Introduction

Duchenne muscular dystrophy (DMD) is a common X-linked progressive muscular disorder that occurs at the frequency of one in 3500 male births. Most of the affected males die of respiratory or heart failure around the third decade of life.^{1, 2, 3} This severe muscle disorder is caused by the absence of a muscle cytoskeleton protein, dystrophin. Restoration of dystrophin protein expression by gene therapy is apparently an important method that may offer a cure for this disease. However, the widespread loss of the dystrophin protein in all muscles of an affected individual represents a major challenge for the development of an effective gene therapy treatment. For this reason, the development of a systemic delivery system, which can ideally deliver dystrophin gene into all muscles, is imperative to the success of gene therapy for DMD.

Gene transfer of dystrophin gene to skeletal muscle has mainly been carried out by intramuscular injection of naked DNA⁴ or viral vector.^{5, 6, 7, 8, 9} Unfortunately, it is unlikely that direct intramuscular injection of naked DNA is a clinically viable method of delivering gene into DMD patients. First of all, the level of gene expression is too low to have any sufficient significance. More importantly, the muscle fibers expressing the transgene product is very spatially limited; gene expression only occurs in a few square centimeters near the injection site.

Neutral polymers, such as polyvinyl pyrrolidone, have been used to enhance gene expression in muscle; however, there is no significant improvement in the area of gene expression.¹⁰ Intravascular delivery of naked DNA to muscle is apparently an attractive approach for overcoming this problem, because it would permit DNA to reach all muscle fibers, thereby leading to more widespread therapeutic gene expression.

Recently, Budker et al¹¹ developed a semisystemic method that could deliver plasmid DNA into a whole leg of a rat They injected DNA solution through the femoral artery of rats while all blood vessels to and from the leg were occluded. Gene expression was detected in all muscle groups of the leg. This method represents a significant advance towards the development of gene therapy for DMD. More importantly, Zhang et al¹² has demonstrated the applicability of this method in primate. However, the applicability of this method to restore dystrophin protein on DMD animal models, such as mouse or dog, remains unknown.

In this study, we report a semisystemic delivery method that can efficiently transfect all muscles in both hind limbs of mice. We demonstrate that it is possible to achieve widespread restoration of dystrophin protein in all hind-limb muscles of mdx mice by intravascular injection of naked DNA encoding the dystrophin gene. We also investigate factors that limit the naked DNA delivery through intravascular injection. These results have important implications with regard to the development of gene therapy for DMD.

Results

Injection route

To examine any difference in gene expression in the muscle after DNA injection through the tail artery or tail vein, 100 g of DNA in 2 ml phosphate-buffered saline (PBS) solution was injected through either the tail artery or tail vein, while the aorta and vena cava were clamped (Figure 1). At 2 days after injection, muscle was collected and luciferase gene expression in all muscle groups was measured. As shown in Figure 2, a relatively high level of gene expression could be detected in all muscle groups in both cases. In addition, the level of gene expression was about 10-fold greater in mice receiving DNA through the tail artery than in those injected through the tail vein. Furthermore, the levels of gene expression in one leg by the tail vein injection could be dramatically enhanced when the blood flow through the other leg was blocked (data not shown). The tail vein injection, technically speaking, is easier to perform than the artery injection.

Figure 1.

Schematic illustration of tail artery and tail vein injection of DNA solution. DNA PBS solution (2 ml) was injected through the tail artery, while the aorta and vena cava were clamped as indicated.

Full figure and legend (23K)

Figure 2.

Effect of intravenous and intra-artery injection on gene expression in muscles. DNA (100 g) in 2 ml PBS solution were injected either through the tail vein (close bar) or the tail artery (open bar), while the blood flow in the aorta and vena cava was occluded. Gene expression in biceps and semitendinous (B), gastrocnemius (G), quadriceps (Q) and tibialis (T) were examined 48 h after injection. Data represent mean s.d. (n=5).

Full figure and legend (21K)

Injection speed, volume, and ischemia time

To optimize the condition of injection for gene transfer, we examine the effects of injection speed, injection volume, and ischemia time on gene expression in muscles. To study the effect of injection speed, 2 ml of DNA solution containing 100 g of luciferase plasmid were injected through the tail artery of mice in 5 or 20 s. Luciferase gene expression in mouse legs was examined 2 days later (Figure 3a). Gene expression in those mice where the DNA solution were injected within 5 s were about a thousand times higher than that in the mice where the DNA solution were injected in 20 s. This result indicates that hydrodynamic pressure induced by injection within a short time period is important for muscle gene transfer.

Figure 3.

