Histone deacetylase 6 inhibition promotes microtubule acetylation and facilitates autophagosome-lysosome fusion in dystrophin-deficient mdx mice

Curation statements for this article:
  • Curated by eLife

    eLife logo

    eLife assessment

    This important study pinpoints nitrite oxide synthase 2 activity and decreased microtubule acetylation as distinct regulators of altered autophagic flux that may contribute to pathogenesis in a mouse model of Duchenne muscular dystrophy. While most of the evidence to support these claims is convincing, the claim that autophagy is improved with increased microtubule acetylation is incompletely supported. This work may be of broad interest to muscle biologists and has translational potential for the treatment of Duchenne muscular dystrophy.

This article has been Reviewed by the following groups

Read the full article See related articles

Abstract

Duchenne Muscular Dystrophy (DMD) is a severe X-linked genetic disorder. Defective autophagy and disorganized microtubule network contributes to DMD pathogenesis, yet the mechanisms by which microtubule alterations regulate autophagy remain elusive. We show decreased acetylated α-tubulin and enhanced histone deacetylase (HDAC6) expression in mdx mice. Pharmacological inhibition of HDAC6 increases tubulin acetylation and enhances Q-SNARE complex formation, leading to improved autophagosome-lysosome fusion. HDAC6 inhibition reduces apoptosis, inflammation, muscle damage and prevents contraction induced force loss. HDAC6 inhibition restores peroxiredoxin (PrxII) by increasing its acetylation and protecting it from hyper-oxidation, hence modulating intracellular redox status in mdx mice. Genetic inhibition of Nox2 activity in mdx mice promotes autophagosome maturation. Our data highlight that autophagy is differentially regulated by redox and acetylation in mdx mice. By restoring tubulin acetylation HDAC6 inhibition enhances autophagy, ameliorates the dystrophic phenotype and improves muscle function, suggesting a potential therapeutic target for treating DMD.

Article activity feed

  1. Author Response

    Reviewer #1 (Public Review):

    It has been previously shown that defective autophagy and disorganized microtubule network contribute to the pathogenesis of Duchenne muscular dystrophy (DMD). The authors previously reported that nitrite oxide synthase 2 (NOX2) regulates these alterations. It was also shown that acetylated tubulin facilitates autophagosome-lysosome fusion and thus autophagy. In the present study, the authors showed that autophagy is differentially regulated by redox and acetylation modifications in dystrophic mdx mice. The ablation of Nox2 in mdx mice activated the autophagosome maturation but not its fusion with the lysosome. On the other hand, the inhibition of histone acetylase 6 (HDAC6) restored microtubule acetylation, promoted autophagosome-lysosome fusion, and improved muscle function in mdx mice. The strength of this paper is the combination of different approaches to decipher the mechanism, including the evaluation of the level and interaction of several proteins involved in the maturation of autophagosomes and in the fusion between autophagosomes and lysosomes.

    This study reveals an important molecular mechanism by which increasing microtubule acetylation improves autophagy and muscle function in dystrophic mice. This has a translational impact on several diseases in which autophagy is impaired. The improvement of autophagosome-lysosome fusion with HDAC6 inhibitor is supported by several data, but some parts merit further analysis:

    1. To add appropriate controls (e.g. without antibodies) to support protein-protein interaction for all co-immunoprecipitation assays.

    Thank you for your valuable suggestion. We appreciate your input and have taken it into consideration. Based on your recommendation, we have conducted an experiment by including IP-IgG as a negative control to support the protein-protein interaction results obtained from the co-immunoprecipitation assays. The results of the negative control have been included in the respective figures. Additionally, to ensure the accuracy of the negative control, we ran the positive controls on the same blot. We have immunoprecipitated the same amount of samples for the negative control as we did for the actual IP samples presented in the manuscript. We believe that the inclusion of the negative control has strengthened the validity of our results and the conclusion drawn from our study.

    1. The simple evaluation of the protein levels of p62 and LC3-II is not sufficient to claim autophagy improvement after HDAC6 inhibition. It would be good to evaluate the autophagic flux in vivo in all groups of mice (to treat the mice with or without autophagy inhibitor and evaluate whether the difference in the level of LC3-II between the two conditions is higher with HDAC6 inhibitor than without in the mdx mice).

    Thank you for your suggestion to further evaluate the role of TubA on autophagic flux in vivo. We have included data using chloroquine to test the effect of TubA on autophagic flux in vivo. We found that chloroquine increased LC3 and p62 in skeletal muscle from mdx and mdx + TubA mice, suggesting. We have now included this information in the revised manuscript.

