gene transfer a more difficult task.1012 Svensson et al10 reported a direct myocardial injection method to deliver adeno-associated virus (AAV) to the adult mouse heart. However, expression was observed only in the injected area. Three methods have been explored to increase transduction efficiency: multiple injections, coronary artery perfusion of the explanted heart, and intracavity gene delivery.1012 Christensen et al12 achieved mouse heart gene transfer by delivering adenovirus directly to the cardiac cavity of embryonic and neonatal mice. This method resulted in up to 6 months of transgene expression without obvious disturbance of cardiomyocyte function.
In the present study, we further optimized the neonatal cardiac gene transfer technique described by Christensen et al12 for the mdx heart gene therapy. More importantly, we examined the therapeutic potential of our recently described
R4-23/
CT microdystrophin (
R4) gene7 in the mdx heart. Both transgenic and AAV-mediated gene transfer studies have suggested that the
R4 gene is the most functional microdystrophin construct for correcting skeletal muscle pathology in the mdx mouse.7 Our results indicated that AAV-mediated neonatal cardiac gene transfer was an efficient method to stably deliver the
R4 gene to the mdx heart. Furthermore,
R4 gene expression was sufficient to restore the dystrophin-glycoprotein complex (DGC) and strengthen the sarcolemma integrity in the mdx heart.
| Methods |
|---|
|
|
|---|
R4 gene contains the N-terminal actin-binding domain, the first 3 and the last spectrin-like repeats, the hinges 1, 2, and 4, and the cysteine-rich domain. The pCisCMV.
R4 was generated by swapping in a 508-bp KpnI/XbaI cytomegalovirus (CMV) promoter fragment to a previously described pCisCK6.
R4.7
Neonatal Cardiac Gene Transfer
All animal experiments were carried out in accordance with institutional guidelines. Heart gene transfer was performed according to a published protocol with modifications.12 Briefly, neonatal mdx pups (<12 hours old) were placed in a custom-designed ice chamber to achieve hypothermic shock. Recombinant AAV (10 µL) (1010 virus particles in PBS) was injected directly into the cardiac cavity by puncturing through the chest wall. Injection micropipette was prepared with capillary tubing (OD, 1 mm; ID, 0.58 mm) on a pipette puller (David Kopf Instruments). A 1-mL rubber bulb was mounted to the micropipette to control virus delivery. Cardiac cavity injection was confirmed by visualization of blood flashback in the micropipette (Figure 1A). After injection, mice were placed on a 34°C warming pad to recover.
|
Reporter Gene Evaluation
LacZ expression was determined as described previously.14 Alkaphos expression was detected on 8-µm tissue sections after heat-inactivation of the endogenous heat-labile alkaphos protein. Staining was performed at 37°C for 10 minutes in 75 mg/mL nitroblue tetrazolium chloride, 50 mg/mL 5-bromo-4-chloro-3-indolyl phosphate p-toluidine salt, 100 mmol/L Tris (pH 9.5), 50 mmol/L MgCl2, and 100 mmol/L NaCl.
Immunofluorescence Staining
The
R4 microdystrophin was revealed by 2 antidystrophin N-terminal antibodies. The polyclonal antiN-terminal antibody was used essentially as described previously.7 The human specific monoclonal antiN-terminal antibody (Dys-3, 1:10, Novocastra) was applied according to a published protocol.15 Additional monoclonal antibodies, including antidystrophin-C-terminal Dys-2 (1:30), antiß-sarcoglycan (1:50), and antiß-dystroglycan (1:50) antibodies (Novocastra) were used the same way as for Dys-3.
In Vivo Cardiomyocyte Sarcolemma Integrity Assay
Evans blue dye (EBD) uptake in the heart was evaluated after ß-isoproterenol challenge as described previously.16 Briefly, ß-isoproterenol and EBD were administered to 10-month-old AAV-infected mdx mice. The sarcolemma integrity was examined 4 hours later under the Texas Red channel by a Nikon E800 fluorescence microscope.
| Results |
|---|
|
|
|---|
A complicated micromanipulator gene delivery system was used by Christensen et al.12 To simplify the procedure, we attached a 1-mL rubber bulb to the micropipette to control virus delivery. To facilitate penetration of the chest, a very small incision was made on the surface skin above the heart. Virus was injected into the heart when the micropipette reached the cardiac cavity (Figure 1A). On average, it took 93±2.7 seconds to complete the entire surgical procedure.
