tTipmemrnwcitwm1tssasGbca14Wopctgi

p

ca

Experimental Neurology 155, 22–30 (1999)Article ID exnr.1998.6973, available online at http://www.idealibrary.com on

0CA

Successful Myoblast Transplantation in Primates Dependson Appropriate Cell Delivery and Induction of

Regeneration in the Host Muscle1

Daniel Skuk, Brigitte Roy, Marlyne Goulet, and Jacques P. TremblayUnite de Recherche en Genetique Humaine, Centre Hospitalier de l’Universite Laval, Quebec, Canada

Received April 8, 1998; accepted October 8, 1998

arpatcri

4dcmddsrmt

rfiDpadibtccbcitmt

Myoblast transplantation (MT) may be a potentialreatment for severe recessive hereditary myopathies.he limited results of MT in clinical trials led us to

mprove this technique in monkeys, an animal modelhylogenetically similar to humans. Three Macacaulata monkeys were used as donors and six as receiv-

rs for MT. Myoblasts were grown in culture fromuscle biopsies of adult monkeys and infected with a

etroviral vector encoding the LacZ gene. Differentumbers of cells (i.e., 4 3 106, 8 3 106, and 24 3 106 cells)ere transplanted into different muscles and 8 3 106

ells (resuspended in a notexin solution) were injectedn one muscle of four monkeys. For these transplanta-ions, the cell suspension (in a volume of about 100 ml)as injected at 35 sites less than 1 mm apart. Two otheronkeys received 100 3 106 myoblasts resuspended inml of HBSS or 1 ml of notexin. For these two monkeys,

he myoblasts were injected at 200–250 sites within amall portion of the muscle.All monkeys were immuno-uppressed with daily injections of FK506. Four weeksfter MT, the transplanted muscle portions were biop-ied and the presence of b-galactosidase-positive (b-al1) muscle fibers was investigated. The number of-Gal1 fibers was 822 6 150 (site grafted with 4 3 106

ells), 1253 6 515 (8 3 106 cells), 1084 6 278 (24 3 106),nd 2852 6 1211 (notexin). In the monkeys grafted with00 3 106 myoblasts, the number of b-Gal1 fibers was850 (site without notexin) and 9600 (site with notexin).e demonstrated that a precise mechanical distribution

f myoblasts into the tissue improves substantially MT inrimates. The presence of notexin with the transplantedells further increased the success of their transplanta-ion. These are the best results obtained either with MT orene therapy in primates and they encourage the possibil-ty to human MT trials. r 1999 Academic Press

Key Words: myoblast; monkey; notexin; FK506; trans-lantation.

1 This work was supported by grants from the Association Fran-¸aise contre les Myopathies, the Muscular Dystrophy Association,

cnd the Muscular Dystrophy Association of Canada.

22014-4886/99 $30.00opyright r 1999 by Academic Pressll rights of reproduction in any form reserved.

INTRODUCTION

Myoblast transplantation (MT) is an experimentalpproach originally proposed to treat severe and deadlyecessive myopathies like Duchenne muscular dystro-hy (34, 44). Other applications also suggested for MTre the replacement of infarcted myocardium (41, 50),he improvement of the levator palpebrae superioris inongenital ptosis (3), and the systemic delivery ofecombinant proteins to treat a variety of acquired andnherited diseases (4, 10).

After the first promising results of MT in mice (34,0, 44), several clinical trials were conducted withystrophic patients (20, 27, 35, 38, 39, 49). All these MTlinical trials were unsuccessful in terms of significantuscle function improvement. Only the histological

emonstration of a significant but small increase inystrophin-positive fibers was observed by some re-earchers (20, 35, 38, 49). Because of the limitedesults, the clinical trials were discontinued and experi-entation in animals was pursued only by some groups

o explain the reasons of the failure.Animal research demonstrated that immune-specific

eactions against the donor cells and hybrid musclebers take place some days after MT (15, 16, 24).ifferent immunosuppressive treatments such as cyclos-orine A (21), cyclophosphamide (53), rapamycin (52),nd FK506 (29) were tested in mice by our team. Theegree of success of MT varied depending on themmunosuppressive treatment (51), the best resultseing obtained with the FK506 (29). It is noteworthyhat cyclosporine A was not sufficient to suppress theellular immune reaction even at high doses (21), andyclophosphamide was toxic for the transplanted myo-lasts (53). This observation was important becauseyclosporine A and cyclophosphamide were the onlymmunosuppressive treatments used in the clinicalrials. FK506 was also observed to be effective for MT inonkeys (30, 32). These results confirmed the role of

he immune-specific reaction in the failure of MT

linical trials and the importance of the choice of the

iwata

sMwwmtAmtsmmtMmjTtgpct3ea

whdMteeacomsi

A

8mrbgx

cCb

C

bBcbmwwcsMrtwbltsb2

M

titp5rc1HaHbissRptmipMpatT

23SUCCESSFUL MYOBLAST TRANSPLANTATION IN PRIMATES

mmunosuppressive treatment. Two other problemsere also signaled as limiting the efficacy of MT: thebsence of migration of myoblasts into the muscularissue (25) and the massive mortality of grafted cellsfter the transplantation (6, 7, 11, 13, 14, 19).In spite of these last two problems, MT was very

uccessfully performed in mice (22, 29, 31). However,T in monkeys was not as successful as in mice evenhen the immune-specific response was controlledith FK506 (30, 32). This low level of the initial MT inonkeys may be due to the transplantation technique

hat was similar to that used in human clinical trials.s already signaled, the conditions in mouse experi-ents were largely different from those used in clinical

rials and this could also explain the difference ofuccess of MT (12). Highly successful experiments inice were performed by injecting hundreds of thousandyoblasts into a very small muscle by multiple injec-

