ATR is essential for preservation of cell mechanics and nuclear integrity during interstitial migration

ATR responds to mechanical stress at the nuclear envelope and mediates envelope-associated repair of aberrant topological DNA states. By combining microscopy, electron microscopic analysis, biophysical and in vivo models, we report that ATR-defective cells exhibit altered nuclear plasticity and YAP delocalization. When subjected to mechanical stress or undergoing interstitial migration, ATR-defective nuclei collapse accumulating nuclear envelope ruptures and perinuclear cGAS, which indicate loss of nuclear envelope integrity, and aberrant perinuclear chromatin status. ATR-defective cells also are defective in neuronal migration during development and in metastatic dissemination from circulating tumor cells. Our findings indicate that ATR ensures mechanical coupling of the cytoskeleton to the nuclear envelope and accompanying regulation of envelope-chromosome association. Thus the repertoire of ATR-regulated biological processes extends well beyond its canonical role in triggering biochemical implementation of the DNA damage response.

In general the conclusions are very fully documented with appropriate controls. The one issue that does need to be further addressed, or at least clarified, is the assumption that the nesprin FRET sensor is sensing "tension". In fact, it is measuring distance, not tension. It is an inference that distance changes reflect changes in tension. If I understand correctly, changes in distance could result from changes in the molecular lengths of linker molecules, rather than changes in mechanical forces that alter their physical lengths. The cited reference only says that there should be mechanosensing and that nesprin should be involved, but it does not say that nesprin per se is the elastic element. Is there really evidence that the FRET construct is a tension sensor? Are there controls to say this? What happens with AFM perturbation where there is a "real time" response with no chance for molecular reorganization? More experiments are not required at this point -it is only required that the actual situation be clearly explained. If it is correct that it is an "inference" that tension is being sensed, then the results are "consistent" with a reduction in tension when ATR is depleted (rather than "demonstrating" such a reduction).
Also, the authors present one experiment showing that certain defects require ATR-mediated kinase activity. It is not appropriate to include more data in the present paper, for many reasons. But the implication of the presented finding should be stated more clearly (see below) and, in future, it would certainly be interesting to have a better idea of whether all depletion phenotypes, including localization defects, are dependent on the kinase function, which will help to know the specific role(s) of ATR beyond the final downstream phosphorylation.
Suggested writing improvements. Note: the number of comments does not imply any problem with content, only presentation.
1. Title change to: "ATR is essential for preservation of cell mechanics and nuclear integrity during..... " The current "preserves" implies direct effects but what is shown is that depletion causes problems.

abstract line 23
Is there some way to make the first sentence of the abstract more faithful to what was shown in previous work? The problem is that the previous work does not show that "ATR is activated". It shows that "ATR responds to mechanical stress at the NE and mediates NE-associated repair of aberrant topological DNA states" 3. abstract line 24 starting with "ATR preserves...." This entire sentence can be omitted. Just talk about the data; generalizations are not meaningful in advance of facts. 4. abstract -latter part. I could suggest the following: "When subjected to mechanical stress or undergoing interstitial migration, ATR-defective nuclei exhibit nuclear collapse, NE ruptures, perinuclear cGAS accumulation, which indicates loss of NE integrity, and aberrant perinuclear chromatin status. ATR-defective cells also are defective in neuronal migration during... (etc). These and other findings indicate that ATR ensures mechanical coupling of the cytoskeleton to the NE and accompanying regulation of NE-chromosome association. Thus: the repertoire of ATR-regulated biological processes extends well beyond its canonical role in triggering biochemical implementation of the DNA damage response." not perfect, but perhaps the general goal is clear 5. Line 38. What is "nuclear plasticity"? It is important to distinguish between between "plasticity" in the generic sense and the mechanical sense. Some more specific word/wording would be better than this jargon. It may well be understood in the field but is nonetheless ambiguous and there should be a better term. 6. Line 40. Do you mean "nucleus" or "nuclear envelope". Ambiguous in first two sentences. Thus perhaps: "the physical properties of the nucleus are modulated in response to inputs from..." would be better. 7. Line 41. Turn sentence around. "The nuclear envelope (NE) plays a critical role in this process by connecting the cytoskeleton and the chromosomes" 8. Line 50. what do you mean by "nuclear mechanics" More jargon. Do you mean responses of the nucleus to mechanical inputs? Sensing and transducing of mechanical forces by the nucleus? be more clear. The term nuclear mechanics should be eliminated everywhere, in my view. 9. Line 50. "We hypothesize" should be omitted. The reader does not care about hypotheses, only about facts. Plus the preceding list of "considerations" are strung together as a list without clear elucidation of the logic connecting them. Example fix: Previous observations suggest that ATR at the NE directly senses mechanical stress at the NE/chromatin interface to effect release of chromatin from the NE. This possibility is supported by the fact that ATR comprises heat repeats, which are elastic connectors. Here we explore the possibility that ATR-mediated mechanical communication are also important for the state of the NE itself and, having obtained evidence to this effect, explore its functional implications".
