676.full

Anti-TNF� (Remicade�) therapy protects dystrophicskeletal muscle from necrosis

MIRANDA D. GROUNDS1 AND JO TORRISISchool of Anatomy and Human Biology, The University of Western Australia,Western Australia, Australia 6009

ABSTRACT Necrosis of skeletal muscle fibers in thelethal childhood myopathy Duchenne muscular dystro-phy (DMD) results from defects in the cell membrane-associated protein, dystrophin. This study tests thenovel hypothesis that the initial sarcolemmal break-down resulting from dystrophin deficiency is exacer-bated by inflammatory cells and that cytokines, specif-ically tumor necrosis factor-� (TNF�), contribute tomuscle necrosis. To block in vivo TNF� bioactivity,young dystrophic mdx mice (a model for DMD) wereinjected weekly from 7 days of age with the anti-TNF�antibody Remicade� before the onset of muscle necro-sis and dystropathology that normally occurs at 21 dayspostnatally. The extent of inflammation, muscle necro-sis, and myotube formation was measured by histolog-ical analysis from 18 to 28 days and muscle damage wasalso visualized by penetration of Evans blue dye intomyofibers. Data from Remicade�-treated and controlmdx mice were compared with mdx/TNF�(�/�) micethat lack TNF�. Pharmacological blockade of TNF�activity with Remicade� clearly delayed and greatlyreduced the breakdown of dystrophic muscle, inmarked contrast to the situation in mdx and mdx/TNF�(�/�) mice. Remicade� had no adverse effecton new muscle formation. Remicade� is a highly spe-cific anti-inflammatory intervention, and clinical appli-cation to muscular dystrophies is suggested by thismarked protective effect against skeletal muscle break-down.—Grounds, M. D., Torrisi, J. Anti-TNF� (Remi-cade�) therapy protects dystrophic skeletal musclefrom necrosis. FASEB J. 18, 676–682 (2004)

Key Words: mdx mice � dystropathology � DMD

Duchenne muscular dystrophy (DMD) is an earlyonset lethal X-linked disorder that occurs in �1 in 3500boys, resulting in severe skeletal muscle degenerationand death, usually by 20 years of age (1). In DMD andthe mdx mouse model for DMD, defects in the subsar-colemmal protein dystrophin initiate muscle pathology(2, 3). There is no successful treatment for DMD. Muchresearch has focused on cell and gene therapies to tryand replace the defective dystrophin gene (4, 5). Acomplementary approach is to try to render the dystro-phic myofibers less susceptible to damage using boostergenes (6) to salvage the muscle integrity (3). Forexample, reduction of dystropathology has been dem-

onstrated in mdx mice by overexpression of IGF-1 (7),utrophin (8), cytotoxic T cell GalNAc transferase (9),calpastatin (10), integrin-� 7 (11), or ADAM-12 (12)and in the absence of myostatin (13, 14). Many of these(and other) observations are derived from experimentsusing genetically engineered mdx mice; difficultiesarise in translating these promising observations intothe clinical situation. Beyond these strategies designedto alter myofiber characteristics or extracellular inter-actions is the approach of targeting inflammatory cyto-kines; this is the focus of the present study.

Damage to the sarcolemma can result in small lesionsin the membrane that may normally be repaired rapidlyby “patching” to prevent myofiber breakdown, and ithas recently been demonstrated that dysferlin, thedefective gene in limb girdle type muscular dystrophy,is a key protein involved in such membrane resealing inskeletal muscle (15, 16). We hypothesize that in dystro-phic muscle, membrane damage caused by the genedefect can be exacerbated by proinflammatory cyto-kines, such as tumor necrosis factor-� (TNF�), that tipthe balance between patching of the minor lesion orfurther breakdown leading to focal necrosis of themyofiber. Strong evidence that inflammatory cells cancontribute to necrosis of healthy muscle cells comesfrom studies investigating the role of neutrophils, mac-rophages, and oxidative damage in vitro (17–20) and invivo (21, 22). It has been proposed that an inappropri-ate or excessive inflammatory response can directlydamage myofibers in myopathic conditions such asdystrophies or myositis (2, 23–25). TNF� is an early andpotent proinflammatory cytokine that stimulates theinflammatory response.

Even minor trauma to muscle will increase levels ofTNF� by release from mast cells; TNF� is also producedby neutrophils, macrophages and lymphocytes thataccumulate rapidly at the site of injury. TNF� increasesrapidly within damaged myofibers and is expressed bymyoblasts and myotubes (26–28). TNF� is greatly ele-vated in damaged myofibers in injured normal (26, 27)and myopathic skeletal muscle (29); it is chemotactic

1 Correspondence: School of Anatomy and Human Biol-ogy, The University of Western Australia, 35 Stirling Hwy.,Crawley, Western Australia, Australia 6009. E-mail:[email protected]

doi: 10.1096/fj.03-1024com

676 0892-6638/04/0018-0676 © FASEB

for myoblasts (30) and is mitogenic for satellite cells invivo, suggesting a direct role in myogenesis of regener-ating muscle (28).