A longer cold-shock may help to extend the virus-heart contact time and increase the transduction efficiency. However, a longer cold-shock may also decrease the survival rate. We compared the recovery after a 2-minute or a 5-minute hypothermic shock. There was no substantial difference in the survival rate between the 2-minute (96%, n=84) and the 5-minute (95%, n=102) groups. The overall recovery time was also very similar in the 2-minute (488±11 seconds) and the 5-minute (486±18 seconds) groups. Therefore, the 5-minute cold-shock was chosen for AAV injection.
We also examined whether virus injection had a negative effect on mouse survival. In the absence of injection, the overall survival rate was 93% (n=23). In AAV-injected mice, the overall survival rate was 96% (n=163). No significant difference in recovery time was observed among different groups (Figure 1B). This result suggested that our modified heart gene transfer procedure was fairly safe.
Effect of the Promoter and Kinetics of Gene Expression
It has been suggested that virus-mediated gene therapy in dystrophic muscle may require tissue-specific promoters.17 The ubiquitous CMV promoter and the muscle-specific CK6 promoter have been used in AAV-mediated microdystrophin gene therapy.7 To determine which promoter worked better for the heart, we compared the expression between AV.CMV.LacZ and AV.CK6.LacZ (n=10 for each group). AAV-2 viruses (1010 particles) were delivered either to the heart or to the left hind limb of the newborn mdx pups, and ß-galactosidase activity was quantified 1 month later. Consistent with our previous study in the adult mdx skeletal muscle,14 the CMV and the CK6 promoter led to similarly high levels of gene expression in the neonatal mdx limb muscle (Figure 2). However, the CMV promoter led to a substantially higher expression in the heart [(2.96±0.26)x106 relative units/mg] than the CK6 promoter [(7.23±0.98)x104 relative units/mg]. The CK6 promoter is derived from mouse muscle creatine kinase gene regulatory elements.18 Transgenic studies have shown that the muscle creatine kinase promoter drives
50-fold higher expression in skeletal muscle than in cardiac muscle.19 Our results agreed with the previous public-ation and highlighted the importance of selecting appropriate gene expression cassettes for DMD heart gene therapy.
|
Long-term gene transfer is required for DMD therapy. To evaluate whether neonatal gene transfer can mediate persistent transgene expression in the heart, we compared AAV-mediated expression at 1 and 6 months after infection. To expand our findings to other transgenes, AV.RSV.Alkaphos was used in this study. The RSV promoter was as effective as the CMV promoter in the mdx muscle.14 Because AAV-5 mediates better transduction in the mdx muscle than AAV-2,13 we decided to use AAV-5 in this and other experiments described later. As shown in Figure 3, quantitative analysis revealed nearly a 3-fold increase in alkaphos-positive heart cells at 6 months after infection. It is possible that multiple copies of virus genomes may have been delivered to some progenitor cells. Subsequent cell division may split the episomal AAV genomes into more cells. Interestingly, in addition to the efficient transduction of the inner myocardium layer, we also observed a fairly efficient transduction of the outer layer (Figure 3). This mirrored the findings of Christensen et al.12 Transduced cells were also observed through the middle myocardium layer (Figure 3).
|
Microdystrophin Expression in the Mdx Heart Restored DGC
Microdystrophin genes are highly truncated versions of the dystrophin gene. We sought to determine whether microdystrophins could be used to treat the mdx cardiomyopathy. As shown in Figure 4, efficient microdystrophin expression was detected at both the inner (Figure 4, A and D) and the outer (Figure 4, B and E) layers of the myocardium at 10 months after infection. In uninfected control mdx heart, we observed only rare revertant cardiomyocytes (Figure 4, C and F). The cardiomyocyte-specific microdystrophin expression was further demonstrated by double immunostaining with antibodies against dystrophin N-terminal and
-actinin (Figure 5). Additional Western blot also confirmed microdystrophin expression in AAV-infected mdx heart (data not shown).
