WASp activity in macrophages prevents mechano-induced inflammation by protecting the nuclear envelope

  1. Roberto Amadio
  2. Giulia Bracchetti
  3. Zahraa Alraies
  4. Giulia Maria Piperno
  5. Lucia Lopez Rodriguez
  6. Mathieu Maurin
  7. Martina Conti
  8. Laura Andolfi
  9. Maria Carmina Castiello
  10. Francesca Ferrua
  11. Anna Villa
  12. Alessandro Aiuti
  13. Ana-Maria Lennon-Dumenil
  14. Federica Benvenuti

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

Mutations in the immune-specific actin regulator WASp induce a proinflammatory state in myeloid cells, whose underlying causes remain poorly defined. Here, we applied microfabricated tools that mimic tissue mechanical forces to explore the role of WASp in connecting mechano-sensing to the activation of inflammatory responses in macrophages. We show that WASp-deficient macrophages carry alterations in nuclear structure and undergo increased blebbing and nuclear rupture when exposed to mechanical confinement. High-resolution imaging indicates that WASp drives the formation of protective perinuclear actin structures in response to confinement. Functionally, a proinflammatory gene signature linked to nuclear envelope rupture is preferentially active in confined WASp null macrophages, which partially depends on the cGAS-STING pathway of cytosolic DNA sensing. Analysis of transcriptional datasets of human and mouse tissue macrophages confirmed elevated inflammatory activation in WASp null cells. Together, these data uncover that WASp restricts pro-inflammatory activation of macrophages by preserving nuclear integrity in confined environments, providing novel clues to understand inflammatory activation in Wiskott-Aldrich Syndrome.

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

    Common reply to all reviewers.

    We sincerely thank the reviewers for their thoughtful evaluation of our work and apologize for the long delay in submitting our response. This delay resulted from the unexpected departure of the post-doc leading the study and the time required to re-establish the necessary expertise.

    In light of these circumstances, and taking the reviewers' comments into account, we substantially revised the manuscript to provide a more focused and coherent story.

    In this revised version we have reorganized original data and added new experiments to support the conclusion that the WASp/Arp2/3 axis controls nuclear resistance to mechanical compression, limiting inflammatory responses arising from nuclear rupture. This study provides the first demostration of an enhanced inflammatory signature related to nuclear mechanics in a WAS disease model and in human patients. Notably, we show that this inflammatory signature is corrected in our own cohort of gene therapy treated patients.

    Efforts to better define the mechanistic link between actin dynamics and nuclear instability did not yield conclusive results, reflecting the complexity of this question, which extends beyond the scope of the present manuscript. As detailed in individual replies and reflected in the text of the revised version, we discuss the potential mechanisms that our data suggest at this stage.

    Reviewer 1

    1.In Figure 1a the authors describe the changes in nuclear area of WKO macrophages. However, in the figure legend it is not mentioned whether the images presented are single z-planes or maximum z-projections. If the former is true, then I would suggest the authors present the maximum z-projections. If the latter is the case, I would suggest the authors to include this information on the figure legend to improve reader's comprehension.

    1. In Line 135 the authors write that the WASp null macrophages display a higher frequency of micronuclei formation and that this is an indication of nuclear disruption. However, the lab of Kenneth Campellone (10.1371/journal.pgen.1010045) has demonstrated that in the absence of Arp-2/3 micronuclei arise from defective chromosomal segregation. I believe the authors should distinguish whether this is the case or it is indeed the nuclear envelope rupture events that produce the micronuclei. The authors could add a sentence in the result and discussion section discussing this possibility.

    2. In figure 1C the authors claim that depletion of WASp results in reduced levels of Lamin-A/C. However, this is very hard to see any difference in the Western Blot provided.

    3. In line 141 the authors mention that loss of WASp results in Lamin-A/C wrinkles and "nuclear irregularities". What are the irregularities mentioned here? Maybe the authors should point them out in the imaging provided in Figure 1 to improve reader comprehension.

    An entire new Figure (revised Figure 2) now presents key nuclear parameters (nuclear areas, Lamin A/C gaps and nuclear protrusions) in confined and unconfined cells. The data show that nuclear envelope ruptures and protrusions are specifically occurring in confined WASp null macrophages.

    Other data that were not conclusive or could not be collected for every condition (micronuclei, DNA damage, Lamin A/C levels by WB) have been excluded from the new version. The possible role of aberrant chromosomal segregation on the nuclear phenotype based on Campellone’s work is at page 11, Discussion.

    In line 177-178 the authors write that there is an increase in nuclear deformation after transwell migration. However, this is not shown. I suggest the authors include a circularity index quantification in the same dataset to support that claim.

    Transwell assay is presented in revised Supplementary Figure 3D and only shows, the rate of cell transmigration. We did not develop further analysis of circularity and DNA damage post-passage as cells were overall too damaged to allow robust quantifications.

    In addition, the difference in the γH2AX foci in post-3µm in wt and WKO conditions appear to be almost identical. I believe it is the post migration difference in DNA damage that should be stressed here. Finally, the authors claim that the WKO macrophages are more nuclear envelope rupture-prone and this results in DNA damage accumulation as has been already described. However, the lab of Jan Lammerding has also demonstrated that nuclear deformation alone can also induce DNA damage accumulation in a cell cycle specific manner via stalled replication forks. The authors should distinguish between the two possibilities with live cell imaging by expressing a DNA damage marker (e.g. 53BP1, coupled with either NLS-GFP or cGAS to assess whether DNA damage accumulates in response to mechanical deformation or nuclear envelope rupture).

    We have not performed additional experiments to document DNA damage in WT and WKO macrophages at steady state or upon confinement. For consistency we excluded the preliminary data on DNA damage as not conclusive at this stage. We do agree that discriminating what comes first in this system is an important question that we have discussed it in the revised version (Page 11).

    In line 204 the authors show that reintroduction of WASp-GFP reduces the frequency in which "Lamin-A/C rupture" events are observed. Then they conclude WASp is important to maintain nuclear stability. While this Lamin-A/C is a good marker for nuclear blebs, I believe that more "classic" nuclear envelope rupture markers should be used to assess the potential compromised nuclear integrity (e.g. NLS-GFP or cGAS). Hence, I would suggest the authors to attempt to replicate this finding with live cell imaging of cells under strong (3µm) confinement and quantify the frequency in which they detect NLS-GFP leakage or perinuclear accumulation of cGAS. Another experiment would be to try and rescue the phenotype observed in the fabricated microchannels used in Figure 2 and observe whether the nuclear envelope rupture events are reduced.

    We agree that showing rescue of NLS cytosolic release upon WASp rescue would be strong. Attempts in this direction were unsuccessful due to limitations in expressing two different exogenous proteins (WASp and NLS) in primary macrophages. We performed rescue experiments in microchannels. Interestingly, overexpression of WASp–GFP in WKO cells strongly inhibited entry into the microchannels, likely because WASp expression exceeded endogenous WT levels (Revised Figure 3E). This finding further supports the conclusion that WASp increases resistance to deformation. However, the marked reduction in the ability to enter channels in cells overexpressing WASp precluded analysis of nuclear ruptures inside microchannels.

    In Figure 4 the authors describe the potential role of WASp in the formation of actin patches in the vicinity of the nucleus. In both Fig4C and Fig4E, the mildly confined (6 µm) condition is missing. I would suggest the authors to quantify the distribution of the different actin structures and the colocalization of WASp/Phalloidin signal also in 6µm confinement. Additionally, in the lines 220-222 the authors claim that the formation of the perinuclear actin ring is absent in wt macrophages, however we still see the perinuclear actin ring in the images

    The actin patches phenotype is now presented in the revised Supplementary Figure 2B, showing only the 3 µm cell height condition, where the phenotype is more pronounced. As we were unable to obtain conclusive evidence linking the presence of actin patches to events of nuclear rupture at the single cell level, these data are now presented as a correlative observations.