The role of the cytokine TNF� in muscular dystrophyhas been investigated using mdx/TNF�(�/�) mice,and histopathological analysis has found that the ab-sence of TNF� in vivo resulted in equivocal findings asopposed to amelioration of muscle pathology as pre-dicted (31), although long-term deletion of TNF�appeared beneficial in older (12 months) mdx/TNF�(�/�) mice (32). Cytokine networks are com-plex, and when TNF� is removed from an in vivosystem, as in mdx/TNF�(�/�) mice, other proinflam-matory cytokines such as IL-12 and IFN� (33) may beup-regulated to overcome the TNF� deficiency. Theconfounding situation of altered cytokine profiles inTNF� knockout mice is further demonstrated by stud-ies of regenerating whole muscle grafts in these mice,where predictions of impaired skeletal muscle regener-ation in the absence of TNF� are not supported (26).

As an alternate approach, we have used the antibodyRemicade� to neutralize activity of the TNF� protein inmdx mice to assess whether blockade of TNF� inpatients with DMD has therapeutic value. It was ex-pected that this strategy to block TNF� activity in vivowould avoid the problems of compensatory up-regula-tion of other proinflammatory cytokines and silencethe inflammatory cell response. Strong support for thein vivo efficacy of Remicade� comes from histologicalanalysis of whole muscle autografts in C57Bl/10SnScmice (the parental strain for mdx) pretreated withRemicade�, where grafts examined at 5 days aftertransplantation contain very few inflammatory cells andconsist almost entirely of persisting necrotic muscletissue with no myotubes, in striking contrast to theadvanced regeneration seen in control grafts wherenecrotic tissue has been removed by inflammatory cellsand largely replaced by myotubes at 5 days (M. Groundset al., unpublished results). However, whereas inflam-matory cell infiltration and associated regeneration aredelayed, they are not prevented since new muscle isformed within 7 days in these Remicade�-treated mice.Clinically, the TNF� neutralizing antibody Remicade�

is highly effective at reducing symptoms of inflamma-tory diseases such as rheumatoid arthritis (34) andCrohn's disease (35), and is being extended to myositis.The clinical success of Remicade� makes it attractive asa potential drug to treat muscular dystrophies, and so itwas tested in the present study on dystrophic mdx micein vivo. It was hypothesized that TNF� plays a role inthe breakdown of dystrophic myofibers and that neu-tralizing TNF� activity in vivo will result in delayedonset and/or decreased intensity of dystropathology inmdx mice.

In mdx mice, the absence of dystrophin results in anabrupt onset of skeletal muscle necrosis at 21 days ofage (36, 37), and our studies focus on this early acutetransitory phase of dystropathology. This critical timeprovides a sensitive assay for factors that shift thethreshold of breakdown of dystrophic muscle. Muscle

breakdown peaks at 28 days, then decreases markedlyto stabilize around 12 wk of age to a relatively low levelof damage (37), with symptoms of dystropathologybeing cumulative and more pronounced in older (15months) mdx mice (38, 39). Any delay in the acuteonset of damage at 19–21 days, indicating a shift in thethreshold of muscle breakdown, is readily detected.The onset of damage is assessed histologically (evi-dence of necrosis and myotubes formed �3 days aftermuscle breakdown) and by injection of Evans blue dye(EBD) as a sensitive and early marker of muscle dam-age/leakiness (40, 41). Although it is often stated thatthe marked difference in severity of pathology betweenmdx mice and DMD boys is due to a greater capacity forregeneration in the mice, it would seem that thereduced muscle breakdown in mature mdx mice (afterthe transitory acute phase has ceased) may be the keyissue to consider. For if muscle necrosis does not occur,regeneration is not required.

MATERIALS AND METHODS

Overview

Mdx mice were injected intraperitoneally once weekly fromday 7 after birth with 10 �g (1 �L) Remicade� (Schering-Plough Pty. Ltd., Sydney, Australia) per gram body weight orequivalent volume of control mouse serum albumin (MSA);all mice were weighed before injection. Remicade�-treatedand control (MSA) mdx mice were also compared withuntreated mdx/TNF�(�/�) mice. At 24 h before sampling,mice were intraperitoneally injected with 1% Evans blue dye(EBD) at 1% volume relative to body mass to identify dam-aged/leaky sarcolemmal membranes (41). Mice were killed(by halothane anesthesia) and TA muscles sampled from 18to 28 days of age and at 12 wk. The TA muscles were eitherfrozen or taken for paraffin processing. Frozen sections wereanalyzed for EBD-positive myofibers. Hematoxylin and eosin(H&E) -stained paraffin sections were analyzed histologicallyfor inflammatory cell infiltration, myofiber necrosis, andmyotube formation and, in 12 wk samples, for myofibers withcentral nuclei.

Mice and treatment groups

Inbred mdx and mdx/TNF�(�/�) mice were used in thisinvestigation. Mdx mice for TNF� neutralization were housedand treated according to the Western Australian Preventionof Cruelties to Animals Act, the National Health and MedicalResearch Council, and the University of Western AustraliaAnimal Ethics Committee. Remicade�-treated mdx mice weresampled 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, and 28 days afterbirth with numbers of mice being 4, 2, 4, 8, 5, 4, 3, 9, 7, 7, and4, respectively, for each time point. Control MSA-injectedmdx mice were sampled 19, 20, 21, 22, 23, 24, 25, 26, and 28days after birth with numbers of mice being 3, 4, 4, 5, 2, 5, 2,4, and 4, respectively, for each sample point. Four additionalmice from both treatment groups were sampled at 12 wk.