    Reviewer #2 (Public Review):

    Agrawal et al. propose an interesting model in which the autophagy pathway in adult mouse skeletal muscle fibers is orchestrated by two independent mechanisms: a) the activity of the NADPH oxidase (Nox) 2 enzyme necessary for autophagosome biogenesis and maturation and b) the level of acetylation of the microtubule (MT) network more selectively responsible for the fusion of the autophagosomes to the lysosomes. Using the well-known mdx mouse, a model for Duchenne muscular dystrophy, the authors perform a quite impressive (but rather traditional) biochemical characterization of the autophagy pathway and found that biogenesis and maturation of the autophagosomes are impaired in mdx mice muscle fibers by means of altered expression of components of the class III phosphatidylinositol 3-kinase complex (PI3K) such as Beclin, VPS15 (both upregulated in mdx mice), ATG14L and VPS34 (both downregulated), and by the reduced expression of JNK and JIP-1, required for the formation of the heterodimer between Beclin and ATG14L-VPS34. In mdx mice, defective nucleation of the phagophore appears to be coupled to altered elongation and expansion as confirmed by decreased expression of WIPI-1, an early marker of autophagosome formation, required for the assembly of the ATG5-12 complex. Clearance of sequestered cytosolic components necessitates the fusion of the autophagosome with the lysosome, a process that the authors found impaired in mdx mice due to altered formation of the SNARE tertiary complex (STX17-SNAP29-VAMP8), as a result of the marked reduction of STX17 expression.

    In a previous work (Pal et al., Nat Commun 2014), the same group described the generation of an mdx-based mouse model where Nox2 activity was abolished by genetic ablation of the p47phox component. These mice presented with a better outcome in terms of dystrophic pathophysiology by means of reduced oxidative stress and improved autophagy. Further characterization of these mice in the present study reveals that in p47-/-/mdx mice abolishment of Nox2 activity restores autophagosome nucleation and maturation thanks to the increased expression of p-JNK, JIP-1 and improved stability of the Beclin-ATG14L complex, but no amelioration is observed on the formation of the SNARE tertiary complex indicating that the biogenesis of autophagosomes is dependent on Nox2 activity but not the fusion between autophagosomes and lysosomes. Given the existing body of evidence in non-muscle cells pointing at alpha-tubulin acetylation as a regulator of MT activity facilitating the fusion of autophagosomes to lysosomes, the authors thought to investigate the level of MT acetylation in mdx mice muscle fibers and found that acetylation is reduced but can be restored by inhibiting the HDAC6 enzyme via the FDA-approved, highly selective pharmacological inhibitor Tubastatin A (Tub A). Treatment of mdx mice at 3 weeks of age (before the onset of pathological manifestations) with Tub A not only restored the normal level of alpha-tubulin acetylation (without altering the organization and density of the MT network) but also curbed the intracellular redox status and improved the autophagic flux by stabilizing the SNARE tertiary complex. Interestingly, treatment of dystrophic mice with Tub A results in substantial improvement of the dystrophic phenotype as confirmed by a reduced level of apoptosis, diminished tissue inflammation, improved sarcolemma integrity, and superior force generation capacity in ex vivo experiments using the diaphragm and Extensor Digitorum Longus (EDL) muscle fibers of Tub A-treated mdx mice compared to untreated mdx and healthy counterparts.

    The in-depth characterization of the steps orchestrating the autophagy pathway in the mdx mouse model on the one hand, and the comprehensive evaluation of the phenotype of the mdx mice treated with the HDAC6 inhibitor Tubastatin A on the other, support the conclusions proposed by the authors. Nonetheless, some aspects deserve consideration.

    1. The effect of increased alpha-tubulin acetylation by means of genetic and pharmacological strategies (i.e., in vivo overexpression of alpha-tubulin acetyltransferase-aTAT1 and treatment with Tubacin or Tubastatin A, respectively) has been previously explored in isolated cardiomyocytes and skeletal muscle fibers and revealed that augmented MT acetylation, due to selective inhibition of HDAC6, increases cytoskeletal stiffness and favors Nox2 activation (Coleman et al., J Gen Physiol 2021).

    We have added a discussion of the work by Coleman and colleagues. In brief, that work was in wild-type cardiac and skeletal muscle and showed that MT acetylation controlled stiffness in control muscle cells. Interestingly, while they did not quantify MT organization, their data suggest that HDAC6 inhibition does not alter organization. Here, we are assessing the role of MT acetylation is a diseased model, mdx. Taken together, our data along with that from Ward and colleagues highlight the importance of a proper balance of tubulin acetylation in order to maintain cellular signaling, which is different between non-diseased and diseased skeletal muscle.