ions very close each other (fractions of a millimeter).ouse muscles were also irradiated to prevent hostyoblast proliferation and myotoxic drugs were in-

ected to trigger a degeneration–regeneration cycle.hese experimental conditions permitted to our groupo obtain 90% of the muscle fibers expressing a reporterene present in the transplanted myoblasts (29). Inatients, large muscles were injected with a signifi-antly lower myoblast-to-muscle volume ratio, by injec-ions significantly far from each other [e.g., 0.5 cm (27,9)]. Only one patient showed 10% of muscle fibersxpressing donor-derived dystrophin in a biopsy (38)nd this was so far the best result of MT in humans.In the present study, we used in monkeys conditionshich were more similar to those which producedighly successful MT in mice. Experiments were con-ucted in monkeys because we think that the results ofT in primates can be extrapolated to humans, due to

heir phylogenetic relation. In the present series ofxperiments, we demonstrated that an effective deliv-ry of myoblasts into the muscle compensates for theirbsence of migration, increasing substantially the suc-ess of their transplantation. Moreover, the injectionsf myoblasts with notexin (one of the most potent of theyotoxic phospholipases) allow us to achieve highly

uccessful MT comparable to the best results obtainedn mouse models.

MATERIALS AND METHODS

nimals

Nine female monkeys (Macaca mulata) among 5 andyears old were included in this study. Three of theseonkeys were used as myoblast donors and six as

eceivers for allotransplantations (monkeys 1 to 6). Foriopsies and transplantations, the animals were undereneral anesthesia using a solution of 13 mg/ml of

ylazine and 87 mg/ml of ketamine. This work was a

onducted according to the guidelines set out by theanadian Council of Animal Care and was authorizedy the Laval University Animal Care Committee.

ell Culture

Primary myoblast cultures were grown from muscleiopsies taken 2 weeks before the transplantation.iopsies from monkeys were performed in the quadri-

eps femoris by opening the skin and aponeurosis. Aiopsy of approximately 0.5 cm3 was taken and theuscle, aponeurosis, and skin were closed separatelyith absorbable sutures. The biopsies were dissociatedith collagenase and trypsin as described (32) and

ultured in MCDB 120 medium with 15% fetal calferum and 10 ng/ml of basic fibroblast growth factor.yoblasts were infected three times in vitro with a

etroviral vector LNPOZ encoding de LacZ gene. Forransplantation, the cells were detached from the flasksith 0.1% trypsin and rinsed three times in Hank’salanced salt solution (HBSS). The percentage of b-ga-actosidase (b-Gal) labeled cells at the time of transplan-ation was estimated by staining cultures or cell suspen-ions with X-Gal as described below. The percentage of-Gal-labeled cells at the time of injection was roughly0% in monkeys 1–4 and 12% in monkeys 5 and 6.

yoblast Transplantation

In monkeys 1–4, four different muscle portions wereransplanted within 2 h, these segments being locatedn both biceps brachii and both quadriceps femoris. Forhe transplantation, the final cell pellets were resus-ended in 100 µl of HBSS or in 100 µl of notexin dilutedµg/ml in HBSS. For each animal, three muscle sites

eceived, respectively, 4 3 106, 8 3 106, and 24 3 106

ells in HBSS and one muscle site was injected with 8 306 cells resuspended in 5 µg/ml of notexin diluted inBSS. To perform the transplantation, the skin andponeurosis were opened to expose the muscle. A 50-µlamilton syringe was used to inject the cells. Myo-lasts were injected obliquely or almost longitudinallynto a segment of 5–8 mm of width just under theurface of the muscle. Injections of about 3 µl of celluspension were done during the needle withdrawal.oughly 35 to 40 injections were performed for eachortion of muscle. The distance between the injectionrajectories was less than 1 mm. As cell injections wereade in a small portion of a voluminous muscle, it was

mportant to insure the appropriate sampling of thatortion at the moment of the biopsy. For this reason,T were performed superficially in the muscle. Two

oints of nonabsorbable suture were placed distallynd proximal to the injection site, comprising exactlyhe region transplanted, to identify it 1 month later.he sutures were placed after the transplantation to

void a possible escape of the injected cells along the

tw

irnCiacasr

I

lmFbtatua

M

msoSc

cswdhbtcpatX

w(wppbsp

ibp3o

imnhfi

I

auptemwccswtdo

I

mpia

M

f

24 SKUK ET AL.

rajectory of the sutures. The aponeurosis and skinere closed separately with absorbable sutures.In monkeys 5 and 6, the right biceps brachii was

njected with 100 3 106 myoblasts. Myoblasts wereesuspended in 1 ml of HBSS containing 5 µg/mlotexin (monkey 5) or without notexin (monkey 6).ells were injected as in monkeys 1–4 but 200–250

njections of about 5 µl were performed less than 1 mmpart with a 100 µl Hamilton syringe. These injectionsovered almost 1 cm in the width of the muscle andbout 0.8 cm in depth. Two points of nonabsorbableuture were placed as indicated above and the aponeu-osis and skin were closed with absorbable sutures.

mmunosuppression

The monkeys were immunosuppressed intramuscu-arly with FK506 (a generous gift from Fujisawa Phar-

aceutical Co., Ltd., Osaka, Japan). In monkeys 1–4,K506 was given at 0.5 mg/kg/day starting 3 daysefore the transplantation to 3 days after and thereforehe dose was reduced to 0.15 mg/kg/day. Taking intoccount that lymphocyte infiltration was observed inhe biopsies of these monkeys, the FK506 was contin-ed at 0.5 mg/kg/day during the 4 weeks in monkeys 5nd 6.

uscle Examination

Four weeks after the transplantation, the grafteduscle portions were biopsied under general anesthe-

ia. The muscle biopsies were placed in sucrose 30%vernight at 4°C and then frozen in liquid nitrogen.erial sections of 30 and 10 µm were performed in aryostat at 225°C.