Mdx/TNF�(�/�) and control mdx mice for TNF� geneknockout analysis were housed and sampled at the Universityof California, Los Angeles, using guidelines for animal use inthe United States. Duplicate mdx/TNF�(�/�) and controlmdx mice were sampled 19, 20, 21, 22, 24, 26, and 28 daysafter birth.

677ANTI-TNF� (REMICADE�) THERAPY

Histological assay

Inflammatory cell infiltration, muscle necrosis, and myotubeformation were analyzed using H&E stained 5 �m frozenand/or paraffin sections using light microscopy. TA musclefor each parameter was examined in transverse section usingthe following scale: 1 indicating 1–5% of total tissue areaaffected, 2 representing 6–15%, 3 for 16–30%, 4 for 31–60%,and 5 for 61–100%. Inflammatory cell infiltration was identi-fied by cells with basophilic nuclear staining and little cyto-plasm located within the endomysium and/or necrotic tissue.Muscle necrosis was identified by breakdown of sarcolemmaand fragmented sarcoplasm, and myotubes (indicating regen-eration) were identified as plump cells with central nuclei. Inolder (12 wk) mice, myofibers with persisting central nucleiare a cumulative index of previous cycles of necrosis/regen-eration (39).

RESULTS

Overall, there was no significant difference in weightgain between Remicade� injected and control MSAinjected mdx mice during the study period. The meangram body weight (and standard error) for the Remi-cade� and control mdx mice, respectively, was 7.25(�/�0.07) and 8.04 (�/�1.83) on day 7, 8.25(�/�0.22) and 9.26 (�/�2.27) on day 14, 10.27(�/�0.30) and 6.62 (�/�1.79) on day 21, 14.44(�/�0.83) and 10.9 (�/�2.05) on day 28, and 32.03(�/�0.41) and 30.83 (�/�0.6) at 12 wk. Only day 21

showed significant difference between the two groups,with slightly higher values for the Remicade�-treatedmice (t test P�0.05). Therefore, neither the growthrate of the young mdx mice nor the adult body weightswere reduced by Remicade� treatment.

The histological appearance of muscles on days 21and 24 is shown in Fig. 1; the overall patterns ofinflammatory cell infiltration, necrosis and myotubeformation from days 18 to 28, for control (MSA) mdx,Remicade�-treated mdx and mdx/TNF�(�/�) mus-cles are summarized in Fig. 2. Detailed quantitation ofthe tissue area occupied by muscle necrosis and myo-tubes at 21 and 28 days of age for control (MSA) andRemicade�-treated mdx mice (n�4 mice for each sam-ple) is presented in Fig. 3: these data strongly supportthe semiquantitative analysis in Fig. 2. Inflammatorycells were conspicuous in control (MSA) mdx musclesat 20 days (Fig. 2a) and pronounced at 21 days through-out 50% of the mdx tissue section (Fig. 1a, Fig. 2a).The marked increase in inflammatory cells at 20 dayspreceded by 1 day the acute phase of necrosis thatoccupied up to 30% of the tissue section on day 21 (Fig.1a, Fig. 2a). In marked contrast, in Remicade�-treatedmdx mice, there was little inflammation and minimalnecrosis (0–5%) at these times (Fig. 1c, Fig. 2b). Themaximum extent of necrosis in control mdx mice (seenon day 21) was much greater than in Remicade�-treatedmdx mice (seen on day 25) (Figs. 1 and 2) and theextent of inflammatory cell infiltration were consis-

Figure 1. Histological appearance of TA muscles frommice aged 21 and 24 days after birth. a, b) Control(MSA injected) mdx mice, c, d) Remicade�-treatedmdx mice and e, f) mdx/TNF�(�/�) mice. In con-trol mdx mice on a) day 21 necrosis was predominantand on b) day 24 myotubes were prominent (b, inset:high magnification shows myotubes with central nu-clei). In contrast, necrosis is not marked in Remi-cade�-treated mice on c) day 21 but is apparent at d)day 24. The mdx/TNF�(�/�) mice at e) days 21 andf) 24 were similar to control (MSA injected) mdxmice. Asterisks indicate sites of necrosis. Arrow indi-cates myotubes. Scale bar, 100 �m.

