Antisense oligonucleotide-mediated Dnm2 knockdown prevents and reverts myotubular myopathy in mice

Abstract

Centronuclear myopathies (CNM) are non-dystrophic muscle diseases for which no effective therapy is currently available. The most severe form, X-linked CNM, is caused by myotubularin 1 (MTM1) loss-of-function mutations, while the main autosomal dominant form is due to dynamin2 (DNM2) mutations. We previously showed that genetic reduction of DNM2 expression in Mtm1 knockout (Mtm1KO) mice prevents development of muscle pathology. Here we show that systemic delivery of Dnm2 antisense oligonucleotides (ASOs) into Mtm1KO mice efficiently reduces DNM2 protein level in muscle and prevents the myopathy from developing. Moreover, systemic ASO injection into severely affected mice leads to reversal of muscle pathology within 2 weeks. Thus, ASO-mediated DNM2 knockdown can efficiently correct muscle defects due to loss of MTM1, providing an attractive therapeutic strategy for this disease.

Figure 1. In vitro validation of ASOs targeting Dnm2.

(a) Location of ASO-targeting sequences in Dnm2 mouse pre-mRNA. (b) Dnm2 mRNA level in ASO-treated C2C12 myoblast cells was determined by qRT-PCR and standardized to Hprt. Cells were electroporated with ASO control (ctrl), #1, #2 or #3 at 0.015, 0.06, 0.25 or 1 μM and collected 24 h later. (c) Representative western blot from the same C2C12 cells for DNM2 and the GAPDH loading control. Below, DNM2 protein levels was determined by densitometry and standardized to GAPDH. n=3 per each group. Data represent an average of three independent experiments±s.e.m. *P<0.05, ***P<0.001 for ASO Dnm2-treated versus ASO ctrl-treated cells (ANOVA test). kDa, kilodalton; MW, molecular weight.

Figure 2. In vivo validation of ASOs targeting Dnm2 following intramuscular injection.

(a) Representative western blot of 7-week-old WT or Mtm1KO TA muscles injected with 20 μg of ASO control (ctrl), #1, #2 or #3. DNM2 is present as two bands in muscle tissue. DNM2 densitometries were quantified below and standardized to the loading control GAPDH (n=5–7 mice per group). (b) Photography of TA muscles from WT or Mtm1KO mice treated with ASO ctrl or ASO Dnm2 #1. (c) TA muscle weight relative to body weight (n=8). (d) Specific muscle force of the TA (n=5–6 mice per group).(e) TA muscle sections were stained with H&E (left) to visualize nuclei positioning or with SDH (right) for mitochondria oxidative activity distribution. Scale bars, 50 μm. (f) Percentage of fibres with mislocalized nuclei (n=4–6 mice per group). (g) Fibre area was determined in 600–1,000 fibres per sample (n=5 mice per group). Data represent means±s.e.m. NS, not statistically significant. *P<0.05, **P<0.01, ***P<0.001 for TA treated with ASO Dnm2 versus TA treated with ASO ctrl (ANOVA test). kDa, kilodalton; MW, molecular weight.

Figure 3. Dose–response effect of ASO-Dnm2 systemic injections.

Representation of survival percentage (a) and whole-body weight evolution (b) of WT injected with 25 mg kg⁻¹ ASO control (ctrl) or ASO#1, or Mtm1KO injected with 25 mg kg⁻¹ ASO ctrl or different doses of ASO#1. (n=7 mice per group at week 3). (c) Dots represent individual body weight of ASO-treated WT or Mtm1KO mice at week 12 (sacrifice day). n=5–7 mice per group except for Mtm1KO treated with 3.125 mg kg⁻¹ where n=3. **P<0.01, ***P<0.001 for mice treated with ASO#1 versus WT treated with ASO control (ANOVA test followed by post hoc Bonferroni).

Figure 4. Physiological effects of ASO-Dnm2 systemic injections at 7 weeks of age.

(a) Following 5-weekly injections of ASO, mice were killed and TA and gast. muscles were weighed (n=5–7 mice per group). (b) TA specific muscle force was measured after sciatic nerve stimulation. The specific muscle force was calculated by dividing the absolute force by the TA weight (n=5–7 mice per group). (c) TA muscle sections were stained for H&E or SDH. Sarcomere and triads (yellow arrows) ultrastructure was assessed by TEM. Scale bars: 50 μm (H&E and SDH) or 500 nm (TEM) images. (d) Percentage of fibres with mislocalized nuclei was determined in 1,000 fibres (n=5–6 mice per group). (e) TA muscle fibre area was calculated on 300–600 fibres per sample (n=5 mice per group). Data represent mean±s.e.m. *P<0.05, **P<0.01, ***P<0.001 for mice treated with ASO Dnm2 versus WT treated with ASO ctrl. ^$P<0.05, ^$$P<0.01, ^$$$P<0.001 for mice treated with ASO Dnm2 versus Mtm1KO treated with ASO ctrl (two-way ANOVA followed by post hoc Bonferroni).

