ACTA1 H40Y mutant iPSC-derived skeletal myocytes display mitochondrial defects in an in vitro model of nemaline myopathy

Abstract

Nemaline myopathies (NM) are a group of congenital myopathies that lead to muscle weakness and dysfunction. While 13 genes have been identified to cause NM, over 50% of these genetic defects are due to mutations in nebulin (NEB) and skeletal muscle actin (ACTA1), which are genes required for normal assembly and function of the thin filament. NM can be distinguished on muscle biopsies due to the presence of nemaline rods, which are thought to be aggregates of the dysfunctional protein. Mutations in ACTA1 have been associated with more severe clinical disease and muscle weakness. However, the cellular pathogenesis linking ACTA1 gene mutations to muscle weakness are unclear To evaluate cellular disease phenotypes, iPSC-derived skeletal myocytes (iSkM) harboring an ACTA1 H40Y point mutation were used to model NM in skeletal muscle. These were generated by Crispr-Cas9, and include one non-affected healthy control (C) and 2 NM iPSC clone lines, therefore representing isogenic controls. Fully differentiated iSkM were characterized to confirm myogenic status and subject to assays to evaluate nemaline rod formation, mitochondrial membrane potential, mitochondrial permeability transition pore (mPTP) formation, superoxide production, ATP/ADP/phosphate levels and lactate dehydrogenase release. C- and NM-iSkM demonstrated myogenic commitment as evidenced by mRNA expression of Pax3, Pax7, MyoD, Myf5 and Myogenin; and protein expression of Pax4, Pax7, MyoD and MF20. No nemaline rods were observed with immunofluorescent staining of NM-iSkM for ACTA1 or ACTN2, and these mRNA transcript and protein levels were comparable to C-iSkM. Mitochondrial function was altered in NM, as evidenced by decreased cellular ATP levels and altered mitochondrial membrane potential. Oxidative stress induction revealed the mitochondrial phenotype, as evidenced by collapsed mitochondrial membrane potential, early formation of the mPTP and increased superoxide production. Early mPTP formation was rescued with the addition of ATP to media. Together, these findings suggest that mitochondrial dysfunction and oxidative stress are disease phenotypes in the in vitro model of ACTA1 nemaline myopathy, and that modulation of ATP levels was sufficient to protect NM-iSkM mitochondria from stress-induced injury. Importantly, the nemaline rod phenotype was absent in our in vitro model of NM. We conclude that this in vitro model has the potential to recapitulate human NM disease phenotypes, and warrants further study.

Keywords: Mitochondria; Nemaline myopathy; Skeletal myotube; Stress injury; iPSC.

Figure 1:. Examining NM-iSkM for evidence of cellular aggregate phenotype.

Following secondary differentiation, cells were fixed and stained for ACTA1 or ACTN2 to evaluate formation of cellular aggregates. A) Images were taken at 20x or B) 100x magnification. No protein aggregates were observed in NM-iSkM vs. C-iSkM. N=3 differentiations, Scale bar represents 100 μM.

Figure 2:. ACTA1 and ACTN transcript levels are consistent between NM-iSkM.

Following secondary differentiation, RNA was harvested to assess any differences in ACTA1 and ACTN2 levels in the NM cell lines. No differences in either A) ACTA1, B) ACTN2 orACTN3 levels were observed in either NM-iSkM line compared to C-iSkM. N=3 differentiations, One-way ANOVA, n.s.

Figure 3:. ATP levels are decreased in NM-iSkM and addition of ATP delays mPTP formation.

Lysates were collected and subject to plate reading assays to examine ATP, ADP and free phosphate (Pi) levels. A) Reduction of ATP concentration is observed in NM –iSkM vs. C-iSkM. B) No differences in the levels of ADP and C) phosphate were observed across NM-iSkM vs.C-iSkM. N=3 differentiations, *p<0.05 vs. C-iSkM. One-way ANOVA with Welch’s correction.

Figure 4:. NM-iSkM display mitochondrial dysfunction and elevated stress injury compared to C-iSkM.

A) TMRE-labeled mitochondria were imaged for mean fluorescence as an indicator of mitochondrial membrane potential and no change in mitochondrial fluorescence was seen in NM-iSkM vs. C-iSkM at baseline. B) A decrease in mitochondrial membrane potential in NM vs. C-iSkM was seen when stress was added to cells. C) Cells were subject to laser-scanning confocal microscopy to induce oxidative stress injury to TMRE-labeled mitochondria. The mPTP formed much sooner in NM-iSkM mitochondria vs. C-iSkM, indicative of mitochondrial dysfunction. N=3 differentiations, One-way ANOVA, **p<0.001. D) Cells were treated with 10 μM ATP 30 minutes prior to stress induction. NM1-iSkM showed a significant delay in mPTP formation, and NM2-iSkM displayed a slight delay in mPTP formation. N=3 technical triplicates, One-way ANOVA, *p<0.05; **p<0.001.

Figure 5:. Induction of oxidative stress reveals stress phenotype in NM-iSkM.

To induce oxidative stress in NM-iSkM, cells were exposed to 1 hr 100 μM H₂0₂ in basal media, and supernatants were collected to measure LDH release as a marker for cell injury. A) Oxidative stress led to a significant increase in LDH release for both NM-iSkM lines, but not C-iSkM, suggestive of enhanced vulnerability to cell injury. N=3 differentiations, One-way ANOVA, *p<0.05, **p<0.001. B) To assess the level oxidative stress, cells were stained with DHE for cytosolic superoxide levels and imaged. NM-iSkM had higher levels of staining compared to C-iSkM, indicative of increased levels of superoxide. N=3 differentiations, One-way ANOVA, *p<0.05.