Metabolic Rewiring by α-Synuclein Enables Mitohormetic Protection

  1. Shay Geula
  2. Inna Grosheva
  3. Yehudit Zaltsman
  4. Limor Regev
  5. Sabita Chourasia
  6. Atan Gross

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

α-Synuclein (αSyn) is widely associated with Parkinson’s disease pathology, yet its physiological impact on cellular metabolism remains unclear. Here we show that stable αSyn expression in HEK293 cells induces coordinated metabolic remodeling that enhances mitochondrial resilience. αSyn interacts with lactate dehydrogenase A (LDHA) and αSyn–expressing clones exhibit elevated LDHA activity. Moreover, these clones exhibit increased lactate secretion, enhanced glycolysis, and reduced mitochondrial-reactive oxygen species, coupling metabolic rewiring to improved mitochondrial adaptation. Strikingly, following chronic low-dose rotenone preconditioning, αSyn– expressing clones acquire robust resistance to subsequent respiratory complex I inhibition, revealing a potent αSyn–dependent mitohormetic response. These findings identify αSyn as a conditional metabolic modulator that supports mitochondrial adaptation under sustained stress.

Article activity feed

  1. Review Commons

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

    Learn more at Review Commons

    Reply to the reviewers

    We thank you and the reviewers for the thoughtful and constructive evaluation of our manuscript, “Metabolic Rewiring by α-Synuclein Enables Mitohormetic Protection.” We appreciate the reviewers’ recognition that the study addresses an important and relatively underexplored aspect of αSyn biology, namely the physiological contribution of αSyn to cellular metabolism and adaptation to mitochondrial stress.

    In response to the reviewers’ comments, we have substantially revised the manuscript, added new experimental data, and clarified several points of interpretation. Importantly, we now include: (i) comparison of αSyn expression in the HEK293T αSyn-expressing clones with endogenous αSyn levels in mouse brain lysates; (ii) additional analysis of αSyn-immunoreactive high-molecular-weight species following chronic mild-dose rotenone treatment; (iii) a non-crosslinked co-immunoprecipitation control; (iv) PRX6-Flag as a negative control for the LDHA-Flag pull-down; (v) proximity ligation assay data in HEK293T cells; (vi) mitochondrial DNA and mitochondrial membrane potential measurements; (vii) Seahorse analysis of ECAR reserve capacity and OCR following acute and chronic mild-dose rotenone treatment.

    We also revised the wording throughout the manuscript to avoid overinterpretation. In particular, we now refer to LDHA as an αSyn-associated protein and to the PLA data as evidence for close proximity, rather than as definitive proof of direct binding. We also clarify that the growth advantage observed under chronic rotenone represents adaptation to chronic mild mitochondrial stress, not protection from acute severe toxicity.

    In addition, we made a substantial effort to address the relevance of our findings for PD and could successfully demonstrate, using proximity ligation assay, that endogenous αSyn associates with endogenous LDHA in cultured mouse primary hippocampal neurons. The fact that endogenous aSyn and LDHA associate with each other also in neurons argues that this association is likely to be relevant to PD. On the other hand, showing that overexpression of aSyn protects neurons from mitochondrial stress was more challenging, and so far, we could not find an experimental setting to demonstrate this. None-the-less, we laid the foundation for a longer-term approach, and we are in the process of expanding C57BL6 SNCA+/+ and SNCA−/− mouse colonies. These mice, and cultured primary neurons prepared from them, will be the ultimate approach to address the relevance of our findings to PD.

    We were also asked to contextualize our findings against contradictory literature, and thus added the following paragraph to the Discussion: “How do our findings contextualize against contradictory literature? For example, Lee et al reported that αSyn overexpression exacerbates rotenone-induced ATP loss, suggesting that αSyn indirectly sensitizes mitochondria to complex I inhibition by endogenous/exogenous stressors in cells [41]. Our findings challenge this view that αSyn is intrinsically toxic. We demonstrate that αSyn promotes mitohormesis: mild energetic stress induces adaptive bioenergetic remodeling. Based on our findings we propose that αSyn acts as a stress-response protein recruited during increased energetic demand to maintain bioenergetic homeostasis. This response preserves cellular function, whereas repeated demand may exhaust the program, leading to mitochondrial dysfunction, ATP loss, pathological αSyn conversion and cell death.”

