Dystrophins carrying spectrin-like repeats 16 and 17 anchor nNOS to the sarcolemma and enhance exercise performance in a mouse model of muscular dystrophy

Abstract

Sarcolemma-associated neuronal NOS (nNOS) plays a critical role in normal muscle physiology. In Duchenne muscular dystrophy (DMD), the loss of sarcolemmal nNOS leads to functional ischemia and muscle damage; however, the mechanism of nNOS subcellular localization remains incompletely understood. According to the prevailing model, nNOS is recruited to the sarcolemma by syntrophin, and in DMD this localization is altered. Intriguingly, the presence of syntrophin on the membrane does not always restore sarcolemmal nNOS. Thus, we wished to determine whether dystrophin functions in subcellular localization of nNOS and which regions may be necessary. Using in vivo transfection of dystrophin deletion constructs, we show that sarcolemmal targeting of nNOS was dependent on the spectrin-like repeats 16 and 17 (R16/17) within the rod domain. Treatment of mdx mice (a DMD model) with R16/17-containing synthetic dystrophin genes effectively ameliorated histological muscle pathology and improved muscle strength as well as exercise performance. Furthermore, sarcolemma-targeted nNOS attenuated alpha-adrenergic vasoconstriction in contracting muscle and improved muscle perfusion during exercise as measured by Doppler and microsphere circulation. In summary, we have identified the dystrophin spectrin-like repeats 16 and 17 as a novel scaffold for nNOS sarcolemmal targeting. These data suggest that muscular dystrophy gene therapies based on R16/17-containing dystrophins may yield better clinical outcomes than the current therapies.

Figure 1. Sarcolemmal nNOS localization depends on the rod, but not the C-terminal, domain of dystrophin.

(A) Schematic outline of the full-length and synthetic dystrophin constructs. “Yes” indicates that nNOS is recruited to the sarcolemma by the construct; “No” indicates that the construct cannot restore nNOS to the sarcolemma. Dotted boxes denote deleted regions. (B) The DH2–R15 minidystrophin anchors nNOS to the sarcolemma. Representative photomicrographs of dystrophin immunostaining, nNOS activity staining, and nNOS immunostaining in the DH2–R15 minigene plasmid–transfected mdx muscle. Dystrophin was revealed by epitope-specific antibodies (H1, R11, R16, R17, R18, and H3). The H1 antibody only recognizes human dystrophin (Hum Dys). Other dystrophin antibodies recognize both human and mouse dystrophin. Asterisks indicate a minigene-transfected myofiber; arrows, revertant myofibers. Scale bar: 20 μm. (C) Sarcolemmal nNOS localization does not require the C-terminal domain, nor is it dependent on the repeats adjacent to R16/17. Representative photomicrographs of human dystrophin/nNOS double immunostaining and nNOS activity staining on serial sections from the DH2–R15/DR18–19, DH2–R15/DC, and DR3–15/DR18–23/DC plasmid–transfected mdx muscles. Asterisks indicate dystrophin plasmid–transfected fibers; crosses, revertant myofibers. Scale bar: 50 μm.

Figure 2. Both R16 and R17 are indispensable for sarcolemmal nNOS localization.

(A) Schematic outline of the microdystrophin constructs. “Yes” indicates that nNOS is recruited to the sarcolemma by the construct; “No” indicates that nNOS is not recruited. Dotted boxes denote deleted regions. (B and C) Both R16 and R17 are required for sarcolemmal nNOS localization. Representative serial immunostaining and nNOS activity staining photomicrographs of mdx muscles infected with AV.CMV.DR3–15/DR17–23/DC (B) or AV.CMV.DR2–15/DR18–23/DC (C). Sarcolemmal nNOS localization was observed only in microdystrophin carrying both R16 and R17 (C) but not in microdystrophin carrying only R16 (B). Scale bars: 500 μm. See Supplemental Figure 4 for high-magnification images. (D) Representative yeast two-hybrid assay results (from 5 independent experiments) reveal interaction between R16/17 and the nNOS PDZ domain. Individual repeat R16 alone or R17 alone does not bind the nNOS PDZ domain. (E) Relative β-galactosidase activity from the quantitative yeast two-hybrid assay. n = 6 for each group.

Figure 3. The R16/17-containing microgene ameliorates muscle disease and improves muscle force following AAV gene transfer.