Effect of injection speed, volume and ischemia time on gene expression in mouse legs. DNA (100 g) in 2 ml PBS solution was injected through the tail artery in 5 or 20 s (a) or DNA in 1 or 2 ml PBS was injected in 5 s (b), while the aorta and vena cava were clamped for 15 s, 5 or 10 min, respectively (c). Gene expression in biceps and semitendinous (B), foot (F), gastrocnemius (G), quadriceps (Q) and tibialis (T) were examined 48 h after injection. Data represent means.d. (n=5).

Full figure and legend (65K)

In another series of experiment, the effect of injection volume was examined. A measure of 1, 2 or 3 ml of DNA solution was injected into mice within 5 s and the luciferase gene expression was examined 2 days later (Figure 3b). Injection of 1 ml DNA solution resulted in gene expressions more than 10-fold lower than those injected with 2 ml DNA solution. Further increase in injection volume did not result in significant increase in gene expression (result not shown for 3 ml injection). This result together with the result of injection speed indicates that increased hydrodynamic pressure is critical for naked DNA delivery to the muscle.

To examine whether high pressure should be maintained after the injection to achieve high levels of gene expression, the blood flow was blocked for different periods of time. Figure 3c shows that there was no significant difference in gene expression in muscles whether the blood flow was blocked for 15 s, 5, or 10 min after DNA was injected. This result indicates that maintaining a high pressure is probably not a critical factor once the DNA is delivered to the muscle tissue.

Effect of histamine on gene expression

As the extravasation of plasmid DNA might be limited by the presence of endothelium, we tested whether histamine, which increases the endothelium permeability, could increase the gene expression in the muscle (Figure 4). In mice preinjected with histamine, gene expressions in all different muscle groups were more than 10-fold higher than in mice preinjected with PBS. This result indicates that the endothelium is one of the major barriers for naked DNA delivery to the muscle.

Figure 4.

Effect of histamine on gene expression in mouse legs. DNA (2 ml) in PBS solution were injected through the tail artery, while the aorta and vena cava were clamped. At 5 min prior to the DNA injection, 1 ml PBS (control) or 1 ml histamine (histamine) in PBS were injected through the tail artery. Gene expressions were examined 48 h after injection. Data represent means.d. of luciferase activity in biceps and semitendinous (B), foot (F), gastrocnemius (G), quadriceps (Q) and tibialis (T) (n=5).

Full figure and legend (18K)

DNA dose–response

Three doses of luciferase DNA were tested for gene expression in muscles. Mice were injected with 25, 100, or 200 g luciferase DNA in 2 ml PBS solution in 5 s and muscles were collected 2 days after of injection. The results are shown in Figure 5. Gene expression in all muscle groups increased with increasing amount of DNA injected. Even with the injection of 200 g of DNA, no saturation has observed. It is known that in intramuscular injection, the uptake of DNA normally saturates at 100 g of DNA.¹³ The reason for this discrepancy could be due to the fact that intra-artery injection can deliver plasmid to many more muscle fibers, therefore the cell surface available for DNA uptake is much greater than the case for intramuscular injection. In addition, our method of occlusion does not guarantee that the DNA solution only flows to the muscle; some could be delivered to the abdominal organs.

Figure 5.

Dose–response of luciferase gene expression in different groups of mouse muscles. Biceps and semitendinous (B), gastrocnemius (G), quadriceps (Q) and tibialis (T) of mice were examined for luciferase gene expression 2 days after the mice were injected with different amounts of luciferase plasmid in 2 ml PBS solution through the tail artery. Data represent means.d. (n=4).

Full figure and legend (51K)

Time course of DNA expression

Intramuscular injection of plasmid DNA results in a relatively long period of gene expression.¹⁴ To examine whether this is the same with intra-artery injection, luciferase expression was measured at various time points after intra-artery injection of 100 g of DNA in 2 ml of saline solution. The results are shown in Figure 6. The gene expression reached its peak between 1 and 2 weeks. After 2 weeks, gene expression started to decline. At 1 month, gene expression in muscles was about 50% of the peak level. This result suggests that the duration of gene expression after intra-artery injection is comparable to that of the intramuscular injection.

Figure 6.

Time course of gene expression in mouse legs after intra-artery injection. DNA (100 g) in 2 ml PBS solution were injected through the tail artery, while the aorta and vena cava were clamped. Gene expression in biceps and semitendinous (b), gastrocnemius (G), quadriceps (Q) and tibialis (T) were examined at 1, 2, 7, 14 and 28 days after the injection. Data represent means.d. (n=4).

