Multiomics study of CHCHD10 S59L -related disease reveals energy metabolism downregulation: OXPHOS and β-oxidation deficiencies associated with lipids alterations

  1. Blandine Madji Hounoum
  2. Rachel Bellon
  3. Emmanuelle C Genin
  4. Sylvie Bannwarth
  5. Antoine Lefevre
  6. Lucile Fleuriol
  7. Delphine Debayle
  8. Anne-Sophie Gay
  9. Agnès Petit-Paitel
  10. Sandra Lacas-Gervais
  11. Hélène Blasco
  12. Patrick Emond
  13. Veronique Paquis-Flucklinger
  14. Jean-Ehrland Ricci

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Mutations in the coiled-coil-helix-coiled-coil-helix domain containing 10 ( CHCHD10 ) gene have been associated with a large clinical spectrum including myopathy, cardiomyopathy and amyotrophic lateral sclerosis (ALS). Herein, we analyzed the metabolic changes induced by the p.S59L CHCHD10 mutation to identify new therapeutic opportunities. Using metabolomic, lipidomic and proteomic analysis we observed a strong alteration of metabolism in plasma and heart of Chchd10 S59L/+ mice compared to their wild type littermates at pre-symptomatic and symptomatic stages. In plasma, levels of phospholipids were decreased while those of carnitine derivatives and most of amino acids were increased. The cardiac tissue from Chchd10 S59L/+ mice showed a decreased Oxidative Phosphorylation (OXPHOS) and β-oxidation proteins levels as well as tricarboxylic acid cycle (TCA) intermediates and carnitine pathway metabolism. In parallel, lipidomics analysis reveals a drastic change in the lipidome, including triglycerides, cardiolipin and phospholipids. Consistent with this energetic deficiency in cardiac tissue, we show that L-acetylcarnitine supplementation improves the mitochondrial network length in IPS-derived cardiomyocytes from a patient carrying the CHCHD10 S59L/+ mutation. These data indicate that a bioenergetic intermediate such as L-acetylcarnitine may restore mitochondrial function in CHCHD10 -related disease, due to the reduction in energy deficit that could be compensated by carnitine metabolic pathways.

Article activity feed

  1. Review Commons

    Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    the answers and strategy plans are details in the "revision plan" document attached to this submission

  2. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    This study used metabolomic, lipidomic and proteomic approaches to analyze the metabolic changes induced by the p.S59L CHCHD10 mutation in mice. Altered metabolism in plasma and heart of Chchd10S59L/+ mice was seen compared to their wild type littermate. In plasma, levels of phospholipids were decreased while those of carnitine derivatives and most of amino acids were increased. In cardiac tissue, Chchd10S59L/+ mice showed a decreased Oxidative Phosphorylation (OXPHOS) and -oxidation proteins levels as well as tricarboxylic acid cycle (TCA) intermediates and carnitine pathway metabolism. In parallel, lipidomics analysis revealed changes in the lipidome, including triglycerides, cardiolipin and phospholipids. Consistent with this energetic deficiency in cardiac tissue, the authors show that L-acetylcarnitine supplementation improved the mitochondrial network length in IPS-derived cardiomyocytes from a patient carrying the CHCHD10S59L/+ mutation. It is concluded that L-acetylcarnitine may restore mitochondrial function in CHCHD10-related disease, due to the reduction in energy deficit that could be compensated by carnitine metabolic pathways.

    General Comments:

    The authors had already generated knock-in (KI) mice (Chchd10S59L/+) that developed cardiomyopathy associated with morphological abnormalities of mitochondria, severe oxidative phosphorylation (OXPHOS) deficiency and multiple defects of respiratory chain (RC) complexes activity in several tissues in the late stage. Other authors have seen similar results (see reference 5) As a result, the insights provided by this further metabolomic, lipidomic and proteomic analysis are not that novel. The data is very descriptive and often the authors misinterpret the meaning of the results. Addionally the alterations in either metabolites or proteins have in their mutant CHCHD10S59L/+ mice are often not accurately represented.