    1. Altered organization and density of the MT network in mdx FDB muscle fibers with loss of vertical directionality is not a novelty as well and it has been reported by others (see Randazzo et al., Hum Mol Genet 2019), who also observed that overexpression of a single beta-tubulin (tubb6) in normal Flexor Digitorum Brevis (FDB) muscle fibers mimic the disruption to the MT network of mdx FDB fibers, increases the level of detyrosinated tubulin and increases Nox2 activity (through elevated expression of gp91phox). Conversely, downregulation of the same beta-tubulin restores normal MT organization in mdx FDB. Previous work from the authors (Loehr et al., eLife 2018) reported that in p47-/-/mdx mice MT organization in diaphragm muscle fibers is normalized and autophagy improved. Accordingly, it is puzzling that increased alphatubulin acetylation determines such a wide range of ameliorations in terms of physiological and morphological aspects in dystrophic skeletal muscle fibers treated with Tubastatin A whereas no improvement in the overall MT organization is observed, as reported by Agrawal and colleagues.

    Our findings are also supported by Coleman et al who show that HDAC6 inhibition did not alter levels of DT-tubulin. Although that group did not specifically measure MT organization viewing and analyzing their representative images of alpha-tubulin (Figure 1D, control and tubacin) shows that HDAC6 inhibition does not alter MT organization in wild-type FDBs

    1. Given that p47-/-/mdx mice present with levels of acetylated alpha-tubulin and HDAC6 expression comparable to mdx while showing significant improvement of the dystrophic phenotype despite partial rescue of the autophagic flux (as reported in Loehr et al., eLife 2018), it would have been of great interest to investigate the effect of HDAC6 inhibition in p47-/-/mdx mice as well.

    We would like to thank the reviewer for acknowledging our in-depth characterization of the steps orchestrating the autophagy pathway in the mdx mouse model and the comprehensive evaluation of the phenotype of the mdx mice treated with the HDAC6 inhibitor Tubastatin A. While we believe these experiments are of interest, we think that they merit a detailed investigation that is beyond the scope of the current work

  2. eLife assessment

    This important study pinpoints nitrite oxide synthase 2 activity and decreased microtubule acetylation as distinct regulators of altered autophagic flux that may contribute to pathogenesis in a mouse model of Duchenne muscular dystrophy. While most of the evidence to support these claims is convincing, the claim that autophagy is improved with increased microtubule acetylation is incompletely supported. This work may be of broad interest to muscle biologists and has translational potential for the treatment of Duchenne muscular dystrophy.

  3. Reviewer #1 (Public Review):

    It has been previously shown that defective autophagy and disorganized microtubule network contribute to the pathogenesis of Duchenne muscular dystrophy (DMD). The authors previously reported that nitrite oxide synthase 2 (NOX2) regulates these alterations. It was also shown that acetylated tubulin facilitates autophagosome-lysosome fusion and thus autophagy. In the present study, the authors showed that autophagy is differentially regulated by redox and acetylation modifications in dystrophic mdx mice. The ablation of Nox2 in mdx mice activated the autophagosome maturation but not its fusion with the lysosome. On the other hand, the inhibition of histone acetylase 6 (HDAC6) restored microtubule acetylation, promoted autophagosome-lysosome fusion, and improved muscle function in mdx mice. The strength of this paper is the combination of different approaches to decipher the mechanism, including the evaluation of the level and interaction of several proteins involved in the maturation of autophagosomes and in the fusion between autophagosomes and lysosomes.

    This study reveals an important molecular mechanism by which increasing microtubule acetylation improves autophagy and muscle function in dystrophic mice. This has a translational impact on several diseases in which autophagy is impaired. The improvement of autophagosome-lysosome fusion with HDAC6 inhibitor is supported by several data, but some parts merit further analysis:

    1. To add appropriate controls (e.g. without antibodies) to support protein-protein interaction for all co-immunoprecipitation assays.
    2. The simple evaluation of the protein levels of p62 and LC3-II is not sufficient to claim autophagy improvement after HDAC6 inhibition. It would be good to evaluate the autophagic flux in vivo in all groups of mice (to treat the mice with or without autophagy inhibitor and evaluate whether the difference in the level of LC3-II between the two conditions is higher with HDAC6 inhibitor than without in the mdx mice).
  4. Reviewer #2 (Public Review):

    Agrawal et al. propose an interesting model in which the autophagy pathway in adult mouse skeletal muscle fibers is orchestrated by two independent mechanisms: a) the activity of the NADPH oxidase (Nox) 2 enzyme necessary for autophagosome biogenesis and maturation and b) the level of acetylation of the microtubule (MT) network more selectively responsible for the fusion of the autophagosomes to the lysosomes. Using the well-known mdx mouse, a model for Duchenne muscular dystrophy, the authors perform a quite impressive (but rather traditional) biochemical characterization of the autophagy pathway and found that biogenesis and maturation of the autophagosomes are impaired in mdx mice muscle fibers by means of altered expression of components of the class III phosphatidylinositol 3-kinase complex (PI3K) such as Beclin, VPS15 (both upregulated in mdx mice), ATG14L and VPS34 (both downregulated), and by the reduced expression of JNK and JIP-1, required for the formation of the heterodimer between Beclin and ATG14L-VPS34. In mdx mice, defective nucleation of the phagophore appears to be coupled to altered elongation and expansion as confirmed by decreased expression of WIPI-1, an early marker of autophagosome formation, required for the assembly of the ATG5-12 complex. Clearance of sequestered cytosolic components necessitates the fusion of the autophagosome with the lysosome, a process that the authors found impaired in mdx mice due to altered formation of the SNARE tertiary complex (STX17-SNAP29-VAMP8), as a result of the marked reduction of STX17 expression.

    In a previous work (Pal et al., Nat Commun 2014), the same group described the generation of an mdx-based mouse model where Nox2 activity was abolished by genetic ablation of the p47phox component. These mice presented with a better outcome in terms of dystrophic pathophysiology by means of reduced oxidative stress and improved autophagy. Further characterization of these mice in the present study reveals that in p47-/-/mdx mice abolishment of Nox2 activity restores autophagosome nucleation and maturation thanks to the increased expression of p-JNK, JIP-1 and improved stability of the Beclin-ATG14L complex, but no amelioration is observed on the formation of the SNARE tertiary complex indicating that the biogenesis of autophagosomes is dependent on Nox2 activity but not the fusion between autophagosomes and lysosomes. Given the existing body of evidence in non-muscle cells pointing at alpha-tubulin acetylation as a regulator of MT activity facilitating the fusion of autophagosomes to lysosomes, the authors thought to investigate the level of MT acetylation in mdx mice muscle fibers and found that acetylation is reduced but can be restored by inhibiting the HDAC6 enzyme via the FDA-approved, highly selective pharmacological inhibitor Tubastatin A (Tub A). Treatment of mdx mice at 3 weeks of age (before the onset of pathological manifestations) with Tub A not only restored the normal level of alpha-tubulin acetylation (without altering the organization and density of the MT network) but also curbed the intracellular redox status and improved the autophagic flux by stabilizing the SNARE tertiary complex. Interestingly, treatment of dystrophic mice with Tub A results in substantial improvement of the dystrophic phenotype as confirmed by a reduced level of apoptosis, diminished tissue inflammation, improved sarcolemma integrity, and superior force generation capacity in ex vivo experiments using the diaphragm and Extensor Digitorum Longus (EDL) muscle fibers of Tub A-treated mdx mice compared to untreated mdx and healthy counterparts.

    The in-depth characterization of the steps orchestrating the autophagy pathway in the mdx mouse model on the one hand, and the comprehensive evaluation of the phenotype of the mdx mice treated with the HDAC6 inhibitor Tubastatin A on the other, support the conclusions proposed by the authors. Nonetheless, some aspects deserve consideration.

    1. The effect of increased alpha-tubulin acetylation by means of genetic and pharmacological strategies (i.e., in vivo overexpression of alpha-tubulin acetyltransferase-aTAT1 and treatment with Tubacin or Tubastatin A, respectively) has been previously explored in isolated cardiomyocytes and skeletal muscle fibers and revealed that augmented MT acetylation, due to selective inhibition of HDAC6, increases cytoskeletal stiffness and favors Nox2 activation (Coleman et al., J Gen Physiol 2021).

    2. Altered organization and density of the MT network in mdx FDB muscle fibers with loss of vertical directionality is not a novelty as well and it has been reported by others (see Randazzo et al., Hum Mol Genet 2019), who also observed that overexpression of a single beta-tubulin (tubb6) in normal Flexor Digitorum Brevis (FDB) muscle fibers mimic the disruption to the MT network of mdx FDB fibers, increases the level of detyrosinated tubulin and increases Nox2 activity (through elevated expression of gp91phox). Conversely, downregulation of the same beta-tubulin restores normal MT organization in mdx FDB. Previous work from the authors (Loehr et al., eLife 2018) reported that in p47-/-/mdx mice MT organization in diaphragm muscle fibers is normalized and autophagy improved. Accordingly, it is puzzling that increased alpha-tubulin acetylation determines such a wide range of ameliorations in terms of physiological and morphological aspects in dystrophic skeletal muscle fibers treated with Tubastatin A whereas no improvement in the overall MT organization is observed, as reported by Agrawal and colleagues.

    3. Given that p47-/-/mdx mice present with levels of acetylated alpha-tubulin and HDAC6 expression comparable to mdx while showing significant improvement of the dystrophic phenotype despite partial rescue of the autophagic flux (as reported in Loehr et al., eLife 2018), it would have been of great interest to investigate the effect of HDAC6 inhibition in p47-/-/mdx mice as well.