Full figure and legend (25K)

Study of LacZ gene expression in muscle

To examine the percentage of muscle fibers transfected by intra-artery injection of plasmid, 400 g of plasmid containing the lacZ gene in 2 ml PBS solution were injected. Figure 7 shows that large numbers of myofibers expressing the lacZ gene product could be observed 7 days after the injection of plasmid DNA. Unlike gene expression using intramuscular injection, lacZ gene expression after intra-artery administration was much more widely spread. This is evident in all muscle groups (data not shown). However, the percentage of muscle fibers transfected was modest; about 10% of muscle fibers were transfected as defined by LacZ expression.

Figure 7.

LacZ gene expression in the muscle after intraartery injection. LacZ activity staining was carried out on the quadriceps muscles 7 days after intra-artery injection of 400 g LacZ DNA in PBS. Magnification 200 (a); 100 (b).

Full figure and legend (196K)

Plasmid distribution in muscle tissue

As the percentage of muscle fibers transfected was modest by intra-artery injection of plasmid, we wondered whether plasmid DNA had reached all muscle cells by this method of delivery. To answer the question, plasmid DNA was labeled with rhodamine-labeled peptide nucleic acid (PNA) so that the plasmid DNA still maintained a supercoiled conformation after labeling.¹⁵ We examined the plasmid distribution at 5 and 30 min after intra-artery injection. As shown in Figure 8, at 5 min after injection, almost all muscle fibers were surrounded by red-colored plasmid DNA. At 30 min, plasmid DNA started to be seen inside some of the cells, but the majority of DNA still remained outside the cells. This result indicates that plasmid DNA can overcome the endothelial barrier by injection through the tail artery. However, the muscle cell membrane seemed to become the main barrier for the uptake of naked DNA delivered intravascularly.

Figure 8.

Naked DNA distribution in muscles after intra-artery injection. Rhodamine-labeled DNA (100 g) in 2 ml PBS solution were injected through the tail artery. Muscles were collected at 5 min (a, c) and 30 min (b, d). Photomicrographs were taken at 200 (a, b) and 100 magnifications (c, d).

Full figure and legend (309K)

Effect of intra-artery injection on muscle integrity

Large volume of injection solution is required to increase the hydrodynamic pressure inside the blood vessel so that plasmid DNA can be forced to pass through the endothelial wall. One major concern of this method of delivery is whether the injection will cause severe damage to muscle cells. To address this issue, we examined the morphology of muscle fibers by hematoxylin and eosin (H&E) staining after intra-artery injection of 2 ml DNA solution. Muscles were collected and examined 2 and 7 days after injection. As shown in Figure 9, muscles showed slight widening of the interstitial endomysial space 2 days after injection. However, there was no evidence of muscle damage at this time point. By 7 days after injection, muscles displayed a normal histological appearance (Figure 9). No inflammation, nor lymphocyte or leukocyte infiltration, was evident in both time points, indicating that intra-artery injection of plasmid DNA does not induce adverse immunological reaction in the muscle.

Figure 9.

Histology of mouse quadriceps. Quadriceps were collected from untreated mouse (a) or mice 2 days (b, c) or 7 days (d) after injection with 100 g luciferase plasmid in 2 ml PBS solution through the tail artery. Magnification 200. Arrows indicate the widening of space between interstitial space.

Full figure and legend (214K)

Restoration of full-length dystrophin protein

The efficacy of expression of full-length dystrophin cDNA by intra-artery (Figure 10a) or tail vein injection (Figure 10b) was examined 7 days after injection of 500 g of plasmid DNA containing full-length murine dystrophin cDNA into the mdx or wild-type C57 mice (Figure 10). Immunofluorescence staining of normal skeletal muscle in C57 mice revealed dystrophin as a continuous staining along the plasma membrane of every muscle fiber. In untreated mdx mice, this pattern of staining was not observed. On the other hand, the dystrophin-positive fibers were clearly observed in mdx mice injected with full-length dystrophin plasmid. Even though the level of dystrophin expressed were lower than that of C57 mice, more than 30% of muscle fibers in all muscle groups are dystrophin positive (Table 1). In treated mdx mice, some vasculatures (arrows in Figure 10a) were also stained due to nonspecificity of the method used. The data for intravenous injection (Figure 10b) is very similar to that of the intra-artery injection (Figure 10a). The quantitative analysis of the bands of Western blot in Figure 10c by NIH Imager 1.6 showed about 31% of dystrophin expression, compared to the C57/BL mice, when mdx mice were transferred with dystrophin gene.

Figure 10.

