Endosome-generated interior actin network regulates mitotic fidelity and organelle homeostasis downstream of P-cadherin

Endosome-generated interior actin network regulates mitotic fidelity and organelle homeostasis downstream of P-cadherin | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Biological Sciences - Article Endosome-generated interior actin network regulates mitotic fidelity and organelle homeostasis downstream of P-cadherin Pakorn Kanchanawong, Kedsarin Fong-Ngern, Sumit Gupta, Alexandra Kaminer, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9256156/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Accurate chromosome segregation during somatic cell division depends on precise assembly of the mitotic spindle, a process traditionally regarded as microtubule-centric1,2. Recent works in acentrosomal systems have implicated actin in spindle function3,4, but whether spindle-associated actin is a broader feature of centrosomal mitosis remains unclear1. Here we show that the mitotic spindle is cradled by a surprisingly delicate cytoplasmic interior actin meshwork that requires optimized preservation and super-resolution imaging5 for its visualization. We show that this mitotic interior actin network contributes to mitotic fidelity and is generated by early endosomes containing P-cadherin, a cell–cell adhesion receptor frequently dysregulated in cancer6. Mechanistically, P-cadherin undergoes cell-cycle-coupled trafficking through Rab5 endosomes and, together with Cdc42, MRCKβ, and FMNL2, drives assembly of dynamic actin cables. Loss of P-cadherin causes widespread spindle defects and impaired chromosome congression. Strikingly, P-cadherin depletion simultaneously collapses this endosome-derived actin network and enhances mitochondria-associated actin7,8, giving rise to peri-mitochondrial actin shells and mitochondrial dysfunctions, consistent with reciprocal regulation between distinct interior actin pools. Our findings reveal a cadherin-dependent, endosome-generated actin system that dynamically organizes the mitotic cell interior to ensure faithful chromosome segregation, with implications for epithelial homeostasis and cancer. Biological sciences/Cell biology/Cell division/Mitosis Biological sciences/Cell biology/Cell division/Mitotic spindle Biological sciences/Cell biology/Cytoskeleton/Actin Biological sciences/Cell biology/Cell adhesion/Cadherins Biological sciences/Cell biology/Cellular imaging/Super-resolution microscopy P-cadherin mitotic spindle mitotic fidelity mitosis endosomes actin mitochondria endosomal signaling Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction High fidelity segregation of genetic materials during cell division requires the accurate assembly of the microtubule-based spindle apparatus, which orchestrates the precise spatial organization and subsequent equipartition of the chromosomes 2,9,10 . Compromised integrity of the spindles gives rise to deleterious consequences ranging from cell death to developmental defects and cancer 11-13 . In recent years, substantial plasticity in mechanisms regulating the spindles have been recognized. For instance, although the centrosomes serve as dominant spindle microtubule-organizing centers in many somatic cells, robust spindle assembly proceeds without centrosomes in diverse contexts including mouse preimplantation embryos (acentrosomal mitosis) or in oocyte meiosis (meiosis I and meiosis II) 14,15 . Furthermore, in such acentrosomal spindles the actin cytoskeleton plays essential and multifaceted roles governing the fidelity of chromosome segregation, including forming an integral part of the spindle architecture in the oocyte meiotic spindles 4,16 or serving as a peri-spindle actin shell in mouse preimplantation embryos 3 . In contrast, centrosomal mitosis has conventionally been considered to be primarily microtubule-driven, in significant part due to the observed absence of actin in the spindle proximity 1 . While there have been multiple studies supporting the centrosomes as active actin polymerization sites during mitosis 17,18 , the contribution of actin to faithful chromosome segregation during mitosis remains incompletely understood. The involvement of actin in mitotic cell division is most extensively characterized in terms of its roles in regulating cell geometry and cell mechanics 19,20 . This is primarily organized by the actin cortex, which actively drives cell shape changes 21 , coordinates cell division polarity 22 , and mediates cytokinesis 23 . Beyond the cortical pool, a recently discovered actin population in the cellular interior contributes to proper mitosis by ensuring equipartition of endomembrane organelles such as mitochondria between daughter cells 24,25 . Unlike the robust cortical actin, such interior actin networks consist of delicate meshes of transient actin cables that oscillate as ‘waves’ or form ‘comet tails’ trailing mitochondria, and thus its characterization require demanding high-resolution imaging conditions. At present, to what extent such dynamic organelle-associated actin contributes to other aspects of mitosis remains unexplored. In epithelial tissues, mitosis must be tightly regulated to preserve tissue integrity, maintain the physical barrier, and prevent tumorigenesis 26 . E-cadherin, along with its associated catenin complexes, mediate the physical connection between neighbouring cells and organize the cortical actin architecture at the epithelial cell-cell junctions 27-30 . Importantly, E-cadherin is also directly involved in mitotic spindle orientation 31 , contributes to contact inhibition via catenin-dependent signalling 32 , and is broadly considered a tumour suppressor 33 . However, during development and disease, E-cadherin expression is often supplanted or supplemented by other classical cadherins, whose functional distinctions have not been fully understood 34 . E-cadherin and P-cadherin are close paralogs encoded by tandem, immediately adjacent genes within the classical cadherin cluster on human chromosome 16q22.1 (separated by ~3.5 kb intergenic interval) and co-expressed in the epithelia of multiple tissues 35,36 . While aberrant P-cadherin expression is strongly associated with poor prognosis in a wide variety of cancers (Paredes et al., 2012; Wu et al., 1993), P-cadherin also fulfils context-specific roles in normal tissues, ranging from mammary gland epithelium and epidermis 37,38 to placenta, where it was originally discovered 39 . Despite the clinical relevance of P-cadherin as a tumour promoter and tumour marker, the molecular mechanisms governing its functions and distinguishing it from E-cadherin remain poorly understood. In this study, while investigating the functional distinction between E- and P-cadherin, we uncovered an essential role of P-cadherin in safeguarding mitotic fidelity in epithelial cells. This is mediated by a delicate actin meshwork in the cytoplasmic interior, which is common in many cell types but requires optimized super-resolution imaging conditions for its visualization. Depletion of P-cadherin causes widespread mitotic spindle defects and severely reduces the interior actin meshwork. We showed that such interior actin is generated by P-cadherin-positive early endosomes through a Cdc42/MRCKβ/FMNL2 pathway and serves to facilitate chromosome congression and support mitotic spindle organization. Interestingly, the disruption of interior actin upon P-cadherin depletion is accompanied by a significant amplification of mitochondria-associated actin, which results in actin shells encasing the mitochondria in conjunction with impaired mitochondrial morphology and functions, indicative of reciprocal regulation. The mitosis-supporting roles of the endosome-generated interior actin networks may represent a new paradigm for how the internal architecture of cells are dynamically organized during cell division to ensure proper mitotic fidelity. Results P-cadherin is essential for mitotic fidelity To dissect the differential functions of E- and P-cadherin, we used MDCK II epithelial cells, known to express both cadherins 40,41 ( Extended Data Fig. 1a-b, h ). We first assessed their junction-forming ability by plating cells on substrates coated with F c -conjugates of the extracellular domains of E- or P-cadherin 42 . Wild-type (WT) cells spread and formed cadherin-based adhesions on both E- and P-cadherin-F c substrates ( Extended Data Fig. 1c-d, f-g ). We next generated biallelic P-cadherin knock-out (KO) clones using CRISPR/Cas9 ( Extended Data Fig. 1h-j ) and compared them with previously described E-cadherin KO MDCK 43,44 . As expected, E-cadherin KO cells failed to attach to E-cad-F c coated substrates, while P-cad KO cells failed to attach to P-cadherin-F c coated substrates ( Extended Data Fig. 1e ), confirming that both cadherins are capable of forming only homophilic interaction. We next examined their differential localization. Endogenous E- and P-cadherin were both observed at the cell-cell contacts, but P-cadherin exhibited lower overall expression and a higher proportion of cytoplasmic puncta ( Extended Data Fig. 1a-b ). Notably, dividing cells displayed a significant increase in P-cadherin cytoplasmic puncta compared with neighboring non-dividing cells, with puncta densely clustered near the mitotic spindle poles ( Extended Data Fig. 1a-c ). Immunoblot analysis of synchronized MDCK WT populations revealed that P-cadherin protein level increased during M phase, whereas E-cadherin levels remained constant ( Extended Data Fig. 1d ). Strikingly, confocal imaging of proliferating monolayers revealed a strong impairment of mitotic fidelity in P-cadherin KO cells compared with WT ( Extended Data Fig. 1e-f) . Observed defects included spindle mis-orientation ( Extended Data Fig. 1h-k , P-cad KO: 21.91 % vs WT: 3.26 %), chromosome misalignment (P-cad KO: 42.38 % vs WT: 12.56 %), spindle distortion (P-cad KO: 28.75 % vs WT: 2.68%), and multipolar spindles (P-cad KO: 3.01% vs WT: 0%). In contrast, the mitotic fidelity of E-cadherin KO cells was comparable to WT ( Extended Data Fig. 1f, 2c-d ). These findings were validated using additional P-Cadherin KO clones, siRNA-mediated knock-down of P-cadherin in MDCK, and in other P-cadherin expressing cell lines, all of which exhibited comparable mitotic defects ( Extended Data Fig. 1k-m, 2f-m ). Moreover, re-expression of P-cadherin-GFP in P-cadherin-KO cells restored mitotic fidelity to WT level ( Extended Data Fig. 1f-g) . Together, these results indicate that P-cadherin is essential for ensuring mitotic fidelity in multiple epithelial cell models. Mitotic spindles are nestled within delicate P-cadherin-dependent interior actin network s Confocal imaging of dividing cells revealed that P-cadherin KO cells exhibited prominent actin-rich clumps in the vicinity of the mitotic spindles, in contrast to the relatively homogeneous actin intensity observed in MDCK WT cells ( Extended Data F ig. 2a ). Motivated by recent AiryScan and structured illumination super-resolution microscopy studies showing that a dynamic lattice of actin cables permeates the cytoplasmic interior of dividing HeLa cells and regulates mitochondrial inheritance 24,25 , we asked whether the actin perturbations caused by P-cadherin ablation might correspond to defects in the cytoplasmic actin network. Previous visualizations of actin filaments associated with spindles or centrosomes have often relied on glutaraldehyde fixation, suggesting that additional stabilization is required to preserve these structures 17,25,45 . To preserve epitopes for immunostaining while maintaining cytoplasmic actin integrity, we optimized a paraformaldehyde-based fixation protocol ( Extended Data Fig. 2b ). In particular, we found it essential to minimize mechanical disturbance during fixation, maintain the specimen at 37 °C, and perform imaging shortly after fixation. Under these conditions, delicate meshes of cytoplasmic actin filaments became apparent using multi-channel spinning-disc confocal microscopy coupled with ~120-nm super-resolution enhancement 5,46 , whereas sub-optimal fixation or prolonged delays between fixation and imaging markedly reduced actin cable visibility ( Extended Data F ig. 2a ). These stringent requirements indicate that such interior actin networks are unusually labile, which may help explain why they have received limited attention in previous investigations despite being present in a wide range of commonly used cell types ( Extended Data Fig. 2f-m ) 47 . Using our optimized protocol, we observed an extensive mesh of actin cables extending throughout the cytoplasm at various mitotic stages ( Extended Data Fig. 2a ). During early prophase and nuclear envelope breakdown (NEBD), the actin cables can be seen emanating from the vicinity of the spindle poles ( Fig. 2b ). Such dense actin cable networks cradle the mitotic spindle, becoming highly enriched at both spindle poles during metaphase, and persisting through telophase and cytokinesis ( Fig. 2a ). At our imaging resolution, the interior actin networks in MDCK cells appear to largely involve elongated actin cables, thus distinguishing it from the recently discovered peri-spindle actin based on Arp2/3-dependent branched networks that encase acentrosomal mitotic spindles in mouse preimplantation embryos 3 . In P-cadherin KO cells, we found that the interior actin networks were drastically depleted, although the actin cortex remained intact ( Fig. 2d ). To quantify this effect, we developed an analysis approach to map interior actin density relative to the centrosomes and normalized against cortical actin ( Fig. 2c, e, Extended Data Fig. 3a-c ). This revealed nearly 50% reduction of the interior actin density in P-cadherin KO compared to WT ( Fig. 2e, Extended Data Fig. 2f-m ). Similar depletion was observed after siRNA-mediated knockdown (KD) of P-cadherin, ruling out long-term adaptation or off-target KO effects ( Extended Data Fig. 1k-m ). We also verified that the expression level of β-actin remains similar between WT and KO ( Extended Data Fig. 1i ). KD of P-cadherin in other epithelial and cancer cell lines also depleted the interior actin network and impaired mitotic fidelity ( Fig. 2f-m ). Conversely, re-expression of P-cadherin-GFP in P-cadherin KO cell restored the interior actin networks as well as mitotic spindle organization ( Fig. 1f-k ). These results indicate that P-cadherin is essential for the integrity of the interior actin network, which is in turn implicated in mitotic fidelity regulation. P-cadherin-dependent interior actin network facilitates mitotic spindle functions Actin-based steric cages and nuclear membrane remnants have been proposed to physically constrain chromosome scattering, thereby promoting efficient chromosome capture and segregation during cell division 48-50 . To assess how the interior actin network contributes to proper mitosis, we next performed 4-color live-cell super-resolution imaging of WT and P-cadherin KO cells, to simultaneously track actin, microtubules, mitochondria, and chromosomes.In WT cells,actin cables were generated in the vicinity of the spindle poles immediately after nuclear envelope breakdown (NEBD) and continued into prometaphase, as the centrosomes moved toward each other ( Fig. 3a , Extended Data Fig. 3e , Movie 1 ). In contrast, in P-cadherin KO cells, actin cable formation was diminished near the spindle but instead became enriched peripherally near mitochondria ( Fig. 3b, Extended Data Fig. 3e, Movies 2 , and further below). Particle Image Velocimetry (PIV) 51 analysis revealed that in WT cells actin cables moved at an average speed of ~3.3 µm/min immediately after NEBD, while slowing to ~2.5 µm/min within 120 s ( Fig. 3c, e-f ). These dynamics coincided with inward spindle pole motion and chromosome congression ( Fig. 3h-j ). On the contrary, in P-cadherin KO cells the average actin speed was significantly slower, decelerating within a narrower range of 2-2.5 µm/min ( Fig. 3d-f ), while the total actin cable length in the nuclear region was significantly reduced compared to WT ( Fig. 3k-l ). Quantification of chromosome motion at NEBD and at metaphase revealed a significant reduction in chromosome speed in P-cadherin KO cells (Fig. 3g, Extended Data Fig. 3f ), while inward centrosome migration and chromosomes capture by spindle microtubules, as quantified by clustered chromosome areas, were delayed ( Fig. 3i ). In conjunction with the depleted interior actin network, the central positioning of the metaphase spindles as well as the in-plane spindle orientation expected for mitotic epithelial cells were impaired in P-cadherin KO ( Extended Data Fig. 3g-h ). Together, these results link the aberrant interior actin organization in P-cadherin KO cells to defects in multiple mitotic steps that collectively contribute to impaired mitotic fidelity. Loss of P-cadherin enhances peri-mitochondrial actin and impairs mitochondrial morphology and functions Alongside the depleted interior actin networks, P-cadherin KO cells also showed sporadic clusters of dense actin shells surrounding organelle-like structures ( Fig. 2d, 3b ). Because cytoplasmic actin cables drive mitochondrial motion during HeLa mitosis 25 , we hypothesized that these structures may correspond to peri-mitochondrial actin. To test this, we made use of an actin-chromobody probe (AC-Mito) that preferentially labels peri-mitochondrial actin 52,53 . The AC-Mito probe was expressed at the same level in both WT and P-cadherin KO cells; a level that showed no detectable effect on mitochondrial morphology in WT. As shown in Fig. 4a-b (see also Movie 3 ),we found that peri-mitochondrial actin was significantly increased in P-cadherin KO cells, seen as co-fluctuating with mitochondria in the kymograph ( Fig. 4b ). Consistently, live-cell imaging of P-cadherin KO cells expressing LifeAct-mEmerald 54 revealed prominent halos around mitochondria, which were not observed in WT ( Fig. 4c ). Furthermore, while the mitochondria in WT were largely excluded from the mitotic spindle zone during metaphase, in P-cadherin KO a significant fraction of mitochondria were found overlapping with the metaphase spindle ( Fig. 4d-e ) indicative of defective mitotic organelle exclusion 55 . During interphase in particular, we observed that P-cadherin KO cells exhibited fragmented and anisotropically distributed mitochondria, compared to WT ( Fig. 4f ). Morphometric analysis showed that mitochondria in P-cadherin KO cells were significantly smaller, shorter, and more circular ( Fig. 4g ). Concomitantly, P-cadherin KO cells contained a greater number of mitochondria but a lower overall mitochondria area coverage, indicating a more heterogeneous distribution and increased aggregation ( Fig. 4g ). Live-cell tracking further revealed a marked reduction in mitochondrial motility in P-cadherin KO cells ( Fig. 4k ). To assess functional consequences, we employed live-cell indicators for mitochondrial membrane potential (TMRM, Fig. 4l ) and mitochondrial superoxide (MitoSox, Fig. 4m ). P-cadherin KO cells displayed a significant decrease in membrane potential together with an increase in reactive oxygen species, indicative of impaired mitochondrial functions. Taken together, the increase in peri-mitochondrial actin shells coupled with the loss of interior actin networks suggests that P-cadherin is required to maintain a homeostatic balance between organelle-associated and cytoplasmic interior actin pools, evocative of the inter-network competition framework 56 . We propose that the depletion of P-cadherin disrupts a competing cytoplasmic actin polymerization pathway, thereby favouring mitochondrial actin polymerization 8 and driving the observed changes in mitochondrial morphology and function. Consistent with this model, overexpression of P-cadherin–GFP in P-cadherin KO cells restored highly elongated mitochondrial morphologies ( Fig. 4j ). Further in-depth studies will be required to delineate the molecular mechanisms and broader implications of these P-cadherin–dependent mitochondrial phenotypes. P-cadherin is trafficked in Rab5 endosomes in association with dynamic actin cables To dissect the mechanism underlying mitotic functions of P-cadherin, we next performed live-cell imaging of fluorescently-tagged P-cadherin re-expressed in KO cells. This revealed that in dividing cells P-cadherin-GFP was localized in vesicles clustered near the centrosomes, as labeled by SiR-tubulin, and these vesicles appeared to be transported on the spindle microtubules ( Fig. 5a-c ). Furthermore, in live-imaging of KO cells co-expressing P-cadherin and LifeAct, we observed actin cables trailing behind P-cadherin-positive vesicles in the cytoplasm of both interphase and mitotic cells (interphase, Fig. 5d: P-cadherin-tdTomato and LifeAct-mEmerald, Movie 4 ; prometaphase, Fig. 5e : P-cadherin-GFP and LifeAct-mScarlet-I , Movie 5 ). Altogether, these findings implicate P-cadherin-containing endosomal vesicles in generating the actin cables that are incorporated into interior actin networks during mitosis, which may in turn contribute to proper mitotic spindle functions described above. We next sought to identify the endosomal carriers of P-cadherin. Different Rab GTPases are known for distinct localization patterns during various stages of mitosis 57 . The enrichment of P-cadherin vesicles at the mitotic spindle poles ( Fig. 5a ) is consistent with a subset of endosomal markers. Co-localization analysis of P-cadherin-GFP with mCherry fusions of candidate Rab GTPases ( Fig. 5g ) revealed an extensive co-localization between P-cadherin and the early endosome marker Rab5 in both interphase and mitotic cells ( Fig. 5i-j ). Immunostaining for endogenous P-cadherin and Rab5 further confirmed their colocalization ( Fig. 5h ). Cytoplasmic domain of P-cadherin determines trafficking in Rab5 endosomes and is essential for interior actin integrity and mitotic fidelity The requirement of P-cadherin, but not E-cadherin ( Fig. 1f, Extended Data Fig. 2d ), for mitotic fidelity is unexpected given their very close sequence similarity. To dissect this difference, we generated chimeric constructs swapping their cytoplasmic tails. In P-cadherin KO cells, expression of ecto-E-cad::cyto-P-cad-GFP rescued mitotic fidelity, whereas ecto-P-cad::cyto-E-cad-GFP did not ( Fig. 5l ). Quantification confirmed that ecto-E-cad::cyto-P-cad-GFP significantly reduced chromosome misalignment compared to ecto-P-cad::cyto-E-cad-GFP ( Fig. 5m ). Importantly, a dense interior actin network was restored by ecto-E-cad::cyto-P-cad-GFP but not by ecto-P-cad::cyto-E-cad-GFP ( Fig. 5k ). Additionally, the ecto-E-cad::cyto-P-cad-GFP construct localized extensively to endosomal vesicles in the cytoplasmic interior, co-localizing with Rab5 similarly to P-cadherin-GFP, while ecto-P-cad::cyto-E-cad-GFP remained largely confined to cell-cell contacts ( Extended Data Fig. 4b-c ). These observations identify the cytoplasmic tail of P-cadherin as the determinant of both its mitotic and interior actin regulatory functions, mediated through its endosomal localization rather than canonical cell-cell junctional roles. Endocytosed population of P-cadherin contributes to interior actin network that drives chromosome congression We next asked whether the mitotic functions of P-cadherin require endocytic recycling. Cadherin cytoplasmic domains provide binding sites for catenins, with p120-catenin stabilizing plasma-membrane presentation by masking an endocytic motif 30,58 . Consistent with this, we found that β-catenin co-localized extensively with P-cadherin-GFP at both cell-cell contacts and endosomes in KO rescue cells ( Extended Data Fig. 4d-f ), while p120-catenin co-localized with P-cadherin exclusively at cell-cell contacts and was absent from the endosomes ( Extended Data Fig. 4g-h ). We next inhibited clathrin-mediated endocytosis using dynasore 59 . As shown in Extended Data Fig. 4i , we found thatdynasore treatment gave rise to a severe depletion of interior actin network in association with defective spindle organization in WT as well as in P-cadherin KO cells rescued with P-cadherin-GFP or ecto-E-cad::cyto-P-cad-GFP. In conjunction, quantification of mitotic fidelity