Local and distant transfection of mdx muscle fibers with dystrophin and LacZ genes delivered in vivo by synthetic microspheres

Article metrics

Abstract

Patterns of dystrophin and β-galactosidase expression were examined in mdx mice after i.m. injections of synthetic microspheres (MF-2) loaded with full-length (pHSADy) or mini-dystrophin gene (pSG5dys) cDNA plasmid constructs or with LacZ marker gene (pCMV-LacZ). A single injection of 25 μg pHSADy into quadriceps femoris muscle resulted in 6.8% of dystrophin positive myofibers (DPM) in a given muscle; 8.4% of DPM in glutaeus muscle and 4.3% of DPM in quadriceps femoris muscle of contralateral limb on day 21 after exposure compared with only 0.6% DPM in intact (non-injected) mdx mice. A high proportion of DPM (17.6% and 10.8%, respectively) was registered in both injected and contralateral muscles after mini- gene cDNA administration. MF-2/dystrophin cDNA par- ticles were detected by FISH analysis in about 60–70% of myofiber nuclei in muscles of injected and contralateral limbs 7 days after application. The presence of human dystrophin cDNA and its products in all skeletal muscles and in different internal organs was proven by PCR and RT-PCR analysis. Patches of β-galactosidase expression were abundant in injected muscle, and frequent in the contralateral and other skeletal muscles as well as in diaphragm, heart and lungs. High levels of dystrophin cDNA expression, and an efficient distant transfection effect with preferential intranuclei inclusion of MF-2 vehicle, are very encouraging for the development of a new constructive strategy in gene therapy trials of DMD.

Introduction

Duchenne muscular dystrophy (DMD) is the most prevalent of all neuromuscular disorders affecting one in 3500 male births. It results from mutations in an exceptionally large gene which encodes the sarcolemma-bound protein dystrophin.1,2 Myofibers lacking dystrophin are abnormally susceptible to contraction-induced sarcolemma damage with subsequent myofiber dysfunction, necrosis and regeneration leading ultimately to the replacement of the lost fibers by adipose and connective tissue and premature death often associated with cardiomyopathy.3

The absence of any efficient pharmaceutical or biological (myoblast transplantation) methods for the treatment of DMD makes the development of gene therapy approaches for widespread systemic delivery of dystrophin gene to skeletal muscles very urgent for the correction and management of this primary lethal genetic disorder. Traditional approaches for gene delivery to skeletal muscles have focused on the intramuscular injection of viral vectors (adenoviruses, retroviruses, Herpes virus type 1).4 Although highly efficient, all viral vectors tested so far are associated with a number of problems such as immunogenicity, toxicity, potential recombination or complementation.5,6 Much effort, therefore, has been devoted to the development of nonviral gene delivery into skeletal muscles by means of naked plasmid DNA,7 cationic liposomes,8 liposomes combined with polylysin,9 viral oligopeptide complexes10 and even by direct ballistic transfection by means of a gene gun.11 Here we describe the results of our recent findings in mdx mice (a biological model of DMD) after intramuscular gene delivery by means of synthetic microspheres (a novel type of transfer vehicle). Currently in use for drug delivery and targeting, these novel types of biomimetic particles are very promising for drug or gene delivery.12,13

We report here for the first time efficient transfection of mdx mice myofibers with dystrophin gene constructs delivered in vivo by the microsphere particles MF-2. Original findings never previously reported in dystrophin studies concern gene expression in non-injected muscle throughout the body, the so-called ‘distant transfection effect’. High preferential intranuclei incorporation of gene-loaded microsphere particles was observed.

Results

Time and dose-dependent gene expression

The levels of dystrophin gene expression in different skeletal muscles as a function of time was examined in mdx mice after a single i.m. administration of microspheres loaded with 25 μg of full-length dystrophin cDNA (MF2/pHSADy). As shown in Table 1, the proportion of dystrophin positive myofibers (DPM) on cryostat sections of injected quadriceps femoris muscles gradually increased up to an average 5.3% of DPM on day 21 after injection and still remained five- to seven-fold more than the control DPM level (0.6–0.7%) after 2 months exposure. There were similar profiles of DPM counts in the glutaeus muscle of the injected side and quadriceps femoris muscle of the non-injected contralateral side with maximal scores of DPM of about 8% in the former and 4% in the latter at 3 weeks after gene administration. DPM counts in two mdx mice injected three times with 25 μg of pHSADy construct packed in MF2 particles was somewhat less than after a single application of the same dosage and was in the range 2.4–3.9% for all muscles studied.

