Scientific Problems with Animal Models of Duchenne
Muscular Dystrophy
Eric Dunayer, V.M.D
Introduction: The Human Disease
Muscular dystrophy is a disorder characterized by progressive muscle atrophy. In humans, the most common form is Duchenne muscular dystrophy (DMD), which is linked to a recessive gene on the X chromosome, afflicts approximately 1 in 3,500 males,(1) and shows no racial or ethnic specificity.(2)
Diagnosis of DMD is based on increased levels of serum creatine phosphokinase (an enzyme, normally present in muscle, that leaks into the blood from damaged muscle) and a muscle biopsy that reveals a combination of necrotic, regenerating, and fibrotic tissue. Initial signs of muscle weakness usually appear as difficulty in walking or running, at two to three years of age. Cardiac muscle may also be affected. In addition, DMD patients may suffer from mental retardation.(1,2) In all DMD patients, skeletal muscle undergoes a chronic cycle of degeneration and regeneration until fibrosis has destroyed the muscle's ability to regenerate.(3) Progressive muscle atrophy usually leads to wheelchair confinement by age twelve. DMD patients generally die in their late teens or early twenties due to respiratory failure.(2)
The primary genetic defect in DMD is a mutation in the gene that codes for the structure of the protein dystrophin.(4) The dystrophin gene, located on the X chromosome, is one of the most unstable genes in humans; 30% of DMD cases arise from a de novo mutation of the gene in patients with no familial history of muscular dystrophy.(5)
In normal humans, the gene is expressed in all muscle types, although dystrophin levels in skeletal and cardiac muscle are much higher than in smooth muscle;(6) dystrophin is attached to the interior of muscle plasma membrane.(5) Dystrophin also exists in nervous tissue in large quantity.(6) Dystrophin's function in muscle tissue is yet to be elucidated, but it may relate to muscle calcium homeostasis;(4) dystrophin's function in nervous tissue is unknown.(5) In DMD, dystrophin is greatly reduced, often to undetectable levels.(1)
Animal Models: Pathologies Induced by Surgery and/or Drugs
Experimenters have created non-inherited pathologies in nonhuman animals as "models" of DMD. One such model involves aortic ligation followed by treatment with a vasopressor. Like DMD, this experimental procedure causes muscle necrosis followed by regeneration. Another model involves use of imprimine to induce necrosis. Neither model has gained wide acceptance.(7) While mimicking DMD's gross pathological changes, such models cannot elucidate DMD's etiology since they are not based on inheritance and inadequate dystrophin.
Animal Models: Inherited Muscular Dystrophies
To date, most animal models of DMD have been inherited muscular dystrophies in mice, hamsters, chickens, and dogs. All of these models have had serious flaws. As J. Elbrink et al. have pointed out, there are major problems in extrapolating from a nonhuman animal model to a human genetic muscular disease. For one thing, even if symptoms in the model and the naturally occurring human disease resemble each other, the genetic defects may differ. In addition, muscles with corresponding anatomical locations in their respective species may nevertheless have very different fiber compositions, metabolisms, and functions, particularly in bipeds versus quadrupeds.(7)
Differences Between DMD and Animal Models Based on Autosomal Inheritance
Although DMD is linked to a recessive gene on the X chromosome, three of the most commonly used animal models of DMD are diseases inherited as autosomal recessive traits: in dy/dy2j mice, BIO 40.45 strain Syrian hamsters, and New Hampshire strain chickens. Because of the non-X-linked nature of these diseases, the primary defect cannot be a lack of dystrophin, as in DMD. As pathologist B. J. Cooper has noted, "Because of the strong conservation of genetic information on the X chromosome, it is unlikely that any model not inherited as an X-linked trait will have similar molecular basis to DMD."(8) These models are, therefore, unlikely to yield information on DMD's etiology. As Cooper has further noted, each of these models has "significant phenotypic dissimilarities to DMD.(8)
Dy/dy2j mice first develop signs of muscular dystrophy at about two weeks of age, starting with hind-limb weakness. The disease progresses rapidly; affected individuals die by about six months of age. While histological studies have shown that muscle-membrane breakdown is one of the first occurrences in DMD and the disease suffered by dy/dy2j mice, the mouse disease shows much more extensive membrane deterioration.(9) Also, fatty replacement of muscle fibers is much less evident in the mouse disease than in DMD.(3)
However, by far the largest discrepancy between DMD and dy/dy2j dystrophy is in the two diseases' underlying etiologies. While DMD is myogenic in origin, dy/dy2j dystrophy is neurogenic. The mice have a peripheral neuropathy characterized by lack of myelin in some spinal nerve roots and inadequate myelination of the peripheral nervous system;(10) the number of myelinated nerves may be reduced to as little as 5% of normal and may be nearly zero in the vicinity of blood vessels.(11)
