Author response:
The following is the authors’ response to the previous reviews
Public Reviews:
Reviewer #1 (Public review):
Summary:
In this paper, the authors investigate the effects of Miro1 on VSMC biology after injury. Using conditional knockout animals, they provide the important observation that Miro1 is required for neointima formation. They also confirm that Miro1 is expressed in human coronary arteries. Specifically, in conditions of coronary diseases, it is localized in both media and neointima and, in atherosclerotic plaque, Miro1 is expressed in proliferating cells.
However, the role of Miro1 in VSMC in CV diseases is poorly studied and the data available are limited; therefore, the authors decided to deepen this aspect. The evidence that Miro-/- VSMCs show impaired proliferation and an arrest in S phase is solid and further sustained by restoring Miro1 to control levels, normalizing proliferation. Miro1 also affects mitochondrial distribution, which is strikingly changed after Miro1 deletion. Both effects are associated with impaired energy metabolism due to the ability of Miro1 to participate in MICOS/MIB complex assembly, influencing mitochondrial cristae folding. Interestingly, the authors also show the interaction of Miro1 with NDUFA9, globally affecting super complex 2 assembly and complex I activity.
Finally, these important findings also apply to human cells and can be partially replicated using a pharmacological approach, proposing Miro1 as a target for vasoproliferative diseases.
Strengths:
The discovery of Miro1 relevance in neointima information is compelling, as well as the evidence in VSMC that MIRO1 loss impairs mitochondrial cristae formation, expanding observations previously obtained in embryonic fibroblasts.
The identification of MIRO1 interaction with NDUFA9 is novel and adds value to this paper. Similarly, the findings that VSMC proliferation requires mitochondrial ATP support the new idea that these cells do not rely mostly on glycolysis.
The revised manuscript includes additional data supporting mitochondrial bioenergetic impairment in MIRO1 knockout VSMCs. Measurements of oxygen consumption rate (OCR), along with Complex I (ETC-CI) and Complex V activity, have been added and analyzed across multiple experimental conditions. Collectively, these findings provide a more comprehensive characterization of the mitochondrial functional state. Following revision, the association between MIRO1 deficiency and impaired Complex I activity is more robust.
Although the precise molecular mechanism of action remains to be fully elucidated, in this updated version, experiments using a MIRO1 reducing agent are presented with improved clarity
Although some limitations remain, the authors have addressed nearly all the concerns raised, and the manuscript has substantially improved
Weaknesses:
Figure 6: The authors do not address the concern regarding the cristae shape; however, characterization of the cristae phenotype with MIRO1 ΔTM would have strengthened the mechanistic link between MIRO1 and the MIB/MICOS complex
Although the authors clarified their reasoning, they did not explore in vivo validation of key biochemical findings, which represents a limitation of the current study. While their justification is acknowledged, at least a preliminary exploratory effort could have been evaluated to reinforce the translational relevance of the study.
Finally, in line with the explanations outlined in the rebuttal, the Discussion section should mention the limits of MIRO1 reducer treatment.
Reviewer #2 (Public review):
Summary:
This study identifies the outer‑mitochondrial GTPase MIRO1 as a central regulator of vascular smooth muscle cell (VSMC) proliferation and neointima formation after carotid injury in vivo and PDGF-stimulation ex vivo. Using smooth muscle-specific knockout male mice, complementary in vitro murine and human VSMC cell models, and analyses of mitochondrial positioning, cristae architecture and respirometry, the authors provide solid evidence that MIRO1 couples mitochondrial motility with ATP production to meet the energetic demands of the G1/S cell cycle transition. However, a component of the metabolic analyses are suboptimal and would benefit from more robust methodologies. The work is valuable because it links mitochondrial dynamics to vascular remodelling and suggests MIRO1 as a therapeutic target for vasoproliferative diseases, although whether pharmacological targeting of MIRO1 in vivo can effectively reduce neointima after carotid injury has not been explored. This paper will be of interest to those working on VSMCs and mitochondrial biology.