678 Vol. 18 April 2004 GROUNDS AND TORRISIThe FASEB Journal

tently lower in Remicade�-treated mice (Fig. 2b) than incontrol mdx mice (Fig. 2a). After the initial acute peakof necrosis, there appeared to be a lull in muscledamage; such a cyclic pattern of necrosis/regenerationwas also noted in earlier studies in mdx mice (36). Thepattern of muscle breakdown in Remicade�-treatedmice was markedly different from that seen in mdx/

TNF�(�/�) mice, where the overall extent and pat-tern of damage resembled the control mdx mice (al-though the peak of inflammation and necrosis wasdelayed by several days) (Fig. 2, compare panels b, c,and a). The striking delay in onset of muscle break-down in the Remicade�-treated mdx mice was furthersupported by analysis of EBD staining in TA musclesfrom 22-day-old mice, where many EBD-positive myofi-bers (indicating membrane damage) were conspicuousin control mdx mice; such myofibers but were rare inthe Remicade�-treated mice (Fig. 4). Statistical analysisof the quantitative data related to the area occupied bynecrotic tissue (Fig. 3) shows highly significant(P�0.05) reduced necrosis in Remicade�-treated dys-trophic muscles at 21 days compared with control(MSA) mdx muscles.

The onset of necrosis was followed �3 days later bythe appearance of myotubes, as expected, since myo-tubes are not formed until 2–3 days after regenerationstarts (42). In control mdx mice (Fig. 2a), myotubeswere dramatically increased on day 24, 3 days after thepeak of necrosis (at 21 days). In Remicade�-treatedmdx mice, myotube numbers increased gradually afterday 24 and the rise in numbers seen at 27 days with afurther increase at 28 days is 3–4 days after (the muchsmaller) peaks of necrosis evident at 23 and 25 days.The timing of myotube formation (with respect tomyofiber necrosis) appears normal and there is no hintof impaired myogenesis in the Remicade�-treated mdxmice. This is further endorsed by the many regenerateddystrophic myofibers (identified by central myonuclei)in 12-wk-old Remicade�-treated mdx TA muscle

In 12-wk-old mice, the Remicade�-treated musclesgenerally had features similar to control mdx muscleswith many central nuclei and small areas of necrosissometimes associated with small myotubes. The extentof necrosis was low and similar in muscles of bothgroups, being only 0.185% (�/�0.19) and 0.245%(�/�0.294), respectively, for control mdx and Remi-cade�-treated mice. There was no difference in EBDstaining between control and Remicade�-treated mdxmice with only a few individual or small groups ofEBD-positive myofibers in TA muscle sections; these

Figure 2. Comparison of graphs summarizing the time courseand extent of phagocyte infiltration, necrosis, and myotubeformation in a) control (MSA injected) mdx mice, b) mdxmice treated with Remicade�, and c) mdx/TNF�(�/�) mice.No difference in the relative time of onset of myotubeformation in response to necrosis was seen between thesegroups. Each time point shows the average grading for all TAmuscles (muscle from both legs was analyzed, n�2, with totaln ranging from 4 to 16; see details in Materials and Methods).

Figure 3. Quantitative analysis of the ex-tent of necrosis and of myotube forma-tion in TA muscles (n�4 mice) fromcontrol and Remicade�-treated mice areshown for samples at 21 and 28 days.These data strongly endorse the semi-quantitative data in Fig. 2. *The lack ofnecrosis at 21 days in Remicade�-treatedcompared with control mdx mice ishighly significant (P�0.05).


stained myofibers generally corresponded to necroticfoci visualized by H&E staining. To test the possibilitythat the size of necrotic lesions might be more limitedin Remicade�-treated mice, the area of all necrotic fociwas measured at 28 days and 12 wk of age in controlmdx and Remicade�-treated mdx mice. Visually therewas no striking difference in necrotic foci size betweenthe two groups of mice. Morphometric analysis (Im-agePro Plus 4.0 software, Microsoft) showed that at 28days the mean size of the necrotic foci (expressed aspercentage of total section area) was small being only0.08% (�/�0.008%) (n�5) and 0.19% (�/�0.12%)(n�4) for Remicade�-treated and control mdx mice,respectively; the relatively high variation in the 28 daycontrols was accounted for by one of the four micehaving significantly larger necrotic foci (maximumvalue was 1.04%) than the other three mice (maximumof 0.32%). At 12 wk, the mean size of the necrotic fociwas 0.02% (�/�0.001) (n�4) and 0.04 (�/�0.014)(n�5) for Remicade�-treated and control mdx mice,respectively. Whereas these appeared to be �fourfoldsmaller than at 28 days, there was no significant differ-ences overall between the two times of examination(P�0.05). The trend for larger necrotic foci in controlmdx vs. Remicade�-treated mice at both time points wasnot statistically significant (t test analysis, P�0.05).

Further quantitative analysis of the area occupied bymyofibers with central myonuclei (a cumulative mea-sure of muscle damage and repair) indicated less

affected muscle in the Remicade�-treated (53.31%�/�14.847) than control (67.11% �/�17.03) mdxmice. The trend (only four mice were analyzed for eachsample) supports the proposal that Remicade� treat-ment protects dystrophic muscle from necrosis andreduces the dystropathology; however, this apparentdifference was not statistically significant.

DISCUSSION

The reduced dystropathology in Remicade�-treated mdxmice in the acute phase from 21 to 28 days supports thehypothesis that TNF� contributes to the early breakdownof dystrophic muscle. It is concluded that although a lackof dystrophin results in sarcolemmal weakness leading toinitial breakdown of dystrophic muscle (2), the extent ofnecrosis is exacerbated by TNF�, which likely triggers theassociated inflammatory cascade (23, 25).