Sections were stained to evaluate different histologi-al features. Hematoxylin-eosin was performed to ob-erve the general histology of the grafted muscle. b-Galas revealed by an histochemical method previouslyescribed (29). Briefly, sections were fixed in glutaralde-yde 0.25%, washed three times with phosphate-uffered saline (PBS) and incubated overnight at roomemperature with a solution of 0.4 M 5-bromo-4-hloro-3 indolyl-b-D-galactopyranoside (X-Gal), 3 mMotassium ferrocyanide, 3 mM potassium ferricyanide,nd 1 mM magnesium chloride in PBS, pH 7.4. Hema-oxylin-eosin slides were mounted in Permount and-Gal slides mounted in a glycerin-gelatin medium.Immunohistochemistry for T-cytotoxic lymphocytesas performed with a mouse anti-human CD8 antibody

Exalpha, Boston, MA) previously known to cross-reactith monkey CD81 lymphocytes (32). The endogenouseroxidase activity was blocked with 1% hydrogeneroxide in PBS for 10 min. Nonspecific binding waslocked by a 30-min incubation with 10% fetal bovineerum in PBS. The sections were incubated 1 h with the

rimary antibody diluted 1/200, followed of a 30-min H

ncubation with 1/150 biotinylated anti-mouse anti-ody and a 30-min incubation in 1/200 streptavidin-eroxidase. The peroxidase activity was revealed with,38 diaminobenzidine (0.25 mg/ml) and hydrogen per-xide (0.03%). The slides were mounted in glycerin-gelatin.The number of b-Gal expressing fibers was counted

n the two sections where they were apparently theost numerous and the highest value was taken. The

umber of fibers with internal nuclei was counted inematoxylin-eosin preparations. In monkey 5, the sur-

ace covered by b-Gal1 fibers was measured in anmage analyzer.

RESULTS

njection of Different Numbers of Myoblasts

Implantation of allogenic b-Gal-labeled myoblasts inprecise pattern in the muscles of the six monkeys

sed in this study showed large quantities of b-Gal-ositive (b-Gal1) fibers 4 weeks after the transplanta-ion. The number of b-Gal1 muscle fibers counted inach muscle are shown in Table 1. The results foronkeys 1–4 are represented graphically in the Fig. 1,here the global effect of grafting different numbers of

ells or injecting the myoblasts resuspended in notexinould be seen more clearly. Injection of 8 3 106 cellseems to increase the number of b-Gal fibers by 50%hen compared to the injection of 4 3 106 cells. Never-

heless, injection of 24 3 106 cells in the same mannerid not result in more b-Gal1 fibers than the injectionf 8 3 106 myoblasts in the same muscle volume.

njection of Myoblasts with Notexin

We confirmed that notexin was not toxic to primateyoblasts in vitro before injecting myoblasts resus-

ended in this substance. The cells were resuspendedn 5 µg/ml of notexin diluted in HBSS or in HBSS alonend lead 4 to 12 h at 4°C. After this incubation, the

TABLE 1

onkey

Number of transplanted myoblasts

4 3 106 8 3 106 24 3 1068 3 106

1 notexin 100 3 106100 3 106

1 notexin

1 636 485 1354 14442 996 1434 1285 37823 863 1570 905 39384 795 1523 791 22445 96006 4850

Note. Numbers of b-Gal1 muscle fibers in monkeys 1–6, as aunction of the number of injected myoblasts resuspended in only

BSS or HBSS with notexin.

vesoH

ofionfiabwsbIda

q(c

S

h2qfiawicoswmc

M

mmwfiba(tfiwcdmpdhit

hnrsbn(nFwfia

w11

25SUCCESSFUL MYOBLAST TRANSPLANTATION IN PRIMATES

iability of these cells was evaluated in an hemacytom-ter using Trypan blue as a dye exclusion test. Noignificant increase in the number of dead cells wasbserved in notexin exposed myoblasts in reference toBSS-resuspended myoblasts.Injection of 8 3 106 myoblasts resuspended in 5 µg/ml

f notexin more than doubled the number of b-Gal1bers considering the same number of cells injectednly in HBSS (Fig. 1). Indeed, in two animals theumber of b-Gal1 fibers was as high as 3782 and 3938bers, including almost all the fibers in the grafted areand comprising fascicles composed almost entirely by-Gal1 fibers (Fig. 2A). The level of b-Gal expressionas variable from one fiber to another, but the exten-

ion of the region expressing the highest density of-Gal1 fibers covered 3–4 mm of length in the muscle.n the sections of muscles not injected with notexin, theistribution of b-Gal1 fibers was not as homogeneouss in notexin injected sites. b-Gal1 fibers were fre-

FIG. 2. Transverse section of monkey muscles 1 month after traematoxylin-eosine and F and H were immunostained for CD81 lymotexin (A) show 3900 b-Gal1 fibers homogeneously distributed in thesuspended only in HBSS (B), less than 1500 b-Gal fibers were presurface (up) to the depth of the muscle (bottom). In a transversal secti-Gal labeled) in the presence of notexin, all fascicles are composedotexin, b-Gal1 fibers are arranged in parallel bands showing the p

left) to the deep tissue (right). Internal nuclei are observed in most ootexin in monkey 5 (E). Few isolated CD81 cells are observed in theK506 at 0.5 mg/kg/day during the 4 weeks after MT. When FK506ere observed between the b-Gal1 fibers (G), many of them correspobers surrounded by lymphocytes may indicate cell lysis and release o

FIG. 1. Mean number of b-Gal1 muscle fibers and muscle fibersith internal nuclei in transverse sections corresponding to monkeys–4, 1 month after the transplantation of 4 3 106, 8 3 106, and 24 306 myoblasts and 8 3 106 cells in the presence of notexin.

