ATR is essential for preservation of cell mechanics and nuclear integrity during interstitial migration AbstractATR 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. IntroductionMechanical properties of the nucleus and nuclear mechanosensing affects genome integrity, nuclear architecture, gene expression, cell migration, and differentiation1,2. The physical properties of the nucleus are conveniently modulated following the inputs from the cell microenvironment or from chromatin dynamics. The nuclear envelope (NE) plays a critical role in this process by connecting the cytoskeleton and the chromatin1,2.Ataxia Telangiectasia and Rad3-related protein (ATR)聽regulates the DNA damage response (DDR)3 and protects genome integrity by regulating multiple pathways4. ATR mutations cause the Seckel syndrome, an autosomal recessive disorder characterized by growth retardation, dwarfism, and microcephaly with mental retardation5. We previously reported that ATR directly senses mechanical stress at the NE/chromatin interface and facilitates release of chromatin from the NE6. This possibility is supported by the fact that ATR comprises HEAT (huntingtin, elongation factor 3, A subunit of protein phosphatase 2A, and TOR1) repeats, which are elastic connectors, ideal to sense mechanical stimuli7,8.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.ResultsATR is enriched at membranes and actin filaments around the nucleusWe visualized ATR distribution in exponentially growing HeLa cells by electron microscope (EM) and found ATR in the nucleus, cytosol, and other organelles, including聽endoplasmic reticulum (ER), Golgi, and mitochondria (Supplementary Fig.聽1a). ATR (18.8%) was bound to actin filaments, particularly in the proximity of the NE and more than 20% was bound to cellular membranes (Fig.聽1a). Membrane fractionation analysis confirmed that 17% of ATR co-fractionated with membranes, also when nucleic acids were degraded by Benzonase treatment (Supplementary Fig.聽1b). We used TopBp1, a chromatin-bound protein, tubulin, a cytoplasmic protein, and Nup133, a NE protein, as controls in our fractionation experiments (Supplementary Fig.聽1b). The Kyte and Doolittle9, and the SOSUI and WoLF PSORT analyses10, which recognize hydrophobic and membrane-associated domains, respectively, identified seven putative membrane binding and hydrophobic regions in ATR (Supplementary Fig.聽1c).Fig. 1: ATR interacts with membranes and preserves nuclear morphology.a Images from routine 60鈥塶m EM section with enrichment of nano-gold-labeled ATR in the perinuclear region. Colored arrowheads indicate actin-associated (black), chromatin-associated (red), and membrane-associated ATR (yellow), respectively (scale bar 500鈥壩穖 for the right and 200鈥壩穖 for the left panel). b Defective nuclear morphology of ATR-depleted HeLa cells visualized with immunofluorescence of Lamin B1 (NE) and DAPI (DNA) (scale bar鈥?鈥?0 渭m). c Quantification of nuclear deformations by manual sorting based on their degree of deformation: Normal, mildly deformed and severely deformed (n鈥?鈥?15, 525, 229 cells for shCtrl, shATR1, and ShATR2, respectively). Quantifications of d micronuclei (n鈥?鈥?15, 525, and 229 cells for shCtrl, shATR1, and ShATR2, respectively) and e nuclear circularity index (n鈥?鈥?88, 253, and 215 cells for shCtrl, shATR1, and ShATR2, respectively; N鈥?鈥? independent experiments). f EM images of the nucleus from control and ATR-depleted HeLa cells. Arrowheads indicate invaginations with nucleoli attachment (scale bar 5鈥壩糾). Bar graphs presented as mean鈥壜扁€塖EM and box plot whiskers and outliers plotted with Tukey鈥檚 method. p-values calculated using two-way ANOVA test for c or for d. e Ordinary one-way ANOVA test with Dunnett鈥檚 multiple comparisons test (****P鈥?lt;鈥?.0001, ***P鈥?lt;鈥?.001, **P鈥?lt;鈥?.01, *P鈥?lt;鈥?.05; NS, not significant). Source Data file contains source data and all additional details of statistical analysis.Full size imageATR depletion results in multiple nuclear membrane defectsShort hairpin RNA (shRNA)-mediated ATR depletion in HeLa cells caused 80% reduction of ATR (Supplementary Fig.聽1d) and no obvious cell cycle anomalies (Supplementary Fig.聽1e). Immunofluorescence (IF) analysis showed that shATR cells have compromised nuclear morphology, characterized by altered nuclear circularity index, invaginations, and micronuclei (Fig.聽1b鈥揺). Similar defects were observed in ATR-depleted U2OS cells (Supplementary Fig.聽1f), human ATR Seckel fibroblasts, and non-cycling primary neurons isolated from humanized Seckel mice11 (Supplementary Fig.聽1g, h). ATRflox/鈭?/i> HCT116 cells, which have reduced ATR levels12 (Supplementary Fig.聽1i), also displayed compromised nuclear morphology (Supplementary Fig.聽1j, k). We transfected ATRflox/鈭?/i> cells with wild-type green fluorescent protein (GFP)聽tagged聽ATR (GFP-ATR)聽or with a kinase inactive version of GFP-ATR. Although wild-type GFP-ATR rescued the nuclear defects, the mutant form did not (Supplementary Fig.聽1l, m). We then performed EM analysis of shATR nuclei (Fig.聽1f, Supplementary Fig.聽1n鈥搒, and Supplementary Video.聽1). ATR-depleted cells exhibited NE invaginations of type II (outer and inner membranes invaginations) and type I (inner membranes invaginations)13, associated with condensed chromatin and/or nucleoli (Fig.聽1f and Supplementary Fig.聽1n鈥搑). NE invaginations also associated with nucleoli forming nucleolar canals that represent intermediates in rRNA export through the NE14,15 (Fig.聽1f and Supplementary Fig.聽1r). We also found, within the nucleus, inner membrane invaginations/fragments attached to chromatin and micronuclei (Supplementary Fig.聽1r, s).ATR depletion alters nuclear mechanical propertiesNE abnormalities can affect the mechanical properties of the nucleus1,16. When we measured the elastic modulus of ATR-depleted cells by atomic force microscopy (AFM)17, we found a reduced elasticity compared to controls (Fig.聽2a). As the nucleus is the stiffest organelle in the cell18, we performed the same analysis on isolated ATR-defective nuclei and found, again, a reduced elasticity, compared to controls (Fig.聽2b). Acute treatment with ATR inhibitors for 4鈥塰 did not alter nuclear stiffness (Supplementary Fig.聽2a). Hence, the reduced nuclear elasticity results from chronic ATR depletion.Fig. 2: ATR preserves nuclear mechanics.a, b Elastic modulus measurements using AFM. a Cellular stiffness (n鈥?鈥?71, 161, and 179 measurements for shCtrl, shATR1, and ShATR2, respectively) and b stiffness of isolated nuclei from control and ATR-depleted cells (n鈥?鈥?44, 110, and 98 measurements; N鈥?鈥? independent experiments). c Membrane phospholipid composition analysis; total cell and nuclear membrane PC/PE ratio, nuclear PC/PE ratio of individual species (from three biological and two technical replicates; two-way ANOVA test; Bonferroni鈥檚 multiple comparisons test; N鈥?鈥? independent experiments). d FLIM-FRET analysis of cells expressing H2B-GFP and H2B-mCherry; sample images of Fluorescent lifetime (FLIM-FRET index) and overall Lifetime in Hela cells infected with shATR (n鈥?鈥?3) or control (n鈥?鈥?), or treated with DMSO (n鈥?鈥?0) or ATR inhibitor for 4鈥塰 (n鈥?鈥?). e Examples of FRET signals at the NE as measured by Nesprin-2 FRET sensor in control, shATR, and control cells in the presence of ATR inhibitor (3鈥塰), and quantification of FRET signals (n鈥?鈥?12, 88, and 96 measurements for shCtrl, shATR1, and ShATR2 respectively; data pooled from two or three independent experiments). f Immunofluorescent images of YAP cellular distribution in control and shATR cells, (below) quantifications of nuclear to cytoplasmic YAP signal ratio (n鈥?鈥?9, 66, and 58 shCtrl, shATR1, and ShATR2, respectively). g Western blotting of ser127-phosphorylated YAP (N鈥?鈥? independent experiments. Uncropped images available in Source Data file). Scale bar is 20鈥壩糾 in all images. Bar graphs presented as mean鈥壜扁€塖EM and box plot whiskers and outliers plotted using Tukey鈥檚 method in prism7 software. p-values calculated using one-way ANOVA test with Tukey鈥檚 multiple comparisons test (for d) or Dunnett鈥檚 multiple comparisons test (for a, b, e, f). (****P鈥?lt;鈥?.0001; ***P鈥?lt;鈥?.001; **P鈥?lt;鈥?.01; *P鈥?lt;鈥?.05; n.s., not significant).Full size imageLipid composition of the NE is altered in ATR-defective cellsNuclear stiffness is influenced by lamins and nuclear membrane fluidity. We did not find significant alterations in Lamins protein levels or in their relative cellular localization in ATR-depleted cells (Supplementary Fig.聽2b). However, when we measured the lipid composition of isolated nuclear membranes from ATR-depleted and control cells, out of 855 lipid species analyzed, we observed significant differences in the phosphatidylcholine (PC)/phosphatidylethanolamine (PE) ratio (Fig.聽2c). In particular, we observed a specific altered ratio in the 18-carbon and 17-carbon lipid species that represent the most common lipids in membranes (Fig.聽2c). Of note, when we performed a whole cell metabolic analysis, we did not observe specific alterations at the levels of PC/PE ratios.ATR depletion alters chromatin organizationChromatin conformation and distribution can also influence mechanical properties of the nucleus19. We first performed the DNAse I sensitivity assay20 to analyze the chromatin state of control and ATR-depleted cells (Supplementary Fig.聽2c). DNAse preferentially cleaves euchromatin, which is more accessible than heterochromatin to its enzymatic activity21. We found that at early time points, ATR-defective cells exhibited a higher level of undigested DNA that failed to migrate in the gel and likely resulted from heterochromatin accumulation (Supplementary Fig.聽2c). Perinuclear chromatin is generally in the heterochromatic state and is influenced by the levels of H3K9 trimethylation22. ATR-defective cells exhibited a 20% increase in K9 trimethylated histone H3 compared to control (Supplementary Fig.聽2d), implying that ATR depletion promotes increased heterochromatization. This conclusion is confirmed by the fluorescence energy transfer (FRET)-based fluorescent lifetime imaging microscopy (FLIM) assay utilizing GFP- and mCherry-tagged histone H2B to measure chromatin compaction23. We found that, under unperturbed conditions, loss of ATR causes a reduction in the fluorescent lifetime signal, indicating that H2B histones are more compacted (Fig.聽2d). However, we did not observe changes at the level of FRET signal or H3K9 trimethylation upon short-term inhibition of ATR using kinase inhibitors (Fig.聽2d and Supplementary Fig.聽2e), suggesting that the aberrant chromatin state owing to ATR depletion results from long-term effects.ATR depletion affects LINC-mediated nuclear-cytoskeleton connectionsAnother parameter affecting the mechanical properties of the nucleus is the connection between the NE and the cytoskeleton, which is mediated by the Linker of Nucleoskeleton and Cytoskeleton聽(LINC) complex24. We measured this parameter in cells with or without ATR, using a FRET sensor of Nesprin 2G25, a component of the LINC complex24. This assay measures relative changes in the molecular length between the actin-binding domain and the membrane-binding domain of Nesprin 2G. A low FRET signal means that the two domains are far apart, implying that Nesprin 2 is connected to the cytoskeleton and the NE. Cells lacking functional ATR, either through acute inhibition or following long-term depletion, exhibited an increased FRET signal at the NE (Fig.聽2e). This result indicates that the actin-binding and the membrane-binding domains of the linker molecule are in close proximity, presumably due to an altered NE-cytoskeleton connection. This finding is also consistent with Nesprin 2 influencing the NE tension, thus suggesting that ATR defects may also cause a reduced NE tension.ATR depletion causes accumulation of YAP in the cytoplasmThe LINC-mediated mechanical coupling between the NE and the cytoskeleton also influences the nuclear accumulation of YAP, a key mechanosensing transcriptional activator26,27. We investigated whether ATR depletion affected YAP nuclear accumulation (Fig.聽2f). We found that, although in control cells YAP distribution was mostly nuclear, the YAP cytoplasmic fraction increased significantly in ATR-defective cells (Fig.