Immunofluorescence localization of dystrophin and Western blot for dystrophin assay. pFDPC (500 g) containing the full-length dystrophin cDNA were injected through the aorta (Figure 10A). Quadriceps (c) and gastrocnemius (d) of the treated mdx mice were collected 7 days after the injection. Controls are cross-sections of quadriceps from untreated mdx (a) and C57/BL10 (b) mice. Green fluorescence indicates dystrophin staining and blue fluorescence indicates nuclei stained with DAPI. Arrows indicate possible staining of the vasculature. The same amount of dystrophin cDNA was injected through the tail vein flowing clamping the aorta and vena cava and occluding the blood flow through one of legs (Figure 10B,C). The dystrophin staining was carried out by using a polyclonal antidystrophin antibody 6–10. Controls were the sections of quadriceps from C57/BL10 (a) and untreated mdx (b). Quadriceps (c) and gastrocnemius (d) of the treated mdx mice were collected 7 days after the injection. Magnification 100. Top panel in Figure 10c: Western blotting of total protein extracted from the mice quadraceps treated with saline or dystrophin gene Lower panel: Bands of myosin (200 kDa) on the post-transfer Commossie gel serving as loading control.

Full figure and legend (272K)

Table 1 - Percentage of myofibers expressing dystrophin in different muscle groups^*.

Full table

Discussion

The pathological severeness of DMD has prompted extensive research on developing gene therapy methods for the treatment of this chronic but deadly disease. Different methods including myoblast transplantation¹⁶ and gene transfer have been explored. Among the gene transfer methods, viral vectors such as herpes simplex virus (HSV),¹⁷ adenovirus^{18, 19, 7} and adeno-associated virus^{20, 8, 9} or nonviral vectors including HVJ liposome²¹ and naked DNA have been tested. Although viral vectors can achieve high degrees of gene transfer efficiency, their use for the delivery of dystrophin has encountered many problems. For example, although HSV and adenoviral vectors can deliver the full-length dystrophin gene, they both suffer from strong immune response in the host and loss of infectivity when muscle fibers are mature.²² Adeno-associated viral vectors, on the other hand, are capable of efficiently infecting both young and mature muscle fibers, but their small packaging limit makes it impossible to deliver full-length dystrophin gene. Compared to viral vectors, the use of naked DNA to treat DMD has certain advantages. First, naked DNA appears to have no significant side effects when delivered to muscle. Although there are some reports with regard to inflammatory responses arising from the CpG motifs on the plasmid DNA,^{23, 24} intra-artery injection of naked DNA does not seem to cause any significant inflammation as demonstrated in our study (Figure 9). This may be because of the absence of cationic carrier, which normally enhance the inflammatory response.²⁵ Furthermore, skeletal muscle seems to internalize naked DNA more efficiently than other types of tissue.²⁶

Our study demonstrates that it is possible to achieve widespread gene expression in skeletal muscles by intravascular injection of naked DNA. Gene expression was detected in all muscle groups in either one or both hind limbs of mice (Figures 2, 3, 4, 5 and 10, Table 1). There are apparently two major barriers for intravascular delivery of naked DNA to skeletal muscles. The first barrier is the endothelium wall, which apparently limits the extravasation of naked DNA to reach the muscle tissue. It was reported that the endothelium in muscle is different from the endothelium in liver or spleen. The latter is noncontinuous and fenestrated, containing many large pores. The former is of continuous and nonfenestrated type.²⁷ Despite its small size, naked DNA apparently cannot pass through the endothelium freely. In fact, the endothelium barrier is not only a barrier for naked DNA but also a barrier for viral vectors such as adenoviral and adeno-associated viral vectors as well.^{28, 29} For gene therapy to be successful for systemic muscular disorders such as DMD, overcoming the endothelium barrier becomes the first prerequisite. Increasing the hydrodynamic pressure by injecting a large volume of DNA solution within a short time period seems to be an effective method to overcome this barrier. This method has been applied for both naked DNA- and viral vector-mediated gene transfer to muscles.^{11, 28, 29} A relatively large volume of solution in the vasculature may expand the endothelium and stretch the pores on the endothelium. Second, the elevated hydrodynamic pressure in the vasculature may force the DNA solution to pass through the endothelium and disperse it in the muscle tissue. Our results also showed a significant difference between intravenous injection and intra-artery injection. This significant difference may be simply due to the fact that arteries are less distensible than veins and so the plasmid solution is transmitted directly to the capillaries, where the hydrostatic pressure forces the solution into the interstitial space. In contrast, delivery via the venous system would lead to pooling of the plasmid solution in the veins as they can expand to accommodate very large volumes. In addition, major veins have valves that act to prevent blood flowing back towards the capillaries and these may have a barrier effect when venous delivery to the hind limbs is attempted. Other methods have also been used to enhance the extravasation of DNA. For example, Budker et al¹¹ have combined vasodilator and collagenase, which can digest the basal membrane of the endothelium, to increase the permeability of the endothelium and they observed further increase of gene expression. In our study, we have used histamine to increase the permeability of the endothelium. In histamine-treated mice, gene expression was enhanced about 10-fold in all muscle groups of the mouse legs. The application of histamine to increase the extravasation of adenoviral vector has also been tested and similar level of enhancement in gene expression was observed.²⁹ The combination of large volume injection with some agents that can increase the endothelium permeability seems to be an effective method to overcome the endothelium barrier.