BIO 40.45 strain Syrian hamsters show muscle degeneration similar to that in dy/dy2j mice. Like the mouse disease, the hamster disorder shows significant differences from DMD. Breakdown of muscle plasma membrane occurs over a larger area than in DMD. In the hamster disease, membrane breakdown occurs concurrently with other muscle-fiber changes rather than before them, as in DMD; calcium deposits in the muscle fibers are more extensive; and there is detectable mitochondria mineralization, which does not occur in DMD. Muscle necrosis and phagocytic activity are also more extensive in the hamster disease.(9) Unlike DMD patients, dystrophic hamsters develop very little fibrous tissue or fat infiltration;(9) muscle regeneration is much more complete than in DMD.(8) Cardiac muscle, on the other hand, is affected to a much greater degree in hamsters. Eventually, degeneration of cardiac muscle leads to heart failure, the major cause of death in dystrophic hamsters.(3)
Like dy/dy2j mice and dystrophic Syrian hamsters, New Hampshire strain chickens suffer from a muscular dystrophy (GMD) based on autosomal rather than sex-linked inheritance. GMD differs from DMD in numerous other significant ways as well. Unlike DMD, GMD has an inheritance pattern more complicated than that of a simple recessive trait.(12) Also, mammalian and avian muscle differ in fundamental ways. In a mammal, muscle fibers are fairly uniform in size; but in a chicken, muscle-fiber sizes vary greatly.(3) In contrast to DMD, GMD affects a fairly specific muscle fiber type. (GMD largely affects muscles composed primarily of fast-twitch α-white fibers and largely spares slow, tonic "red" fibers.)(13) Both hypertrophy and necrosis appear in dystrophic chickens, but hypertrophy is not a feature of DMD. In chickens, there is no evidence of a plasma membrane defect as seen in DMD patients.(9) Finally, the muscles of GMD chickens contain dystrophin.(5)
Although the mouse, hamster, and chicken models have been extensively studied, none is homologous to DMD. Referring to these models, J. B. Harris and C. R. Slater of Great Britain's Muscular Dystrophy Group Research Laboratories have acknowledged that "it may be unrealistic to look to animals to provide a detailed model of any of the specific forms of human muscular dystrophy."(12)
Differences Between DMD and Animal Models Based on X-Llnked Inheritance
In the mid-1980s, a new model for DMD emerged: the dystrophy of mdx mice. Unlike the previously described muscular dystrophies, the disorder in mdx mice is X-linked and arises from a defect in the dystrophin gene.(6) Nevertheless, researchers using this model have found that the mouse disease, while originating in a genetic defect similar to that of DMD, is phenotypically quite different from the human disease.
Both mdx mice and DMD patients lack dystrophin in their muscles,(6) and both show muscle necrosis with elevated serum creatine phosphokinase levels.(14) In the mice, however, massive muscle necrosis with an influx of phagocytic macrophages is seen at about three weeks of age, followed by a period of rapid regeneration with almost complete recovery. By five weeks of age, the mice are essentially normal,(8) with almost their full muscle mass regenerated.(15) Muscle fiber replacement by fat and fibrotic tissue is virtually absent in the mice. The regenerated muscle appears to be more stable than the original muscle fibers and resists further degeneration (although degeneration/regeneration on a much smaller scale can continue up to a year of age). As mentioned earlier, in DMD the muscle undergoes a chronic cycle of degeneration and regeneration until fibrosis has destroyed the muscle's ability to regenerate.(8)
Researchers report that mdx mice are only slightly weaker than normal mice or even apparently normal. The mouse disease has no effect on lifespan or ability to reproduce.(8) DMD, in contrast, causes severe disability and greatly reduces lifespan.(2) So, despite their similar genotypic origins, DMD and mdx have very different manifestations. Also, the fact that mdx mice experience spontaneous recovery largely, if not wholly, invalidates using this model to determine possible treatments for DMD. J. Dangain and G. Vrbova have described the mdx model as "unlikely to be related to Duchenne muscular dystrophy,"(15) and B. J. Cooper et al. have termed it a "poor clinical model for DMD."(16)
A canine X-linked muscular dystrophy (CXMD) has now been identified in a line of golden retrievers.(16,17) These dogs develop severe weakness and muscle atrophy at about six to eight weeks of age. Degeneration of muscle fibers with calcification and regeneration are seen histologically. Myocardial necrosis is also present.(17) The dogs' signs stabilize at about six months. As in DMD patients, the muscles of CXMD dogs lack dystrophin.(16) However, whereas humans with DMD live only one-third to one-fourth of a normal lifespan and die of respiratory failure,(2) most CXMD dogs live at least half a normal lifespan and die of cardiomyopathy.(16)
Even though CXMD appears to closely resemble DMD in etiology and progression, studies have already shown that a treatment's potential to benefit or harm DMD patients cannot be determined by observing that treatment's effect on CXMD dogs. As discussed in the section to follow, a treatment that now appears quite promising in DMD patients was unsuccessful in CXMD dogs.