revealed a significant impairment ( Extended Data Fig. 4j-k ). We interpret these outcomes as arising from the inhibition of P-cadherin endosomal internalization, supported by the severe depletion of GFP-positive vesicles upon dynasore treatment compared to control ( Extended Data Fig. 4i ). We next combined pharmacological inhibition of endocytosis with cell cycle synchronization to test whether endosomal P-cadherin is required for the assembly of the interior actin network that supports chromosome congression. As shown in Extended Data Fig. 4l , we synchronized LifeAct-mEmerald-expressing WT cells at G2/M using RO3306 treatment for 20 h. At 1 h prior to G2/M release, dynasore was added to inhibit clathrin-mediated endocytosis. Wash-out of RO3306 was performed in the presence of dynasore to enable mitotic progression in the presence of endocytosis blockade. Mitosis was allowed to progress for 1.5 h following G2/M release, followed by S-trityl-L-cysteine (STLC) treatment, at which point the cells became enriched at late prometaphase with depleted interior actin network. Then, dynasore wash-out was performed to initiate P-cadherin endocytosis, and live-cell imaging commenced at 30 min post-washout. As shown in Movie 6 and Extended Data Fig. 4l, we observed that the interior actin cables became more prominent over time, forming nest-like meshes around the clustering chromosomes. Quantification of chromosome area following the wash-out revealed a progressively decreasing trend, indicative of chromosome congression ( Extended Data Fig. 4m ). Together, these results support a model in which P-cadherin internalization via clathrin-mediated endocytosis drives interior actin network generation that facilitates chromosome segregation and ensures mitotic fidelity ( Fig. 5f ). Mitotic interior actin network depends on Cdc42- and formin-dependent actin polymerization During mitosis, both the activity and centrosomal localization of the small GTPase Cdc42 peak at metaphase, reminiscent of our observed P-cadherin endosome distribution 60,61 . Therefore, we tested whether P-cadherin functions are coupled to Cdc42 activity. Using the mRFP-wGBD biosensor to probe for active Cdc42 62 , we observed its extensive co-localization with P-cadherin-GFP at endosomal vesicles in interphase cells and at the spindle poles of dividing cells, while no co-localization was observed at the cell-cell contacts ( Fig. 6a-b, Extended Data Fig. 5a-b ). Further, when mRFP-wGBD is expressed in un-rescued P-cadherin KO, we observed that fluorescence intensity appears largely cytosolic with a significant reduction in the number of active Cdc42 vesicles compared to WT cells or P-cadherin KO rescued with P-cadherin-GFP ( Fig. 6d, Extended Data Fig. 5c-d ). Next, we pharmacologically inhibited Cdc42 activity using either ML141 or CASIN 63,64 . Inhibition of Cdc42 resulted in extensive disruption of the interior actin network integrity as well as defects in mitotic fidelity ( Fig. 6e-g ). Additionally, live-cell imaging of WT cells expressing LifeAct-mEmerald and mRFP-wGBD revealed transient actin enrichment associated with the active Cdc42 endosome ( Fig. 6c, Extended Data Fig. 5e) , resembling actin cables generated by P-cadherin vesicles ( Fig. 5d-e ). In contrast, in KO cells actin association becomes strongly diminished in the few remaining active Cdc42 endosomes ( Extended Data Fig. 5f ). Taken together, these findings indicate that P-cadherin endosomes serve as platforms for Cdc42 activity and implicate Cdc42 as a key effector controlling the endosome-generated interior actin network. Since Cdc42 regulates actin via both Arp2/3 and formins 65 , we tested which pathway contributed to the polymerization of the endosome-generated interior actin network. As shown in Fig. 6h-j , inhibition of formins by SMIFH2 significantly depleted interior actin network and impaired mitotic fidelity in WT cells, whereas Arp2/3 inhibition with CK666 increased interior actin density, likely reflecting the competitive balance between the actin polymerization machineries 56 , but did not compromise mitosis ( Fig. 5a-c ). We also observed that CK666 treatment of P-cadherin KO cells also increased the interior actin network density compared to DMSO control. However, the CK666-induced increase in actin density still remains below WT level and mitotic fidelity is not fully restored ( Extended Data Fig. 5g-h ). Altogether, these results implicate formin as the primary actin polymerase assembling the interior actin network in MDCK, in contrast to the Arp2/3-based networks that generated the peri-spindle actin structures in mouse preimplantation embryos or centrosomal actin in non-epithelial cells 3,17,25 . P-cadherin mediates interior actin network assembly via Cdc42-MRCKβ-FMNL2 pathway To pinpoint the relevant Cdc42-dependent formin 66 and given the known off-target effect of SMIFH2 67 , we next made use of siRNA knockdown. This identified FMNL2, but not FMNL3, as being essential for both interior actin integrity and mitotic fidelity, as its KD phenocopied P-cadherin KD ( Fig. 6k-m ). Live-cell imaging showed that FMNL2-EGFP was localized at cell-cell contacts and on endosomal vesicles in WT cells and P-cadherin KO rescued by P-cadherin-tdTomato ( Extended Data Fig. 5j ), In dividing cells, endosomal FMNL2 can readily be seen localized to spindle poles ( Extended Data Fig. 5l ) and co-localized with actin cables ( Extended Data Fig. 5k ). In contrast, in P-cadherin KO cells, FMNL2 lost its endosomal localization, becoming largely diffuse in the cytosol ( Extended Data Fig. 5m ), suggesting that FMNL2 requires P-cadherin for endosomal targeting. FMNL2 is auto-inhibited via the DAD domain 68,69 . Indeed, the expression of a constitutively active FMNL2 mutant (ΔDAD) in P-cadherin KO cells restored endosomal localization as well as interior actin networks, along with partially rescuing mitotic fidelity. In contrast, P-cadherin KO expressing FMNL2-EGFP still exhibited significant interior actin depletion as well as mitotic defects ( Extended Data Fig. 5m-n, 7a-c ). Thus, FMNL2 emerges as the primary actin cable generator downstream of P-cadherin and Cdc42, whose activation and endosome localization in turn depend on P-cadherin. To further delineate signaling partners, we performed proximity biotinylation followed by proteomics based on TurboID fused to the cytoplasmic tails of either E- or P-cadherin 70 ( Extended Data Fig. 6 ). The TurboID constructs were expressed in the human cell line A431D, which is devoid of classical cadherins 71 , and in which P-cadherin expression affects interior actin similarly to MDCK cells ( Extended Data Fig. 6a ). Candidate proteins enriched in the P-cad-TurboID sample relative to the E-cad-TurboID sample were identified and classified based on STRING classification 72 . Importantly, P-cad-TurboID-enriched proteins included endosomal markers such as Rab5b and various mitotic spindle proteins ( Supplementary Table 1, Extended Data Fig. 6d-h ). Validation of top candidates was performed in MDCK cells by assessing mitotic fidelity and interior actin integrity following siRNA-mediated KD ( Fig. 6n ). We found that KD of MRCKβ (gene name: CDC42BPB), a Cdc42-dependent kinase with known association to FMNL2 73 , significantly impaired mitotic fidelity and depleted interior actin, phenocopying the KD of P-cadherin or FMNL2 ( Fig. 6k-n ). We next validated the subcellular localization of MRCKβ, observing that in WT cells its mScarlet3 fusion 74 localized prominently to cell-cell junctions as well as endosomal vesicles clustered near centrosomes in dividing cells (Fig. 7a-d, Extended Data Fig. 7e ). Likewise, MRCKβ is found to co-localize extensively at the endosomes with P-cadherin ( Fig. 7e-f ), FMNL2 ( Fig. 7g ), and Rab5 ( Fig. 7h ). In contrast, in P-cadherin KO cells MRCKβ localization at cell-cell junctions and endosomes was largely abrogated, becoming diffuse and cytosolic ( Fig. 7a-d ), indicating that its subcellular localization depends on P-cadherin. In turn, we found that SiRNA-mediated KD of MRCKβ in WT cells severely abrogate FMNL2 endosomal localization ( Extended Data Fig. 7d, f ), thus implicating its role in potentiating FMNL2 function at the endosomes. Corroborating this, pharmacological inhibition of MRCKβ using BDP9066 75 disrupted the interior actin networks while also impairing mitotic fidelity ( Fig. 7i-k ), albeit with the caveat that this inhibitor also affects MRCKα. Altogether, our results define a core molecular mechanism whereby endocytosed P-cadherin recruits an actin-generation machinery consisting of Cdc42, MRCKβ, and FMNL2 to Rab5-positive endosomes to assemble the interior actin network that organizes intracellular architecture and safeguards mitotic fidelity. Discussion Accurate chromosome segregation involves a tightly choreographed remodelling of the sub-cellular cytoskeletal architecture. Studies across diverse tissues and cell types have underscored the plasticity of the cell division machinery 19 . In most somatic cells, spindle assembly is thought to depend primarily on centrosomes, whereas in cells lacking centrosomes, such as oocytes or mouse pre-implantation embryos, both actin and microtubules collaborate in spindle organization 1,3,4,48 . Although microtubules have long been considered the canonical driver of centrosomal mitosis, several studies – especially those optimizing actin visualization – have reported actin involvement 17,45,47 , suggesting that the role of actin in centrosomal mitosis may be broader than previously assumed. Here, through the optimization of preservation and observation methods, we show that P-cadherin-dependent endosome-generated interior actin networks are essential for centrosomal mitosis in common epithelial cell types. Moreover, such actin network influences mitochondrial morphology through competition with the peri-mitochondrial actin machinery. Consequently, loss of P-cadherin affects not only mitotic fidelity but also gives rise to excess actin encapsulation of mitochondria, leading to encapsulation of mitochondria, their fragmentation, and defective distribution in cells. Mechanistically, our data reveal an endosome-based pathway that ensures proper chromosome segregation ( Fig. 7l, Extended Data Fig. 7h-i ). Rather than being degraded by Hakai after endocytosis, as established for E-cadherin 76 , internalized P-cadherin is trafficked in Rab5-positive endosomes, where it recruits Cdc42, MRCKβ, and FMNL2 to polymerize actin cables. As P-cadherin-containing endosomes are accumulated in the centrosomal vicinity, they polymerize actin into an interwoven network that extends throughout the cytoplasmic interior. This meshwork then serves as a physical constraint limiting chromosome scattering after NEBD, promoting timely capture by kinetochore microtubules, while the dense mass of actin meshes at each spindle pole may also contribute to the stabilization of the mitotic spindle position and orientation ( Extended Data Fig. 7h ). Accordingly, P-cadherin loss drastically reduces the density of the interior actin network, thereby reducing chromosome congression, prolonging chromosome scattering, and destabilizing the spindle positioning. This role is reminiscent of actin functions in oocyte meiosis 48 , where FMNL2 depletion disrupts spindle and endomembrane organization 77 . Because both E- and P-cadherin are expressed in oocytes 78 , it will be interesting to test whether the P-cadherin–dependent endosomal mechanism may also operate there. Conversely, the formin-dependent mechanism described here is distinct from the Arp2/3- and myosin-X-dependent actin remodelling and contraction involved in acentrosomal mitosis of mouse preimplantation embryos 3 . Collectively, these comparisons suggest that actin-dependent contributions to mitosis are more diverse than previously appreciated. While validated across several epithelial and carcinoma cell lines, the in vivo significance of the P-cadherin-based mechanism remains to be established. Although P-cadherin expression is pronounced in embryonic tissues 39 , its ablation results in phenotypes far less severe than the embryonic lethality associated with E-cadherin loss 38 . Whether and how interior actin networks are regulated in cells lacking P-cadherin, for example through alternative endosomal scaffolds or distinct regulatory mechanisms remain to be investigated. Our data also indicates that P-cadherin-dependent interior actin networks persist in interphase, balancing different organelle-associated actin pools. Moore et al. described dynamic cytoplasmic actin cables that undergo wave-like motion in HeLa cells 24,25 , which we did not observed in MDCK, consistent with cell-type specific differences in actin regulators. Nevertheless, in both cell types the integrity of the cytoplasmic actin networks were linked to mitochondrial morphology. Since mitochondrial fragmentation is associated with metabolic dysfunction 79 , future work should clarify whether actin monomer availability, shared nucleators, or signalling feedback underlies the reciprocal relationship between P-cadherin endosomes and mitochondria, and how it influences mitochondrial inheritance and function. This study also highlights a previously unrecognized non-junctional role of P-cadherin. While classical cadherins are best known for stabilizing cell–cell junctions and transmitting mechanical cues through cortical actin, we show that P-cadherin, through its cytoplasmic domain, also serves as a structural and signalling platform on Rab5-positive endosomes, driving the assembly of endosome-generated interior actin network. Endosomal signalling has been described for integrins and growth factor receptors 80,81 , and cadherin-independent signalling of junctional proteins such as α-catenin and β-catenin from Rab11 endosomes has been reported 82 . However, to the best of our knowledge, cadherin endosomal signalling has not been reported previously. E-cadherin and P-cadherin have a high degree of sequence similarity (68% identity for ectodomain, ~95% identity for cytoplasmic domain) and are co-expressed in a broad range of epithelial tissues 83 . The E- versus P-cadherin specialization may reflect unique mechanical constraints of epithelial mitosis, where cells must preserve junctional integrity and tissue topology while simultaneously segregating chromosomes. We propose that E-cadherin primarily organizes cortical and junctional actin, whereas P-cadherin, when internalized and stabilized, regulates internal actin to support the spindle and surrounding organelles. Consistent with this, the fate of endocytosed E-cadherin and P-cadherin appear to drastically diverge, with internalized P-cadherin preferentially stabilized in a cell-cycle dependent manner. Notably, endocytosis of P-cadherin is required for interior actin integrity ( Extended Data Fig. 4i-m ), suggesting that additional layers of regulation via junctional turnover or cadherin cleavage may also be operational. The unique molecular features of the P-cadherin cytoplasmic domain underlying its mitotic function remain to be identified. Finally, P-cadherin depletion affects multiple aspects of spindle organization, including spindle distortion and multipolarity, implying additional contributions to microtubule regulation. Likely, endosomal P-cadherin, in addition to regulating actin, also helps localize factors that regulate the microtubule-based machinery. These can be identified by further analysis of the differential proximity proteomics data. More broadly, the pleiotropic effects of P-cadherin on fundamental cellular processes such as mitosis and organelle homeostasis raise the possibility that its dysregulation contributes to aneuploidy or metabolic dysfunctions. P-cadherin is druggable 84 . Given the correlations between P-cadherin dysregulation and multiple cancers 6 , further investigation of these mechanisms may provide valuable insights into disease progression. Methods Mammalian cell culture MDCK II (Madin-Darby canine kidney II) used were originally obtained from Sigma-Aldrich (ECACC, Cat. No. 85011435) and gifted by Huang Cheng-Kuang (Mechanobiology Institute, Singapore). E-cadherin KO MDCK II is described previously and gifted by Rene-Marc Mége, Institut Jacques Monod, France. Cadherin-null A431D (human epidermoid carcinoma) cells is a gift from Ada Cavalcanti-Adam (Max Planck Institute for Medical Research). MCF10A (non-malignant human breast epithelial) cells were gifted by Chwee Teck Lim (Mechanobiology Institute, Singapore). HaCaT (spontaneously immortalized human keratinocyte), HK-2 (human proximal tubular epithelial) and Caco-2 (human colon adenocarcinoma) cells were gifted by Jennifer Young, Yusuke Toyama and Selwin Wu (Mechanobiology Institute, Singapore), respectively. A431 cells were purchased from the American Type Culture Collection (ATCC, Cat. No. CRL-1555). U2OS (human osteosarcoma) was gifted by Clare Waterman (National Institutes of Health). ARPE-19 (immortalized human retinal pigment epithelial) cells was gifted by Xinyi Su, Institute of Molecular and Cellular Biology A*STAR, Singapore. All cells were maintained in an incubator at 37°C in a humidified atmosphere containing 5% CO 2 and verified to be free from mycoplasma infection, and tested monthly. MDCK II, A431, A431D, U2OS and HaCaT cells were cultivated in Dulbecco’s modified Eagle’s medium (DMEM, Gibco™, Cat. No. 10566016) containing 10% (v/v) fetal bovine serum (FBS, Gibco™, Cat. No. 10437028), 1% (v/v) Pen/Strep (Gibco™, Cat. No. 15070063), 1% (v/v) sodium pyruvate (Gibco™, Cat. No. 11360070) and passaged using 0.25% (w/v) trypsin-EDTA solution (Gibco™, Cat. No. 25200072) every 2-3 days. HK-2 cells were grown in DMEM/F12 (Gibco™, Cat. No. 11330032) supplemented with 10% (v/v) FBS, 1% Pen/Strep, 1% (v/v) sodium pyruvate, and split with 0.05% trypsin-EDTA solution (Gibco™, Cat. No. 25300062) every 3-4 days. Caco-2 cells were maintained in DMEM containing 10% (v/v) FBS, 1% (v/v) Pen/Strep, 1% (v/v) sodium pyruvate and 1% (v/v) MEM non-essential amino acids (MEM NEAA, Thermo Fisher, Cat. No. 11140050). Cells were passaged using 0.05% (w/v) trypsin-EDTA solution every 3-4 days. MCF10A cells were maintained in DMEM/F12 supplemented with 5% (v/v) horse serum (Invitrogen, Cat. No. 16050-122), 20 ng/ml EGF (Thermo Fisher, Cat. No. AF-100-15), 0.5 mg/ml hydrocortisone (Sigma-Aldrich, Cat. No. H0888), 100 ng/ml cholera toxin (Sigma-Aldrich, Cat. No. C8052), 10 µg/ml insulin (Sigma-Aldrich, Cat. No. I0516), 1% (v/v) Pen/Strep and passaged using 0.05% trypsin-EDTA solution every 3-4 days. ARPE-19 cells were cultured in minimum essential medium, alpha modification (MEM-α, Sigma-Aldrich, Cat. No. M4526) containing 10% (v/v) FBS, 1% (v/v) Pen/Strep, 1% GlutaMAX (Thermo Fisher, Cat. No. 35050079), 1% (v/v) MEM NEAA, 1% (v/v) N1 medium supplement (Sigma-Aldrich, Cat. No. N6530), 10 mM Nicotinamide (Sigma-Aldrich, Cat. No. N5535), 0.25 mg/ml Taurine (Sigma-Aldrich, Cat. No. T0625), 20 ng/ml hydrocortisone and 0.013 ng/ml 3,3′,5-Triiodo-L-thyronine sodium salt (Sigma-Aldrich, Cat. No. T5516). Cells were split using TrypLE Express (Thermo Fisher, Cat. No. 12604013) every 4-5 days. CRISPR/Cas9-mediated knockout (KO) of P-cadherin P-cadherin KO MDCK cells were generated by CRISPR/Cas9-mediated genomic editing using dual guide RNAs to induce exon skipping, including sgRNA-A (5'-ACCCGAAAAT CGTGATCGTG-3') and sgRNA-B (5'-CGATAGAGGC GCTAAAGAGT-3'), targeting exon 2 and downstream of exon 4 of dog CDH3 gene, respectively. The oligo sequences were designed using the CHOPCHOP web tool (https://chopchop.cbu.uib.no). Oligonucleotides were synthesized by Integrated DNA Technologies. Each guide sgRNA sequence was separately introduced into the pSpCas9(BB)-2A-Puro (PX459) V2.0 plasmid backbone, which was a gift from Feng Zhang (Addgene # 62988). The vector contains Cas9 gene with BbsI cloning site under the control of U6 promoter. The hU6-F primer (5'-GAGGGCCTATTTCCCATGATT-3') was used for sequencing to confirm the presence of sgRNA oligos after cloning (Axil Scientific Sequencing). siRNA-mediated knockdown For negative siRNA control, a non-targeting siRNA (Invitrogen™, Cat. No. 4390846) was used, P-cadherin (siRNA #10694 and #S2774), RAB5B (siRNA #s11683 and #s11681), KIF15 (siRNA #s32546 and #s32548), ECT2 (siRNA #146755 and #S4444), profilin-2 (PFN2, siRNA #S10380), FMNL2 (siRNA #128873 and #128874), FMNL3 (siRNA #131854 and #HSS132081), and MRCKβ (CDC42BPB, siRNA #748 and #S18402) were purchased from Thermo Fisher. Stock solutions of each siRNA (25 µM) were generated by dissolving siRNA powder in RNase-free water and stored at -20°C until usage. Two siRNA oligonucleotides that target distinct regions of P-cadherin, RAB5B, KIF15, ECT2, FMNL2, FMNL3, or MRCKβ transcripts were combined to prepare siRNA mixture using for electroporation. Sequences of the siRNA oligonucleotides were verified to be capable of binding fully complementary regions in mRNA transcripts of both human and dog targets. For knock-down experiments, cells were maintained in growth media without antibiotics overnight to reach 60-70% confluency. Then, 5x10 5 -7.5x10 5 cells were suspended and electroporated with siRNAs (30 pmol for MDCK II or 750 pmol for other cells) using Neon™ Transfection System (100 μl Kit, Thermo Fisher, Cat. No. MPK10096) with pulse rate of 1650V, 20 ms pulse width and 1 pulse. Transfected cells were transferred into 27 mm glass bottom dish containing warm medium and grown for 48 h before performing immunofluorescence staining or Western blotting. E- and P-cadherin constructs P-cadherin-tdTomato, P-cadherin chimera-GFP (ecto-P-cad::cyto-E-cad) and E-cadherin chimera-GFP (ecto-E-cad::cyto-P-cad) were generated as follows. UniProtKB and NCBI databases were used to annotate the DNA sequence and plan the precise swapping of domains in SnapGene software (San Diego, CA). Plasmids containing cDNA of E-cadherin GFP (Addgene plasmid #28009) and P-cadherin (Addgene plasmid #47502) were obtained from Addgene repository. The pcDNA3.1 vector backbone from E-cadherin GFP plasmid was used as the backbone for further cloning. Primers for amplification of DNA fragments via Polymerase chain reaction (PCR) contained an identical 5’ end to an adjacent segment and a 3’ end that anneals to the target sequence were ordered from Integrated DNA technology (IDT). The following primers were used for construct assembly: P-cad tdTomato insert, forward 5′-GGTGGCGGGG AGGACGACTC CGGACTCAGA GTGAGCAAGG GCGAGGAGGTC-3′ and reverse 5′-TGACACTATA GAATAGGGCC CTCTAGACTA CTTGTACAGC TCGTCCATGCCG-3′; P-cad tdTomato vector, forward 5′-CTGTACGGCA TGGACGAGCT GTACAAGTAG TCTAGAGGGC CCTATTCTAT AGTGTC-3′ and reverse 5′-CTCCTCGCCC TTGCTCACTC TGAGTCCGGA GTCGTCCTCC CCGCC-3′; P-cad extracellular domain insert, forward 5′-CCTGCCCTCG CTCGGCGTCC CCGGCCAGCC ATGGGGCTCC CTCGTGGAC-3′ and reverse 5′-AGCAAGAATT CCTCCAAGAA TCCCCAGAAT ACCTCCCTTC CAGGGTCCAG-3′; P-cad extracellular domain vector, forward 5′-GAAACCTGCC CTGGACCCTG GAAGGGAGGT ATTCTGGGGA TTCTTGGAGG-3′ and reverse 5′-AGACGCGAGA GGTCCACGAG GGAGCCCCAT GGCTGGCCGG GGACGCC-3′; E-cad extracellular domain insert, forward 5′-CCTGTCGAAG CAGGATTGCA AATTCCTGCC TTCATCCTCC CTGTGCTGGGG-3′ and reverse 5′-CACCCCGGTG AACAGCTCCT CCCCCTTGCT GTCGTCCTCC CCGCCAC-3′; E-cad extracellular domain vector, forward 5′-GCAGACATGT ACGGTGGCGG GGAGGACGAC AGCAAGGGGG AGGAGCTG-3′ and reverse 5′-CAGGACAGCCC CCAGCACAGG GAGGATGAAG GCAGGAATTT GCAATCC-3′. DNA electrophoresis was performed for PCR products in order to separate them by length in 1% agarose gel in TAE buffer (2 M Tris-HCl, 0.6% acetic acid, 50 mM EDTA pH 8.0) to ensure the accuracy of previous steps. After purification of PCR products using Gel and PCR clean-up kit (Macherey-Nagel), fragments were combined in Gibson assembly reaction using the Gibson Cloning Master Mix (New England Biolabs), which included three enzymes: T5 Exonuclease, Phusion DNA Polymerase, and Taq DNA Ligase. The Gibson reaction was then transformed into XL-1 Blue competent bacteria and the correct clones was selected by PCR. E-Cad-TurboID, P-Cad-TurboID, and TurboID sequences were cloned into the pLL3.7 lentiviral vector backbone (Addgene plasmid #11795). Expression vectors mRFP-wGBD plasmid was obtained