Strengths:
The strength of the study lies in its comprehensive approach assessing the role of MIRO1 in VSMC proliferation in vivo, ex vivo and importantly in human cells. The subject provides mechanistic links between MIRO1-mediated regulation of mitochondrial mobility and optimal respiratory chain function to cell cycle progression and proliferation. Finally, the findings are potentially clinically relevant given the presence of MIRO1 in human atherosclerotic plaques and the available small molecule MIRO1.
Weaknesses:
(1) High-resolution respirometry (Oroboros) to determine mitochondrial ETC activity in permeabilized VSMCs would be informative.
(2) Therapeutic targeting of MIRO1 failed to prevent neointima formation, however, the technical difficulties of such an experiment is appreciated.
Reviewer #3 (Public review):
Summary:
This study addresses the role of MIRO1 in vascular smooth muscle cell proliferation, proposing a link between MIRO1 loss and altered growth due to disrupted mitochondrial dynamics and function. While the findings are useful for understanding the importance of mitochondrial positioning and function in this specific cell type, the main bioenergetic and mechanistic claims are not strongly supported.
Strengths:
This study focuses on an important regulatory protein, MIRO1, and its role in vascular smooth muscle cell (VSMC) proliferation, a relatively underexplored context.
This study explores the link between smooth muscle cell growth, mitochondrial dynamics, and bioenergetics, which is a significant area for both basic and translational biology.
The use of both in vivo and in vitro systems provides a useful experimental framework to interrogate MIRO1 function in this context.
Weaknesses:
The proposed link between MIRO1 and respiratory supercomplex biogenesis or function is not clearly defined.
Completeness and integration of mitochondrial assays is marginal, undermining the strength of the conclusions regarding oxidative phosphorylation.
We thank the reviewers for their thoughtful and constructive feedback. We appreciate their recognition of our work’s value and the improvements made in this revised version.
We are particularly grateful to Reviewer 3 for their detailed and insightful comments, which identified errors we (and other reviewers) had unfortunately overlooked. To address these concerns and ensure the manuscript meets the high standards of clarity and rigor we aim for, we have made additional corrections and refinements.
As part of this process, we conducted a thorough review of the original source files. This was especially important given that the project spanned from 2018 to 2025, and many co-authors have since left their previous positions.
We appreciate the opportunity to resubmit this manuscript and are confident that these updates fully address the concerns raised by the reviewer and the editorial team.
Reviewer #3 (Recommendations for the authors):
(1) I still do not see the data in WB 2G reflecting the quantification in 2H and 2I. Moreover, the authors state they performed 1 additional experiment, but it appears not to have been included in the analysis of 2H and 2I since the graphs remained the same from the last version of the manuscript.
We apologize for this oversight. The additional experiment has now been incorporated into the analysis for Figures 2H and 2I, and the graphs have been updated accordingly. While we had uploaded the new blot, we inadvertently forgot to update the analysis graphs. Thank you for bringing this to our attention.
(2) The authors talk several times about "supercomplexes 1 and 2" without testing their precise composition (there is a ton of literature about SC species in several mouse cell types, and separate BN-PAGE immunoblotting of individual MRC complexes would precisely define them in this context)
We agree with the reviewer that this is an important point. However, structural differences between supercomplexes were outside the scope of this paper, and we did not perform such analyses. That said, examining the precise composition of supercomplexes could be a valuable direction for future work.
(3) Steady-state levels of MRC subunits do not match the observations from BN-PAGE results. That might be potentially interpreted and explained by the possible accumulation of intermediates but this is not explored.
We appreciate the reviewer’s observation. There is indeed a strong possibility that differences in the expression of structural components of mitochondrial complexes exist between WT and Miro1 -/- cells. However, in this study, we chose to focus on assessing potential differences in the enzymatic activities of the complexes rather than examining their structural composition. Exploring the accumulation of intermediates and structural differences could be an interesting avenue for future investigations.