uently in parallel bands when notexin was omittedFig. 2B), these bands corresponding with the mechani-al damage produce by the injections.

igns of Muscle Regeneration

Internal nuclei were observed in many fibers inematoxylin-eosin-stained sections of grafted sites (Fig.E). Since nuclear internalization is considered a conse-uence of recent muscle regeneration, the number ofbers with internal nuclei was counted in the sectionsdjacent to those where the number of b-Gal1 fibersas estimated. The number of muscle fibers with

nternal nuclei was high in all transplanted muscles, asan be seen in Fig. 1. No significant differences werebserved between the muscles grafted without notexin,uggesting that roughly the same mechanical damageas produced. On the other hand, the number ofuscle fibers with internal nuclei was largely in-

reased in muscle portions injected with notexin (Fig. 1).

assive Transplantation of Myoblasts

Monkeys 5 and 6 were grafted with 100 3 106

yoblasts, from which 12% were b-Gal1 at the mo-ent of transplantation. In monkey 5, where myoblastsere transplanted with notexin, the number of b-Gal1bers was 9600, covering an area of 48 mm2. The-Gal1 fibers were distributed homogeneously in thisrea, including almost the totality of the muscle fibersFig. 2C). This pattern of b-Gal staining was observedhrough 6 mm along the muscle and many b-Gal1bers extended further. In monkey 6, where notexinas not used, approximately 4850 b-Gal1 fibers were

ounted in a single transversal section, and the sameensity of fibers was observed through 4 mm along theuscle. In this monkey, b-Gal1 fibers were disposed in

arallel bands arranged from the muscle surface to theeep (Fig. 2D). In monkey 5, injected with notexin,ematoxylin-eosin-stained sections showed internal-

zed nuclei in almost all the fibers present in theransplanted muscle portion (Fig. 2E). In monkey 6,

lantation. b-Gal was revealed in A to D and G; E was stained withocytes. In monkey 2, a site injected with 8 3 106 cells in presence ofrafted tissue. In the contralateral muscle injected with 8 3 106 cellst and they showed an alignment in parallel bands oriented from thecorresponding to the monkey 5 (C), grafted with 100 3 106 cells (12%b-Gal1 fibers. In monkey 6 (D), grafted as monkey 5 but omitting

ern of the mechanical damage of injections from the muscle surfacee muscle fibers in the site injected with myoblasts in the presence ofnnective tissue of transplanted muscles immunosupressed with i/mreduced to 0.15 mg/kg/day (3 days after MT), focal cell infiltrations

ing to CD81 lymphocytes (H). The diffuse staining for b-Gal aroundtracellular b-Gal (G, arrow). Original magnification was 403 (A to D)

nspphe genonbyattf thco

wasndf in

nd 2003 (E to H).

26 SKUK ET AL.

nm

C

ttmMtswifiiiLiaFt

P

acepmTttthp

lttiaicctbtwlgci

li

tbstpt5pmckcnna4tbm

R

dsfiscntdvgwnicfNcgtbuimesnmn

27SUCCESSFUL MYOBLAST TRANSPLANTATION IN PRIMATES

uclear internalization was considerably less than inonkey 5.

ellular Infiltration in the Transplanted Sites

The presence of cell infiltration was evident in hema-oxylin-eosin stained sections for monkeys 1–4. Inhese monkeys, FK506 was administered i.m. at 0.15g/kg/day after a 6 days period at 0.5 mg/kg/day.ononuclear cell infiltration was observed only in

ransplanted areas as diffuse infiltration in perimy-ium and endomysium and focal infiltrates, most ofhich were centered by vascular elements. These focal

nfiltrates were comprising and surrounding b-Gal1bers, which frequently were invaded by the infiltrat-

ng cells (Fig. 2G). Immunohistochemically, many of thenfiltrating cells expressed the CD8 antigen (Fig. 2H).ymphocyte infiltration was almost completely abol-

shed in monkeys 5 and 6, where FK506 was given i.m.t 0.5 mg/kg/day during the 4 weeks of the experiment.ew CD81 cells were observed scattered in the intersti-

ial tissue and near vascular elements (Fig. 2F).

DISCUSSION

arameters Limiting the Success of MTin Large Animals

Up to date, experiments of MT were characterized byduality of results between mouse experiments and

linical trials as well as large animals models. Veryfficient grafts were obtained in mice (22, 29, 31), whileoor results were observed after MT in dogs (26),onkeys (30, 32), and humans (20, 27, 35, 38, 39, 49).he poor results of MT in large animals were attributedo different problems like the absence of migration ofransplanted cells (25) and the high rate of mortality ofransplanted myoblasts (6, 11, 13, 19), which our groupad hypothesized to be more pronounced in dogs andrimates than in mice (13).Without neglecting the importance of the last prob-

em, these different results could also be explained byhe different techniques of myoblast delivery into theissue. Successful MT in mice were obtained by inject-ng the cells at multiple sites, fraction of a millimeterpart, into a very small muscle preconditioned byrradiation and the injection of notexin (29). We canonsider that the poor migration of myoblasts wasompensated by a precise mechanical distribution intohe tissue and favored by the muscle damage inducedy notexin. In humans, voluminous muscles not condi-ioned by irradiation or myotoxic agents were injectedith a number of myoblasts significantly smaller for a

arge muscle and the injections were performed atreater distance from each other than in mice [e.g., 0.5m (27, 39)]. Considering that myoblasts do not migrate

nto the muscle (25), their only possibility under these n

ast conditions is to be incorporated into the fibersmmediately next to the sites of injection.