    Point-by-point description of the revisions

    Reviewer 1

    Comment 1. Overexpression of α-synuclein does not seem to affect cell growth, but is its level in the range that might occur in living organisms?

    Response: We thank the reviewer for raising this important point. To address it, we added a new Western blot comparison between αSyn expression in HEK293T αSyn-expressing clones and endogenous αSyn levels in mouse brain lysates. This analysis is now shown in Supplementary Fig. 1B. As expected, αSyn expression in the HEK293T stable clones is higher than in the endogenous tissue reference. We now explicitly state this in the Results section and have adjusted the interpretation accordingly. Importantly, despite this elevated expression, αSyn stable overexpression did not affect basal cell growth or transgene stability under standard culture conditions. Thus, while this model is not intended to reproduce endogenous neuronal αSyn levels, it provides a controlled gain-of-function system to uncover αSyn-dependent metabolic effects.

    Changes made:

    We added Supplementary Fig. 1B and revised the Results section to state that αSyn expression in HEK293T clones is higher than in mouse brain lysates. We also added the corresponding Methods section describing preparation of whole-brain lysates.

    Comment 2. Protein-protein interactions of α-synuclein are shown under treatment with supramaximal complex I inhibition. To show relevance of the interaction in resistance to prolonged and mild complex I inhibition, it would be important to show the protein complexes under those conditions.

    Response: We agree with the reviewer that it is important to distinguish between the effects of acute high-dose rotenone and chronic mild-dose rotenone. We therefore analyzed αSyn-immunoreactive high-molecular-weight species after chronic mild-dose (40nM) rotenone treatment. These data are now included in Supplementary Fig. 4. Interestingly, chronic mild-dose rotenone did not increase the intensity of the αSyn-immunoreactive HMW species. This suggests that increased formation of these HMW species is not required for the adaptive growth resilience observed after chronic mild-dose rotenone treatment. We have revised the Results section to discuss this point more explicitly. We also note that chronic treatment may alter the composition, localization, or functional state of αSyn-associated complexes, even if the overall abundance of DSG-captured HMW species is not increased.

    Changes made:

    We added Supplementary Fig. 4 and revised the Results section to clarify that acute and chronic high-/low-dose rotenone has distinct effects on αSyn-immunoreactive HMW species.

    Comment 3. Would it be possible to test the physiological relevance of α-synuclein by silencing/knockout strategy?

    Response: We agree that loss-of-function or endogenous models are important to support physiological relevance. We made a large effort to identify a cell line in which we could detect endogenous aSyn, to knock it out, and were not successful. We could detect endogenous aSyn only in mouse brain lysates (shown in Supplementary Fig. 1B), and therefore we are in the process of expanding C57BL6 SNCA+/+ and SNCA−/− mouse colonies. These mice, and cultured primary neurons prepared from them, will be the ultimate approach to test the physiological relevance of αSyn.

    Comment 4. Table 1 is difficult to understand and is not explained well by the legend.

    Response: We thank the reviewer for pointing this out. We have rewritten the legend to Table 1 to explain the LC-MS/MS analysis, the meaning of the columns, and the interpretation of peptide-spectrum matches, unique peptides, and protein coverage.

    Changes made:

    We revised the Table 1 legend to improve clarity.

    Comment 5. Fig. 2A scale bar 20uM has to be changed to 20µm.

    Response: We thank the reviewer for noting this error. The scale bar label has been corrected to 20 µm.

    Changes made:

    We corrected the scale bar label in Fig. 2A and checked additional figure panels for similar unit-formatting issues.

    Reviewer 2

    Comment 1. For one, the proposed interaction of LDHA and αSyn was only studied by coimmunoprecipitation of tagged and crosslinked proteins after massive transient overexpression in HEK cells. As HEK cells only express very low and negligible amounts of endogenous αSyn this overexpression probably results in vast amounts of mislocalized αSyn. No attempts are described to verify this interaction in a more relevant cellular model with endogenous proteins. I would ask for native co-immunoprecipitation with and without crosslinking and proximity ligation assays from at least something like SH-SY5Y cells which express endogenous αSyn.