10. The paper would benefit from more frequent subheadings that state observations, not conclusions. for example: -line 52 ATR in the cytoplasm localizes preferentially to membranes and actin fillaments and in proximity to the nucleus/NE. -line 64 Depletion of ATR results in multiple nuclear membrane defects.
-line 83 Depletion of ATR alters the response of the nucleus to external force -line 92 Depletion of ATR alters diverse molecular features related to nuclear envelope status subheadings: lipid composition; heterochromatin formation; inner/outer membrane distancenesprin length; YAP accumulation 11. line 83: suggestion: Abnormalities in the NE and/or chromatin/chromosomes can affect the mechanical properties of the nucleus 12. line 87: (Fig 2b), consistent with the fact that the nucleus is the stiffest component of the cell (refs). move this statement to this position in the text from farther down.
13. line 87 ff. " The irregular spikes in the force plots (Fig. 2c) in shATR cells suggests that nuclei collapse due to their inability to sustain the mechanical pressure. The abnormal force plots increased to nearly 50% in shATR cells (Fig. 2c), well correlating with the percentage of deformed nuclei (Fig.  1c)." The experiment needs to be described more clearly so one doesn't have to look at the figure. Is there a plot of force vs time? over what times? Spikes of what? And separate "in principle" what spikes mean from difference in normal and shATR cells. That is: spikes mean nuclear collapse. See X spikes per time (or probability per cell?) in normal cellsand Y% in shATR. And what is the % of nuclear collapse in Figure 1? don't make the reader go look.
14. line 92. New subheading (see above) 15. line 107: ATR does not "cause heterochromatinization"...it "promotes increased heterochromatinization". 21. line 142. You need to say: ATR is a molecule whose presence is required for normal response of the nucleus to mechanical forces. Then, since responses of the cell are dominated by effects on the nucleus, you can generalize to "the cell". Need to make cell/nucleus distinction clearer.
22. line 157. Make this a new section specifically discussing neurogenesis and metastasis. 23. line 159. As above, noone cares about your "reasoning". Just say that many cellular functions involving moving through tight spaces and the observations in the previous section suggest that depletion of ATR should severely compromise the ability of cells to do such moving. To explore this possibility you examined to well-known cases: neurogenesis and metastasis. In the case of neurogenesis, nuclear stresses and accompanying chromatin compaction are prominent features. (There is also information from Discher re metastasis; could be mentioned). Then go on with description of experiments.
24. line 175. tension issue again 25. Line 178 "likely due to their intrinsic defects in nuclear mechanics".. the recurring issue is "what is nuclear mechanics". Be more specific. 26. Line 183: "By analysing the cGAS-GFP foci distribution, we found that in ATR184 defective cells cGAS-GFP foci appeared much earlier than controls at the leading tip of nucleus engaged in the constriction (Fig. 3f)." This finding should be presented below in the context of other data regarding leading/lagging ends of the nucleus.
27. Line 193 Both control and ATR-depleted cells exhibited sporadic NE ruptures (Fig. 3h), in accordance with previous reports 21,22. This is ambiguous. Previous papers do NOT talk about ATR, only about control cells. ATR role has not been previously analyzed, yes? Need to make this clear so current data are appreciated.
28. Line 206 While the intrinsic defects in nuclear mechanics in ATR-defective cells may not affect cell viability under normal conditions, the consequences of nuclear collapse following mechanical stress certainly contribute to cell lethality. "nuclear mechanics"....fix Meaning unclear......cell lethality when cells are forced through narrow passages, not "cell lethality in general" ????
29. line 211. Omit first sentence. Just say: Cell migration for neurogenesis and for metastastasis involve going through narrow places. Above in vitro suggests ATR important. Consistent with this, neurogenesis involves mechanical stress and previous reports say that state of nuclear membrane is important 30. 225/226. "we predicted" As above.......Rephrase. 31. Line 235 Title: known to influence (better wording) mechanical responses of the nucleus (not the cell) 32. line 246 We identified several ATR interactors able to influence nuclear mechanics and contribute, at least in part, to some of the phenotypes observed in ATR-depleted cells fix to: We identified several ATR interactors for which previous studies have identified roles in the mechanical properties of the nucleus and whose depletion mimics, at least in part, some of the phenotypes observevd in.....