Experiments in monkeys seem to be of great impor-ance to elucidate the feasibility of MT in humansecause their phylogenetic relation. In the presenttudy in monkeys, we tried to approach the conditionshat gave successful MT results in mice. MT wereerformed with a Hamilton syringe, allowing the injec-ion of very small quantities of cell suspension (i.e., 3 to

µl in our experiments). Indeed, the injections werelaced fractions of a millimeter apart. This preciseyoblast distribution into the muscle tissue was suffi-

ient to improve substantially the MT success in mon-eys. The number of b-Gal1 fibers was highly in-reased in reference to our previous experiments, whereo more than 100 b-Gal1 fibers were obtained whenotexin was not injected (30, 32). In our present studynd in the sites where notexin was not used, between85 and 1570 b-Gal1 fibers were obtained in sitesransplanted with 4 to 24 3 106 myoblasts and 4850-Gal1 fibers in a site transplanted with 100 3 106

yoblasts.

elationship between the Number of Myoblastsand Transplantation Success

Our experiment was also conducted to elucidate ififferent numbers of myoblasts injected in a site neces-arily produced differences in the number of b-Gal1bers 1 month later. Although a difference was ob-erved between the injection of 4 3 106 and 8 3 106

ells, more b-Gal1 fibers were not obtained when theumber of injected cells was increased to 24 3 106 inhe same muscle volume. This indicates that whenifferent numbers of cells are injected in the sameolume of muscle, there is a limit in the increase ofraft success (evaluated in term of number of fibers). Itas previously observed in mice that when differentumbers of myoblasts were implanted with a single

njection, the number of b-Gal1 fibers increases to aertain level and thereafter more myoblasts did noturther increase the number of b-Gal1 fibers (46).evertheless, although in the present study an in-

rease in the number of myoblasts transplanted in aiven site did not increase the number of b-Gal1 fibers,his increased the b-Gal staining intensity, probablyecause there were more nuclei coding for b-Gal pernit length of muscle fiber. It may be important to

nvestigate in monkeys how to optimize the number ofyoblasts needed to cover a certain volume of muscle in

ventual clinical trials. It should, however, be empha-ized that the number of myoblasts may depend on theuclear domain of a given protein (i.e., the length ofuscle fiber over which a protein coded by a single

ucleus is expressed). We have recently found that the

uclear domain of b-Gal in mouse muscle fibers is

a(

I

fsst2snan

fIi2enicdnbipTcmSfimotpd

eumtmtfiomc

I

stfi

nctiSrwdwmrcit

dreraertt

I

uTuaicircapdpteptciclmpl

N

28 SKUK ET AL.

round 900 µm while that of dystrophin is only 500 µm28).

mproved Transplantation Success with Notexin

Notexin is a potent myotoxic phospholipase purifiedrom the venom of the Australian tiger snake Notechiscutatus scutatus (18, 45, 54). This protein was used totudy the muscle regeneration (18, 48) and to improvehe success of myoblast transplantation in mice (22,9). Notexin damages and induces the necrosis of thekeletal muscle fibers, preserving the satellite cells,erves, and vascular elements (17). This specific effectllows complete and rapid regeneration after muscleecrosis (48).The injection of myoblasts resuspended in notexin

urther improved the success of their transplantation.ndeed, the concomitant injection of notexin was moremportant than the threefold increase (8 3 106 to4 3 106) in the number of transplanted myoblasts. Inarlier experiments in monkeys (30, 32), injection ofotexin the day before the cell implantation did not

mprove the success of MT as in the present study. Thisould be explained because in large muscles it isifficult to insure that cell implantation 1 day afterotexin injection is exactly in the same areas damagedy the myotoxic agent (33). However, if myoblasts arenjected resuspended in notexin, these cells will beresent in the areas where muscle necrosis is induced.his allows them to efficiently participate in the pro-ess of regeneration, proliferating together with hostyoblasts and incorporating in regenerating fibers.ince as indicated above, notexin damages the musclebers but not the satellite cells (17), we expected thatyoblasts could be resuspended in notexin without lost

f viability. In agreement with those in vivo observa-ions, we have confirmed that notexin was not toxic torimate myoblasts in vitro at the concentration used toamage the skeletal muscle fibers.To be applicable in clinical trials, it is important to

lucidate if MT could be effectively performed in vol-mes larger than the small muscle of a mouse. Twoonkeys were used in our study to test the possibility

o transplant large volumes of muscle. One of these twoonkeys was injected with cells resuspended in no-

exin while the other received cells only in HBSS. Therst of these monkeys gave the best results of MTbtained in a primate up to date, where almost 10,000uscle fibers expressed the reporter gene, covering a

ross area of 0.5 cm2 for many millimeters in length.

mportance of Muscle Damage to Improve the MT

The distribution of the b-Gal1 muscle fibers demon-trates the importance of the muscle damage to inducehe fusion of transplanted myoblasts into the host

bers. b-Gal1 muscle fibers in sites not injected with p

otexin were arranged in bands, resembling the me-hanical damage produced by the injections. In con-rast, b-Gal1 fibers were homogeneously distributednto the transplanted sites when notexin was used.ince nuclear internalization is considered a sign ofecent muscle regeneration, the high number of fibersith internal nuclei confirms that larger areas ofegeneration–regeneration were induced in the muscleshere notexin was injected. These results suggest thatost of the transplanted myoblasts was recruited in

egenerating fibers, these regenerating fibers being theonsequence of the localized mechanical damage ofnjections or the extended muscle destruction by no-exin.