    Response: We agree with the reviewer, and below is a description of the new experiments we performed and data added to the revised manuscript:

    • We used PRX6-Flag as a FLAG-tagged negative control protein in the co-immunoprecipitation experiment, and showed that αSyn-HA was detected in the LDHA-Flag pull-down but not in the PRX6-Flag control pull-down, supporting specificity of the LDHA-associated αSyn signal (Fig 1C).
    • we performed co-immunoprecipitation without crosslinkers in HEK293T cells co-expressing LDHA-Flag and αSyn-HA. In the absence of DSG, αSyn was not detected in LDHA-Flag immunoprecipitates )supp Fig. 2B), suggesting that the LDHA-αSyn association is not efficiently preserved under conventional lysis and immunoprecipitation conditions. We envision that the αSyn-LDHA complex requires the intact cellular setting to remain a complex, and DSG preserves this setting.
    • We performed a proximity ligation assay (PLA; [31]) in intact HEK293T cells and in intact mouse primary hippocampal neurons.
    • HEK293T cells: Using antibodies against LDHA and aSyn, we detected a high PLA signal in aSyn-expressing cells, and a very low signal in vector control cells (Supp Fig. 2C).
    • Mouse primary hippocampal neurons: Using antibodies against LDHA and aSyn, we detected endogenous LDHA and endogenous αSyn, without tagged protein overexpression, in neurons prepared from ICR-SNCA+/+ and C57BL6JHUK-SNCA−/− mice. In this experiment we observed a robust PLA signal in ICR-SNCA+/+ neurons and a very low signal in C57BL6JHUK-SNCA−/− neurons (Supp Fig. 3B, C). These new results add important support that aSyn associates with LDHA, but additional data are needed to confirm direct binding, e.g., AlphaFold-based structural predictions to model potential binding interfaces. Therefore, we revised the wording throughout the manuscript, and we now use “associates with”, “αSyn-associated” and “close proximity” where appropriate, and we avoid using “direct interaction”.

    Changes made:

    We added PRX6-Flag control data in Fig. 1C, non-crosslinked co-IP in Supplementary Fig. 2B, PLA in Supplementary Fig. 2C and Supplementary Fig. 3, and revised the wording throughout the manuscript.

    Comment 2. αSyn was expressed with an IRES-GFP but the control contained only GFP which is not ideal. It is unclear why rotenone results in increased levels of αSyn. Less degradation, increased expression? At least a qPCR could help.

    Response: We agree that the vector design should be considered when interpreting the data. The αSyn construct expresses αSyn-IRES-GFP, whereas the control expresses GFP from the corresponding empty IRES-GFP vector. We emphasize that the major metabolic comparisons were performed across multiple independent αSyn-expressing and vector-control clones, reducing the likelihood that the observed effects reflect a single clonal artifact.

    Regarding the 2nd part: “It is unclear why rotenone results in increased…

    Response: We thank the reviewer for this important point. We have revised the text to clarify that we do not interpret the increased αSyn signal after chronic mild-dose rotenone as evidence for transcriptional induction or altered degradation. Because the αSyn construct is linked to IRES-GFP, the shift toward higher GFP intensity after chronic low-dose rotenone strongly suggests enrichment or selective expansion of cells with higher transgene expression, rather than necessarily increased expression within each individual cell. This interpretation is supported by both fluorescence microscopy and flow cytometry showing enrichment of high-GFP cells specifically in the αSyn-expressing population.

    We agree that qPCR could distinguish between transcriptional upregulation and selection of high-expressing cells. However, because the FACS data already show a population shift in GFP intensity, and because the main conclusion is selection or enrichment of high αSyn-expressing cells under chronic mild mitochondrial stress, we have revised the wording to avoid claiming a specific mechanism of increased αSyn expression.

    Changes made:

    We revised the Results and the Fig. 2 legend to describe enrichment of high-GFP and high-αSyn-expressing cells, rather than implying transcriptional induction.

    Comment 3. For the claim that αSyn overexpression shifts metabolism toward increased glycolysis and OXPHOS, important controls are missing. What about cell numbers, mitochondrial mass, mitochondrial membrane potential, etc. Were these experiments done with or without rotenone preconditioning?