from Addgene (plasmid # 26733 and 26734, gifted by William Bement). mEmerald-Lifeact-7 (Addgene plasmid # 54148) and Histone H2B-mCherry (Addgene #20972) were gifts from Michael Davidson (The Florida State University). mCherry-Rab5b was obtained from Addgene (plasmid # 49201, deposited by Gia Voeltz). Genetically-encoded mitochondria actin probe (AC-Mito), pSIN CMV AC-GFP-Fis1, was as described previously by Uri Manor (UC San Diego) 52 . FMNL2-Delta-DAD-EGFP and FMNL2-GFP plasmids were developed in the laboratory of Klemes Rottner 69 and kindly provided by Maddy Parsons (King’s College London). MRCK-β-mScarlet3 was synthesized by GentleGen Inc. (China). iRFP-FRB-Rab5 was obtained from Addgene (plasmid # 51612, deposited by Tamas Balla). LifeAct-mScarlet-I was obtained from Addgene (plasmid # 85056, deposited by Dorus Gadella). Expression vector for human-P-cadherin-GFPspark, referred to as P-cadherin-GFP for short, was purchased from Sino Biological (Cat. No. HG10305-ACG). Transfection was performed using 5x10 5 suspended cells per reaction and 0.5-2 µg plasmids using Neon™ Transfection System (100 μl, Thermo Fisher). Transfected cells were transferred into 12 mm or 27 mm glass bottom dish containing warm medium and allowed to grow for 24-48 h before performing immunofluorescence staining or live-cell imaging. When necessary, NucBlue™ (Hoechst 33342, Thermo Fisher) reagent was used as nuclear counterstain prior to live-cell imaging. Stable cell line generation Generation of P-cadherin KO MDCK cells MDCK cells (5x10 5 cells per transfection) was electroporated with 5 µg of either plasmids containing sgRNA-A (5'-ACCCGAAAATCGTGATCGTG-3') or sgRNA-B (5'-CGATAGAG GCGCTAAAGAGT-3'), by utilizing a Neon™ Transfection System as described above. The transfected cells were grown for 48 h before selecting using 2 µg/mL puromycin (Sigma-Aldrich, Cat. No. P8833). After puromycin selection, single cell clones were sorted into 96-well plates using the benchtop SH800S cell sorter (Sony). Three single KO clones were successfully verified by PCR, immunofluorescence staining and Western blotting. Generation of stable cell lines expressing fluorescently tagged proteins Transfection of plasmids with fluorescently tagged proteins was performed as described above. After maintaining transfected cells for 24-48 h, the stable cell lines were selected under antibiotics (Geneticin (G418) at 600 µg/ml or Hygromycin B at 200 µg/ml), according to the antibiotic resistance genes presented in plasmids for at least 2 weeks. Medium to high expression stable cell lines were sorted (SH800S cell sorter, Sony) to generate monoclonal populations, which were subsequently used for further experiments. Preparation of biomimetic cadherin–F c chimera substrate Coverglasses (no. 1.5, 18 mm diameter) were washed in absolute ethanol overnight and rinsed three times with sterile water and PBS, respectively before exposure to UV in sterile laminar flow hood for 15 min. Sterile coverglasses were silanized by incubation with 3-glycidoxy-propyl-dimethoxymethylsilane (Sigma-Aldrich, Cat. No. 539252) (0.045% in 100% ethanol) for 1 h on a shaker at room temperature and then heated at 110°C for 1 h. Silanized substrates were rinsed with 70% ethanol and distilled water before drying with nitrogen gas. Subsequently, the substrates were then incubated with AffiniPure ® goat anti-human IgG, Fcγ fragment specific (Jackson ImmunoResearch Inc., Cat. No. 109-005-008) at 1 µg/cm 2 in 0.1 M borate buffer, pH 8.0 and incubated at 4°C overnight in a humidity chamber. After rinsing with PBS, the substrates were neutralized by NaHCO 3 (100 mM, pH 8.3) and 2-(2-aminoethoxy)-ethanol (Sigma-Aldrich, Cat. No. A54059) for 1 h. The substrates were washed twice with PBS and then incubated with recombinant human P-cadherin F c (R&D system, Cat. No. 861-PC) or E-cadherin F c (R&D system, Cat. No. 648-EC) chimera protein in PBS containing Ca 2+ and Mg 2+ at 1 µg/cm 2 for 2 h. After rinsing, the substrates were blocked with 0.2 % pluronic acid (Sigma-Aldrich, Cat. No. P2443) in PBS for 20 min at room temperature. The planarized cadherin substrates were washed twice with PBS before cell seeding in serum free medium. Cell adhesion on planarized cadherin–F c chimera substrat e To assay cadherin engagement, MDCK II WT, E-cadherin KO, or P-cadherin KO cells were dissociated with 0.05% Trypsin-EDTA solution and pelleted by centrifugation. Cell pellets were washed twice with serum-free DMEM medium. Single cell suspensions were prepared from the pellets in serum-free DMEM medium and then pre-incubated in 37° C tissue culture incubator for 30 min prior to seeding on the coated substrates. The cells on substrates were then fixed with fresh 4% (v/v) paraformaldehyde in PBS+ to preserve cell adhesion. Identification of P-cadherin-specific interactome by proximity biotinylation Production of lentiviral particles using pLL3.7 vector backbone Lentiviral particles were produced by transient transfection of HEK293 FT cells, as described previously. Briefly, cells were seeded at a density of 1-1.5x10 6 cells/well and cultivated in DMEM supplemented with 10% FBS, 1% Pen Strep, 1% L-Glutamine, 1% sodium pyruvate, 1% sodium bicarbonate). At 80% confluency, cells were replated on poly-L-lysine coated 6-well plates. Plasmids (E-Cad-TurboID, P-Cad-TurboID, and TurboID cloned into the pLL3.7 vector backbone) were co-transfected with envelop plasmid VSV-G (Invitrogen), and packaging plasmids pLP1 and pLP2 (Invitrogen) using PEI (Polyethylenimine, high molecular weight, water free, Sigma-Aldrich, Cat. No. 408727). On the next day, medium was switched with the growing medium (DMEM+10% FBS 1.5 ml/well) without any washing and cells were grown for another 30 h. After 30 h, supernatant was collected, and viral particles in the supernatant were concentrated at 1:100 by ultracentrifugation for 90 min at 25,000 rpm (SW28 rotor, Beckman Coulter) and resuspended in PBS. Titers of viral particles ranged between 10 5 and 10 6 multiplicity of infection. Generation of stable cell lines expressing Turbo ID, E-Cad fused to TurboID, and P-Cad fused to TurboID Cadherin-null A431D (human epidermoid carcinoma) cells were seeded at 3x10 5 cells/ml into 6-well plates. Respective lentiviral particles and 8 μg/ml polybrene (Sigma-Aldrich) were added to the culture and centrifuged for 90 min at 30°C. Supernatant was removed immediately after infection and replaced with DMEM medium containing 10% FBS in all experiments. The efficiency of transduction was approximately 70-80% as determined by immunostaining. The respective protein expression (Turbo ID, E-Cad-TurboID fusion gene, and P-Cad-TurboID fusion gene) was confirmed by Western blotting with an anti-FLAG antibody. For further validation, these cell lines were cultured in 24-well plates and subjected to biotinylation reaction (incubation with DMEM medium supplemented with 330 μM Biotin (Sigma-Aldrich, Cat. No. B4639)) for a duration of 10 minutes. Next, biotinylated cells were fixed using ice-cold methanol and then immunostained for the respective transgene and streptavidin signal. Sample preparation for Mass spectrometry Monolayers of the three cell lines described above were grown in a 10 cm dish to more than 90% confluency, incubated with 330 µM biotin for 10 min, washed with PBS buffer thrice and then lysed with RIPA buffer. As a negative control, all three cell lines were also processed identically without Biotin. Cell lysates were quantified for the protein concentration using BCA assay (Thermo Fisher, Cat. No. 23225). An equal amount of cell lysate (700 µg) from each of the six samples was incubated with streptavidin magnetic beads (Thermo Fisher, Cat. No. 88817) overnight at 4°C and then eluted. The eluate obtained from all six samples was loaded on SDS-PAGE gel and subsequently imaged with Western blotting using streptavidin-HRP antibody (Thermo Fisher, Cat. No. N100). As expected, the samples incubated with biotin were stained intensely with streptavidin-HRP antibody whereas the samples incubated without biotin were stained weakly. This experiment was conducted in triplicate and the nine samples prepared for the MS analysis were generated in an identical fashion and sent to the mass spectrometry core facility at Technion – Israel Institute of Technology. Proteolysis The proteins were brought to 8.5 M Urea, 100 mM ammonium bicarbonate (ABC), reduced with 2.8 mM DTT (60ºC for 30 min), modified with 8.8 mM iodoacetamide in 100 mM ABC in the dark at room temperature for 30 min, and digested in 1.5 M Urea, 25 mM ABC with modified trypsin (Promega) at a 1:50 enzyme-to-substrate ratio for overnight at 37°C. Mass spectrometry analysis The tryptic peptides were desalted using C18 tips (Empore™ C18 Solid Phase Extraction Disks (SPE): 2215), dried and re-suspended in 0.1% formic acid. The peptides were further cleaned from detergents using strong cation exchange tips (Empore™ cation disk 2251). The peptides were resolved by reverse-phase chromatography on 0.075 X 180-mm fused silica capillaries (J&W) packed with Reprosil reversed phase material (Dr. Maisch GmbH, Germany). The peptides were eluted with linear 60 min gradient of 5% to 28% and 15 min gradient of 28% to 95% and 15 min at 95% acetonitrile with 0.1% formic acid in water at flow rates of 0.15 μl/min. Mass spectrometry was performed by Q Exactive™ Plus Hybrid Quadrupole-Orbitrap™ Mass Spectrometer (Thermo Fisher) in a positive mode using repetitively full MS scan followed by high collision dissociation (HCD) of the 10 most dominant ions selected from the first MS scan. The mass spectrometry data was analyzed using the MaxQuant software 1.5.2.8 for peak picking and identification using the Andromeda search engine, searching against the Homo sapiens proteome from the UniProtKB database with mass tolerance of 6 ppm for the precursor masses and the fragment ions. Oxidation on methionine, biotinylation and protein N-terminus acetylation were accepted as variable modifications and carbamidomethyl on cysteine was accepted as static modifications. Minimal peptide length was set to seven amino acids and a maximum of two mis-cleavages was allowed. The data was quantified by label free analysis using the same software. Peptide- and protein-level false discovery rates (FDRs) were filtered to 1% using the target-decoy strategy. Protein tables were filtered to eliminate the identifications from the reverse database, and common contaminants and single peptide identifications. Statistical analysis of the identification and quantization results was done using Perseus 1.6.7.0 software developed by Mathias Mann's group. Mass Spectrometric analysis was performed at the Smoler Proteomics Center at the Technion-Israel Institute of Technology, Israel. Protein filtering and subcellular localization enrichment analysis A total of 3,766 proteins were identified from three independent biological replicates extracted from cells expressing TurboID, E-cadherin–TurboID, or P-cadherin–TurboID. Potential contaminants, ribosomal proteins, and proteins identified solely by site or by a single unique peptide were excluded from further analysis. Notably, CAV2, CDC42BPB, DDR1, FCHO2, PFN2, RAP1B, and VPS26B were retained despite being identified by a single peptide, owing to their specific enrichment, biological relevance, and unique association with the P-cadherin–TurboID dataset. Proteins exhibiting a log₂ fold-change of less than 1 between TurboID and either P-cadherin–TurboID or E-cadherin–TurboID samples were additionally excluded from the final protein list. Subcellular localization enrichment analysis was evaluated using STRING version 12.0 (https://string-db.org/). Western blotting Cells cultured in a 6-well plate were extracted by scraping in 200 µL RIPA buffer (Thermo Fisher, Cat. No. 89900) containing 1X Protease Inhibitor Cocktail (Roche, Cat. No. 11836170001) followed by incubation on ice for 30 min. After high-speed centrifugation, protein concentrations of supernatants were determined by Bradford protein assay (BIO-RAD, Cat. No. 5000201). Equal amounts of proteins from each sample were resolved using SDS-PAGE procedure with 4–15% Mini-PROTEAN ® TGX™ Precast Protein Gels (BIO-RAD, Cat. No. 4561084). Thereafter, proteins from gel were electroblotted using Trans-Blot Turbo System (Bio-Rad Laboratories). After washing with TBS and 15 min blocking in EveryBlot Blocking Buffer (Bio-Rad Laboratories), membranes were sequentially incubated with primary antibody (1:1000) in EveryBlot Blocking Buffer for overnight at 4ºC and secondary antibody conjugated to horseradish peroxidase (HRP) (1:2000, 1h at room temperature) after washing. After each incubation, membranes were washed 3 times in TBS containing 0.2% (v/v) Tween 20 (TBST), and final rinsed with TBS before developing with Chemiluminescence and detecting protein bands using iBright Imaging Systems (Thermo Fisher). Cell cycle synchronization Synchronization at G1/S-border Thymidine (Santa Cruz Biotechnology, Cat. No. sc-296542), was used to arrest cells at G1/S-border at a final concentration of 2 mM by incubating with cells for 24 h at 37°C. Synchronization at G2/M-border Cells were arrested at G2/M-border by using a Cdk1 inhibitor, RO-3306, (Sigma-Aldrich, Cat. No. SML0569) at a final concentration of 10 μM for 20 h at 37°C. To release the cells from G2/M, RO-3306 was removed from the cells by 5 washes in warm media. Synchronization at mitosis (prometaphase or metaphase) STLC ((+)-S-trityl-L-cysteine) (Tocris Bioscience, Cat. No. 2191), which is an Eg5 (Kinesin-5) inhibitor, was used to arrest cells at metaphase. Cells were initially synchronized at G2/M-border by treating with 10 μM RO-3306 for 20 h at 37°C. Upon washout of RO-3306, STLC (50 μM) was added after RO-3306 removal for 1 h to stall at prometaphase or 1.5 h for metaphase. Endocytosis inhibition by dynasore treatment For endocytosis inhibition, MDCK II cells were treated with 80 μM Dynasore (Selleck Chemicals, Cat. No. S8047) in growth medium for 24 h at 37°C. After washing with PBS, cells were fixed and then performed immunofluorescence staining. To examine mitotic rescue upon Dynasore washout, MDCK II cells were initially synchronized at G2/M by treating with 10 μM RO-3306 as described above. Dynasore (80 μM) was added to the cells 1 h before RO-3306 release by 5 washes with warm medium. Cells were allowed to enter prometaphase for 1.5 h in the presence of Dynasore. After dynasore washout, the cells were stalled at prometaphase using 50 μM STLC and endocytosis was allowed to resume for 30 min before image acquisition. Pharmacological inhibition To inhibit Arp2/3 complex-driven actin assembly, cells were treated by adding a 100 μM CK-666 (MedChemExpress, Cat. No. HY-16926) prepared in warm medium and incubated for 1 h at 37°C before fixation. To inhibit formin-driven actin assembly, we treated the cells by adding a 25 μM SMIFH2 (Sigma-Aldrich, Cat. No. 344092) prepared in warm medium and incubated for 1 h at 37°C. Cells were fixed and subsequently performed immunostaining. To inhibit Cdc42 activity, we treated the cells by adding 25 μM ML141 (MedChemExpress, Cat. No.: HY-12755) or 10 μM CASIN (MedChemExpress, Cat. No.: HY-12874) prepared in warm medium and incubated for 2 h at 37°C. Cells were fixed and subsequently performed immunostaining. To inhibit MRCKb activity, confluent monolayer of MDCK II WT was treated by 5 μM BDP9066 (MedChemExpress, Cat. No.: HY-111424) and incubated for 2 h at 37°C. Cells were fixed and subsequently performed immunostaining. Immunofluorescence staining Cells grown on 12 mm or 27 mm glass-bottom dish (IWAKI) were fixed with fresh 4% (v/v) paraformaldehyde (PFA, Electron Microscopy Sciences) in DPBS containing calcium and magnesium (PBS+, Gibco™, Cat. No. 14040117) to preserve cell adhesion for 15 min at room temperature. After 3 washes with PBS, cells were sequentially permeabilized with 0.1% (v/v) Triton X-100 in PBS for 10 min and blocked non-specific binding with 2.5% (w/v) BSA in PBS for 30 min. After rinsing, cells were incubated with primary antibody (1:100) in 1% (w/v) BSA in PBS for overnight at 4°C on shaker and then incubated with respective secondary antibody conjugated to Alexa Fluor 488, 568 or 647 at a dilution of 1:200 (Invitrogen) in 1% (w/v) BSA in PBS for 1 h at room temperature. When necessary, DAPI (1:400, Thermo Fisher) and Phalloidin–Atto 647N (1:120, Sigma-Aldrich, Cat. No. 65906) were used to stain nuclei and F-actin, respectively. Samples were rinsed and performed image acquisition in PBS. Optimized fixation for preserving astral actin networks For interior actin structure visualization, great care must be taken to minimize mechanical disturbance during medium exchange. Culture medium was carefully aspirated from cells by leaving a small volume (~100 µl) sufficient to cover the cells. Cells were then fixed with fresh pre-warmed 4% (v/v) PFA in PBS+ by continuously dropping fixative solution slowly along the walls of the dish to avoid direct contact with the cells. Cells were incubated for 15 min at 37°C to ensure structural preservation. After three washes with PBS, cells were permeabilized with 0.1% (v/v) Triton X-100 in PBS for 7-8 min. Subsequent antibody staining steps were done as described above. Actin images were taken by confocal microscope on the same day of staining. Antibodies Rabbit polyclonal anti-Rab5 (ab218624), anti-clathrin (ab21679), anti-α-tubulin (ab18251), anti-γ-tubulin (ab179503), anti-α-catenin (ab51032), anti-β-catenin (ab2365), mouse monoclonal anti-α-tubulin (ab7290), mouse polyclonal anti-DNMBP (TUBA, ab88534) and rabbit polyclonal anti-Intersectin-2 (ab237509) primary antibodies were purchased from Abcam. Rabbit polyclonal anti-FMNL2 (PA5-52148), anti-FMNL3 (PA5-23317) and mouse monoclonal anti-CDC42BPB (MRCKβ, H00009578-M03) primary antibodies were purchased from Thermo Fisher. Rabbit polyclonal anti-Rab11 (71-5300), anti-Rab35 (PA5-31674), anti-ECT2 (PA5-65769) and mouse monoclonal anti-P-cadherin (32-4000) primary antibodies were obtained from Invitrogen. Mouse monoclonal antibody against P-cadherin extracellular domain (sc-74545) and GAPDH (sc-365062) were purchased from Santa Cruz. Rat monoclonal anti-E-cadherin (MAB726) antibody was purchased from EMD Millipore. Mouse monoclonal anti-p120 catenin (610133) antibody was obtained from BD Biosciences. Rabbit polyclonal antibody to Intersectin-2 was purchased from Antibodies.com. Mouse monoclonal anti-β-actin (MCA5775GA) antibody was obtained from Bio-Rad. Mitochondria, microtubule, and DNA labelling for live-cell imaging Mitochondria labelling MitoTracker™ Orange CMTMRos (Thermo Fisher, Cat. No. M7512) and MitoTracker™ Green FM (Thermo Fisher, Cat. No. M7514) were prepared in warm cell culture medium at a final concentration of 100 nM, and then added to cells for 15 min at 37°C. The cells were washed twice with warm medium before subsequently incubating with other probes or imaging in FluoroBrite medium (Thermo Fisher). For mitochondrial membrane potential imaging, the indicator Image-iT™ TMRM (Thermo Fisher, Cat. No. I3436), was added to cells at a concentration of 100 nM and incubated for 30 min at 37°C prior to imaging. For the detection of mitochondrial superoxide in cells, MitoSOX™Red Mitochondrial Superoxide Indicator (Thermo Fisher, Cat. No. M36007) was incubated with live cells at a concentration of 5 µM at 37°C prior to imaging. Microtubule labelling SiR-Tubulin (Cytoskeleton Inc., Cat. No. CY-SC002) was freshly prepared at a final concentration of 1 µM in warm FluoroBrite (Thermo Fisher) medium. The probes were added to the cells followed by incubating for 1-3 h at 37°C. DNA labelling NucBlue™ (Hoechst 33342) reagent was dropped into the cells according to the manufacturer's instructions and incubated for at least 15 min for nuclear visualization before performing live cell-imaging without washing step. Fluorescence microscopy Fixed cells were imaged with a Nikon CFI Plan Apochromat ×100/1.45 NA oil objective using a spinning disk confocal unit (CSU-W1, Yokogawa) on a Nikon Eclipse Ti-E inverted microscope (Nikon Instruments), equipped with 405 nm (100 mW), 488 nm (150 mW), 561 nm (100 mW) and 642 nm (100 mW) diode lasers, and corresponding dichroic (Chroma Technology) on a Prime 95B sCMOS camera (1200x1200 pixels, Teledyne Photometrics), with acquisition control by MetaMorph Version 7.10.1.161 software. The 405 nm, 488 nm, 561 nm, and 642 nm laser powers were set at ~40% with exposure times of ~500 ms. Super-resolution imaging was performed by activating the Live-SR (Roper Scientific) module with pixel resolution at 100X magnification equal to 0.06587 µm/pixel. Z-stacks were taken using a Piezo stage (Plano® Z, Physik Instrument) in which the step size was 0.2 or 0.5 µm for a total thickness of ~10-15 µm. For time-lapse or live-cell imaging, the on-stage incubator (LCI CU-501 Temperature Controller) was stabilized to 37°C with 5% CO2 (LCI FC-5N CO2 mixer) in humidified chamber before image acquisition. The 405 nm, 488 nm, 561 nm, and 642 nm laser powers for live-cell imaging were set at 6-8%, 10%, 9.5% and 13% with exposure times of 400 ms, 400 ms, 400 ms and 800 ms, respectively. Image processing and analyses Confocal images and movies were processed and analyzed using Imaris (Andor Inc.), ImageJ, MATLAB software, or custom-written python code for astral density quantification. All fluorescence images shown in Figures are from a single z-slice unless stated otherwise. Mitotic defect quantification Quantifications of mitotic abnormality phenotypes were performed from metaphase cells imaged under confocal microscope after fixation and staining for α-tubulin and DAPI to visualize mitotic spindle morphology and chromosomes, respectively. Metaphase cells displayed at least one uncongressed/unaligned chromosomes were counted as chromosome misalignment mitotic cells. Metaphase cells in which the spindle axis is bended instead of straight were counted as spindle distortion mitotic cells. Spindle misorientation was counted when spindle poles were not oriented parallel to the cell-substrate adhesion plane. Multipolar spindles were counted when more than two spindle poles are present. Colocalization analysis The JACoP ImageJ plugin was used to calculate the Pearson's correlation coefficient (PCC, r) to analyze colocalization. Single z slice images were background subtracted and denoised (1 pixel) in ImageJ. Individual cells were cropped and specified cytoplasm as the region to use for colocalization analysis and applied threshold to create a mask for each channel before running JACoP. Mitochondria morphological quantification To quantify mitochondria morphology, the raw images of mitochondria (MitoTracker channel) and nuclei (DAPI channel) were denoised (1 pixel) and performed Gaussian filter with sigma radius of 1.5. Single slice of nucleus channel was applied thresholding to create binary mask. Then find maxima algorithm with the segmented particles option (prominence >100) was done to generate individual cell mask. Mitochondria diameter, length, area and number of each cell were quantified using ImageJ. Mitochondria area coverage quantification To quantify mitochondria area coverage, maximum projection of MitoTracker (mitochondria) and DAPI (nuclei) channels were denoised (1 pixel) and subjected to Gaussian filtering (sigma radius = 1), followed by thresholding to generate mitochondria and individual cell masks, respectively. The total mitochondria area per cell was measured using the Analyze particles function in ImageJ. Individual cell area was measured to calculate the percentage of cell area occupied by mitochondria. To quantify mitochondria area coverage within the spindle region, maximum projection of a-tubulin (mitotic spindle) channel in metaphase cells was denoised (1 pixel) and Gaussian filtered (sigma radius = 1), followed by thresholding to generate a spindle mask. The mitochondria area within the spindle region was measured and expressed as the percentage of spindle area occupied by mitochondria. Mitochondria dynamics quantification To quantify mitochondrial dynamics, raw images from the MitoTracker channel acquired by live-imaging (2 s per frame) were cropped to isolate individual cells, denoised (1 