    Adding on my previous comment, if the perinuclear actin patches are not present in the absence of confinement, can the authors display the formation of these patches through live cell imaging utilizing probes such as LifeAct or the Actin Chromobody in wild type macrophages?

    As noted above, actin patches, their mechanism of formation, and their causal relationship to nuclear instability were not investigated further and are no longer a primary focus of the revised manuscript.

    While it is intuitive that if WASp is important for the formation of perinuclear actin patches, then this has to be through Arp-2/3 this should be tested. I would suggest the authors to either pharmacologically inhibit Arp2/3 with CK-666 or genetically manipulating the system by either overexpressing a dominant negative version of the Arp2/3 subunit or silencing an essential subunit.

    The actin patches forming under compression are Arp2/3 dependent in dendritic cells, as demonstrated in DCs by a joined effort with our collaborators (Alraies et al, Nature Immunology 2025). We have performed preliminary experiments indicating that this is the case also in macrophages, however these data are not presented in the new version as explained in the previous comments.

    The authors interestingly discovered the upregulated expression of certain inflammatory cytokines. While the authors nicely show that the mRNA levels increase, they do not show whether this translates also in increased protein levels. I would suggest the authors to include an experiment in which they assess the protein levels of a few of the upregulated chemo-/cytokines via immunoblotting or ELISAs. This is also the case for IL-6 mRNA levels showed in Figure 5.

    We agree that measuring protein levels would strengthen the mRNA data. However, the experimental setup does not allow for this. First, the number of cells recovered after confinement is not sufficient for downstream protein analysis. Second, RNA is collected after short periods of confinement (1-4 hrs), as longer confinement cause damage to the cells.

    In FigS1 the authors display that in WKO cells have increased "cellular stiffness" when WASp is depleted from macrophages in the absence of confinement. However, since the cantilever is placed on top of the nuclear area, I am not sure whether the authors were measuring nuclear rigidity or cortical rigidity. In any case, in Fig5 they show that in the absence of confinement factors responsible for branched actin nucleation are downregulated (If I am not wrong, Arp2/3 is a major factor contributing to cortical actin nucleation - along with mDia1, although there might be a compensation mechanism), while the expression of factors related to nuclear mechanics is not altered. This is not in line with the result from the AFM experiment presented in FigS1. I am not sure what does this AFM experiment add to the message of the manuscript. The authors could possibly add a sentence to make the connection to the rest of the manuscript clearer.

    We do agree that this initial attempt to characterize stiffness is not sufficiently developed and we excluded it from the revised version.

    The authors beautifully show that WKO macrophages and AMO's show an upregulation of inflammation related genes. However, I believe it is important to assess whether the upregulation of these signature genes upon confinement can be rescued by the re-introduction of wild-type WASp (and possibly a mutant that is not able to activate Arp-2/3 - to display that is indeed the Arp-2/3 mediated nucleation of these actin patches that limit inflammation).

    We carefully considered performing these experiments. However, both general and study-specific considerations ultimately prevented us from doing so. We are working with primary cells in which transduction efficiency is approximately 50% and positive cells cannot be enriched by sorting, as the procedure induces cellular activation. Therefore, all experiments must be performed on bulk cell populations, in which only around half of the cells are expected to display a rescued phenotype. Given that the increase in IL-6 expression observed in WASp-deficient cells is relatively modest, achieving any significant conclusion would have required a substantial number of biological replicates. This is still challenging with the confiner approach because of inherent variability of the system and limited access to reagents to assemble the device. Overall, while the confiner approach is highly informative for single-cell imaging and can be used for RNA isolation, as we have done here, it is not yet sufficiently developed to allow high-content downstream analyses across multiple experimental conditions.

    The finding that CRISPR/Cas9-mediated depletion of WASp in wild-type cells recapitulates the increase in IL-6 expression (Figure 5I) is consistent with a causal link between WASp deficiency and inflammatory activation.

    In the discussion (line 333-335), the authors mention that the WASp (and I presume the produced actin patches) act as a barrier to prevent nuclear envelope rupture events. From the images provided in the manuscript I get the impression that both WASp and the perinuclear actin patches display a polarized nature. For example, in Figure 4C the actin patch is located at the left of the nucleus. WASp in general is more abundant at the trailing side of the cytoplasm in the migrating macrophage displayed in Fig S4B. Finally, WASp has higher abundancy in a specific perinuclear area in Figure 4E. How these seemingly polarized structures act as barriers? Do they stop nuclear blebs from protruding through lamin B1 gaps? Are nuclear envelope rupture events in migrating macrophages happening specifically in the migrating or trailing end of the nucleus?

    These are all pertinent and logical questions. As explained, being unable to further document the relation between actin patches and NE stability at this stage, the

    occurrence of actin patches in WT is now presented in revised Supplementary Figure 2B, as a correlative observation.

    There has been a report by the lab of Andrea Ablasser (10.1126/science.aaw6421) in which loss of nucleocytoplasmic compartmentalization does not activate the expression of cytokines, as BAF can compete against cGAS for binding of leaked chromatin thus limiting inflammation. If there is not a cell-type specific (or lack of) expression of BAF on BMDMs, how do the authors explain their results in relation to the aforementioned publication?

    We thank the reviewer for raising this important point. We believe the different outcomes are most likely explained by differences in both the experimental model and the confinement protocol. Ablasser and colleagues performed their experiments in HeLa cells, whereas our study uses primary bone marrow-derived macrophages. In addition, their analysis was performed after recovery from confinement, whereas we isolate RNA immediately after release to capture the early transcriptional response to mechanical compression. Consistent with the importance of these kinetics, we find that extending confinement from 1 to 4 hours largely abolishes the inflammatory response in BMDMs. Finally, as discussed in relation to the STING-deficient experiments, we cannot exclude that additional inflammatory pathways contribute to the phenotype observed in WASp-deficient macrophages but are not engaged in HeLa cells.

    Reviewer #2 (Evidence, reproducibility and clarity (Required)):

    Summary: This study from Roberto Amadio and coworkers significantly expands our understanding of WASp by identifying its role in nuclear mechanotransduction, linking actin defects to nuclear instability and inflammation, specifically for macrophages. Although there are several studies related to Dendritic Cells and T cells for example, linking the role of the protein to pro-inflammatory pathways and expression of cytokines, this is the first study that thoroughly links WASp, nuclear instability and downstream pro-inflammatory activation. These findings have broader implications for immune disorders and actinopathies, highlighting WASp as a key regulator of mechanosensing and inflammatory control. Future research should validate these results in vivo, further investigate the actin-based nuclear support system, and explore therapeutic strategies targeting nuclear integrity or cGAS-STING signaling in WAS and related conditions.

    The study provides strong evidence supporting a new role for WASp in nuclear integrity and inflammation control in macrophages. The findings are backed by multiple experimental approaches, including high-resolution imaging, mechanotransduction assays, gene expression profiling, and live-cell tracking, which consistently show increased nuclear deformation, rupture, and inflammatory activation in WASp-deficient macrophages. Reproducibility is reinforced through the use of different experimental models, including genetic reconstitution (WASp-GFP rescue), CRISPR-Cas9 knockout validation, and transcriptomic analysis in both murine and human macrophages, as well as PBMC data from WAS patients pre- and post-gene therapy. The study is clearly structured, with systematic quantification of nuclear defects, inflammation markers, and transcriptional changes. However, some dense data presentation (e.g., transcriptomic heatmaps) and limited discussion of alternative pathways could make certain sections more accessible. Despite these minor limitations, the study's conclusions are well-supported, reproducible, and provide clear mechanistic insights into how WASp regulates nuclear stability and inflammatory signalling.

