Myonuclear Domain-Associated and Central Nucleation-Dependent Spatial Restriction of Dystrophin Protein Expression

The restoration of uniformly distributed dystrophin protein expression is an important consideration for the development of advanced therapeutics for Duchenne muscular dystrophy (DMD).

We have generated a novel genetic mouse model (mdx52-Xist^Δhs) that expresses variable and nonuniformly distributed dystrophin protein from birth as a consequence of skewed X-chromosome inactivation. mdx52-Xist^Δhs myofibers are heterokaryons containing a mixture of myonuclei expressing either wild-type or mutant dystrophin alleles in a mutually exclusive manner, resulting in dystrophin protein being spatially restricted to corresponding dystrophin-expressing myonuclear domains. This phenotype models the situation in female dystrophinopathy and dystrophic muscle in which dystrophin has been incompletely restored by partially effective experimental therapeutics. Dystrophin distribution was assessed in mdx52-Xist^Δhs muscle sections and isolated single myofibers by immunostaining and RNA-FISH analysis.

Total dystrophin expression increased by ~2.8-fold (p < 0.010) in aged (60-week-old) mdx52-Xist^Δhs mice relative to 6-week-old adults, suggestive of an aging-associated accumulation of dystrophin-expressing myonuclei through positive selection, although this was insufficient to resolve sarcolemmal dystrophin patchiness. Nonuniformly distributed dystrophin conferred partial protection against pathology-related muscle turnover in an expression-level-dependent manner in both adult and aged mdx52-Xist^Δhs mice compared to mice expressing no dystrophin. Isolated mdx52-Xist^Δhs myofibers exhibited patchy 'zebra-like' banding of dystrophin sarcolemmal coverage that colocalized with β-dystroglycan but not neuronal nitric oxide synthase, which was uniformly distributed. Systematic classification of isolated mdx52-Xist^Δhs myofibers revealed unexpected and profound differences associated with central nucleation, with dystrophin found to be absent in centrally nucleated myofibers and myofiber segments. Muscle injury alone was insufficient to recapitulate this phenomenon, suggesting that it is a feature of the dystrophic environment. Dmd mRNA was found to be present throughout centrally nucleated segments, and proteins such as titin and F-actin were uniformly distributed, suggesting that dystrophin is specifically repressed at the protein level in these regions. The microtubule network was moderately disrupted in mdx52-Xist^Δhs 'zebra-banded' fibers, but this effect was not different between dystrophin-positive and dystrophin-negative myofiber subdomains. By contrast, centrally nucleated mdx52-Xist^Δhs myofibers exhibited severe microtubule network disruption.

These findings reveal new insights into the importance of dystrophin spatial localization and identify a previously unappreciated barrier to effective therapeutic dystrophin restoration, particularly within regenerated or centrally nucleated myofibers.