The second barrier for gene delivery to skeletal muscle appears to be the cellular uptake of naked DNA. Using fluorescence-labeled plasmid DNA, we clearly demonstrate that our method can deliver plasmid DNA to almost all muscle fibers in the tissue (Figure 8). However, only a relatively small percentage of muscle fibers showed gene expression when LacZ was used as a reporter gene. Therefore, muscle fibers seem to resist naked DNA uptake. This barrier appears to be the major barrier for gene delivery to skeletal muscle. Overcoming this barrier will have significant implication for gene therapy for DMD.

Compared to intramuscular injection, in which only a few percent of muscle fibers are been transfected, our method represents a significant advancement. Although in our LacZ study, the percentage of muscle fibers expressing LacZ is still low, the gene was expressed in all the muscle groups in mouse legs. Furthermore, it was reported that LacZ staining normally underestimates the gene expression about three-fold.³⁰ Indeed, our immunostaining with dystrophin gene expression revealed a much wider area of gene expression (Figure 10 and Table 1). However, it is not clear whether muscle function can be restored at this expression level. Function studies are required to assess the effect. Furthermore, data shown in Figure 10 seem to indicate that dystrophin was also expressed in the vasculature cells of the treated mouse. This could be due to the fact that the dystrophin gene used in this study was not controlled by a muscle-specific promoter. With the development of more efficient DNA vectors such as the ones with a targeting ligand attached,³¹ and with the use of a muscle-specific gene expression system, specific gene expression may be further increased.

In summary, the results of our study demonstrates the possibility of systemic delivery of naked DNA and achieve widespread expression of dystrophin gene in multiple skeletal muscles in mdx mice. However, the practical application of this method to treat DMD is still limited by the relatively low efficiency of DNA uptake by muscle cells and relatively short period of transgene expression. In addition, injecting large volume of solution locally may be applicable to human, as demonstrated by Zhang et al, ¹² in monkeys. However, the application of this method to the diaphragm would be problematic. Different delivery method for the naked DNA to the diaphragm has recently been reported by this lab.³² It is possible to treat DMD by treating different organs separately. However, before this can happen, problems such as transfection efficiency and short-term expression have to be overcome.

Materials and methods

Plasmids

pCMV-Luc plasmid containing firefly luciferase cDNA driven by CMV immediate-early promoter was constructed by introducing the luciferase cDNA into pNGVL3 plasmid (National Gene Vector Laboratory, University of Michigan, MI, USA). pCMVLacZ plasmid was obtained from Invitrogen (Palo Alto, CA, USA). pFDPC plasmid containing full-length dystrophin cDNA under the control of CMV promoter was constructed by digesting pSRDMD (a gift from Dr Paula Clemens) with NotI and ligating to the NotI site of pcDNA 3 (Invitrogen, Palo Alto, CA, USA). For plasmid distribution study, pWIZ-Lux plasmid obtained from Gene Therapy System (GTS, San Diago, CA, USA) was labeled with rhodamine-labeled PNA according to a previous report.¹⁵ All plasmids were amplified in Escherichia coli and purified by Qiagen Giga plasmid preparation kit (Qiagen, Valencia, CA, USA).