The reduced inflammatory response did not appear tohave any adverse effects on regeneration since myotubeformation was not impaired in Remicade�-treated mdxmice in response to necrosis during the acute phase. Thatnew muscle can be readily formed in Remicade�-treatedmice was further confirmed by the presence of manyregenerated myofibers (identified by the presence ofcentral nuclei) in TA muscles from 12 wk mdx mice. Themarked inhibition of inflammation (with an associateddelay in regeneration) seen in our whole muscle grafts at5 days in Remicade�-treated C57Bl/10Sn mice (data notshown) did not impair muscle formation at later times.This graft model is a useful biological assay to demon-strate that this regime of Remicade� does indeed reduceinflammation in mice in vivo. However, the large mass ofnecrotic avascular tissue in the graft is an extreme situa-tion compared with the relatively small foci of damagenear a good blood supply that occurs in dystrophicmuscle, where probably few inflammatory cells are re-quired to enable regeneration to proceed. The effect ofRemicade� on whole muscle grafts contrasts with anotherstudy, where TNF� neutralizing antibody appeared tohave no effect on the early histological events of regener-ation in response to freeze injury in C57Bl/6 mice; thispattern corresponded to results when TNF� function wasblocked in TNF� double receptor knockout mice (43),although some changes in gene expression and musclestrength were reported. Clearly, the precise tool used toblock TNF� function is of critical importance.

A striking effect was seen with Remicade� during theearly acute phase of muscle necrosis (3–4 wk) indystrophic mice. Although the data at 12 wk suggestedthat sustained Remicade� treatment reduced the dys-tropathology of TA muscles over a longer period oftime, this effect was not statistically significant. Al-though it is widely considered that a cycle of necrosisstabilizes and protects a dystrophic myofiber fromsubsequent necrosis, this is clearly not the case (44).Therefore, whereas the presence of central myonucleiwithin a myofiber indicates that regeneration has oc-curred, it does not indicate how many times this

Figure 4. Evans blue dye (EBD) staining as a marker of muscledamage is shown in TA muscles of day 22 a) control (MSAinjected) and b) Remicade�-treated mdx mice. On day 22after birth, the TA muscle of a) control mdx mice showmassive necrosis and are strongly positive for EBD. This is incontrast to b) Remicade�-treated mice where the onset ofnecrosis is delayed and only a few myofibers show weakstaining. Scale bar, 50 �m.


individual myofiber has been subjected to repeatednecrosis/regeneration, and this should be taken intoaccount. Furthermore, although the presence of EBDwithin myofibers certainly indicates sarcolemmal dam-age and leakiness, there is not always a close correlationwith histological myofiber necrosis (Shavlakadze et al.,unpublished results) (40, 41). There is also greatvariation between the pathology of different musclesand individual mdx mice (40, 44, 45), and thereforea detailed study of many muscles up to 20 wk of ageis in progress to thoroughly assess the impact ofRemicade�. On the basis of this available informationin the mdx mouse (where the pathology is mild), it isdifficult to predict the potential overall benefit ofRemicade� on the severe pathology seen in the dogmodels of DMD (46, 47) or in clinical dystrophiesand dysferlinopathies (48).

The reduced breakdown of dystrophic muscle withRemicade� treatment suggests that this highly specificanti-inflammatory drug may be a useful clinical inter-vention to treat DMD and other myopathies with apronounced inflammatory component such as dysferli-nopathies (15, 48). The high specificity of Remicade� isadvantageous compared with the existing use of anti-inflammatory corticosteroids such as Prednisolone andDeflazacort to treat DMD as these steroids are associ-ated with severe adverse side effects such as weight gainand osteoporosis (49, 50), and steroids should beavoided for dysferlinopathies due to unrecoverable lossof strength (48). The precise mode of action of thesesteroidsOwhether they act by preventing muscle break-down, suppressing inflammation, increasing the sizeand strength of myofibers, enhancing regeneration, orreducing fibrosisOremains unclear (50). In contrast,the precise mode of action of Remicade� is known: itdoes not cause generalized suppression of the immunesystem and has been used clinically to treat inflamma-tory disorders (34). Although it can have mild adverseeffects such as rash and infusion-related reactions,there are few instances of severe reactions such as sepsisor pneumonia (35, 51). Remicade� has been used inpediatric patients to treat disorders such as juvenileidiopathic arthritis (52, 53), and Crohn's disease (54,55) in children as young as 3 years (53), and children asyoung as 18 months are now being treated.

The present study shows that treatment of mdxmice with Remicade� reduces the onset and intensityof necrosis and the dystropathology. This presents anew approach for using highly specific anti-cytokinetherapies to treat DMD. Further studies are requiredto evaluate long-term benefits of Remicade� treat-ment and to compare them with other clinicallyrelevant anti-cytokine drugs (34) such as the use ofsoluble receptors that bind TNF� (Enbrel�) or inter-leukin-1 (56, 57). This striking protective effect com-bined with high specificity and few serious adverseside effects for Remicade� makes this simple strategya serious contender for the clinical treatment ofdystropathologies.