The Failure of Animal Models to Reveal Effective Therapies
To date, researchers have developed no cure for DMD or intervention known to delay the disease's progression.(4) However, transplantation of myoblasts (primitive muscle cells) has yielded some encouraging results.(18) Myoblast transplantation involves injecting normal myoblasts into diseased muscle. In successful transplantation, the myoblasts fuse with muscle cells and/or replace them; they also produce dystrophin. Researchers hope that this process will restore normal muscle function in DMD patients.(18)
Myoblast transplantation to correct defective muscle was first developed in vitro and reported in 1984. Cultured muscle from muscular dysgenic mice (whose muscles congenitally fail to develop normally) was fused with normal myoblasts, and the muscle's ability to contract was restored.(19) Subsequent in vivo tests using nonhuman animals have failed to produce any results that are both clearly interpretable and in agreement with human data.
In 1988 researchers reported that myoblast transfer into the muscles of dy/dy2j mice returned muscle function to nearly normal, greatly improved the overall health of the mice, and significantly increased their lifespan.(20) However, as pathologist T. A. Partridge has pointed out, these results are inexplicable. The defect in dy/dy2j mice is primarily neurogenic (a matter of inadequate myelination). How could transplanted myoblasts--which act only on the muscle itself and do not affect myelination or nerve function--cure a neurogenic disease?(18,19)
Another group of researchers injected myoblasts into the muscle of mdx mice. (As previously mentioned, these mice lack dystrophin yet spontaneously recover nearly complete muscle function following a single bout of massive muscle degeneration.) After fusing with the muscle, the transplanted myoblasts produced dystrophin. However, because mdx mice do not require dystrophin for recovery of muscle function, the experiment cannot provide information on dystrophin's capacity to bring about such recovery. Because of this fundamental problem in using the mdx model to determine dystrophin's effect, the same group of researchers then varied the experimental procedure. Before myoblast transfer, they now exposed the mdx mice to Xrays. As the researchers had intended, irradiation prevented spontaneous recovery following muscle atrophy. Injection of myoblasts then restored the atrophied muscle to nearly normal function.(18,19) Irradiation and blockage of regeneration, however, constitute confounding variables since neither occurs in DMD. This experiment cannot provide data from which to predict the efficacy of myoblast transfer in DMD patients.
Researchers have also experimented with myoblast transfer in CXMD dogs, One laboratory treated two-week-old CXMD puppies with the immunosuppressant cyclosporine-A. Nevertheless, the muscles injected with myoblasts became mineralized and massively infiltrated with giant cells (fused macrophages)--indicating that the dogs' immune systems had rejected the transplanted myoblasts. Using a higher dose of cyclosporine-A, another laboratory prevented this immune reaction in CXMD dogs; myoblast transfer, however, still failed to induce dystrophin production.(21)
In any case, myoblast transfer is already being tested in humans. In contrast to the studies in CXMD dogs, preliminary studies in DMD patients have been promising. Injection of myoblasts into DMD patients' toe muscle has led to the appearance of dystrophin-containing muscle cells.(18) Currently, however, it remains unknown whether the presence of dystrophin can reduce or cure DMD symptoms and whether myoblast transfer will ever prove a safe, effective, and feasible mode of therapy.