聽2f). Moreover, following ATR depletion, cells accumulated the cytoplasmic form of YAP, phosphorylated in Serine 127 (Fig.聽2g). To exclude the possibility that the change in YAP localization could be a nonspecific consequence of YAP leakage due to NE disruption, we tested whether interfering with Exportin1 (with the Leptomycin-B specific inhibitor), essential for YAP nuclear exit under physiological conditions, could counteract the effect of ATR depletion. Treatment of ATR-depleted cells with Leptomycin-B rescued nuclear YAP levels, with concomitant decrease of phosphorylated YAP (Supplementary Fig.聽2f, g). Following an acute treatment with an ATR inhibitor for 3鈥塰, again, we found enrichment of YAP in the cytoplasm and this effect was abolished by Leptomycin-B treatment (Supplementary Fig.聽2h, i).ATR-defective nuclei collapse following mechanical compressionThe previous results suggest that ATR is required for the normal response of the nucleus to mechanical forces and, as the responses of the cell are dominated by effects on the nucleus, that ATR may also influence cell mechanoresponsiveness. We therefore analyzed the response of ATR-defective cells to mechanical stress using a microfluidic device able to compress cells with a linear step-wise increment of mechanical forces (see 鈥淢ethods鈥? (Fig.聽3a). GFP-tagged cGAS, a cytoplasmic DNA-binding protein, was used as an NE damage marker28. We compressed cells expressing cGAS-GFP using the microfluidic device by applying ten steps of compression with a range of pressure spanning 0 to 50鈥塵Bar, with an increment of 5鈥塵Bar at each step. ATR-depleted cells accumulated perinuclear cGAS-GFP foci in the range between 15 and 30鈥塵Bar, whereas in control cells cGAS-GFP foci started to appear only at 45鈥塵Bar (Fig.聽3b鈥揹). We also analyzed the recovery from cell compression and found that control cells restored the initial nuclear size following de-compression, but ATR-defective nuclei did not completely revert back to the original size (Fig.聽3b, c). We conclude that ATR-defective cells, experiencing compression forces, undergo irreversible nuclear deformation and collapse followed by NE ruptures, thus exposing DNA into the cytoplasm, and accumulating perinuclear cGAS.Fig. 3: ATR-defective cells undergo nuclear collapse upon mechanical compression.Mechanical compression using a microfluidic compression device. a Design and b graph of compression-induced nuclear deformation with respect to the pressure applied, represented as ratio of its initial uncompressed area. In inlet, graph representing step-wise increase of pressure used for the experiment. c Time-lapse images of perinuclear cGAS foci formation in control and ATR-depleted cells under compression. d Amount of pressure applied through the microfluidic pump during the time of NE rupture (formation of first foci of cGAS) (n鈥?鈥?9, 24, and 21 shCtrl, shATR1, and ShATR2, respectively; data pooled from N鈥?鈥? independent replicates). All graphs presented as mean鈥壜扁€塖EM. p-values are calculated using two-way ANOVA test and Tukey鈥檚 multiple comparisons test for b and one-way ANOVA test with Dunnett鈥檚 multiple comparisons test for d (**** if P鈥?lt;鈥?.0001, *** if P鈥?lt;鈥?.001).Full size imageATR defects impair interstitial migrationWhen migrating through tight spaces, nuclei experience tremendous mechanical stress that causes chromatin compaction and nuclear ruptures. Cell survival under these conditions depends on cellular pathways controlling nuclear integrity and nuclear membrane repair28,29,30. The previous sections suggest that ATR defects should compromise the ability of cells to migrate through narrow pores. To explore this possibility, we analyzed the contribution of ATR on interstitial migration using in vitro experiments. Moreover, as nuclear stress and chromatin compaction are prominent features during neurogenesis and metastasis2,31,32, we also addressed whether ATR influences neurogenesis and metastasis by in vivo experiments.We analyzed H2B-mCherry-expressing shATR and control HeLa cells migrating through 4鈥壩糾-wide and 15鈥壩糾-long micro-fabricated constrictions28 (Fig.聽4a). The fraction of cells undergoing nuclear collapse and cell death (Fig.聽4a, see also Supplementary Movie.聽2) while engaging the constrictions were increased in the absence of ATR (Fig.聽4b). Similar results were observed in cells treated with ATR inhibitors (Supplementary Fig.聽3a). We measured DNA damage occurrence by counting 53BP1 foci of cells in the constrictions and found comparable level of foci numbers between normal and ATR-defective cells (Fig.聽4c). Given the well-established links between ATR activity and cell cycle checkpoints, we tested whether the migration defects of ATR-depleted cells were also connected to the cell cycle. Using U2OS-FUCCI cells that mark different phases of cell cycle33, we found that cell death in shATR cells was not dependent on the cell cycle phase (Supplementary Fig.聽3b). We found comparable range of 53BP1-GFP foci in ATR-inhibited cells and in control cells; moreover, the presence of 53BP1 foci did not correlate with cell death (Fig.聽4c). This finding is in agreement with previous reports28,29 showing that 53BP1 foci accumulate in a Ataxia-Telangiectasia mutated protein (ATM)-dependent manner. We note that ATM is fully functional in ATR-defective cells. Therefore, nuclear collapse and cell death of ATR-defective cells in constrictions does not correlate with increased DNA damage or cell cycle stage. Using the Nesprin 2G FRET sensor, we found lower FRET intensity at the leading edge of the nucleus and higher intensity at the lagging edge (Fig.聽4d, e), suggesting that cells respond to migration-induced mechanical stress by modulating the cytoskeleton-NE connection at the leading edge of the NE. ATR inhibition caused a general increase in FRET signal (Supplementary Fig.聽3c, d) and abolished the differential coupling of the cytoskeleton-NE connection in the nuclei engaged in the constrictions (Fig.聽4e). As a control, we used a headless form of Nesprin 2G sensor, which cannot bind cytoskeleton, and found that it did not exhibit significant changes in FRET signals in control or ATR-defective cells (Supplementary Fig.聽3e).Fig. 4: ATR-defective nuclei are inefficient in migrating through narrow pores.a Snapshots of H2B-mCherry labeled control and shATR nuclei passing through constriction. b Cell death measured as the percentage of engaged cells that burst at the constriction (n鈥?鈥?22 and 88 cells for shCtrl and shATR1; data pooled from three independent experiments). c Quantification of 53BP1-GFP foci generated due to constriction in HeLa cells expressing 53BP1-GFP in the presence of DMSO or ATR inhibitor, VE-821 (n鈥?鈥?4 and 17 for DMSO and ATRinh; pooled from two independent experiments). Cells that undergo cell death in the constriction are highlighted in red (for ATRinh). d, e FRET signal measurements of cells engaged in constrictions. d Images of FRET signal at various stages of migration through the constriction and measurement of signal ratio between front (leading half of the nucleus) and back (lagging half of the nucleus) of a nuclei at various stages of migration (n鈥?鈥?1, 10, 10, and 6, respectively). e Ratio of front to back FRET signal in migrating cells (inside or outside the constriction) in the presence DMSO or ATRinh (n鈥?鈥?1, 10, 9, and 10; data from 2 to 3 experiments). f Quantification of nuclear position in the constriction during the first cGAS foci formation (n鈥?鈥?7, 43, and 28; numbers pooled from 3 experiments). g EM images of control and shATR nuclei in constriction (routine 200鈥塶m EM sections). Arrowheads indicate invaginations and NE attached chromatin or nucleoli. h 3D reconstruction of NE at the leading edge from control nucleus in constriction. Green color indicates inner nuclear membrane (INM) and yellow indicates outer nuclear membrane (ONM). i 3D reconstruction of NE section from leading edge of shATR nucleus in constriction. j Quantification of ratio between number of inner nuclear membrane breaks to that of the outer membrane (n鈥?鈥?5 and 13). Scale bar for a, d is 20鈥壩糾, for g is 9鈥壩糾, and for i, h is 200鈥壩穖). Bar graphs presented as mean鈥壜扁€塖EM and dot-plot as mean鈥壜扁€塖D. P-value calculated using two-tailed Student鈥檚 t-test for b, c, j. One-way ANOVA for d, e with Tukey鈥檚 or Sidak鈥檚 multiple comparisons test, and two-way ANOVA for f (****P鈥?lt;鈥?.0001, ***P鈥?lt;鈥?.001, **P鈥?lt;鈥?.01, and *P鈥?lt;鈥?.05; n.s., not significant).Full size imageWe then analyzed the cGAS-GFP foci distribution during interstitial migration (Fig.聽4f). We found that, compared to normal cells, in ATR-defective cells cGAS-GFP foci appeared much earlier at the leading tip of nuclei engaged in the constrictions, suggesting that ATR defects enhance NE fragility during interstitial migration. To directly analyze the integrity of the NE experiencing interstitial migration, we performed EM analysis of cells migrating across constrictions with or without ATR (Fig.聽4g and Supplementary Fig.聽3f鈥搃). Control cells approached the constrictions by deforming their nuclei at the leading edge without aberrant NE structures (Fig.聽4g and Supplementary Fig.聽3f) or, occasionally, with well-organized NE invaginations parallel to the direction of migration (Supplementary Fig.聽3g). Conversely, most of ATR-depleted nuclei were deformed, exhibiting several NE invaginations in the front part of the nucleus, often associated with semicondensed chromatin or nucleoli (Fig.聽4g and Supplementary Fig.聽3f, h). Both control and ATR-depleted cells exhibited sporadic NE ruptures (Fig.聽4h). Our EM analysis allowed us to establish that control cells accumulated extensive outer membrane ruptures and fewer inner membrane ruptures at the leading edge of the nucleus (Fig.聽4h), whereas ATR-depleted cells accumulated extensive NE ruptures involving both the outer and inner membranes, with a higher frequency of inner membrane damage (Fig.聽4i, j). Moreover, we noticed that the leading edge of ATR-defective nuclei exhibited NE portions with a disorganized distribution of outer and inner membranes, likely due to aberrant nuclear membrane remodeling. The NE at the rear part of the nucleus was normal and comparable in control and in ATR-depleted cells (Supplementary Fig.聽3i). Hence, as soon as ATR-defective cells engage narrow constrictions, they fail to adequately respond to the mechanical stress arising at the leading edge of the nucleus and undergo NE deformation and extensive NE damage, which, in turn, cause nuclear collapse and cell death. We failed to observe NE ruptures in ATR-defective cells grown under normal conditions, suggesting that they represent a consequence of mechanical compression. Although the intrinsic alterations of the mechanical properties of ATR-defective nuclei may not affect cell viability under normal conditions, the consequences of nuclear collapse following mechanical stress certainly contribute to cell lethality when cells are forced through narrow passages.ATR influences neurogenesis and metastatic disseminationDuring neurogenesis and metastasis, cells migrate through narrow places. The in vitro observations described above suggest that ATR may play a relevant role in these processes. We analyzed the contribution of ATR in neurogenesis by performing a transwell migration assay of neuroprogenitors isolated from ATR-conditional knockout mouse brain (E13.5 days) (Fig.聽5a). ATR depletion impaired migration of neurosphere-derived cells through 3 or 8鈥壩糾 pore size membranes and, as expected, the defect was more pronounced in the smaller pore size. We next depleted ATR in vivo, in a developing mouse brain. GFP-tagged shRNAs against ATR were electroporated into a developing brain (at day 14.5), to selectively deplete ATR in a subpopulation of migrating neurons (Fig.聽5b). Cortical plates from these brains were analyzed in later stages (E18.5) of embryonic development. By using two independent shRNAs against ATR, we observed a compromised neuronal migration in the cortical plate: although many of the neurons transfected with Luciferase (control) reached the top layers of cortical plate, ATR-depleted neurons were stuck in lower layers (Fig.