Figure 5. Physiological effects of ASO-Dnm2 systemic injections at 12 weeks of age.

(a) TA and gast. muscle weights from these mice were measured and the ratio relative to the body weight was represented. (b) The TA in situ muscle force was measured after sciatic nerve stimulation. The specific muscle force was calculated by dividing the absolute force by the TA weight. (c) TA muscle transversal sections stained with H&E (left panel) and SDH (middle panel) at 12 weeks (10 weeks of treatment). Sarcomere and triads (yellow arrows) ultrastructure assessed by TEM (right panel). Scale bars for H&E and SDH is 50 μm and 500 nm for TEM pictures. (d) The percentage of fibres with mislocalized nuclei was counted on 1,000 fibres per group. (e) TA muscle fibre area was calculated on 300–600 fibres per sample. For all tests cited above, n=5–7 mice per group except for Mtm1KO treated with 3.125 mg kg⁻¹ where n=3. Data represent means±s.e.m. NS, no statistical significance, *P<0.05, **P<0.01, ***P<0.001 for mice treated with ASO#1 versus WT treated with ASO control (ctrl) (ANOVA test followed by post hoc Bonferroni).

Figure 6. Diaphragm histology and ultrastructure organization.

(a) Hematoxylin and eosin (H&E) staining of cross or longitudinal diaphragm sections of WT or Mtm1KO mice treated with ASO control (ctrl) or ASO#1 at different ages (7 or 12 weeks). Scale bars, 100 μm. These longitudinal sections have been immunostained with antibodies against RYR1 and DHPR. Scale bars, 10 μm. The ultrastructure of diaphragm was analysed using electronic microscopy to assess the sarcomere organization and triad formation and shape (yellow arrows). Scale bars, 500 nm. (b) Graph depicting mean±s.e.m. of quantification of diaphragm transversal sections thickness. n=4–5 samples. NS, not statistically significant, *P<0.05, **P<0.01 (ANOVA test followed by post hoc Bonferroni).

Figure 7. DNM2 knockdown and ASO concentration in skeletal muscles.

(a,b) DNM2 and GAPDH (loading control) expression in TA muscles of ASO-treated WT or Mtm1KO mice at 7 weeks (a) (n=6–7 mice per group) or at 12 weeks of age (b) (n=5–7 mice per group except for Mtm1KO treated with 3.125 mg per kg where n=3). (c,d) ASO concentration was determined in gast. muscle of 7- (c) or 12- (d) week-old mice using mass spectrometry and normalized to muscle weight (n=4 mice per group). Data represent mean±s.e.m. NS, not statistically significant, *P<0.05, **P<0.01, ***P<0.001 for mice treated with ASO Dnm2 versus WT treated with ASO ctrl. ^$$P<0.01, ^$$$P<0.001 for mice treated with ASO Dnm2 versus Mtm1KO mice treated with ASO ctrl (two-way ANOVA followed by post hoc Bonferroni). kDa, kilodalton; MW, molecular weight.

Figure 8. ASO-Dnm2 systemic injections revert installed muscle defects in Mtm1 KO mice.

(a) Chronology of CNM phenotype onset and evolution in Mtm1KO mice. (b) Survival of WT (n=5) or Mtm1KO (n=16) mice upon ASO treatment started at 5 weeks. (c) Disease severity score (DSS) evolution of Mtm1KO mice that died during the first week of treatment or have been rescued after ASO treatment started at week 5. (d) Radar chart representing average values of the six main CNM features in Mtm1KO mice at 5 weeks old (before ASO treatment, left panel) and their evolution at week 12 (after 8 injections, right panel). (e) Body weight evolution of WT treated with ASO control (n=5), Mtm1KO mice treated with ASO control (n=8) or ASO#1 (n=8 for dead or rescued Mtm1KO groups). (f) Hanging test performance of WT (n=5) or Mtm1KO (n=8) rescued by ASO treatment. (g–i) Clinical tests done at 11 weeks of age, WT or Mtm1KO mice underwent different clinical tests to assess whole body strength, fine motor coordination, balance and resistance to fatigue using rotarod (g), string test (h) and grip test (i). NS, not statistically significant (t-test).

Figure 9. Physiological improvements of CNM phenotype in severely affected Mtm1 KO mice.

(a) TA and gast. muscles weight and (b) in situ specific muscle force at week 12 and following 8 weeks of ASO treatment. (c) TA muscle sections were stained with H&E or SDH. Scale bar, 50 μm for histology and 500 nm for TEM. Yellow arrows point to organized triads. (d) Fibre size average was quantified in 400–600 fibres (n=5). (e) TA muscle fibre area distribution of Mtm1KO mice injected with ASO#1 versus WT injected with ASO control (n=5). (f) Fibres with abnormal position of nuclei (n=5). (g) Ratio of triads per sarcomere. (h) DNM2 and GAPDH (loading control) protein levels were measured in TA muscles by western blot. Data represent means±s.e.m. NS, not statistically significant, *P<0.05, **P<0.01, ***P<0.001 (t-test). kDa, kilodalton; MW, molecular weight.