(A) Results from adult mdx muscle (local delivery; n = 4–8 for each assay). Top: Serial sections with H&E and nNOS activity staining (see immunostaining in Supplemental Figure 5); scale bar: 100 μm. Bottom left: Evans blue dye (EBD) uptake; scale bar: 400 μm. An R16-specific antibody revealed the microgene. Bottom right: Central nuclear quantification. (B) Specific tetanic force in the mdx EDL muscle following systemic neonatal microgene therapy. ^#P ≤ 0.033 compared with all other groups; ^†P ≤ 0.001 compared with mdx only. (C) Systemic delivery of AV.CMV.DR2–15/DR18–23/DC to adult MyoD/dystrophin double-knockout (M-dko) mice improved skeletal muscle function and reduced the serum CK level. Sample size for muscle function assay: n = 4 for uninfected littermate controls; n = 14 for AV.CMV.DR2–15/DR18–23/DC–infected mice. Sample size for the CK assay: n = 3 for uninfected littermate controls; n = 6 for AV.CMV.DR2–15/DR18–23/DC–infected mice. *P ≤ 0.002, ^‡P = 0.015 compared with controls.

Figure 4. The R16/17-containing minigene (DH2–R15) restores sarcolemmal nNOS, normalizes muscle force to wild-type levels, and reduces eccentric contraction–associated muscle injury.

(A) Representative Western blots of whole muscle lysates from the TA muscles of BL10, mdx, DH2–R15 transgenic mdx, DH2–R19 transgenic mdx, and nNOS-knockout mice. GAPDH was used as the loading control. (B) Representative Western blots of microsomal preparations from the quadriceps femoris muscles. (C) Representative photomicrographs of dystrophin immunostaining (H1, R16, R19) and nNOS activity staining from DH2–R15 and DH2–R19 transgenic mdx mice. Scale bar: 100 μm. (D) The DH2–R15 minigene normalizes muscle-specific force and prevents eccentric contraction-induced injury. *P ≤ 0.021 compared with transgenic mdx mice.

Figure 5. The DH2–R15 minigene attenuates norepinephrine-mediated vasoconstriction in contracting muscle.

Decreases in femoral vascular conductance were recorded following intraarterial injection of 2 doses of norepinephrine (NE) into the resting and contracting hind limbs of male HSA(human skeletal α-actin promoter).DH2–R15 and HSA.DH2–R19 transgenic mdx mice. (A) Results from DH2–R15 transgenic mdx mice. (B) Results from DH2–R19 transgenic mdx mice. AUC is presented in arbitrary units. Asterisks indicate significant differences.

Figure 6. The DH2–R15 minigene transgenic mdx mice show better muscle perfusion during exercise.

Muscle and internal organ perfusion at rest and during exercise was quantified using radiolabeled microspheres in male HSA.DH2–R15 and HSA.DH2–R19 transgenic mdx mice. (A) Tissue perfusion in selected muscles at rest. (B) Kidney perfusion at rest. (C) Tissue perfusion in selected muscles during exercise. (D) Kidney perfusion during exercise. Gastro., gastrocnemius muscle. Asterisks indicate significant differences.

Figure 7. The DH2–R15 minigene prevents ischemic injury and enhances performance during intensive exercise.

(A) Body weight–normalized daily running distance in HSA.DH2–R15 and HSA.DH2–R19 transgenic mdx mice (6-month-old male mice). *P ≤ 0.02. (B and C) Representative photomicrographs of general histopathology (H&E staining), inflammation (macrophage infiltration), nNOS activity, and minidystrophin expression (immunostaining with epitope-specific antibodies) in the limb muscle harvested after 8 days’ intensive treadmill running. (B) Results from DH2–R15 transgenic mdx mice. Arrowhead indicates a single regenerating myofiber. (C) Results from DH2–R19 transgenic mdx mice. Arrows indicate macrophages. Scale bars: 50 μm. See Supplemental Figure 10 for low-magnification images.

Figure 8. Schematic outline of sarcolemmal nNOS recruitment by an R16/17-containing microdystrophin gene.

(A) The N-terminal domain (N) of the DR2–15/DR18–23/DC microgene interacts with γ-actin. The cysteine-rich domain (CR) of this microgene interacts with the dystroglycan complex (DG). Syntrophin (Syn) is brought to the sarcolemma by dystrobrevin (Dbr), which binds to the sarcoglycan complex (SG). nNOS recruitment requires 2 independent interactions, one between the nNOS PDZ β-finger and the syntrophin PDZ domain and the other between the nNOS PDZ groove and R16/17. Numbers denote spectrin-like repeats; H1 and H4, hinge 1 and hinge 4. (B) The hypothetical nNOS-R16/17 interaction model. The C-terminal helix of R16 joins the N-terminal helix of R17 to form a long helix. The R16/17 dimer then binds to the groove in the nNOS PDZ domain. Yellow, R16; blue, R17; dotted, flanking repeats.