pixel), Gaussian filtered (sigma radius = 1) and followed by background subtraction. Mitochondria were automatically thresholded to generate binary masks and tracked using the TrackMate plugin in ImageJ. Mean mitochondrial track speed (mm/s) per cell was calculated for comparison. MitoSOX and TMRM intensity quantitation Images of MitoSOX or TMRM were denoised (1 pixel) and Gaussian filtered (sigma radius = 1), followed by background subtraction. MitoTracker images were used to generate mitochondria masks, which were applied via the ROI Manager in ImageJ. Mitochondria MitoSOX or TMRM intensity was measured per cells and used for comparison. Actin cable length quantification Actin cable length within the clustered chromosome region was measured manually using the line tool in ImageJ. The DAPI channel was used to define the clustered chromosome area. Chromosome congression analysis To quantify chromosome dynamics, individual mitotic cells were cropped and DAPI channel from spinning disc confocal images were used. Gaussian blurring with 1 pixel radius was used for noise reduction, followed by background subtraction with 50 pixel radius in ImageJ. Otsu thresholding was then performed with threshold calculated for each frame to create binary mask. Chromosome cluster area from each frame was extracted in FIJI. Spindle pole migration quantification To quantify spindle pole migration, individual mitotic cells were cropped and Sir-tubulin channel images were used. Gaussian blurring with 0.5-1 pixel radius was used for noise reduction in ImageJ. MTrackJ plugin was used to manually locate each spindle pole in every frame of time-lapse image series and recorded the distance between them frame by frame over time. Endo-MRCK b density quantification Single z-slices of MRCKb images were denoised (1 pixel) and smoothed before detecting endosomal-associated MRCKb puncta in the cytoplasm using the Find maxima with the point selection option in ImageJ. DAPI channel images were used to generate nuclear and cell masks to measure nuclear and total cell areas. Cytoplasmic area was calculated as cell area minus nucleus area and used to determine endosomal MRCKb density. Particle image velocimetry (PIV) analysis To analyze motion of cytoplasmic actin, chromosomes and mitochondria, PIV analysis was performed using a MATLAB toolbox, MatPIV 1.6.1 (http://urn.nb.no/URN:NBN:no-27806). For each movie, single cell was cropped from the raw data and background region was cleared so that only region of interest was analyzed. Single-pass PIV with window size of 32x32 pixels and 50% overlapping was applied. For each frame (interval), average speed (µm/min) was computed by averaging all the local instantaneous speeds. Normalization by actin cortex intensity To facilitate comparison of actin density, the F-actin intensity in confocal image is normalized by the average cortical actin intensity of each cell of interest. To determine the average cortical actin intensity, the cell boundary was interactively traced in custom-written python code, guided by the super-resolution confocal slice. A margin of ±10 pixels (.65 µm) was defined around the cell boundary to define the cortical zone. A histogram of pixel intensity within the cortical zone is calculated, and the Otsu threshold was calculated to define high-density cortical region. Average intensity of thresholded pixels were then calculated and used to normalize the F-actin intensity in each cell. Interior actin density quantification To enable comparison of interior actin network density, custom-written python code was used to generate a standardized map for ease of registration between cells. From the microtubule channel of a multi-color super-resolution confocal stacks, the position of each centrosome of a metaphase cell is interactively defined. The spindle axis is then calculated as the vector linking the 2 metaphase centrosomes. Polar grids were then defined with 60 azimuthal bins, and 10 radial bins (6˚ per azimuthal bin and 10 pixel or .65 µm per radial bin). Polar grids within ± 90˚ relative to the spindle axis are considered to encompass the interior actin. Subsequently, from the cortex-normalized F-actin density described above, a binary mask of the cortex region is applied to reject pixels outside of the cell. The density map is then calculated as the average of all pixels within each bin. Statistical analysis Statistical analyses were performed in GraphPad Prism 10.0 software using two-tailed unpaired Student’s t-tests and one-way ANOVA for multiple comparison. All data were shown as means ± SD, unless stated otherwise. A value of p < 0.05 was considered statistically significant. Declarations ACKNOWLEDGEMENTS We thank Timothy Mitchison (Harvard Medical School), Gregg Gundersen (Columbia University), and Jin Zhu (Mechanobiology Institute, Singapore) for helpful discussion. We acknowledge funding support from Ministry of Education Singapore Academic Research Fund Tier 2 (MOE-T2-EP3-0124-0012, to P.K.), Ministry of Education Singapore Academic Research Fund Tier 3 (MOE-T3-2020-01, to P.K.), National Research Foundation Singapore (NRF-MSG-2023-0001, to P.K.), National Research Foundation Singapore-Israel Science Foundation joint grant (NRF2019-NRF-ISF003-2952 to P.K.), Israel Science Foundation grant 2952/19 to R.Z-B, Chan Zuckerberg Initiative DAF CZI Imaging Scientist Award DOI:10.37921/694870itnyzk (UM), the Goeddel Family Technology Sandbox, and National Science Foundation NSF NeuroNex Award 2014862 (UM). We thank the microscopy, IT, high-throughput molecular genetics, and wet lab core facilities of the Mechanobiology Institute (MBI) for helpful discussion and infrastructure support. Author contributions: Conceptualization: PK, RZB, KF Methodology: KF, AK, SG, RZB, PK, UM, CS, HTO, PR, RN, KYEL, SSC, AV Investigation: KF, AK, SG Visualization: KF, PK Funding acquisition: PK, RZB Project administration: PK, RZB Supervision: PK, RZB Writing – original draft: KF, PK, RZB Writing – review & editing: KF, PK, RZB, UM Data and Materials Availability Data, materials, and code that support the plots and figures within this paper and other findings of study are available from the corresponding author upon reasonable request. Competing interests Authors declare that they have no competing interests. References Sandquist, J. C., Kita, A. M. & Bement, W. M. And the dead shall rise: actin and myosin return to the spindle. Developmental cell 21 , 410-419 (2011). Lechler, T. & Mapelli, M. Spindle positioning and its impact on vertebrate tissue architecture and cell fate. Nature Reviews Molecular Cell Biology 22 , 691-708 (2021). Hernandez, B. et al. Actin organizes chromosomes and microtubules to ensure mitotic fidelity in the preimplantation embryo. Science (New York, N.Y 388 , eads1234 (2025). Mogessie, B. & Schuh, M. Actin protects mammalian eggs against chromosome segregation errors. Science (New York, N.Y 357 , eaal1647 (2017). Schulz, O. et al. Resolution doubling in fluorescence microscopy with confocal spinning-disk image scanning microscopy. Proceedings of the National Academy of Sciences 110 , 21000-21005 (2013). Vieira, A. F. & Paredes, J. P-cadherin and the journey to cancer metastasis. Molecular cancer 14 , 178 (2015). Moore, A. S., Wong, Y. C., Simpson, C. L. & Holzbaur, E. L. Dynamic actin cycling through mitochondrial subpopulations locally regulates the fission–fusion balance within mitochondrial networks. Nat Commun 7 , 12886 (2016). Fung, T. S., Chakrabarti, R. & Higgs, H. N. The multiple links between actin and mitochondria. Nature Reviews Molecular Cell Biology 24 , 651-667 (2023). Valdez, V. A., Neahring, L., Petry, S. & Dumont, S. Mechanisms underlying spindle assembly and robustness. Nature Reviews Molecular Cell Biology 24 , 523-542 (2023). Prosser, S. L. & Pelletier, L. Mitotic spindle assembly in animal cells: a fine balancing act. Nature Reviews Molecular Cell Biology 18 , 187-201 (2017). Bakhoum, S. F. & Cantley, L. C. The multifaceted role of chromosomal instability in cancer and its microenvironment. Cell 174 , 1347-1360 (2018). Li, R. & Zhu, J. Effects of aneuploidy on cell behaviour and function. Nature Reviews Molecular Cell Biology 23 , 250-265 (2022). Bharadwaj, R. & Yu, H. The spindle checkpoint, aneuploidy, and cancer. Oncogene 23 , 2016-2027 (2004). Dumont, J. & Desai, A. Acentrosomal spindle assembly and chromosome segregation during oocyte meiosis. Trends in cell biology 22 , 241-249 (2012). Schuh, M. & Ellenberg, J. Self-organization of MTOCs replaces centrosome function during acentrosomal spindle assembly in live mouse oocytes. Cell 130 , 484-498 (2007). Roeles, J. & Tsiavaliaris, G. Actin-microtubule interplay coordinates spindle assembly in human oocytes. Nat Commun 10 , 4651 (2019). Plessner, M., Knerr, J. & Grosse, R. Centrosomal actin assembly is required for proper mitotic spindle formation and chromosome congression. IScience 15 , 274-281 (2019). Farina, F. et al. Local actin nucleation tunes centrosomal microtubule nucleation during passage through mitosis. Embo J 38 , e99843 (2019). Lancaster, O. M. & Baum, B. in Semin Cell Dev Biol. 109-115 (Elsevier). Stewart, M. P., Helenius, J., Toyoda, Y., Ramanathan, S. P., Muller, D. J. & Hyman, A. A. Hydrostatic pressure and the actomyosin cortex drive mitotic cell rounding. Nature 469 , 226-230 (2011). Salbreux, G., Charras, G. & Paluch, E. Actin cortex mechanics and cellular morphogenesis. Trends in Cell Biology 22 , 536-545 (2012). https://doi.org:10.1016/j.tcb.2012.07.001 Kunda, P. & Baum, B. The actin cytoskeleton in spindle assembly and positioning. Trends in cell biology 19 , 174-179 (2009). Green, R. A., Paluch, E. & Oegema, K. Cytokinesis in animal cells. Annual review of cell and developmental biology 28 , 29-58 (2012). Coscia, S. M., Moore, A. S., Thompson, C. P., Tirrito, C. F., Ostap, E. M. & Holzbaur, E. L. An interphase actin wave promotes mitochondrial content mixing and organelle homeostasis. Nat Commun 15 , 3793 (2024). Moore, A. S. et al. Actin cables and comet tails organize mitochondrial networks in mitosis. Nature 591 , 659-664 (2021). Ragkousi, K. & Gibson, M. C. Cell division and the maintenance of epithelial order. Journal of Cell Biology 207 , 181-188 (2014). Charras, G. & Yap, A. S. Tensile forces and mechanotransduction at cell–cell junctions. Current Biology 28 , R445-R457 (2018). Campàs, O., Noordstra, I. & Yap, A. S. Adherens junctions as molecular regulators of emergent tissue mechanics. Nature Reviews Molecular Cell Biology , 1-18 (2023). Nelson, W. J. Regulation of cell–cell adhesion by the cadherin–catenin complex. Biochem Soc Trans 36 , 149 (2008). Takeichi, M. Dynamic contacts: rearranging adherens junctions to drive epithelial remodelling. Nature Reviews Molecular Cell Biology 15 , 397-410 (2014). Gloerich, M., Bianchini, J. M., Siemers, K. A., Cohen, D. J. & Nelson, W. J. Cell division orientation is coupled to cell-cell adhesion by the E-cadherin/LGN complex. Nat Commun 8 , 13996 (2017). https://doi.org:10.1038/ncomms13996 Benham-Pyle, B. W., Pruitt, B. L. & Nelson, W. J. Mechanical strain induces E-cadherin–dependent Yap1 and β-catenin activation to drive cell cycle entry. Science (New York, N.Y 348 , 1024-1027 (2015). Mendonsa, A. M., Na, T. Y. & Gumbiner, B. M. E-cadherin in contact inhibition and cancer. Oncogene 37 , 4769-4780 (2018). https://doi.org:10.1038/s41388-018-0304-2 Wheelock, M. J., Shintani, Y., Maeda, M., Fukumoto, Y. & Johnson, K. R. Cadherin switching. Journal of cell science 121 , 727-735 (2008). Yates, A. D. et al. Ensembl 2026. Nucleic Acids Res 54 , D1053-D1060 (2026). Van Roy, F. Beyond E-cadherin: roles of other cadherin superfamily members in cancer. Nature Reviews Cancer 14 , 121-134 (2014). Andrews, J. L., Kim, A. C. & Hens, J. R. The role and function of cadherins in the mammary gland. Breast cancer research 14 , 203 (2012). Samuelov, L., Sprecher, E. & Paus, R. The role of P-cadherin in skin biology and skin pathology: lessons from the hair follicle. Cell and tissue research 360 , 761-771 (2015). Nose, A. & Takeichi, M. A novel cadherin cell adhesion molecule: its expression patterns associated with implantation and organogenesis of mouse embryos. The Journal of cell biology 103 , 2649-2658 (1986). Wu, J.-C., Gregory, C. W. & DePhilip, R. M. P-cadherin and E-cadherin are co-expressed in MDCK cells. Biochem Biophys Res Commun 195 , 1329-1335 (1993). Tran, S. K. et al. P-cadherin dependent adhesions are required for single lumen formation and HGF-mediated cell protrusions during epithelial morphogenesis. iScience (2025). Gavard, J., Marthiens, V., Monnet, C., Lambert, M. & Mege, R. M. N-cadherin activation substitutes for the cell contact control in cell cycle arrest and myogenic differentiation: involvement of p120 and beta-catenin. J Biol Chem 279 , 36795-36802 (2004). https://doi.org:10.1074/jbc.M401705200 Bauer, D. E., Canver, M. C. & Orkin, S. H. Generation of genomic deletions in mammalian cell lines via CRISPR/Cas9. JoVE (Journal of Visualized Experiments) , e52118 (2015). Balasubramaniam, L. et al. Investigating the nature of active forces in tissues reveals how contractile cells can form extensile monolayers. Nat Mater 20 , 1156-1166 (2021). Maupin, P. & Pollard, T. D. Arrangement of actin filaments and myosin-like filaments in the contractile ring and of actin-like filaments in the mitotic spindle of dividing HeLa cells. Journal of ultrastructure and molecular structure research 94 , 92-103 (1986). Azuma, T. & Kei, T. Super-resolution spinning-disk confocal microscopy using optical photon reassignment. Optics Express 23 , 15003-15011 (2015). Kita, A. M., Swider, Z. T., Erofeev, I., Halloran, M. C., Goryachev, A. B. & Bement, W. M. Spindle–F-actin interactions in mitotic spindles in an intact vertebrate epithelium. Molecular biology of the cell 30 , 1645-1654 (2019). Harasimov, K. et al. Actin-driven chromosome clustering facilitates fast and complete chromosome capture in mammalian oocytes. Nature cell biology 25 , 439-452 (2023). Booth, A. J. et al. Contractile acto-myosin network on nuclear envelope remnants positions human chromosomes for mitosis. Elife 8 , e46902 (2019). Zheng, Y. A membranous spindle matrix orchestrates cell division. Nature reviews Molecular cell biology 11 , 529-535 (2010). Stamhuis, E. & Thielicke, W. PIVlab–towards user-friendly, affordable and accurate digital particle image velocimetry in MATLAB. Journal of open research software 2 , 30 (2014). Schiavon, C. R. et al. Actin chromobody imaging reveals sub-organellar actin dynamics. Nature methods 17 , 917-921 (2020). Gatti, P., Schiavon, C., Cicero, J., Manor, U. & Germain, M. Mitochondria-and ER-associated actin are required for mitochondrial fusion. Nat Commun 16 , 451 (2025). Riedl, J. et al. Lifeact: a versatile marker to visualize F-actin. Nature methods 5 , 605-607 (2008). Schweizer, N., Pawar, N., Weiss, M. & Maiato, H. An organelle-exclusion envelope assists mitosis and underlies distinct molecular crowding in the spindle region. Journal of Cell Biology 210 , 695-704 (2015). Suarez, C. & Kovar, D. R. Internetwork competition for monomers governs actin cytoskeleton organization. Nature Reviews Molecular Cell Biology 17 , 799-810 (2016). Miserey-Lenkei, S. & Colombo, M. I. Small RAB GTPases regulate multiple steps of mitosis. Frontiers in Cell and Developmental Biology 4 , 2 (2016). Cadwell, C. M., Su, W. & Kowalczyk, A. P. Cadherin tales: Regulation of cadherin function by endocytic membrane trafficking. Traffic 17 , 1262-1271 (2016). Macia, E., Ehrlich, M., Massol, R., Boucrot, E., Brunner, C. & Kirchhausen, T. Dynasore, a cell-permeable inhibitor of dynamin. Developmental cell 10 , 839-850 (2006). Chircop, M. Rho GTPases as regulators of mitosis and cytokinesis in mammalian cells. Small GTPases 5 , e29770 (2014). Mitsushima, M., Toyoshima, F. & Nishida, E. Dual role of Cdc42 in spindle orientation control of adherent cells. Molecular and cellular biology 29 , 2816-2827 (2009). Benink, H. A. & Bement, W. M. Concentric zones of active RhoA and Cdc42 around single cell wounds. The Journal of cell biology 168 , 429-439 (2005). Surviladze, Z. et al. A potent and selective inhibitor of Cdc42 GTPase. Probe Reports from the NIH Molecular Libraries Program [Internet] (2010). Liu, W. et al. Rational identification of a Cdc42 inhibitor presents a new regimen for long-term hematopoietic stem cell mobilization. Leukemia 33 , 749-761 (2019). Watson, J. R., Owen, D. & Mott, H. R. Cdc42 in actin dynamics: An ordered pathway governed by complex equilibria and directional effector handover. Small GTPases 8 , 237-244 (2017). Kühn, S. & Geyer, M. Formins as effector proteins of Rho GTPases. Small GTPases 5 , e983876 (2014). Nishimura, Y. et al. The formin inhibitor SMIFH2 inhibits members of the myosin superfamily. Journal of cell science 134 , jcs253708 (2021). Kühn, S. et al. The structure of FMNL2–Cdc42 yields insights into the mechanism of lamellipodia and filopodia formation. Nat Commun 6 , 7088 (2015). Pfisterer, K. et al. FMNL2 regulates dynamics of fascin in filopodia. Journal of Cell Biology 219 (2020). Cho, K. F., Branon, T. C., Udeshi, N. D., Myers, S. A., Carr, S. A. & Ting, A. Y. Proximity labeling in mammalian cells with TurboID and split-TurboID. Nat Protoc 15 , 3971-3999 (2020). Ollech, D. et al. An optochemical tool for light-induced dissociation of adherens junctions to control mechanical coupling between cells. 11 , 472 (2020). Szklarczyk, D. et al. The STRING database in 2023: protein–protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res 51 , D638-D646 (2023). Fox, S. et al. Identification of an FMNL2 Interactome by Quantitative Mass Spectrometry. International Journal of Molecular Sciences 25 , 5686 (2024). Gadella Jr, T. W. et al. mScarlet3: a brilliant and fast-maturing red fluorescent protein. Nature methods 20 , 541-545 (2023). Birch, J. L. et al. A novel small-molecule inhibitor of MRCK prevents radiation-driven invasion in glioblastoma. Cancer research 78 , 6509-6522 (2018). Fujita, Y. et al. Hakai, a c-Cbl-like protein, ubiquitinates and induces endocytosis of the E-cadherin complex. Nature cell biology 4 , 222-231 (2002). Pan, M.-H. et al. FMNL2 regulates actin for endoplasmic reticulum and mitochondria distribution in oocyte meiosis. Elife 12 , RP92732 (2024). Rufas, O., Fisch, B., Ziv, S. & Shalgi, R. Expression of cadherin adhesion molecules on human gametes. Mol Hum Reprod 6 , 163-169 (2000). Tábara, L.-C., Segawa, M. & Prudent, J. Molecular mechanisms of mitochondrial dynamics. Nature Reviews Molecular Cell Biology 26 , 123-146 (2025). Ivaska, J., Heino, J. J. A. r. o. c. & biology, d. Cooperation between integrins and growth factor receptors in signaling and endocytosis. 27 , 291-320 (2011). Alanko, J. et al. Integrin endosomal signalling suppresses anoikis. Nature cell biology 17 , 1412-1421 (2015). Vassilev, V., Platek, A., Hiver, S., Enomoto, H. & Takeichi, M. Catenins steer cell migration via stabilization of front-rear polarity. Developmental cell 43 , 463-479. e465 (2017). Nollet, F., Kools, P. & Van Roy, F. Phylogenetic analysis of the cadherin superfamily allows identification of six major subfamilies besides several solitary members. J Mol Biol 299 , 551-572 (2000). Zhang, C. C. et al. PF-03732010: a fully human monoclonal antibody against P-cadherin with antitumor and antimetastatic activity. Clinical Cancer Research 16 , 5177-5188 (2010). Additional Declarations There is NO Competing Interest. Supplementary Files ExtendedDataMaterial.docx Movie6m.mp4 Movie 6 DataS120260316.xlsx Dataset 1 Movie4m.mp4 Movie 4 Movie5m.mp4 Movie 5 Movie3m.mp4 Movie 3 Movie1m.mp4 Movie 1 Movie2m.mp4 Movie 2 Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9256156","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Biological Sciences - Article","associatedPublications":[],"authors":[{"id":628475329,"identity":"b18633eb-10da-49f2-aee3-40844ceb9bf4","order_by":0,"name":"Pakorn Kanchanawong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAy0lEQVRIiWNgGAWjYBACA2YGZgaGCghHggQtZ0jSwgDUwthGihZzdubDxoXz7iTOb2A+eJuHoc6ugZAWy2a25OSZ254lbjjAlmzNw3A4maAWg8M8xod5tx1O3MDAYybNw3AgmaDDDA7zfz7MO+cw0GH834Ba6ojRwsOczNtwOLHhAA8bUAuzHUEtQL8YG884dth4w2E2Y8s5BocTCGox5z/8WLqg5rDs/PbmhzfeVNTZE9QCAswI0oAhsYF4LVBAnC2jYBSMglEwogAAufA2uHFVEWcAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-8765-3861","institution":"National University of Singapore","correspondingAuthor":true,"prefix":"","firstName":"Pakorn","middleName":"","lastName":"Kanchanawong","suffix":""},{"id":628475330,"identity":"fcf33c51-ff33-408a-bc47-bf9b742eb9e1","order_by":1,"name":"Kedsarin Fong-Ngern","email":"","orcid":"","institution":"National University of Singapore","correspondingAuthor":false,"prefix":"","firstName":"Kedsarin","middleName":"","lastName":"Fong-Ngern","suffix":""},{"id":628475331,"identity":"e3307f83-a97f-4917-8b6e-a248757d700a","order_by":2,"name":"Sumit Gupta","email":"","orcid":"","institution":"Tel Aviv University","correspondingAuthor":false,"prefix":"","firstName":"Sumit","middleName":"","lastName":"Gupta","suffix":""},{"id":628475332,"identity":"58280522-5ed5-4e5a-8e38-ac79461f8ad4","order_by":3,"name":"Alexandra Kaminer","email":"","orcid":"https://orcid.org/0009-0009-6516-7008","institution":"Goethe-University Frankfurt","correspondingAuthor":false,"prefix":"","firstName":"Alexandra","middleName":"","lastName":"Kaminer","suffix":""},{"id":628475333,"identity":"d5a672ac-a9d4-4630-9d67-74abed137bcd","order_by":4,"name":"Pratchaya Rukthanapitak","email":"","orcid":"","institution":"National University of Singapore","correspondingAuthor":false,"prefix":"","firstName":"Pratchaya","middleName":"","lastName":"Rukthanapitak","suffix":""},{"id":628475334,"identity":"941e90da-aeb5-425e-8ed2-0d0c282a5fa0","order_by":5,"name":"Ryosuke Nishimura","email":"","orcid":"","institution":"National University of Singapore","correspondingAuthor":false,"prefix":"","firstName":"Ryosuke","middleName":"","lastName":"Nishimura","suffix":""},{"id":628475335,"identity":"6304b887-fe1a-43a2-8235-7bae69960858","order_by":6,"name":"Kyna Lim","email":"","orcid":"","institution":"National University of Singapore","correspondingAuthor":false,"prefix":"","firstName":"Kyna","middleName":"","lastName":"Lim","suffix":""},{"id":628475336,"identity":"07b0b3d6-0910-44cb-84b0-878ec18b698e","order_by":7,"name":"Samantha Chang","email":"","orcid":"https://orcid.org/0009-0005-5469-3601","institution":"National University of Singapore","correspondingAuthor":false,"prefix":"","firstName":"Samantha","middleName":"","lastName":"Chang","suffix":""},{"id":628475337,"identity":"d1cce898-bfe2-4f99-ae93-0e330026db23","order_by":8,"name":"Arulchelvan Vijayakumar","email":"","orcid":"","institution":"National University of Singapore","correspondingAuthor":false,"prefix":"","firstName":"Arulchelvan","middleName":"","lastName":"Vijayakumar","suffix":""},{"id":628475338,"identity":"5a60d420-ee54-4a60-86d5-750d88850b5e","order_by":9,"name":"Hui Ting Ong","email":"","orcid":"https://orcid.org/0000-0002-3434-4683","institution":"Mechanobiology Institute, National University of Singapore, 117411, Singapore.","correspondingAuthor":false,"prefix":"","firstName":"Hui","middleName":"Ting","lastName":"Ong","suffix":""},{"id":628475339,"identity":"e809e072-3815-48c6-87d8-e7c9c155dd60","order_by":10,"name":"Cara Schiavon","email":"","orcid":"https://orcid.org/0000-0002-9311-2145","institution":"UCSD","correspondingAuthor":false,"prefix":"","firstName":"Cara","middleName":"","lastName":"Schiavon","suffix":""},{"id":628475340,"identity":"0c352d35-90e6-4b87-8730-71d66f971db5","order_by":11,"name":"Uri Manor","email":"","orcid":"https://orcid.org/0000-0002-9802-1955","institution":"UC San Diego","correspondingAuthor":false,"prefix":"","firstName":"Uri","middleName":"","lastName":"Manor","suffix":""},{"id":628475341,"identity":"b75d11f7-a259-47ad-b226-a69cb3e334be","order_by":12,"name":"Ronen Zaidel-Bar","email":"","orcid":"https://orcid.org/0000-0002-1374-5007","institution":"Tel-Aviv University","correspondingAuthor":false,"prefix":"","firstName":"Ronen","middleName":"","lastName":"Zaidel-Bar","suffix":""}],"badges":[],"createdAt":"2026-03-29 03:25:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9256156/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9256156/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108007928,"identity":"2a53d1ec-c624-4eae-b752-5763942eccfe","added_by":"auto","created_at":"2026-04-28 13:04:45","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1697854,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eP-cadherin is essential for mitotic fidelity.