Editor's synthesis & decision
Minor RevisionRevision requests
- Provide a more explicit discussion on whether the 'up' and 'down' binding poses are energetically equivalent in the absence of geometric strain, and clarify if the observed randomness is intrinsic or a result of the reconstitution.
- Refine the interpretation of the 'interface score' and its relationship to bundle termination, ensuring the distinction between a correlation and a causal mechanism is clearly articulated.
- Clarify the G2 inhibition mechanism by discussing whether it acts as a total block or a shift in binding equilibrium, and provide additional context on the structural 'clash' model (e.g., whether local side-chain relaxation was considered).
- Include standard deviations or error bars for the rotational phase shift measurements in Figure 4 to demonstrate the precision of the analysis across the particle population.
Where this could be published
At its current quality
High-impact structural and molecular biology journals
- Nature Structural & Molecular Biology
- Cell Reports
- Structure
Then you could realistically aim for
High-impact structural and molecular biology journals
- Nature Structural & Molecular Biology
- Nature Communications
- eLife
The manuscript is already at a high level of technical rigor and is well-aligned with the scope of your target venue. The reach band remains consistent because the work is already near its ceiling; the primary improvement needed is a more nuanced discussion of the biological implications of your structural findings. This guidance is not a guarantee of acceptance, and you should perform a final literature search to ensure the novelty of your findings remains current.
edited by model: gemini-flash-lite-latest
Reviewer comments — reproduced in full, unabridged
Reviewer 1
gemini-flash-lite-latestThis manuscript presents a comprehensive structural and organizational analysis of fascin-mediated actin bundling, bridging scales from atomic-level crossbridge interactions to micron-scale bundle architecture. Using cryo-electron microscopy and cryo-electron tomography, the authors reveal that fascin employs two distinct, asymmetric actin-binding sites and undergoes significant conformational plasticity to accommodate the geometric mismatches inherent in packing helical actin filaments into hexagonal arrays. A novel neural-network-based denoising approach is employed to resolve the molecular details of these bundles in situ, demonstrating that fascin's structural flexibility and variable binding modes are essential for bundle formation. The authors further propose a model where the accumulation of energetically unfavorable crosslinks at specific inter-filament interfaces intrinsically limits bundle size. This work provides a significant advance in understanding the mechanical and structural principles of cytoskeletal assembly, with clear implications for the development of fascin-targeted cancer therapeutics.
Strengths
- The use of neural-network-based denoising and semantic segmentation is a powerful and well-validated methodological contribution that enables molecular-resolution analysis of complex, non-helical cytoskeletal bundles.
- The integration of high-resolution single-particle analysis with cryo-ET allows for a coherent, multi-scale model of how fascin's conformational plasticity directly influences the mesoscale architecture of actin bundles.
- The structural explanation for the allosteric inhibition of fascin by G2, supported by the comparison of the prebound, inhibitor-bound, and actin-bound states, is compelling and provides a clear mechanism for drug action.
- The analysis of the energetic cost of bundle expansion, using Mahalanobis distances from subtomogram variability analysis, provides a sophisticated and quantitative framework for understanding the size restriction of filopodia.
Weaknesses
- The manuscript relies on a reconstituted in vitro system, and while it successfully recapitulates many features of filopodia, the discussion could more explicitly address the potential role of other cellular factors (e.g., molecular crowding, other crosslinkers) in modulating bundle size in vivo.
- The claim that "the accumulation of unfavorable crosslinks that limit bundle size" is an emergent property is well-supported by the data, but the link to the specific size distribution observed in cellular filopodia remains somewhat speculative.
- The manuscript mentions that "the 55 kDa F-actin crosslinking protein fascin is critical for biogenesis of filopodia," but the discussion of how the identified structural plasticity might be modulated by post-translational modifications other than S39 phosphorylation is limited.
Detailed comments
The structural characterization of the fascin-actin interface is exceptionally rigorous. The authors state, "the two binding interfaces are highly asymmetric despite the overall shape complementarity between the two fascin ABS and the opposed F-actin clefts." This asymmetry is a key finding that explains how fascin can bridge filaments with varying rotational phases. The detailed breakdown of the residue-level interactions for both ABS1 and ABS2, including the unusual cation-pi interaction between fascin R149 and actin R28, provides a high level of confidence in the model. The authors should ensure that the implications of this asymmetry for the overall bundle symmetry are fully explored in the discussion.
The application of the neural-network denoising approach is a highlight of the study. The authors note, "our denoising approach has limitations, notably the requirement for prior structural knowledge of all system components." This is a fair and important caveat. It would be beneficial for the authors to briefly discuss how this method might be extended to more heterogeneous or multi-component systems, as this would be of broad interest to the structural biology community. The validation of this approach using synthetic data and the subsequent subtomogram averaging results are convincing and demonstrate the robustness of the method.