Animal procedures

All experiments were performed using CD-1 mice (4–6 weeks old) or X-linked muscular dystrophy (mdx) mice (4–6 weeks old). Prior to any surgical procedure, the animals were anesthetized by intraperitoneal injection of 2,2,2-tribromoethanol (500 mg/kg). For plasmid injection, the tail artery or tail vein was cannulated with a 25 G butterfly needle connected to a polyethylene tubing (Harvard Apparatus, Holliston, MA, USA). The needle was subsequently fixed to the tail with instant adhesive (Loctite 404, Applied Industrial Technology, Pittsburgh, PA, USA). The abdomen was then opened to expose the aorta and the vena cava. Prior to injection, a microvascular clamp (Harvard Apparatus, Holliston, MA, USA) was placed on the aorta and vena cava at the location just below the kidneys (Figure 1) so that the DNA solution can pass through the common iliac artery and reach the leg muscles after injection from the tail artery or tail vein. In a typical experiment, 2 ml of PBS solution containing 100 g of luciferase plasmid or 400 g of -galactosidase plasmid was injected in 5 s. Alternatively, DNA solution was injected to mice through the tail vein following clamping the aorta, vena cava and occlusion of the blood flow through one of the legs using modified artery forceps. The clamps were removed 15 s after injection and mice were allowed to recover from the effect of anesthesia. To examine the effect of histamine, 1 ml of PBS with or without histamine (10 mM) was first injected, and 5 min later, 2 ml of luciferase plasmid solution was then injected.

For the administration of dystrophin plasmid, the aorta of mdx mice was cannulated with a microcannula (Harvard Apparatus, Holliston, MA, USA), 1.5 ml of 10 mM histamine solution in PBS containing 500 g of the full-length dystrophin plasmid was injected in 5 s. After injection, the incision on aorta was closed with suture and mice were allowed to recover from the effect of anesthesia.

-Galactosidase expression analysis

At 1 week after the intravascular injection of pCMVLacZ, mice were killed by cervical dislocation. All muscle groups were harvested and flash-frozen in isopentane cooled in liquid nitrogen. Serial cross-sections (12 m) were placed onto glass slides (Super-frost plus; Fisher Scientific, Pittsburgh, PA, USA) and kept at –80°C. X-gal staining was carried out using X-galactosidase staining kit (Invitrogen, Palo Alto, CA, USA) according to the manufacture's procedure. Briefly, muscle sections were fixed in 1% glutaldehyde for 10 min and then rinsed with PBS. Sections were then stained overnight for -galactosidase activity by incubation at 37°C with 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-gal, 1 mg/ml), 1 mM MgCl₂, 5 mM potassium ferrocyanide, and 5 mM potassium ferricyanide in PBS. After washing in PBS, the slides were counterstained in alcoholic eosin and then dehydrated and mounted. Photographs were taken with a Nikon TE-300 fluorescence microscope. To quantitate the percentage of LacZ-positive fibers, the number of positive cells and the total number of cells were counted from 15 photographs covering different areas of muscle section.

Luciferase gene expression analysis

For luciferase assay, the entire muscle groups were collected from each mouse leg 2 days after injection, except for the time-course study. The muscles were homogenized in lysis buffer (1% Triton X-100, 2 mM EDTA, 0.1 M Tris, pH 7.8) using a tissue tearer (Biospec Products, Bartlesville, OK, USA). The homogenates were centrifuged at 14 000 g for 10 min at 4°C and 10 l of the supernatant was analyzed for luciferase activity using an LB 953 luminometer (Berthod). The results were expressed as relative light units per mg of tissue protein. Protein assay was carried out with the Coomassie plus reagent (Pierce, Rockford, IL, USA).

Plasmid distribution and histological examination of the muscle damage

PBS (2 ml) solution containing 100 g of luciferase plasmid labeled with rhodamine-PNA was injected through the tail artery as described above. At 5 and 30 min after injection, the quadriceps were collected, flash-frozen and sectioned. The slides were examined and photographed using a Nikon T-300 fluorescence microscope. For the muscle damage study, 2 ml of PBS solution containing luciferase plasmid was injected as above. Quadriceps were collected at 2 and 7 days after injection, sectioned and stained with H&E.

Immunofluorescence detection of dystrophin gene expression

Serial crosscryosections (6 m) were collected. Immunostaining of dystrophin was performed with the Mouse-on-Mouse Kit (Vector Laboratories) according to the manufacturer's protocol without fixing the muscle sections. Primary monoclonal antibody against the C-terminus of dystrophin (NCL-Dys2; 1:20 dilution) was purchased from NovoCastra Laboratories (Burlingame, CA, USA). Muscle cell nuclei were counterstained with a mounting medium containing 4', 6'-diamidino-2-phenylindole (DAPI) (Vector Laboratories). For quantitation of dystrophin-positive fibers, the number of dystrophin-positive cells and the total number of cells were counted from 15 photographs covering different areas of muscle section. For intravenously injected animals, the sections were preincubated for 1 h at room temperature with 10% horse serum in PBS (pH 7.4) and then incubated overnight with affinity purified rabbit polyclonal antidystrophin antibody 6–10 (a gift from EP Hoffman).^{33, 34} After four rinses in 10% horse serum/PBS, the sections were incubated with Alexa Fluor 568 goat anti-rabbit antibody (Molecular Probes, 1:200 dilution) for 1 h. As controls, the muscle sections from C57BL/10 and untreated mdx mice were similarly processed. The sections were examined and photographed with a Nikon TE-300 fluorescence microscope.