We thank Dr. Melissa Spencer and Alexandra Potts of theDepartment of Pediatrics and Duchenne Muscular DystrophyResearch Centre, University of California, Los Angeles, USA, forproviding all of the muscle samples from the mdx/TNF�(�/�)mice for this analysis and for critical comments on the manu-script. The technical assistance and helpful comments of Mari-lyn Davies, Jason White, Monique Berendse, and StuartHodgetts (UWA) are gratefully acknowledged.

REFERENCES

1. Emery, E. H. (2002) The muscular dystrophies. Lancet 359,687–695

2. Porter, J. D., Khanna, S., Kaminski, H. J., Rao, J. S., Merriam,A. P., Richmonds, C. R., Leahy, P., Li, J., Guo, W., and Andrade,F. H. (2002) A chronic inflammatory response dominates theskeletal muscle molecular signature in dystrophin-deficient mdxmice. Hum. Mol. Genet. 11, 263–272

3. Tseng, B. S., Zhao, P., Pattison, J. S., Gordon, S. E., Granchelli,J. A., Madsen, R. W., Folk, L. C., Hoffman, E. P., and Booth,F. W. (2002) Regenerated mdx mouse skeletal muscle showsdifferential mRNA expression. J. Appl. Physiol. 93, 537–545

4. Wells, D. J., Ferrer, A., and Wells, K. E. (2002) Immunologicalhurdles in the path to gene therapy for Duchenne musculardystrophy. Exp. Rev. Mol. Med. 4, 1–23

5. Skuk, D., and Tremblay, J. P. (2003) Myoblast transplantation:the current status of a potential therapeutic tool for myopathies.J. Musc. Res. Cell Motil. 24, 285–300

6. Engvall, E., and Wewer, U. M. (2003) The new frontier inmuscular dystrophy research: booster genes. FASEB J. 17, 1579–1584

7. Barton, E. R., Morris, L., Musaro, A., Rosenthal, N., andSweeney, H. L. (2002) Muscle-specific expression of insulin-likegrowth factor I counters muscle decline in mdx mice. J. Cell Biol.157, 137–148

8. Khurana, T. S., and Davies, K. E. (2003) Pharmacologicalstrategies for muscular dystrophy. Nat. Rev. Drug Discov. 2,379–390

9. Nguyen, H. H., Jayasinha, V., Xia, B., Hoyte, K., and Martin,P. T. (2002) Overexpression of the cytotoxic T cell GalNActransferase in skeletal muscle inhibits muscular dystrophy inmdx mice. Proc. Natl. Acad. Sci. USA 99, 5616–5621

10. Spencer, M. J., and Mellgren, R. L. (2002) Overexpression of acalpastatin transgene in mdx muscle reduces dystrophic pathol-ogy. Hum. Mol. Genet. 11, 2645–2655

11. Burkin, D. J., Wallace, G. Q., Nicol, K. J., Kaufman, D. J., andKaufman, S. J. (2001) Enhanced expression of the a7B1 integrinreduces muscular dystrophy and restores viability in dystrophicmice. J. Cell Biol. 152, 1207–1218

12. Moghadaszadeh, B., Albrechtsen, R., Guo, L., Zaik, M., Kawagu-chi, N., and Borup, R. H. FKronquist, P., Schroder, H. D.,Davies, K. E., Voit, T., Neielson, F. C., Engvall, E., and Wewer,U. M. (2003) Compensation for dystrophin-deficiency: ADAM12 overexpression in skeletal muscle results in increased {alpha}7 integrin, utrophin, and associated glycoproteins. Hum. Mol.Genet. 12, 2467–2479

13. Wagner, K. R., McPherron, A. C., Winik, N., and Lee, S. J.(2002) Loss of myostatin attenuates severity of muscular dystro-phy in mdx mice. Ann. Neurol. 52, 832–836

14. Bogdanovich, S., Krag, T. O., Barton, E. R., Morris, L. D.,Whittemore, L. A., Ahima, R. S., and Khurana, T. S. (2002)Functional improvement of dystrophic muscle by myostatinblockade. Nature (London) 420, 418–421

15. Bansal, D., Miyake, K., Vogel, S. S., Groh, S., Chen, C. C.,Williamson, R., McNeil, P. L., and Campbell, K. P. (2003)Defective membrane repair in dysferlin-deficient muscular dys-trophy. Nature (London) 423, 168–172

16. Doherty, K. R., and McNally, E. M. (2003) Repairing the tears:dysferlin in muscle membrane repair. Trends Mol. Med. 9,327–330

17. Pizza, F. X., McLoughlin, T. J., McGregor, S. J., Calomeni, E. P.,and Gunning, W. T. (2001) Neutrophils injure cultured skeletalmyotubes. Am. J. Physiol. 281, C335–C341


18. Xing, J., Yu, Y., and Rando, T. A. (2002) The modulation of cellularsusceptibility to oxidative stress: protective and destructive actionsof Cu,Zn-superoxide dismutase. Neurobiol. Dis. 10, 234–246