(4) Citrate synthase normalization of kinetic enzyme activities is claimed, yet it is not shown in any graph and no description of the method is provided.
We sincerely thank the reviewer for pointing out this discrepancy. Upon careful review, we realized that our statement regarding citrate synthase normalization of kinetic enzyme activities in the last revised version was made in error. This was a miscommunication between co-authors, and we did not perform citrate synthase normalization. Instead, the normalization was performed against protein concentration, determined by the BCA assay as described in the manuscript. We regret this oversight and appreciate the opportunity to clarify this.
(5) Complex I activity is still wrongfully described as NADPH oxidation in the methods
We corrected this error.
(6) The authors state 'Thank you for this comment. We believe this is due to a technical issue. Complex IV can be challenging to detect consistently, as its visibility is highly dependent on sample preparation conditions. In this specific case, we suspect that the buffer used during the isolation process may have influenced the detection of Complex IV'. I do not understand this, I find this justification insufficient and not substantiated by any experimental evidence. What buffer has been used for isolation? There are hundreds of protocols for isolation of intact mitochondria and MRC complexes. Also, DDM and digitonin are the gold-standard detergents for MRC complexes isolation and separation via BN-PAGE.
We thank the reviewer for raising this important point. We have revised the response to clarify the exact experimental conditions and to provide supporting data.
For BN-PAGE, mitochondrial fractions purified from cultured VSMCs or aortic tissue were prepared using a standard protocol (now explicitly detailed in the Methods). Briefly, mitochondria were resuspended in 6-aminocaproic acid (ACA) buffer containing 750 mM ACA, 50 mM Bis-Tris (pH 7.0), and protease inhibitors. Forty micrograms of mitochondrial protein were solubilized with 1.5% digitonin, using a final detergent-to-protein ratio of 8:1, and incubated on ice for 20 minutes prior to clarification by centrifugation at 16,000 g for 30 minutes at 4°C. Thus, consistent with established standards, digitonin—one of the gold-standard detergents for MRC complex solubilization and BN-PAGE—was used throughout.
Despite using these widely accepted conditions, we found that detection of fully assembled Complex IV by BN-PAGE was inconsistent, a limitation that has been reported by others and is known to be sensitive to mitochondrial source, tissue type, and solubilization efficiency. To address this directly and avoid over-interpretation, we assessed Complex IV integrity by examining core subunits. As shown in Figure 6—figure supplement 1 (panels B and C), expression levels of MTCO1 and MTCO2, both essential core components of Complex IV, do not differ significantly between WT and Miro1-/- cells, supporting the conclusion that Complex IV abundance is not altered.
We have revised the manuscript to clarify these methodological details and to explicitly state that conclusions regarding Complex IV are based on subunit analysis rather than BN-PAGE visualization alone.
(7) Complex V IGA also does not seem to reflect its quantification.
Thank you for highlighting this concern. To address it, we will include the numerical data alongside the figures to ensure clarity and alignment with our findings. We hope this will provide a more comprehensive understanding and resolve any ambiguity.
(8) Figure 6 supplement 1, the authors state 'we concentrated on ETC1 and 5 and performed experiments in cells after expression of MIRO1 WT and MIRO1 mutants'. I do not understand, what background is being used? what mutants are being expressed? all the figures refer to Miro1 -/- which is, according to standard genetic nomenclature, a loss-of-function allele (KO).
Thank you for your comment. To clarify, we first infected MIRO1fl/fl VSMCs with an adenovirus expressing the DNA recombinase Cre or a control adenovirus. Cells infected with the adenovirus expressing Cre are labeled as MIRO1-/- cells. In these MIRO1-/- cells, we then introduced MIRO1 wild type (WT) and MIRO1 mutants via adenoviral expression.
The mutants include one lacking the transmembrane domain (MIRO1-ΔTM), and another in which the two EF hands of MIRO1 were point-mutated (MIRO1-KK). MIRO1-WT is denoted as Ad WT, the mutant MIRO1-KK as Ad KK, and MIRO1-ΔTM as Ad ΔTM in the figures. We hope this explanation clarifies the experimental background and nomenclature used.
