The golden retriever model of Duchenne muscular dystrophy

Abstract

Duchenne muscular dystrophy (DMD) is an X-linked disease caused by mutations in the DMD gene and loss of the protein dystrophin. The absence of dystrophin leads to myofiber membrane fragility and necrosis, with eventual muscle atrophy and contractures. Affected boys typically die in their second or third decade due to either respiratory failure or cardiomyopathy. Despite extensive attempts to develop definitive therapies for DMD, the standard of care remains prednisone, which has only palliative benefits. Animal models, mainly the mdx mouse and golden retriever muscular dystrophy (GRMD) dog, have played a key role in studies of DMD pathogenesis and treatment development. Because the GRMD clinical syndrome is more severe than in mice, better aligning with the progressive course of DMD, canine studies may translate better to humans. The original founder dog for all GRMD colonies worldwide was identified in the early 1980s before the discovery of the DMD gene and dystrophin. Accordingly, analogies to DMD were initially drawn based on similar clinical features, ranging from the X-linked pattern of inheritance to overlapping histopathologic lesions. Confirmation of genetic homology between DMD and GRMD came with identification of the underlying GRMD mutation, a single nucleotide change that leads to exon skipping and an out-of-frame DMD transcript. GRMD colonies have subsequently been established to conduct pathogenetic and preclinical treatment studies. Simultaneous with the onset of GRMD treatment trials, phenotypic biomarkers were developed, allowing definitive characterization of treatment effect. Importantly, GRMD studies have not always substantiated findings from mdx mice and have sometimes identified serious treatment side effects. While the GRMD model may be more clinically relevant than the mdx mouse, usage has been limited by practical considerations related to expense and the number of dogs available. This further complicates ongoing broader concerns about the poor rate of translation of animal model preclinical studies to humans with analogous diseases. Accordingly, in performing GRMD trials, special attention must be paid to experimental design to align with the approach used in DMD clinical trials. This review provides context for the GRMD model, beginning with its original description and extending to its use in preclinical trials.

Keywords: Animal models; Duchenne muscular dystrophy (DMD); Golden retriever muscular dystrophy (GRMD); Preclinical studies.

Fig. 1

Characteristic myofiber membrane lesion in DMD. Electron photomicrograph demonstrates lack of continuity of the myofiber (sarcolemmal) membrane (arrows), while the basal lamina remains intact. The myofiber architecture is disrupted subjacent to the membrane lesion. Original magnification 26,000. From reference [8]

Fig. 2

Spontaneous EMG activity in muscle disease. Activity, termed pseudomyotonic bursts, that begins and ends abruptly, in the hypertrophied calf muscle of a DMD patient (a, b), contrasts with waxing and waning myotonic activity from a forearm extensor of a myotonic dystrophy patient (c, d). The activity in a was recorded from two fibers discharging at 22 and 11/s over a 45-s period with the three bursts (a–c) separated by 20 s. Bursts in b had a frequency of 15/s. Increasing and decreasing activity in c (maximum 50/s) and d (maximum 35/s) were induced by movement of the concentric recording electrode and direct stimuli, respectively. From reference [12]

Fig. 3

GRMD colony founder dog, Rusty, at North Carolina State University at ~3 years of age. Note the relatively mild phenotype characterized primarily by a plantigrade stance

Fig. 4

Myofiber with focal necrosis and membrane lesion in GRMD dog. Electron photomicrographs demonstrating a wedge-shaped subsarcolemmal area of disrupted architecture (a) and higher magnification (b) of the area identified with the asterisk where the myofiber membrane is absent (arrow) and the basal lamina is intact. Original magnification 4500 in a and 13,500 in b. Modified from reference [93]

Fig. 5

Canine dystrophin protein (Ensembl protein ID ENSCAFP00000031637), with mutation information for 10 dog breeds with dystrophinopathies. The breeds are Pembroke Welsh Corgi [98], Labrador retriever [97], Tibetan terrier [97], Cocker spaniel [97], golden retriever [96], Japanese Spitz [99], Norfolk terrier [100], German shorthaired pointer [101, 102], two distinct mutations in the Cavalier King Charles spaniel [103, 104], and Rottweiler [105]. CH indicates actin-binding calponin homology domains. The “WWP” domain binds proline-rich polypeptides and is the primary interaction site for dystrophin and dystroglycan. EF indicates members of the EF-hand family domain that stabilizes the dystrophin-dystroglycan complex. ZNF represents a putative zinc-binding domain, ZnF_ZZ is present in dystrophin-like proteins and may bind to calmodulin. All 79 exons are represented. Exons and protein domains are approximately shown to scale. Insertion and deletion mutations are shown above the exons. At the bottom of the figure, the German shorthaired pointer (GSHP) DMD gene deletion and point mutations are identified with a hatched line and arrows, respectively. Modified from reference [70]

Fig. 6

Comparative disease course of GRMD based on the relative equivalency of the first year of a golden retriever’s life and initial 20 years of a human’s. The two periods are divided into quartiles, e.g., 0–3 months of GRMD paralleling 0–5 years of DMD, with signs of the skeletal myopathy (SM) and cardiomyopathy (CM) listed for each period [–130]. Note, the GRMD clinical course from 0–6 months largely parallels that of DMD over the 0–10 year period. However, the GRMD and DMD phenotypes then dramatically diverge, with GRMD dogs often stabilizing and DMD continuing to progress. SM skeletal myopathy, CM cardiomyopathy, LVEDV and LVSDV left ventricular end diastolic and systolic volumes, LVEF left ventricular ejection fraction. Modified from reference [75]

Fig. 7

Averaged MRI segmentation of dogs with variable GRMD and myostatin (Mstn) genotypes. T2-FS MRI images of pelvic limb muscles in the transverse plane at the level of the midthigh are shown in non-dystrophic control (a), dystrophic GRMD, wild-type Mstn ^+/+ (b) and GRMD, heterozygous null Mstn ^+/− (GRippet) (c) dogs. Note the proportional enlargement of the sartorius and hamstring muscles and the associated atrophy/hypoplasia of the quadriceps femoris of the GRMD, wild-type Mstn ^+/+ dogs, relative to the non-dystrophic control dogs, and the even more dramatic differential size of these muscles in the GRippet Mstn ^+/− dogs. From reference [158]

Fig. 8

T2-weighted MR images of GRMD pelvic limb muscles 16 weeks after AAV9-CMV-mini-dystrophin vector intravenous injection. Transverse (left) and sagittal (right) images of two different dogs (dog 1: a, b, e, and f; dog 2: c, d, g, and h) are seen. The images in e–h have been segmented and color-coded to outline individual muscles. Signal intense lesions that persisted with fat saturation, most likely representing fluid due to inflammation or edema, are particularly pronounced in the vastus heads of the quadriceps and adductor muscles. From reference [195]