Dystrophin assay by Western blot

Methods

Mice quadriceps were sectioned and extracted at TEE buffer that contains 20 mM Tris pH 8.0, 1 mM EDTA, 1 mM EGTA and then mixed with same volume of 10% SDS and incubate in ice for 30 min. Protein concentration was determined using BCA protein assay kit (Biorad). The samples were then subjected to electrophoresis of 7.5% SDS-PAGE gel, transferred to nitrocellulose membrane. Dystrophin expression was then detected using a high-affinity antidystrophin antibody anti 6–10, a generous gift from Dr LM Kunkel (Children's hospital and Harvard Medical School, Boston, MA, USA), and visualized with anti-rabbit secondary antibody linked to horseradish peroxidase and chemiluminescence (Amersham).

References

1. Hoffman EP, Brown Jr RH, Kunkel LM. Dystrophin: the protein product of the Duchenne muscular dystrophy locus. Cell 1987; 51: 919–928. | Article | PubMed | ISI | ChemPort |
2. Zubrzycka-Gaarn EE et al. The Duchenne muscular dystrophy gene product is localized in sarcolemma of human skeletal muscle. Nature 1988; 333: 466–469. | Article | PubMed | ChemPort |
3. Campbell KP, Kahl SD. Association of dystrophin and an integral membrane glycoprotein. Nature 1989; 338: 259–262. | Article | PubMed | ISI | ChemPort |
4. Acsadi G et al. Human dystrophin expression in mdx mice after intramuscular injection of DNA constructs [see comments]. Nature 1991; 352: 815–818. | Article | PubMed | ISI | ChemPort |
5. Deconinck N et al. Functional protection of dystrophic mouse (mdx) muscles after adenovirus-mediated transfer of a dystrophin minigene. Proc Natl Acad Sci USA 1996; 93: 3570–3574. | Article | PubMed | ChemPort |
6. Yang L et al. Adenovirus-mediated dystrophin minigene transfer improves muscle strength in adult dystrophic (MDX) mice. Gene Therapy 1998; 5: 369–379. | Article | PubMed |
7. Gilbert R et al. Prolonged dystrophin expression and functional correction of mdx mouse muscle following gene transfer with a helper-dependent (gutted) adenovirus-encoding murine dystrophin. Hum Mol Genet 2003; 12: 1287–1299. | Article | PubMed | ISI | ChemPort |
8. Harper SQ et al. Modular flexibility of dystrophin: implications for gene therapy of Duchenne muscular dystrophy. Nat Med 2002; 8: 253–261. | Article | PubMed | ISI | ChemPort |
9. Watchko J et al. Adeno-associated virus vector-mediated minidystrophin gene therapy improves dystrophic muscle contractile function in mdx mice. Hum Gene Ther 2002; 13: 1451–1460. | Article | PubMed | ISI | ChemPort |
10. Mumper RJ et al. Protective interactive noncondensing (PINC) polymers for enhanced plasmid distribution and expression in rat skeletal muscle. J Control Rel 1998; 52: 191–203. | Article |
11. Budker V et al. The efficient expression of intravascularly delivered DNA in rat muscle. Gene Therapy 1998; 5: 272–276. | Article | PubMed | ISI | ChemPort |
12. Zhang G et al. Efficient expression of naked DNA delivered intraarterially to limb muscles of nonhuman primates. Hum Gene Ther 2001; 12: 427–438. | Article | PubMed | ISI | ChemPort |
13. Levy MY et al. Characterization of plasmid DNA transfer into mouse skeletal muscle: evaluation of uptake mechanism, expression and secretion of gene products into blood. Gene Therapy 1996; 3: 201–211. | PubMed | ISI | ChemPort |
14. Wolff JA et al. Long-term persistence of plasmid DNA and foreign gene expression in mouse muscle. Hum Mol Genet 1992; 1: 363–369. | Article | PubMed | ChemPort |
15. Zelphati O et al. Gene chemistry: functionally and conformationally intact fluorescent plasmid DNA. Hum Gene Ther 1999; 10: 15–24. | Article | PubMed | ISI | ChemPort |
16. Karpati G et al. Dystrophin is expressed in mdx skeletal muscle fibers after normal myoblast implantation. Am J Pathol 1989; 135: 27–32. | PubMed | ISI | ChemPort |