Comments on figures
- Figure 2D: The detail view of the clash between NP-G2-029 and beta-trefoil 1 is very clear. However, it would be helpful to include a brief explanation in the figure legend or text of how the inhibitor-bound state (PDB 6B0T) was aligned to the actin-bound state, as the orientation of the beta-trefoil 1 domain is the central point of the allosteric inhibition model.
- Figure 4E: The diagram of peripheral filament rotational phase shifts is informative. Could the authors clarify if the rotational phase shifts are measured relative to the central filament in a single bundle element, or if these are averaged values across multiple bundle elements? The variance in these measurements is critical to the claim of structural heterogeneity.
- Figure 7C: The plots of graph metrics versus linkage distance are excellent for demonstrating the stability of different cluster sizes. The authors should clarify if the 'linkage distance' is in arbitrary units or if it has a physical interpretation related to the energy of the crossbridge interfaces.
Referee report rubric — reproduced exactly as submitted
Reviewer 2
gemini-3.1-flash-liteThis manuscript presents a multi-scale structural analysis of fascin-mediated actin bundling, bridging high-resolution cryo-EM of the fascin-actin interface with cryo-ET of reconstituted bundles. The authors demonstrate that fascin’s structural plasticity—specifically, its ability to adopt two distinct binding poses and undergo inter-domain rearrangements—allows it to accommodate the geometric mismatches inherent in packing helical actin filaments into hexagonal arrays. Furthermore, they develop a neural-network-based denoising approach to visualize these assemblies in situ, revealing that bundle size is constrained by the accumulation of energetically unfavorable crosslinks. This work provides a sophisticated, mechanistic explanation for long-standing questions regarding how fascin builds and regulates actin bundles, offering a clear structural basis for its inhibition by G2-type small molecules. The study is highly rigorous, technically innovative, and provides a compelling model for how nanoscale protein dynamics dictate mesoscale cytoskeletal architecture.
Strengths
- Elegant integration of high-resolution single-particle cryo-EM with cryo-ET to bridge molecular mechanisms and mesoscale bundle architecture.
- Development and validation of a neural-network-based denoising and segmentation pipeline that enables molecular-resolution analysis of tomograms.
- Clear structural explanation for the allosteric inhibition of fascin by G2, providing a concrete framework for future drug optimization.
- Rigorous analysis of bundle heterogeneity, moving beyond static models to explain how fascin accommodates symmetry mismatches in actin bundles.
Weaknesses
- The manuscript relies on a specific in vitro reconstituted system, and while it recapitulates many features of filopodia, the degree to which these findings generalize to the complex, multi-protein environment of cellular filopodia remains to be fully established.
- The discussion of bundle size restriction is largely based on energetic favorability derived from subtomogram variability, which, while compelling, remains a correlation that lacks direct experimental validation (e.g., via titration or kinetic studies).
- The reliance on a custom neural-network denoising pipeline, while powerful, introduces a potential 'black box' element that requires careful interpretation of the resulting density maps.
Detailed comments
The authors claim that 'fascin’s two F-actin binding poses and crossbridge conformational flexibility facilitate the incorporation of helical actin filaments into a hexagonal bundle lattice.' This conclusion is well-supported by the 3D classification of bundle elements and the variability analysis. However, it would strengthen the manuscript to explicitly address whether these two poses ('up' and 'down') are energetically equivalent in the absence of geometric strain. The authors state that 'the two modes distributed throughout the bundle element without discernable regularity,' which suggests a lack of cooperativity. A brief discussion on whether this randomness is a result of the reconstitution process or an intrinsic property of the fascin-actin interaction would be beneficial.
Regarding the inhibition mechanism, the authors state that 'G2 binds in a cleft between β-trefoils 1 and 2 distal from both ABS, we infer that it acts as a molecular wedge, which allosterically blocks the inter-domain rearrangements.' This is a logical inference based on the structural comparison provided in Figure 2. To further solidify this, it would be helpful if the authors could comment on whether this 'wedge' mechanism is consistent with the known kinetics of G2 inhibition, or if there is any evidence that G2 might also compete for the actin-binding interface directly, as some small-molecule inhibitors of cytoskeletal proteins do.
The authors note that 'the accumulation of unfavorable inter-filament interfaces restricts bundle size.' This is a fascinating model for size control. However, the manuscript states that 'this analysis does not provide a quantitative explanation for the precise size distribution.' Given the high quality of the cryo-ET data, could the authors provide a more direct correlation between the 'interface scores' and the observed bundle thickness distributions? Specifically, is there a threshold interface score below which filament dissociation becomes statistically probable?
Comments on figures
- Figure 4B: The end-on view and the actin plane 6-0-3 side view clearly illustrate the 'up' and 'down' poses. It would be helpful if the authors could clarify if the 'up' and 'down' designations are strictly defined by the orientation of ABS1 relative to the actin filament, or if there are subtle conformational differences in the fascin molecule itself that correlate with these poses beyond the rigid-body rotation.