It may be important to investigate whether theamage in a dystrophic muscle could be effectivelyecovered by grafting normal myoblasts, without wors-ning its already reduced regenerative capacity. It wasecently demonstrated that MT into irreversibly dam-ged muscles (1, 23) or in a regenerating muscle (2) canffectively restore force in mice. Reproduction of similaresults in large animals should be important to answerhis question and experiments in dystrophic dogs seemo be the best way to do it.

mmunosuppression

In this study, we changed our FK506 oral protocolsed in former studies for intramuscular injections.he protocol of intramuscular FK506 administrationsed for monkeys 1–4 was taken from a study of lungllografts in monkeys (42). In the present study, focalnfiltrates of mononuclear cells (many of them beingytotoxic T lymphocytes) in areas of b-Gal1 fibersndicated a cellular rejection. Our team previouslyeported that the rejection phenomenon is alreadyompleted and lymphocyte infiltration was reduced orbsent 4 weeks after MT in monkeys not immunosup-ressed (32). Therefore, the high success of our graftsespite this cell infiltration could be explained by theossibility that our doses of FK506 were not sufficiento completely control the immune response but suchnough to slow down the cellular rejection and toermit success 1 month after transplantation. Sincehe same protocol gave mild or no perivascular lympho-yte infiltration in transplanted lungs (42), it couldndicate that rejection after MT is more difficult toontrol than after organ graft rejection, as we postu-ated previously (26). In monkeys 5 and 6, FK506 was

aintained at 0.5 mg/kg/day during the whole periodreceding the muscle biopsy. With this protocol, cellu-ar rejection was not observed.

otexin Toxicity

Although our results suggest that MT could be a

otential therapy when cells are implanted in a regen-

etseipi4mkrtwlt

djlbotomtclosm

tgtit

W

1

1

1

1

1

1

1

1

1

1

2

2

2

2

29SUCCESSFUL MYOBLAST TRANSPLANTATION IN PRIMATES

rating muscle, this raises questions over the toxicity ofhe method. Notexin is a constituent of the venom of thenake Notechis scutatus scutatus (18) and its principalffect is the induction of myonecrosis. In snake poison-ng where myonecrosis is the predominant event, theathogenic mechanism involved in the morbimortalitys the acute renal failure caused by myoglobinuria (9,7). The massive myolisis releases large quantities ofyoglobin to the circulation that are filtered in the

idneys inducing acute pathological changes in theenal tissue. In the present study, a total of 5 µg (lesshan 2 µg/kg) of notexin was injected in monkey 5ithout any ostensible effects. This dose was largely

esser than the lethal dose for dogs and cats, respec-ively, 30 and 100 µg/kg (5).

To avoid the use of notexin, other drugs that wereescribed as causing specific tissue damage when in-ected locally in muscles, sparing satellite cells, vascu-ar elements, and nerves (8, 36, 37, 43) could eventuallye used for MT. Nevertheless, if muscle damage is thenly option to incorporate the transplanted myoblasts,he risk of myoglobinuria could be present with anyther agent used to damage the muscle. As the risk ofyoglobinuria will depend on the volume of muscle

issue damaged, maybe a whole body MT in a patientould be performed in several different sessions underocal anesthesia. That will avoid the release of danger-us quantities of myoglobin or the risk of a highystemic absorption of the substance used to induce theuscle damage.Finally, it must be emphasized that these results are

he best achieved up to date to incorporate a foreignene into a host muscle in primates, both by cell or geneherapy. These results showed that MT is a real possibil-ty for patients if properly done and firmly encourageshe possibility to resuming human trials.

ACKNOWLEDGMENTS

The authors thank Mr. Steve Brochu for his technical assistance.e also acknowledge the support from the AFM and MDA.

REFERENCES

1. Allamedine, H. S., J. P. Louboutin, M. Dehaupas, A. Sebille, andM. Fardeau. 1994. Functional recovery induced by satellite cellgrafts in irreversibly injured muscles. Cell Transplant. 3: 3–14.

2. Arcila, M. E., B. T. Ameredes, J. F. Derosimo, C. H. Washabaugh,J. Yang, P. C. Johnson, and M. Ontell. 1997. Mass and functionalcapacity of regenerating muscle is enhanced by myoblast trans-fer. J. Neurobiol. 33: 185–198.

3. Baker, R. S., P. H. Bonner, J. D. Porter, M. N. Madhat, and J.Gross. 1993. Myoblast transfer therapy in the treatment ofptosis: A preliminary study. J. Pediatr. Ophthalmol. Strabismus30: 113–117.

4. Barr, E., and J. M. Leiden. 1991. Systemic delivery of recombi-nant proteins by genetically modified myoblasts. Science 254:

1507–1509.

5. Barr, S. C. 1984. Clinical features, therapy and epidemiology oftiger snake bite in dogs and cats. Aust. Vet. 61: 208–212.

6. Beauchamp, J. R., C. N. Pagel, and T. A. Partridge. 1997. A dualmarker system for quantitative studies of myoblast transplanta-tion in the mouse. Transplantation 63: 1794–1797.

7. Beauchamp, J. R., J. E. Morgan, C. N. Pagel, and T. A.Partridge. 1994. Quantitative studies of efficacy of myoblasttransplantation. Muscle Nerve 4(Suppl.): 261.

8. Caldwell, C. J., D. L. Mattey, and R. O. Weller. 1990. Role of thebasement membrane in the regeneration of skeletal muscle.Neuropathol. Appl. Neurobiol. 16: 225–238.

9. Chugh, K. S. 1989. Snake-bite-induced acute renal failure inIndia. Kidney Int. 35: 891–907.

0. Dhawan, J., L. C. Pan, G. K. Pavlath, M. A. Travis, A. M.Lanctot, and H. M. Blau. 1991. Systemic delivery of humangrowth hormone by injection of genetically engineered myo-blasts. Science 254: 1509–1512.

1. Fan, Y., M. Maley, M. Beilharz, and M. Grounds. 1996. Rapiddeath of injected myoblasts in myoblast transfer therapy. MuscleNerve 19: 853–860.