|
|
The biological function of dystrophin is performed in conjunction with the DGC (reviewed by Blake et al20 and Rando21). In the absence of dystrophin, the DGC is destabilized, and this loss of DGC contributes greatly to the muscle pathology. Consequently, DGC restoration is a critical therapeutic end point (reviewed by Chamberlain4 and Straub and Campbell22). The DGC in cardiac muscle is similar to that in skeletal muscle.23 An immunoelectron microscopic study in cardiac muscle revealed a close interaction between the dystrophin cysteine-rich and C-terminal domains and ß-dystroglycan.24 These 2 domains are important for recruiting the other DGC components (reviewed in Reference 20). Because the C-terminal domain is completely deleted in our construct, we investigated whether AAV-mediated microdystrophin expression could restore the DGC in the mdx heart. As shown in Figure 6A,
R4 microdystrophin was detected only in AAV-infected mdx heart by a human-specific antidystrophin N-terminal antibody. Importantly, other components of DGC (such as ß-sarcoglycan and ß-dystroglycan) were also detected in AAV-infected but not mock-infected mdx heart. Additional immunostainings were performed on serial sections of AAV-infected mdx heart. Consistent with the data shown in Figure 6A,
R4 microdystrophin expression successfully restored ß-sarcoglycan and ß-dystroglycan in virus-transduced cardiomyocytes (Figure 6B). These data demonstrated that AAV-mediated microdystrophin expression was sufficient to restore the DGC in the mdx heart.
|
Microdystrophin Expression in the Mdx Heart Improved Sarcolemma Integrity
EBD uptake is a hallmark for dystrophin-deficient myofibers and reflects the susceptibility of these fibers to mechanical stress. ß-Isoproterenol was administered to the virus-infected mdx mouse to create a mechanical challenge to the heart.16 As shown in Figure 7, microdystrophin expression indeed improved the sarcolemma integrity and prevented EBD uptake. Quantitative examination of 4 independent AAV-infected mdx heart (
300 cells/heart were evaluated) demonstrated that all the EBD-positive cells were negative for microdystrophin and none of the AAV-infected cells were leaky to EBD.
|
| Discussion |
|---|
|
|
|---|
The posterior left ventricle wall is the most commonly affected area in DMD.26 It is currently unclear whether gene therapy of DMD cardiomyopathy requires efficient gene transfer to every cardiomyocyte in the heart or whether transduction of a subset of cells at the posterior left ventricular wall is sufficient. Consistent with the findings by Christensen et al,12 we also observed a preferential transduction at the inner and the outer layers of the myocardium (Figures 3 and 4
). The inner myocardium transduction may have derived from a passive diffusion of AAV across endocardium. It is currently unclear how intracavity delivery leads to expression at the outer layer. Occasionally, transduced cells were observed in the middle myocardium layer. Because these cells cannot be reached by simple diffusion, we suspect that the transduction may have occurred through coronary artery distribution. It is likely that further optimization of gene transfer techniques will be necessary to target a greater proportion of the heart cells to achieve therapeutic effects in patients.
In this study, we also evaluated the effect of the promoter and the persistence of gene transfer. The muscle creatine kinase promoter and its derivatives (such as CK6) have been used extensively to drive muscle-specific expression (reviewed in Reference 4). Our data suggest that the CK6 promoter is less optimal in cardiac muscle than it is in skeletal muscle (Figure 2). Novel promoters need to be developed for high-level expression in both skeletal and cardiac muscle. Kinetic study at 1 and 6 months after injection revealed a persistent transduction. Interestingly, we also observed a 3-fold increase in the absolute number of transgene-positive cells over this period (Figure 3C). We suspect that the increase in the number of AAV-transduced cells may have resulted from the limited proliferation of AAV-infected progenitor cells.