19. Nguyen, H. X., and Tidball, J. G. (2003) Interactions betweenneutrophils and macrophages promote macrophage killing ofrat muscle cells in vitro. J. Physiol. 547, 125–132

20. McLoughlin, T. J., Tsivitse, S. K., Edwards, J. A., Aiken, B. A.,and Pizza, F. X. (2003) Deferoxamine reduces and nitric oxidesynthase inhibition increases neutrophil-mediated myotube in-jury. Cell Tissue Res. In press

21. Nguyen, H. X., and Tidball, J. G. (2003) Expression of amuscle-specific, nitric oxide synthase transgene prevents musclemembrane injury and reduces muscle inflammation duringmodified muscle use in mice. J. Physiol. 550, 347–356

22. Cheung, E. V., and Tidball, J. G. (2003) Administration of thenon-steroidal anti-inflammatory drug ibuprofen increases mac-rophage concentrations but reduces necrosis during modifiedmuscle use. Inflamm. Res. 52, 170–176

23. Tews, D. S., and Goebel, H. H. (1998) Cell death and oxidativedamage in inflammatory myopathies. Clin. Immunol. Immuno-pathol. 87, 240–247

24. Lundberg, I. E. (2000) The role of cytokines, chemokines, andadhesion molecules in the pathogenesis of idiopathic inflamma-tory myopathies. Curr. Opin. Rheumatol. 2, 216–224

25. Spencer, M. J., and Tidball, J. G. (2001) Do immune cellspromote the pathology of dystrophin-deficient myopathies?Neuromusc. Disorders 11, 556–564

26. Collins, R. A., and Grounds, M. D. (2001) The role of tumournecrosis factor-alpha (TNF-�) in muscle regeneration: studies inTNF�(�/�) and TNF�(�/�)/LT�(�/�) mice. J. Histochem.Cytochem. 49, 989–1001

27. Warren, G. L., Hulderman, T., Jensen, N., McKinstry, M.,Mishra, M., Luster, M. I., and Simeonova, P. P. (2002) Physio-logical role of tumor necrosis factor alpha in traumatic muscleinjury. FASEB J. 16, 1630–1632

28. Li, Y.-P. (2003) TNF-� is a mitogen in skeletal muscle. Am. J.Physiol. 285, C370–C376

29. Kuru, S., Inukai, A., Kato, T., Liang, Y., Kimura, S., and Sobue,G. (2003) Expression of tumor necrosis factor-alpha in regen-erating muscle fibers in inflammatory and non-inflammatorymyopathies. Acta Neuropathol. (Berlin) 105, 217–224

30. Torrente, Y., El Fahime, E., Carson, N. J., Del Bo, R., Belicchi,M., Pisati, F., Tremblay, J. P., and Bresolin, N. (2003) Tumornecrosis factor-alpha (TNF-alpha) stimulates chemotactic re-sponse in mouse myogenic cells. Cell Transplant. 12, 91–100

31. Spencer, M. J., Marino, M. W., and Winckler, W. M. (2000)Altered pathological progression of diaphragm and quadricepsmuscle in TNF-deficient, dystrophin-deficient mice. Neuromusc.Disord. 10, 612–619

32. Gosselin, L. E., Barkley, J. E., Spencer, M. J., McCormick, K. M.,and Farkas, G. A. (2003) Ventilatory dysfunction in mdx mice:impact of tumor necrosis factor-alpha deletion. Muscle Nerve Inpress

33. Hodge-Dufour, J., Marino, M. W., Horton, M. R., Jungbluth, A.,Burdick, M. D., Strieter, R. M., Noble, P. W., Hunter, C. A., andPure, E. (1998) Inhibition of interferon y induced interleukin12 production: a potential mechanism for the anti-inflammatoryactivities of tumor necrosis factor. Proc. Natl. Acad. Sci. USA 95,13806–13811

34. Gabay, C. (2002) Cytokine inhibitors in the treatment ofrheumatoid arthritis. Expert Opin. Biol. Ther. 2, 135–149

35. van Deventer, S. J. H. (1999) Review article: targetting TNF� asa key cytokine in the inflammatory processes of Crohn'sdiseaseOthe mechanisms of action of infliximab. Aliment. Phar-macol. Ther. 13, 3–8

36. Beilharz, M. W., Lareu, R. R., Garrett, K. L., Grounds, M. D., andFletcher, S. (1992) Quantitation of muscle precursor cell activityin skeletal muscle by northern analysis of MyoD and myogeninexpressionOapplication to dystrophic (mdx) mouse muscle.Mol. Cell. Neurosci. 3, 326–331

37. McGeachie, J. K., Grounds, M. D., Partridge, T. A., and Morgan,J. E. (1993) Age-related changes in replication of myogenic cellsin mdx mice: quantitative autoradiographic studies. J. Neurol.Sci. 119, 169–179

38. Lefaucheur, J. P., Pastoret, C., and Sebille, A. (1995) Phenotypeof dystrophinopathy in old mdx mice. Anat. Rec. 242, 70–76

39. Coulton, G. R., Morgan, J. E., Partridge, T. A., and Sloper, J. C.(1988) The mdx mouse skeletal muscle myopathy: I. A histolog-ical, morphometric and biochemical investigation. Neuropathol.Appl. Neurobiol. 14, 53–70