Table 1 Proportion of dystrophin-positive myofibers (DPM) in the muscles of injected (inj) and non-injected (contrlat) hind limbs of mdx mice at different time intervals (days) following a single treatment with 25 μg of full-length dystrophin cDNA in MF2/pHSADy complexes

High local and distant transfection activity of minigene cDNA

As shown in Figure 1, 25 μg of minigene cDNA construct resulted in 18% of DPM in injected quadriceps muscles (intervals 10–25%) compared with 13.4% of DPM after a double dose (50 μg) of the same construct MF2/pSG5dys and to only 5.8% DPM after MF2/pHSADy application. A significant increase of DPM counts (9.4–10.8%) was registered in quadriceps femoris muscles of the opposite side. Expression of both dystrophin gene constructs after 3 weeks remained within 8–11% limits in the glutaeus muscles of injected limbs.

Figure 1
figure1

Percentage of dystrophin-positive fibers in the skeletal muscles of injected (inj) and non-injected (conrlat) hindlimbs of the mdx mice 3 weeks after a single intramuscular injection of full-length or minigene cDNA expression constructs (pHSADy and pSG5dys, respectively) loaded in MF-2 microspheres.

Effect of human dystrophin constructs on dystrophic muscle histopathology

Cryostat section analysis of all experimental mice revealed two types of DPMs. The major portion of DPM was represented by the fibers with bright presarcolemma staining (ss-DPM) (skeletal muscles (Figure 2a), heart (Figure 2b)) and much less frequent were DPM with intensive cytoplasmic staining (csDPM) (Figure 2c). csDPM were absent or rare in the muscle injected with 25 μg of mini- or full-length dystrophin gene cDNA constructs, but their number significantly increased after administration of 50 μg of pSG5dys (pSG5dys/MF2). As might be inferred from the data shown in Table 2 cross-sectional areas of ssDPM closely approximated the relevant values for muscle fibers in normal C57Bl10J+/+ mice, and were significantly larger than dystrophin negative myofibers (DNM) of the same mdx muscle. Cross-sectional areas of csDPM varied in shape and size and often presented features of degeneration. Massive mononuclear infiltration was typical for the clusters of csDPM after Giemsa staining and indicated muscle damage in these particular areas (Figure 2d). Whereas myonuclei are normally located peripherally in healthy muscle fibers, centrally located nuclei are typical for mdx mouse myofibers that have undergone necrosis and subsequent regeneration. Since dystrophin constructs in these experiments were administered to adult mdx muscle that had already undergone cycles of necrosis,14 central nucleation was present within a major proportion of myofibers at the time of gene transfer. No obvious shift in the number of DPM with peripheral location of the nuclei was detected.

Figure 2
figure2

Immunofluorescence (a, b, c) and immunoperoxidase detection (d) of dystrophin in cross-sections of quadriceps femori muscles (a, c, d) and the heart (b) from mdx mice 3 weeks after intramuscular injection of MF2/pSG5dys complexes. (c, d) Myofibers with sarcolemma (ssDPM) and cytoplasmic (csDPM) dystrophin localization. (c) csDPM marked with arrows. (d) ssDPMs, arrows, csDPMs with mononuclear infiltration, asterisk.

Table 2 Transverse section areas (in μm2) of dystrophin negative (DNM) and dystrophin positive (DPM) myofibers after a single intramuscular injection of 25 and 50 μg of pSG5dys construct packed in synthetic microspheres