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003ea\u003c/strong\u003e) Enrichment of P-cadherin puncta near spindle poles during mitotic stages in MDCK II cells. Main panels: enlarged 2-color composites of P-cadherin (green) and centrosomes (γ-tubulin, magenta), with mitotic cell boundaries outlined (white dashed lines). Side panels: DNA/γ-tubulin or three-color composites (P-cadherin, green; γ-tubulin, magenta; F-actin, orange; DNA, cyan). (\u003cstrong\u003eb-c\u003c/strong\u003e) Quantification of P-cadherin total intensity, cytoplasmic intensity, and cytoplasmic/total ratio in interphase versus mitotic cells. (\u003cstrong\u003ed\u003c/strong\u003e) Immunoblot of synchronized cells showing mitotic enrichment of P-cadherin, whereas E-cadherin levels remain constant. GAPDH serves as loading control. (\u003cstrong\u003ee\u003c/strong\u003e) Representative spindle morphologies in WT and P-cadherin KO. Upper panels: 3-color composites (α-tubulin, orange; γ-tubulin, magenta; DNA, cyan), with magnified insets as 2-color composites. (\u003cstrong\u003ef\u003c/strong\u003e) Quantification of mitotic defects in WT, P-cadherin KO, E-cadherin KO, and P-cadherin-rescue cells. (\u003cstrong\u003eg\u003c/strong\u003e) Re-expression of P-cadherin rescues mitotic defects. 4-color datasets are decomposed into separate views for clarity: P-cadherin-GFP alone (left), 3-color composite (P-cadherin, green; α-tubulin, orange; DNA, cyan; center left), 2-color composite (α-tubulin, orange; DNA, cyan; center right), and 3-color composite (α-tubulin, orange; F-actin, blue; DNA, cyan; right). (\u003cstrong\u003eh–k\u003c/strong\u003e) Frequencies of mitotic abnormalities: (h) spindle misorientation, (i) chromosome misalignment, (j) spindle distortion, and (k) multipolar spindles. ***: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001, ****: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001. Scale bars, 10 μm (a, e, g).\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/e8483f67ae5b3280870e29c9.png"},{"id":108007894,"identity":"3e4dfd08-b0ef-4eb9-a7f7-dad5f89fb63a","added_by":"auto","created_at":"2026-04-28 13:04:14","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1736208,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eP-cadherin regulates mitotic interior actin network.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003ea\u003c/strong\u003e) Super-resolution imaging of interior actin networks across various mitotic stages in MDCK II WT. Top panels: 3-color composite (F-actin, orange; α-tubulin, blue; DNA, magenta). Middle and lower panels: 2-color composite of the spindles (α-tubulin and DNA) and F-actin, respectively. (\u003cstrong\u003eb\u003c/strong\u003e) Enlarged view of boxed region in (a) showing actin cables extending from spindle pole during NEBD (arrows). (\u003cstrong\u003ec\u003c/strong\u003e) Quantification of interior actin density. (Leftmost) Definition of polar coordinate on mid-spindle plane centered at each metaphase centrosome, from which the 2D heatmap (angle versus distance relative to spindle pole) of actin density are calculated (center left, normalized against cortical actin). Interior actin density map (azimuthal angle: -90° to 90° relative to spindle axis) is shown in center right panel. (Rightmost) Plot of normalized actin density as a function of distance from spindle pole. (see \u003cstrong\u003eExtended Data Fig.3d\u003c/strong\u003e for P-cadherin KO). (\u003cstrong\u003ed\u003c/strong\u003e) Comparison of metaphase actin architecture between WT and P-cadherin KO cells. Left: 3-color composite (F-actin, orange; α-tubulin, blue; DNA, magenta). Center: F-actin channel. Right: enlarged view around metaphase spindle pole (boxed region in left panels). White arrows indicate peri-mitochondrial actin. (\u003cstrong\u003ee\u003c/strong\u003e) Semi-polar heatmap of interior actin density in WT and P-cadherin KO cells. Average of 2D heatmap from (c) is calculated from indicated number of cells. Scale bars: 10 µm (a, d), 5 µm (c).\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/02b6d5d187b0b5c45cc6ed6f.png"},{"id":108008708,"identity":"3768d730-5121-4e1d-b58b-a7ce3c005751","added_by":"auto","created_at":"2026-04-28 13:08:03","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1515284,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eP-cadherin promotes interior actin dynamics facilitating chromosome congression.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003ea-b\u003c/strong\u003e) Time-lapse live-cell super-resolution imaging of WT or P-cadherin KO cells during NEBD, showing decreased actin cable formation and increased mitochondria-actin association in KO (see also \u003cstrong\u003eMovies 1-2\u003c/strong\u003e). 4-color datasets are decomposed into separate views for clarity: (upper panels) 3-color composite (LifeAct-mEmerald, orange; SiR-tubulin, green; DNA, magenta), and boxed regions with 2-color composite (MitoTracker, cyan; LifeAct-mEmerald, orange) shown in insets. (Center panels) Montages of spindle poles (1 and 2) as inverted grayscale (actin) and 2-color composite (LifeAct-mEmerald, orange; SiR-tubulin, green), respectively. (Lower panels) Kymographs along white dashed lines (spindle pole 2), showing association of actin cables emanating from spindle pole with the mitotic spindle in WT (a), and its absence in KO (b). (\u003cstrong\u003ec-d\u003c/strong\u003e) Particle Image Velocimetry (PIV) analysis of actin dynamics during NEBD. Representative images correspond to WT (c) or P-cadherin KO (d) cells depicted in (a) or (b) respectively, with PIV vector field overlaid on LifeAct-mEmerald channel used for PIV calculation. (\u003cstrong\u003ee\u003c/strong\u003e) Mean cytoplasmic actin cables velocity following NEBD comparing WT and P-cadherin KO, corresponding to (c) and (d). (\u003cstrong\u003ef-g\u003c/strong\u003e) Average speed of cytoplasmic actin cables (f) or chromosomes (g) during NEBD. (\u003cstrong\u003eh-j\u003c/strong\u003e) Delayed spindle pole migration and chromosome congression in P-cadherin KO. White dashed lines denote spindle positions at \u003cem\u003et\u003c/em\u003e=0, yellow dashed lines the positions at \u003cem\u003et\u003c/em\u003e=120 s, overlaid on still images from (a) and (b) datasets (SiR-tubulin, green; DNA, magenta), with quantification of clustered chromosome area in (i) and inter spindle pole distance in (j). (\u003cstrong\u003ek-l\u003c/strong\u003e) Reduced interior actin cable formation in P-cadherin KO. Inverted grayscale frame of LifeAct-mEmerald channel from (a) and (b) datasets, with clustered chromosomes areas marked by magenta dashed lines, with quantification of total actin cable length in (l). Scale bars: 10 µm (a, b). Two-tailed unpaired \u003cem\u003et\u003c/em\u003e test (f, g), *: \u003cem\u003eP\u003c/em\u003e\u0026lt; 0.05, ****: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/c17c80d867effc05b00983d8.png"},{"id":108007919,"identity":"7c94e8c6-e985-4262-a161-1d64e2c5d45e","added_by":"auto","created_at":"2026-04-28 13:04:39","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1625160,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLoss of P-cadherin enhances peri-mitochondrial actin and impairs mitochondrial morphology and functions.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003ea\u003c/strong\u003e) Peri-mitochondrial actin visualized by AC-mito probe in WT (upper) and P-cadherin KO (lower) (see also \u003cstrong\u003eMovie 3\u003c/strong\u003e). Left: composite 3-color data (AC-mito, green; MitoTracker, magenta; DNA: blue). Inner left: 2-color composites of AC-mito (green) and MitoTracker (magenta). Inner right: AC-mito channel. Right: enlarged views corresponding to boxed regions in left panels. Intensity profiles correspond to selected regions in zoomed views. (\u003cstrong\u003eb\u003c/strong\u003e) Enhanced peri-mitochondria in P-cadherin KO. Kymographs corresponding to yellow boxed regions in zoomed view in (a). (\u003cstrong\u003ec\u003c/strong\u003e) Live-cell super-resolution confocal images of interphase WT cells (top) and P-cadherin KO (bottom) with stable expression of LifeAct-mEmerald (green), showing actin encapsulation of mitochondria (magenta: MitoTracker) in P-cadherin KO. (Right) Decomposed individual channels corresponding to boxed regions on left-hand panels. (\u003cstrong\u003ed\u003c/strong\u003e) Live-cell super-resolution confocal images of metaphase cells at mid-spindle plane, comparing MDCK II WT (top) and P-cadherin KO (bottom) with stable expression of LifeAct-mEmerald, showing spindle-encroachment by mitochondria in P-cadherin KO. 4-color composite (orange, LifeAct-mEmerald; green, SiR-tubulin; cyan, MitoTracker; magenta, DNA). (\u003cstrong\u003ee-f\u003c/strong\u003e) Quantification of mitochondria coverage as percent of total spindle area in metaphase cells (e) or total cell area in interphase cells (f). (\u003cstrong\u003eg, k\u003c/strong\u003e) Quantification of mitochondria properties: area, length, circularity, density, fractional area coverage, and tracked speed (k), comparing WT and P-cadherin KO. (\u003cstrong\u003eh-j\u003c/strong\u003e) Maximum intensity projection of live-cell confocal images\u003cstrong\u003e \u003c/strong\u003eof confluent epithelial monolayers, comparing mitochondria morphology (yellow: MitoTracker), between WT ( h), P-cadherin KO (i), and P-cadherin KO rescued with P-cadherin-GFP overexpression (j), showing significant aggregation and fragmentation in P-cadherin KO, and hyper-elongation in P-cadherin overexpression. Enlarged views of boxed regions in top panels (h, i) shows 2-color composite of MitoTracker (yellow) and histone-H2B-mCherry (cyan) and overlay of MitoTracker on DIC channel (grayscale), or (j) individual channel for MitoTracker and P-cadherin-GFP. (Bottom, h, i). Enlarged views corresponding to arrows in center panels, showing uneven distribution of mitochondria in P-cadherin KO. (\u003cstrong\u003el, m\u003c/strong\u003e) Effects of P-cadherin depletion on mitochondria membrane potential and mitochondrial superoxide as probed by TMRM (l) or MitoSox (m), respectively. (Left) Maximum intensity projection of (top) TMRM (l) or MitoSox (m) and (bottom) 2-color composites (yellow, MitoTracker; cyan, DNA), comparing WT (left) and P-cadherin KO (right). (Right) Quantification of TMRM or MitoSox intensity. Scale bars: 5 µm (a, c, d); 10 µm (h-j, l, m). ***: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001; ****: \u003cem\u003eP\u003c/em\u003e\u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/857f54222d546db52d242e40.png"},{"id":108008520,"identity":"b6f45a08-8a4a-4209-b1d9-bb601efd00bb","added_by":"auto","created_at":"2026-04-28 13:07:11","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1690499,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eP-cadherin is trafficked in Rab5 endosomes in association with actin cables.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003ea-c\u003c/strong\u003e) P-cadherin is localized in endosomal vesicles in mitotic spindle vicinity. Live-cell 3-color super-resolution imaging of MDCK P-cadherin KO rescued with P-cadherin-GFP during NEBD (a) or prometaphase (b-c) (P-cadherin-GFP, green; SiR-tubulin, magenta; DNA, blue). (a, right) Enlarged 2-color composites corresponding to boxed region (left) showing P-cadherin endosomes clustering near the centrosomes during NEBD. (b: right panels) 2-color composite (GFP, green; SiR-tubulin, magenta) and individual channels, showing co-localization of P-cadherin endosomes with centrosomes (white arrows). (c) Time-lapse 3-color montage corresponding to yellow-dashed region in (b), showing P-cadherin endosome movement on mitotic spindle (white arrow) during prometaphase. (\u003cstrong\u003ed-e\u003c/strong\u003e) P-cadherin endosomes are associated with actin cables. Time-lapse 3-color live-cell super-resolution imaging in P-cadherin KO rescued with P-cadherin-tdTomato (d, interphase cell, \u003cstrong\u003eMovie 4\u003c/strong\u003e) or P-cadherin-GFP (e, prometaphase cell, \u003cstrong\u003eMovie 5\u003c/strong\u003e) and expressing LifeAct probes (d: LifeAct-mEmerald, e: LifeAct-mScarlet-I). (d, right) Montage of 2-color composite corresponding to boxed region, showing actin comets trailing P-cadherin endosomes. (e, right) Zoomed view corresponding to boxed region (left) showing actin cables associated with P-cadherin. (\u003cstrong\u003ef\u003c/strong\u003e) Schematic diagram depicting endocytic trafficking of P-cadherin. (\u003cstrong\u003eg\u003c/strong\u003e) Quantification of endosomal co-localization between P-cadherin-GFP and various endosomal markers (see also \u003cstrong\u003eExtended Data Fig. 4a\u003c/strong\u003e), indicative of Rab5 co-trafficking. (\u003cstrong\u003eh\u003c/strong\u003e) Co-localization of endogenous P-cadherin and Rab5. Left: 2-color composite image (anti P-cadherin, green; anti-Rab5, magenta; DNA, blue). Right panels: Zoomed view corresponding to boxed region denoted 1 or 2 in left panels, with arrows highlighting co-localized endosomes. (\u003cstrong\u003ei, j\u003c/strong\u003e) P-cadherin co-localizes with Rab5 primarily in endosomes. Live-cell confocal images of WT cells co-expressing P-cadherin-GFP and mCherry-Rab5b during interphase (i) or mitosis (j), with arrows marking endosomes containing both P-cadherin and Rab5b. (\u003cstrong\u003ek-m\u003c/strong\u003e) Rescue of mitotic fidelity in P-cadherin KO by chimeric cadherin containing P-cadherin cytoplasmic domain. (k) Super-resolution confocal image of P-cadherin KO (left), and KO cells expressing P-cadherin-GFP, ecto-P-cad::cyto-E-cad-GFP, or ecto-E-cad::cyto-P-cad-GFP. Upper panels show composite 3-channel image for P-cad KO or 4-channel images for chimera expressing cells (GFP, green; F-actin, orange; α-tubulin, blue; DNA, magenta). Lower panels: decomposed views. Quantification of normal metaphase and chromosome misalignments are shown in (l) and (m), respectively. Scale bars: 5 µm (a, b, e, h-k), 10 µm (d). *: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05, ****: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/07ae8e5e7d5861ded804b401.png"},{"id":108007918,"identity":"5b6480d0-be95-4819-b0b4-a27b23ad1031","added_by":"auto","created_at":"2026-04-28 13:04:39","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1625163,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eP-cadherin-dependent interior actin networks are generated by FMNL2 formin downstream of active Cdc42.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003ea-b, d\u003c/strong\u003e) P-cadherin co-localizes with active Cdc42 primarily in endosomes. Live-cell confocal images of MDCK P-cadherin KO expressing P-cadherin-GFP and the active Cdc42 biosensor, mRFP-wGBD, during interphase (a) or mitosis (b, see also \u003cstrong\u003eExtended Data Fig. 5a-b\u003c/strong\u003e). (Right panels) Zoomed view of boxed region with arrows indicating endosomes with co-localization. Pearson’s coefficient provided for indicated cells. (d) Quantification of active Cdc42 positive vesicles, suggesting that P-cadherin regulates the endosomal activity of Cdc42. (\u003cstrong\u003ec\u003c/strong\u003e) Visualization of endosomal actin polymerization by active Cdc42- vesicles. Time-lapse frames of live-cell 2-color confocal composites (LifeAct-mEmerald, blue; mRFP-wGBD, magenta). (Lower panels) zoomed montage with arrows denoting transient actin polymerization on active Cdc42 endosomes. (\u003cstrong\u003ee-g\u003c/strong\u003e) Pharmacological inhibition of Cdc42 impairs interior actin network and mitotic fidelity. Super-resolution confocal images (g) of MDCK II WT treated with ML141 or CASIN to inhibit Cdc42 activity. Left panels show composite 3-channel image (F-actin, orange; α-tubulin, blue; DNA, magenta). Center and right panels: F-actin channel and zoomed views of boxed spindle regions, respectively. (e, f) Quantification of mitotic fidelity and chromosome misalignment respectively, upon Cdc42 inhibition. (\u003cstrong\u003eh-j\u003c/strong\u003e) Formin inhibition depletes interior actin network. Super-resolution confocal images (h) of MDCK II WT treated with the formin inhibitor, SMIFH2, or Arp2/3 inhibitor, CK666. Left panels show composite 3-channel image (F-actin, orange; α-tubulin, blue; DNA, magenta). Center and right panels: 2-color composite of spindle and inverted contrast F-actin channel, respectively. (i) Intensity profile along the spindle axis (highlighted band in diagram), comparing control with CK666 or SMIFH2 treatment and (j) quantification of mitotic fidelity. (\u003cstrong\u003ek-n\u003c/strong\u003e) FMNL2 and MRCKβ regulates mitotic fidelity and interior actin network. Super-resolution confocal images (k) of MDCK II WT comparing siRNA depletion of P-cadherin with FMNL2, FMNL3, and MRCKβ (see also \u003cstrong\u003eFig. 7\u003c/strong\u003e and \u003cstrong\u003eExtended Data Fig. 6\u003c/strong\u003e). Quantification of siRNA depletion and mitotic fidelity of FMNL2 and FMNL3 (l) and mitotic fidelity (m). (n) Quantification of mitotic fidelity upon siRNA depletion of additional candidates identified in Extended Data Fig. 6. Scale bars: 5 µm (b, g, h, k), 10 µm (a, c). *: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05; ***: P\u0026lt; 0.001; ****: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/9ae2b9fbef3d6f80d79c1c0a.png"},{"id":108008511,"identity":"fe0d3f14-6f18-410f-a3e9-b68ea3ed4bc8","added_by":"auto","created_at":"2026-04-28 13:07:03","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1339658,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCdc42/MRCKβ/FMNL2 pathway mediates P-cadherin-dependent interior actin network to promote mitotic fidelity and organelle homeostasis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003ea-d\u003c/strong\u003e) MRCKβ localization to cell-cell junction and endosomes depends on P-cadherin. 3-color confocal datasets of MDCK II WT (upper) or P-cadherin KO cells (lower) expressing MRCKβ-mScarlet3 (magenta) and stained for α-tubulin (green) and DNA (blue) during interphase (a) or metaphase (d) (see \u003cstrong\u003eExtended Data Fig. 7e\u003c/strong\u003e for prometaphase). Arrows (a) indicate MRCKβ localization to endosomes (endo) or adherens junctions (AJ) localization in WT, contrasting with diffuse cytoplasmic localization in P-cadherin KO. (b-c) Quantification of MRCKβ localization at AJ (b) and endosome localization (c) comparing WT versus P-cadherin KO. (\u003cstrong\u003ee-f\u003c/strong\u003e) Endosome co-localization of P-cadherin and MRCKβ. Confocal 2-color composite image of P-cadherin-GFP (green) re-expressed in P-cadherin KO cells and MRCKβ-mScarlet3 (magenta) with enlarged view and intensity line profiles in (f) corresponding to dashed lines. (\u003cstrong\u003eg\u003c/strong\u003e) Endosomal co-localization of FMNL2 and MRCKβ. Left: confocal 2-color composite image of FMNL2-GFP (green) and MRCKβ-mScarlet3 (magenta) expressed in MDCK WT cells. Right: enlarged views corresponding to boxed region. Arrows denote co-localized endosomes. (\u003cstrong\u003eh\u003c/strong\u003e) Endosomal co-localization of MRCKβ, Rab5, and actin in MDCK WT. 3-color confocal images are decomposed with left and top-right as 2-color composite image of MRCKβ-mScarlet3 (magenta) and iRFP-FRB-Rab5 (blue). Lower left: 2-color composite of mScarlet3 (magenta) and LifeAct-mEmerald (green). Enlarged views corresponding to boxed region (left), with arrows denoting Rab5 endosome localization of MRCKβ (top right) or actin cables associated with MRCKβ-containing endosome (lower right). (\u003cstrong\u003ei-k\u003c/strong\u003e) Pharmacological inhibition of MRCKβ impairs interior actin network and mitotic fidelity. Super-resolution confocal images (i) of MDCK II WT comparing DMSO with BDP9066 treatment (MRCKβ inhibitor). Zoomed (right panel) indicates F-actin channel, with arrow indicating abnormal actin clump in BDP9066-treated cell. (j-k) Quantification of mitotic fidelity and chromosome misalignment, respectively. (\u003cstrong\u003el\u003c/strong\u003e) Working model depicting the P-cadherin-dependent signalling pathways driving endosome-generated interior actin network assembly to maintain mitotic fidelity and organelle homeostasis, operating through Rab5 endosomes and Cdc42-MRCKβ-FMNL2 axis (see also \u003cstrong\u003eExtended Data Fig. 7i\u003c/strong\u003e). Scale bars: 5 µm (a, d, e, g-i). ***: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001, ****: \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/077d20471c19839e9f8854f0.png"},{"id":108010341,"identity":"acbda998-d12d-4cb5-a090-f2cbeeff4f2b","added_by":"auto","created_at":"2026-04-28 13:13:13","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":11904936,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/e8796efe-0caf-48d9-913d-877813f3e198.pdf"},{"id":108007917,"identity":"d2e7267e-4da2-4704-bc42-ab379652f83a","added_by":"auto","created_at":"2026-04-28 13:04:39","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":5798911,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedDataMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/d604977de0c3fc803aa0a603.docx"},{"id":108007924,"identity":"c67c0b2a-7647-4f9b-a5c8-fb3646551364","added_by":"auto","created_at":"2026-04-28 13:04:44","extension":"mp4","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":139924,"visible":true,"origin":"","legend":"Movie 6","description":"","filename":"Movie6m.mp4","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/1588f29a4cc3a8bef46de84b.mp4"},{"id":108008516,"identity":"52ef936f-1837-4ed1-a764-4f0f0890377b","added_by":"auto","created_at":"2026-04-28 13:07:04","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":335654,"visible":true,"origin":"","legend":"Dataset 1","description":"","filename":"DataS120260316.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/bcd86bf8dc6a2fb1f06c2105.xlsx"},{"id":108008652,"identity":"ec926e0b-89b5-4d5c-8d1e-922d1d812803","added_by":"auto","created_at":"2026-04-28 13:07:47","extension":"mp4","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":2003471,"visible":true,"origin":"","legend":"Movie 4","description":"","filename":"Movie4m.mp4","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/2f43c1de7cc24e49473f6a46.mp4"},{"id":108007927,"identity":"7cfa0f93-9b38-4ce2-a4b8-cc61b700cf53","added_by":"auto","created_at":"2026-04-28 13:04:45","extension":"mp4","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":1387361,"visible":true,"origin":"","legend":"Movie 5","description":"","filename":"Movie5m.mp4","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/497d1a92ddb9a2bdef72e878.mp4"},{"id":108007920,"identity":"09c9e218-08ad-49bb-a6f9-5c6c2d6752b0","added_by":"auto","created_at":"2026-04-28 13:04:40","extension":"mp4","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":1261635,"visible":true,"origin":"","legend":"Movie 3","description":"","filename":"Movie3m.mp4","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/033b0105c06f2092a215157c.mp4"},{"id":108008728,"identity":"8771bcc9-66bb-4a23-9ce0-9d76d7ad77f5","added_by":"auto","created_at":"2026-04-28 13:08:13","extension":"mp4","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":3833631,"visible":true,"origin":"","legend":"Movie 1","description":"","filename":"Movie1m.mp4","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/a13e0d6604210460acfaa122.mp4"},{"id":108007921,"identity":"38713e6c-e9e9-4a5f-864b-4f56983d5765","added_by":"auto","created_at":"2026-04-28 13:04:40","extension":"mp4","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":4539167,"visible":true,"origin":"","legend":"Movie 2","description":"","filename":"Movie2m.mp4","url":"https://assets-eu.researchsquare.com/files/rs-9256156/v1/7b25d2d908988235d707cf79.mp4"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Endosome-generated interior actin network regulates mitotic fidelity and organelle homeostasis downstream of P-cadherin","fulltext":[{"header":"Introduction","content":"\u003cp\u003eHigh fidelity segregation of genetic materials during cell division requires the accurate assembly of the microtubule-based spindle apparatus, which orchestrates the precise spatial organization and subsequent equipartition of the chromosomes\u003csup\u003e2,9,10\u003c/sup\u003e. Compromised integrity of the spindles gives rise to deleterious consequences ranging from cell death to developmental defects and cancer\u003csup\u003e11-13\u003c/sup\u003e. \u0026nbsp;In recent years, substantial plasticity in mechanisms regulating the spindles have been recognized. For instance, although the centrosomes serve as dominant spindle microtubule-organizing centers in many somatic cells, robust spindle assembly proceeds without centrosomes in diverse contexts including mouse preimplantation embryos (acentrosomal mitosis) or in oocyte meiosis (meiosis I and meiosis II)\u003csup\u003e14,15\u003c/sup\u003e. \u0026nbsp;Furthermore, in such acentrosomal spindles the actin cytoskeleton plays essential and multifaceted roles governing the fidelity of chromosome segregation, including forming an integral part of the spindle architecture in the oocyte meiotic spindles\u003csup\u003e4,16\u003c/sup\u003e or serving as a peri-spindle actin shell in mouse preimplantation embryos\u003csup\u003e3\u003c/sup\u003e.