2. Grounds, M. D. 1996. Commentary on the present state ofknowledge for myoblast transfer therapy. Cell Transplantation5: 431–433.

3. Guerette, B., D. Skuk, F. Celestin, M. Entman, F. Tardif, I.Asselin, B. Roy, M. Goulet, C. Huard, and J. P. Tremblay. 1997.Prevention by anti-LFA-1 of acute myoblast death followingtransplantation. J. Immunol. 159: 2522–2531.

4. Guerette, B., I. Asselin, D. Skuk, M. Entman, and J. P. Trem-blay. 1997. Control of inflammatory damage by anti-LFA-1:Increase success of myoblast transplantation. Cell Transplant.6: 101–107.

5. Guerette, B., I. Asselin, J. T. Vilquin, R. Roy, and J. P. Tremblay.1994. Lymphocyte infiltration following allo and xeno myoblasttransplantation in mdx mice. Muscle Nerve 18: 39–51.

6. Guerette, B., R. Roy, M. Tremblay, I. Asselin, I. Kinoshita, andJ. P. Tremblay. 1995. Increased granzyme B mRNA followingalloincompatible myoblast transplantation. Transplantation 60:1011–1016.

7. Harris, J. B., and M. J. Cullen. 1990. Muscle necrosis caused bysnake venoms and toxins. Electr. Microsc. Rev. 3: 183–211.

8. Harris, J. B., and M. A. Johnson. 1978. Further observations onthe pathological responses of rat skeletal muscle to toxinsisolated from the venom of the Australian tiger snake, Notechisscutatus scutatus. Clin. Exp. Pharmacol. Physiol. 5: 587–600.

9. Huard, J., G. Acsadi, A. Jani, B. Massie, and G. Karpati. 1994.Gene transfer into skeletal muscles by isogenic myoblasts.Hum. Gene Ther. 5: 949.

0. Huard, J., J. P. Bouchard, R. Roy, F. Halouin, G. Dansereau, C.Labrecque, N. Albert, C. L. Richards, B. Lemieux, and J. P.Tremblay. 1992. Human myoblast transplantation: Preliminaryresults of 4 cases. Muscle Nerve 15: 550–560.

1. Huard, J., R. Roy, B. Guerette, S. Verreault, G. Tremblay, andJ. P. Tremblay. 1994. Human myoblast transplantation inimmunodeficient and immunosuppressed mice: Evidence ofrejection. Muscle Nerve 17: 224–234.

2. Huard, J., S. Verreault, R. Roy, M. Tremblay, and J. P. Tremblay.1994. High efficiency of muscle regeneration after humanmyoblast clone transplantation in SCID mice. J. Clin. Invest.93: 586–599.

3. Irintchev, A., M. Langer, M. Zweyer, R. Theisen, and A. Wernig.1997. Functional improvement of damaged adult mouse muscleby implantation of primary myoblasts. J. Physiol. 500.3: 775–

785.

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

4

4

4

4

4

4

4

4

5

5

5

5

5

30 SKUK ET AL.

4. Irintchev, A., M. Zweyer, and A. Wernig. 1995. Cellular andmolecular reactions in mouse muscles after myoblast implanta-tion. J. Neurocytol. 24: 1–13.

5. Ito, H., and J. P. Tremblay. 1998. Prior culture with concanava-lin A increases intramuscular migration of transplanted myo-blast. Muscle Nerve 21: 291–297.

6. Ito, H., J. T. Vilquin, D. Skuk, B. Roy, M. Goulet, S. Lille, F. J.Dugre, I. Asselin, R. Roy, M. Fardeau, and J. P. Tremblay. 1998.Myoblast transplantation in non-dystrophic dog. Neuromuscu-lar Disorders 8: 95–110.

7. Karpati, G., D. Ajdukovic, D. Arnold, R. B. Gledhill, R. Gutt-mann, P. Holland, P. A. Koch, E. Shoubridge, D. Spence, M.Vanasse, G. V. Watters, M. Abrahamamowicz, C. Duff, and R.Worton. 1993. Myoblast transfer in Duchenne muscular dystro-phy. Ann. Neurol. 34: 8–17.

8. Kinoshita, I., I. Asselin, J. T. Vilquin, J. Chamberlain, and J. P.Tremblay. 1998. Transplantation of myoblasts from a transgenicmouse overexpressing dystrophin produced only a relativelysmall increase of dystrophin-positive membrane. Muscle Nerve.21: 91–103.

9. Kinoshita, I., J. T. Vilquin, B. Guerette, I. Asselin, R. Roy, andJ. P. Tremblay. 1994. Very efficient myoblast allotransplantationin mice under FK-506 immunosupression. Muscle Nerve 17:1407–1415.

0. Kinoshita, I., J. T. Vilquin, C. Gravel, R. Roy, and J. P. Tremblay.1995. Myoblast allotransplantation in primates. Muscle Nerve18: 1217–1218.

1. Kinoshita, I., J. T. Vilquin, and J. P. Tremblay. 1995. Pretreat-ment of myoblast cultures with basic fibroblast growth factorincreases the efficacy of their transplantation in mdx mouse.Muscle Nerve 18: 834–841.

2. Kinoshita, I., R. Roy, F. J. Dugre, C. Gravel, B. Roy, M. Goulet, I.Asselin, and J. P. Tremblay. 1996. Myoblast transplantation inmonkeys: Control of immune response by FK506. J. Neuro-pathol. Exp. Neurol. 55: 687–697.

3. Kornegay, J. N., S. M. Prattis, D. J. Bogan, N. J. H. Sharp, R. J.Bartlett, H. S. Allamedine, and M. J. Dykstra. 1992. Results ofmyoblast transplantation in a canine model of muscle injury. InDuchenne Muscular Dystrophy: Animals Models and GeneticManipulation (B. A. Kakulas, J. M. Howell, and A. D. Roses,Eds.), pp. 203–212. . Raven Press, New York.