    Minor comments: For better readability of text, can the authors include the following editions into their main text:

    1. Please define BMDM at its earliest reference (page 5, line 187)
    2. Can the authors provide information regarding how the image processing was performed for heterochromatin content, its distribution and chromocenter analysis?
    3. It would be great if they can explain how the roundedness of nuclei was characterized
    4. Rather than calling it cell mechanics, I would be more comfortable to read it cell stiffness or cell elasticity on line 150, page 4.
    5. For more clarity and quantitative information, can the authors provide numbers everywhere in the main text where comparisons between two datasets are cited to? For example: WKO cells were 20 +/- 10% stiffer than WT ones, etc.

    We thank the reviewer for the positive assessment of our work. As outlined in the introductory comments, we substantially reorganized the manuscript to improve its overall flow while maintaining the main message as highlighted by this reviewer “A new role for WASp in nuclear integrity and inflammation control in macrophages”.

    We have added new data that strengthen the link between WASp, the Arp2/3 pathway, and nuclear stability in confined macrophages. At the same time, we have reduced the emphasis on the role of actin patches in regulating nuclear stability, as the current evidence is not yet sufficiently conclusive.

    Reviewer 3

    In this manuscript cells of WASp null mice bone marrow derived macrophages. They find that lamin A/C levels and RNA decreased while lamin A/C wrinkles increased and chromatin measured indirectly did not change. Next WKO cell were more likely to migrate through constricted channels and had more nuclear rupture events as well as those cells/nuclei undergoing multiple ruptures. Under low artificial confinement WT and WKO nuclei had similar low percentage of nuclear blebbing but upon high artificial confinement WKO nuclei measured a drastic increase in nuclear blebbing and ruptures while blebbing could be partially rescued by expression of WASp-GFP. The manuscript then suggests that perinuclear actin dependent on WASp is the mechanism due the occurrence of what appear to be random patches of actin. Finally, the paper reports upregulation of a subset of genes different from WT and WKO due to confinement. The data are largely of interest to the field but there remains no conclusive evidence to support a clear mechanism for why WKO nuclei undergo nuclear blebbing, rupture, and differential gene expression under artificial confinement.

    The manuscript does not do a sufficient job detailing the cause of the nuclear mechanical changes between lamin A/C and actin. There is data that lamin A/C decreases (Figure1) why is the mechanism not just loss of lamin A/C? Instead random actin patches protect the nucleus? However, the manuscript does not disrupt these actin patches in another manner to show they are indeed important to resisting artificial confinement. An actin depolymerizer with compression might show this? Alternatively, it might not be about actin structures but acto-myosin contraction known to be essential to causing nuclear blebbing and rupture in artificially confinement experiments (Mistriotis et al., 2019 JCB). The fact that lamin A/C is upregulated in WT upon confinement and not in WKO suggests mechanotransduction is not occurring properly as lamin A/C are upregulated under tension which is lost in WKO. This suggests that actin incorporation with the nucleus might be broadly flawed through possibly the LINC complex. The wrinkling of the nuclear lamina in WKO suggests it is actually under less tension, possibly supporting this idea. This continues to point to the fact that this paper, while it has a lot of interesting data, does not appear to have a conclusive understanding of the mechanism occurring in WKO.

    We thank the reviewer for this well-focused review.

    In this revised version, we present novel data adding to the original findings in a format that we believe conveys a more coherent message. Specifically, our data support the conclusion that the nuclei of WASp-deficient macrophages are less resistant to mechanical compression, resulting in the activation of inflammatory pathways in both mouse and human systems.

    We also attempted to define the mechanistic link between actin dynamics and nuclear instability. However, these efforts did not yield conclusive results, reflecting the complexity of this question, which extends beyond the scope of the present manuscript. Accordingly, for consistency, data on actin patches are now presented as correlative evidence (Supplementary Figure 2B) rather than a causal mechanism. Lamin A/C instability and acto-myosin contractility as potential mechanism underlying nuclear fragility are discussed ad proposed by the reviewer (Results page 8, Discussion page 10,11).

    An entire new Figure (revised Figure 2) now presents key nuclear parameters (nuclear areas, Lamin A/C gaps and nuclear protrusions) in confined and unconfined WT and WASp null cells and in cells treated with Arp2/3 inhibitor. Results show that nuclear envelope ruptures and protrusions are specifically occurring in confined WASp null macrophages and even more in cells treated with Arp2/3 inhibitors. Rescue with WASp overexpression corrects the phenotype, reinforcing that the WASp-Arp2/3 axis controls nuclear integrity under mechanical challenge.

    Revised Figure 4 presents results in microchannels, now complemented by new data showing that Arp2/3 inhibitors increase entry rate (phenocopying WASp deficiency) and that rescue with WASp-GFP brings back entry levels below those of WT cells (likely due to overexpression).

    Reviewer #4 (Evidence, reproducibility and clarity (Required)):

    Amadio and co-authors propose a mechanism that links the immune-specific actin regulator WASp with nucleus integrity/mechanosensing and the proinflammatory phenotype observed in WASp-null cells. This work arises from the following three points/previous findings:

    the WAS syndrome, caused by mutations of the WAS protein (WASp), is associated with autoimmune and autoinflammatory manifestations. WASp is an activator of the famous Arp2/3 complex which controls branched actin polymerization.

    immune cells need to migrate in confined and challenging environments and therefore mechanisms of correct mechanosensing are required for their survival and to ensure a working immune system.

    nucleus integrity loss and DNA damage upon external force, sensed by cGAS-STING in the cytosol, can trigger senescence, death or autoimmunity.

    Therefore, the question of how WASp connects mechanosensing to the inflammatory response appears natural and it is indeed a very interesting question. This study would clearly help the understanding of the WAS autoimmune syndrome and explain processes beyond it, as other members of the WASp family might act in similar ways in other contexts such as cancer.

    The paper is well written and the figures are nicely presented. The proposed mechanism is intriguing, however it is not fully clear and not entirely supported by the presented data. Some functional experiment would be needed to demonstrate it, or the statements adapted to what the data support. Also, sometimes statistics are quite weak and some data could be further analyzed.

    The authors suggest the following mechanism, depicted in Figure 5J, here summarized with percentages of cells. The majority (90%) of WT cells sustain mechanical compression without visible nuclear blebs (Figure 3B) and low (but unknown, to be provided) percentage of NE rupture (Figure 3D), thanks to the appearance of an actin-WASp-rich patch in the cytoplasm, close to the cell nucleus, visible in the 30% of the cells (Figure 4B). On the other hand, cells that lack WASp, show nuclear blebs in the 35% of the case (Figure 3B), three times more NE rupture compared to WT (Figure 3D) and the actin patch only in 10% of the cells (Figure 3B, or less? Figure 4D). Therefore, it looks like only a small minority of cells shows what the model proposes as a general working mechanism. Considering that statistical tests are often poor and rescue experiments do not show any clear (or statistically different) result, the authors should support the work with additional experiments and discussion.

    First, physically, how does the presence of an intracellular (stiff?) actin patch (observed in 30% of WT cells) prevent a nucleus from damage? Could the author further explain/comment on this point? More detailed quantifications could be provided, together with a clearer discussion or some key experiment (see later) to demonstrate this idea.

    For example, is the mechanoresponse of cells with/without patch different? Do the nuclei of cells with patch show a different area increase, less blebbing and rupture? This would suggest less "force" transmitted to the nuclei.

    Otherwise, as many WT cells do not show the patch, but the nucleus deals fine with the compression, could a global cytosolic/cortical stiffening explain the mechanism? The patch could be the extreme outcome of this stiffening, therefore observed in fewer cells. In this second option, cytosolic stiffening should be quantified (see AFM point below) and the model explained better.

    We fully recognize the validity of the reviewer's comments and acknowledge the limitations of the original version of the manuscript. As commented in reply to Reviewer 3, we now present a deeply revised version to provide a more coherent message, that is supported by the data.