\u003cem\u003e\u0026nbsp;\u003c/em\u003e In contrast, centrosomal mitosis has conventionally been considered to be primarily microtubule-driven, in significant part due to the observed absence of actin in the spindle proximity\u003csup\u003e1\u003c/sup\u003e. While there have been multiple studies supporting the centrosomes as active actin polymerization sites during mitosis\u003csup\u003e17,18\u003c/sup\u003e, the contribution of actin to faithful chromosome segregation during mitosis remains incompletely understood.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe involvement of actin in mitotic cell division is most extensively characterized in terms of its roles in regulating cell geometry and cell mechanics\u003csup\u003e19,20\u003c/sup\u003e. This is primarily organized by the actin cortex, which actively drives cell shape changes\u003csup\u003e21\u003c/sup\u003e, coordinates cell division polarity\u003csup\u003e22\u003c/sup\u003e, and mediates cytokinesis\u003csup\u003e23\u003c/sup\u003e. Beyond the cortical pool, a recently discovered actin population in the cellular interior contributes to proper mitosis by ensuring equipartition of endomembrane organelles such as mitochondria between daughter cells\u003csup\u003e24,25\u003c/sup\u003e. Unlike the robust cortical actin, such interior actin networks consist of delicate meshes of transient actin cables that oscillate as \u0026lsquo;waves\u0026rsquo; or form \u0026lsquo;comet tails\u0026rsquo; trailing mitochondria, and thus its characterization require demanding high-resolution imaging conditions. At present, to what extent such dynamic organelle-associated actin contributes to other aspects of mitosis remains unexplored.\u003c/p\u003e\n\u003cp\u003eIn epithelial tissues, mitosis must be tightly regulated to preserve tissue integrity, maintain the physical barrier, and prevent tumorigenesis\u003csup\u003e26\u003c/sup\u003e. E-cadherin, along with its associated catenin complexes, mediate the physical connection between neighbouring cells and organize the cortical actin architecture at the epithelial cell-cell junctions\u003csup\u003e27-30\u003c/sup\u003e. Importantly, E-cadherin is also directly involved in mitotic spindle orientation\u003csup\u003e31\u003c/sup\u003e, contributes to contact inhibition via catenin-dependent signalling\u003csup\u003e32\u003c/sup\u003e, and is broadly considered a tumour suppressor\u003csup\u003e33\u003c/sup\u003e. However, during development and disease, E-cadherin expression is often supplanted or supplemented by other classical cadherins, whose functional distinctions have not been fully understood\u003csup\u003e34\u003c/sup\u003e. E-cadherin and P-cadherin are close paralogs encoded by tandem, immediately adjacent genes within the classical cadherin cluster on human chromosome 16q22.1 (separated by ~3.5 kb intergenic interval) and co-expressed in the epithelia of multiple tissues\u003csup\u003e35,36\u003c/sup\u003e. While aberrant P-cadherin expression is strongly associated with poor prognosis in a wide variety of cancers (Paredes et al., 2012; Wu et al., 1993), P-cadherin also fulfils context-specific roles in normal tissues, ranging from mammary gland epithelium and epidermis\u003csup\u003e37,38\u003c/sup\u003e to placenta, where it was originally discovered\u003csup\u003e39\u003c/sup\u003e. \u0026nbsp;Despite the clinical relevance of P-cadherin as a tumour promoter and tumour marker, the molecular mechanisms governing its functions and distinguishing it from E-cadherin remain poorly understood.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; In this study, while investigating the functional distinction between E- and P-cadherin, we uncovered an essential role of P-cadherin in safeguarding mitotic fidelity in epithelial cells. \u0026nbsp;This is mediated by a delicate actin meshwork in the cytoplasmic interior, which is common in many cell types but requires optimized super-resolution imaging conditions for its visualization. Depletion of P-cadherin causes widespread mitotic spindle defects and severely reduces the interior actin meshwork. We showed that such interior actin is generated by P-cadherin-positive early endosomes through a Cdc42/MRCK\u0026beta;/FMNL2 pathway and serves to facilitate chromosome congression and support mitotic spindle organization. Interestingly, the disruption of interior actin upon P-cadherin depletion is accompanied by a significant amplification of mitochondria-associated actin, which results in actin shells encasing the mitochondria in conjunction with impaired mitochondrial morphology and functions, indicative of reciprocal regulation. The mitosis-supporting roles of the endosome-generated interior actin networks may represent a new paradigm for how the internal architecture of cells are dynamically organized during cell division to ensure proper mitotic fidelity.\u0026nbsp;\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eP-cadherin is essential for mitotic fidelity\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo dissect the differential functions of E- and P-cadherin, we used MDCK II epithelial cells, known to express both cadherins\u003csup\u003e40,41\u003c/sup\u003e (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 1a-b, h\u003c/strong\u003e). We first assessed their junction-forming ability by plating cells on substrates coated with F\u003csub\u003ec\u003c/sub\u003e-conjugates of the extracellular domains of E- or P-cadherin \u003csup\u003e42\u003c/sup\u003e. Wild-type (WT) cells spread and formed cadherin-based adhesions on both E- and P-cadherin-F\u003csub\u003ec\u003c/sub\u003e substrates (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 1c-d, f-g\u003c/strong\u003e). We next generated biallelic P-cadherin knock-out (KO) clones using CRISPR/Cas9 (\u003cstrong\u003eExtended\u003c/strong\u003e \u003cstrong\u003eData\u003c/strong\u003e \u003cstrong\u003eFig. 1h-j\u003c/strong\u003e) and compared them with previously described E-cadherin KO MDCK \u003csup\u003e43,44\u003c/sup\u003e. As expected, E-cadherin KO cells failed to attach to E-cad-F\u003csub\u003ec\u003c/sub\u003e coated substrates, while P-cad KO cells failed to attach to P-cadherin-F\u003csub\u003ec\u003c/sub\u003e coated substrates (\u003cstrong\u003eExtended\u003c/strong\u003e \u003cstrong\u003eData\u003c/strong\u003e \u003cstrong\u003eFig. 1e\u003c/strong\u003e), confirming that both cadherins are capable of forming only homophilic interaction. \u003c/p\u003e\n\u003cp\u003eWe next examined their differential localization. Endogenous E- and P-cadherin were both observed at the cell-cell contacts, but P-cadherin exhibited lower overall expression and a higher proportion of cytoplasmic puncta (\u003cstrong\u003eExtended\u003c/strong\u003e \u003cstrong\u003eData\u003c/strong\u003e \u003cstrong\u003eFig. 1a-b\u003c/strong\u003e). Notably, dividing cells displayed a significant increase in P-cadherin cytoplasmic puncta compared with neighboring non-dividing cells, with puncta densely clustered near the mitotic spindle poles (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 1a-c\u003c/strong\u003e). Immunoblot analysis of synchronized MDCK WT populations revealed that P-cadherin protein level increased during M phase, whereas E-cadherin levels remained constant (\u003cstrong\u003eExtended \u003c/strong\u003e\u003cstrong\u003eData\u003c/strong\u003e \u003cstrong\u003eFig. 1d\u003c/strong\u003e). Strikingly, confocal imaging of proliferating monolayers revealed a strong impairment of mitotic fidelity in P-cadherin KO cells compared with WT (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 1e-f)\u003c/strong\u003e. Observed defects included spindle mis-orientation (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 1h-k\u003c/strong\u003e, P-cad KO: 21.91 % vs WT: 3.26 %), chromosome misalignment (P-cad KO: 42.38 % vs WT: 12.56 %), spindle distortion (P-cad KO: 28.75 % vs WT: 2.68%), and multipolar spindles (P-cad KO: 3.01% vs WT: 0%). In contrast, the mitotic fidelity of E-cadherin KO cells was comparable to WT (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 1f, 2c-d\u003c/strong\u003e). These findings were validated using additional P-Cadherin KO clones, siRNA-mediated knock-down of P-cadherin in MDCK, and in other P-cadherin expressing cell lines, all of which exhibited comparable mitotic defects (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 1k-m, 2f-m\u003c/strong\u003e). Moreover, re-expression of P-cadherin-GFP in P-cadherin-KO cells restored mitotic fidelity to WT level (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 1f-g)\u003c/strong\u003e. Together, these results indicate that P-cadherin is essential for ensuring mitotic fidelity in multiple epithelial cell models. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMitotic spindles are nestled within delicate P-cadherin-dependent interior actin network\u003c/strong\u003e\u003cstrong\u003es\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConfocal imaging of dividing cells revealed that P-cadherin KO cells exhibited prominent actin-rich clumps in the vicinity of the mitotic spindles, in contrast to the relatively homogeneous actin intensity observed in MDCK WT cells (\u003cstrong\u003eExtended \u003c/strong\u003e\u003cstrong\u003eData\u003c/strong\u003e \u003cstrong\u003eF\u003c/strong\u003e\u003cstrong\u003eig. 2a\u003c/strong\u003e). Motivated by recent AiryScan and structured illumination super-resolution microscopy studies showing that a dynamic lattice of actin cables permeates the cytoplasmic interior of dividing HeLa cells and regulates mitochondrial inheritance\u003csup\u003e24,25\u003c/sup\u003e, we asked whether the actin perturbations caused by P-cadherin ablation might correspond to defects in the cytoplasmic actin network. Previous visualizations of actin filaments associated with spindles or centrosomes have often relied on glutaraldehyde fixation, suggesting that additional stabilization is required to preserve these structures\u003csup\u003e17,25,45\u003c/sup\u003e. To preserve epitopes for immunostaining while maintaining cytoplasmic actin integrity, we optimized a paraformaldehyde-based fixation protocol (\u003cstrong\u003eExtended\u003c/strong\u003e \u003cstrong\u003eData\u003c/strong\u003e \u003cstrong\u003eFig. 2b\u003c/strong\u003e). In particular, we found it essential to minimize mechanical disturbance during fixation, maintain the specimen at 37 \u0026deg;C, and perform imaging shortly after fixation. Under these conditions, delicate meshes of cytoplasmic actin filaments became apparent using multi-channel spinning-disc confocal microscopy coupled with ~120-nm super-resolution enhancement\u003csup\u003e5,46\u003c/sup\u003e, whereas sub-optimal fixation or prolonged delays between fixation and imaging markedly reduced actin cable visibility (\u003cstrong\u003eExtended \u003c/strong\u003e\u003cstrong\u003eData\u003c/strong\u003e \u003cstrong\u003eF\u003c/strong\u003e\u003cstrong\u003eig. 2a\u003c/strong\u003e). These stringent requirements indicate that such interior actin networks are unusually labile, which may help explain why they have received limited attention in previous investigations despite being present in a wide range of commonly used cell types (\u003cstrong\u003eExtended \u003c/strong\u003e\u003cstrong\u003eData\u003c/strong\u003e\u003cstrong\u003eFig. 2f-m\u003c/strong\u003e) \u003csup\u003e47\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eUsing our optimized protocol, we observed an extensive mesh of actin cables extending throughout the cytoplasm at various mitotic stages (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 2a\u003c/strong\u003e). During early prophase and nuclear envelope breakdown (NEBD), the actin cables can be seen emanating from the vicinity of the spindle poles (\u003cstrong\u003eFig. 2b\u003c/strong\u003e). Such dense actin cable networks cradle the mitotic spindle, becoming highly enriched at both spindle poles during metaphase, and persisting through telophase and cytokinesis (\u003cstrong\u003eFig. 2a\u003c/strong\u003e). At our imaging resolution, the interior actin networks in MDCK cells appear to largely involve elongated actin cables, thus distinguishing it from the recently discovered peri-spindle actin based on Arp2/3-dependent branched networks that encase acentrosomal mitotic spindles in mouse preimplantation embryos\u003csup\u003e3\u003c/sup\u003e. \u003c/p\u003e\n\u003cp\u003eIn P-cadherin KO cells, we found that the interior actin networks were drastically depleted, although the actin cortex remained intact (\u003cstrong\u003eFig. 2d\u003c/strong\u003e). To quantify this effect, we developed an analysis approach to map interior actin density relative to the centrosomes and normalized against cortical actin (\u003cstrong\u003eFig. 2c, e, Extended Data\u003c/strong\u003e \u003cstrong\u003eFig. 3a-c\u003c/strong\u003e). This revealed nearly 50% reduction of the interior actin density in P-cadherin KO compared to WT (\u003cstrong\u003eFig. 2e, Extended Data\u003c/strong\u003e \u003cstrong\u003eFig. 2f-m\u003c/strong\u003e). Similar depletion was observed after siRNA-mediated knockdown (KD) of P-cadherin, ruling out long-term adaptation or off-target KO effects (\u003cstrong\u003eExtended Data Fig. 1k-m\u003c/strong\u003e). We also verified that the expression level of \u0026beta;-actin remains similar between WT and KO (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 1i\u003c/strong\u003e). KD of P-cadherin in other epithelial and cancer cell lines also depleted the interior actin network and impaired mitotic fidelity (\u003cstrong\u003eFig. 2f-m\u003c/strong\u003e). Conversely, re-expression of P-cadherin-GFP in P-cadherin KO cell restored the interior actin networks as well as mitotic spindle organization (\u003cstrong\u003eFig. 1f-k\u003c/strong\u003e). These results indicate that P-cadherin is essential for the integrity of the interior actin network, which is in turn implicated in mitotic fidelity regulation. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eP-cadherin-dependent interior actin network facilitates mitotic spindle functions \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eActin-based steric cages and nuclear membrane remnants have been proposed to physically constrain chromosome scattering, thereby promoting efficient chromosome capture and segregation during cell division\u003csup\u003e48-50\u003c/sup\u003e. To assess how the interior actin network contributes to proper mitosis, we next performed 4-color live-cell super-resolution imaging of WT and P-cadherin KO cells, to simultaneously track actin, microtubules, mitochondria, and chromosomes.In WT cells,actin cables were generated in the vicinity of the spindle poles immediately after nuclear envelope breakdown (NEBD) and continued into prometaphase, as the centrosomes moved toward each other (\u003cstrong\u003eFig. 3a\u003c/strong\u003e, \u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 3e\u003c/strong\u003e, \u003cstrong\u003eMovie 1\u003c/strong\u003e). In contrast, in P-cadherin KO cells, actin cable formation was diminished near the spindle but instead became enriched peripherally near mitochondria (\u003cstrong\u003eFig. 3b, Extended Data\u003c/strong\u003e \u003cstrong\u003eFig. 3e,\u003c/strong\u003e \u003cstrong\u003eMovies 2\u003c/strong\u003e, and further below). Particle Image Velocimetry (PIV)\u003csup\u003e51\u003c/sup\u003e analysis revealed that in WT cells actin cables moved at an average speed of ~3.3 \u0026micro;m/min immediately after NEBD, while slowing to ~2.5 \u0026micro;m/min within 120 s (\u003cstrong\u003eFig. 3c, e-f\u003c/strong\u003e). These dynamics coincided with inward spindle pole motion and chromosome congression (\u003cstrong\u003eFig. 3h-j\u003c/strong\u003e). On the contrary, in P-cadherin KO cells the average actin speed was significantly slower, decelerating within a narrower range of 2-2.5 \u0026micro;m/min (\u003cstrong\u003eFig. 3d-f\u003c/strong\u003e), while the total actin cable length in the nuclear region was significantly reduced compared to WT (\u003cstrong\u003eFig. 3k-l\u003c/strong\u003e). Quantification of chromosome motion at NEBD and at metaphase revealed a significant reduction in chromosome speed in P-cadherin KO cells \u003cstrong\u003e(Fig. 3g, Extended Data Fig. 3f\u003c/strong\u003e), while inward centrosome migration and chromosomes capture by spindle microtubules, as quantified by clustered chromosome areas, were delayed (\u003cstrong\u003eFig. 3i\u003c/strong\u003e). In conjunction with the depleted interior actin network, the central positioning of the metaphase spindles as well as the in-plane spindle orientation expected for mitotic epithelial cells were impaired in P-cadherin KO (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 3g-h\u003c/strong\u003e). Together, these results link the aberrant interior actin organization in P-cadherin KO cells to defects in multiple mitotic steps that collectively contribute to impaired mitotic fidelity. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLoss of P-cadherin enhances peri-mitochondrial actin and impairs mitochondrial morphology and functions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAlongside the depleted interior actin networks, P-cadherin KO cells also showed sporadic clusters of dense actin shells surrounding organelle-like structures (\u003cstrong\u003eFig. 2d, 3b\u003c/strong\u003e). Because cytoplasmic actin cables drive mitochondrial motion during HeLa mitosis\u003csup\u003e25\u003c/sup\u003e, we hypothesized that these structures may correspond to peri-mitochondrial actin. To test this, we made use of an actin-chromobody probe (AC-Mito) that preferentially labels peri-mitochondrial actin\u003csup\u003e52,53\u003c/sup\u003e. The AC-Mito probe was expressed at the same level in both WT and P-cadherin KO cells; a level that showed no detectable effect on mitochondrial morphology in WT. As shown in \u003cstrong\u003eFig. 4a-b \u003c/strong\u003e(see also \u003cstrong\u003eMovie 3\u003c/strong\u003e),we found that peri-mitochondrial actin was significantly increased in P-cadherin KO cells, seen as co-fluctuating with mitochondria in the kymograph (\u003cstrong\u003eFig. 4b\u003c/strong\u003e). Consistently, live-cell imaging of P-cadherin KO cells expressing LifeAct-mEmerald\u003csup\u003e54\u003c/sup\u003e revealed prominent halos around mitochondria, which were not observed in WT (\u003cstrong\u003eFig. 4c\u003c/strong\u003e). Furthermore, while the mitochondria in WT were largely excluded from the mitotic spindle zone during metaphase, in P-cadherin KO a significant fraction of mitochondria were found overlapping with the metaphase spindle (\u003cstrong\u003eFig. 4d-e\u003c/strong\u003e) indicative of defective mitotic organelle exclusion \u003csup\u003e55\u003c/sup\u003e. \u003c/p\u003e\n\u003cp\u003eDuring interphase in particular, we observed that P-cadherin KO cells exhibited fragmented and anisotropically distributed mitochondria, compared to WT (\u003cstrong\u003eFig. 4f\u003c/strong\u003e). Morphometric analysis showed that mitochondria in P-cadherin KO cells were significantly smaller, shorter, and more circular (\u003cstrong\u003eFig. 4g\u003c/strong\u003e). Concomitantly, P-cadherin KO cells contained a greater number of mitochondria but a lower overall mitochondria area coverage, indicating a more heterogeneous distribution and increased aggregation (\u003cstrong\u003eFig. 4g\u003c/strong\u003e). Live-cell tracking further revealed a marked reduction in mitochondrial motility in P-cadherin KO cells (\u003cstrong\u003eFig. 4k\u003c/strong\u003e). To assess functional consequences, we employed live-cell indicators for mitochondrial membrane potential (TMRM, \u003cstrong\u003eFig. 4l\u003c/strong\u003e) and mitochondrial superoxide (MitoSox, \u003cstrong\u003eFig. 4m\u003c/strong\u003e). P-cadherin KO cells displayed a significant decrease in membrane potential together with an increase in reactive oxygen species, indicative of impaired mitochondrial functions. Taken together, the increase in peri-mitochondrial actin shells coupled with the loss of interior actin networks suggests that P-cadherin is required to maintain a homeostatic balance between organelle-associated and cytoplasmic interior actin pools, evocative of the inter-network competition framework\u003csup\u003e56\u003c/sup\u003e. We propose that the depletion of P-cadherin disrupts a competing cytoplasmic actin polymerization pathway, thereby favouring mitochondrial actin polymerization\u003csup\u003e8\u003c/sup\u003e and driving the observed changes in mitochondrial morphology and function. Consistent with this model, overexpression of P-cadherin\u0026ndash;GFP in P-cadherin KO cells restored highly elongated mitochondrial morphologies (\u003cstrong\u003eFig. 4j\u003c/strong\u003e). Further in-depth studies will be required to delineate the molecular mechanisms and broader implications of these P-cadherin\u0026ndash;dependent mitochondrial phenotypes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eP-cadherin is trafficked in Rab5 endosomes in association with dynamic actin cables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo dissect the mechanism underlying mitotic functions of P-cadherin, we next performed live-cell imaging of fluorescently-tagged P-cadherin re-expressed in KO cells. This revealed that in dividing cells P-cadherin-GFP was localized in vesicles clustered near the centrosomes, as labeled by SiR-tubulin, and these vesicles appeared to be transported on the spindle microtubules (\u003cstrong\u003eFig. 5a-c\u003c/strong\u003e). Furthermore, in live-imaging of KO cells co-expressing P-cadherin and LifeAct, we observed actin cables trailing behind P-cadherin-positive vesicles in the cytoplasm of both interphase and mitotic cells (interphase, \u003cstrong\u003eFig. 5d: \u003c/strong\u003eP-cadherin-tdTomato and LifeAct-mEmerald, \u003cstrong\u003eMovie 4\u003c/strong\u003e; prometaphase, \u003cstrong\u003eFig. 5e\u003c/strong\u003e: P-cadherin-GFP and LifeAct-mScarlet-I\u003cstrong\u003e,\u003c/strong\u003e \u003cstrong\u003eMovie 5\u003c/strong\u003e). Altogether, these findings implicate P-cadherin-containing endosomal vesicles in generating the actin cables that are incorporated into interior actin networks during mitosis, which may in turn contribute to proper mitotic spindle functions described above.