40. Straub, V., Rafael, J. A., Chamberlain, J. S., and Campbell, K. P.(1997) Animal models for muscular dystrophy show differentpatterns of sarcolemmal disruption. J. Cell Biol. 139, 375–385

41. Hamer, P. W., McGeachie, J. M., Davies, M. J., and Grounds,M. D. (2002) Evans blue dye as an in vivo marker of myofibrefragility and damage: optimising parameters for detecting initialmyofibre permeability. J. Anat. 200, 69–79

42. Mitchell, C. A., McGeachie, J. K., and Grounds, M. D. (1992)Cellular differences in the regeneration of murine skeletalmuscle: a quantitative histological study in SJL/J and BALB/cmice. Cell Tissue Res. 269, 159–166

43. Warren, G. L., Hulderman, T., Jensen, N., McKinstry, M.,Mishra, M., Luster, M. I., and Simeonova, P. P. (2002) Physio-logical role of tumor necrosis factor a in traumatic muscleinjury. FASEB J. 16, 1630–1632

44. McGeachie, J. K., and Grounds, M. D. (1999) The timingbetween skeletal muscle myoblast replication and fusion intomyotubes, and the stability of regenerated dystrophic myofibres:an autoradiographic study in mdx mice. J. Anat. 194, 287–295

45. Straub, V., Donahue, K. M., Allamand, V., Davisson, R. L., Kim,Y. R., and Campbell, K. P. (2000) Contrast agent-enhancedmagnetic resonance imaging of skeletal muscle damage inanimal models of muscular dystrophy. Magn. Reson. Med. 44,655–659

46. Cozzi, F., Cerletti, M., Luvoni, G. C., Lombardo, R., Brambilla,P. G., Faverzani, S., Blasevich, F., Cornelio, F., Pozza, O., andMora, M. (2001) Development of muscle pathology in canineX-linked muscular dystrophy. 11. Quantitative characterizationof histopathological progression during postnatal skeletal mus-cle development. Acta Neuropathol. (Berlin) 101, 469–478

47. Schatzberg, S. J., Olby, N. J., Breen, M., Anderson, L. V.,Langford, C. F., Dickens, H. F., Wilton, S. D., Zeiss, C. J., Binns,M. M., Kornegay, J. N., Morris, G. E., and Sharp, N. J. (1999)Molecular analysis of a spontaneous dystrophin "knockout" dog.Neuromusc. Disord. 9, 289–295

48. Hoffman, E. P., Rao, D., and Pachman, L. M. (2002) Clarifyingthe boundaries between the inflammatory and dystrophic my-opathies: insights from molecular diagnostics and microarrays.Rheumat. Dis. Clin. N. Am. 28, 743–757

49. Wong, B. L., and Christopher, C. (2002) Corticosteroids inDuchenne muscular dystrophy: a reappraisal. J. Child Neurol. 17,183–190

50. Muntoni, F., Fisher, I., Morgan, J. E., and Abraham, D. (2002)Steroids in Duchenne muscular dystrophy: from clinical trials togenomic research. Neuromusc. Disord. 12, S162–S165

51. Elliott, M. J., Maini, R. N., Feldmann, M., Kalden, J. R., Antoni, C.,Smolen, J. S., Leeb, B., Breedveld, F. C., Macfarlane, J. D., Bijl, H.,et al. (1994) Randomised double-blind comparison of chimericmonoclonal antibody to tumour necrosis factor alpha (cA2) versusplacebo in rheumatoid arthritis. Lancet 344, 1105–1110

52. Billiau, A. D., Cornillie, F., and Wouters, C. (2002) Infliximabfor systemic onset juvenile idiopathic arthritis: experience in 3children. J. Rheumatol. 29, 1111–1114

53. Lahdenne, P., Vahasalo, P., and Honkanen, V. (2003) Inflix-imab or etanercept in the treatment of children with refractoryjuvenile idiopathic arthritis: an open label study. Ann. RheumaticDis. 62, 245–247

54. Serrano, M. S., Schmidt-Sommerfeld, E., Kilbaugh, T. J., Brown,R. F., Udall, J. N., Jr., and Mannick, E. E. (2001) Use ofinfliximab in pediatric patients with inflammatory bowel dis-ease. Ann. Pharmacol. 35, 823–828

55. Hyams, J. S., Markowitz, J., and Wyllie, R. (2000) Use ofinfliximab in the treatment of Crohn's disease in children andadolescents. J. Pediatr. 137, 192–196

56. Arend, W. P. (2002) The mode of action of cytokine inhibitors.J. Rheumatol. Suppl. 65, 16–21

57. Economides, A. N., Carpenter, L. R., Rudge, J. S., Wong, V.,Koehler-Stec, E. M., Hartnett, C., Pyles, E. A., Xu, X., Daly, T. J.,Young, M. R., et al. (2003) Cytokine traps: multi-component,high-affinity blockers of cytokine action. Nat. Med. 9, 47–52

Received for publication September 24, 2003.Accepted for publication December 10, 2003.


Documents

676.full