    Response: We agree that these controls are essential. ECAR and OCR measurements were normalized to cell number per well, as described in the Methods. In addition, we added new analyses of mitochondrial DNA content and mitochondrial membrane potential. These data are presented in the Supplementary Fig. 5C and 5D and show no differences between vector and αSyn-expressing clones in mitochondrial DNA content or mitochondrial membrane potential, either under basal conditions or after chronic 40nM rotenone treatment. These data support the conclusion that the observed metabolic differences are not simply explained by increased mitochondrial number or altered mitochondrial membrane potential.

    In the revised manuscript, we show that αSyn-expressing clones exhibit higher LDHA activity and increased lactate secretion both under basal conditions and after chronic low/mild-dose (40nM) rotenone treatment. We also added Seahorse analysis under acute and chronic mild-dose rotenone conditions. These data show that chronic mild-dose rotenone treatment abolished mitochondrial respiration in both aSyn- and vector clones (Supp Fig 6). In glycolysis, aSyn clones showed a larger increase in glycolytic capacity as compared to vector clones following chronic mild-dose rotenone treatment (Fig 4A, B).

    Importantly, we also detected differences between the aSyn- and vector clones in glycolytic reserve, which is the value obtained by the subtraction of the glycolysis capacity (max glycolysis) from the basal rate of glycolysis [34]. aSyn–expressing clones preserve high glycolytic reserve following either acute or chronic mild-dose rotenone treatment, whereas similar treatments abolish the glycolytic reserve in vector clones (Fig 4C). These results are consistent with the idea that the presence of aSyn enables cells to continue to proliferate by maintaining the glycolytic energy reserve when mitochondria energy production is abolished.

    Changes made:

    We added Supplementary Fig. 5C, Supplementary Fig. 5D, Fig. 4A and Supplementary Fig. 6, and revised the Seahorse Methods section.

    Comment 4. MitoSOX measures mitochondrial hydrogen peroxide and not mitoROS.

    Response: We agree that the readout should be described with greater precision. According to Thermo Fisher/Invitrogen, MitoSOX-Red is a mitochondrial superoxide indicator. They describe it as a mitochondria-targeted dye whose oxidation is by mitochondrial superoxide, O₂•⁻. MitoSOX-Red fluorescence is commonly used as a mitochondrial superoxide-sensitive signal, but it should not be interpreted as a comprehensive measurement of all mitochondrial reactive oxygen species. To avoid overinterpretation, we revised the relevant text to describe the signal more cautiously as MitoSOX-Red fluorescence or mitochondrial superoxide-sensitive ROS signal, where appropriate, rather than as a broad measure of total mitoROS.

    Changes made:

    We revised the Results, Methods, and figure legend terminology to define the MitoSOX-Red assay more precisely and to avoid overgeneralization.

    Reviewer 3

    Major comment 1. Figure 1C: The co-IP lacks important controls, such as cells expressing an empty vector or a FLAG-tagged non-interacting control protein (e.g., cytosolic GFP-FLAG). Also, IP alone does not sufficiently support a direct interaction between αSyn and LDHA. While co-association in a complex is evident, additional data are needed to confirm direct binding, e.g., AlphaFold-based structural predictions to model potential binding interfaces.

    Response: We agree with the reviewer, and below is a description of the new experiments we performed and data added to the revised manuscript:

    • We used PRX6-Flag as a FLAG-tagged negative control protein in the co-immunoprecipitation experiment, and showed that αSyn-HA was detected in the LDHA-Flag pull-down but not in the PRX6-Flag control pull-down, supporting specificity of the LDHA-associated αSyn signal (Fig 1C).
    • We performed co-immunoprecipitation without crosslinkers in HEK293T cells co-expressing LDHA-Flag and αSyn-HA. In the absence of DSG, αSyn was not detected in LDHA-Flag immunoprecipitates )Supp Fig. 2B), suggesting that the LDHA-αSyn association is not efficiently preserved under conventional lysis and immunoprecipitation conditions. We envision that the αSyn-LDHA complex requires the intact cellular setting to remain a complex, and DSG preserves this setting.
    • We performed a proximity ligation assay (PLA; [31]) in intact HEK293T cells and in intact mouse primary hippocampal neurons.
    • HEK293T cells: Using antibodies against LDHA and aSyn, we detected a high PLA signal in aSyn-expressing cells, and a very low signal in vector control cells (Supp Fig. 2C).
    • Mouse primary hippocampal neurons: Using antibodies against LDHA and aSyn, we detected endogenous LDHA and endogenous αSyn, without tagged protein overexpression, in primary neurons prepared from ICR-SNCA+/+ and C57BL6JHUK-SNCA−/− mice. In this experiment we observed a robust PLA signal in ICR-SNCA+/+ neurons and a very low signal in C57BL6JHUK-SNCA−/− neurons (Supp Fig. 3B, C). Please find below further information about this experiment and our planned future experiments in our reply to Major comment 4.
    • We generated an AlphaFold model of the αSyn-LDHA complex but it did not reach a high enough confidence score, and thus did not include it in the revised manuscript. These new results add important support that aSyn associates with LDHA, but we agree with the reviewer that additional data is needed to confirm direct binding, e.g., AlphaFold-based structural predictions to model potential binding interfaces. Therefore, we revised the wording throughout the manuscript, and we now use “associates with”, “αSyn-associated” and “close proximity” where appropriate, and we avoid using “direct interaction”.

    Changes made:

    We added PRX6-Flag control data in Fig. 1C, non-crosslinked co-IP in Supplementary Fig. 2B, PLA in Supplementary Fig. 2C and Supplementary Fig. 3, and revised the wording throughout the manuscript.

    Major comment 2. Figure 2D shows that preconditioned αSyn-expressing cells exhibit a growth advantage under rotenone treatment, but broader metabolic consequences are absent. How does low-dose rotenone preconditioning impact glycolysis versus OXPHOS? What about lactate secretion or other metabolic readouts? Also, could the authors contextualize these findings against contradictory literature in their Discussion? For example, an old study (doi: 10.1074/jbc.M105326200) reports that both wild-type and mutant αSyn expression exacerbate rotenone-induced mitochondrial membrane potential loss, suggesting αSyn indirectly sensitizes mitochondria to complex I inhibition by endogenous/exogenous stressors in cells.

    Response: We agree with the reviewer that it is important to connect the growth phenotype to metabolic adaptation. In the revised manuscript, we show that αSyn-expressing clones exhibit higher LDHA activity and increased lactate secretion both under basal conditions and after chronic mild-dose (40nM) rotenone treatment. We also added Seahorse analysis under acute and chronic mild-dose rotenone conditions. These data show that chronic mild-dose rotenone treatment abolished mitochondrial respiration in both aSyn- and vector clones (Supp Fig 6). In glycolysis, aSyn clones showed a larger increase in glycolytic capacity as compared to vector clones following chronic mild-dose rotenone treatment (Fig 4A, B).

    Importantly, we also detected differences between the aSyn- and vector clones in glycolytic reserve, which is the value obtained by the subtraction of the glycolysis capacity (max glycolysis) from the basal rate of glycolysis [34]. aSyn–expressing clones preserve high glycolytic reserve following either acute or chronic mild-dose rotenone treatment, whereas similar treatments abolish the glycolytic reserve in vector clones (Fig 4C). These results are consistent with the idea that the presence of aSyn enables cells to continue to proliferate by maintaining the glycolytic energy reserve when mitochondria energy production is abolished.

    Changes made:

    We revised the Results to integrate LDHA activity, lactate secretion, ECAR, glycolytic reserve, and OCR after acute/chronic mild-dose rotenone treatment. Please see the revised Fig. 4 and Supplementary Fig. 6.

    Regarding the reviewer’s request to contextualize our findings against contradictory literature in the Discussion.

    Response: We thank the reviewer for this important suggestion. We have added this reference to the Discussion (Ref 41 in the revised manuscript) and wrote the following paragraph on p. 16: “How do our findings contextualize against contradictory literature? For example, Lee et al reported that aSyn overexpression exacerbates rotenone-induced ATP loss, suggesting that αSyn indirectly sensitizes mitochondria to complex I inhibition by endogenous/exogenous stressors in cells [41]. Our findings challenge this view that αSyn is intrinsically toxic. We demonstrate that aSyn promotes mitohormesis: mild energetic stress induces adaptive bioenergetic remodeling. Based on our findings we propose that αSyn acts as a stress-response protein recruited during increased energetic demand to maintain bioenergetic homeostasis. This response preserves cellular function, whereas repeated demand may exhaust the program, leading to mitochondrial dysfunction, ATP loss, pathological αSyn conversion and cell death.”