\u003c/p\u003e\n\u003cp\u003eWe next sought to identify the endosomal carriers of P-cadherin. Different Rab GTPases are known for distinct localization patterns during various stages of mitosis\u003csup\u003e57\u003c/sup\u003e. The enrichment of P-cadherin vesicles at the mitotic spindle poles (\u003cstrong\u003eFig. 5a\u003c/strong\u003e) is consistent with a subset of endosomal markers. Co-localization analysis of P-cadherin-GFP with mCherry fusions of candidate Rab GTPases (\u003cstrong\u003eFig. 5g\u003c/strong\u003e) revealed an extensive co-localization between P-cadherin and the early endosome marker Rab5 in both interphase and mitotic cells (\u003cstrong\u003eFig. 5i-j\u003c/strong\u003e). Immunostaining for endogenous P-cadherin and Rab5 further confirmed their colocalization (\u003cstrong\u003eFig. 5h\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCytoplasmic domain of P-cadherin determines trafficking in Rab5 endosomes and is essential for interior actin integrity and mitotic fidelity \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe requirement of P-cadherin, but not E-cadherin (\u003cstrong\u003eFig. 1f, Extended \u003c/strong\u003e\u003cstrong\u003eData\u003c/strong\u003e\u003cstrong\u003e Fig. 2d\u003c/strong\u003e), for mitotic fidelity is unexpected given their very close sequence similarity. To dissect this difference, we generated chimeric constructs swapping their cytoplasmic tails. In P-cadherin KO cells, expression of ecto-E-cad::cyto-P-cad-GFP rescued mitotic fidelity, whereas ecto-P-cad::cyto-E-cad-GFP did not (\u003cstrong\u003eFig. 5l\u003c/strong\u003e). Quantification confirmed that ecto-E-cad::cyto-P-cad-GFP significantly reduced chromosome misalignment compared to ecto-P-cad::cyto-E-cad-GFP (\u003cstrong\u003eFig. 5m\u003c/strong\u003e). Importantly, a dense interior actin network was restored by ecto-E-cad::cyto-P-cad-GFP but not by ecto-P-cad::cyto-E-cad-GFP (\u003cstrong\u003eFig. 5k\u003c/strong\u003e). Additionally, the ecto-E-cad::cyto-P-cad-GFP construct localized extensively to endosomal vesicles in the cytoplasmic interior, co-localizing with Rab5 similarly to P-cadherin-GFP, while ecto-P-cad::cyto-E-cad-GFP remained largely confined to cell-cell contacts (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 4b-c\u003c/strong\u003e). These observations identify the cytoplasmic tail of P-cadherin as the determinant of both its mitotic and interior actin regulatory functions, mediated through its endosomal localization rather than canonical cell-cell junctional roles.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEndocytosed population of P-cadherin contributes to interior actin network that drives chromosome congression\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe next asked whether the mitotic functions of P-cadherin require endocytic recycling. Cadherin cytoplasmic domains provide binding sites for catenins, with p120-catenin stabilizing plasma-membrane presentation by masking an endocytic motif\u003csup\u003e30,58\u003c/sup\u003e. Consistent with this, we found that \u0026beta;-catenin co-localized extensively with P-cadherin-GFP at both cell-cell contacts and endosomes in KO rescue cells (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 4d-f\u003c/strong\u003e), while p120-catenin co-localized with P-cadherin exclusively at cell-cell contacts and was absent from the endosomes (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 4g-h\u003c/strong\u003e). We next inhibited clathrin-mediated endocytosis using dynasore\u003csup\u003e59\u003c/sup\u003e. As shown in \u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 4i\u003c/strong\u003e, we found thatdynasore treatment gave rise to a severe depletion of interior actin network in association with defective spindle organization in WT as well as in P-cadherin KO cells rescued with P-cadherin-GFP or ecto-E-cad::cyto-P-cad-GFP. In conjunction, quantification of mitotic fidelity revealed a significant impairment (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 4j-k\u003c/strong\u003e). We interpret these outcomes as arising from the inhibition of P-cadherin endosomal internalization, supported by the severe depletion of GFP-positive vesicles upon dynasore treatment compared to control (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 4i\u003c/strong\u003e). \u003c/p\u003e\n\u003cp\u003eWe next combined pharmacological inhibition of endocytosis with cell cycle synchronization to test whether endosomal P-cadherin is required for the assembly of the interior actin network that supports chromosome congression. As shown in \u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 4l\u003c/strong\u003e, we synchronized LifeAct-mEmerald-expressing WT cells at G2/M using RO3306 treatment for 20 h. At 1 h prior to G2/M release, dynasore was added to inhibit clathrin-mediated endocytosis. Wash-out of RO3306 was performed in the presence of dynasore to enable mitotic progression in the presence of endocytosis blockade. Mitosis was allowed to progress for 1.5 h following G2/M release, followed by S-trityl-L-cysteine (STLC) treatment, at which point the cells became enriched at late prometaphase with depleted interior actin network. Then, dynasore wash-out was performed to initiate P-cadherin endocytosis, and live-cell imaging commenced at 30 min post-washout. As shown in \u003cstrong\u003eMovie 6\u003c/strong\u003e and \u003cstrong\u003eExtended Data Fig. 4l, \u003c/strong\u003ewe observed that the interior actin cables became more prominent over time, forming nest-like meshes around the clustering chromosomes. Quantification of chromosome area following the wash-out revealed a progressively decreasing trend, indicative of chromosome congression (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 4m\u003c/strong\u003e). Together, these results support a model in which P-cadherin internalization via clathrin-mediated endocytosis drives interior actin network generation that facilitates chromosome segregation and ensures mitotic fidelity (\u003cstrong\u003eFig. 5f\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMitotic interior actin network depends on Cdc42- and formin-dependent actin polymerization\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDuring mitosis, both the activity and centrosomal localization of the small GTPase Cdc42 peak at metaphase, reminiscent of our observed P-cadherin endosome distribution\u003csup\u003e60,61\u003c/sup\u003e. Therefore, we tested whether P-cadherin functions are coupled to Cdc42 activity. Using the mRFP-wGBD biosensor to probe for active Cdc42\u003csup\u003e62\u003c/sup\u003e, we observed its extensive co-localization with P-cadherin-GFP at endosomal vesicles in interphase cells and at the spindle poles of dividing cells, while no co-localization was observed at the cell-cell contacts (\u003cstrong\u003eFig. 6a-b, Extended Data Fig. 5a-b\u003c/strong\u003e). Further, when mRFP-wGBD is expressed in un-rescued P-cadherin KO, we observed that fluorescence intensity appears largely cytosolic with a significant reduction in the number of active Cdc42 vesicles compared to WT cells or P-cadherin KO rescued with P-cadherin-GFP (\u003cstrong\u003eFig. 6d, Extended Data\u003c/strong\u003e \u003cstrong\u003eFig. 5c-d\u003c/strong\u003e). Next, we pharmacologically inhibited Cdc42 activity using either ML141 or CASIN\u003csup\u003e63,64\u003c/sup\u003e. Inhibition of Cdc42 resulted in extensive disruption of the interior actin network integrity as well as defects in mitotic fidelity (\u003cstrong\u003eFig. 6e-g\u003c/strong\u003e). Additionally, live-cell imaging of WT cells expressing LifeAct-mEmerald and mRFP-wGBD revealed transient actin enrichment associated with the active Cdc42 endosome (\u003cstrong\u003eFig. 6c, Extended Data\u003c/strong\u003e \u003cstrong\u003eFig. 5e)\u003c/strong\u003e, resembling actin cables generated by P-cadherin vesicles (\u003cstrong\u003eFig. 5d-e\u003c/strong\u003e). In contrast, in KO cells actin association becomes strongly diminished in the few remaining active Cdc42 endosomes (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 5f\u003c/strong\u003e). Taken together, these findings indicate that P-cadherin endosomes serve as platforms for Cdc42 activity and implicate Cdc42 as a key effector controlling the endosome-generated interior actin network.\u003c/p\u003e\n\u003cp\u003eSince Cdc42 regulates actin via both Arp2/3 and formins\u003csup\u003e65\u003c/sup\u003e, we tested which pathway contributed to the polymerization of the endosome-generated interior actin network. As shown in \u003cstrong\u003eFig.\u003c/strong\u003e \u003cstrong\u003e6h-j\u003c/strong\u003e, inhibition of formins by SMIFH2 significantly depleted interior actin network and impaired mitotic fidelity in WT cells, whereas Arp2/3 inhibition with CK666 increased interior actin density, likely reflecting the competitive balance between the actin polymerization machineries \u003csup\u003e56\u003c/sup\u003e, but did not compromise mitosis (\u003cstrong\u003eFig. 5a-c\u003c/strong\u003e). We also observed that CK666 treatment of P-cadherin KO cells also increased the interior actin network density compared to DMSO control. However, the CK666-induced increase in actin density still remains below WT level and mitotic fidelity is not fully restored (\u003cstrong\u003eExtended Data Fig. 5g-h\u003c/strong\u003e). Altogether, these results implicate formin as the primary actin polymerase assembling the interior actin network in MDCK, in contrast to the Arp2/3-based networks that generated the peri-spindle actin structures in mouse preimplantation embryos or centrosomal actin in non-epithelial cells\u003csup\u003e3,17,25\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eP-cadherin mediates interior actin network assembly via Cdc42-MRCK\u0026beta;-FMNL2 pathway\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo pinpoint the relevant Cdc42-dependent formin\u003csup\u003e66\u003c/sup\u003e and given the known off-target effect of SMIFH2\u003csup\u003e67\u003c/sup\u003e, we next made use of siRNA knockdown. This identified FMNL2, but not FMNL3, as being essential for both interior actin integrity and mitotic fidelity, as its KD phenocopied P-cadherin KD (\u003cstrong\u003eFig. 6k-m\u003c/strong\u003e). Live-cell imaging showed that FMNL2-EGFP was localized at cell-cell contacts and on endosomal vesicles in WT cells and P-cadherin KO rescued by P-cadherin-tdTomato (\u003cstrong\u003eExtended Data Fig. 5j\u003c/strong\u003e), In dividing cells, endosomal FMNL2 can readily be seen localized to spindle poles (\u003cstrong\u003eExtended Data Fig. 5l\u003c/strong\u003e) and co-localized with actin cables (\u003cstrong\u003eExtended Data Fig. 5k\u003c/strong\u003e). In contrast, in P-cadherin KO cells, FMNL2 lost its endosomal localization, becoming largely diffuse in the cytosol (\u003cstrong\u003eExtended Data Fig. 5m\u003c/strong\u003e), suggesting that FMNL2 requires P-cadherin for endosomal targeting. FMNL2 is auto-inhibited via the DAD domain\u003csup\u003e68,69\u003c/sup\u003e. Indeed, the expression of a constitutively active FMNL2 mutant (\u0026Delta;DAD) in P-cadherin KO cells restored endosomal localization as well as interior actin networks, along with partially rescuing mitotic fidelity. In contrast, P-cadherin KO expressing FMNL2-EGFP still exhibited significant interior actin depletion as well as mitotic defects (\u003cstrong\u003eExtended Data Fig. 5m-n, 7a-c\u003c/strong\u003e). Thus, FMNL2 emerges as the primary actin cable generator downstream of P-cadherin and Cdc42, whose activation and endosome localization in turn depend on P-cadherin. \u003c/p\u003e\n\u003cp\u003eTo further delineate signaling partners, we performed proximity biotinylation followed by proteomics based on TurboID fused to the cytoplasmic tails of either E- or P-cadherin\u003csup\u003e70\u003c/sup\u003e (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 6\u003c/strong\u003e). The TurboID constructs were expressed in the human cell line A431D, which is devoid of classical cadherins\u003csup\u003e71\u003c/sup\u003e, and in which P-cadherin expression affects interior actin similarly to MDCK cells (\u003cstrong\u003eExtended Data\u003c/strong\u003e \u003cstrong\u003eFig. 6a\u003c/strong\u003e). Candidate proteins enriched in the P-cad-TurboID sample relative to the E-cad-TurboID sample were identified and classified based on STRING classification\u003csup\u003e72\u003c/sup\u003e. Importantly, P-cad-TurboID-enriched proteins included endosomal markers such as Rab5b and various mitotic spindle proteins (\u003cstrong\u003eSupplementary Table 1, Extended Data\u003c/strong\u003e \u003cstrong\u003eFig. 6d-h\u003c/strong\u003e). Validation of top candidates was performed in MDCK cells by assessing mitotic fidelity and interior actin integrity following siRNA-mediated KD (\u003cstrong\u003eFig. 6n\u003c/strong\u003e). We found that KD of MRCK\u0026beta; (gene name: CDC42BPB), a Cdc42-dependent kinase with known association to FMNL2\u003csup\u003e73\u003c/sup\u003e, significantly impaired mitotic fidelity and depleted interior actin, phenocopying the KD of P-cadherin or FMNL2 (\u003cstrong\u003eFig. 6k-n\u003c/strong\u003e). \u003c/p\u003e\n\u003cp\u003eWe next validated the subcellular localization of MRCK\u0026beta;, observing that in WT cells its mScarlet3 fusion\u003csup\u003e74\u003c/sup\u003e localized prominently to cell-cell junctions as well as endosomal vesicles clustered near centrosomes in dividing cells\u003cstrong\u003e (Fig. 7a-d, Extended Data\u003c/strong\u003e \u003cstrong\u003eFig. 7e\u003c/strong\u003e). Likewise, MRCK\u0026beta; is found to co-localize extensively at the endosomes with P-cadherin (\u003cstrong\u003eFig. 7e-f\u003c/strong\u003e), FMNL2 (\u003cstrong\u003eFig. 7g\u003c/strong\u003e), and Rab5 (\u003cstrong\u003eFig. 7h\u003c/strong\u003e). In contrast, in P-cadherin KO cells MRCK\u0026beta; localization at cell-cell junctions and endosomes was largely abrogated, becoming diffuse and cytosolic (\u003cstrong\u003eFig. 7a-d\u003c/strong\u003e), indicating that its subcellular localization depends on P-cadherin. In turn, we found that SiRNA-mediated KD of MRCK\u0026beta; in WT cells severely abrogate FMNL2 endosomal localization (\u003cstrong\u003eExtended Data Fig. 7d, f\u003c/strong\u003e), thus implicating its role in potentiating FMNL2 function at the endosomes. Corroborating this, pharmacological inhibition of MRCK\u0026beta; using BDP9066\u003csup\u003e75\u003c/sup\u003e disrupted the interior actin networks while also impairing mitotic fidelity (\u003cstrong\u003eFig. 7i-k\u003c/strong\u003e), albeit with the caveat that this inhibitor also affects MRCK\u0026alpha;. Altogether, our results define a core molecular mechanism whereby endocytosed P-cadherin recruits an actin-generation machinery consisting of Cdc42, MRCK\u0026beta;, and FMNL2 to Rab5-positive endosomes to assemble the interior actin network that organizes intracellular architecture and safeguards mitotic fidelity. \u003c/p\u003e\n"},{"header":"Discussion","content":"\u003cp\u003eAccurate chromosome segregation involves a tightly choreographed remodelling of the sub-cellular cytoskeletal architecture. Studies across diverse tissues and cell types have underscored the plasticity of the cell division machinery\u003csup\u003e19\u003c/sup\u003e. In most somatic cells, spindle assembly is thought to depend primarily on centrosomes, whereas in cells lacking centrosomes, such as oocytes or mouse pre-implantation embryos, both actin and microtubules collaborate in spindle organization\u003csup\u003e1,3,4,48\u003c/sup\u003e. Although microtubules have long been considered the canonical driver of centrosomal mitosis, several studies \u0026ndash; especially those optimizing actin visualization \u0026ndash; have reported actin involvement\u003csup\u003e17,45,47\u003c/sup\u003e, suggesting that the role of actin in centrosomal mitosis may be broader than previously assumed.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHere, through the optimization of preservation and observation methods, we show that P-cadherin-dependent endosome-generated interior actin networks are essential for centrosomal mitosis in common epithelial cell types. Moreover, such actin network influences mitochondrial morphology through competition with the peri-mitochondrial actin machinery. Consequently, loss of P-cadherin affects not only mitotic fidelity but also gives rise to excess actin encapsulation of mitochondria, leading to encapsulation of mitochondria, their fragmentation, and defective distribution in cells.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMechanistically, our data reveal an endosome-based pathway that ensures proper chromosome segregation (\u003cstrong\u003eFig. 7l, Extended\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eData\u003c/strong\u003e \u003cstrong\u003eFig. 7h-i\u003c/strong\u003e). Rather than being degraded by Hakai after endocytosis, as established for E-cadherin\u003csup\u003e76\u003c/sup\u003e, internalized P-cadherin is trafficked in Rab5-positive endosomes, where it recruits Cdc42, MRCK\u0026beta;, and FMNL2 to polymerize actin cables. As P-cadherin-containing endosomes are accumulated in the centrosomal vicinity, they polymerize actin into an interwoven network that extends throughout the cytoplasmic interior. This meshwork then serves as a physical constraint limiting chromosome scattering after NEBD, promoting timely capture by kinetochore microtubules, while the dense mass of actin meshes at each spindle pole may also contribute to the stabilization of the mitotic spindle position and orientation (\u003cstrong\u003eExtended\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eData\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eFig. 7h\u003c/strong\u003e). Accordingly, P-cadherin loss drastically reduces the density of the interior actin network, thereby reducing chromosome congression, prolonging chromosome scattering, and destabilizing the spindle positioning. This role is reminiscent of actin functions in oocyte meiosis\u003csup\u003e48\u003c/sup\u003e, where FMNL2 depletion disrupts spindle and endomembrane organization\u003csup\u003e77\u003c/sup\u003e. Because both E- and P-cadherin are expressed in oocytes\u003csup\u003e78\u003c/sup\u003e, it will be interesting to test whether the P-cadherin\u0026ndash;dependent endosomal mechanism may also operate there. Conversely, the formin-dependent mechanism described here is distinct from the Arp2/3- and myosin-X-dependent actin remodelling and contraction involved in acentrosomal mitosis of mouse preimplantation embryos\u003csup\u003e3\u003c/sup\u003e. Collectively, these comparisons suggest that actin-dependent contributions to mitosis are more diverse than previously appreciated. While validated across several epithelial and carcinoma cell lines, the \u003cem\u003ein vivo\u003c/em\u003e significance of the P-cadherin-based mechanism remains to be established. Although P-cadherin expression is pronounced in embryonic tissues\u003csup\u003e39\u003c/sup\u003e , its ablation results in phenotypes far less severe than the embryonic lethality associated with E-cadherin loss\u003csup\u003e38\u003c/sup\u003e. Whether and how interior actin networks are regulated in cells lacking P-cadherin, for example through alternative endosomal scaffolds or distinct regulatory mechanisms remain to be investigated.\u003c/p\u003e\n\u003cp\u003eOur data also indicates that P-cadherin-dependent interior actin networks persist in interphase, balancing different organelle-associated actin pools. Moore et al. described dynamic cytoplasmic actin cables that undergo wave-like motion in HeLa cells\u003csup\u003e24,25\u003c/sup\u003e, which we did not observed in MDCK, consistent with cell-type specific differences in actin regulators. Nevertheless, in both cell types the integrity of the cytoplasmic actin networks were linked to mitochondrial morphology. Since mitochondrial fragmentation is associated with metabolic dysfunction\u003csup\u003e79\u003c/sup\u003e, future work should clarify whether actin monomer availability, shared nucleators, or signalling feedback underlies the reciprocal relationship between P-cadherin endosomes and mitochondria, and how it influences mitochondrial inheritance and function.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study also highlights a previously unrecognized non-junctional role of P-cadherin. While classical cadherins are best known for stabilizing cell\u0026ndash;cell junctions and transmitting mechanical cues through cortical actin, we show that P-cadherin, through its cytoplasmic domain, also serves as a structural and signalling platform on Rab5-positive endosomes, driving the assembly of endosome-generated interior actin network. Endosomal signalling has been described for integrins and growth factor receptors\u003csup\u003e80,81\u003c/sup\u003e, and cadherin-independent signalling of junctional proteins such as \u0026alpha;-catenin and \u0026beta;-catenin from Rab11 endosomes has been reported\u003csup\u003e82\u003c/sup\u003e. However, to the best of our knowledge, cadherin endosomal signalling has not been reported previously.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eE-cadherin and P-cadherin have a high degree of sequence similarity (68% identity for ectodomain, ~95% identity for cytoplasmic domain) and are co-expressed in a broad range of epithelial tissues\u003csup\u003e83\u003c/sup\u003e. The E- versus P-cadherin specialization may reflect unique mechanical constraints of epithelial mitosis, where cells must preserve junctional integrity and tissue topology while simultaneously segregating chromosomes. We propose that E-cadherin primarily organizes cortical and junctional actin, whereas P-cadherin, when internalized and stabilized, regulates internal actin to support the spindle and surrounding organelles. Consistent with this, the fate of endocytosed E-cadherin and P-cadherin appear to drastically diverge, with internalized P-cadherin preferentially stabilized in a cell-cycle dependent manner. Notably, endocytosis of P-cadherin is required for interior actin integrity (\u003cstrong\u003eExtended\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eData\u003c/strong\u003e \u003cstrong\u003eFig. 4i-m\u003c/strong\u003e), suggesting that additional layers of regulation via junctional turnover or cadherin cleavage may also be operational. The unique molecular features of the P-cadherin cytoplasmic domain underlying its mitotic function remain to be identified.\u003c/p\u003e\n\u003cp\u003eFinally, P-cadherin depletion affects multiple aspects of spindle organization, including spindle distortion and multipolarity, implying additional contributions to microtubule regulation. Likely, endosomal P-cadherin, in addition to regulating actin, also helps localize factors that regulate the microtubule-based machinery. These can be identified by further analysis of the differential proximity proteomics data. More broadly, the pleiotropic effects of P-cadherin on fundamental cellular processes such as mitosis and organelle homeostasis raise the possibility that its dysregulation contributes to aneuploidy or metabolic dysfunctions. P-cadherin is druggable\u003csup\u003e84\u003c/sup\u003e. Given the correlations between P-cadherin dysregulation and multiple cancers\u003csup\u003e6\u003c/sup\u003e, further investigation of these mechanisms may provide valuable insights into disease progression.\u0026nbsp;\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eMammalian cell culture\u003c/p\u003e\n\u003cp\u003eMDCK II (Madin-Darby canine kidney II) used were originally obtained from Sigma-Aldrich (ECACC, Cat. No. 85011435) and gifted by Huang Cheng-Kuang (Mechanobiology Institute, Singapore). E-cadherin KO MDCK II is described previously and gifted by Rene-Marc M\u0026eacute;ge, Institut Jacques Monod, France. Cadherin-null A431D (human epidermoid carcinoma) cells is a gift from Ada Cavalcanti-Adam (Max Planck Institute for Medical Research). MCF10A (non-malignant human breast epithelial) cells were gifted by Chwee Teck Lim (Mechanobiology Institute, Singapore). HaCaT (spontaneously immortalized human keratinocyte), HK-2 (human proximal tubular epithelial) and Caco-2 (human colon adenocarcinoma) cells were gifted by Jennifer Young, Yusuke Toyama and Selwin Wu (Mechanobiology Institute, Singapore), respectively. A431 cells were purchased from the American Type Culture Collection (ATCC, Cat. No. CRL-1555). U2OS (human osteosarcoma) was gifted by Clare Waterman (National Institutes of Health). ARPE-19 (immortalized human retinal pigment epithelial) cells was gifted by Xinyi Su, Institute of Molecular and Cellular Biology A*STAR, Singapore. All cells were maintained in an incubator at 37\u0026deg;C in a humidified atmosphere containing 5% CO\u003csub\u003e2\u003c/sub\u003e and verified to be free from mycoplasma infection, and tested monthly.