聽5b, c).Fig. 5: Loss of ATR dampens neuronal migration and tumor cell circulation.a Relative migration of neuroprogenitors isolated from Atr-iKO mice brain were plated into 8鈥壩糾 (n鈥?鈥?, 5 cell lines) or 3鈥壩糾 (n鈥?鈥?, 5 cell lines) polycarbonate membrane insert (ThinCert鈩?, allowed to migrate for 20鈥塰, then fixed and counted (neuroprogenitors isolated from 13 embryos from 2 pregnant mice independently). b, c In vivo neuronal migration: b images of cortical plates from E18.5 embryos. c Percentage of GFP-positive (GFP+) cells present in different equally divided segments of E18.5 brain cortex (n鈥?鈥? animals, 5 section, 2270 GFP+ cells for shLuc; n鈥?鈥? animals, 5 section, 2475 GFP+ cells for shAtr-4; n鈥?鈥? animals, 3 section, 730 GFP+ cells for shAtr-6. Statistical comparisons performed between shLuc individually with shAtr-4 or shAtr-6 from the same layer). d Scheme of in vivo homing assay. Control and shATR HeLa cells labeled with vital dye were injected into the tail vein of immune-compromised mice. Sections of the lung were collected after 2 and 48鈥塰, respectively. e Images of lung surface with labeled HeLa cells residing on them. f Quantifications of cells/field at 2 and 48鈥塰, respectively (n鈥?鈥?5 images from 5 mice and 15 images from 3 mice (for 2 and 48鈥塰) for control; 20 images from 4 mice (for 2鈥塰) and 25 images from 5 mice (for 48鈥塰) for shATR). Scale bar is 100鈥壩糾 in all images. Bar graphs presented as mean鈥壜扁€塖EM and box plot whiskers and outliers plotted using Tukey鈥檚 method in prism-7 software. p-values calculated using two-tailed Student鈥檚 t-test or one-way ANOVA test with Tukey鈥檚 multiple comparisons test (****P鈥?lt;鈥?.0001, ***P鈥?lt;鈥?.001, **P鈥?lt;鈥?.01, *P鈥?lt;鈥?.05, n.s., not significant).Full size imageWe then examined the contribution of ATR in the migration of cancer cells. We injected equal number of shRNA control and ATR-depleted HeLa cells labeled with a vital dye into the tail vein of immunocompromised mice and recovered lung disseminated cells at 2 and 48鈥塰 after injection (see scheme in Fig.聽5d). We found a significant reduction of fluorescent-positive shATR cells in the lung 48鈥塰 after injection compared to controls (Fig.聽5e, f), indicating that ATR is essential to allow cells to sustain the harsh mechanical environment imposed by blood flow and extravasation.ATR interactors known to influence nuclear mechano-responseTo identify potential ATR interactors and targets contributing to nuclear mechanics and dynamics, we performed high-resolution mass spectrometry screens (IP-liquid chromatography (LC)-MS/MS) in exponentially growing U2OS cells expressing GFP-ATR. Combining data from three SILAC (stable isotopes labeling with amino acids in cell culture) quantitative approaches (Supplementary Fig.聽4a), we obtained 479 unique interactors of ATR (analysis details in 鈥淢ethods鈥?and Supplementary Data聽1). Our ATR interactome exhibited a 55% enrichment (265/479) for proteins with (S/T-Q) motif, a potential targets of ATR phosphorylation, while such proteins represent only 22% (5137/17522) in the total phospho-proteins of the PhosphoSitePlus database34 (Supplementary Fig.聽4b). Several of these interactors were reported to be phosphorylated during S phase, mitosis35 (71/479) and in response to DNA damage when ATR is hyperactive4 (48/479) (Supplementary Fig.聽4c).We found 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 observed in ATR-defective cells (Fig.聽6a and Supplementary Fig.聽4d鈥揻). TOPII, an ATR/ATM phosphorylation target4, is involved in modulating DNA topology in S phase and in prophase to deal with the mechanical stress caused by chromosome dynamics. Moreover, genetic evidence36 suggests that Top2 activity is restrained by Mec1ATR. We also found HDAC2 and CHD4, which are members of the Nuclear Remodeling and Deacetylating (NRD) complex, previously identified as ATR interactors37. An ATR-mediated regulation of the TOP2 and NRD complex could account for the phenotypes associated with the heterochromatic and condensed chromatin observed in ATR-defective cells. The screen identified four proteins of the nuclear pore complex (Nup50, 107, 133, 160), regulating nuclear transport, centrosome attachment to the NE during mitosis, as well as YAP mechanotransduction26. In addition, we identified several transport proteins including Exportin1 (XPO1/CRM1), an ATR/ATM phosphorylation target4, involved in rRNA transport38, which might be connected to the accumulation of nucleolar canals described in this study. We also identified Nesprin-2. Intriguingly, Nesprin-2-defective cells also exhibit NE invaginations, and chromatin architecture and nuclear mechanics dysfunctions39,40. Moreover, loss of Nesprin-2 leads to defective neuronal migration in developing mice brain41. We confirmed the ATR鈥揘esprin-2 interaction by immunoprecipitation (IP) followed by western blotting (Fig.聽6b and Supplementary Fig.聽4h) and proximity ligation assay (PLA) (Fig.聽6c). PLA showed that the number of ATR鈥揘esprin-2 foci at the NE increased in cells undergoing chromatin condensation in prophase (Fig.聽6d and controls in Supplementary Fig.聽4i). This observation, combined with the previous result showing that ATR influences nesprin 2 function (Fig.聽2e), suggests that ATR and Nesprin-2 dynamically interact at the NE in response to mechanical stress. ATR depletion did not affect Nesprin-2, protein levels, or intracellular localization (Supplementary Fig.聽4g). In sum, the ATR-phosphointeractome reveals a number of ATR targets involved in the response of the NE to mechanical strains. Although certain targets might be directly regulated by ATR to allow cells to properly respond to mechanical stress, a set of ATR interactors might mediate the recruitment of ATR to the NE or even promote ATR activation when nuclei are stressed. Of note, we did not find ATR interactors involved in lipid metabolism.Fig. 6: ATR interactors.a Schematic summary of ATR interactors and their relative roles at the nuclear envelope and perinuclear areas b Validation of Nesprin-2 and ATR interaction using GFP-ATR immunoprecipitation and western blot analysis (full blots of Nesprin-2 with molecular weights included in Supplementary Fig.聽4h and in Source data file). c Example of ATR鈥揘esprin-2 interaction foci generated by proximity ligation assay (PLA) in interphase and prophase cells (scale bar鈥?鈥?0鈥壩糾). d Quantification of total number of foci present in interphase and prophase cells (n鈥?鈥?5 for interphase and 10 for mitosis, pooled from 2 independent experiments). Graph presented as mean鈥壜扁€塖EM. p-value calculated using two-tailed Student鈥檚 t-test.Full size imageDiscussionOur findings unravel a non-canonical role for ATR in maintaining the normal properties of the NE and in mediating a communication between the cytoplasm and the nuclear interior (chromatin/chromosomes). These observations, together with previous findings showing that (i) ATR is a giant HEAT repeat protein, therefore ideal for sensing mechanical stress7,8; (ii) ATR reliefs the mechanical stress at the NE exerted by the mRNA export machinery42; and (iii) ATR relocalizes at the NE following mechanical stress6,43, suggest that ATR might directly sense and transduce mechanical stress in general and at the NE.We described a variety of nuclear defects owing to ATR depletion or inactivation, some of which are evident already in unchallenged cells; others become obvious when cells (and thus their nuclei, which are their primary load-bearing element) experience mechanical deformations. Although some of the nuclear defects likely represent a direct consequence of ATR inactivation/depletion, others may be related to long-term adaptive responses to limiting ATR. We also identified a set of ATR interactors that 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. Finally, we show that ATR is essential for neurogenesis and cancer cell migration, two processes in which cells and their nuclei must squeeze through tight spaces.Nuclear defects in unchallenged ATR-depleted cellsUnder normal conditions, ATR-defective cells exhibit NE invaginations tethered with semicondensed chromatin and nucleoli. These nuclear defects appear only upon long-term depletion of ATR. Considering that (i) condensation of perinuclear chromatin and nucleolar canals generates transient NE invaginations14,44; (ii) ATR-defective cells are unable to coordinate NEBD with chromatin condensation, as they exhibit a slow condensation process6; and (iii) ATR has been involved in coordinating RNA export through the NE with topological stress45, the most likely possibility is that in ATR-defective cells NE remains deformed due to the inability to efficiently separate from condensed chromatin and to complete the export of nucleolar RNA species.NE invaginations are rare in normal cells, as the NE efficiently recovers its original shape. This process requires the separation of condensed and transcribed chromatin from the NE, and is influenced by NE remodeling activities and by proteins modulating the topological and epigenetic context of chromatin and nuclear transport. ATR regulates type II A and B topoisomerases and condensins4, as well as the NRD complex37 that controls chromatin epigenetics. It is possible that, in ATR-defective cells, deregulated topological and condensation activities may cause nuclear membranes invaginations and ruptures, nuclear fragmentation, and micronuclei formation; the heterochromatic and heavily condensed chromatin could instead result from the deregulation of the NRD complex46. Recent observations47 showed that H3K9 trimethylation marked heterochromatin levels rearrange in response to mechanical stress at the NE and the recovery of the nucleus. In this scenario, our finding that a long-term depletion of ATR accumulates hypercompacted chromatin at the nuclear periphery and elevated levels of H3K9 trimethylation may therefore reflect the inability of ATR-depleted cells to properly recover from nuclear stress. The hypercompacted chromatin at the nuclear periphery and the consequent reduction in nucleosome packaging density in the rest of the nucleoplasm might contribute to explain the low nuclear stiffness of ATR-defective cells.Nuclear membrane defects owing to ATR depletionThe lack of coordination between chromatin condensation and NEBD during cell division in ATR-defective cells causes accumulation of semicondensed chromatin attached to NE fragments6. Moreover, our EM analysis showed that ATR-depleted cells accumulate membrane ruptures already under unperturbed conditions and, following nuclear deformation during interstitial migration, they exhibit massive breakage of the outer nuclear membrane and aberrant membrane remodeling. The ESCRTIII complex plays a key role in sealing membrane holes in the reforming NE during mitotic exit48 and in repairing the NE upon migration-induced rupture28,29. It is possible that the activity of the ESCRTIII complex becomes limiting in ATR-defective cells, due to the massive damage of nuclear membranes. The extensive NE damage and remodeling in ATR-defective cells may also represent the primary cause of the aberrant phospholipid composition of their nuclear membranes. In agreement with this hypothesis, the aberrant phospholipid composition of the NE in ATR-defective cells does not reflect a direct metabolic problem and we failed to identify ATR interactors involved in lipid metabolism.Altered NE-cytoskeleton coupling upon ATR inactivationA key finding of this work is that ATR is a component of the cell mechanotransduction machinery by ensuring appropriate mechanical coupling of the cytoskeleton to the NE. The NE is exposed to forces acting in opposite directions: forces deriving from chromatin dynamics (as outlined above) and opposite forces generated by extracellular matrix (ECM) attachment and conveyed to the NE through the LINC complex and the NE-associated cytoskeleton. ATR orchestrates the integration of all these mechanical inputs, by regulating at once NE dynamics, chromatin condensation, and the LINC function as the mechanosensory properties of the entire cell; this is visualized by the here discovered ATR-YAP mechano-signaling axis.Although abnormalities in nuclear shape and mechanics can impact on genome integrity by generating chromatin fragmentation36 and fork collapse42, under normal conditions, the nuclei remain relatively stable as