    The central message of the revised manuscript is that, for the first time, we demonstrate that nuclei of WASp-deficient macrophages are less resistant to mechanical compression, leading to the activation of inflammatory pathways in both mouse and human systems. While we sought to define the mechanism by which altered actin dynamics gives rise to nuclear instability, these efforts were not conclusive at this stage, reflecting the complexity of the underlying biology.

    For consistency, data on actin patches are now presented as correlative evidence (Supplementary Figure 2B) rather than a mechanism.

    Related to both options, functional experiments to prove that the expression of WASp is sufficient to prevent NE rupture, are required. The authors already perform a rescue experiment (that is a very elegant way to prove a mechanism), by over-expressing WASp-GFP in WKO cells, but statistics and numbers are too low. Figure 3F-G shows a reduction in blebbing nuclei in confined cells upon WASp rescue (see below comments about this plot), but it is not clear if it statistically reduces NE ruptures (Figure 3I too weak, low N, no statistical difference shown), if it rescues the formation of the actin patch or if it has any further effect.

    In the rescue experiment (now revised Figure 3A) data show a robust rescue of the blebbing phenotype upon reconstitution (data on n=83 WKO control and n=99 WKO reconstituted cells in N=3 independent experiments). Rescue of Lamin A/C rupture in rescued cells (Revised 3B) was analyzed in fewer cells because of variable reconstitution efficiencies, labeling resolution and confinement efficiency. We now comment this set of data as a trend and not as conclusive evidence.

    It is also not obvious why the authors discarded a more central role of nuclear mechanics in this entire process. It is widely accepted and shown in many studies (Lomakin et al. 2020, Earle et al. 2020 https://doi-org.sire.ub.edu/10.1038/s41563-019-0563-5 ; Cho et al. 2019 https://doi.org/10.1016/j.devcel.2019.04.020) that nuclear envelope composition in general controls nuclear mechanosensing and the ability of nuclei to sustain mechanical force. The authors show a statistically significant downregulation of LaminA/C, but not LaminB, in WKO compared to WT cells. Therefore, WKO nuclei should be softer, as further supported by the higher levels of H3K9me2. This interpretation, that could be supported by AFM indentation of the nuclei, explain why WKO cells enter more easily into the microchannels and potentially the entire mechanism.

    To exclude a direct role of nuclear mechanics in preventing NE break, some experiments could be done. For example, is LaminA/C over-expression in WKO sufficient to rescue the WT phenotype in terms of nuclear blebbing under confinement and NE break? Or, on the other hand, would LaminA/C silencing in WT, lead to increase nucleus blebbing/NE break upon mechanical compression? This set of experiments, together with the ones suggested before, would be key to support the hypothesis proposed in the manuscript and clarify the mechanism, and the over-expression of LaminA/C with transient transfection is not a complicated task.

    We thank the reviewer for raising the role of nuclear mechanics in the process, which we have not excluded. Indeed, we have discussed more extensively this hypothesis in the text of the revised version (Discussion page 10), considering new data showing Lamin A/C rupture (Revised 2C,E) and the original data showing Lamin A/C reduction at steady state and in response to mechanical confinement in WKO (Revised Figure 5F). We have not directly addressed rescue by Lamin A/C overexpression, which will be the objective of future investigations.

    Other major points:

    Related to Figure1: images and quantifications of cell shape are missing. Differences in cell spreading are expected when interfering with actin regulators. As nucleus shape depends on both cell spreading and on the ability of the actin to pull onto the nucleus and flattened it (especially when cells are plated in extremely stiff environments like glass), quantification of cell area and images of the actin cortex would elucidate better the cell phenotype. If the authors already have phalloidin stainings, this analysis is straight forward and does not require additional experiments. An entire new Figure (revised Figure 2) now presents key nuclear parameters (nuclear areas, Lamin A/C gaps and nuclear protrusions) in confined and unconfined WT and WASp null cells, and cells treated with Arp2/3 inhibitor. Results show that nuclear envelope ruptures and protrusions are specifically occurring in confined WASp null macrophages and Arp2/3 treated cells. Rescue with WASp overexpression corrects the phenotype, reinforcing that a WASp-Arp2/3 axis controls nuclear integrity under mechanical challenge. Supplementary Figure 2A shows that cell areas under confinement are not statistically different pointing to a nucleus specific process.

    Revised Figure 4 presents results in microchannels, now complemented by new data showing that Arp2/3 inhibitors increase entry rate (phenocopying WASp deficiency) and that rescue with WASp-GFP brings back entry levels to those of WT cells.

    The nuclei of WKO macrophages do not appear to be much different from WT ones (Figure 1), as claimed in the text, considering a pPresentation of data on nuclear parameters has been deeply revised in new Figure 2 providing different representative images in line with statistical quantifications of key parameters.

    As the authors acquired 3D stacks of nuclei, it would be nice to know if nuclei are different in volume and 3D shape parameters. This would further connect to the first question and clarify if nuclei have a slightly larger area because of a bigger volume or if they are less flattened by the actin cortex. ROBERTO?

    AFM experiments: by indenting a cell of 500nm on top of its nucleus, are cytosolic mechanics being measured or rather cortical/perinuclear ones? Changes in cortical mechanics are expected by knocking-out WASp or any other actin regulator. It is not clear whether the authors are measuring cytosolic mechanics or what. this must be defined better and further discussed. Moreover, as cells are not elastic materials, the physical parameter measured is an "apparent young modulus" and the axis title could be changed. We agree that this assay is not conclusive and in line with the manuscript claim and it has been excluded.

    Regarding the front/back localization of herniations: how do the authors define front/back when imaging only the Hoechst channel in fixed cells? Could it be that some cells start migrating back within the channel? This could be expected during migration. If this was done using live data, nothing has to be changed otherwise the analysis is not robust. We agree with this point. Data are collected from fixed cells and we cannot exclude that cells migrated back. Thus, we excluded this result from the present version.

    • Regarding the NLS-GFP timelapses and kymographs (Figure 2F): why is the NLS-GFP intensity increasing in time, in both WT and WKO, in the first time frames? Shouldn't it be constant in time and change its localization only upon rupture? If the signal increases because the transfection is not at a saturation level, the authors could wait to have a constant signal. To avoid confusion, as the NLS is used as a qualitative tool only and nuclear rupture events should be evident anyway, it would be better to show only parts of the kymographs with a constant baseline intensity. Also, the time scales of images/kymographs are not specified anywhere, this information should be added. This increase in signal is a technical issue we have observed over experiments, linked to instability of the signal acquired by the microscope. As this graphical representation does not add to the message and may create confusion we removed from the present version.

    • Related to the mechanical compression experiments (Figure 3), data regarding the response of WT and WKO nuclei to confinement are missing. Is the area of both population increasing in the same way upon confinement or not (quantify nucleus area before and after confinement)? Are these nuclei unruffling in the same way? These data could be obtained from the already acquired datasets and enriched with some LaminA/C experiment or actin-patch ones (see above).

    Revised Figure 2B shows responses to confinement in WT and WKO with measures showing larger increase in the nucleus area in confined cells WASp ko cells and In Arp2/3 treated cells. The same Figures shows Lamin A/C gaps and ruptures. Ruffling parameters were not captured well in confined cells by technical issues that we could not solve, so the ruffling parameter is no longer part of the revised version.

    Regarding the RNAseq experiments. Why there are three repeats for confined data and two for unconfined ones? Do two confined datasets correspond to the same control or how were the samples acquired? Are the sample batch-corrected? As we were not expecting major differences in the unconfined cells, we kept only two replicates for the control conditions. Samples were not batch-corrected, as were processed together, PCA shows they cluster according to biological conditions, even though confined samples, especially WKO showed greater variability; which is intrinsic of the assay and can be appreciated also in qPCR experiments

    The NER inflammatory score (Figure 5F): the last repeat of WKO 3um shows very high score values, much higher than the other two, clearly affecting global statistics. Also, in Supplementary Figure 5B one point of WKO confined appears quite far from the other two, is this the same one of Figure 5F? Please comment on this.

    The sample showing higher score values is the same that is separated in the PCA. Overall, confiners are difficult to control and inherent variability is expected, especially in bulk downstream approaches. Before running bulk RNA seq, we selected 3 samples with varied Il6 upregulation by RT-PCR. Nevertheless, when looking at the profile of individual genes of the branched actin scores, nuclear mechanical score, NER score and RhoA, the trend is seen in the 3 samples, although to different extents. We validated the prototype inflammatory gene IL-6 in several replicates, finding a robust statistic.

    Overall, even if we acknowledge the intrinsic variability of this assay, we believe this reflect that actual biological variation and does not affect our conclusion. Moreover, we do observe a similar trend in gene corrected WAS patients.

    The fact that WASp shows a milder effect than the complete blockade of Arp2/3 or LaminA/C (discussion lines 372 and following), could be due by the overexpression of other regulators of branched actin? It would not be surprising if cells over-express other actin regulators to compensate the loss of WASp, either to promote branched actin via Arp2/3 or filamentous structures via formins. As the authors performed RNA sequencing, it would be interesting and not complicated to check for the expression of these factors in WKO vs WT cells. We show upregulation of Rac2/RhoA related genes in RNA-seq (Revised Supplementary Figure 4D) and we have discussed that overexpression of compensatory mechanism may be causal to the altered phenotype in WASp null cells.

    Regarding the human dataset analysis, wasn't it known that WKO cells or people with the WAS, have a proinflammatory signature? Could the author clarify the novelty of the analysis? As the reviewer correctly highlights, it is well known that WAS patients show enhanced production of inflammatory cytokines in their peripheral blood.

    However, to the best of our knowledge, this is the first dataset to report a complete transcriptomic analysis of PBMCs data in WAS patients. Most importantly we show a unique cohort of patients before and after gene therapy.

    Minor comments:

    Replies are only for the data that are part of the revised version

    In the figures is not always clear if the representative images and the associated quantifications come from live or fixed cells. This must be clarified for reproducibility.

    Line 81, is the citation of Thiam et al. 2016 correct?

    Line 99 missing citations.

    Line 134: "nuclei are more irregular": apart from the discussion above, an explanation such as "as quantified by nucleus roundness" could be added.

    In Figure 1 and in the text, why do the authors focus on Emerin? This point is not explained at all. Is it only to visualize the nuclear envelope or for any specific reason? A small sentence of explanation should be added.

    As stated in lines 164-165: "the entrance rate of WKO macrophages was consistently higher, especially at smaller constrictions". This is not precisely true, it is rather at the intermediate/larger ones (6-5um), but not at 4 or 3um (at least not statistically significant, or p-values are missing). OK correct in text

    • Figure 2E: fixation and staining for what? Aren't the experiments and analysis done on live cells? This has been corrected

    • Supplementary Figure 2B missing statistical test or all not significant?

    Supplementary Figure 2D why the pre-conditions have such long bars in the upper values? Is it due to a single outlier cell?

    Line 185 "to more faithfully recapitulate deformations events experienced by macrophages in tissues, we next applied a vertical confiner device". The confinement tool is certainly a great system in mechanobiology and can mimic certain types of confinement cells experience in vivo, but it is probably not the most "faithful recapitulation" of an in vivo tissue, right? I would downgrade/rephrase this sentence. Corrected

    • Figure 3E: provide for times (minutes, minutes after confinement is applied) instead of frame numbers in images. Higher magnification would allow to visualize rupture/blebbing better.

    Figure 4D missing statistics? Error bars (doing % in each experiment, mean and standard deviation..)

    Percentages in panel B and D of Figure 4 are not exactly the same or consistent, it is not clear why.

    Supplementary Figure 4B: Why is the average intensity first shown with the three blocks (back-center-front) and then shown across the cell (lower panel)? If the lower panel is an average of n=42 cells, the first panel is not needed. Otherwise, if the lower cell is a single representative cell, what does the label "Avg WASp intensity" refer to? Finally, the plot of the normalized intensity, is done for 1 cell or is it done from the average? Clarify.

    • The nuclear mechanics score includes many genes/proteins, whose regulation would affect nuclear mechanics in different ways. Could the author comment on this or on some specific finding? The fact that nuclear mechanics are altered by confinement, and that this depends on WASp, is very interesting but not discussed. We added discussion of nuclear mechanics at page 10, 11.

    Figure 5 and supplementary Figure 5: color scale miss numbers in several panels.

    • Line 333 cite.

    Line337 in vitro tools mimic, do not recapitulate forces experienced by cells in vivo.

  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 #4

    Evidence, reproducibility and clarity

    R. Amadio and co-authors propose a mechanism that links the immune-specific actin regulator WASp with nucleus integrity/mechanosensing and the proinflammatory phenotype observed in WASp-null cells. This work arises from the following three points/previous findings:

    • the WAS syndrome, caused by mutations of the WAS protein (WASp), is associated with autoimmune and autoinflammatory manifestations. WASp is an activator of the famous Arp2/3 complex which controls branched actin polymerization.
    • immune cells need to migrate in confined and challenging environments and therefore mechanisms of correct mechanosensing are required for their survival and to ensure a working immune system.
    • nucleus integrity loss and DNA damage upon external force, sensed by cGAS-STING in the cytosol, can trigger senescence, death or autoimmunity.
      Therefore, the question of how WASp connects mechanosensing to the inflammatory response appears natural and it is indeed a very interesting question. This study would clearly help the understanding of the WAS autoimmune syndrome and explain processes beyond it, as other members of the WASp family might act in similar ways in other contexts such as cancer.

    The paper is well written and the figures are nicely presented. The proposed mechanism is intriguing, however it is not fully clear and not entirely supported by the presented data. Some functional experiment would be needed to demonstrate it, or the statements adapted to what the data support. Also, sometimes statistics are quite weak and some data could be further analyzed.
    The authors suggest the following mechanism, depicted in Figure 5J, here summarized with percentages of cells. The majority (90%) of WT cells sustain mechanical compression without visible nuclear blebs (Figure 3B) and low (but unknown, to be provided) percentage of NE rupture (Figure 3D), thanks to the appearance of an actin-WASp-rich patch in the cytoplasm, close to the cell nucleus, visible in the 30% of the cells (Figure 4B). On the other hand, cells that lack WASp, show nuclear blebs in the 35% of the case (Figure 3B), three times more NE rupture compared to WT (Figure 3D) and the actin patch only in 10% of the cells (Figure 3B, or less? Figure 4D). Therefore, it looks like only a small minority of cells shows what the model proposes as a general working mechanism. Considering that statistical tests are often poor and rescue experiments do not show any clear (or statistically different) result, the authors should support the work with additional experiments and discussion.

    First, physically, how does the presence of an intracellular (stiff?) actin patch (observed in 30% of WT cells) prevent a nucleus from damage? Could the author further explain/comment on this point? More detailed quantifications could be provided, together with a clearer discussion or some key experiment (see later) to demonstrate this idea. For example, is the mechanoresponse of cells with/without patch different? Do the nuclei of cells with patch show a different area increase, less blebbing and rupture? This would suggest less "force" transmitted to the nuclei.

    Otherwise, as many WT cells do not show the patch, but the nucleus deals fine with the compression, could a global cytosolic/cortical stiffening explain the mechanism? The patch could be the extreme outcome of this stiffening, therefore observed in fewer cells. In this second option, cytosolic stiffening should be quantified (see AFM point below) and the model explained better. Related to both options, functional experiments to prove that the expression of WASp is sufficient to prevent NE rupture, are required. The authors already perform a rescue experiment (that is a very elegant way to prove a mechanism), by over-expressing WASp-GFP in WKO cells, but statistics and numbers are too low. Figure 3F-G shows a reduction in blebbing nuclei in confined cells upon WASp rescue (see below comments about this plot), but it is not clear if it statistically reduces NE ruptures (Figure 3I too weak, low N, no statistical difference shown), if it rescues the formation of the actin patch or if it has any further effect.
    It is also not obvious why the authors discarded a more central role of nuclear mechanics in this entire process. It is widely accepted and shown in many studies (Lomakin et al. 2020, Earle et al. 2020 https://doi-org.sire.ub.edu/10.1038/s41563-019-0563-5 ; Cho et al. 2019 https://doi.org/10.1016/j.devcel.2019.04.020) that nuclear envelope composition in general controls nuclear mechanosensing and the ability of nuclei to sustain mechanical force. The authors show a statistically significant downregulation of LaminA/C, but not LaminB, in WKO compared to WT cells. Therefore, WKO nuclei should be softer, as further supported by the higher levels of H3K9me2. This interpretation, that could be supported by AFM indentation of the nuclei, explain why WKO cells enter more easily into the microchannels and potentially the entire mechanism. To exclude a direct role of nuclear mechanics in preventing NE break, some experiments could be done. For example, is LaminA/C over-expression in WKO sufficient to rescue the WT phenotype in terms of nuclear blebbing under confinement and NE break? Or, on the other hand, would LaminA/C silencing in WT, lead to increase nucleus blebbing/NE break upon mechanical compression? This set of experiments, together with the ones suggested before, would be key to support the hypothesis proposed in the manuscript and clarify the mechanism, and the over-expression of LaminA/C with transient transfection is not a complicated task.

    Other major points:

    • Related to Figure1: images and quantifications of cell shape are missing. Differences in cell spreading are expected when interfering with actin regulators. As nucleus shape depends on both cell spreading and on the ability of the actin to pull onto the nucleus and flattened it (especially when cells are plated in extremely stiff environments like glass), quantification of cell area and images of the actin cortex would elucidate better the cell phenotype. If the authors already have phalloidin stainings, this analysis is straight forward and does not require additional experiments.
    • The nuclei of WKO macrophages do not appear to be much different from WT ones (Figure 1), as claimed in the text, considering a p<0.05 with more than 2000 cells for the area plot and a not significant plot for roundness, while claiming "more irregular nuclei" in the text. Claims in the text must follow the statistical difference of the data as well as representative images (instead of a very elongated nucleus in WKO when the plot says another thing). If 3 images are needed to show 3 types of nuclei, it is maybe recommended to divide the nuclei in 3 categories like round/elongated (roundness <0.7?)/with micronuclei, quantify the % of each category in WT vs WKO and their area separately. Also, are nuclei with micronuclei smaller than the ones without, as it looks from the image? If this is the case, quantifying nuclear area and roundness for each category might support the statistics. As the authors acquired 3D stacks of nuclei, it would be nice to know if nuclei are different in volume and 3D shape parameters. This would further connect to the first question and clarify if nuclei have a slightly larger area because of a bigger volume or if they are less flattened by the actin cortex.
    • AFM experiments: by indenting a cell of 500nm on top of its nucleus, are cytosolic mechanics being measured or rather cortical/perinuclear ones? Changes in cortical mechanics are expected by knocking-out WASp or any other actin regulator. It is not clear whether the authors are measuring cytosolic mechanics or what. this must be defined better and further discussed. Moreover, as cells are not elastic materials, the physical parameter measured is an "apparent young modulus" and the axis title could be changed.
    • Regarding the front/back localization of herniations: how do the authors define front/back when imaging only the Hoechst channel in fixed cells? Could it be that some cells start migrating back within the channel? This could be expected during migration. If this was done using live data, nothing has to be changed otherwise the analysis is not robust.
    • Regarding the NLS-GFP timelapses and kymographs (Figure 2F): why is the NLS-GFP intensity increasing in time, in both WT and WKO, in the first time frames? Shouldn't it be constant in time and change its localization only upon rupture? If the signal increases because the transfection is not at a saturation level, the authors could wait to have a constant signal. To avoid confusion, as the NLS is used as a qualitative tool only and nuclear rupture events should be evident anyway, it would be better to show only parts of the kymographs with a constant baseline intensity. Also, the time scales of images/kymographs are not specified anywhere, this information should be added.
    • Related to the mechanical compression experiments (Figure 3), data regarding the response of WT and WKO nuclei to confinement are missing. Is the area of both population increasing in the same way upon confinement or not (quantify nucleus area before and after confinement)? Are these nuclei unruffling in the same way? These data could be obtained from the already acquired datasets and enriched with some LaminA/C experiment or actin-patch ones (see above).
    • Regarding the RNAseq experiments. Why there are three repeats for confined data and two for unconfined ones? Do two confined datasets correspond to the same control or how were the samples acquired? Are the sample batch-corrected? The NER inflammatory score (Figure 5F): the last repeat of WKO 3um shows very high score values, much higher than the other two, clearly affecting global statistics. Also, in Supplementary Figure 5B one point of WKO confined appears quite far from the other two, is this the same one of Figure 5F? Please comment on this.
    • The fact that WASp shows a milder effect than the complete blockade of Arp2/3 or LaminA/C (discussion lines 372 and following), could be due by the overexpression of other regulators of branched actin? It would not be surprising if cells over-express other actin regulators to compensate the loss of WASp, either to promote branched actin via Arp2/3 or filamentous structures via formins. As the authors performed RNA sequencing, it would be interesting and not complicated to check for the expression of these factors in WKO vs WT cells.
    • Regarding the human dataset analysis, wasn't it known that WKO cells or people with the WAS, have a proinflammatory signature? Could the author clarify the novelty of the analysis?

    Minor comments:

    • In the figures is not always clear if the representative images and the associated quantifications come from live or fixed cells. This must be clarified for reproducibility.
    • Line 81, is the citation of Thiam et al. 2016 correct?
    • Line 99 missing citations.
    • Line 134: "nuclei are more irregular": apart from the discussion above, an explanation such as "as quantified by nucleus roundness" could be added.
    • In Figure 1 and in the text, why do the authors focus on Emerin? This point is not explained at all. Is it only to visualize the nuclear envelope or for any specific reason? A small sentence of explanation should be added.
    • As stated in lines 164-165: "the entrance rate of WKO macrophages was consistently higher, especially at smaller constrictions". This is not precisely true, it is rather at the intermediate/larger ones (6-5um), but not at 4 or 3um (at least not statistically significant, or p-values are missing).
    • Figure 2E: fixation and staining for what? Aren't the experiments and analysis done on live cells?
    • Supplementary Figure 2B missing statistical test or all not significant?
    • Supplementary Figure 2D why the pre-conditions have such long bars in the upper values? Is it due to a single outlier cell?
    • Line 185 "to more faithfully recapitulate deformations events experienced by macrophages in tissues, we next applied a vertical confiner device". The confinement tool is certainly a great system in mechanobiology and can mimic certain types of confinement cells experience in vivo, but it is probably not the most "faithful recapitulation" of an in vivo tissue, right? I would downgrade/rephrase this sentence.
    • Figure 3E: provide for times (minutes, minutes after confinement is applied) instead of frame numbers in images. Higher magnification would allow to visualize rupture/blebbing better.
    • Figure 4D missing statistics? Error bars (doing % in each experiment, mean and standard deviation..)
    • Percentages in panel B and D of Figure 4 are not exactly the same or consistent, it is not clear why.
    • Supplementary Figure 4B: Why is the average intensity first shown with the three blocks (back-center-front) and then shown across the cell (lower panel)? If the lower panel is an average of n=42 cells, the first panel is not needed. Otherwise, if the lower cell is a single representative cell, what does the label "Avg WASp intensity" refer to? Finally, the plot of the normalized intensity, is done for 1 cell or is it done from the average? Clarify.
    • The nuclear mechanics score includes many genes/proteins, whose regulation would affect nuclear mechanics in different ways. Could the author comment on this or on some specific finding? The fact that nuclear mechanics are altered by confinement, and that this depends on WASp, is very interesting but not discussed.
    • Figure 5 and supplementary Figure 5: color scale miss numbers in several panels.
    • Line 333 cite.
    • Line337 in vitro tools mimic, do not recapitulate forces experienced by cells in vivo.

    Referees cross-commenting

    Agree with the questions of other reviewers, especially on the points needed to prove the mechanisms such as the point on nuclear envelope break vs DNA damage, or the role of Arp2/3 raised by Reviewer #1 and the LaminA/C vs actin patch question raised by Reviewer #3. In my opinion, not all the suggestions made by the 4 reviewers are strictly required to consider this work for publication, but the mechanism has to be supported by more experiments, including some functional ones.

    Significance

    R. Armadio and co-authors propose a cellular mechanism that controls, at the single cell level, nuclear integrity in confined environments and, at the tissue level, inflammation response. To show this new role of WASp in controlling nucleus shape and integrity, the authors used various tools to compress cells in vitro and provide an extensive RNA profile of both WT and WKO cells, with or without compression. This work fits together with other recent papers, cited by the authors, like Delgado et al. 2024, that bridge extremely relevant mechanobiology findings, with more physiological problems and diseases. Therefore the mechanism, if demonstrated properly, can be interesting for the biophysics community, the general mechanobiology field as well as for people with a more medical background (because of the implications in the syndrome). In mechanobiology this work opens new questions regarding the role of other members of the WASp family in controlling cellular mechano-sensing and nucleus integrity in other tissues or different DNA damage sensing mechanisms. From a physical perspective, it is very intriguing to think a (stiff?) patch in the cytosol of cells could protect the nucleus from damage and to think at possible other implications of this idea. As evident from the comments, this is written from a biophysics/mechanobiology background.

  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 #3

    Evidence, reproducibility and clarity

    In this manuscript cells of WASp null mice bone marrow derived macrophages. They find that lamin A/C levels and RNA decreased while lamin A/C wrinkles increased and chromatin measured indirectly did not change. Next WKO cell were more likely to migrate through constricted channels and had more nuclear rupture events as well as those cells/nuclei undergoing multiple ruptures. Under low artificial confinement WT and WKO nuclei had similar low percentage of nuclear blebbing but upon high artificial confinement WKO nuclei measured a drastic increase in nuclear blebbing and ruptures while blebbing could be partially rescued by expression of WASp-GFP. The manuscript then suggests that perinuclear actin dependent on WASp is the mechanism due the occurrence of what appear to be random patches of actin. Finally, the paper reports upregulation of a subset of genes different from WT and WKO due to confinement. The data are largely of interest to the field but there remains no conclusive evidence to support a clear mechanism for why WKO nuclei undergo nuclear blebbing, rupture, and differential gene expression under artificial confinement.

    The manuscript does not do a sufficient job detailing the cause of the nuclear mechanical changes between lamin A/C and actin. There is data that lamin A/C decreases (Figure1) why is the mechanism not just loss of lamin A/C? Instead random actin patches protect the nucleus? However, the manuscript does not disrupt these actin patches in another manner to show they are indeed important to resisting artificial confinement. An actin depolymerizer with compression might show this? Alternatively, it might not be about actin structures but acto-myosin contraction known to be essential to causing nuclear blebbing and rupture in artificially confinement experiments (Mistriotis et al., 2019 JCB).

    The fact that lamin A/C is upregulated in WT upon confinement and not in WKO suggests mechanotransduction is not occurring properly as lamin A/C are upregulated under tension which is lost in WKO. This suggests that actin incorporation with the nucleus might be broadly flawed through possibly the LINC complex. The wrinkling of the nuclear lamina in WKO suggests it is actually under less tension, possibly supporting this idea. This continues to point to the fact that this paper, while it has a lot of interesting data, does not appear to have a conclusive understanding of the mechanism occurring in WKO.

    Measuring cytoplasm stiffness makes no sense in Sup Fig 1.

    Significance

    This study provides unique evidence that WASp is important in nuclear mechanobiology though the mechanism is not clear.

    This data will be a great interest to the field of mechanobiology, but revisions will be required to clarify the underlying mechanism of WASp action in maintaining nuclear integrity.

  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 #2

    Evidence, reproducibility and clarity

    Summary: This study from Roberto Amadio and coworkers significantly expands our understanding of WASp by identifying its role in nuclear mechanotransduction, linking actin defects to nuclear instability and inflammation, specifically for macrophages. Although there are several studies related to Dendritic Cells and T cells for example, linking the role of the protein to pro-inflammatory pathways and expression of cytokines, this is the first study that thoroughly links WASp, nuclear instability and downstream pro-inflammatory activation. These findings have broader implications for immune disorders and actinopathies, highlighting WASp as a key regulator of mechanosensing and inflammatory control. Future research should validate these results in vivo, further investigate the actin-based nuclear support system, and explore therapeutic strategies targeting nuclear integrity or cGAS-STING signaling in WAS and related conditions.

    The study provides strong evidence supporting a new role for WASp in nuclear integrity and inflammation control in macrophages. The findings are backed by multiple experimental approaches, including high-resolution imaging, mechanotransduction assays, gene expression profiling, and live-cell tracking, which consistently show increased nuclear deformation, rupture, and inflammatory activation in WASp-deficient macrophages. Reproducibility is reinforced through the use of different experimental models, including genetic reconstitution (WASp-GFP rescue), CRISPR-Cas9 knockout validation, and transcriptomic analysis in both murine and human macrophages, as well as PBMC data from WAS patients pre- and post-gene therapy. The study is clearly structured, with systematic quantification of nuclear defects, inflammation markers, and transcriptional changes. However, some dense data presentation (e.g., transcriptomic heatmaps) and limited discussion of alternative pathways could make certain sections more accessible. Despite these minor limitations, the study's conclusions are well-supported, reproducible, and provide clear mechanistic insights into how WASp regulates nuclear stability and inflammatory signalling.

    Minor comments: For better readability of text, can the authors include the following editions into their main text:

    1. Please define BMDM at its earliest reference (page 5, line 187)
    2. Can the authors provide information regarding how the image processing was performed for heterochromatin content, its distribution and chromocenter analysis?
    3. It would be great if they can explain how the roundedness of nuclei was characterized
    4. Rather than calling it cell mechanics, I would be more comfortable to read it cell stiffness or cell elasticity on line 150, page 4.
    5. For more clarity and quantitative information, can the authors provide numbers everywhere in the main text where comparisons between two datasets are cited to? For example: WKO cells were 20 +/- 10% stiffer than WT ones, etc.

    Significance

    Strengths: This study uncovers a novel function of WASp in nuclear integrity maintenance and inflammation control, expanding its known role beyond actin polymerization at the cell cortex. The findings are supported by rigorous experimental approaches, including high-resolution imaging, live-cell tracking, gene expression profiling, and mechanotransduction assays. Additionally, human data from WAS patients validate the clinical relevance, showing that gene therapy can partially reverse inflammation. The study's multi-model approach, including genetic reconstitution (WASp-GFP rescue) and CRISPR validation, ensures reproducibility and robustness.

    Limitations: Despite strong in vitro evidence, alternative pathways contributing to nuclear rupture, such as RhoA/ROCK signaling, are not fully explored. Finally, while transcriptomic data from WAS patients reinforce key findings, sample variability limits generalization.

    This study provides a conceptual shift in our understanding of WASp's function, demonstrating that it is not only an actin regulator but also a key protector of nuclear stability under mechanical stress. It links mechanosensing to immune activation, showing that nuclear rupture in WASp-deficient macrophages triggers cGAS-STING-mediated inflammation. This establishes a new connection between cytoskeletal defects and nuclear-driven inflammatory pathways, offering a mechanistic explanation for the chronic inflammation seen in Wiskott-Aldrich Syndrome.

    This work will be cruicial for biologists, in particular from the field of oncology and immunology, and biophysicists to explore the link between nuclear mechanics and WAS protein expression in health and disease.

    I am a biophysicist and experimental physicist with background in cell mechanics and immunology.

  5. 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

    In the manuscript entitled "WASp activity in macrophages prevents mechano-induced inflammation by 1 protecting the nuclear envelope" Amadio and colleagues demonstrate that WASp-mediated actin polymerization at perinuclear areas protect against nuclear envelope rupture events and subsequent upregulation of inflammation signature genes.

    I found this study intriguing but I do have some concerns and some missing experiments that if addressed by the authors, I would recommend the paper for publication.

    Comments:

    1. In Figure 1a the authors describe the changes in nuclear area of WKO macrophages. However, in the figure legend it is not mentioned whether the images presented are single z-planes or maximum z-projections. If the former is true, then I would suggest the authors present the maximum z-projections. If the latter is the case, I would suggest the authors to include this information on the figure legend to improve reader's comprehension.
    2. In Line 135 the authors write that the WASp null macrophages display a higher frequency of micronuclei formation and that this is an indication of nuclear disruption. However, the lab of Kenneth Campellone (10.1371/journal.pgen.1010045) has demonstrated that in the absence of Arp-2/3 micronuclei arise from defective chromosomal segregation. I believe the authors should distinguish whether this is the case or it is indeed the nuclear envelope rupture events that produce the micronuclei. The authors could add a sentence in the result and discussion section discussing this possibility.
    3. In figure 1C the authors claim that depletion of WASp results in reduced levels of Lamin-A/C. However, this is very hard to see any difference in the Western Blot provided.
    4. In line 141 the authors mention that loss of WASp results in Lamin-A/C wrinkles and "nuclear irregularities". What are the irregularities mentioned here? Maybe the authors should point them out in the imaging provided in Figure 1 to improve reader comprehension.
    5. In line 177-178 the authors write that there is an increase in nuclear deformation after transwell migration. However, this is not shown. I suggest the authors include a circularity index quantification in the same dataset to support that claim. In addition, the difference in the γH2AX foci in post-3µm in wt and WKO conditions appear to be almost identical. I believe it is the post migration difference in DNA damage that should be stressed here. Finally, the authors claim that the WKO macrophages are more nuclear envelope rupture-prone and this results in DNA damage accumulation as has been already described. However, the lab of Jan Lammerding has also demonstrated that nuclear deformation alone can also induce DNA damage accumulation in a cell cycle specific manner via stalled replication forks. The authors should distinguish between the two possibilities with live cell imaging by expressing a DNA damage marker (e.g. 53BP1, coupled with either NLS-GFP or cGAS to assess whether DNA damage accumulates in response to mechanical deformation or nuclear envelope rupture).
    6. In line 204 the authors show that reintroduction of WASp-GFP reduces the frequency in which "Lamin-A/C rupture" events are observed. Then they conclude WASp is important to maintain nuclear stability. While this Lamin-A/C is a good marker for nuclear blebs, I believe that more "classic" nuclear envelope rupture markers should be used to assess the potential compromised nuclear integrity (e.g. NLS-GFP or cGAS). Hence, I would suggest the authors to attempt to replicate this finding with live cell imaging of cells under strong (3µm) confinement and quantify the frequency in which they detect NLS-GFP leakage or perinuclear accumulation of cGAS. Another experiment would be to try and rescue the phenotype observed in the fabricated microchannels used in Figure 2 and observe whether the nuclear envelope rupture events are reduced.
    7. In Figure 4 the authors describe the potential role of WASp in the formation of actin patches in the vicinity of the nucleus. In both Fig4C and Fig4E, the mildly confined (6 µm) condition is missing. I would suggest the authors to quantify the distribution of the different actin structures and the colocalization of WASp/Phalloidin signal also in 6µm confinement. Additionally, in the lines 220-222 the authors claim that the formation of the perinuclear actin ring is absent in wt macrophages, however we still see the perinuclear actin ring in the images.
    8. Adding on my previous comment, if the perinuclear actin patches are not present in the absence of confinement, can the authors display the formation of these patches through live cell imaging utilizing probes such as LifeAct or the Actin Chromobody in wild type macrophages?
    9. While it is intuitive that if WASp is important for the formation of perinuclear actin patches, then this has to be through Arp-2/3 this should be tested. I would suggest the authors to either pharmacologically inhibit Arp2/3 with CK-666 or genetically manipulating the system by either overexpressing a dominant negative version of the Arp2/3 subunit or silencing an essential subunit. This would make it clear that these actin patches are indeed Arp-2/3 driven.
    10. The authors interestingly discovered the upregulated expression of certain inflammatory cytokines. While the authors nicely show that the mRNA levels increase, they do not show whether this translates also in increased protein levels. I would suggest the authors to include an experiment in which they assess the protein levels of a few of the upregulated chemo-/cytokines via immunoblotting or ELISAs. This is also the case for IL-6 mRNA levels showed in Figure 5.
    11. In FigS1 the authors display that in WKO cells have increased "cellular stiffness" when WASp is depleted from macrophages in the absence of confinement. However, since the cantilever is placed on top of the nuclear area, I am not sure whether the authors were measuring nuclear rigidity or cortical rigidity. In any case, in Fig5 they show that in the absence of confinement factors responsible for branched actin nucleation are downregulated (If I am not wrong, Arp2/3 is a major factor contributing to cortical actin nucleation - along with mDia1, although there might be a compensation mechanism), while the expression of factors related to nuclear mechanics is not altered. This is not in line with the result from the AFM experiment presented in FigS1. I am not sure what does this AFM experiment add to the message of the manuscript. The authors could possibly add a sentence to make the connection to the rest of the manuscript clearer.
    12. The authors beautifully show that WKO macrophages and AMO's show an upregulation of inflammation related genes. However, I believe it is important to assess whether the upregulation of these signature genes upon confinement can be rescued by the re-introduction of wild-type WASp (and possibly a mutant that is not able to activate Arp-2/3 - to display that is indeed the Arp-2/3 mediated nucleation of these actin patches that limit inflammation).
    13. In the discussion (line 333-335), the authors mention that the WASp (and I presume the produced actin patches) act as a barrier to prevent nuclear envelope rupture events. From the images provided in the manuscript I get the impression that both WASp and the perinuclear actin patches display a polarized nature. For example, in Figure 4C the actin patch is located at the left of the nucleus. WASp in general is more abundant at the trailing side of the cytoplasm in the migrating macrophage displayed in Fig S4B. Finally, WASp has higher abundancy in a specific perinuclear area in Figure 4E. How these seemingly polarized structures act as barriers? Do they stop nuclear blebs from protruding through lamin B1 gaps? Are nuclear envelope rupture events in migrating macrophages happening specifically in the migrating or trailing end of the nucleus?
    14. There has been a report by the lab of Andrea Ablasser (10.1126/science.aaw6421) in which loss of nucleocytoplasmic compartmentalization does not activate the expression of cytokines, as BAF can compete against cGAS for binding of leaked chromatin thus limiting inflammation. If there is not a cell-type specific (or lack of) expression of BAF on BMDMs, how do the authors explain their results in relation to the aforementioned publication?

    Significance

    In the manuscript entitled "WASp activity in macrophages prevents mechano-induced inflammation by 1 protecting the nuclear envelope" Amadio and colleagues demonstrate that WASp-mediated actin polymerization at perinuclear areas protect against nuclear envelope rupture events and subsequent upregulation of inflammation signature genes. the idea is interesting but the evidence is somewhat preliminary. see my suggestions.

  6. Version published to 10.1101/2024.12.23.630061 on bioRxiv