Distant transfection phenomena studies

Transfection of contralateral muscles after a single administration of human dystrophin cDNA has never been reported in the numerous gene therapy studies performed in mdx mice with different vectors, different vehicles and different routes of gene delivery. Four experimental approaches were carried out to clarify the nature of this so-called distant transfection phenomenon: (1) Expression and localization of β-galactosidase in MF2/pCMVLacZ-injected mdx mice; (2) PCR analysis of human dystrophin cDNA in different tissues of mdx mice; (3) RT-PCR analysis of dystrophin mRNA in muscle tissues; and (4) FISH analysis of tissue distribution of injected complexes. (1) Wholemount X-gal staining was used to assess gross three-dimensional patterns of LacZ expression in quadriceps femoris muscles, biceps muscles of the forelimb from the opposite side as well as in other internal organs of mdx mice 21 days after a single intramuscular injection of 25 μg of LacZ cDNA with a hCMV promoter (MF2/pCMVLacZ complex). An intensive massive blue staining of myofibers was observed in injected muscle (Figure 3a). The staining appeared as strands that often extended across two-thirds of the muscle length. No preferential expression was observed near the injection site. Intensive blue patches or blue spots were also registered in wholemount preparations of the muscles from the forelimb of the opposite side (Figure 3b). The same type of local blue foci was clearly visible in superficial layers of heart muscle (Figure 3c). These results correspond well to the finding of dystrophin-positive cardiomyocyte clusters after MF2/pSG5dys complex administration into mdx skeletal muscle (Figure 2b). Besides superficial blue spotting and extended patches of β-gal positive tissue, wholemount preparations of lungs after clearing in glycerol and KOH treatment, revealed specific bright blue staining of all respiratory airways. Cryostat sections of the lungs showed large numbers of β-gal positive airway epithelium cells and substantial positive staining of mesenchymal and interstitial cells (not shown). (2) PCR analysis of human dystrophin cDNA demonstrates prolonged survival of human dystrophin cDNA in the muscles and internal organs of mdx mice. A specific amplification product corresponding in size to exons 52 and 53 of the human dystrophin gene, which never gives a positive amplification signal in control mdx mice, could easily be detected in MF2/pHSADy injected quadriceps femoris muscles, as well as in the same muscle from the opposite side at least 2 months after gene delivery (Figure 4a). After 3 weeks exposure, this specific amplification product was detected in all tissue samples studied including heart, lungs, blood and to our surprise in brain, placenta and 18-day-old fetus (Figure 4b). (3) The results of RT-PCR studies shown in Figure 4c unequivocally prove that a specific 325 bp cDNA fragment corresponding to exons 52–53 of the human dystrophin gene could be identified in the RNA samples extracted from experimental and contralateral quadriceps femoris muscles, as well as from the diaphragm muscles of mdx mouse 3 weeks after a single i.m. injection of MF-2 microspheres loaded with the human dystrophin minigene construct pSG5dys. (4) These PCR and RT-PCR findings are in good agreement with relevant FISH analysis with biotin-labeled pHSADy as a probe in different tissues of mdx mice. Microspheres injected into the right quadriceps femoris muscle are quickly distributed throughout the whole body of mdx mice including all skeletal muscles, blood, heart and lungs, as well as brain, placenta and fetal tissues (Figure 5 a–d). The most conspicuous feature of MF-2 microsphere transportation concerns their highly specific targeting of the cell nuclei of all relevant tissues. As shown in Figure 6, 70% of all myofiber nuclei of injected hind limb and 60% of myofiber nuclei in the opposite limb contained brightly FITC-labeled MF2 particles 7 days after a single MF2/pHSADy administration. The proportion of MF-2-positive myonuclei of the same muscles was somewhat reduced, but remained high (about 40%), 3 weeks after injection. Progressive fragmentation of microspheres with a simultaneous increase in the number of small FITC marked spots is already visible after 3 weeks and becomes conspicuous after 2 months of exposure.

Figure 3
figure3

Wholemount β-galactosidase staining of injected (right) (a) and non-injected (left) (b) quadriceps femoris muscle and heart (c) 2 weeks after a single i.m. injection of MF-2 synthetic microspheres containing LacZ cDNA into mdx mice. Lung sample was cleared by KOH-glycerol treatment after modified Dawson technique. Note widespread (a, b) and intensive local (c) blue staining.

Figure 4
figure4

Detection of dystrophin cDNA fragment (325 kb, exons 52–53) by PCR (a, b) and its specific RNA product by RT-PCR (c) after a single intramuscular injection of MF2/pHSADy (a, b) or MF/pSG5dys complexes (c). (a) At different time intervals (1, 7, 21, 60 days) after injection: lanes 1, 3, 5, 7, injected quadriceps femori muscles; 2, 4, 6, 8, DNA samples from contralateral muscle. (b) In different tissues on day 21 after injection: injected muscle (1), heart (2), lungs (3), brain (4), blood (5), fetal membranes (6), placenta (7), 18-day fetus (8), intact mdx muscle (9), positive control (10), molecular weight marker (11). Note the absence of amplification product in intact mdx muscle (9). (c) Detection of specific RNA transcripts (325 bp) by RT-PCR technique 3 weeks after a single i.m. injection of MF-2 microspheres loaded with human dystrophin minigene cDNA construct pSG5dys: injected muscle (1); non-injected (contralateral) muscle of the same mdx mouse (2), diaphragm (3), intact muscle of control mdx mouse-negative control (4), quadriceps femori muscle from C57Bl/6j mouse (positive control) (5), molecular marker (6).