well as the NE. However, at raising levels of mechanical strain, cells must promptly respond to mechanical stress. Here we show that ATR is critical also for the nuclear response to more severe mechanical challenges. ATR-defective cells fail to properly respond to sub-lethal compression forces and undergo extensive nuclear collapse characterized by NE ruptures. In turn, this imposes an additional stress at the level of nuclear membrane remodeling, as revealed by the presence of mixed portions of the outer and inner membranes at the NE engaged in the constrictions. NE fragmentation under high level of mechanical stress exposes nuclear DNA into the cytoplasm, leading to activation of the cGAS-STING pathway49. The functional consequences of the ATR and cGAS connection remain unexplored but may hold relevant pathological implications, particularly in tissues undergoing mechanical stress.Pathological consequences owing to ATR defectsOur observations indicate that, when the nucleus engages the narrow constrictions, the LINC complexes at the leading edge of the nucleus are tightly bound to the cytoskeleton and the mechanical strain generates extensive ruptures at the outer nuclear membranes. Hence, the nucleus during interstitial migration is polarized at the level of NE. ATR-defective cells fail to maintain the coupling between Nesprin-2 and the cytoskeleton, and to polarize the NE under these conditions and accumulate NE invaginations and extensive ruptures at both nuclear membranes at the leading edge of the nucleus. Moreover, the extensive NE invaginations tethered with semicondensed chromatin may hinder efficient nuclear squeezing and prevent an efficient repair of the nuclear membranes.Cancer cell migration through the ECM requires nuclear deformability, particularly when cells must meander through dense and highly crosslinked collagen type I-rich stroma, extravagate, and sustain the harsh conditions of the blood circulation before extravasating, passing through pores as small as 2鈥壩糾 in diameter50. In fact, altered NE morphology is typical of cancer cells and crucial in the tumor grade assessment, and correlates with prognosis. Cancer cells can adapt to metastatic migration by deregulating the expression of Lamins, but a certain degree of NE stiffness is required to allow a productive migration and to prevent massive NE ruptures50. Our observations suggest that ATR activity might be therefore beneficial for cancer cell migration, thus implying that ATR might play opposite roles in cancer progression, by preventing genome instability and by promoting metastasis. Along this idea, it is interesting to note that our experiments indicate that ATR depletion impairs three-dimensional (3D) invasion and lung homing of cancer cells.Our findings describe a variety of nuclear defects and their pathological consequences (Fig.聽7). However some of this defects such as the altered mechanical coupling between cytoskeleton and NE, and the YAP cytoplasmic retention occur soon after ATR catalytic inactivation, suggesting that these two phenotypes represent a direct consequence of ATR deregulation. Considering that during development, organogenesis requires that stem/progenitor cells migrate towards destination tissues, our observations may contribute to explain some of the developmental defects of Seckel patients bearing genetic defects in ATR. Our results might also explain the increased cell death in non-proliferating neuroprogenitors and neurons of ATR-knockout mice51,52, which cannot be directly ascribed to the role of ATR in replication stress. Moreover, the delocalization of YAP might also contribute to a variety of pathological outcomes such as loss of stem cells and cardiomyopathies27,53. Intriguingly, ATR-conditional knockout mice exhibit a progeroid phenotype that has been related to stem cell loss54.Fig. 7: Graphical summary of ATR defects affecting nuclear morphology and mechanics and relative pathological consequences.In the absence of external stimuli ATR coordinates chromatin processes (such as RNA export, epigenetic and topological transitions, and chromatin condensation in prophase) with NE dynamics, influencing nuclear morphology. ATR defects lead to nuclear deformation, NE remodeling, altered nuclear mechanics, and YAP delocalization. In response to mechanical stress, ATR-defective nuclei collapse leading to NE ruptures and cGAS recruitment at the nuclear periphery. Pathological consequences are also described. See text for details.Full size imageMethodsPlasmidsATR shRNA1 and control (pLKO1) plasmids were from Dr. O.F. Capetillo (CNIO, Spain); ATR shRNA2 was purchased from Sigma (TRCN0000219647); the GFP-ATR plasmid was from Dr. R.Tibbetts (Wisconsin, USA)55. GFP-AU1-ATR plasmid was digested with BamHI to excise out GFP cDNA. The BamHI-digested GFP cDNA insert was cloned into FLAG-ATR-KD plasmid56 also linearized with BamHI, followed by transformation in Escherichia coli. Positive clones after transformations were screened by PCR and BamHI restriction digestion, finally sequenced (list of primers used are provided in Supplementary Table聽1). pTRIP-CMV-GFP-FLAG-cGAS (Plasmid #86675)28, Nesprin tension sensor (pcDNA nesprin TS; plasmid #68127), and nesprin headless control (pcDNA nesprin HL; plasmid #68128)25 were acquired from Addgene plasmid repository. Antibody Source Catalog number Dilution 1. ATR Cell Signal 2790 1鈥?鈥?000 (WB); 1鈥?鈥?00 (IF) 2. TopBP1 Abcam ab2402 1鈥?鈥?000 (WB) 3. Nup133 SantaCruz sc-27392 1鈥?鈥?00 (WB) 4. Tubulin Sigma T5168 1鈥?鈥?000 (WB) 5. Lamin B1 Abcam ab16048 1鈥?鈥?0,000 (WB); 1鈥?鈥?000 (IF) 6. Lamin A/C SantaCruz sc-7292 1鈥?鈥?00 (WB); 1鈥?鈥?00 (IF) 7. Nesprin 2 Thermo Scientific MA5-18075 1鈥?鈥?00 (WB); 1鈥?鈥?00 (IF) 8. Histone H3(tri-met K9) Abcam ab8898 1鈥?鈥?000 (WB) 9. Total Histone H3 Abcam ab1791 1鈥?鈥?000 (WB) 10. Total YAP(63.7) SantaCruz sc-101199 1鈥?鈥?00 (WB); 1鈥?鈥?00 (IF) 11. Phospho YAP (Ser127) Cell Signal 4911S 1鈥?鈥?000 (WB) Secondary antibodies were obtained from IFOM imaging facility:Polyclonal Donkey anti-mouse AlexaFluor-488 AB_2340846 (Jackson ImmunoResearch).Polyclonal Donkey anti-mouse AlexaFluor-594 AB_2340854 (Jackson ImmunoResearch).Polyclonal Donkey anti-mouse AlexaFluor-Cy3 AB_2340813 (Jackson ImmunoResearch).Polyclonal Donkey anti-rabbit AlexaFluor-488 AB_2313584 (Jackson ImmunoResearch).Polyclonal Donkey anti-rabbit AlexaFluor-594 AB_2340621 (Jackson ImmunoResearch).Polyclonal Donkey anti-rabbit AlexaFluor-Cy3 AB_2307443 (Jackson ImmunoResearch). Other reagents Source Catalog number 1. ATR inhibitors 聽聽聽聽鈥僂TP46464 Calbiochem 500508 聽聽聽聽鈥傾Z-20 Tocris 5198 聽聽聽聽鈥僔E-821 Selleckchem S8007 2. Benzonase Sigma E1014 3. Anti-GFP mAb-Magnetic beads MBL, Japan D153-11 4. Mem-PER Plus Kit Thermo Fisher 89842 5. Duolink庐 PLA kit Sigma DUO92102 6. BCA Protein Assay Thermo Fisher 23227 7. RTV615 (PDMS) Momentive Perf. Materials RTV615 8. Leptomycin-B Sigma L2913-.5UG 9. DNAse I Promega M6101 Cell linesU2OS cells stably expressing GFP-ATR and HeLa cells stably expressing mCherry-H2B were reported previously6. U2OS cells expressing the FUCCI reporter was a kind gift from Libor Mac暖rek33. Human primary Seckel fibroblasts (GM18366) and IMR90 were from Coriell Cell Repository. HCT116 and ATRflox/鈭?/i> cells were from The American Type Culture Collection.Cell culture, transfection, and inhibitor treatmentsHeLa and U2OS cells were maintained in Dulbecco鈥檚 modified Eagle鈥檚 medium (DMEM) with GlutaMAX (Life Technologies) supplemented with 10% (vol/vol) fetal bovine serum (FBS, Biowest) and penicillin鈥搒treptomycin (Microtech). Human primary fibroblasts derived from Seckel patient were maintained in DMEM supplemented with 15% FBS (not activated, Sigma-Aldrich) and IMR90 were grown in 10% FBS (not activated). HCT116 and ATRflox/鈭?/i> cells were grown in McCOY鈥檚 5A media. All cells were grown in a humidified incubator atmosphere at 37掳 and 5% CO2.We used Lipofectamine 2000 (Invitrogen) for transfecting plasmids into cells, using the protocol recommended by the manufacturer.HEK293T cells were transfected with shRNA plasmids and viral packaging plasmids to generate lentiviral particles. Desired cell lines were then infected for 16鈥塰 followed by 2鈥壩糶/ml puromycin selection for 24鈥塰. Infected cells were cultured in 1鈥壩糶/ml puromycin containing media and were utilized for experiments up to 10 days after infection.Cells were treated with ATR inhibitors (2鈥壩糓 ETP46464, 10鈥壩糓 VE-821, or 1鈥壩糓 AZ-20) 1鈥塰 before (unless mentioned otherwise) starting the experiment and were maintained in the media throughout the course of the experiment.For cell cycle analysis, cells were fixed with ice-cold ethanol, DNA was labeled with propidium iodide, and quantified using FACS calibur (BD bioscience) system.Membrane fractionation using Mem-Per Plus kitMembrane fractionations were performed following protocol provided by the vendor. Briefly, cells were trypsinized, washed with cell wash solution, resuspended in permeabilization buffer (with or without Benzonase), and incubated for 30鈥塵in at 4鈥壜癈 with constant mixing. Permeabilized cells were then centrifuged for 15鈥塵in at 16,000鈥壝椻€?i>g, soluble fraction was collected, and pellet resuspended and incubated in solubilization buffer for 30鈥塵in at 4鈥壜癈. Samples were then centrifuged for 15鈥塵in at 16,000鈥壝椻€?i>g and supernatant was collected as a membrane fraction.Cell lysis and immunoblottingTotal cell lysates were prepared in lysis buffer (50鈥塵M Tris-HCl pH 8.0, 1鈥塵M MgCl2, 200鈥塵M NaCl, 10% Glycerol, 1% NP-40) Protease (Roche) and Phosphatase inhibitors (Sigma) were added at the time of experiment, and Benzonase (50鈥塙/ml) was added if degradation of nucleic acid was needed. Cell lysates boiled with Laemmli buffer were (20鈥?0鈥壩糶) resolved using NuPAGE庐 (Invitrogen) or Mini-PROTEAN庐 (Biorad) precast gels, transferred to nitrocellulose membrane, and probed as with primary (2鈥塰 at roomtemperature (RT) or overnight at 4鈥壜癈) and secondary antibodies (1鈥塰 at RT), and acquired using ChemiDoc imaging system (Image Lab v5.0). Image intensity measurements were performed using ImageJ.IF assays and quantificationsBriefly, cells were fixed with 4% formaldehyde (15鈥塵in), permeabilized with 0.2% Triton X-100 in phosphate-buffered saline (PBS) (15鈥塵in), blocked with 1% bovine serum albumin in PBS for 1鈥塰 (blocking buffer), incubated with primary antibodies (diluted in blocking buffer) for 1鈥塰 in RT, followed by three PBS washes and then incubated in secondary antibodies (1鈥?鈥?00 in blocking solution) for 1鈥塰 in the dark at RT followed by three PBS washes. Samples were mounted with VectaShield mounting medium containing 4鈥?6-diamidino-2-phenylindole (DAPI). Image acquisition was performed using Leica TCS SP2 confocal scanning microscope, equipped with a 脳63/1.4 numerical aperture (NA) objective. Single optical sections of the images or maximum projections (step size 0.5鈥壩糾) were processed using ImageJ and smoothed to reduce the background noise.Quantification of nuclear morphology, YAP localization: images from random fields (upto 50) were acquired from coverslips stained with DAPI and Lamin or YAP on a UltraVIEW VoX spinning-disc confocal unit with Velocity software (PerkinElmer). Nuclei from each field were manually binned into normal, mild (blebs, invaginations, wrinkles, micronuclei, and multi-nuclei), and severely deformed (with multiple defects), as well as with or without micronuclei alone (in case if field has 20 cells it is combined with next field), which then are averaged to perform statistical analysis. Circularity index was calculated on central section of Lamin staining using ImageJ particle analysis tool and ABsnake plugin. YAP localization analysis was performed following the method described in Elosegui-Artola et al.26. Ratio was calculated between gross intensity measurements from a circular region of 30 pixel diameter in the nucleus and in the cytoplasm from individual cell.Electron microscopyThe staff of EM facility at IFOM performed all the EM analysis. EM examination, Immuno-EM gold labeling based on pre-embedding, EM tomography, and correlative light-electron microscopy (CLEM) were performed exactly as it has been reported previously6,57,58. A brief description of each process is described below.Embedding: cells grown on MatTek dishes (MatTek Corporation, USA) were fixed with of 4% paraformaldehyde and 2.5% glutaraldehyde (EMS, USA) mixture in 0.2鈥塎 sodium cacodylate buffer (pH 7.2) for 2鈥塰 at RT, followed by six washes in 0.2鈥塎 sodium cacodylate buffer at RT. Then cells were incubated in 1鈥?鈥? mixture of 2% osmium tetraoxide and 3% potassium ferrocyanide for 1鈥塰 at RT followed by six times rinsing in 0.2鈥塎 sodium cacodylate buffer (pH 7.2). Then the samples were sequentially treated with 0.3% Thiocarbohydrazide (in 0.2鈥塎 sodium cacodylate buffer) for 10鈥塵in and 1% OsO4 (in 0.2鈥塎 cacodylate buffer, pH 6.9) for 30鈥塵in. Samples were then rinsed with 0.1鈥塎 sodium cacodylate (pH 6.9) buffer until all traces of the yellow osmium fixative have been removed. Then samples were washed in de-ionized water, treated with 1% uranyl acetate (in distilled water) for 1鈥塰 and washed in water again57,59. The samples were subsequently embedded in Epoxy resin at RT and polymerized for at least 72鈥塰 in a 60鈥壜癈 oven. Embedded samples were then sectioned with diamond knife (Diatome, Switzerland) using the ultramicrotome (LeicaEM UC7, Leica Microsystem, Vienna). Sections were analyzed with a Tecnai20 EM (FEI, Thermo Fisher Scientific, Eindhoven, The Netherlands) operating at 200鈥塳V58.Nano-gold labeling: cells grown on MatTeks were fixed with a mixture of 4% paraformaldehyde and 0.05% glutaraldehyde (0.15鈥塎 Hepes buffer, pH 7.2) for 5鈥塵in at RT and then replaced with 4% paraformaldehyde (in 0.15鈥塎 Hepes buffer, pH 7.2) for 30鈥塵in. Cells were washed six times in PBS and incubated with blocking solution for 30鈥塵in at RT. Then cells were incubated with primary antibody diluted in blocking solution overnight at 4鈥壜癈. On the following day, the cells were washed six times with PBS and incubated with goat anti-rabbit Fab鈥?fragments coupled to 1.4鈥塶m gold particles (diluted in blocking solution 1鈥?鈥?00) for 2鈥塰 and washed six times with PBS. Meanwhile, the activated GoldEnhanceTM-EM was prepared according to the manufacturer鈥檚 instructions and 100鈥壩糽 of it was added into each sample well. The reaction was monitored by a conventional light microscope and was stopped after 5鈥?0鈥塵in when the cells had turned 鈥渄ark enough鈥?by washing several times with PBS. Osmification followed: the cells were incubated for 1鈥塰 at RT with a 1鈥?鈥? mixture of 2% osmium tetraoxide (in water) and 3% potassium ferrocyanide (in 0.2鈥塎 sodium cacodylate pH 7.4) and then rinsed six times with PBS and then with distilled water. The samples were then dehydrated: 3鈥壝椻€?0鈥塵in in 50% ethanol; 3鈥壝椻€?0鈥塵in in 70% ethanol; 3鈥壝椻€?0鈥塵in in 90% ethanol; 3鈥壝椻€?0鈥塵in in 100% ethanol. The samples were subsequently incubated for 2鈥塰 in 1鈥?鈥? mixture of 100% ethanol and Epoxy resin (Epon.EMS) at RT; the mixture was then removed with a pipette and finally samples were embedded for 2鈥塰 in Epoxy resin at RT. The resin was polymerized for at least 10鈥塰 at 60鈥壜癈 in an oven.Tomography: two-step CLEM based on the analysis of tomographic reconstructions acquired under low magnification with consecutive reacquisition of EM tomo box under high magnification (脳60,000) and its re-examination was used exactly as described previously60. Briefly, an ultratome (LeicaEM UC7; Leica Microsystems, Vienna) was used to cut 60鈥塶m serial thin sections and 200鈥塶m serial semi-thick sections. Sections were collected onto 1% Formvar films adhered to slot grids. Both sides of the grids were labeled with fiduciary 10鈥塶m gold (PAG10, CMC, Utrecht, The Netherlands). Tilt series were collected from the samples from 卤65掳 with 1掳 increments at 200鈥塳V in Tecnai20 EMs (FEI, Thermo Fisher Scientific, Eindhoven, Tthe Netherlands). Tilt series were recorded at a magnification of 脳20,000 or 脳60,000 using software supplied with the instrument. The nominal resolution in our tomograms was 4鈥塶m, based upon section thickness, the number of tilts, tilt increments, and tilt angle range. The IMOD package and its newest viewer, 3DMOD 4.0.11, were used to construct individual tomograms and for the assignment of the outer leaflet of organelle membrane contours, and best-fit sphere models of the outer leaflet were used for vesicle measurements. Videos were made in 3DMOD and assembled in QuickTime Pro 7.5 (Apple) and the video size was reduced by saving videos at 480p in QuickTime. CLEM was performed exactly as described previously59.FIBSEM: FIBSEM analysis was performed using a FEI Helios NanoLab 660 FEGSEM or G3 equipped with SEM Multi-Detector and ICD detector at accelerating voltage 2.0鈥塳V. Access to both of which was kindly provided by FEI, Co. (FEI, Thermo Fisher Scientific, Eindhoven, The Netherlands). For all high-resolution EFSEM images, a primary beam energy of 2.0鈥塳V was used with a working distance of 1鈥塵m, 3鈥塵s dwell time, and tube bias of 140鈥塚. An Auriga 60 FIB-SEM (Zeiss) microscope with Atlas3D software (FIBICS) was additionally used to collect the 3D data of two cells. Acquisitions were performed according to instructions of the manufacturers.Quantification and statistics: all acquired images were aligned using the TrakEM2 plugin of FIJI. Images were segmented by thresholding with Amira ((FEI, Thermo Fisher Scientific, Eindhoven, The Netherlands). The number of ATR-tagged gold particles in different compartments of the cell was counted and percentages were calculated. The labeling density of ATR on different cellular structures was assessed and calculated as described in ref. 61. For this we used the following criteria: gold particles were considered to label the NE, ER, or mitochondria when these particles were observed over lumens or membranes of these compartments; gold particles were considered as a label of the PM when these particles were observed over the PM. Normality of variant distribution was assessed with Shapiro鈥揥ilk tests. Cumulative probability distributions were compared using the Kolmogorov鈥揝mirnov test. Estimation of the minimal set of samples was performed according to ref. 62. Correlation between two variables was calculated using Pearson鈥檚 product moment correlation.Analysis of cells in channels: we embedded Poly-di-methyl-siloxane (PDMS) molds on MatTek dishes, loaded cell, and incubated for 24鈥塰 to facilitate cell migration into the channels. Cells migrating within the channels were examined under the UltraVIEW VoX spinning-disc confocal unit (PerkinElmer) and acquired images of cells suitable for the future CLEM analysis. The remaining cells from the loading wells were eliminated and 0.05% glutaraldehyde鈥?鈥?% formaldehyde solution (in 0.1鈥塎 cacodylate buffer, pH 7.2) was added to the dish for 5鈥塵in. Cells were then fixed with 2.5% glutaraldehyde鈥?鈥?% formaldehyde (in 0.2鈥塎 cacodylate buffer, pH 7.2) for 10 days, to make cell bodies resistant to the process of the mechanical detachment of PDMS from the MatTek. Then PDMS mold was detached from the MatTek dish and the cells attached to the dishes were processed for EM analysis as described above. After mold detachment, cells were additionally stained with 1% methylene blue in PBS for 3鈥塵in at RT and again examined under a light microscope to confirm the presence of selected cells on the MatTek glass.DNAse I sensitivity assayCells were trypsinized, washed in ice-cold PBS, and resuspended in 2鈥塵l of ice-cold cell lysis buffer (300鈥塵M sucrose, 10鈥塵M Tris pH 7.4, 15鈥塵M NaCl, 5鈥塵M MgCl, 0.5% NP-40, 0.5鈥塵M dithiothreitol, protease inhibitor (Complete, Roche)). After 30鈥塵in, the lysed cells were centrifuged at 500鈥壝椻€?i>g for 5鈥塵in at 4鈥壜癈 and supernatant was discarded. The nuclei were gently resuspended in appropriate amount of reaction buffer (30鈥壩糽 per DNAse I condition). Separate 30鈥壩糽 aliquots were then taken and gently mixed with 70鈥壩糽 of DNase I mix (of varying units) on ice. It was incubated for 15鈥塵in at 25鈥壜癈 and then 700鈥壩糽 of nuclei lysis buffer (100鈥塵M Tris-HCL pH 8, 5鈥塵M EDTA pH 8, 200鈥塵M NaCl, 0.2% SDS) was added to each sample with 50鈥塵g proteinase K. Samples were incubated at 55鈥壜癈 for 1鈥塰; RNaseA (10鈥塵g) was added and again incubated at 37鈥壜癈 for 30鈥塵in. DNA was then extracted using standard phenol鈥揷hloroform technique and was resuspended in 200鈥壩糽 of 0.1 TE, quantified, and were run on 1% agarose gels.AFM measurementsThe AFM measurements were performed using Nanowizard 3 (JPK Instruments, Germany) and a modified silicon nitride cantilever (NovaScan, USA) with a spring constant of 0.03鈥塏/m and a 5鈥壩糾 diameter polystyrene bead adhered at the tip. Central region of the cell was indented with a loading rate of 1.5鈥壩糾/s. The ramp size was 3鈥壩糾 was used. All the measurements were performed as previously described in ref. 17. Nuclei were isolated by treating cells with 1鈥塵l of a 0.01% Igepal CA-630 (a non-ionic detergent, Sigma) and 1% citric acid solution in water for 5鈥塵in. Expelled nuclei from the adherent cells were collected, washed with 5鈥塵l PBS, centrifuged at 800鈥壝椻€?i>g for 5鈥塵in, resuspended in PBS, and dropped onto coverslip for AFM experiments.FRET image acquisition and analysisCells grown on coverslips were injected with the Nesprin 2G-TS construct (50鈥塶g/cell) and the following day, imaged using a DeltaVision Elite imaging system using an Olympus 脳60/1.42 Plan Apo N oil-immersion objective. Three images were collected in sequence at each point: Cyan Fluorescent Protein聽(CFP) (for mTFP1) (ex: 438/24鈥塶m, em: 470/24鈥塶m), FRET (ex: 438/24鈥塶m, em: 559/38鈥塶m), and Yellow Fluorescent Protein (for mVenus) (ex: 513/17鈥塶m, em: 528/38鈥塶m). A single-plane image was background corrected, realigned, converted into 32 bits, and analyzed using an in-house macro in ImageJ. The nuclear membrane of each cell was manually selected as a region of interest and average FRET/CFP ratios calculated for the nuclear membrane region. Approximately 30 cells/conditions were analyzed for n鈥?鈥? experiments. ATR inhibitor was added (ETP46464 2鈥壩糓) 3鈥塰 before image acquisition. ATR inhibitor VE-821 could not be used for this measurements, as they exhibited auto fluorescence. HeLa cells stably expressing Nesprin 2G sensor or the Headless control sensor were generated by Lipofectamine 2000 transfection, Neomycin (G418) selection, and single-cell fluorescence-activated cells sorting of the mVenus/mTFP1-positive population. These cells were loaded onto the channels in the presence of ATR inhibitor or聽DMSO. Single stack image acquired for each field of view every 2鈥塰 for 10鈥塰 duration. Image acquisition parameters and analysis were similar to the above-mentioned experiment.FLIM-FRET analysisFor the acquisition, we used the Leica TCS SP8 confocal microscope with White light laser as excitation source tuned at 488鈥塶m and HC PL APO CS2 脳63/1.40 oil-immersion objective, everything managed by Leica Application Suite X software, ver. 3.5.2.18963. For the lifetime measurements, the above system was implemented with PicoQuant Pico Harp 300 TCSPC module and picosecond event timer, managed by PicoQuant software (SymPho Time 64, ver. 2.4). Data were imported and analyzed using in-house ImageJ macro.Micro-fabricated cell compression chamberA custom-made cell compression device has been invented based on movement of thin membrane attached with a piston, which is precisely controlled by air pressure regulator. The cell compression device was designed using Solidworks and device components were 3D-printed using Dental SG resin (Formlabs) for its biocompatibility. All the components were printed and then washed with IPA for 20鈥塵in, followed by post processing in ultraviolet chamber as suggested by Formlabs. A 20鈥塵m diameter coverslip was stick on the top center of the cell compression device. Silicon membrane was sticked with a piston and then clamped to the bottom of the cell compression device by clamping tools. The assembled cell compression device was then connected to the air pressure regulator. Cells were plated on glass-bottom petridish and maintained in cell incubator. Before the experiment, a cell compression device was capped and locked on the cell culture dish. Images were acquired using 脳40 oil lens (NA鈥?鈥?.3) in PerkinElmer spinning disk microscope.Analysis of cell migration in micro-fabricated channelsWe followed the protocol established previously28 for PDMS channels preparation. Briefly, Polymer and crosslinking agent (RTV615 kit) (mixed in 1鈥?鈥?0 ratio) was used to prepare PDMS channels. These are then plasma treated and embedded onto a glass-bottom dish or a two-chamber LAB-TEK II dish (Thermo Fisher). Then channels were fibronectin-coated and cells were loaded the day before the beginning of time lapse. We chose 15鈥壩糾-long, 4鈥壩糾-wide constriction for experiments involving HeLa and U2OS cell lines. Time-lapse images were acquired (every 10 or 15鈥塵in, with z-stacks) on a UltraVIEW VoX spinning-disc confocal unit with Velocity software (PerkinElmer), equipped with an Eclipse Ti inverted microscope (Nikon) and a C9100-50 electron-multiplying CCD (charge-coupled device) camera (Hamamatsu) or Confocal Spinning Disk microscope (Olympus) equipped with IX83 inverted microscope provided with an IXON 897 Ultra camera (Andor) with OLYMPUS cellSens Dimension software, or on a DeltaVision Elite imaging system using 脳40 oil-immersion (for 53BP1 foci counting) or 脳20 dry objective for a duration of 18鈥?4鈥塰. The images were processed using ImageJ and smoothed to reduce the background noise. All the quantifications were performed manually. Number of cells reaching the constriction within the experimental period was considered as a total cell number. Number of cell death and cell passing were counted per field to calculate the percentages. Fields with no cell migration or death were discarded from analysis. 53BP1 foci were counted manually using ImageJ. Difference was calculated by subtracting number of Foci before engaging the constriction from the number of foci present in the constriction.IP, MS, and data analysisU2OS cells expressing GFP-ATR or GFP alone are cultured in SILAC medium containing light or heavy-labeled l-lysine and l-arginine for 5 days ensuring adequate incorporation of isotopes. These cells were collected with Lysis buffer (50鈥塵M Tris-HCl pH 8.0, 1鈥塵M MgCl2, 200鈥塵M NaCl, 10% Glycerol, 1% NP-40鈥?+鈥塒rotease, and Phosphatase inhibitors) containing Benzonase (50鈥塙/ml) and incubated for 1鈥塰 on ice. Lysates were pre-cleaned by incubating with Protein-A beads for 1鈥塰 at 4鈥壜癈. GFP and GFP-ATR (4鈥塵g of protein lysate per sample) were immunoprecipitated by incubating the lysates with 200鈥壩糽 of anti-GFP-conjugated magnetic beads overnight on a rotor at 4掳. Beads were washed (with Lysis buffer) and eventually pooled before elution with sample buffer. Proteins were then resolved onto a 4鈥?2% NuPAGE庐 precast gel (Invitrogen) and stained by Coomassie colloidal blue. The gel lane was cut into eight or ten slices each of which has been reduced, alkylated, and digested with trypsin as reported elsewhere63. Peptide mixtures were desalted and concentrated on a home-made C18 desalting tip, then peptides were injected in a nanoHPLC (EasyLC Proxeon, Denmark). Peptides separation occurred onto a 25鈥塩m-long column, reverse-phase spraying fused silica capillary column (75鈥壩糾 i.d.) packed with 3鈥壩糾 ReproSil AQ C18 (Dr. Maisch GmbH, Germany). A gradient of eluents A (high-performance liquid chromatography-grade water with 0.1% v/v formic acid) and B (Acetonitrile) with 20% v/v water with 0.1% v/v formic acid) were used to achieve separation, from 7 to 60% of B in 30鈥塵in, at a constant flow rate of 250鈥塶l/min. The LC system was connected to a QExactiveHF mass spectrometer (Thermo Scientific, Bremen, Germany) equipped with a nanoelectrospray ion source (Proxeon Biosystems, Odense, Denmark). Full scan mass spectra were acquired in the LTQ Orbitrap mass spectrometer with the resolution set to 60,000 (@200鈥塵/z) accumulating ions to a target value of 6,000,000. The acquisition mass range for each sample was from m/z 300鈥?650鈥塂a and the analyses were made in duplicates. The 15 most intense doubly and triply charged ions were automatically selected and fragmented in the ion trap after accumulation to a 鈥渢arget value鈥?of 15,000. Target ions already selected for the MS/MS were dynamically excluded for 20鈥塻. Identification and quantification of peptides and proteins were performed with MaxQuant 1.5.2.8 against the human Uniprot complete proteome set, having identified a protein with at least two peptides (one unique), six amino acids of minimal length, false discovery rate鈥?lt;1%, and quantified with at least two ratio counts. Significant outliers scores were calculated using Perseus 1.5.2.664 and those with a p-value鈥?lt;鈥?.05 have been selected for further analysis. For label-free analysis, the procedure was the same but immunoprecipitated samples were kept separate and loaded separately, then digested and analyzed by MS. Proteins were identified using Mascot (v. 2.3.02) and quantification was done using Scaffold (v. 4.3.4). Exclusive unique peptide count was selected to evaluate changes in proteins abundance.For analysis, we pooled data from all experiments, selected candidates that were found as ATR-GFP interactors in at least two experiments, and which were significantly enriched (fourfold) over GFP control in at least one SILAC experiment. We performed Gene Ontology (GO) analysis using DAVID, to generate enriched terms for cellular compartments (GOTERM_CC) and biological processes (GOTERM_BP) (p-value with Benjamini correction鈥?lt;鈥?.05). Revigo tool was utilized to simplify GO terms (resulting list size: 0.7, database: Homo sapiens, semantic similarity measure: SimRel) and R Studio for plotting of Revigo output (size鈥?鈥塴og size, color鈥?鈥塴og10 p-value). Candidates where then manually curated to generate non-overlapping sub-categories of interest (for this study). A network was generated for each sub-category using STRING interaction analysis and the output was plotted using cytoscape.The MS proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE65 partner repository with the dataset identifier PXD020622 (Project Name: ATR interactome in H. sapiens bone osteosarcoma U2OS cells). Files are named SILAC1, SILAC2, and SILAC3.Lipidomic analysis(I) Lipid extraction: Nuclei were isolated according to the protocol66. Nucleus and total cell samples were resuspended in 150鈥塵M ammonium bicarbonate and passed through a 26鈥塆 syringe needle to fragment nucleic acids. Samples were centrifuged at 10,000鈥壝椻€?i>g for 10鈥塵in at 4鈥壜癈 to eliminate cell debris. Lipids were extracted starting from an sample size equivalent of 50鈥壜礸 of proteins, using a two-step extraction protocol (Folch method) with methanol and chloroform in different proportions67. Organic phase fractions were then dried out and resuspended in 50鈥壜礚 of 95% phase A (CH3CN鈥?鈥塇2O 40鈥?鈥?0; 5鈥塵M NH4COOCH3; 0.1% FA) plus 5% phase B (IPA鈥?鈥塇2O 90鈥?鈥?0; 5鈥塵M NH4COOCH3; 0.1% FA) for subsequent analysis. Before extraction, samples were spiked in with 16 internal standards: PC (12鈥?鈥?/13鈥?鈥?) 40鈥塸mol, PE (12鈥?鈥?/13鈥?鈥?) 52鈥塸mol, phosphatidylglycerol (PG) (12鈥?鈥?/13鈥?鈥?) 7.5鈥塸mol, phosphatidylserine (PS) (12鈥?鈥?/13鈥?鈥?) 43鈥塸mol, phosphatidylinositol (PI) (12鈥?鈥?/13鈥?鈥?) 54鈥塸mol, Cer (d18鈥?鈥?/25鈥?鈥?) 100鈥塸mol, cholesterol ester (CE) (19鈥?鈥?) 100鈥塸mol, GlcCer (d18鈥?鈥?/12鈥?鈥?) 50鈥塸mol, LacCer (d18鈥?鈥?/12鈥?鈥?) 50鈥塸mol, sphinganine (d17鈥?鈥?) 50鈥塸mol, sphingosine-1-P (d17鈥?鈥?) 100鈥塸mol, sphingosine (d17鈥?鈥?) 50鈥塸mol, Galactosyl(尾) Sphingosine-d5 20鈥塸mol, d5-TG ISTD Mix I 20鈥塸mol, d5-DG ISTD Mix I 20鈥塸mol, and cholesterol (d7) 800鈥塸mol. (II) Protein quantification: proteins were extracted form 20鈥壜礚 of ammonium bicarbonate resuspended fractions by adding 5鈥壜礚 of lysis buffer (10% NP-40, 2% SDS in PBS) and quantified by BCA protein assay kit (Thermo Fisher Scientific). (III) Lipid profiling data acquisition: lipid extracts were diluted 1鈥?鈥? and 1鈥壜礚 injected on a LC system nLC Ekspert nanoLC400 (Eksigent, 5033460C; Singapore) coupled with a Triple TOF 6600 (AB Sciex, Singapore). Chromatography was performed using an in-house packed nanocolumn Kinetex EVO C18, 1.7鈥壜祄, 100鈥堿, 0.75鈥壝椻€?00鈥塵m. The mobile phases A (CH3CN鈥?鈥塇2O 40鈥?鈥?0; 5鈥塵M NH4COOCH3; 0.1% FA) and B (IPA鈥?鈥塇2O 90鈥?鈥?0; 5鈥塵M NH4COOCH3; 0.1% FA) were used in positive mode. The gradient elution was initially started from 5% B, linearly increased to 100% B in 5鈥塵in, maintained for 45鈥塵in, then returned to the initial ratio in 2鈥塵in, and maintained for 8鈥塵in. Acquisition in MS was performed in positive with the following parameters: mass over charge (m/z) range 100鈥?700, T source 80鈥壜癈, Ion Spray Voltage 2000, declustering potential 80, fixed collision energy 40鈥塚 (+). For Information Dependent Acquisition analysis (Top 8), range of m/z was set as 200鈥?800 in positive ion mode; target ions were excluded for 20鈥塻 after two occurrences (Analyst TF 1.7.1). (IV) Data processing: Lipidview workstation (version 1.3 beta, AB SCIEX, USA) was for lipids identification and quantification. Lipid identification was based on exact mass, retention time, and MS/MS pattern. Lipid species based on precursor fragment ion pairs were determined using a comprehensive target list in LipidView (Sciex). Lipid species identification was performed using the mass tolerance of 0.05 in MS and 0.02 in MS/MS, s/n of 3, and % peak intensity鈥?gt;0 for positive ion mode. Lipid classes included for statistics and downstream analysis were cholesterol ester (CE), sphingomyelin, diacylglycerol, triacylglycerol, ceramide (Cer), PC, PE, PG, PI, PS and lysophosphatidylcholine, lysophosphatidylethanolamine, lysophosphatidylglycerol, lysophosphatidylinositol, lysophosphatidylserine, hexosylceramide, dihexosylceramide, trihexosylceramide, sulphatides (SGalCer), and Cer-phosphate, in positive mode. All statistical analysis was performed using Metaboanalyst 4.0 web tool68 (https://www.metaboanalyst.ca/MetaboAnalyst/faces/home.xhtml). Three experimental and two technical replicates were measured for each condition and 854 lipid metabolites were detected in total. Missing metabolite intensities were imputed with half of the lowest detectable value. Intensities were then normalized by the median of the 322 technically most reliable metabolites (detectable in ~95% of all samples) and averaged over two technical replicates. The total intensities of PC and PE were calculated as sum of all individual phospholipids with choline and ethanolamine head groups.Proximity ligation assayExperiment was performed using rabbit polyclonal ATR antibody (1鈥?鈥?00) and mouse monoclonal Nesprin-2 antibody (1鈥?鈥?50), following the protocol from the manufacturer (Sigma, Duolink PLA Technology).AnimalsWild-type B6/CBAF1 mice used for in utero electroporation were purchased from The Jackson Laboratory. Mice lines (Atr-CER, 129/Sv, and C57BL/6 mixed background) with inducible deletion of Atr were generated as reported in ref. 51. All animals were maintained in the Specific Pathogen-Free聽facility and experiments were conducted according to German animal welfare legislation. All animals were housed at a temperature 22鈥壜扁€?鈥壜癈, humidity 30鈥?0%, and 12鈥塰/12鈥塰 dark/light cycle. Animal experiment protocol was approved by Th眉ringen Landesamt f眉r Verbraucherschutz, Germany.Primers used for genotyping are as follows:ATR10 (5鈥?CTATTTTTTGTTGCTGGTTTTG-3鈥?ATR15 (5鈥?CTTCTAATCTTC-CTCCAGAATTGTAAAAGG-3鈥?Cre1 (5鈥?CGGTCGATGCAACGAGTGATG-3鈥?Cre2 (5鈥?CCAGA-GACGGAAATCCATCGC-3鈥?.Transwell membrane migration assayNeuroprogenitors were isolated from E13.5 embryonic brain of Atr-inducible deletion (ATR-CER) mouse line brain and cultured in neurosphere medium (DMEM/F12 supplemented with B-27, penicillin/streptomycin, 10鈥塶g/ml epidermal growth factor (EGF), and 20鈥塶g/ml basic fibroblast growth factro (bFGF)) for 1~2 days. Then 4-hydroxytamoxifen (4-OHT) was added to induce Atr deletion. Four days after tamoxifen or 4-OHT treatment, neurosphere were trypsinized and resuspended in EGF- and bFGF-free medium. Cells were plated into polycarbonate membrane insert (ThinCert鈩? Greiner-Bio-One GmbH, Frickenhausen Germany) coated with poly-l-lysine. The cells were allowed to migrate for 20鈥塰 at 37鈥壜癈 in 5% CO2 before fixation with paraformaldehyde and staining with DAPI. All cells in the upper side of membrane were carefully removed with a cotton ball before mounting the membrane to glass slice with coverslip on the top. Cells on the underside of filter membrane were counted under a fluorescence microscope and migration activity was calculated by average number of cells per object (脳20) field.In vivo neuronal migrationThe construction of shRNA expression vectors was reported in ref. 69. All oligonucleotides contained the hairpin loop sequence 5鈥?TTCAAGAGA-3鈥? The targeting sequences are as follows:shLuc: 5鈥?GGCTTGCCAGCAACTTACA-3鈥?shATR-4: 5鈥?GGACCTAAACATGTCAGTTCT-3鈥?shATR-6: 5鈥?GCATGCCATCAGTACCCAAGA-3鈥?The efficiency of these shRNA was screened with mouse embryonic fibroblasts cells and Neuro 2A cells. In utero electroporation was performed as described in ref. 70. One microgram of plasmid DNA in PBS was injected into the lateral ventricle of E14.5 embryos followed by electroporation. The embryos were isolated 4 days after electroporation (at E18.5) and processed for cryo-section. Images were acquired with Zeiss ApoTome (Carl Zeiss, Germany) after immunostaining with DAPI. GFP-tagged shRNA vectors were electroporated into wild-type embryonic brain ventricles at E14.5. The embryonic brain is analyzed by imaging at E18.5. Brain cortex was equally divided into ten segmentations and percentage of GFP-positive (GFP+) cells from each segmentation were quantified based on 1~2 sections from indicated number of animals for each plasmid.Short-term lung colonization assayControl HeLa cells (5鈥壝椻€?05) and shATR HeLa cells (5鈥壝椻€?05) were labeled with E-Fluor 670 (Molecular Probes), mixed in 200鈥壩糽 PBS, and injected intravenously. Mice were then sacrificed after 2 and 48鈥塰. The lungs were isolated and fixed in 4% phosphate-buffered formalin. Micrometastases were visualized using a confocal microscope and counted. All animal experiments were approved by the OPBA (Organisms for the well-being of the animal) of IFOM and Cogentech. All experiments complied with national guidelines and legislation for animal experimentation. All animal experiments were performed in accordance with national and international laws and policies. Mice were bred and housed under pathogen-free conditions in our animal facilities at Cogentech Consortium at the FIRC Institute of Molecular Oncology Foundation under the authorization from the Italian Ministry of Health (Autorizzazione Number 604-2016).Statistics and reproducibilityStatistical calculations and graphs were generated with GraphPad Prism5 and Prism7 or using Microsoft Excel (2011) software. All bar graphs are represented as mean 卤 s.e.m. Box plot, whiskers, and outliers are plotted in Graphpad Prism7 using Tukey鈥檚 method. Each dot on the box plots represents a measurement from a single cell. P-values were calculated by Student鈥檚 t-test, one-way or two-way analysis of variance with Sidek鈥檚, Tukey鈥檚, Dunnett鈥檚, or Bonferroni multiple comparisons as indicated in the figure legends of the respective figure. All detailed report of statistical analysis for each graph is included in the source data file. All the experiments presented in this manuscript are successfully reproduced at least in two independent experiments. Exact numbers of replicates are included in the figure legend.Reporting summaryFurther information on research design is available in the聽Nature Research Reporting Summary linked to this article. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE65 partner repository with the dataset identifier PXD020622. All other data supporting the findings of this study are available in main text, the Supplementary Material or in Source Data file. Any additional information can be made available personally upon reasonable request to the corresponding author.聽Source Data are provided with this paper. References1.Cho, S., Irianto, J. Discher, D. E. Mechanosensing by the nucleus: from pathways to scaling relationships. J. Cell Biol. 216, 305鈥?15 (2017).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 2.Kirby, T. J. Lammerding, J. Emerging views of the nucleus as a cellular mechanosensor. Nat. Cell Biol. 20, 373鈥?81 (2018).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 3.Branzei, D. Foiani, M. Regulation of DNA repair throughout the cell cycle. Nat. Rev. Mol. Cell Biol. 9, 297鈥?08 (2008).CAS聽 PubMed聽 Article聽Google Scholar聽 4.Matsuoka, S. et al. ATM and ATR substrate analysis reveals extensive protein networks responsive to DNA damage. Science 316, 1160鈥?166 (2007).ADS聽 CAS聽 PubMed聽 Article聽Google Scholar聽 5.O鈥橠riscoll, M., Ruiz-Perez, V. L., Woods, C. G., Jeggo, P. A. Goodship, J. A. A splicing mutation affecting expression of ataxia-telangiectasia and Rad3-related protein (ATR) results in Seckel syndrome. Nat. Genet. 33, 497鈥?01 (2003).PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 6.Kumar, A. et al. ATR mediates a checkpoint at the nuclear envelope in response to mechanical stress. Cell 158, 633鈥?46 (2014).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 7.Perry, J. Kleckner, N. The ATRs, ATMs, and TORs are giant HEAT repeat proteins. Cell 112, 151鈥?55 (2003).CAS聽 PubMed聽 Article聽Google Scholar聽 8.Grinthal, A., Adamovic, I., Weiner, B., Karplus, M. Kleckner, N. PR65, the HEAT-repeat scaffold of phosphatase PP2A, is an elastic connector that links force and catalysis. Proc. Natl Acad. Sci. USA 107, 2467鈥?472 (2010).ADS聽 CAS聽 PubMed聽 Article聽Google Scholar聽 9.Kyte, J. Doolittle, R. F. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105鈥?32 (1982).CAS聽 PubMed聽 Article聽Google Scholar聽 10.Mitaku, S., Hirokawa, T. Tsuji, T. Amphiphilicity index of polar amino acids as an aid in the characterization of amino acid preference at membrane-water interfaces. Bioinformatics 18, 608鈥?16 (2002).CAS聽 PubMed聽 Article聽Google Scholar聽 11.Murga, M. et al. A mouse model of ATR-Seckel shows embryonic replicative stress and accelerated aging. Nat. Genet. 41, 891鈥?98 (2009).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 12.Cortez, D., Guntuku, S., Qin, J. Elledge, S. J. ATR and ATRIP: partners in checkpoint signaling. Science 294, 1713鈥?716 (2001).ADS聽 CAS聽 PubMed聽 Article聽Google Scholar聽 13.Malhas, A., Goulbourne, C. Vaux, D. J. The nucleoplasmic reticulum: form and function. Trends Cell Biol. 21, 362鈥?73 (2011).CAS聽 PubMed聽 Article聽Google Scholar聽 14.Bourgeois, C. A., Hemon, D. Bouteille, M. Structural relationship between the nucleolus and the nuclear envelope. J. Ultrastruct. Res. 68, 328鈥?40 (1979).CAS聽 PubMed聽 Article聽Google Scholar聽 15.Duheron, V., Chatel, G., Sauder, U., Oliveri, V. Fahrenkrog, B. Structural characterization of altered nucleoporin Nup153 expression in human cells by thin-section electron microscopy. Nucleus 5, 601鈥?12 (2014).PubMed聽 PubMed Central聽 Article聽Google Scholar聽 16.Dahl, K. N. et al. Distinct structural and mechanical properties of the nuclear lamina in Hutchinson-Gilford progeria syndrome. Proc. Natl Acad. Sci. USA 103, 10271鈥?0276 (2006).ADS聽 CAS聽 PubMed聽 Article聽Google Scholar聽 17.Li, Q. S., Lee, G. Y., Ong, C. N. Lim, C. T. AFM indentation study of breast cancer cells. Biochem. Biophys. Res. Commun. 374, 609鈥?13 (2008).CAS聽 PubMed聽 Article聽Google Scholar聽 18.Guilak, F., Tedrow, J. R. Burgkart, R. Viscoelastic properties of the cell nucleus. Biochem. Biophys. Res. Commun. 269, 781鈥?86 (2000).CAS聽 PubMed聽 Article聽Google Scholar聽 19.Mazumder, A. Shivashankar, G. V. Emergence of a prestressed eukaryotic nucleus during cellular differentiation and development. J. R. Soc. Interface 7, S321鈥揝330 (2010).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 20.Krajewski, W. A. Becker, P. B. Reconstitution of hyperacetylated, DNase I-sensitive chromatin characterized by high conformational flexibility of nucleosomal DNA. Proc. Natl Acad. Sci. USA 95, 1540鈥?545 (1998).ADS聽 CAS聽 PubMed聽 Article聽Google Scholar聽 21.Weintraub, H. Groudine, M. Chromosomal subunits in active genes have an altered conformation. Science 193, 848鈥?56 (1976).ADS聽 CAS聽 PubMed聽 Article聽Google Scholar聽 22.Cavalli, G. Misteli, T. Functional implications of genome topology. Nat. Struct. Mol. Biol. 20, 290鈥?99 (2013).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 23.Lleres, D., James, J., Swift, S., Norman, D. G. Lamond, A. I. Quantitative analysis of chromatin compaction in living cells using FLIM-FRET. J. Cell Biol. 187, 481鈥?96 (2009).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 24.Meinke, P. Schirmer, E. C. LINC鈥檌ng form and function at the nuclear envelope. FEBS Lett. 589, 2514鈥?521 (2015).CAS聽 PubMed聽 Article聽Google Scholar聽 25.Arsenovic, P. T. et al. Nesprin-2G, a component of the nuclear LINC complex, is subject to myosin-dependent tension. Biophys. J. 110, 34鈥?3 (2016).ADS聽 CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 26.Elosegui-Artola, A. et al. Force triggers YAP nuclear entry by regulating transport across nuclear pores. Cell 171, 1397鈥?410 e1314 (2017).CAS聽 PubMed聽 Article聽Google Scholar聽 27.Panciera, T., Azzolin, L., Cordenonsi, M. Piccolo, S. Mechanobiology of YAP and TAZ in physiology and disease. Nat. Rev. Mol. Cell Biol. 18, 758鈥?70 (2017).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 28.Raab, M. et al. ESCRT III repairs nuclear envelope ruptures during cell migration to limit DNA damage and cell death. Science 352, 359鈥?62 (2016).ADS聽 CAS聽 PubMed聽 Article聽Google Scholar聽 29.Denais, C. M. et al. Nuclear envelope rupture and repair during cancer cell migration. Science 352, 353鈥?58 (2016).ADS聽 CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 30.Lammerding, J. Wolf, K. Nuclear envelope rupture: actin fibers are putting the squeeze on the nucleus. J. Cell Biol. 215, 5鈥? (2016).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 31.Zwerger, M., Ho, C. Y. Lammerding, J. Nuclear mechanics in disease. Annu. Rev. Biomed. Eng. 13, 397鈥?28 (2011).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 32.Isermann, P. Lammerding, J. Nuclear mechanics and mechanotransduction in health and disease. Curr. Biol. 23, R1113鈥揜1121 (2013).CAS聽 PubMed聽 Article聽Google Scholar聽 33.Macurek, L. et al. Downregulation of Wip1 phosphatase modulates the cellular threshold of DNA damage signaling in mitosis. Cell Cycle 12, 251鈥?62 (2013).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 34.Hornbeck, P. V. et al. PhosphoSitePlus, 2014: mutations, PTMs and recalibrations. Nucleic Acids Res. 43, D512鈥揇520 (2015).CAS聽 PubMed聽 Article聽Google Scholar聽 35.Olsen, J. V. et al. Quantitative phosphoproteomics reveals widespread full phosphorylation site occupancy during mitosis. Sci. Signal. 3, ra3 (2010).PubMed聽 Article聽 CAS聽Google Scholar聽 36.Hashash, N., Johnson, A. L. Cha, R. S. Topoisomerase II- and condensin-dependent breakage of MEC1ATR-sensitive fragile sites occurs independently of spindle tension, anaphase, or cytokinesis. PLoS Genet. 8, e1002978 (2012).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 37.Schmidt, D. R. Schreiber, S. L. Molecular association between ATR and two components of the nucleosome remodeling and deacetylating complex, HDAC2 and CHD4. Biochemistry 38, 14711鈥?4717 (1999).CAS聽 PubMed聽 Article聽Google Scholar聽 38.Zemp, I. Kutay, U. Nuclear export and cytoplasmic maturation of ribosomal subunits. FEBS Lett. 581, 2783鈥?793 (2007).CAS聽 PubMed聽 Article聽Google Scholar聽 39.Luke, Y. et al. Nesprin-2 Giant (NUANCE) maintains nuclear envelope architecture and composition in skin. J. Cell Sci. 121, 1887鈥?898 (2008).PubMed聽 Article聽 CAS聽Google Scholar聽 40.Banerjee, I. et al. Targeted ablation of nesprin 1 and nesprin 2 from murine myocardium results in cardiomyopathy, altered nuclear morphology and inhibition of the biomechanical gene response. PLoS Genet. 10, e1004114 (2014).PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 41.Zhang, X. et al. SUN1/2 and Syne/Nesprin-1/2 complexes connect centrosome to the nucleus during neurogenesis and neuronal migration in mice. Neuron 64, 173鈥?87 (2009).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 42.Bermejo, R. et al. The replication checkpoint protects fork stability by releasing transcribed genes from nuclear pores. Cell 146, 233鈥?46 (2011).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 43.Kidiyoor, G. R., Kumar, A. Foiani, M. ATR-mediated regulation of nuclear and cellular plasticity. DNA Repair (Amst), https://doi.org/10.1016/j.dnarep.2016.05.020 (2016).44.Bozler, J., Nguyen, H. Q., Rogers, G. C. Bosco, G. Condensins exert force on chromatin-nuclear envelope tethers to mediate nucleoplasmic reticulum formation in Drosophila melanogaster. G3 5, 341鈥?52 (2015).CAS聽 Article聽Google Scholar聽 45.Ghavidel, A. et al. Impaired tRNA nuclear export links DNA damage and cell-cycle checkpoint. Cell 131, 915鈥?26 (2007).CAS聽 PubMed聽 Article聽Google Scholar聽 46.Basta, J. Rauchman, M. The nucleosome remodeling and deacetylase complex in development and disease. Transl. Res. 165, 36鈥?7 (2015).CAS聽 PubMed聽 Article聽Google Scholar聽 47.Nava, M. M. et al. Heterochromatin-driven nuclear softening protects the genome against mechanical stress-induced damage. Cell, https://doi.org/10.1016/j.cell.2020.03.052 (2020).48.Olmos, Y., Hodgson, L., Mantell, J., Verkade, P. Carlton, J. G. ESCRT-III controls nuclear envelope reformation. Nature 522, 236鈥?39 (2015).ADS聽 CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 49.Bakhoum, S. F. Cantley, L. C. The multifaceted role of chromosomal instability in cancer and its mmicroenvironment. Cell 174, 1347鈥?360 (2018).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 50.Bell, E. S. Lammerding, J. Causes and consequences of nuclear envelope alterations in tumour progression. Eur. J. Cell Biol. 95, 449鈥?64 (2016).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 51.Zhou, Z., Bruhn, C. Wang, Z. Q. Differential function of NBS1 and ATR in neurogenesis. DNA Repair 11, 210鈥?21 (2012).CAS聽 PubMed聽 Article聽Google Scholar聽 52.Lee, Y. et al. ATR maintains select progenitors during nervous system development. EMBO J. 31, 1177鈥?189 (2012).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 53.Wang, J., Liu, S., Heallen, T. Martin, J. F. The Hippo pathway in the heart: pivotal roles in development, disease, and regeneration. Nat. Rev. Cardiol. 15, 672鈥?84 (2018).CAS聽 PubMed聽 Article聽Google Scholar聽 54.Ruzankina, Y. et al. Deletion of the developmentally essential gene ATR in adult mice leads to age-related phenotypes and stem cell loss. Cell Stem Cell 1, 113鈥?26 (2007).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 55.Tibbetts, R. S. et al. Functional interactions between BRCA1 and the checkpoint kinase ATR during genotoxic stress. Genes Dev. 14, 2989鈥?002 (2000).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 56.Tibbetts, R. S. et al. A role for ATR in the DNA damage-induced phosphorylation of p53. Genes Dev. 13, 152鈥?57 (1999).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 57.Mironov, A. A. et al. Dicumarol, an inhibitor of ADP-ribosylation of CtBP3/BARS, fragments golgi non-compact tubular zones and inhibits intra-golgi transport. Eur. J. Cell Biol. 83, 263鈥?79 (2004).CAS聽 PubMed聽 Article聽Google Scholar聽 58.Fusella, A., Micaroni, M., Di Giandomenico, D., Mironov, A. A. Beznoussenko, G. V. Segregation of the Qb-SNAREs GS27 and GS28 into Golgi vesicles regulates intra-Golgi transport. Traffic 14, 568鈥?84 (2013).CAS聽 PubMed聽 Article聽Google Scholar聽 59.Beznoussenko, G. V. et al. Transport of soluble proteins through the Golgi occurs by diffusion via continuities across cisternae. Elife 3, https://doi.org/10.7554/eLife.02009 (2014).60.Beznoussenko, G. V., Ragnini-Wilson, A., Wilson, C. Mironov, A. A. Three-dimensional and immune electron microscopic analysis of the secretory pathway in Saccharomyces cerevisiae. Histochem. Cell Biol. 146, 515鈥?27 (2016).CAS聽 PubMed聽 Article聽Google Scholar聽 61.Lucocq, J. M. et al. A rapid method for assessing the distribution of gold labeling on thin sections. J. Histochem. Cytochem. 52, 991鈥?000 (2004).CAS聽 PubMed聽 Article聽Google Scholar聽 62.Mironov, A. A. Jr. Mironov, A. A. Estimation of subcellular organelle volume from ultrathin sections through centrioles with a discretized version of the vertical rotator. J. Microsc. 192, 29鈥?6 (1998).PubMed聽 Article聽Google Scholar聽 63.Restuccia, U. et al. pI-based fractionation of serum proteomes versus anion exchange after enhancement of low-abundance proteins by means of peptide libraries. J. Proteomics 72, 1061鈥?070 (2009).CAS聽 PubMed聽 Article聽Google Scholar聽 64.Cox, J. Mann, M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotechnol. 26, 1367鈥?372 (2008).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 65.Perez-Riverol, Y. et al. The PRIDE database and related tools and resources in 2019: improving support for quantification data. Nucleic Acids Res. 47, D442鈥揇450 (2019).CAS聽 PubMed聽 Article聽Google Scholar聽 66.Korfali, N., Fairley, E. A., Swanson, S. K., Florens, L. Schirmer, E. C. Use of sequential chemical extractions to purify nuclear membrane proteins for proteomics identification. Methods Mol. Biol. 528, 201鈥?25 (2009).CAS聽 PubMed聽 Article聽Google Scholar聽 67.Folch, J., Lees, M. Sloane Stanley, G. H. A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem. 226, 497鈥?09 (1957).68.Chong, J. et al. MetaboAnalyst 4.0: towards more transparent and integrative metabolomics analysis. Nucleic Acids Res. 46, W486鈥揥494 (2018).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 69.Zhou, Z. et al. PRMT5 regulates Golgi apparatus structure through methylation of the golgin GM130. Cell Res. 20, 1023鈥?033 (2010).CAS聽 PubMed聽 Article聽Google Scholar聽 70.Gruber, R. et al. MCPH1 regulates the neuroprogenitor division mode by coupling the centrosomal cycle with mitotic entry through the Chk1-Cdc25 pathway. Nat. Cell Biol. 13, 1325鈥?334 (2011).CAS聽 PubMed聽 Article聽Google Scholar聽 Download referencesAcknowledgementsWe thank O. Capetillo (CNIO, Spain), R. Tibbetts (Wisconsin) for reagents, Y. Schwab and J.M. Serra Lleti (EMBL, Germany), Letian Li and Anthony Burgess (Thermo Fisher Scientific, The Netherlands) for assistance in EM analysis, Michele Giannattasio and the members of MF and GS laboratories for assistance and suggestions. We thank IFOM cell culture and imaging facility and Cogentech sequencing services for technical assistance. Work in MF鈥檚 laboratory is supported by grants from Associazione Italiana per la Ricerca sul Cancro, AIRC, Italy (16770, 21416, 22759), Association for International Cancer Research, AICR (ref-14-0338),聽Telethon-Italy (GGP12171), Centro europeo di Nanomedicina (ref-EP002), Ministero dell鈥橧struzione, dell鈥橴niversita e della Ricerca (MIUR-PRIN- 2015SJLMB9), and the European Commission (ref-316390). G.R.K. is supported by Marie Curie Initial Training Networks (ITN), (FP7 鈥渁DDRess鈥? Project number: 316390) fellowship and Italian Association for Cancer Research (AIRC) fellowship (Ref. 19464). A.K. is supported by Wellcome Trust/DBT India Alliance intermediate fellowship, SERB (Department of Science and Technology), and Department of Biotechnology, Government of India. J.B. by is supported by the Danish Cancer Society, the Danish Council for Independent Research, the Novo Nordisk Foundation, the Swedish Research Council, and CancerFonden. S.P. receives support from Denovostem ERC 670126 and FARE-MIUR grants. T.P. is a Cariparo-Foundation fellow.Author informationAffiliationsIFOM- FIRC Institute of Molecular Oncology, Milan, ItalyGururaj Rao Kidiyoor,聽Qingsen Li,聽Giulia Bastianello,聽Christopher Bruhn,聽Irene Giovannetti,聽Adhil Mohamood,聽Galina V. Beznoussenko,聽Alexandre Mironov,聽Umberto Restuccia,聽Vittoria Matafora,聽Angela Bachi,聽Sara Barozzi,聽Dario Parazzoli,聽Emanuela Frittoli,聽Andrea Palamidessi,聽Stefano Piccolo,聽Giorgio Scita,聽Paolo Maiuri,聽Kristina M. Havas聽 聽Marco FoianiInstitut Curie/CNRS, Paris, FranceMatthew Raab聽 聽Matthieu PielUniversity of Padova, Padova, ItalyTito Panciera聽 聽Stefano PiccoloUniversity of Milan, Milan, ItalyGiorgio Scita聽 聽Marco FoianiLeibniz Institute on Aging, Fritz Lipmann Institute, Jena, GermanyZhong-Wei Zhou聽 聽Zhao-Qi WangSchool of Medicine, Sun Yat-Sen University, Shenzhen, ChinaZhong-Wei ZhouGenome and Cell Integrity Lab, CSIR-Indian Institute of Toxicology Research, Lucknow, IndiaAmit KumarDanish Cancer Society Research Center, Copenhagen, DenmarkJiri BartekKarolinska Institute, Stockholm, SwedenJiri BartekFriedrich-Schiller University, Jena, GermanyZhao-Qi WangAuthorsGururaj Rao KidiyoorView author publicationsYou can also search for this author in PubMed聽Google ScholarQingsen LiView author publicationsYou can also search for this author in PubMed聽Google ScholarGiulia BastianelloView author publicationsYou can also search for this author in PubMed聽Google ScholarChristopher BruhnView author publicationsYou can also search for this author in PubMed聽Google ScholarIrene GiovannettiView author publicationsYou can also search for this author in PubMed聽Google ScholarAdhil MohamoodView author publicationsYou can also search for this author in PubMed聽Google ScholarGalina V. BeznoussenkoView author publicationsYou can also search for this author in PubMed聽Google ScholarAlexandre MironovView author publicationsYou can also search for this author in PubMed聽Google ScholarMatthew RaabView author publicationsYou can also search for this author in PubMed聽Google ScholarMatthieu PielView author publicationsYou can also search for this author in PubMed聽Google ScholarUmberto RestucciaView author publicationsYou can also search for this author in PubMed聽Google ScholarVittoria MataforaView author publicationsYou can also search for this author in PubMed聽Google ScholarAngela BachiView author publicationsYou can also search for this author in PubMed聽Google ScholarSara BarozziView author publicationsYou can also search for this author in PubMed聽Google ScholarDario ParazzoliView author publicationsYou can also search for this author in PubMed聽Google ScholarEmanuela FrittoliView author publicationsYou can also search for this author in PubMed聽Google ScholarAndrea PalamidessiView author publicationsYou can also search for this author in PubMed聽Google ScholarTito PancieraView author publicationsYou can also search for this author in PubMed聽Google ScholarStefano PiccoloView author publicationsYou can also search for this author in PubMed聽Google ScholarGiorgio ScitaView author publicationsYou can also search for this author in PubMed聽Google ScholarPaolo MaiuriView author publicationsYou can also search for this author in PubMed聽Google ScholarKristina M. HavasView author publicationsYou can also search for this author in PubMed聽Google ScholarZhong-Wei ZhouView author publicationsYou can also search for this author in PubMed聽Google ScholarAmit KumarView author publicationsYou can also search for this author in PubMed聽Google ScholarJiri BartekView author publicationsYou can also search for this author in PubMed聽Google ScholarZhao-Qi WangView author publicationsYou can also search for this author in PubMed聽Google ScholarMarco FoianiView author publicationsYou can also search for this author in PubMed聽Google ScholarContributionsM.F. and G.R.K. designed all the experiments and wrote the manuscript. G.V.B. and A.M. performed all the EM analysis. G.R.K. and Q.L. performed AFM and compression device measurements. G.B., P.M., D.P., and S.B. contributed towards FRET analysis. Proteomic and lipidomic analysis was done by G.R.K., U.R., V.M., K.H., A.K., and C.B. A.M. and I.G. contributed towards migration assays. E.F. and A.P. performed tail vein injection assays. Z.Z. performed in vitro and in vivo neuronal migration experiments. M.R. and M.P. provided support for micro-fabricated migration assays. S.P. and T.P. provided support for YAP experiments and reagents, and A.K., S.P., G.S., P.M., J.B., Z.Q.W. and A.B. provided constructive inputs and revised the MS.Corresponding authorCorrespondence to Marco Foiani.Ethics declarations Competing interests The authors declare no competing interests. Additional informationPeer review information Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.Publisher鈥檚 note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Supplementary informationSupplementary InformationPeer Review FileDescription of Additional Supplementary FilesSupplementary Data 1Supplementary Movie 1Supplementary Movie 2Reporting SummarySource data Source聽DataRights and permissions Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article鈥檚 Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article鈥檚 Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Reprints and PermissionsAbout this articleCite this articleKidiyoor, G.R., Li, Q., Bastianello, G. et al. ATR is essential for preservation of cell mechanics and nuclear integrity during interstitial migration. Nat Commun 11, 4828 (2020). https://doi.org/10.1038/s41467-020-18580-9Download citationReceived: 06 June 2019Accepted: 25 August 2020Published: 24 September 2020DOI: https://doi.org/10.1038/s41467-020-18580-9 CommentsBy submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate. Sign up for the Nature Briefing newsletter 鈥?what matters in science, free to your inbox daily.