    Major comment 3. Figure 4. αSyn clones display elevated basal respiration and maximal respiratory capacity. This phenotype is underexplored and should be integrated earlier rather than sidelined until the Discussion.

    Response: We agree with the reviewer. We revised the Results to more explicitly discuss the observation that αSyn-expressing clones exhibit increased OCR in addition to increased ECAR. We now frame this as evidence that αSyn increases metabolic flexibility and energetic capacity, rather than acting only through a simple shift from OXPHOS to glycolysis. We also added mitochondrial DNA and membrane potential measurements to show that increased respiration is not explained by major changes in mitochondrial DNA content or mitochondria membrane potential. Finally, we added OCR analysis after acute and chronic rotenone exposure, showing that mitochondrial respiration is suppressed by rotenone in both vector and αSyn-expressing clones, supporting the idea that the adaptive advantage under chronic rotenone depends on preserved glycolytic reserve rather than maintained mitochondrial respiration.

    Changes made:

    We revised the Results and Discussion, added Supplementary Fig. 5C, Supplementary Fig. 5D, and Supplementary Fig. 6, and updated the Seahorse Methods section.

    Major comment 4. Considering the relevance of the proposed mechanism for PD, how and would the observed effects hold in neurons? And how certain readouts, e.g. the proliferative advantage observed upon mitohormesis, would apply to post-mitotic neurons? To this end, it would be critical to recapitulate some of the effects in more relevant systems, e.g. patient-derived iPSCs with SNCA multiplications/mutations, αSyn-silenced primary neurons, or through reanalysis of publicly available PD patient datasets.

    Response: We thank the reviewer for raising this important point. In the past six months, we have setup cultures of neurons to begin to address the relevance of our findings for PD.

    As mentioned above, we initially asked whether the association between endogenous LDHA and endogenous aSyn occurs in intact primary neurons. For this purpose, we cultured primary hippocampal neurons derived from ICR )SNCA+/+) mice and from C57BL6JHUK (SNCA−/−) mice (Ref 32 in revised manuscript). We first validated aSyn expression in ICR (SNCA+/+) brain lysates, and its absence from C57BL6JHUK (SNCA−/−) brain lysates, by Western blot analysis using anti-aSyn Abs (Supp Fig 3A). Mouse primary hippocampal neurons prepared from both mouse strains were cultured, and proximity ligation assay (PLA) was performed 12 days-post culture using anti-LDHA and anti-aSyn Abs. This analysis revealed a strong PLA signal in multiple ICR-SNCA+/+ primary neurons, which was largely absent from C57BL6JHUK-SNCA−/− primary neurons (Supp Fig 3B, C). The fact that endogenous aSyn and LDHA associate with each other in neurons argues that this association is likely to be relevant to PD.

    On the other hand, showing that overexpression of aSyn protects neurons from mitochondrial stress was more challenging, and so far, we could not find an experimental setting to demonstrate this. None-the-less, we laid the foundation for a longer-term approach, and we are in the process of expanding C57BL6 SNCA+/+ and SNCA−/− mouse colonies. We envision that these mice, and cultured primary neurons prepared from them, will be the ultimate approach to address the relevance of our findings to PD.

    Minor comment 1. loading control for the Western blot in Figure 1B is missing.

    Response: We thank the reviewer for noting this point. In the revised manuscript, we replaced the original panel with a new subcellular fractionation experiment that more directly addresses the point we intended to make. Specifically, our aim was to determine whether the DSG-cross-linked αSyn-immunoreactive high-molecular-weight (HMW) species are present only in the cytosolic fraction or also in the mitochondria-enriched heavy membrane fraction. To this end, cells from the aSyn-clone #2 were treated with DSG and then subcellularly-fractionated into the cytosolic S100 and the mitochondria-enriched heavy membrane fractions. The results show that both p17-aSyn and the HMW-cross-linked aSyn-immuno-reactive bands/complexes are detected only in the cytosolic S100 fractions (Fig 1B). These results are consistent with the idea that the HMW-cross-linked aSyn-immuno-reactive bands/complexes detected in whole cells are comprised of cytosolic protein(s).