17. Akkaraju GR et al. Herpes simplex virus vector-mediated dystrophin gene transfer and expression in MDX mouse skeletal muscle. J Gene Med 1999; 1: 280–289. | Article | PubMed | ISI | ChemPort |
18. Acsadi G et al. Dystrophin expression in muscles of mdx mice after adenovirus-mediated in vivo gene transfer. Hum Gene Ther 1996; 7: 129–140. | PubMed | ISI | ChemPort |
19. Jani A et al. Generation, validation, and large scale production of adenoviral recombinants with large size inserts such as a 63 kb human dystrophin cDNA. J Virol Methods 1997; 64: 111–124. | Article | PubMed | ChemPort |
20. Wang B, Li J, Xiao X. Adeno-associated virus vector carrying human minidystrophin genes effectively ameliorates muscular dystrophy in mdx mouse model. Proc Natl Acad Sci USA 2000; 97: 13714–13719. | Article | PubMed | ChemPort |
21. Yanagihara I et al. Expression of full-length human dystrophin cDNA in mdx mouse muscle by HVJ-liposome injection. Gene Therapy 1996; 3: 549–553. | PubMed |
22. Feero WG et al. Viral gene delivery to skeletal muscle: insights on maturation-dependent loss of fiber infectivity for adenovirus and herpes simplex type 1 viral vectors. Hum Gene Ther 1997; 8: 371–380. | PubMed | ChemPort |
23. Krieg AM et al. CpG motifs in bacterial DNA trigger direct B-cell activation. Nature 1995; 374: 546–549. | Article | PubMed | ISI | ChemPort |
24. McMahon JM et al. Inflammatory responses following direct injection of plasmid DNA into skeletal muscle. Gene Therapy 1998; 5: 1283–1290. | Article | PubMed | ChemPort |
25. Whitmore M, Li S, Huang L. LPD lipopolyplex initiates a potent cytokine response and inhibits tumor growth. Gene Therapy 1999; 6: 1867–1875. | Article | PubMed | ISI | ChemPort |
26. Davis HL et al. Plasmid DNA is superior to viral vectors for direct gene transfer into adult mouse skeletal muscle. Hum Gene Ther 1993; 4: 733–740. | PubMed | ISI | ChemPort |
27. Taylor A, Granger D. The cardiovascular system: microcirculation. In: Renkin E, Michel CC, Geiger SR (eds). Handbook of Physiology, Am. Physiol. Soc.: Bethesda, MD 1984: 467–520.
28. Cho WK et al. Modulation of Starling forces and muscle fiber maturity permits adenovirus-mediated gene transfer to adult dystrophic (mdx) mice by the intravascular route. Hum Gene Ther 2000; 11: 701–714. | Article | PubMed | ISI | ChemPort |
29. Greelish JP et al. Stable restoration of the sarcoglycan complex in dystrophic muscle perfused with histamine and a recombinant adeno-associated viral vector. Nat Med 1999; 5: 439–443. | Article | PubMed | ISI | ChemPort |
30. Couffinhal T et al. Histochemical staining following LacZ gene transfer underestimates transfection efficiency. Hum Gene Ther 1997; 8: 929–934. | PubMed | ChemPort |
31. Liang KW, Hoffman EP, Huang L. Targeted delivery of plasmid DNA to myogenic cells via transferrin-conjugated peptide nucleic acid. Mol Ther 2000; 1: 236–243. | Article | PubMed | ChemPort |
32. Liu F et al. Transfer of full-length Dmd to the diaphragm muscle of Dmd(mdx/mdx) mice through systemic administration of plasmid DNA. Mol Ther 2001; 4: 45–51. | Article | PubMed | ISI | ChemPort |
33. Koenig M, Kunkel LM. Detailed analysis of the repeated domain of dystrophin reveals 4 potential hinge regions that may confer flexibility. J Biol Chem 1990; 265: 4560–4566. | PubMed | ISI | ChemPort |
34. Hoffman EP et al. Somatic reversion/suppression of the mouse mdx phenotype in vivo. J Neurol Sci 1990; 99: 9–25. | Article | PubMed | ISI | ChemPort |

Acknowledgements

This work was supported in part by Muscular Dystrophy Association grant and NIH grant PO1 AR45925 (L. Huang) and grant from the Uehara memeriol Foundation (M. Nishikawa). We thank Dr Paula Clemens for providing the dystrophin plasmid and mdx mice. Kenneth W Liang and Makiya Nishikawa were postdoctoral fellows of Duchenne Muscular Dystrophy Research Center, University of Pittsburgh.