Figure 5
figure5

Detection of FITC-stained complexes MF-2/pHSADy in cryostat sections of myofibers (a), blood leukocytes (b), brain (c), fetal lungs (d) 1 week after a single i.m. injection into mdx mice. Note preferential intranuclei localization of MF2/pHSADy complexes in the muscle fibers (a), its substantial amount in the brain and fetal lung tissues (c, d).

Figure 6
figure6

Relative numbers of mdx myofiber nuclei containing MF2/pHSA-Dy particles 7 and 21 days after a single intramuscular administration.

Discussion

It is known that muscles contribute to about 30–50% of total body mass in humans.16 It is also postulated that phenotypic correction of DMD requires restoration of about 20% of normal dystrophin levels in affected dystrophic muscles.15,17 This level of dystrophin gene expression in limb muscles has recently been achieved in mdx mice transduced with a new type of adenoviral vector4 and this approach looks especially promising when supplemented with transient immunosuppression.5 Retroviral transduction or nonviral dystrophin gene delivery resulted in much less conspicuous transgene expression.6

In spite of some promising findings7 the overall results with nonviral dystrophin gene delivery in mdx muscle still remain rather discouraging. Our recent data with ballistic transfection,11 lipofectamine, naked plasmid and synthetic oligopeptide complexes (SOC)10 imply only moderate transfection efficiency of these vectors. It should be noted, however, that both ballistic transfection and SOC accounted for about 15–17% of DPM in mdx muscles, but only 2–3% of these DPM were represented by typical presarcolemma-stained dystrophin fibers while the rest showed intense cytoplasmic staining of dystrophin. The latter type of DPM is sometimes interpreted as myofibers with overproduction of dystrophin.17,18 Meanwhile histopathological changes in csDPM (mononuclear infiltration, great variations in cross- sectional area with unusually small cross-sections of some fibers, fragmentation of some csDPM, and the absence of contact with neighboring fibers) observed after ballistic transfection, SOC vector delivery and after maximal dosages of MF-2 constructs tested in this study, provide evidence for their interpretation as degenerative myofibers.10,11 It has been postulated that overexpression of transgenic dystrophin corrects dystrophic symptoms in mdx mice without causing deleterious side-effects.16 Thus the actual nature and properties of csDPM still remain obscure and deserve more detailed analysis. The average number of DPM in these experiments using the MF2/pHSADy construct was only about 8%, but all of them could be unambiguously attributed to the ssDPM type. Administration of minigene resulted in a high proportion (up to 25%) of ssDPM; this is comparable to the transfection efficiency of the most successful experiments with dystrophin gene transfer into mdx mice.4,5,16

Absence of any signs of degeneration proved the viability of all these transfected fibers. Higher efficiency of minigene transfection compared with full-length gene cDNA might be partly explained by almost twice the difference in their molecular weights (6.3 and 12 kb) and thus by the higher copy number of minigene cDNA in the same volume of microspheres. Meanwhile, increasing the transgene dosage does not correlate with transfection efficiency. A double dose (50 μg) of the same minigene construct was less effective compared with a single dose (25 μg) and resulted in a significant proportion of csDPM with typical histopathological injuries. It is known that high doses of plasmid DNA (over 250 μg) can completely arrest transfected gene expression.9 The arresting effect with the dystrophin gene in these studies was evident both after triple (25 μg × 3) or double (50 μg) doses of dystrophin constructs delivered by MF-2 particles. Whether this transfection reduction is due to the muscle damage caused by increased vehicle volume in injected muscles or by direct toxic effect of human dystrophin gene overexpression remains to be clarified. Another possible explanation for dosage-related toxicity of MF-2 could be the unexpectedly high inclusion of these rather big constructs into cell nuclei. More details of these will be given in the next section.

At present, it would appear logical that further improvements in transfection efficiency with dystrophin cDNA delivered by microspheres might be achieved by the creation of more efficient plasmid constructs supplemented with more sophisticated regulatory elements such as myosin heavy chain promoter and enhancer.18

A significant increase of DPM count in the mdx muscles not subjected to gene transfection was first reported in our studies with ballistic transfection and synthetic oligopeptide complexes.10,11 Full-length dystrophin cDNA constructs (pHSADy) delivered by these vectors to the right quadriceps femoris muscle resulted in 10- to 20-fold elevation of DPM score in the muscles of the opposite limb. A distant transfection effect with some marker genes has recently been shown for some cationic liposomes,19 but not for the novel adenovirus vectors.4 The distant transfection effect with dystrophin constructs delivered by MF-2 microspheres deserves special consideration. Distant transfection has been shown for all experimental series, irrespective of the type of plasmid constructs, the size of dystrophin cDNA, their doses and even time check interval, although its value was usually below that found in this study of injected muscle. No muscle lesions or any histopathological effects including csDPM were identified in the opposite limb. These immunocytochemical results with dystrophin gene transfection are quite in line with identification of specific fragments of human dystrophin cDNA by PCR analysis and its specific mRNA product by RT-PCR studies in different muscles and non-muscle tissues. Detection of minigene expression product in non-injected quadriceps femoris muscles, as well as in the heart muscle but not in the relevant samples of the control mdx mouse is of special importance. Expression of marker gene LacZ delivered into mdx mice with MF-2 vehicle might also be considered as clear evidence in favor of the distant transfection effect. Unambiguous expression of LacZ gene was registered not only in injected muscle but also in the muscles of the opposite limb and in different internal organs. Of special value is the β-galactosidase-positive reaction in the heart muscle where dystrophy and functional insufficiency is known to be one of the primary causes of DMD patients mortality. Specific expression of the LacZ gene in respiratory epithelium of the lungs might also be of significant interest for gene therapy studies. Elevation of dystrophin and β-galactosidase-positive myofibers in non-injected mdx muscle advocate in favor of a distant transfection phenomenon.

Penetration of the DNA-loaded microspheres into the muscle blood stream is feasible with subsequent redistribution throughout the body and incorporation into the cells and nuclei of other tissues and organs. This could include distantly located skeletal muscles that account for the distant transfection effect. To prove this, experiments with direct i.v. application of dystrophin-gene loaded microspheres are now in progress. Meanwhile, tissue distribution of injected transgenes renders more support for this phenomenon.

Detailed analysis of cell and tissue distribution of MF-2 complexes with biotin-labeled pHSADy DNA resulted in two unusual findings. First, FITC-marked microspheres were abundant in all tissue samples studied including muscles, heart, diaphragm and lungs. They were also rather numerous in the brains, placenta and even in the tissues of the intra-uterine fetus of mdx mice injected on the 18th day of gestation. The detailed mechanism for such a high invasiveness of these microspheres remains unknown and needs further analysis. Second, the majority of muscle tubes as well as cells from other tissues incorporated MF-2 particles in their nuclei. As shown for cationic liposomes successful transfection of mammalian cells requires efficient transportation of exogenous DNA in the nuclei through endosomal and lysosomal machinery.20 Large sizes of MF-2 particles (about 2 μm) are hardly suitable for the cellular processing of this vehicle by standard intranuclear trafficking. None the less direct visualization of these microspheres proves that their size does not interfere with their efficient incorporation into the muscle and other cell nuclei. Whether the internalization of such large particles with a substantial genetic load might interfere with normal nuclear functions remains obscure. Meanwhile slight toxic effects of multiple or high doses of dystrophin gene constructs carried by MF-2 vehicle in the muscle nuclei might be attributed to impairment of normal nucleus function. Direct ultramicroscopic studies are needed to clarify the mechanisms of MF2/plasmid DNA particles transportation and to demonstrate their internalization in the nuclei.

Efficient transfection of muscle nuclei with dystrophin cDNA in combination with a clearcut distant transfection effect and high penetration of biological membranes make MF-2 microspheres a promising vehicle for systemic delivery of dystrophin gene into affected myofibers and thus for gene therapy of DMD.

Materials and methods

DNA constructs

Both full-length (12 kb) and Becker type minigene (6.3 kb) cDNAs were used throughout the study. Plasmid containing a full-length human dystrophin cDNA pHSADy (pBSSK+ vector, human alpha-skeletal actin promoter) was obtained from Professor G Dickson (Royal Holloway University, London, UK),21 recombinant plasmid containing Becker-type human dystrophin cDNA was pSG5dys (pSG5 vector, SV40 promoter) (courtesy of Dr F Mavilio, TIGET, Milan, Italy). Plasmid DNA of all the human dystrophin expression constructs as well as of a plasmid containing bacterial β-galactosidase driven by a human cytomegalovirus immediate–early promoter (pCMVLacZ) was isolated by standard lysozyme/alkaline lysis and purified by cesium chloride centrifugation. After ethanol precipitation plasmid DNA was resuspended in sterile TE buffer, divided into aliquots and stored at −20°C until required.

Preparation of DNA and synthetic polymer complexes

Microspheres MF-2 loaded with DNA constructs were prepared by the solvent–evaporation method. To begin, two water-in-oil (W/O) emulsions were prepared ex tempore. The first one was prepared from an aqueous solution of the plasmid DNA. The second W/O emulsion contained aqueous solution of synthetic polymer (an original polymer was provided by Biosystem XXI, biosystem@ctinet.ru). Both W/O emulsions were thoroughly mixed at the desired ratio, stirred and evaporated. Polymer microspheres were collected by centrifugation and washed three times with organic solvent. Washing and subsequent drying procedures were performed under sterile conditions. The applied technology resulted in globular DNA–polymeric complexes about 2 microns in size.

In vivo transfection of mdx mouse muscles

Experimental and matched control mdx mice (4–6 weeks old) (kindly provided by Professor T Partridge, Hammersmith Hospital, London, UK) were used throughout the studies. C57BL/10J+/+ mice of relevant age were used as a dystrophin-positive control. Quadriceps femoris muscles and glutaeus superficialis muscles of the right hindlimb were opened surgically under Nembutal anesthesia (P3393; Sigma, St Louis, MO, USA) and repeatedly injected with MF-2 particles loaded with plasmid DNA constructs (25 or 50 μg of DNA) in 100 μl of sterile physiological saline. Positive and negative matched control mice were injected with 100 μl of pure saline or with saline plus non-loaded MF-2 particles. The skin was stitched and the animals were returned to the animal house. All animals were killed by cervical dislocation at different intervals after injections (1 day–2 months). All tissue samples were snap-frozen in liquid nitrogen and processed for subsequent immunocytochemical, molecular or FISH analysis.

Dystrophin and β-galactosidase expression analysis

For dystrophin expression assessment serial-cross cryosections (5–7 μm) from injected and non-injected muscles were obtained, collected on Superfrost slides (Fisher Scientific, Pittsburgh, PA, USA) and kept at −70°C. For immunohistochemical detection slides were first incubated 5 min in 1 × PBS. Nonspecific reactivity was blocked by 30 min incubation with 1 × PBS/3% BSA. Sections were then processed for 1 h at room temperature with the affinity purified rabbit polyclonal antibodies (P6) raised against the C-terminus of human dystrophin (courtesy of Dr Peter Strong and Dr Caroline Sewry, Hammersmith Hospital, London, UK) in 1% BSA in 1 × PBS (final dilution 1:600). FITC or horseradish peroxidase-conjugated sheep anti-rabbit IgG (DAKO, Carpinteria, CA, USA; Sigma) were used as a secondary antibody (dilution 1:200 in 1% BSA in 1 × PBS). Incubation time was 40 min at room temperature. Following the histochemical reaction, the sections were dehydrated through a series of ethanols, mounted with Antifade solution (90% glycerol, 0.02 M Tris-HCl pH 8.0, 2.3% DABCO, 0.5 μg/ml propidium iodide) and used for microscopy. Direct light microscopy and epifluorescence illumination with Zeiss Axiophot photomicroscope (Oberkochen, Germany) were used to visualize dystrophin-positive myofibers (DPM) after horseradish peroxidase or FITC staining, respectively. After fluorescent analysis Antifade solution was washed out by 1 × PBS, sections counterstained with hematoxylin and eosin and used for precise counting of total number of myofibers per section area (×400) as well as for estimation of fibers harboring centrally located nuclei (×400). DPM were counted on randomly chosen cross-sections of the midportion of every muscle, all fibers of these cross- sections (mean of 2050 fibers) were included in the analysis. The proportion of DPM was quantified as a percentage of DPM in the total number of fibers. Cross-section areas of dystrophin-positive and dystrophin-negative myofibers were calculated from the measurements of long and short diameters of each myofiber in horseradish peroxidase stained preparations.

X-gal staining

Two weeks after the i.m. injections of pCMVLacZ formulated into microspheres, the mice were killed by cervical dislocation. Samples of injected and non-injected muscles as well as heart, lung and other organs were harvested, rinsed in cooled PBS, fixed in 4% paraformaldehyde in 1 × PBS (3 h at room temperature), washed out in PBS for 1 h and incubated in a staining solution (1 mg/ml 5-bromo-4-chloro-3-indolyl-β-D-galactoside (Promega, Madison, WI, USA), 2 mM MgCl2, 5 mM K3Fe(CN)6, 5 mM K4Fe(CN)6 × 3 H2O in 1 × PBS. After overnight incubation at 37°C, the samples were washed three times × 10 min in 1 × PBS and stored in 1 × PBS at +4°C. Wholemount samples were analyzed and photodocumented under dissecting microscope. After staining some of these samples were used for preparation of 10-μm thick paraffin sections.

Detection of human dystrophin cDNA constructs and RNA transcripts by PCR and RT-PCR technique

Genomic DNA samples from the muscles, heart and lungs as well as from the other internal organs including placentae and 18 day embryos were obtained after proteinase K digestion, phenol-chloroform extraction and ethanol precipitation. Upstream primer at the 5′ end of exon 52 (52F-5′-AAT GCA GGA TTT GGA ACA GAG GCG TCC–3′) and downstream primer at the 3′ end of exon 53 (53R-5′-CTT GGT TTC TGT GAT TTT CTT TTG GAT TG–3′) of the human dystrophin gene were selected. Amplification conditions were: 94°C, 10 min; (94°C, 48 s; 58°C, 48 s; 72°C, 3 min) × 30 cycles; 72°C, 5 min. The resulting amplification product subjected to electrophoresis in a 5% polyacrylamide gel could be easily identified as a 325 bp DNA fragment.

For RT-PCR analysis, total RNA was isolated by means of Promega RNA extraction kit (Z5110). First strand cDNA from RNA template was generated using Pharmacia Ready-To-Go You-Prime First-Strand Beads kit (27–9264–01) (Pharmacia, Uppsala, Sweden) with a primer DMD8d (5′-CTCCTGGTAGAGTTTCTCTAG-3′).22 Subsequent PCR reaction was carried out according to a standard procedure with the pair of oligonucleotide primers 52F and 53R as indicated before (see PCR).

DNA–DNA in situ hybridization

FISH technique was used for analysis of distribution and intracellular localization of microspheres loaded with plasmid DNA in mouse tissues. Standard (5–7 μm) cryosections of the skeletal muscles and other mouse samples (heart, diaphragm, brain, lungs) were digested with RNase A (200 μg/ml for 2–3 h), washed three times in fresh 2 × SSC (pH 7.0), rinsed 5 min in 1 × PBS and incubated 5–20 min in pepsin solution (10 mg/100 ml 0.01 M HCl). After 5 min rinsing in 1 × PBS the sections were post-fixed in 3% paraformaldehyde, rinsed in 1 × PBS, dehydrated by graded ethanols and air-dried. DNA denaturation of section samples was performed in 70% formamide, 2 × SSC (pH, 7.0) for 3 min at 73°C. Slides were quenched immediately in cold 70% ethanol and dehydrated in an ethanol series. DNA probes were labeled as indicated using standard nick translation protocol and used for in situ hybridization according to the standard method. The hybridization reaction was incubated at 37°C overnight. After post-hybridization washes in three changes of 50% formamide, 2 × SSC and three changes of 2 × SSC, nonspecific binding was blocked with 3% BSA, 4 × SSC and hybridization signal was detected by one layer of avidin-FITC conjugate. After the last washing in three changes of 4 × SSC the slides were dehydrated in an ethanol series, air dried and then mounted in Antifade solution. After in situ hybridization the cryosection samples were analyzed by fluorescence microscopy on Zeiss Axiophot Photomicroscope. The proportion of FITC marked MF2/plasmid DNA complexes in cytoplasm and cell nuclei was recorded.

References

  1. 1

    Emery AEH . Duchenne Muscular Dystrophy, 2nd edn Oxford Monograph on Medical Genetics 24: Oxford University Press: Oxford 1993 p 157

  2. 2

    Partridge TA . Models of dystrophinopathy, pathological mechanisms and assessment of therapies. In: Brown SC, Lucy JA (eds) Dystrophin: Gene, Protein and Cell Biology Cambridge University Press: Cambridge 1997 pp 310–331

  3. 3

    Engel AG, Yamamoto M, Fischbeck KH . Dystrophinopathies. In: Engel AG, Franzini-Armstrong C (eds) Myology McGraw-Hill: New York 1994 pp 1133–1187

  4. 4

    Clemens PR et al. In vivo muscle gene transfer of full-length dystrophin with an adenoviral vector that lacks all viral genes Gene Therapy 1996 3: 965–972