Conclusion
A number of nonhuman animal models have been used in an attempt to characterize DMD and develop treatments. All of the models have differed from DMD in etiology, pathology, and/or clinical manifestation. They have also failed to yield information that can confidently be applied to the development of therapies. The evidence indicates that successful therapies for DMD will derive from in vitro and clinical studies, rather than from animal models.
References
1. Kunkel LM, Beggs AH, Hoffman HP: Molecular genetics of Duchenne and Becker muscular dystrophy: Emphasis on improved diagnosis. Clin Chem 1989;35:B21-B24.
2. Kunkel LM: Muscular dystrophy: a time of hope. Proc R Soc Lond B 1989:237:1-9.
3. Mrak RE: Muscular dystrophy, in Mrak RE (ed): Muscle Membranes in Diseases of Muscles. Boca Raton, CRC Press, 1985.
4. Kingston WJ, Moxley RT: Treatment of muscular dystrophies. Gen Pharmac 1989;20:263-268.
5. Meng G, Kress W, Bettecken T, Feichtinger W, Schempp W, Schmid M, Muller CR: A comparison of the dystrophin gene structure in primates and lower vertebrates, in Angelini C, Danieli GA, Fontanari D (eds): Muscular Dystrophy Research From Molecular Diagnosis toward Therapy. New York, Excerpta Medica. 1991.
6. Hoffman EP, Brown RH, Jr., Kunkel LM: Dystrophin: The protein product of the Duchenne muscular dystrophy locus. Cell 1987;51:919-928.
7. Elbrink J, Malhotra SK, Hunter EG: Duchenne muscular dystrophy: Assessment of experimental data from animals in relation to the human disease. Medical Hypotheses 1987;23:131-136.
8. Cooper BJ: Animal models of Duchenne and Becker muscular dystrophy. Brit Med Bull 1989;45:703-718.
9. Mendell JR, Higgins R, Sahenk Z, Cosmos E: Relevance of genetic animal models of muscular dystrophy to human muscular dystrophies. Ann NY Acad Sci 1979;319:409-430.
10. Bradley WG, Jaros E: Involvement of peripheral and central nerves in murine dystrophy. Ann NY Acad Sci 1979;319:132-141.
11. Carnwath JW, Shotton DM: Muscular dystrophy in the mdx mouse: Histopathology of the soleus and extensor digitorum longus muscles. J Neuro Sci 1987;80:39-54.
12. Harris JB, Slater CR: Animal models: What is their relevance to the pathogenesis of human muscular dystrophy? Brit Med Bull 1980;36:193-197.
13. Wilson BW: Developmental and maturational aspects of inherited avian myopathies. Proc Soc Exp Bio Med 1990;194:87-91.
14. Bulfield G, Siller WG, Wight PAL, Moore KJ: X chromosome-linked muscular dystrophy (mdx) in the mouse. Proc Natl Acad Sci USA 1984;81:1189-1192.
15. Dangain J, Vrbova G: Muscle development in mdx mutant mice. Muscle & Nerve 1984;7:700-704.
16. Cooper BJ, Winand NJ, Stedman H, et al.: The homologue of the Duchenne locus is defective in X-linked muscular dystrophy of dogs. Nature 1988;334:154-156.
17. Valentine BA, Cooper BJ, Cummings JF, deLahunta A: Progressive muscular dystrophy in a golden retriever dog: Light microscope and ultrastructural features at 4 and 8 months. Acta Neuropathol 1986;71:301-310.
18. Partridge TA: Myoblast transfer: A possible therapy for inherited myopathies? Muscle & Nerve 1991;14:197-212.
19. Partridge T: Use of normal myogenic cells to repair, replace, and rescue mdx mouse muscle from necrosis, in Angelini C, Danieli GA, Fontanari D (eds): Muscular Dystrophy Research: From Molecular Diagnosis toward Therapy. New York, Excerpta Medica, 1991.
20. Law PK, Goodwin TG, Li HJ, Ajamoughli G, Chen M: Myoblast transfer improves muscle genetics/structure/function and normalizes the behavior and life-span of dystrophic mice, in Griggs R, Karpati G (eds): Myoblast Transfer Therapy. New York, Plenum Press, 1990.
21. Discussion of Drs. Kornegay's, Bartlett's and Cooper's papers, in Griggs R, Karpati G (eds): Myoblast Transfer Therapy. New York, Plenum Press, 1990.