    The administration of acetylcarnitine to reverse the metabolic defects in iPSC cardiomyoctytes derived from a from a patient carrying the CHCHD10S59L/+ mutation is interesting, although the authors did not determine if similar benefits of acetylcarnitine were seen in the CHCHD10S59L/+ mice. They msy also misinterprete the role of acetylcarnitine in the mitochondria.

    Specific Comments:

    1. The authors point out that in vitro and in vivo studies have shown that CHCHD10 mutants cause disease by a toxic gain-of-function mechanism rather than by loss of function. It would be helpful to better define what the effects of the CHCHD10S59L/+ mutation represent.
    2. An aim of this study is to understand how mitochondria dysfunction associated with CHCHD10 mutations triggers altered global metabolism. This aim has not really be accomplished in this study.
    3. Figure 3C: The authors put FABP3 in the ß-oxidation pathwayt. This is incorrect. The figure also does not highlight the ketothiolase (ACAT1)? It is also not clear why CPT1 and CPT2 are place together. There is a carnitine translocase situation in between these enzymes. It should also be pointed out that CACT also has role in long chain acylcarnitine translocation.
    4. The main direction of CrAT in heart mitochondria is the conversion of acetyl CoA to acetylcarnitine. This is not recognized by the authors, and has important implications on interpreting acetylcarnitine therapy.
    5. Figure 3C: Why is just palmitate and linoeate shown as fatty acids?
    6. The authors state that "In cells, long-chain fatty acids dependent on esterification with L-carnitine to form acetyl-carnitine for transport from the cytoplasm to the mitochondrial matrix for oxidation and energy production." This sentence should be re-written and broken down into clearer statements.
    7. The authors state that "In lipid biosynthesis pathway, protein levels of long-chain- fatty-acid-CoA ligase 1 (Acsl1; as well as mRNA expression) and mitochondrial short-chain specific acyl-CoA dehydrogenase (Acads) in the hearts of symptomatic Chchd10S59L/+ mice were significantly lower than those of age-matched wild-type mice (with adjusted p < 0.05) (Fig. 3A-B). Please clarify and re-write this sentence. ACSl1 is primarily producing acyl CoA for CPT1. ACADs is involved in ß-oxidation of short chain fatty acids. Not lipid biosynthesis.
    8. The authors state that "With respect to the lipid oxidation, protein levels of carnitine-O-Acetyltransferase (Crat; as well as mRNA expression)...". Crat is not involved in fatty acid oxidation.
    9. It is stated that "Interestingly, a significant decrease in fatty acid biosynthesis intermediates, such as malonate and ethyl-malonate, was also observed in the hearts of symptomatic Chchd10S59L/+ mice (Fig. 3C). I do no not see this data. Where is this?
    10. Figure 3G.; Data is incomplete
    11. The authors state that: Altogether, those results suggest that CHCHD10S59L mutation induces branched-chain amino acids catabolic defects and increased non-essential amino acids synthesis which may contribute to the elevated levels of amino acids metabolites observed in plasma and heart of Chchd10S59L/+ mice (Fig.1D, Fig. 3D)." It should be recognized that the heart is a small contributor to plasma amino acid changes.
    12. The IPSC-derived cardiomyocytes seem to involve one single patient.
    13. The authors state that "As highlighted by our study, Chchd10S59L/+ mice displayed markedly elevated levels of cholesteryl esters, some glycerophospholipids and reduced concentrations of triglycerides species in heart at the symptomatic stage. Changes in cholesteryl ester metabolism were not reflected in changes in plasma total cholesterol pools (Table 1). In other words, it appears that cholesterol synthesis is upregulated but does not result in accumulation of free cholesterol." It should be recognized that the heart is not really involved in the synthesis of cholesterol.
    14. The authors state that "Increased plasma concentrations of acylcarnitines (which formed from carnitine and acyl-CoAs) are suggested as a marker of metabolism disorders related to cardiovascular diseases [43, 44]. Based on our data and the literature, we suggested that targeting carnitine metabolism pathway could counterbalance the metabolic disturbances, ameliorate mitochondrial functions, and therefore delay CHCHD10-related disease progression." This statement is not supported by data. There are also inaccuracies with the discussion of metabolism.

    Significance

    The authors had already generated knock-in (KI) mice (Chchd10S59L/+) that developed cardiomyopathy associated with morphological abnormalities of mitochondria, severe oxidative phosphorylation (OXPHOS) deficiency and multiple defects of respiratory chain (RC) complexes activity. Other authors have seen similar results (see reference 5) As a result, the insights provided by this further metabolomic, lipidomic and proteomic analysis are not that novel. The data is very descriptive and often the authors misinterpret the meaning of the results. They often misrepresent the significance of the observed alterations in metabolites and proteins in their mutant CHCHD10S59L/+ mice.

    The administration of acetylcarnitine to reverse the metabolic defects in iPSC cardiomyoctytes derived from a from a patient carrying the CHCHD10S59L/+ mutation is interesting, although the authors did not determine if similar benefits of acetylcarnitine were seen in the CHCHD10S59L/+ mice.

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    This manuscript by Madji Hounoum et al presents a study of the metabolic changes associated with the p.S59L CHCHD10 mutation that underlies a toxic gain-of-function in a number of disease phenotypes, including ALS and mitochondrial myopathy. The authors have used a multi-omics approach in their mouse model to demonstrate significant decreases in proteins required for oxidative phosphorylation (OXPHOS) and -oxidation, as well as metabolites involved in the Kreb's cycle and carnitine metabolism, along with significant changes in lipid profiles. The authors also provide evidence that L-acetylcarnitine appears to ameliorate mitochondrial function.

    Major comments:

    The manuscript is, in general, well-written and

    1. The conclusions drawn are mostly justified on the basis of the data presented (see comment 2).
    2. The authors have used a multi-omics approach, which is very powerful in allowing a cell-(or organism-) wide analysis of multiple systems. However, the key drawback for the proteomics approach is that low abundance proteins, such as the OXPHOS proteins, are often not accurately assessed. In this regard, it seems critical that the authors provide further evidence for the changes suggested by the heatmaps - or justify that the technical approach they have taken can account for low abundance. Western blotting would fit the bill nicely - there are antibodies available to many of the OXPHOS complex subunits.
    3. There is a very rich and deep literature, particularly from the '80s and '90s, regarding the metabolic disturbances identified in mitochondrial disease patients, both for OXPHOS and fatty acid oxidation defects, that should be cited. In addition, the literature delineating the mitochondrial 'cofactor cocktail' should at least be mentioned.

    Minor comments:

    1. Some attention to grammar would help clarify the meaning for the reader. Some sentences are incomplete.
    2. The authors are encouraged to check their references, as a number are incomplete.
    3. Adding the scale to the scale bar for the MitoTracker Red confocal images would be handy. The right-most micrographs of the S59L/+ cells (Fig 4C) are not nearly as convincing as the middle pair of micrographs - and detract from the message. In addition, it would be nice to have the authors comment on the very intense staining of the mitochondria in the S59L/+ cardiomyocytes in panel A of Figure 4.
    4. The final paragraph of the Discussion is not as impactful as it could be. Specifically, referencing the mitochondrial disease literature would aloow a more fulsome discussion as to the mechanism underlying the action of ALCAR i.e. is the anti-oxidant function most important? Or is it simply it's action as a metabolite?

    Significance

    This study is an important contribution to the literature examining neurodegenerative diseases, such as ALS, that remain poorly understood. While the advance seems incremental, especially in light of the many studies in recent years that have document mitochondrial dysfunction in the mutant mice, the ALCAR treatment is novel and worthy of dissemination. The audience that this manuscript would interest would span from neurologists and other clinicians, such as medical geneticists, to basic biomedical researchers in both academic and industry.

    Reviewer area of expertise: mitochondrial function in health and disease; insufficient expertise to evaluate the details of the 'omics' methodologies.

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    Summary

    In the manuscript titled "Multiomics study of CHCHD10 S59L -related disease reveals energy metabolism downregulation: OXPHOS and beta-oxidation deficiencies associated with lipids alterations," Hounoum et al. performed metabolomics, lipidomics, and proteomics analyses of symptomatic and presymptomatic CHCHD10 S59L murine hearts. The results are largely descriptive. Levels of carnitine acyltransferases and carnitine palmitoyltransferase 2 are lower in S59L hearts. Many proteins of lipid biosynthesis and beta-oxidation pathways are also reduced. Authors concluded that utilization of fatty acids for energy production is deficient in the mutants. They treated CHCHD10 S59L iPS-derived cardiomyocytes with acetyl-L-carnitine and observed a rescue of mitochondrial length and cristae organization in vitro. This paper contains a few grammatical, structural, and organizational issues. The paper confirms several findings recently published by Sayles et al. (Cell Reports, 2022) while providing some additional information through lipidomics analysis in the hearts and metabolomics analysis in the plasma. Overall, it is largely descriptive apart from an experiment in the figure 3 in which the authors use iPSC-derived cardiomyocytes to show partial rescue of a mitochondrial morphology defect following acetyl-L-carnitine supplementation. However, these data seem preliminary and the correlation between the iPSC and mouse model is not established. Overall, these data are likely to be of interest to the immediate field but lack novelty for a general audience.

    Major comments

    1. In Figure 3, authors treated CHCHD10 S59L iPSC-derived cardiomyocytes with acetyl-L-carnitine for 72 hours and observed a rescue of mitochondrial length and cristae organization. However, it is not established that the defect in mitochondrial length is important to the pathology in the mouse heart. A rescue of the mitochondrial functional defects would have greater relevance. Thus, the relevance of rescuing this phenotype is not clear. It is not clear a priori whether increasing acetyl-L-carnitine levels would be protective. While certainly there is evidence for downregulation of fatty acid beta-oxidation as well as carnitine-related metabolites, if this a compensatory response to an OXPHOS defect, increasing acetyl-L-carnitine levels could make the cardiomyopathy worse. Supporting this alternative interpretation, levels of carnitine derivatives are already higher in mutant plasma. Thus, it is important to demonstrate this rigorously and in the most appropriate model. For instance, showing rescue of cardiomyopathy in the in vivo mouse model would be stronger than showing rescue of a minor defect in mitochondrial morphology in iPSC derived cardiomyocytes. Additionally, the EM images provided in the paper are in low magnification, and cristae structures are not clear. Quantifications of the cristae defects are not provided. Authors also do not provide any information regarding how they quantify mitochondrial length. Mitochondria is a 3D structure. The length of mitochondria on a 2D transmission EM image may not represent the true length of the mitochondria.

    Minor comments

    1. Title. Recommend changing "lipids alterations" to "lipid alterations."
    2. Page 7, line 13, 14, and 23. Some characters appear as squares.
    3. Page 12, line 8. Recommend changing "After 72 treated with Acetyl-L-carnitine" to "After treated with acetyl-L-carnitine for 72 hours."
    4. Page 14. It is unclear what the significance of the electrolyte measurements is. A brief discussion of the implications should be added.
    5. Page 15, last sentence. Consider rephrasing the sentence "Very interestingly, most of the amino acids detected in the plasma were increased in CHCHD10 S59L/+ mice at both stages, as well as carnitine, O-acetyl-L-carnitine and deoxycarnitine increased at symptomatic stage."
    6. Page 16, line 10. "Longitudinal proteomics" implies that the pre-symptomatic and symptomatic heart samples came from the same subjects. If this is not the case, I would recommend deleting the word "longitudinal."
    7. Page 17, line 14. Authors show changes in levels of proteins and metabolites in beta-oxidation, but there is no evidence presented to show that beta-oxidation is impaired in CHCHD10 S59L/+ hearts.
    8. Page 17, line 19. Please indicate where Cyb5r1 levels are shown.
    9. Page 18, line 15. Please indicate where Acss1 levels are shown.
    10. Page 18, line 10. Abbreviation Plaa has already been spelled out in line 2.
    11. Page 18, line 10, it is stated here that Plaa levels are lower in the CHCHD10 S59L KO mice but 3 sentences earlier and on the figure they are higher. Please resolve this discrepancy.
    12. Page 18, line 18. Recommend changing "lipids alterations" to "lipid alterations."
    13. Are enzymes involved in amino acid metabolism (PHGDH, PSAT1, and ASNS) upregulated in your proteomics analysis?
    14. Page 20, line 6. Recommend changing "Catabolic enzymes are mainly located in the mitochondrial matrix such as the branched-chain alpha-keto dehydrogenase complex (Bckdha, Bckdhb) and branched-chain-amino-acid aminotransferase (Bcat2)" to "Catabolic enzymes, such as the branched-chain alpha-keto dehydrogenase complex (Bckdha, Bckdhb) and branched-chain-amino-acid aminotransferase (Bcat2), are mainly located in the mitochondrial matrix."
    15. Page 20, line 12. The authors interpret the increase in non-essential amino acid synthesis as a consequence of the changes in branch-chain amino acids. An alternative interpretation is that this is a consequence of the mitochondrial integrated stress response, which has been demonstrated to be upregulated in symptomatic CHCHCD10 S59L hearts. Specifically, expression of proline synthesis, serine metabolism, and asparagine synthesis enzymes is increased in a manner that depends on the OMA1-DELE1-eIF2alpha signaling axis. This alternative explanation for these results should be discussed and the relevant studies cited (PMID: 35700042 and 32338760).
    16. Page 22, line 12. An alternative explanation for lower creatinine levels is decreased body weight in the CHCHD10 S59L mice. This should be discussed in addition to noting changes in creatinine in ALS. Additionally, it seems more likely that these changes in the CHCHD10 S59L mouse relate more to the myopathy/cardiomyopathy in this model than motor neuron disease, which is not a prominent feature in the model.
    17. Page 23, line 10. Recommend showing Uchl1 levels in Results and/or Figures.
    18. Page 23, line 15. Myh6 levels are unchanged or decreased not increased in cardiomyopathy.
    19. Page 24, line 17. Recommend changing "mitochondrial dysfunctional" to "mitochondrial dysfunction."
    20. Page 27, line 10. Recommend changing "ameliorate mitochondrial function" to "ameliorate mitochondrial dysfunction."
    21. Page 31, line 22. Formatting error.
    22. Figure and Figure legend 3D. Recommend changing "not regulated" and "not modulated" to "unchanged" or "not significantly different."
    23. Figure 4C and 4D. Please provide images of higher magnification to clearly show mitochondrial network and cristae. Please provide quantification for 4C.

    Significance

    This paper contains a few grammatical, structural, and organizational issues. The paper confirms several findings recently published by Sayles et al. (Cell Reports, 2022) while providing some additional information through lipidomics analysis in the hearts and metabolomics analysis in the plasma. Overall, it is largely descriptive apart from an experiment in the figure 3 in which the authors use iPSC-derived cardiomyocytes to show partial rescue of a mitochondrial morphology defect following acetyl-L-carnitine supplementation. However, these data seem preliminary and the correlation between the iPSC and mouse model is not established. Overall, these data are likely to be of interest to the immediate field but lack novelty for a general audience.

  5. Version published to 10.1101/2023.01.19.524672v1 on bioRxiv