    Changes made:

    We revised Fig. 1B and the corresponding figure legend.

    Minor comment 2. "P17-αSyn" abbreviation (intro, page 6) has never been introduced nor explained.

    Response: We thank the reviewer for pointing this out. We revised the text to introduce this notation clearly as the monomeric approximately 17-kDa αSyn band. Where possible, we also simplified the wording to “monomeric αSyn” to avoid unnecessary confusion.

    Changes made:

    We revised the Results to clarify the meaning of p17-αSyn.

    Minor comment 3. Labels in figures are sometimes not properly explained, e.g., shades of red for aSyn clones in Figures 3 and 4 are not keyed to specific clones.

    Response: We thank the reviewer for this helpful comment. We revised the figures to include a separate legend for each clone, better explain the clone labeling and experimental groups.

    Changes made:

    We revised Figures 3A, 3D, 4A, 5A and 6 to include each clone separately.

  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

    In this manuscript the authors aim to further elucidate the physiological role of aSyn, a protein highly enriched at synaptic terminals that supports key neuronal functions and contributes prominently to Parkinson's disease (PD) pathology when SNCA gene mutations alter/ increase its expression. Based on previous accounts that a) monomeric aSyn alone is not sufficient to promote PD pathology in vivo and b) α-syn can exert cytoprotective effects during oxidative stress or provide neuroprotection in hypoxia-challenged mice when overexpressed, the authors seek to uncover α-syn's metabolic functions.

    Using HEK293T stable clones expressing aSyn and exposed to the complex I inhibitor rotenone, the authors provide evidence that aSyn expression enhances LDHA activity, boosts mitochondrial respiration, and lowers mitochondrial ROS levels. Interestingly, pre-treatment of HEK293T with rotenone discloses a mito-hormetic effect of aSyn which confers a proliferative advantage over control cells. Based on these data the authors conclude that aSyn functions as a metabolic rheostat enabling cells adaptation during stress. While these findings suggest a novel role for aSyn in metabolism with potential implications for PD pathology, stronger mechanistic evidence is needed. Critically, despite aSyn's central role in PD, the study provides no data demonstrating relevance in neuronal cells.

    Major points:

    • Figure 1C: The co- IP lacks important controls, such as cells expressing an empty vector or a FLAG-tagged non-interacting control protein (e.g., cytosolic GFP-FLAG). Also, IP alone does not sufficiently support a direct interaction between aSyn and LDHA. While co-association in a complex is evident, additional data are needed to confirm direct binding, e.g., AlphaFold-based structural predictions to model potential binding interfaces.

    • Figure 2D shows that pre-conditioned aSyn-expressing cells exhibit a growth advantage under rotenone treatment but broader metabolic consequences are absent. How does low-dose rotenone preconditioning impact glycolysis vs OXPHOS? What about lactate secretion or other metabolic readouts? Also, could the authors contextualize these findings against contradictory literature in their Discussion? For example, an old study (doi: 10.1074/jbc.M105326200) reports that both wild-type and mutant aSyn expression exacerbate rotenone-induced mitochondrial membrane potential loss, suggesting aSyn indirectly sensitizes mitochondria to complex I inhibition by endogenous/exogenous stressors in cells.

    • Figure 4. All aSyn clones display elevated basal respiration and maximal respiratory capacity, adding complexity to the narrative of aSyn as direct interactor of LDHA, and leaving also this phenotype rather underexplored: what does it imply for mitochondrial function or bioenergetics? Along these lines, this aSyn-driven increased mitochondrial respiration is introduced but sidelined until the Discussion section, where it's framed as aSyn boosting "overall energetic capacity." This aligns better with the data than overemphasizing LDHA/glycolysis, and I would suggest the authors to integrate and discuss this earlier in the manuscript.

    • Considering the relevance of the proposed mechanism for PD, how and would the observed effects hold in neurons? And how certain readouts, e.g. the proliferative advantage observed upon mitohormesis, would apply to post-mitotic neurons? To this end, it would be critical to recapitulate some of the effects in more relevant systems, e.g. patient-derived iPSCs with SNCA multiplications/mutations, aSyn-silenced primary neurons, or through reanalysis of publicly available PD patient datasets.