Whether the lack of dystrophin in the heart is directly responsible for the DMD cardiomyopathy has been debated. Although cardiomyopathy has been described in old mdx mice (>20 months old),27 studies in young mdx mice (<30 weeks old) revealed infrequent pathological changes in the heart.28 These observations have been partially attributed to the compensatory upregulation of the dystrophin homologue utrophin in the mdx mouse (reviewed in Reference 20). However, no apparent histological abnormalities were observed in hearts of young mice (<7 weeks old) lacking both utrophin and dystrophin.29 Even in relatively old dystrophin/utrophin double-knockout mice (7 to 11 weeks old), only 50% to 60% of the mice showed cardiac pathology.29 Megeney et al30 recently reported an extensive skeletal muscle dystrophy and severe cardiomyopathy in the MyoD-knockout mdx mouse. Because MyoD is expressed only in skeletal muscle, Megeney et al proposed that the secondary response to the severe skeletal muscle damage, rather than the loss of dystrophin itself, might be crucial for the cardiomyopathy in DMD. This hypothesis is challenged by several recent studies showing a direct link between the loss and/or remodeling of dystrophin and the nondystrophic cardiomyopathy.31,32 To understand the pathogenesis of DMD cardiomyopathy, it will be important to determine whether dystrophin expression in skeletal or heart muscle alone can rescue cardiomyopathy. A suggestive answer to this question is provided by a recent transgenic study in the
-sarcoglycandeficient mouse.33 The lack of
-sarcoglycan leads to the limb girdle muscular dystrophy 2C in humans. Zhu et al33 found that the genetic rescue of skeletal muscle pathological conditions by skeletal musclespecific expression of
-sarcoglycan had no beneficial effect on the cardiomyopathy. Taken together, these recent results indicate that the loss of dystrophin is probably the most important pathogenic factor for the cardiomyopathy in DMD. Restoration of the missing dystrophin with a functional homologue such as a microdystrophin not only in skeletal muscle but also in the heart will be needed to eventually cure DMD. Our study provides an important first step toward this goal.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. Li, Y. Yue, and D. Duan Preservation of Muscle Force in Mdx3cv Mice Correlates with Low-Level Expression of a Near Full-Length Dystrophin Protein Am. J. Pathol., May 1, 2008; 172(5): 1332 - 1341. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Jung, A. S. Martins, E. Niggli, and N. Shirokova Dystrophic cardiomyopathy: amplification of cellular damage by Ca2+ signalling and reactive oxygen species-generating pathways Cardiovasc Res, March 1, 2008; 77(4): 766 - 773. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Bostick, Y. Yue, C. Long, and D. Duan Prevention of Dystrophin-Deficient Cardiomyopathy in Twenty-One-Month-Old Carrier Mice by Mosaic Dystrophin Expression or Complementary Dystrophin/Utrophin Expression Circ. Res., January 4, 2008; 102(1): 121 - 130. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Duan Challenges and opportunities in dystrophin-deficient cardiomyopathy gene therapy Hum. Mol. Genet., October 15, 2006; 15(suppl_2): R253 - R261. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Zhu, L. Zhou, S. Mori, Z. Wang, C. F. McTiernan, C. Qiao, C. Chen, D. W. Wang, J. Li, and X. Xiao Sustained Whole-Body Functional Rescue in Congestive Heart Failure and Muscular Dystrophy Hamsters by Systemic Gene Transfer Circulation, October 25, 2005; 112(17): 2650 - 2659. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.W. Ashford Jr, W. Liu, S.J. Lin, P. Abraszewski, S.D. Caruthers, A.M. Connolly, X. Yu, and S.A. Wickline Occult Cardiac Contractile Dysfunction in Dystrophin-Deficient Children Revealed by Cardiac Magnetic Resonance Strain Imaging Circulation, October 18, 2005; 112(16): 2462 - 2467. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Yue, J. W. Skimming, M. Liu, T. Strawn, and D. Duan Full-length dystrophin expression in half of the heart cells ameliorates {beta}-isoproterenol-induced cardiomyopathy in mdx mice Hum. Mol. Genet., August 1, 2004; 13(15): 1669 - 1675. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. A. Lapidos, R. Kakkar, and E. M. McNally The Dystrophin Glycoprotein Complex: Signaling Strength and Integrity for the Sarcolemma Circ. Res., April 30, 2004; 94(8): 1023 - 1031. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2003 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |