Gene Therapy (1997) 4, 401–408 1997 Stockton Press All rights reserved 0969-7128/97 $12.00

Mini-dystrophin gene transfer in mdx4cv diaphragmmuscle fibers increases sarcolemmal stability

A Decrouy1, J-M Renaud1, HL Davis1,2,3, JA Lunde1, G Dickson4 and BJ Jasmin1

1Department of Physiology, Faculty of Medicine and 2Program in Physiotherapy, Faculty of Health Sciences, University of Ottawa;3Loeb Medical Research Institute, Ottawa Civic Hospital, Ottawa, Ontario, Canada; and 4Division of Biochemistry, School ofBiological Sciences, Royal Holloway, University of London, Egham, Surrey, UK

To date, all dystrophin gene transfer studies have been lemma. Most importantly, this level of dystrophinperformed on mdx hindlimb skeletal muscles which in com- expression was sufficient to protect all fibers present withinparison to the severe deficits seen in muscles from patients these diaphragm muscle bundles from the damagingafflicted with Duchenne muscular dystrophy (DMD), exhibit effects of repetitive lengthening contractions. In addition,only modest morphological and functional changes. Since dystrophin expression partially restored the ability of trans-the mdx diaphragm muscle presents the same pathophysi- duced mdx4cv muscle bundles to generate isometric tetanicological alterations characteristic of DMD muscles, we tension following lengthening contractions. These resultstherefore injected recombinant plasmid DNA encoding the show that mini-dystrophin expression leads to rapid anddystrophin mini-gene (pRSVdy-B) into diaphragm muscles significant functional improvements in diaphragm musclesof 10-week-old mdx4cv mice and examined the physiologi- of mdx4cv mice. Although these data provide encouragingcal consequences of dystrophin expression in a muscle results for future therapeutic strategies aimed at curingthat has undergone a phase of massive degeneration and DMD, additional work will none the less be necessary toregeneration. Immunoperoxidase and immunofluoresc- determine the full impact of dystrophin gene replacement.ence experiments revealed that 1 and 3 weeks following In this context, it is clear from the data presented here thatgene transfer, approximately 17% of the fibers in a bundle the diaphragm muscle of the mdx mouse is an invaluableof diaphragm muscle expressed dystrophin at the sarco- model system to address this critical issue.

Keywords: gene therapy; Duchenne muscular dystrophy; sarcolemma

organization has led to the suggestion that dystrophinIntroductionplays an important mechanical role in stabilizing and pre-

Duchenne muscular dystrophy (DMD) is the most preva- serving the integrity of the sarcolemma. In support oflent of all neuromuscular diseases affecting one in 3500 this, it has been demonstrated that the absence of dystro-male births. This disease is characterized by repeated phin results in an excessively fragile sarcolemma10,11

cycles of muscle fiber degeneration and regeneration highly susceptible to rupture in response to osmoticwith eventual failure to regenerate. Concomitantly, there shock12 and lengthening contractions.13–15 More recentis an inexorable loss of muscle mass and function, and a studies have revealed that dystrophin may also partici-replacement of myofibers by adipose and connective pate indirectly in various signal transduction pathwaystissues. The disease progresses rapidly, such that DMD since members of the dystrophin complex associate withpatients are usually wheelchair-bound by adolescence Grb216 as well as with nitric oxide synthase (NOS).17–19

and die in their second or third decade of life, most often At present, there is no effective treatment or cure foras a result of respiratory failure. The genetic defect DMD. However, since it is caused by a single recessiveunderlying DMD is located on the short arm of the X gene defect, this disease is a candidate for gene therapy.chromosome and prevents the production of dystrophin, In recent years, the feasibility of transferring full-lengtha large cytoskeletal protein of the spectrin superfamily.1,2

or mini-dystrophin gene constructs into hindlimb skeletalPrevious studies have shown that in skeletal muscle fib- muscles of mdx mice has been examined. The mdx mouseers, dystrophin is located at the cytoplasmic face of the is genetically homologous to DMD since mutation of thesarcolemma3,4 where it associates with F-actin.5,6 Bio- dystrophin gene also leads to an absence of dystrophinchemical experiments have subsequently demonstrated in muscle fibers of these mice.20,21 Using various experi-that, in fact, dystrophin links the intracellular cytoskele- mental approaches including viral vectors22–24 and intra-ton network to the extracellular matrix via a complex muscular injection of plasmid DNA,25,26 several studiesof dystrophin-associated proteins.7–9 Such subcellular have shown expression of full-length or truncated dystro-

phin27 at the sarcolemma of muscle fibers from mdx mice,yet only one report has so far examined the physiologicalCorrespondence: BJ Jasmin, Department of Physiology, Faculty of Medi-consequences of dystrophin gene replacement.28 More-cine, University of Ottawa, 451 Smyth Road, Ottawa, Ontario, Canadaover, all gene transfer studies performed to date haveK1H 8M5

Received 17 October 1996; accepted 2 January 1997 used mdx hind limb skeletal muscles, which in

Direct gene transfer into mdx diaphragm muscleA Decrouy et al

402 comparison to the severe deficits seen in DMD muscles, fibers contained within the bundles. Indeed, 1 week afterintramuscular injection of the mini-dystrophin gene con-exhibit only modest morphological and functional

changes. A more appropriate model is the diaphragm struct, approximately 4% of the fibers present in injectedbundles of mdx4cv diaphragm muscles were strongly dec-muscle which, in the mdx mouse, presents the same

pathophysiological alterations characteristic of DMD orated by the anti-dystrophin antibody. In addition tothese strongly labelled fibers, numerous fibers were alsomuscles including for instance, an increased amount of

connective tissue as well as a significant loss of func- faintly stained (see example in Figure 2c). In thesebundles, the total number of dystrophin-positive fibers,tion.29–31 Therefore, it would be most useful to perform

dystrophin gene transfer in diaphragm muscles of adult ie strongly plus faintly labelled, was approximately 17%(Figure 3). This pattern of dystrophin labelling as well asmdx mice to determine the functional impact of dystro-

phin expression at the sarcolemma of this respiratory the number of dystrophin-positive fibers following injec-tion with pRSVdy-B was similar at 1 and 3 weeks aftermuscle. This is a particularly important issue if gene ther-

apy is to be envisaged as a putative therapeutic approach injection.for DMD.

The objectives of the present study were therefore to Assessment of sarcolemmal stabilityTo determine the impact of dystrophin gene transfer ondetermine the relative efficiency of dystrophin gene

replacement in diaphragm muscles from mdx mice and sarcolemmal stability of muscle fibers from mdx4cv mice,we subjected dissected bundles of diaphragm muscles toexamine the physiological consequences of dystrophin

expression on the stability and integrity of the sarco- a series of lengthening contractions. Lengthening contrac-tions are known to induce significant damage to musclelemma during a series of lengthening contractions. To

this end, we used a model of intramuscular DNA injec- fibers since they disrupt the integrity of the sarcolemmaas a result of sudden muscle lengthening during tetaniction recently developed32 to introduce mini-dystrophin

gene constructs into diaphragm muscle fibers. We chose contractions.15,36 Owing to this disruption, the ability ofmuscle fibers to generate tension decreases dramaticallyto study 10-week-old mdx mice, since at this developmen-

tal stage diaphragm muscles have already undergone a and, concomitantly, the sarcolemma becomes highlypermeable to low molecular mass molecules such asphase of massive degeneration and regeneration as evi-

denced by approximately 80% of the fibers exhibiting procion orange.Using this experimental approach, we observed thatcentral nucleation.33,34 A preliminary account of this work

has previously appeared in abstract form.35 following 12 lengthening contractions bundles of dia-phragm muscles from control C57Bl/6 mice showed ahigh resistance to lengthening contractions since only a

Results few fibers (approximately 1%) were stained with procionorange (Figures 4 and 5). In contrast, bundles of dia-phragm muscles from mdx4cv mice showed significantlyDystrophin expression in diaphragm muscle following

direct gene transfer more procion orange-positive fibers, thereby reflectingthe greater sensitivity of these fibers to lengthening con-We examined the pattern of dystrophin expression in dis-

sected bundles (used to test the impact of repetitive tractions (P , 0.05; Figures 4 and 5). One week followinginjection of pRSVdy-B, bundles of mdx4cv diaphragmlengthening contractions; see below) of diaphragm

muscle fibers removed from mdx4cv mice following intra- muscles were clearly more resistant to lengthening con-tractions since the number of fibers which had incorpor-muscular injection of recombinant plasmid DNA enco-

ding the 6.3 kb Becker-like human dystrophin mini-gene ated the dye decreased dramatically (P , 0.05; Figures 4and 5). In fact, the number of procion orange-positiveunder the control of the Rous sarcoma virus promoter

(pRSVdy-B).25 Expression of dystrophin was examined 1 fibers in these diaphragm bundles was not statisticallydifferent (P . 0.05) from that observed in bundles fromand 3 weeks following intramuscular injection of the

mini-dystrophin gene construct by both immunoperoxid- control C57Bl/6 mice (Figure 5). In separate experiments,we verified that this protective effect of dystrophin genease and immunofluorescence staining. In diaphragm

muscle of control C57Bl/6 mice, both approaches replacement was not related to the surgical or injectionprocedures per se by injecting a nonexpressing controlshowed, as expected, that dystrophin was expressed at

the cytoplasmic face of the sarcolemma of all muscle fib- vector (pBluescript phagemid) into diaphragm musclesof mdx4cv mice. In these diaphragm muscles, the numberers (Figures 1a and 2a). Parallel experiments demon-

strated that bundles of diaphragm muscles from mdx4cv of procion orange-positive fibers after 12 lengtheningcontractions was similar to those observed in diaphragmmice contained no dystrophin-positive fibers (Figures 1b

and 2b) except for a few revertants which accounted for muscles from noninjected mdx4cv mice (data not shown).In contrast to these findings which indicate that dystro-less than 1% of the total number of fibers present in these

bundles. One week after injection of pRSVdy-B in dia- phin gene replacement clearly led to enhanced stabilityof the sarcolemma (see Figures 4 and 5), intramuscularphragm muscles from mdx4cv mice, several fibers

expressed significant levels of dystrophin (Figure 1c). The injection of pRSVdy-B into mdx4cv diaphragm muscleonly partially restored the ability of muscle bundles todistribution of dystrophin in these muscle bundles was

heterogeneous as dystrophin-positive fibers appeared in generate tension following repetitive lengthening con-tractions. The maximal tension generated by bundles ofsmall clusters mostly confined near the injection site.

The number of dystrophin-positive fibers present in fibers from the diaphragm of control mice decreased byapproximately 17% ± 7 (X ± s.d.) after 12 lengthening con-bundles of injected and noninjected diaphragm muscles

from mdx4cv mice was quantified by immunofluoresc- tractions whereas this protocol had a more severe effect(P , 0.05) on the ability of muscle bundles from mdx4cvence. In these experiments, we observed that the level of

dystrophin staining was highly variable between positive mice. Indeed, tetanic tension generated by these dia-

Direct gene transfer into mdx diaphragm muscleA Decrouy et al

403

Figure 1 Immunoperoxidase detection of dystrophin in cross-sections of diaphragm muscles from C57Bl/6 (a) and mdx4cv (b) mice as well as frommdx4cv mice 1 week following intramuscular injection of pRSVdy-B (c). Note that the distribution of dystrophin in transduced muscle bundles washeterogeneous as dystrophin-positive fibers appeared in small clusters mostly confined near the injection site. Bar = 65 mm.

phragm muscle bundles was reduced by 38% ± 12 muscles.29–31 Our results show that after injection ofpRSVdy-B, expression of dystrophin in approximately(X ± s.d.). In injected bundles of mdx4cv diaphragm

muscles, maximal tetanic tension was decreased by 17% of the fibers significantly reduced the number of pro-cion orange-positive fibers in muscle bundles following24% ± 10 (X ± s.d.); a value statistically not different (P .

0.05) from that obtained with bundles of diaphragm a series of lengthening contractions. Remarkably, thefunctional sarcolemmal improvement induced by themuscles from noninjected mdx4cv mice. Together, these

results therefore suggest that the increase in the stability replacement of dystrophin was observed as early as 1week after injection of pRSVdy-B. Although dystrophinof the sarcolemma following dystrophin gene transfer

(see Figure 5) cannot entirely prevent the marked expression increased the stability of the plasma mem-brane, it failed however to restore fully the ability of mus-decrease in tetanic force that occurs during a series of

lengthening contractions in diaphragm muscle from cle bundles to generate tetanic tension following a seriesof lengthening contractions.mdx4cv mice.

Our immunofluorescence and immunoperoxidaseexperiments revealed that approximately 17% of the fib-Discussion ers expressed significant levels of dystrophin followingintramuscular injection of pRSVdy-B into the diaphragmIn the present study, we examined the physiologicalmuscle. This result indicates that the efficiency of trans-consequences of dystrophin gene replacement in skeletalduction with plasmid DNA in the diaphragm muscle ismuscle fibers of mdx4cv mice. We carried out gene trans-relatively high compared with other muscles. Earlierfer on the diaphragm muscle since previous studies havestudies have shown, for example, that direct gene trans-established that this respiratory muscle displays patho-

physiological alterations characteristic of DMD fer into the quadriceps muscle group resulted in

Direct gene transfer into mdx diaphragm muscleA Decrouy et al

404

Figure 2 Immunofluorescence localization of dystrophin in cross-sections of diaphragm muscles from C57Bl/6 (a) and mdx4cv (b) mice as well as frommdx4cv mice 1 week following intramuscular injection of pRSVdy-B (c). Note that in transduced muscle bundles some fibers displayed strong immuno-reactivity (large fiber to the right) whereas others were faintly decorated by the anti-dystrophin antibody (arrows). Bar = 35 mm.

expression of dystrophin in only 1% of the fibers.25,26 The that dystrophin expression in approximately 17% of thefibers protected essentially all fibers present within thehigher efficiency of gene transfer into diaphragm muscle

fibers may result from a greater diffusion of the injected diaphragm muscle bundle from the damaging effects ofrepetitive lengthening contractions is of paramountsolution due to a thinner endomysium and to the con-

tinuous rhythmical activity of the muscle during and importance. Indeed, the number of procion orange-posi-tive fibers observed in transduced diaphragm muscleafter the surgical procedure. It is interesting to note that

our results obtained using intramuscular injection of bundles from mdx4cv was similar to that seen in musclebundles from control mice. This indicates that the sarco-pRSVdy-B into adult diaphragm muscles compare favor-

ably with those obtained by others with a recombinant lemma recovered its structural integrity as a result of dys-trophin expression. It is particularly important to note inadenovirus used to infect hindlimb muscles of neonatal

mdx mice.23,24,37 this case, that the protective effect was not strictly limitedto fibers expressing dystrophin. These results suggestUntil now, studies performed with transgenic mdx

mice expressing various dystrophin constructs have indi- therefore that the efficiency of transduction may not needto be at, or close to, 100% to obtain significant functionalcated that their skeletal muscle fibers present relatively

normal morphological, biochemical and functional and protective benefits. Possibly, ectopic dystrophinexpression restored not only the interactions between thephenotypes.38–42 However, the applicability of this

approach in a clinical setting remains, for obvious extracellular matrix and the internal cytoskeleton oftransduced muscle fibers, but it also led to an increase inreasons, rather limited. As such, our experiments aimed

at determining the functional consequences of dystrophin the architectural cohesion of the entire muscle bundle.For example, dystrophin-positive fibers could act as rein-gene replacement in adult mice bring novel and essential

information concerning the usefulness of gene therapy as forcing structures absorbing most of the mechanicalstress imposed on the diaphragm bundle during length-a therapeutic strategy for DMD. In comparison to normal

muscles, the sarcolemma of mdx mouse muscle is more ening contractions, thus protecting dystrophin-negativefibers. Alternatively, since we and others37 have observedfragile and therefore, more sensitive to repetitive length-

ening contractions.13–15 In this context, our observation variable intensities in dystrophin labelling following gene

Direct gene transfer into mdx diaphragm muscleA Decrouy et al

405

Figure 3 Percentage of dystrophin-positive fibers in diaphragm musclebundles following intramuscular injection of pRSVdy-B. Values aremeans + s.e.m. for control diaphragm from C57Bl/6 mice (CTL), untreateddiaphragm from mdx4cv mice (MDX4CV), and injected diaphragm frommdx4cv mice (MDX4CV (pRSVdy-B)).

transfer, it is also possible that a large number of fibersexpressing low levels of dystrophin were not detected bythe immunofluorescent staining procedure.

Despite these encouraging results, we still observedafter intramuscular injection of the mini-dystrophin geneconstruct into diaphragm muscles of mdx4cv, a substantialdecrease in isometric tetanic force following a series oflengthening contractions. Although a clear tendency forfunctional recovery was observed, the efficiency of dys-trophin gene replacement appeared none the less insuf-ficient to normalize the capacity of transduced musclebundles to sustain tension following lengthening contrac-tions. This indicates that a percentage greater than 17%of transduced muscle fibers may be necessary for fullrecovery of contractile properties. Taken together, ourresults thus highlight a dissociation between stability ofthe sarcolemma and ability to generate tetanic tensionduring repetitive lengthening contractions. This, in fact,may not be all that surprising given that the integrity ofthe sarcolemma is only one of the key factors influencing Figure 4 Assessment of sarcolemmal stability following repetitive length-the capacity of muscle fibers to generate tetanic tension. ening contractions. Incorporation of procion orange into fibers was moni-

tored in diaphragm muscles from control C57Bl/6 (a) and mdx4cv (b) miceIn a recent study, it was shown that injection of anas well as from mdx4cv mice following intramuscular injection of pRSVdy-adenovirus vector carrying the mini-dystrophin gene intoB (c). Note the greater number of procion orange-positive fibers in mdx4cv

hindlimb muscles of neonatal mdx mice, which have notdiaphragm muscles (arrows in b) and a reduced number following dystro-yet entered the degenerative process, conferred phin gene replacement (c). Arrows in c point to dystrophin-positive fibers

important functional and structural protection against identified by immunoperoxidase staining. Bar = 70 mm.mechanical stress.28 Remarkably, despite significantmethodological differences between the latter study andours, the results are in good agreement and in fact, are eterious process caused by the absence of dystrophin

even following several cycles of degeneration–regener-complementary. Indeed, by injecting plasmid DNA enco-ding the mini-dystrophin gene into the diaphragm ation (this study).

In summary, our data showing that dystrophin genemuscle of adult mdx mice, we reversed some of thephenotypic manifestations of the disease. Taken together, replacement in diaphragm muscles from mdx4cv mice

induces expression of dystrophin in a relatively high per-these results thus indicate that injection of a vector enco-ding the mini-dystrophin gene leads to functional centage of fibers and increases the stability of the sarco-

lemma, provide encouraging results for future thera-improvements of the muscle fibers by either preventingthe physiological impairment28 or by reversing the del- peutic strategies aimed at curing DMD. Additional work

Direct gene transfer into mdx diaphragm muscleA Decrouy et al

406 were housed in groups of two or three, maintained on a12 h:12 h light:dark cycle and provided with food andwater ad libitum. Care of the animals was in accordancewith the guidelines established by the Canadian Councilfor Animal Care. All surgical procedures were approvedby the University of Ottawa Animal Care Committee.

Injection of plasmid DNA into diaphragm muscles wasperformed according to the procedure described byDavis and Jasmin.32 Briefly, with the animals underanaesthesia (sodium pentobarbital; 8 mg/g of body mass,IP), the inferior surface of the right hemidiaphragm wasexposed via a transverse incision made below the lateralcostal margin. While holding the costal margin with for-ceps, 50 ml of DNA solution (4 mg/ml) was directlyinjected at several sites between the inferior surface ofthe muscle and the overlying epimysium. This procedurewas carried out using a 0.3 cc tuberculin syringe undera dissecting microscope. Following DNA injection, theabdominal musculature was sutured and the woundclosed with surgical clips.

Lengthening contraction protocolFigure 5 Percentage of procion orange-stained fibers in diaphragm muscleSeven days following direct gene transfer, physiologicalbundles induced by a series of 12 lengthening contractions. Values areexperiments were performed on injected diaphragmmeans + s.e.m. for control diaphragm from C57Bl/6 mice (CTL), untreated

diaphragm from mdx4cv mice (MDX4CV) and injected diaphragm from muscles of mdx4cv mice as well as on noninjected mdx4cv

mdx4cv mice (MDX4CV (pRSVdy-B)). and control mice. Animals were first anaesthetized withsodium pentobarbital and the diaphragm muscle sub-sequently removed in its entirety. A longitudinal musclewill none the less be necessary to determine, for example, bundle containing approximately 400 fibers was carefullythe full impact of dystrophin gene replacement on vari- dissected from the injected area leaving the ribs as wellous functional and biochemical parameters and the long- as the central tendon attached to the extremities of theterm effects of dystrophin expression. On the basis of the muscle bundle. In preliminary experiments, we determ-findings presented here, it is clear that the diaphragm ined that this dissection procedure ensured integrity andmuscle of the mdx mouse is an invaluable model system optimal contractile performance of the muscle fibers. Fol-to address these critical issues since it constitutes the only lowing dissection, the muscle bundle was rapidly trans-muscle that shows the same pathophysiological alter- ferred to the test chamber where it was continuouslyations typical of DMD muscles. superfused on both sides with Krebs-Ringer solution(118.5 mm NaCl, 4.7 mm KCl; 2.4 mm CaCl2, 3.1 mm

MgCl2, 25 mm Na2HCO3, 1 mm NaH2PO4, 5 mm glucoseMaterials and methodsand 95% O2–5% CO2; pH 7.4) maintained at 37°C. Thetendon end of the muscle bundle was attached to a forceDystrophin expression vector

The 6.3 kb mini-gene used in our experiments corre- transducer while the other extremity was connected to amotor (Cambridge Technology, Boston, MA, USA, modelsponds to the cDNA identified by England et al27 in a

patient suffering a mild Becker muscular dystrophy 350) to allow rapid lengthening of the bundle. Musclelength (Le) was adjusted in order to obtain maximalphenotype despite a very large deletion in the dystrophin

gene where 5106 bp of coding sequence in the central tetanic force during 700 ms stimulating trains of 0.3ms square pulses at a frequency of 120 Hz and suprama-domain are missing. Recombinant plasmid DNA enco-

ding the 6.3 kb Becker-like human dystrophin mini-gene xi-mal voltages (6–8 V). Prior to lengthening contractions,under the control of the Rous sarcoma virus promoter

(pRSVdy-B)25 was prepared by anion exchange chroma- muscle bundles were left to equilibrate for 30 min inKrebs-Ringer solution. Following this, 12 lengtheningtography using the Qiagen mega-prep procedure

(Chatsworth, CA, USA). The DNA was recovered by iso- contractions consisting of 10% lengthening at a velocityof 0.5 Le/s were applied at 15 s intervals during the lastpropanol precipitation, redissolved in sterile 0.1 m PBS

(pH 7.4), aliquoted and stored at −20°C until required 200 ms of each tetanic contraction. In all these experi-ments, 0.2% (w/v) procion orange (Sigma, St Louis, MO,for injection.USA) was added to the Krebs-Ringer solution at thebeginning of the equilibration period to determine theAnimals and surgical procedures

All experimental procedures involving injection of plas- extent of cell membrane damage elicited by the lengthen-ing contraction procedure.14 This dye penetrates into themid DNA were performed with mdx4cv mice since

mutation of the dystrophin gene in these mice leads to sarcoplasm of cells whose membranes have been dam-aged by the lengthening contraction protocol and mayapproximately 10-fold less revertant, dystrophin-positive

fibers in comparison to mdx mice.43 Ten-week-old mice easily be observed in cryostat sections under the lightmicroscope. Following the last contraction, diaphragmwere used for these studies and age-matched controls

consisted of mice from the C57Bl/6ScSn strain. Animals muscle bundles were frozen in melting isopentane preco-

Direct gene transfer into mdx diaphragm muscleA Decrouy et al

407oled with liquid nitrogen and stored at −80°C for Research Council of Canada, the Muscular DystrophyGroup of Great Britain and the UK Medical Researchfurther analysis.Council. HLD is a Career Scientist of the Ontario Ministryof Health. BJJ is a Scholar of the Medical Research Coun-Assessment of dystrophin expression and sarcolemmal

stability cil of Canada.Serial cross-sections (12 mm) from injected and nonin-jected bundles of diaphragm muscles were obtained,collected on Superfrost plus slides (Fisher Scientific, ReferencesPittsburgh, PA, USA) and kept at −80°C. For immunohis-

1 Hoffman EP, Brown RH, Kunkel LM. Dystrophin: the proteintochemical detection of dystrophin, slides were first incu-product of the Duchenne Muscular Dystrophy locus. Cell 1987;bated in 0.5% BSA, 0.15% glycine in PBS (buffer A) for51: 919–928.15 min. Nonspecific binding was blocked by preincu-

2 Koenig M, Monaco AP, Kunkel LM. The complete sequence ofbation of tissue sections for 15 min in buffer A sup- dystrophin predicts a rod-shaped cytoskeletal protein. Cell 1988;plemented with 5% normal goat serum. Sections were 53: 219–228.then incubated for 1 h at room temperature with the 3 Watkins SC, Hoffman EP, Slayter HS, Kunkel LM. Immunoelec-polyclonal rabbit anti-6–10 dystrophin antibody44 diluted tron microscopic localization of dystrophin in myofibres. Nature1:2000 in buffer A. In some experiments, the primary anti- 1988; 333: 863–866.

4 Zubrzycka-Gaarn EE et al. The Duchenne muscular gene pro-body used was a polyclonal antibody raised against dys-duct is localized in sarcolemma of human skeletal muscle. Nat-trophin from Torpedo marmorata45 (1:500 in buffer A).ure 1988; 333: 466–469.Horseradish peroxidase-conjugated goat anti-rabbit IgG

5 Levine BA, Moir AJG, Patchell VB, Perry SV. The interaction of(Kirkegaard and Perry Laboratories, Gaithersburg, MD,actin with dystrophin. FEBS Lett 1990; 263: 159–162.USA) was used as a secondary antibody (1:500 in buffer

6 Way M et al. Expression of the N-terminal of dystrophin in E.A). Following the histochemical reaction and for the coli and demonstration of binding to F-actin. FEBS Lett 1992; 301:observation of procion orange-positive fibers, the sections 243–245.were dehydrated through a series of ethanol solutions, 7 Ohlendieck K, Ervasti JM, Snook JB, Campbell KP. Dystrophin-cleared with xylene and mounted with Permount (Fisher glycoprotein complex is highly enriched in isolated skeletal

muscle sarcolemma. J Cell Biol 1991; 112: 135–148.Scientific). For detection of dystrophin-positive fibers by8 Campbell KP, Kahl SD. Association of dystrophin and an inte-immunofluorescence, procion orange was omitted from

gral membrane glycoprotein. Nature 1989; 338: 259–261.the Krebs-Ringer solution since it interferes with fluor-9 Ervasti JM, Campbell KP. Dystrophin and the membrane skel-escent signals. For these studies, Cy-3 conjugated goat

eton. Curr Opin Cell Biol 1993; 5: 82–87.anti-rabbit IgG (Sigma; 1:500 in buffer A) was used as a10 Pasternak C, Wong S, Elson EL. Mechanical function of dystro-secondary antibody. Following thorough washing of the phin in muscle cells. J Cell Biol 1995; 128: 355–361.

secondary antibody, slides were mounted with CitiFluor 11 Fabbrizio E et al. Dystrophin, the protein that promotes mem-(Canterbury, UK). brane resistance. Biochem Biophys Res Commun 1995; 213: 295–

Slides were observed with a Zeiss Axiophot photomic- 301.roscope (Oberkochen, Germany). Direct light microscopy 12 Menke A, Jockush H. Decreased osmotic stability of dystrophin-

less muscle cells from the mdx mouse. Nature 1991; 349: 69–71.and epifluorescence illumination were used to visualize13 Weller B, Karpati G, Carpenter S. Dystrophin-deficient mdxprocion orange/peroxidase-positive muscle fibers and

muscle fibers are preferentially vulnerable to necrosis inducedthe fluorescent signal corresponding to dystrophin,by experimental lengthening contractions. J Neurol Sci 1990; 100:respectively. Kodak T-Max black and white and Kodak9–13.Gold colour films (Newhaven, CT, USA) were used for

14 Moens P, Baatsen PHWW, Marechal G. Increased susceptibilityphotodocumentation. of EDL muscles from mdx mice to damage induced by contrac-tions with stretch. J Musc Res Cell Mot 1993; 14: 446–451.

Data analysis 15 Petrof BJ et al. Dystrophin protects the sarcolemma from stressesDystrophin- and procion orange-positive fibers present in developed during muscle contraction. Proc Natl Acad Sci USA

1993; 90: 3710–3714.the entire muscle bundles were counted and their num-16 Yang B et al. SHE domain-mediated interaction of dystroglycanber expressed as a percentage of the total number of fib-

and Grb2. J Biol Chem 1995; 270: 11711–11714.ers included in the bundle which represents approxi-17 Brenman JE et al. Nitric oxide synthase complexed with dystro-mately 400 fibers. These analyses were performed on

phin and absent from skeletal muscle sarcolemma in Duchenneseparate occasions by two individuals. One-tail Studentsmuscular dystrophy. Cell 1995; 82: 743–752.t tests were used to locate significant differences between 18 Brenman JE et al. Interaction of nitric oxide synthase with the

individual group means. The level of significance was set postsynaptic density protein PSD-95 and a1-syntrophinat ,0.05. mediated by PDZ domains. Cell 1996; 84: 757–767.

19 Grozdanovic Z, Gosztonyi G, Gossrau R. Nitric oxide synthaseI (NOS-I) is deficient in the sarcolemma of striated muscle fibresAcknowledgementsin patients with Duchenne muscular dystrophy, suggesting anassociation with dystrophin. Acta Histochem 1996; 98: 61–69.We thank Dr Verne M Chapman (Rosewell Park Cancer

20 Bulfield G, Siller WG, Wight PAL, Moore KJ. X chromosome-Institute, Buffalo, NY, USA) for providing breeding pairslinked muscular dystrophy (mdx) in the mouse. Proc Natl Acadof the mdx4cv mice, and Drs Simon C Watkins (UniversitySci USA 1984; 81: 1189–1192.of Pittsburgh, Pittsburgh, PA, USA) and Jean Cartaud

21 Sicinski P et al. The molecular basis of muscular dystrophy in(Institut Jacques Monod, Paris, France) for the generous the mdx mouse: a point mutation. Science 1989; 244: 1578–1580.gift of anti-dystrophin antibodies. The expert technical 22 Dunckley MG, Wells DJ, Walsh FS, Dickson G. Direct retroviral-assistance of Mr Stephane Landry is gratefully acknowl- mediated transfer of a dystrophin minigene into mdx mouseedged. This work was supported by a grant from the muscle in vitro. Hum Mol Genet 1993; 2: 717–723.

23 Ragot T et al. Efficient adenovirus-mediated transfer of a humanMuscular Dystrophy Association of America, the Medical

Direct gene transfer into mdx diaphragm muscleA Decrouy et al

408 minidystrophin gene to skeletal muscle of mdx mice. Nature 35 Decrouy A et al. Direct gene transfer with mini-dystrophin generesults in rapid improvement of physiological properties of mdx1993; 361: 647–650.diaphragm muscle. Mol Biol Cell 1995; 6: 152a.24 Vincent N et al. Long-term correction of mouse dystrophic

36 McCully KK, Faulkner JA. Injury to skeletal muscle fibers ofdegeneration by adenovirus-mediated transfer of a minidystro-mice following lengthening contractions. J Appl Physiol 1985; 59:phin gene. Nat Genet 1993; 5: 130–134.119–126.25 Ascadi G et al. Human dystrophin expression in mdx mice after

37 Alameddine HS et al. Expression of a recombinant dystrophinintramuscular injection of DNA constructs. Nature 1991; 352:in mdx mice using adenovirus vector. Neuromusc Dis 1994; 4:815–818.193–203.26 Danko I et al. Dystrophin expression improves myofiber sur-

38 Wells DJ et al. Human dystrophin expression corrects thevival in mdx muscle following intramuscular plasmid DNAmyopathic phenotype in transgenic mdx mice. Hum Mol Genetinjection. Hum Mol Genet 1993; 2: 2055–2061.1992; 1: 35–45.27 England SB et al. Very mild muscular dystrophy associated with

39 Cox GA et al. Overexpression of dystrophin in transgenic mdxthe deletion of 46% of dystrophin. Nature 1990; 343: 180–182.mice eliminates dystrophic symptoms without toxicity. Nature28 Deconinck N et al. Functional protection of dystrophic mouse1993; 364: 725–729.(mdx) muscles after adenovirus-mediated transfer of a dystro-

40 Matsumura K, Lee CC, Caskey CT, Campbell KP. Restorationphin minigene. Proc Natl Acad Sci USA 1996; 93: 3570–3574.of dystrophin-associated proteins in skeletal muscle of mdx mice29 Stedman HH et al. The mdx mouse diaphragm reproduces thetransgenic for dystrophin gene. FEBS Lett 1993; 320: 276–280.

degenerative changes of Duchenne muscular dystrophy. Nature 41 Phelps SF et al. Expression of full-length and truncated dystro-1991; 352: 536–539. phin mini-genes in transgenic mdx mice. Hum Mol Genet 1995;

30 Dupont-Versteegden EE, McCarter RJ. Differential expression of 4: 1251–1258.muscular dystrophy in diaphragm versus hindlimb muscles of 42 Wells D et al. Expression of human full-length and minidystro-mdx mice. Muscle and Nerve 1992; 15: 1105–1110. phin in transgenic mdx mice: implications for gene therapy of

31 Boland B, Himpens B, Denef JF, Gillis JM. Site-dependent patho- Duchenne muscular dystrophy. Hum Mol Genet 1995; 4: 1245–logical differences in smooth muscles and skeletal muscles of 1250.the adult mdx mouse. Muscle and Nerve 1995; 18: 649–657. 43 Danko I, Chapman V, Wolff JA. The frequency of revertants in

32 Davis HL, Jasmin BJ. Direct gene transfer into mouse dia- mdx mouse genetic models for Duchenne muscular dystrophy.phragm. FEBS Lett 1993; 333: 146–150. Pediatr Res 1992; 32: 128–131.

33 Tanabe Y, Esaki K, Nomura T. Skeletal muscle pathology in X 44 Lidov HGW, Byers TJ, Watkins SC, Kunkel LM. Localization ofchromosome-linked muscular dystrophy (mdx) mouse. Acta dystrophin to postsynaptic regions of central nervous systemNeuropathol (Berlin) 1986; 69: 91–95. cortical neurons. Nature 1990; 348: 725–728.

34 Karpati G, Carpenter S, Prescott S. Small-caliber skeletal muscle 45 Cartaud A et al. Localization of dystrophin and dystrophin-fibers do not suffer necrosis in mdx mouse dystrophy. Muscle related protein at the electromotor synapse and neuromuscular

junction in Torpedo marmorata. Neuroscience 1992; 48: 995–1003.and Nerve 1988; 11: 795–803.