line 269-270
It should be clarified that there are various reasons for interaction with ATR: does stress activate ATR phosphorylation which acts on these targets? And/or do these targets recruit ATR? And/or do these targets participate in ATR activation?
Discussion -in the view of this reviewer, the authors are free to say whatever they wish in the Discussion, but it could be possible to further separate the observations from the conclusions from the inferences while still emphasizing the intrinsic interest of the work. Moreover, the two most important points (see comments at beginning of the review) are somewhat drowned out in the plethora of other information.
Reviewer #2 (Remarks to the Author): The study by Kidiyoor et al. uses microscopy, biophysical and in-vivo approaches to characterize mechanical stress-associated phenotypes that are observed in ATR deficient cells. The authors show that loss of ATR results in nuclear envelop defects that lead to nuclear collapse under mechanical stress. Such mechanical stress occurs during interstitial cell migration, which the authors show is reduced upon loss of ATR function. The study also shows that loss of ATR impacts a variety of mechanical responses, which includes reducing the nuclear localization of the Hippo pathway component YAP. Further potential insight into ATR function is shown from a mass spectrometrybased screening effort to identify ATR-associated proteins.
The manuscript is well written and clear, and provides interesting details into the functions of ATR. The concept of the study has a somewhat moderate impact given the group's prior study, but the manuscript does offer important information that is interesting to the broader research community.
There are a few minor comments that I have, which are listed below: -The conclusion that H3K9me3 is increased in ATR-depleted cells is weak based on the data shown. The western blot images do not show an obvious change and no statistics are provided for the quantitation. Given how central this conclusion is throughout the manuscript, more solid evidence that chromatin dynamics are changing in ATR deficient cells should be provided. -The images in Figure 5b (referenced as 4b in the text) do not appear to show any obvious differences in cell number. Is this the case? One would expect a reduced cell number give that loss of ATR leads to nuclear collapse and cell death during migration.
-No controls are shown for the PLA experiment in Figure 4C (referenced as 5C in the text). It is unclear whether the signals that are shown are above background.
-While it is nice to see a confirmation of the mass spec with the ATR and Nesprin-2 experiments, it is unclear how this interaction (or others) fit into the mechanism by which ATR regulates nuclear mechanics. The impact of the study would be greatly improved if the authors can connect one of the potential ATR binding partners with the altered mechanical properties observed in ATR-depleted cells, such as the changes in nuclear phospholipid composition shown in Figure 2.
Reviewer #3 (Remarks to the Author): In this paper, Kidiyoor and colleagues address an interesting question and investigate the consequences of ATR depletion on nuclear morphology and mechanics. There is a growing body of literature indicating that changes in nuclear mechanical properties can impact cell differentiation and migration. Interestingly, the authors observed that ATR may regulate some of these aspects. The approach is straightforward and the paper contains a large amount of work. The work adds to the body of knowledge about the consequence of ATR deficiency, but in the end, one doesn't have a clearer picture of why ATR affects nuclear mechanics and morphology. Much of the current study is essentially descriptive, and constitutes a succession of observations and does not reveal a molecular mechanism to account for the different phenotypes, which is extremely frustrating. Even though some observations are intriguing, the study as it is presented here seems too preliminary to be published in Nature Communications.
-It is unclear whether the distinct observations made in figure 1 and 2 are connected. The authors observed nuclear envelope defects and decreased nuclear stiffness, but is-it a consequence (or a cause?) of the other observations? Such as: the altered nuclear lipid composition (figure 2d)? Or increased chromatin condensation (figure 2e)? Or LINC mediated tension? (Figure 2f which is unlikely, since the AFM experiments have been done using isolated nuclei) -One of the main question remaining is how does ATR regulate nuclear morphology/structure? They observed that ATR kinase activity was critical in producing these effects. The authors identified many ATR interactors using mass spec (figure 4), however they performed no control experiment to validate these interactions (besides nesprin 2) and they did not test if these proteins were actual ATR substrates (Is nesprin2 phosophorylated by ATR? And if so, what is the consequence of this phosphorylation and can it explain the phenotype?). The authors also show that ATR localize at the membranes (figure 1), suggesting that this could participate to the observed phenotype, but they did not test it (potentially using mutant forms of ATR with or without the putative membrane binding regions).
-AFM data obtained from ATR depleted cells show difference between retraction and approach curve which could be hysteresis, and may be due to viscous components (and not necessarily "nuclei collapse"). The nuclei were isolated using detergent, which alters nuclear membranes and can potentially affect nuclear mechanical properties. -figure 4b. Nesprin 2 is a very large protein which is hard to observe by western blot, what is the molecular weight of the signal showed in this figure? -Surprisingly, the authors observed no difference in the number of 53BP1 foci between control cells and cells depleted for ATR (line 167) and conclude that "cell death of ATR defective cells does not correlate […] with increased DNA damage". This is discordant with previous work from Lammerding's group and Piel's laboratory showing that nuclear envelope rupture leads to DNA damage, do the authors have a potential explanation for this apparent discrepancy? -The authors observed that YAP activity and phosphorylation is affected by ATR depletion. YAP can be activated in response to NE deformation (Elosegui-Artola et al and Aureille et al.), but can the authors explain how a softer nucleus could be responsible for YAP inhibition (and YAP phosphorylation)? -It is unclear what the authors mean by "preserve cell mechanics" in the title (and elsewhere in the manuscript)?
Answers to the reviewers' comments. We thank the reviewers for their suggestions and criticisms. Our answers are in red.

Reviewers' comments:
Reviewer #1 (Remarks to the Author): This paper presents important and interesting new findings relating to the interplay between ATR and communication between the cytoplasm and the nuclear interior (chromatin/chromosomes) as mediated by the nuclear envelope. There are many detailed findings presented that are of interest. Two general points are of primary significance. First, ATR is traditionally known as a mediator of responses to problems in DNA/chromosomes. However, in seminal previous work, this group has shown that one such response involves topological/mechanical stress within the DNA and occurs via ATR at the nuclear envelope (NE); they also showed that ATR localizes to the NE specifically during S-phase when this response will be required; and that ATR hyperlocalizes to the NE in response to externally applied mechanical force or osmostic pressure. These observations, plus the fact that ATR is a giant HEAT repeat protein, raised the possibility that ATR might directly sense and transduce mechanical stress in general and at the NE in particular. The current paper now shows that ATR is actually required for the NE itself to have normal properties, thus revealing an entirely new aspect of ATR functionality. Depletion experiments show that ATR has roles for NE morphology in unchallenged cells; but diverse presented studies show that depletion of ATR has especially profound effects in situations where cells (and thus their nuclei, which are their primary load-bearing element) are under mechanical stress by deformations of various types. This is studied in a microfluidic system where cells are forced through narrow channels. Thus, broadly speaking, ATR is a key component of the system by which nuclei (and thus cells) sense and transduce mechanical forces. This is an entirely new discovery and opens the way to many future studies. This work also identifies several suspects for molecules that might be recruiters, stress-transducers and/or targets of ATR. Second, this paper shows that ATR is essential for two processes in which cells must squeeze through tight spaces (neurogenesis and metastasis), in accord with the above microfluidic studies, and with additional interesting observations regarding the specific problems encountered.
Appreciation of this paper would be significantly improved if it were better written. Multiple specific suggestions are provided below. But the overall issues are (a) to cleanly separate results and interpretation; (b) to reduce overstatements/hype; (c) to better organize both the data and the conclusions. These are challenging tasks because of the large quantity of information and because of the fact that many effects are interrelated. Nonetheless, additional effort in this direction is warranted. We thank the referee for the comments and suggestions. The paper has been extensively edited.
Of special importance is to clarify that (i) what the data show is that ATR is "involved in/required for" many interesting effects but that (ii) the specific role(s) for ATR are unclear. Further (iii) it is attractive to assume that the defects observed imply direct roles of ATR in sensing and transducing mechanical information, for many reasons including economy, HEAT repeats and previous observations in which ATR responds to topological (and thus mechanical) effects of NEassociated replication/transcription collisions. But this is not directly documented. And finally (iv) identified interactions with various molecules could reflect recruitment, mechanosensing and/or transduction of information via ATR-mediated phosphorylation of various targets. The comments are well taken and we therefore described and discussed the results accordingly.
In general the conclusions are very fully documented with appropriate controls. The one issue that does need to be further addressed, or at least clarified, is the assumption that the nesprin FRET sensor is sensing "tension". In fact, it is measuring distance, not tension. It is an inference that distance changes reflect changes in tension. If I understand correctly, changes in distance could result from changes in the molecular lengths of linker molecules, rather than changes in mechanical forces that alter their physical lengths. The cited reference only says that there should be mechanosensing and that nesprin should be involved, but it does not say that nesprin per se is the elastic element. Is there really evidence that the FRET construct is a tension sensor? Are there controls to say this? What happens with AFM perturbation where there is a "real time" response with no chance for molecular reorganization? More experiments are not required at this point -it is only required that the actual situation be clearly explained. If it is correct that it is an "inference" that tension is being sensed, then the results are "consistent" with a reduction in tension when ATR is depleted (rather than "demonstrating" such a reduction). Agree. We modified the text to tune down the "Tension" issue and to clarify the conclusions.
Also, the authors present one experiment showing that certain defects require ATR-mediated kinase activity. It is not appropriate to include more data in the present paper, for many reasons. But the implication of the presented finding should be stated more clearly (see below) and, in future, it would certainly be interesting to have a better idea of whether all depletion phenotypes, including localization defects, are dependent on the kinase function, which will help to know the specific role(s) of ATR beyond the final downstream phosphorylation. We added new experiments and changed the text extensively. We can now classify the phenotypes of ATR-defective cells as direct consequences of ATR catalytic inhibition or in long term effects due to chronic ATR depletion.
Suggested writing improvements. Note: the number of comments does not imply any problem with content, only presentation.
1. Title change to: "ATR is essential for preservation of cell mechanics and nuclear integrity during..... " The current "preserves" implies direct effects but what is shown is that depletion causes problems. Done

abstract line 23
Is there some way to make the first sentence of the abstract more faithful to what was shown in previous work? The problem is that the previous work does not show that "ATR is activated". It shows that "ATR responds to mechanical stress at the NE and mediates NE-associated repair of aberrant topological DNA states" Done 3. abstract line 24 starting with "ATR preserves...." This entire sentence can be omitted. Just talk about the data; generalizations are not meaningful in advance of facts. Done 4. abstract -latter part. I could suggest the following: "When subjected to mechanical stress or undergoing interstitial migration, ATR-defective nuclei exhibit nuclear collapse, NE ruptures, perinuclear cGAS accumulation, which indicates loss of NE integrity, and aberrant perinuclear chromatin status. ATR-defective cells also are defective in neuronal migration during...(etc). These and other findings indicate that ATR ensures mechanical coupling of the cytoskeleton to the NE and accompanying regulation of NE-chromosome association. Thus: the repertoire of ATR-regulated biological processes extends well beyond its canonical role in triggering biochemical implementation of the DNA damage response." not perfect, but perhaps the general goal is clear Done 5. Line 38. What is "nuclear plasticity"? It is important to distinguish between between "plasticity" in the generic sense and the mechanical sense. Some more specific word/wording would be better than this jargon. It may well be understood in the field but is nonetheless ambiguous and there should be a better term. Changed 6. Line 40. Do you mean "nucleus" or "nuclear envelope". Ambiguous in first two sentences. Thus perhaps: "the physical properties of the nucleus are modulated in response to inputs from..." would be better. The sentence has been reorganized as suggested. 9. Line 50. "We hypothesize" should be omitted. The reader does not care about hypotheses, only about facts. Plus the preceding list of "considerations" are strung together as a list without clear elucidation of the logic connecting them. Example fix: Previous observations suggest that ATR at the NE directly senses mechanical stress at the NE/chromatin interface to effect release of chromatin from the NE. This possibility is supported by the fact that ATR comprises heat repeats, which are elastic connectors. Here we explore the possibility that ATR-mediated mechanical communication are also important for the state of the NE itself and, having obtained evidence to this effect, explore its functional implications". Done (much better now!) 10. The paper would benefit from more frequent subheadings that state observations, not conclusions. for example: -line 52 ATR in the cytoplasm localizes preferentially to membranes and actin fillaments and in proximity to the nucleus/NE. -line 64 Depletion of ATR results in multiple nuclear membrane defects.
-line 83 Depletion of ATR alters the response of the nucleus to external force -line 92 Depletion of ATR alters diverse molecular features related to nuclear envelope status subheadings: lipid composition; heterochromatin formation; inner/outer membrane distance-nesprin length; YAP accumulation Done 11. line 83: suggestion: Abnormalities in the NE and/or chromatin/chromosomes can affect the mechanical properties of the nucleus Done 12. line 87: (Fig 2b), consistent with the fact that the nucleus is the stiffest component of the cell (refs). move this statement to this position in the text from farther down. Done 13. line 87 ff. " The irregular spikes in the force plots (Fig. 2c) in shATR cells suggests that nuclei collapse due to their inability to sustain the mechanical pressure. The abnormal force plots increased to nearly 50% in shATR cells (Fig. 2c), well correlating with the percentage of deformed nuclei (Fig. 1c)." The experiment needs to be described more clearly so one doesn't have to look at the figure. Is there a plot of force vs time? over what times? Spikes of what? And separate "in principle" what spikes mean from difference in normal and shATR cells. That is: spikes mean nuclear collapse. See X spikes per time (or probability per cell?) in normal cellsand Y% in shATR. And what is the % of nuclear collapse in Figure  1? don't make the reader go look. We removed the curves from figure and modified  21. line 142. You need to say: ATR is a molecule whose presence is required for normal response of the nucleus to mechanical forces. Then, since responses of the cell are dominated by effects on the nucleus, you can generalize to "the cell". Need to make cell/nucleus distinction clearer. Done 22. line 157. Make this a new section specifically discussing neurogenesis and metastasis. Done 23. line 159. As above, no one cares about your "reasoning". Just say that many cellular functions involving moving through tight spaces and the observations in the previous section suggest that depletion of ATR should severely compromise the ability of cells to do such moving. To explore this possibility you examined to well-known cases: neurogenesis and metastasis. In the case of neurogenesis, nuclear stresses and accompanying chromatin compaction are prominent features. (There is also information from Discher re metastasis; could be mentioned). Then go on with description of experiments. Done 24. line 175. tension issue again Done 25. Line 178 "likely due to their intrinsic defects in nuclear mechanics".. the recurring issue is "what is nuclear mechanics". Be more specific. The sentence has been changed 26. Line 183: "By analysing the cGAS-GFP foci distribution, we found that in ATR184 defective cells cGAS-GFP foci appeared much earlier than controls at the leading tip of nucleus engaged in the constriction (Fig. 3f)." This finding should be presented below in the context of other data regarding leading/lagging ends of the nucleus. Done 27. Line 193 Both control and ATR-depleted cells exhibited sporadic NE ruptures (Fig. 3h), in accordance with previous reports 21,22. This is ambiguous. Previous papers do NOT talk about ATR, only about control cells. ATR role has not been previously analyzed, yes? Need to make this clear so current data are appreciated. Done

Line 206
While the intrinsic defects in nuclear mechanics in ATR-defective cells may not affect cell viability under normal conditions, the consequences of nuclear collapse following mechanical stress certainly contribute to cell lethality. "nuclear mechanics"....fix Meaning unclear......cell lethality when cells are forced through narrow passages, not "cell lethality in general" ???? Fixed 29. line 211. Omit first sentence. Just say: Cell migration for neurogenesis and for metastastasis involve going through narrow places. Above in vitro suggests ATR important. Consistent with this, neurogenesis involves mechanical stress and previous reports say that state of nuclear membrane is important Done, see also the previous paragraph.
30. 225/226. "we predicted" As above.......Rephrase. Done 31. Line 235 Title: known to influence (better wording) mechanical responses of the nucleus (not the cell) Done 32. line 246 We identified several ATR interactors able to influence nuclear mechanics and contribute, at least in part, to some of the phenotypes observed in ATR-depleted cells fix to: We identified several ATR interactors for which previous studies have identified roles in the mechanical properties of the nucleus and whose depletion mimics, at least in part, some of the phenotypes observevd in..... Done 33. line 269-270 It should be clarified that there are various reasons for interaction with ATR: does stress activate ATR phosphorylation which acts on these targets? And/or do these targets recruit ATR? And/or do these targets participate in ATR activation? Done Discussion -in the view of this reviewer, the authors are free to say whatever they wish in the Discussion, but it could be possible to further separate the observations from the conclusions from the inferences while still emphasizing the intrinsic interest of the work. Moreover, the two most important points (see comments at beginning of the review) are somewhat drowned out in the plethora of other information. Agree. The discussion has been extensively modified.

Reviewer #2 (Remarks to the Author):
The study by Kidiyoor et al. uses microscopy, biophysical and in-vivo approaches to characterize mechanical stressassociated phenotypes that are observed in ATR deficient cells. The authors show that loss of ATR results in nuclear envelop defects that lead to nuclear collapse under mechanical stress. Such mechanical stress occurs during interstitial cell migration, which the authors show is reduced upon loss of ATR function. The study also shows that loss of ATR impacts a variety of mechanical responses, which includes reducing the nuclear localization of the Hippo pathway component YAP. Further potential insight into ATR function is shown from a mass spectrometry-based screening effort to identify ATR-associated proteins.
The manuscript is well written and clear, and provides interesting details into the functions of ATR. The concept of the study has a somewhat moderate impact given the group's prior study, but the manuscript does offer important information that is interesting to the broader research community. There are a few minor comments that I have, which are listed below: We thank the referee for the comments and suggestions. The paper has been extensively edited.
-The conclusion that H3K9me3 is increased in ATR-depleted cells is weak based on the data shown. The western blot images do not show an obvious change and no statistics are provided for the quantitation. Given how central this conclusion is throughout the manuscript, more solid evidence that chromatin dynamics are changing in ATR deficient cells should be provided.
We have now included the statistics (see Supplementary Figure 2d). Moreover, we have performed new sets of experiments. The DNAse sensitivity assay (Supplementary Figure 2c)

is routinely used to assess the chromatin state and is based on the rationale that DNAse preferentially cleaves euchromatin. We show that ATR defective cells at early time points are more resistant to DNAse treatment than control cells. We conclude that ATR defective cells exhibit a lower euchromatin/heterochromatin ratio compared to control. Moreover, we compared the accumulation of H3K9trimethylation and chromatin compaction by FLIM-FRET following acute treatments with ATR inhibitors or after chronic ATR depletion; we found that both phenotypes represent long term responses following ATR depletion. Notably, a recent paper (Nava et al. 2020) showed that H3K9-3Me heterochromatic levels rearrange in response to mechanical stress at the NE and when nuclei recover from the stress. Our new results suggest that the accumulation of H3K9-trimethylation may reflect the inability of ATR depleted cells to recover from nuclear stress, rather than representing a direct consequence of ATR inactivation. Most likely, the aberrant nuclear morphology of ATR defective nuclei, which is also a late effect, does not allow the cells to re-form a correct perinuclear NE-heterochromatin organization.
In these months we also tried to map the heterochromatic regions at the genomic levels using antibodies against H3K9-trimethylation, but we experienced some technical difficulties that could not solved also due to the covid crisis that imposed a lock down in the institute. Anyhow, we hope that the new results and the text editing addresses the reviewer's comments.
-No statistics are provided for Figure 3F. do not appear to show any obvious differences in cell number. Is this the case? One would expect a reduced cell number give that loss of ATR leads to nuclear collapse and cell death during migration. We note that the cells may not die since they exit cell cycle and stop migrating at E18.5 (4 days after electroporation). In fact, it is difficult to make any conclusion on cell death from the number of quantified cells since some sections have more cells whereas others have less. As you can see from the images, shATR-6 treated sections have less cells compared to shLuc or shATR-4.
-No controls are shown for the PLA experiment in Figure 4C (referenced as 5C in the text).

Corrected. Now the controls are shown in Supplementary Figure-4h.
It is unclear whether the signals that are shown are above background. In supplementary Figure-4h we show that non-specific background signals appear more spread in the cytoplasm, while the specific foci localize in the proximity of the NE.
-While it is nice to see a confirmation of the mass spec with the ATR and Nesprin-2 experiments, it is unclear how this interaction (or others) fit into the mechanism by which ATR regulates nuclear mechanics. The impact of the study would be greatly improved if the authors can connect one of the potential ATR binding partners with the altered mechanical properties observed in ATR-depleted cells, such as the changes in nuclear phospholipid composition shown in Figure 2.
The new sets of results strongly suggest that following an acute treatment with ATR inhibitors cells experience a mechanical uncoupling of cytoskeleton and NE and accumulate YAP in the cytoplasm. The other phenotypes become obvious much later, when cells experience chronic ATR depletion. In this scenario, the Nesprin-2 results become particularly relevant as it is directly implicated in the coupling between NE and cytoskeleton. However, we wish to point out that the ATR-mediated control of the mechanical properties of the nucleus is not a linear pathway (see below). The proteome data strongly suggest that the phenotypes of ATR defective cells are likely caused by the deregulation of several processes/pathways (see also the scheme presented in the model figure) and some of the factors identified are well known ATR targets. Moreover, we wish to stress that the ATR interactors identified in this study might be involved in the cellular response to mechanical stress by mediating ATR recruitment at the NE, transducing the mechanical stress signals or in triggering short and long term response pathways. Clearly, the characterization of each interaction at a biochemical and genetic level goes beyond the scope of the current work. Our published and unpublished results are in line with our conclusions: 1) Yeast Mlp1 nucleoporin is regulated by ATR/Mec1 and plays a key role in controlling mRNA export and influences chromatin-NE dynamics (Bermejo et al. Cell, 2011); We found that TPR, the human homologue of Mlp1, also in mammals is phosphorylated by ATR and controls mRNA export through the NE; However, TPR contributes only in part to the NE abnormalities observed in ATR defective cells (Kosar and Foiani submitted). For instance, the association of NE with nucleoli does not depend on TPR. 2) We know that Chk1, the primary kinase downstream of ATR, responds to nuclear stress (Kumar et al.2014) and prevents the accumulation of Nuclear Envelope invaginations and YAP de-localization (like ATR); however, Chk1 does not influence the detachment of chromatin and nucleoli from the NE (see Figure 1 for reviewers only). Moreover, Chk1 itself interacts with several proteins which play a role at the NE (Blasius et.al 2011) (see Table 1 for reviewers only), such Lamin A/C, TPR, NUP153 and several importin subunits. Hence, CHK1 defective cells exhibit only part of the nuclear abnormalities of ATR defective cells, implying that the NE abnormalities of ATR defective cells results from the inability of ATR to detach chromatin from the envelope and from the inability of Chk1 to prevent NE invaginations. Regarding the changes in phospholipid composition of NE membranes. We performed additional experiments to address whether alterations in membrane phospholipid composition could phenocopy some of the nuclear abnormalities of ATRdefective cells. As shown in Figure 2 for reviewers only, we treated cells with Bromoenol Lactone (BEL), that inhibits the calcium-independent Phospholipase A2 (iPLA2), leading to the accumulation of high levels of polyunsaturated PC species (Zhang et al., 2007); We found that BEL treatment reduced the nuclear stiffness of control nuclei to the levels of shATR nuclei, whilst it had a modest effect on shATR nuclei (Figure 2 for reviewers only). However, we feel uncomfortable to include these results as, at the moment, it is not clear whether the changes in phospholipid compositions reflect an ATR-mediated response to mechanical stress or an adaptative metabolic rewiring due to extensive NE remodeling. Interestingly, mechanical stress has been shown to induce metabolic rewiring (Park et al 2020). Moreover, we did not find any ATR interactor involved in phospholipid metabolism.
Reviewer #3 (Remarks to the Author): In this paper, Kidiyoor and colleagues address an interesting question and investigate the consequences of ATR depletion on nuclear morphology and mechanics. There is a growing body of literature indicating that changes in nuclear mechanical properties can impact cell differentiation and migration. Interestingly, the authors observed that ATR may regulate some of these aspects. The approach is straightforward and the paper contains a large amount of work. The work adds to the body of knowledge about the consequence of ATR deficiency, but in the end, one doesn't have a clearer picture of why ATR affects nuclear mechanics and morphology. Much of the current study is essentially descriptive, and constitutes a succession of observations and does not reveal a molecular mechanism to account for the different phenotypes, which is extremely frustrating. Even though some observations are intriguing, the study as it is presented here seems too preliminary to be published in Nature Communications. Perhaps in the original version of the paper we did not stress enough the relevance of our findings. To our view, the following findings are of pivotal relevance: 1) ATR influence the mechanical coupling between NE and the cytoskeleton (see also below and the new observations), and 2) the pathological consequences of ATR inactivation/depletion which impact on interstitial migration, neurogenesis and even metastasis. ATR is known as a key DNA damage response regulator that senses ssDNA to protect genome integrity. Our observations clearly extend ATR functions well beyond its role in sensing single stranded DNA. In particular, while our findings can explain the interstitial migration defects of ATR-depleted cells and the progeroid features and the stem cell exhaustion of Seckel patients and ATR conditional k.o.mice (see our scheme/model in the paper), the canonical function of ATR cannot justify these abnormalities. In fact, ATR defects are reminiscent of those features caused by laminopathies and our recent unpublished observations (Giovannetti, Kidyoor and Foiani, manuscript in preparation submission) showing that Seckel patients develop cardiomyopathies due to NE anomalies further reinforce our conclusions. We hope that the referee will find the new version of the paper suitable for Nat. Communications.
-It is unclear whether the distinct observations made in figure 1 and 2 are connected. The authors observed nuclear envelope defects and decreased nuclear stiffness, but is-it a consequence (or a cause?) of the other observations? Such as: the altered nuclear lipid composition (figure 2d)? Or increased chromatin condensation (figure 2e)? Or LINC mediated tension? (Figure 2f which is unlikely, since the AFM experiments have been done using isolated nuclei) The paper has been extensively edited also to clarify this specific point. In general, the phenotypes observed in ATR depleted/inhibited cells can be classified in short term effects (3)(4)

hours after ATR inhibition) and long term effects (2-7 days following ATR depletion). While some of these phenotypes may represent direct consequences of ATR inactivation others may result from long term adaptative responses. This is now extensively discussed in the new version of the paper. Below we provide a table showing which phenotypes can be observed 3-4 hours after ATR inhibition (short term) or 3-7 days following ATR depletion (long term) and discuss the observed phenotypes one by one.
Lipid changes: Altogether our observations suggest that the lipid changes at the level of nuclear membranes represent a long term effect, likely due to the extensive nuclear envelope remodeling as also shown in Figure 5h. This is also consistent with the finding that ATR is essential for NE reformation/repair during mitosis (see also Kumar et al. 2014) (Matsuoka et al.,2007), further re-inforce the conclusion that the loose NE-cytoskeleton connections is a direct consequence of ATR inhibition. (Supplementary fig 2a). These observations suggest that the changes in stiffness likely result from other defects, such as altered lipid composition, aberrant nuclear morphology and chromatin state. Accordingly, we now show that the BEL inhibitor altering the phospholipid metabolism, causes a decrease in stiffness, phenocopying ATR defects (Figure 2 for reviewers only).

Phenotype
Long term Short term