    Minor points:

    • A loading control for the Western blot in Figure 1B is missing.

    • "P17-aSyn" abbreviation (intro, page 6) has never been introduced nor explained.

    • Labels in figures are sometimes not properly explained, e.g., shades of red for aSyn clones in Figures 3 and 4 are not keyed to specific clones.

    Significance

    The study proposes a novel metabolic role for aSyn via LDHA/glycolysis modulation which the authors suggest is relevant for OXPHOS-dependent neurons (as also stated at page 16 in the Discussion), but the study requires additional work to substantiate their proposed model, and the exclusive use of HEK293 cells limits the translational impact of the study.

    Audience: basic research.

    Expertise: mitochondrial metabolism, cellular neuroscience, neurodegeneration

  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

    In their contribution, Geul et al. found that chronic overexpression of a-synuclein (aSyn) in human embryonic kidney (HEK293) cells shifts cellular metabolism towards increased glycolysis AND simultaneously increased oxidative phosphorylation (OxPhos). These changes correlated with a proposed interaction of aSyn and lactate dehydrogenase A (LDHA). aSyn is the key component of Lewy bodies, the histopathological hallmark of Parkinson's disease (PD) and aSyn mutation and increased gene dosage causes PD which makes the project scientifically interesting. Previous work has apparently found that shifting metabolism of dopaminergic neurons, which degenerate in PD, towards glycolysis has some protective effect. However, the work has several substantial flaws.

    • For one, the proposed interaction of LDHA and aSyn was only studied by coimmunoprecipitation of tagged and crosslinked proteins after massive transient overexpression in HEK cells. As HEK cells only express very low and negligible amounts of endogenous aSyn this overexpression probably results in vast amounts of mislocalized aSyn. No attempts are described to verify this interaction in a more relevant cellular model with endogenous proteins. I would ask for native co-immunoprecipitation with and without crosslinking and proximity ligation assays from at least something like SH-SY5Y cells which express endogenous aSyn.

    • aSyn was expressed with an IRES-GFP but the control contained only GFP which is not ideal It is unclear why rotenone results in increased levels of aSyn. Less degradation, increased expression? At least a qPCR could help.

    • For the claim that aSyn overexpression shifts metabolism towards increased glycolysis and OxPhos important controls are missing. What about cell numbers, mito mass, mitochondrial membrane potential etc. Were these experiments done with or without rotenone preconditioning?

    • MitoSox measures mitochondrial hydrogen peroxide and not mitoROS

    Significance

    I have my doubts whether the reported findings are relevant for the understanding of aSyn physiology and aSyn-caused PD.

    My expertise is in mitochondrial dysfunction and neurodegeneration. I feel well qualified in assessing this work.

  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

    Geula et al seek insight into the physiological function of a-synuclein in the context of cell metabolism. Misfolded and aggregated form of a-synuclein in neurodegeneration has been a major focus of research but less is known about the function of the physiological monomeric form. The Authors created a stable a-synuclein overexpression in HEK and demonstrated interaction of a-synuclein with LDHA with increased lactate production and decreased ROS production. They also show that a-synuclein overexpressing cells have favorable response to mild and prolonged complex I inhibition in terms of metabolic adaptation and survival. The Authors conclude that a-synuclein has a role in metabolic adaptation during prolonged mitochondrial stress. It is an interesting study and a well written ms.

    The following points need consideration:

    • Overexpression of a-synuclein does not seem to affect cell growth but is its level in the range that might occur in living organisms? Protein protein interactions of a-synuclein are shown under treatment with supramaximal complex I inhibition. To show relevance of the interaction in resistance to prolonged and mild complex I inhibition (40nM rotenone), it would be important to show the protein complexes under those conditions.

    • Would it be possible to test the physiological relevance of a-synuclein by siliencing/knockout strategy?

    • Table 1 is difficult to understand and is not explained well by the legend. Fig2A scale bar 20uM has to be changed to 20um.

    Significance

    I am expert in mitochondrial biology

  5. Version published to 10.64898/2025.12.05.692538 on bioRxiv