(9) Figure 6 supplement 1B, no normalization is provided (e.g. VDAC, TOM20 etc.). Interestingly, VDAC is then used to normalize the data in C-D-E-F-G. Also, why is MIRO1 detected in lane 4? Is the mutant stable or not? There is zero signal in A.
Thank you very much for pointing out that the immunoblot for VDAC1 was missing in Figure 6—Supplement 1B. This figure has been reviewed several times, and unfortunately, this error was not detected. We sincerely apologize for this oversight. We have now revised the figure to include the immunoblot for VDAC1 to address this issue.
Regarding the detection of MIRO1 in lane 4, we confirm that the "mutant" is not stable. To generate MIRO1 knockout cells, aortic smooth muscle cells from MIRO1fl/fl mice were isolated and cultured, followed by infection with an adenovirus expressing Cre. As these are primary cells and the deletion was induced by Cre expression, the recombination efficiency can vary, which is reflected in the variability observed in lanes 2 and 4 of the immunoblot.
(10) Why are COX4 levels so low in the 2nd replicate in 7A? the authors 'We also performed anti-VDAC immunoblots on the same membranes as alternative loading control (see image below)'. I could not find the image.
Thank you for your comment. The second pair of samples in Figure 7A is from a different preparation of mitochondria. In our experimental design, a control sample and a MIRO1 knockdown sample were processed side by side and run next to each other on the immunoblot.
Regarding the anti-VDAC immunoblot, the image was included in our response to reviewers during the previous revision, as we did not believe it altered the message conveyed by the COX4 blot. However, to ensure clarity and address your concern, we have now included the anti-VDAC immunoblot directly in the figure. We hope this addition resolves any ambiguity and provides further confidence in the data presented.
(11) The proposed interaction between MIRO1 and NDUFA9 is very difficult to reconcile, as the two proteins reside in distinct mitochondrial compartments. MIRO1 is anchored to the outer mitochondrial membrane (OMM), with its functional domains facing the cytosol, whereas NDUFA9 is a matrix-facing accessory subunit of mitochondrial Complex I, positioned at the interface between the N- and Q-modules.
We appreciate the reviewer’s comment and agree that MIRO1 and NDUFA9 occupy distinct mitochondrial compartments. MIRO1 is anchored to the outer mitochondrial membrane with cytosol-facing domains, whereas NDUFA9 is a matrix-facing accessory subunit of Complex I at the N/Q-module interface.
Our data do not suggest a stable, constitutive interaction within intact mitochondria. Rather, the observed association likely reflects an indirect, transient, or context-dependent interaction, potentially occurring during mitochondrial stress, remodeling, or turnover. Such associations may be mediated by multi-protein complexes spanning mitochondrial membranes, dynamic contact sites, or post-lysis interactions detected under experimental conditions. Increasing evidence supports functional coupling between outer mitochondrial membrane proteins and inner membrane or matrix pathways without direct physical binding.
Additional comments:
(12) All the raw data should be provided to the readers (uncropped and annotated WB, IHC images, numerical data with statistics applied).
We agree with the reviewer and appreciate the emphasis on transparency. In accordance with eLife submission requirements, we have provided all raw data. The Source Data files associated with each figure now include uncropped and annotated immunoblots, as well as the numerical source data for all quantified analyses.
During the compilation of these materials, we were unable to locate the original source files for Figure 2A. The control experiment depicted in the previous version, which demonstrates in vitro recombination, was performed in 2018. However, this experiment was repeated several times throughout the project. Therefore, to ensure the manuscript remains complete, we have replaced this panel with a representative immunoblot from a similar experiment. Additionally, during our review, we discovered a labeling error in Figure 3D and G. We have corrected these figures to ensure accuracy.
All source files have been provided and carefully labeled to facilitate independent evaluation.