4. Law, P. K., T. G. Goodwin, and M. Wang. 1988. Normal myoblastinjections provide genetic treatment for murine dystrophy.Muscle Nerve 11: 525–533.

5. Law, P. K., T. E. Bertorini, T. G. Goodwin, M. Chen, Q. Fang,H.-J. Li, D. S. Kirby, J. A. Florendo, H. G. Herrod, and G. S.Golden. 1990. Dystrophin production induced by myoblasttransfer therapy in Duchenne muscular dystrophy. Lancet 336:114–115.

6. Manor, D., and M. Sadeh. 1989. Muscle fiber necrosis induced byintramuscular injection of drugs. Br. J. Exp. Pathol. 70: 457–462.

7. McArdle, A., R. H. Edwards, and M. J. Jackson. 1994. Release ofcreatine kinase and prostaglandin E2 from regenerating skel-etal muscle fibers. J. Appl. Physiol. 76: 1274–1278.

8. Mendell, J. R., J. T. Kissel, A. A. Amato, W. King, L. Signore,T. W. O. Prior, Z. Sahenk, S. Benson, P. E. Mcandrew, R. Rice, H.Nagaraja, R. Stephens, L. Lantry, G. E. Morris, and A. H. M.Burghes. 1995. Myoblast transfer in the treatment of Duch-

enne’s muscular dystrophy. New. Engl. J. Med. 333: 832–838.

9. Miller, R. G., K. R. Sharma, G. K. Pavlath, E. Gussoni, M.Mynhier, P. Yu, A. M. Lanctot, C. M. Greco, L. Steinman, andH. M. Blau. 1997. Myoblast implantation in Duchenne musculardystrophy: The San Francisco study. Muscle Nerve 20: 469–478.

0. Morgan, J. E., E. P. Hoffman, and T. A. Partridge. 1990. Normalmyogenic cells from newborn mice restore normal histology todegenerating muscles of the mdx mouse. J. Cell. Biol. 111:2437–2450.

1. Murry, C. E., R. W. Wiseman, S. M. Schwartz, and S. D.Hauschka. 1996. Skeletal myoblast transplantation for repair ofmyocardial necrosis. J. Clin. Invest. 98: 2512–2523.

2. Nakajima, J. 1993. The immunosuppressive effects of FK506 onsingle lung allotransplantation in Japanese monkey. NipponKyobu Geka Gakkai Zasshi 41: 1172–1183.

3. Olive, M., R. Rivera, and I. Ferrer. 1994. Parvalbumin immuno-cytochemistry and calcium deposition in muscle fiber necrosisand subsequent regeneration following intramuscular injectionof metoclopramide. Muscle Nerve 17: 494–499.

4. Partridge, T. A., J. E. Morgan, G. R. Coulton, E. P. Hoffman, andL. M. Kunkel. 1989. Conversion of mdx myofibers from dystro-phin-negative to -positive by injection of normal myoblasts.Nature 337: 176–179.

5. Pluskal, M. G., J. B. Harris, R. J. Pennington, and D. Eaker.1978. Some biochemical responses of rat skeletal muscle to asingle subcutaneous injection of a toxin (notexin) isolated fromthe venom of the Australian tiger snake Notechis scutatusscutatus. Clin. Exp. Pharmacol. Physiol. 5: 131–141.

6. Rando, T. A., and H. M. Blau. 1994. Primary mouse myoblastpurification, characterization, and transplantation for cell-mediated gene therapy. J. Cell. Biol. 125: 1275–1287.

7. Reid, H. A. 1961. Myoglobinuria and sea-snake-bite poisoning.Br. Med. J. (Clin. Res.) 1: 1284–1289.

8. Sharp, N. J. H., J. N. Kornegay, R. J. Bartlett, W. Y. Hung, andM. J. Dykstra. 1993. Notexin-induced muscle injury in the dog.J. Neurol. Sci. 116: 73–81.

9. Tremblay, J. P., F. Malouin, R. Roy, J. Huard, J. P. Bouchard, A.Sato, and C. Richards. 1993. Results of a triple blind clinicalstudy of myoblast transplantation without immunosupressivetreatment in young boys with Duchenne muscular dystrophy.Cell Transplant. 2: 99–112.

0. Van Meter, C. H. Jr., W. C. Claycomb, J. B. Delcarpio, D. M.Smith, H. Degruiter, F. Smart, and J. L. Ochsner. 1995. Myo-blast transplantation in the porcine model: A potential tech-nique for myocardial repair. J. Thorac. Cardiovasc. Surg. 110:1442–1448.

1. Vilquin, J. T., I. Asselin, B. Guerette, I. Kinoshita, S. Lille, R.Roy, and J. P. Tremblay. 1994. Myoblast allotransplantation inmice: The degree of success varies depending on the efficacy ofvarious immunosupressive treatments. Transpl. Proc. 26: 3372–3373.

2. Vilquin, J. T., I. Asselin, B. Guerette, I. Kinoshita, R. Roy, andJ. P. Tremblay. 1995. Successful myoblast allotransplantation inmdx mice using rapamycin. Transplantation 59: 422–426.

3. Vilquin, J. T., I. Kinoshita, R. Roy, and J. P. Tremblay. 1995.Cyclophosphamide immunosuppression does not permit success-ful myoblast allotransplantation in mouse. Neuromusc. Disord.6: 511–517.

4. Yang, C. C., and L. S. Chang. 1991. Dissociation of lethal toxicityand enzymic activity of notexin from Notechis scutatus scutatus(Australian-tiger-snake) venom by modification of tyrosine resi-

dues. Biochem. J. 280: 739–744.