  5. 5

    Lochmuller H et al. Transient immunosuppression by FK506 permits a sustained high-level dystrophin expression after adenovirus-mediated dystrophin minigene transfer to skeletal muscles of adult dystrophic (mdx) mice Gene Therapy 1996 3: 706–716

  6. 6

    Fassati A, Murphy S, Dickson G . Gene therapy of Duchenne muscular dystrophy Adv Genet 1997 35: 117–153

  7. 7

    Budker V et al. The efficient expression of intravascularly delivered DNA in rat muscles Gene Therapy 1998 5: 272–276

  8. 8

    Dodds E et al. Lipofection of cultured mouse muscle cells: a direct comparison of Lipofectamine and DOSPER Gene Therapy 1998 5: 542–551

  9. 9

    Vitiello L et al. Condensation of plasmid DNA with polylysine improves liposome-mediated gene transfer into established and primary muscle cells Gene Therapy 1996 3: 396–404

  10. 10

    Baranov V et al. Human dystrophin gene expression in mdx muscles after in vivo ballistic transfection, application of synthetic oligopeptide complexes and cationic liposomes In: NATO ASI, Subseries M 1997 105: 219–223

  11. 11

    Zelenin V, Tarasenko O, Kolesnikov VA et al. Bacterial β-galactosidase and human dystrophin genes are expressed in mouse skeletal muscle fibers after ballistic transfection FEBS Lett 1997 414: 319–322

  12. 12

    Mathiowitz E, Jacob JS, Jong YS et al. Biologically erodable microspheres as potential oral drug delivery system Nature 1997 386: 410–414

  13. 13

    Walsh S et al. Simultaneous delivery of drugs and genes by gelatin nanospheres Pediatric Pulmonology 1997 (Suppl. 14): 255–257

  14. 14

    Bockhold KJ, Rosenblatt JD, Partridge TA . Aging normal and dystrophic muscle: analysis of myogenicity in cultures of living single fibers Muscle Nerve 1998 21: 173–183

  15. 15

    Feero WG et al. Selection and use of ligands for receptor-mediated gene delivery to myogenic cells Gene Therapy 1997 4: 664–674

  16. 16

    Cox GA et al. Overexpression of dystrophin in transgenic mdx mice eliminated dystrophic symptoms without toxicity Nature 1993 364: 725–729

  17. 17

    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

  18. 18

    Skarli M et al. Myosin regulatory elements as vector for gene transfer by intramuscular injection Gene Therapy 1998 5: 514–520

  19. 19

    Liu F, Qi H, Huang L, Liu D . Factors controlling the efficiency of cationic lipid-mediated transfection in vivo via intravenous administration Gene Therapy 1997 4: 517–523

  20. 20

    Coonrod A, Li F-Q, Horwitz M . On the mechanism of DNA transfection: efficient gene transfer without viruses Gene Therapy 1997 4: 1313–1321

  21. 21

    Wells D, Wells K, Asante EA et al. Expression of human full length and mini-dystrophin in transgenic mdx mice: implication for gene therapy of Duchenne muscular dystrophy Hum Mol Genet 1995 3: 1245–1250

  22. 22

    Roberts RG, Barby TFM, Manner E et al. Direct detection of dystrophin gene rearrangements by analysis of dystrophin mRNA in peripheral blood lymphocytes Am J Hum Genet 1991 49: 298–310

Download references

Acknowledgements

The authors are grateful to ‘Biosystem XXI, St Petersburg, Russia for the original polymeric vehicle, to Dr Caroline Sewry and Dr Peter Strong (Hammersmith Hospital, London, UK) for the generous gift of P6 antibodies, Professor G Dickson (Royal Holloway University, London, UK) and Dr F Mavilio (TIGET, Milano, Italy) for dystrophin cDNA constructs. Personal gratitude to Professor T Partridge (Hammersmith Hospital, London, UK) for the stock of mdx mice, as well as permanent interest in this work and helpful constructive comments during manuscript preparation. This work was partly supported by the Russian State Program ‘Human Genome’ (grant 48/95).

Author information

Correspondence to V Baranov.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Baranov, A., Glazkov, P., Kiselev, A. et al. Local and distant transfection of mdx muscle fibers with dystrophin and LacZ genes delivered in vivo by synthetic microspheres. Gene Ther 6, 1406–1414 (1999) doi:10.1038/sj.gt.3300954

Download citation

Keywords

  • Duchenne muscular dystrophy
  • dystrophin
  • gene transfer
  • synthetic microspheres
  • distant transfection

Further reading