- Figure 6C: The plot of 'Cumulative relative angle' versus 'Axial shift' is a key piece of data. The authors show a clear trend, but the scatter in the data is significant. Could the authors clarify if the points in 'shades of grey' (mismatched poses) represent transient, less stable states, or if they are simply a consequence of the inherent flexibility of the fascin-actin interface?
- Figure 7A: The mapping of Mahalanobis distances onto the bundle is a sophisticated approach. However, it is not immediately clear how the 'interface score' is calculated from these distances. A more detailed explanation of the weighting function used to derive the interface score from the Mahalanobis distances would be necessary to ensure the reproducibility of this analysis.
Referee report rubric — reproduced exactly as submitted
Reviewer 3
gemini-flash-lite-latestThis manuscript provides a comprehensive multi-scale structural analysis of fascin-mediated actin bundling, bridging high-resolution cryo-EM of the crossbridge interface with cryo-ET of larger bundle assemblies. The authors demonstrate that fascin’s structural plasticity, specifically inter-domain rotations and two distinct binding poses, allows it to accommodate the geometric mismatches between helical F-actin and the hexagonal lattice of bundles. Furthermore, they develop a neural-network-based denoising and segmentation pipeline to analyze these bundles in situ, revealing that bundle size is limited by the accumulation of energetically unfavorable crosslinking interfaces. This work offers a significant mechanistic advance in understanding how cytoskeletal crosslinkers manage geometric frustration to build complex cellular structures. The study is technically rigorous, well-controlled, and provides a clear structural basis for the allosteric inhibition of fascin by clinical candidates.
Strengths
- Elegant integration of single-particle cryo-EM and cryo-ET to bridge molecular and mesoscale structural phenomena.
- Development and validation of a neural-network-based denoising and segmentation pipeline for cryo-ET, which is clearly explained and made available.
- Clear mechanistic explanation of how fascin’s conformational flexibility and dual binding poses resolve geometric frustration in hexagonal bundles.
- Rigorous structural analysis of the allosteric inhibition mechanism by G2, providing a concrete basis for future inhibitor optimization.
Weaknesses
- The manuscript relies on a specific in vitro reconstituted system; while this is standard, the discussion could more explicitly address potential differences in the crowded cellular environment.
- The hierarchical clustering analysis in Figure 7 is a sophisticated approach, but the link between the 'interface score' and actual physical stability could be strengthened by discussing potential alternative interpretations of the modularity peaks.
- The reliance on custom neural networks for image processing, while powerful, creates a potential barrier to reproducibility for labs without significant computational expertise.
Detailed comments
The authors propose that "the accumulation of unfavorable crosslinks... limit bundle size," which is a compelling model. However, the connection between the Mahalanobis distance of individual crossbridges and the global stability of the bundle is an inference that warrants more caution. Specifically, in Figure 7, the authors state that "boundaries between clusters at the modularity peak corresponded to the worst interfaces." It would be beneficial to clarify whether these 'worst interfaces' are a cause of bundle termination or a consequence of the geometric constraints imposed by the bundle lattice as it grows. A brief discussion on whether these interfaces are inherently unstable or simply less optimal would strengthen the conclusion.
Regarding the allosteric inhibition mechanism, the authors state that G2 acts as a "molecular wedge, which allosterically blocks the inter-domain rearrangements required for fascin to adopt its actin-binding conformation." The structural evidence provided in Figure 2 is convincing. However, to fully support this, it would be helpful to explicitly state whether the G2-bound state is completely incapable of binding F-actin or if it merely shifts the equilibrium. The manuscript mentions that "G2 might disrupt formation of the actin-binding conformation," but a more definitive statement on the binding affinity in the presence of G2 would be valuable for researchers interested in the clinical implications of these inhibitors.
Comments on figures
- Figure 2D: The detail view shows a clash between NP-G2-029 and beta-trefoil 1. The authors should clarify if this clash is based on a rigid-body superposition or if any local side-chain relaxation was permitted. If it is a rigid-body model, it would be helpful to know if the clash is severe enough to sterically preclude binding entirely.
- Figure 4E: The rotational phase shifts are classified into two groups. The authors state that "all six peripheral filaments exhibit distinct rotational phase shifts that vary in direction and magnitude." It would be helpful to see the error bars or standard deviation for these measurements to understand the precision of these phase shift assignments across the particle population.
- Figure 7C: The metrics of transitivity and modularity are used to infer cluster stability. The authors should clarify how sensitive these results are to the choice of 'linkage distance' and if similar trends are observed across all 24 analyzed bundles, or if there is significant variability in the 'optimum stability' point.
Referee report rubric — reproduced exactly as submitted
This document was generated by PsiPi, an AI-assisted manuscript review tool. It is not a product of real peer review, was not issued by any journal, and creates no publication record. The reviews and editorial synthesis above were written by large language models and may contain errors, omissions, or hallucinated claims — treat them as a starting point for revision, not as an authoritative expert assessment.