\u003c/p\u003e\n\u003cp\u003eMDCK II, A431, A431D, U2OS and HaCaT cells were cultivated in Dulbecco\u0026rsquo;s modified Eagle\u0026rsquo;s medium (DMEM, Gibco\u0026trade;, Cat. No. 10566016) containing 10% (v/v) fetal bovine serum (FBS, Gibco\u0026trade;, Cat. No. 10437028), 1% (v/v) Pen/Strep (Gibco\u0026trade;, Cat. No. 15070063), 1% (v/v) sodium pyruvate (Gibco\u0026trade;, Cat. No. 11360070) and passaged using 0.25% (w/v) trypsin-EDTA solution (Gibco\u0026trade;, Cat. No. 25200072) every 2-3 days. HK-2 cells were grown in DMEM/F12 (Gibco\u0026trade;, Cat. No. 11330032) supplemented with 10% (v/v) FBS, 1% Pen/Strep, 1% (v/v) sodium pyruvate, and split with 0.05% trypsin-EDTA solution (Gibco\u0026trade;, Cat. No. 25300062) every 3-4 days. Caco-2 cells were maintained in DMEM containing 10% (v/v) FBS, 1% (v/v) Pen/Strep, 1% (v/v) sodium pyruvate and 1% (v/v) MEM non-essential amino acids (MEM NEAA, Thermo Fisher, Cat. No. 11140050). Cells were passaged using 0.05% (w/v) trypsin-EDTA solution every 3-4 days. MCF10A cells were maintained in DMEM/F12 supplemented with 5% (v/v) horse serum (Invitrogen, Cat. No. 16050-122), 20 ng/ml EGF (Thermo Fisher, Cat. No. AF-100-15), 0.5 mg/ml hydrocortisone (Sigma-Aldrich, Cat. No. H0888), 100 ng/ml cholera toxin (Sigma-Aldrich, Cat. No. C8052), 10 \u0026micro;g/ml insulin (Sigma-Aldrich, Cat. No. I0516), 1% (v/v) Pen/Strep and passaged using 0.05% trypsin-EDTA solution every 3-4 days. ARPE-19 cells were cultured in minimum essential medium, alpha modification (MEM-\u0026alpha;, Sigma-Aldrich, Cat. No. M4526) containing 10% (v/v) FBS, 1% (v/v) Pen/Strep, 1% GlutaMAX (Thermo Fisher, Cat. No. 35050079), 1% (v/v) MEM NEAA, 1% (v/v) N1 medium supplement (Sigma-Aldrich, Cat. No. N6530), 10 mM Nicotinamide (Sigma-Aldrich, Cat. No. N5535), 0.25 mg/ml Taurine (Sigma-Aldrich, Cat. No. T0625), 20 ng/ml hydrocortisone and 0.013 ng/ml 3,3\u0026prime;,5-Triiodo-L-thyronine sodium salt (Sigma-Aldrich, Cat. No. T5516). Cells were split using TrypLE Express (Thermo Fisher, Cat. No. 12604013) every 4-5 days.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003e \u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eCRISPR/Cas9-mediated knockout (KO) of P-cadherin\u003c/p\u003e\n\u003cp\u003eP-cadherin KO MDCK cells were generated by CRISPR/Cas9-mediated genomic editing using dual guide RNAs to induce exon skipping, including sgRNA-A (5\u0026apos;-ACCCGAAAAT CGTGATCGTG-3\u0026apos;) and sgRNA-B (5\u0026apos;-CGATAGAGGC GCTAAAGAGT-3\u0026apos;), targeting exon 2 and downstream of exon 4 of dog CDH3 gene, respectively. The oligo sequences were designed using the CHOPCHOP web tool (https://chopchop.cbu.uib.no). Oligonucleotides were synthesized by Integrated DNA Technologies. Each guide sgRNA sequence was separately introduced into the pSpCas9(BB)-2A-Puro (PX459) V2.0 plasmid backbone, which was a gift from Feng Zhang (Addgene # 62988). The vector contains Cas9 gene with BbsI cloning site under the control of U6 promoter. The hU6-F primer (5\u0026apos;-GAGGGCCTATTTCCCATGATT-3\u0026apos;) was used for sequencing to confirm the presence of sgRNA oligos after cloning (Axil Scientific Sequencing).\u003c/p\u003e\n\n\u003cp\u003esiRNA-mediated knockdown\u003c/p\u003e\n\u003cp\u003eFor negative siRNA control, a non-targeting siRNA (Invitrogen\u0026trade;, Cat. No. 4390846) was used, P-cadherin (siRNA #10694 and #S2774), RAB5B (siRNA #s11683 and #s11681), KIF15 (siRNA #s32546 and #s32548), ECT2 (siRNA #146755 and #S4444), profilin-2 (PFN2, siRNA #S10380), FMNL2 (siRNA #128873 and #128874), FMNL3 (siRNA #131854 and #HSS132081), and MRCK\u0026beta; (CDC42BPB, siRNA #748 and #S18402) were purchased from Thermo Fisher. \u003c/p\u003e\n\u003cp\u003eStock solutions of each siRNA (25 \u0026micro;M) were generated by dissolving siRNA powder in RNase-free water and stored at -20\u0026deg;C until usage. Two siRNA oligonucleotides that target distinct regions of P-cadherin, RAB5B, KIF15, ECT2, FMNL2, FMNL3, or MRCK\u0026beta; transcripts were combined to prepare siRNA mixture using for electroporation. Sequences of the siRNA oligonucleotides were verified to be capable of binding fully complementary regions in mRNA transcripts of both human and dog targets.\u003c/p\u003e\n\u003cp\u003eFor knock-down experiments, cells were maintained in growth media without antibiotics overnight to reach 60-70% confluency. Then, 5x10\u003csup\u003e5\u003c/sup\u003e-7.5x10\u003csup\u003e5\u003c/sup\u003e cells were suspended and electroporated with siRNAs (30 pmol for MDCK II or 750 pmol for other cells) using Neon\u0026trade; Transfection System (100 \u0026mu;l Kit, Thermo Fisher, Cat. No. MPK10096) with pulse rate of 1650V, 20 ms pulse width and 1 pulse. Transfected cells were transferred into 27 mm glass bottom dish containing warm medium and grown for 48 h before performing immunofluorescence staining or Western blotting.\u003c/p\u003e\n\n\u003cp\u003eE- and P-cadherin constructs \u003c/p\u003e\n\u003cp\u003eP-cadherin-tdTomato, P-cadherin chimera-GFP (ecto-P-cad::cyto-E-cad) and E-cadherin chimera-GFP (ecto-E-cad::cyto-P-cad) were generated as follows. UniProtKB and NCBI databases were used to annotate the DNA sequence and plan the precise swapping of domains in SnapGene software (San Diego, CA). Plasmids containing cDNA of E-cadherin GFP (Addgene plasmid #28009) and P-cadherin (Addgene plasmid #47502) were obtained from Addgene repository. The pcDNA3.1 vector backbone from E-cadherin GFP plasmid was used as the backbone for further cloning. Primers for amplification of DNA fragments via Polymerase chain reaction (PCR) contained an identical 5\u0026rsquo; end to an adjacent segment and a 3\u0026rsquo; end that anneals to the target sequence were ordered from Integrated DNA technology (IDT). \u003c/p\u003e\n\u003cp\u003eThe following primers were used for construct assembly: P-cad tdTomato insert, forward 5\u0026prime;-GGTGGCGGGG AGGACGACTC CGGACTCAGA GTGAGCAAGG GCGAGGAGGTC-3\u0026prime; and reverse 5\u0026prime;-TGACACTATA GAATAGGGCC CTCTAGACTA CTTGTACAGC TCGTCCATGCCG-3\u0026prime;; P-cad tdTomato vector, forward 5\u0026prime;-CTGTACGGCA TGGACGAGCT GTACAAGTAG TCTAGAGGGC CCTATTCTAT AGTGTC-3\u0026prime; and reverse 5\u0026prime;-CTCCTCGCCC TTGCTCACTC TGAGTCCGGA GTCGTCCTCC CCGCC-3\u0026prime;; P-cad extracellular domain insert, forward 5\u0026prime;-CCTGCCCTCG CTCGGCGTCC CCGGCCAGCC ATGGGGCTCC CTCGTGGAC-3\u0026prime; and reverse 5\u0026prime;-AGCAAGAATT CCTCCAAGAA TCCCCAGAAT ACCTCCCTTC CAGGGTCCAG-3\u0026prime;; P-cad extracellular domain vector, forward 5\u0026prime;-GAAACCTGCC CTGGACCCTG GAAGGGAGGT ATTCTGGGGA TTCTTGGAGG-3\u0026prime; and reverse 5\u0026prime;-AGACGCGAGA GGTCCACGAG GGAGCCCCAT GGCTGGCCGG GGACGCC-3\u0026prime;; E-cad extracellular domain insert, forward 5\u0026prime;-CCTGTCGAAG CAGGATTGCA AATTCCTGCC TTCATCCTCC CTGTGCTGGGG-3\u0026prime; and reverse 5\u0026prime;-CACCCCGGTG AACAGCTCCT CCCCCTTGCT GTCGTCCTCC CCGCCAC-3\u0026prime;; E-cad extracellular domain vector, forward 5\u0026prime;-GCAGACATGT ACGGTGGCGG GGAGGACGAC AGCAAGGGGG AGGAGCTG-3\u0026prime; and reverse 5\u0026prime;-CAGGACAGCCC CCAGCACAGG GAGGATGAAG GCAGGAATTT GCAATCC-3\u0026prime;.\u003c/p\u003e\n\u003cp\u003eDNA electrophoresis was performed for PCR products in order to separate them by length in 1% agarose gel in TAE buffer (2 M Tris-HCl, 0.6% acetic acid, 50 mM EDTA pH 8.0) to ensure the accuracy of previous steps. After purification of PCR products using Gel and PCR clean-up kit (Macherey-Nagel), fragments were combined in Gibson assembly reaction using the Gibson Cloning Master Mix (New England Biolabs), which included three enzymes: T5 Exonuclease, Phusion DNA Polymerase, and Taq DNA Ligase. The Gibson reaction was then transformed into XL-1 Blue competent bacteria and the correct clones was selected by PCR. E-Cad-TurboID, P-Cad-TurboID, and TurboID sequences were cloned into the pLL3.7 lentiviral vector backbone (Addgene plasmid #11795).\u003c/p\u003e\n\n\u003cp\u003eExpression vectors\u003c/p\u003e\n\u003cp\u003emRFP-wGBD plasmid was obtained from Addgene (plasmid # 26733 and 26734, gifted by William Bement). mEmerald-Lifeact-7 (Addgene plasmid # 54148) and Histone H2B-mCherry (Addgene #20972) were gifts from Michael Davidson (The Florida State University). mCherry-Rab5b was obtained from Addgene (plasmid # 49201, deposited by Gia Voeltz). Genetically-encoded mitochondria actin probe (AC-Mito), pSIN CMV AC-GFP-Fis1, was as described previously by Uri Manor (UC San Diego)\u003csup\u003e52\u003c/sup\u003e. FMNL2-Delta-DAD-EGFP and FMNL2-GFP plasmids were developed in the laboratory of Klemes Rottner\u003csup\u003e69\u003c/sup\u003e and kindly provided by Maddy Parsons (King\u0026rsquo;s College London). MRCK-\u0026beta;-mScarlet3 was synthesized by GentleGen Inc. (China). iRFP-FRB-Rab5 was obtained from Addgene (plasmid # 51612, deposited by Tamas Balla). LifeAct-mScarlet-I was obtained from Addgene (plasmid # 85056, deposited by Dorus Gadella). Expression vector for human-P-cadherin-GFPspark, referred to as P-cadherin-GFP for short, was purchased from Sino Biological (Cat. No. HG10305-ACG).\u003c/p\u003e\n\u003cp\u003eTransfection was performed using 5x10\u003csup\u003e5\u003c/sup\u003e suspended cells per reaction and 0.5-2 \u0026micro;g plasmids using Neon\u0026trade; Transfection System (100 \u0026mu;l, Thermo Fisher). Transfected cells were transferred into 12 mm or 27 mm glass bottom dish containing warm medium and allowed to grow for 24-48 h before performing immunofluorescence staining or live-cell imaging. When necessary, NucBlue\u0026trade; (Hoechst 33342, Thermo Fisher) reagent was used as nuclear counterstain prior to live-cell imaging.\u003c/p\u003e\n\n\u003cp\u003eStable cell line generation\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGeneration of P-cadherin KO MDCK cells\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMDCK cells (5x10\u003csup\u003e5\u003c/sup\u003e cells per transfection) was electroporated with 5 \u0026micro;g of either plasmids containing sgRNA-A (5\u0026apos;-ACCCGAAAATCGTGATCGTG-3\u0026apos;) or sgRNA-B (5\u0026apos;-CGATAGAG GCGCTAAAGAGT-3\u0026apos;), by utilizing a Neon\u0026trade; Transfection System as described above. The transfected cells were grown for 48 h before selecting using 2 \u0026micro;g/mL puromycin (Sigma-Aldrich, Cat. No. P8833). After puromycin selection, single cell clones were sorted into 96-well plates using the benchtop SH800S cell sorter (Sony). Three single KO clones were successfully verified by PCR, immunofluorescence staining and Western blotting.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGeneration of stable cell lines expressing fluorescently tagged proteins\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTransfection of plasmids with fluorescently tagged proteins was performed as described above. After maintaining transfected cells for 24-48 h, the stable cell lines were selected under antibiotics (Geneticin (G418) at 600 \u0026micro;g/ml or Hygromycin B at 200 \u0026micro;g/ml), according to the antibiotic resistance genes presented in plasmids for at least 2 weeks. Medium to high expression stable cell lines were sorted (SH800S cell sorter, Sony) to generate monoclonal populations, which were subsequently used for further experiments.\u003c/p\u003e\n\n\u003cp\u003ePreparation of biomimetic cadherin\u0026ndash;F\u003csub\u003ec\u003c/sub\u003e chimera substrate\u003c/p\u003e\n\u003cp\u003eCoverglasses (no. 1.5, 18 mm diameter) were washed in absolute ethanol overnight and rinsed three times with sterile water and PBS, respectively before exposure to UV in sterile laminar flow hood for 15 min. Sterile coverglasses were silanized by incubation with 3-glycidoxy-propyl-dimethoxymethylsilane (Sigma-Aldrich, Cat. No. 539252) (0.045% in 100% ethanol) for 1 h on a shaker at room temperature and then heated at 110\u0026deg;C for 1 h. Silanized substrates were rinsed with 70% ethanol and distilled water before drying with nitrogen gas. Subsequently, the substrates were then incubated with AffiniPure\u003csup\u003e\u0026reg;\u003c/sup\u003e goat anti-human IgG, Fc\u0026gamma; fragment specific (Jackson ImmunoResearch Inc., Cat. No. 109-005-008) at 1 \u0026micro;g/cm\u003csup\u003e2\u003c/sup\u003e in 0.1 M borate buffer, pH 8.0 and incubated at 4\u0026deg;C overnight in a humidity chamber. After rinsing with PBS, the substrates were neutralized by NaHCO\u003csub\u003e3\u003c/sub\u003e (100 mM, pH 8.3) and 2-(2-aminoethoxy)-ethanol (Sigma-Aldrich, Cat. No. A54059) for 1 h. The substrates were washed twice with PBS and then incubated with recombinant human P-cadherin F\u003csub\u003ec\u003c/sub\u003e (R\u0026amp;D system, Cat. No. 861-PC) or E-cadherin F\u003csub\u003ec\u003c/sub\u003e (R\u0026amp;D system, Cat. No. 648-EC) chimera protein in PBS containing Ca\u003csup\u003e2+\u003c/sup\u003e and Mg\u003csup\u003e2+\u003c/sup\u003e at 1 \u0026micro;g/cm\u003csup\u003e2\u003c/sup\u003e for 2 h. After rinsing, the substrates were blocked with 0.2 % pluronic acid (Sigma-Aldrich, Cat. No. P2443) in PBS for 20 min at room temperature. The planarized cadherin substrates were washed twice with PBS before cell seeding in serum free medium.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003e \u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eCell adhesion on planarized cadherin\u0026ndash;F\u003csub\u003ec\u003c/sub\u003e chimera substrat\u003cu\u003ee\u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eTo assay cadherin engagement, MDCK II WT, E-cadherin KO, or P-cadherin KO cells were dissociated with 0.05% Trypsin-EDTA solution and pelleted by centrifugation. Cell pellets were washed twice with serum-free DMEM medium. Single cell suspensions were prepared from the pellets in serum-free DMEM medium and then pre-incubated in 37\u0026deg; C tissue culture incubator for 30 min prior to seeding on the coated substrates. The cells on substrates were then fixed with fresh 4% (v/v) paraformaldehyde in PBS+ to preserve cell adhesion.\u003c/p\u003e\n\n\u003cp\u003eIdentification of P-cadherin-specific interactome by proximity biotinylation\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eProduction of lentiviral particles using pLL3.7 vector backbone\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eLentiviral particles were produced by transient transfection of HEK293 FT cells, as described previously. Briefly, cells were seeded at a density of 1-1.5x10\u003csup\u003e6\u003c/sup\u003e cells/well and cultivated in DMEM supplemented with 10% FBS, 1% Pen Strep, 1% L-Glutamine, 1% sodium pyruvate, 1% sodium bicarbonate). At 80% confluency, cells were replated on poly-L-lysine coated 6-well plates. Plasmids (E-Cad-TurboID, P-Cad-TurboID, and TurboID cloned into the pLL3.7 vector backbone) were co-transfected with envelop plasmid VSV-G (Invitrogen), and packaging plasmids pLP1 and pLP2 (Invitrogen) using PEI (Polyethylenimine, high molecular weight, water free, Sigma-Aldrich, Cat. No. 408727). On the next day, medium was switched with the growing medium (DMEM+10% FBS 1.5 ml/well) without any washing and cells were grown for another 30 h. After 30 h, supernatant was collected, and viral particles in the supernatant were concentrated at 1:100 by ultracentrifugation for 90 min at 25,000 rpm (SW28 rotor, Beckman Coulter) and resuspended in PBS. Titers of viral particles ranged between 10\u003csup\u003e5\u003c/sup\u003e and 10\u003csup\u003e6\u003c/sup\u003e multiplicity of infection.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGeneration of stable cell lines expressing Turbo ID, E-Cad fused to TurboID, and P-Cad fused to TurboID\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eCadherin-null A431D (human epidermoid carcinoma) cells were seeded at 3x10\u003csup\u003e5\u003c/sup\u003e cells/ml into 6-well plates. Respective lentiviral particles and 8 \u0026mu;g/ml polybrene (Sigma-Aldrich) were added to the culture and centrifuged for 90 min at 30\u0026deg;C. Supernatant was removed immediately after infection and replaced with DMEM medium containing 10% FBS in all experiments. The efficiency of transduction was approximately 70-80% as determined by immunostaining. The respective protein expression (Turbo ID, E-Cad-TurboID fusion gene, and P-Cad-TurboID fusion gene) was confirmed by Western blotting with an anti-FLAG antibody. For further validation, these cell lines were cultured in 24-well plates and subjected to biotinylation reaction (incubation with DMEM medium supplemented with 330 \u0026mu;M Biotin (Sigma-Aldrich, Cat. No. B4639)) for a duration of 10 minutes. Next, biotinylated cells were fixed using ice-cold methanol and then immunostained for the respective transgene and streptavidin signal.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSample preparation for Mass spectrometry \u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMonolayers of the three cell lines described above were grown in a 10 cm dish to more than 90% confluency, incubated with 330 \u0026micro;M biotin for 10 min, washed with PBS buffer thrice and then lysed with RIPA buffer. As a negative control, all three cell lines were also processed identically without Biotin. Cell lysates were quantified for the protein concentration using BCA assay (Thermo Fisher, Cat. No. 23225). An equal amount of cell lysate (700 \u0026micro;g) from each of the six samples was incubated with streptavidin magnetic beads (Thermo Fisher, Cat. No. 88817) overnight at 4\u0026deg;C and then eluted. The eluate obtained from all six samples was loaded on SDS-PAGE gel and subsequently imaged with Western blotting using streptavidin-HRP antibody (Thermo Fisher, Cat. No. N100). As expected, the samples incubated with biotin were stained intensely with streptavidin-HRP antibody whereas the samples incubated without biotin were stained weakly. This experiment was conducted in triplicate and the nine samples prepared for the MS analysis were generated in an identical fashion and sent to the mass spectrometry core facility at Technion \u0026ndash; Israel Institute of Technology.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eProteolysis \u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe proteins were brought to 8.5 M Urea, 100 mM ammonium bicarbonate (ABC), reduced with 2.8 mM DTT (60\u0026ordm;C for 30 min), modified with 8.8 mM iodoacetamide in 100 mM ABC in the dark at room temperature for 30 min, and digested in 1.5 M Urea, 25 mM ABC with modified trypsin (Promega) at a 1:50 enzyme-to-substrate ratio for overnight at 37\u0026deg;C.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMass spectrometry analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe tryptic peptides were desalted using C18 tips (Empore\u0026trade; C18 Solid Phase Extraction Disks (SPE): 2215), dried and re-suspended in 0.1% formic acid. The peptides were further cleaned from detergents using strong cation exchange tips (Empore\u0026trade; cation disk 2251). The peptides were resolved by reverse-phase chromatography on 0.075 X 180-mm fused silica capillaries (J\u0026amp;W) packed with Reprosil reversed phase material (Dr. Maisch GmbH, Germany). The peptides were eluted with linear 60 min gradient of 5% to 28% and 15 min gradient of 28% to 95% and 15 min at 95% acetonitrile with 0.1% formic acid in water at flow rates of 0.15 \u0026mu;l/min. Mass spectrometry was performed by Q Exactive\u0026trade; Plus Hybrid Quadrupole-Orbitrap\u0026trade; Mass Spectrometer (Thermo Fisher) in a positive mode using repetitively full MS scan followed by high collision dissociation (HCD) of the 10 most dominant ions selected from the first MS scan. \u003c/p\u003e\n\u003cp\u003eThe mass spectrometry data was analyzed using the MaxQuant software 1.5.2.8 for peak picking and identification using the Andromeda search engine, searching against the Homo sapiens proteome from the UniProtKB database with mass tolerance of 6 ppm for the precursor masses and the fragment ions. Oxidation on methionine, biotinylation and protein N-terminus acetylation were accepted as variable modifications and carbamidomethyl on cysteine was accepted as static modifications. Minimal peptide length was set to seven amino acids and a maximum of two mis-cleavages was allowed. The data was quantified by label free analysis using the same software. Peptide- and protein-level false discovery rates (FDRs) were filtered to 1% using the target-decoy strategy. Protein tables were filtered to eliminate the identifications from the reverse database, and common contaminants and single peptide identifications. Statistical analysis of the identification and quantization results was done using Perseus 1.6.7.0 software developed by Mathias Mann\u0026apos;s group. Mass Spectrometric analysis was performed at the Smoler Proteomics Center at the Technion-Israel Institute of Technology, Israel.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eProtein filtering and subcellular localization enrichment analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eA total of 3,766 proteins were identified from three independent biological replicates extracted from cells expressing TurboID, E-cadherin\u0026ndash;TurboID, or P-cadherin\u0026ndash;TurboID. Potential contaminants, ribosomal proteins, and proteins identified solely by site or by a single unique peptide were excluded from further analysis. Notably, CAV2, CDC42BPB, DDR1, FCHO2, PFN2, RAP1B, and VPS26B were retained despite being identified by a single peptide, owing to their specific enrichment, biological relevance, and unique association with the P-cadherin\u0026ndash;TurboID dataset. Proteins exhibiting a log₂ fold-change of less than 1 between TurboID and either P-cadherin\u0026ndash;TurboID or E-cadherin\u0026ndash;TurboID samples were additionally excluded from the final protein list. Subcellular localization enrichment analysis was evaluated using STRING version 12.0 (https://string-db.org/).\u003c/p\u003e\n\n\u003cp\u003eWestern blotting\u003c/p\u003e\n\u003cp\u003eCells cultured in a 6-well plate were extracted by scraping in 200 \u0026micro;L RIPA buffer (Thermo Fisher, Cat. No. 89900) containing 1X Protease Inhibitor Cocktail (Roche, Cat. No. 11836170001) followed by incubation on ice for 30 min. After high-speed centrifugation, protein concentrations of supernatants were determined by Bradford protein assay (BIO-RAD, Cat. No. 5000201). Equal amounts of proteins from each sample were resolved using SDS-PAGE procedure with 4\u0026ndash;15% Mini-PROTEAN\u003csup\u003e\u0026reg;\u003c/sup\u003e TGX\u0026trade; Precast Protein Gels (BIO-RAD, Cat. No. 4561084). Thereafter, proteins from gel were electroblotted using Trans-Blot Turbo System (Bio-Rad Laboratories). After washing with TBS and 15 min blocking in EveryBlot Blocking Buffer (Bio-Rad Laboratories), membranes were sequentially incubated with primary antibody (1:1000) in EveryBlot Blocking Buffer for overnight at 4\u0026ordm;C and secondary antibody conjugated to horseradish peroxidase (HRP) (1:2000, 1h at room temperature) after washing. After each incubation, membranes were washed 3 times in TBS containing 0.2% (v/v) Tween 20 (TBST), and final rinsed with TBS before developing with Chemiluminescence and detecting protein bands using iBright Imaging Systems (Thermo Fisher).\u003c/p\u003e\n\u003cp\u003e\u003cu\u003e \u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eCell cycle synchronization\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSynchronization at G1/S-border\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThymidine (Santa Cruz Biotechnology, Cat. No. sc-296542), was used to arrest cells at G1/S-border at a final concentration of 2 mM by incubating with cells for 24 h at 37\u0026deg;C.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSynchronization at G2/M-border\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eCells were arrested at G2/M-border by using a Cdk1 inhibitor, RO-3306, (Sigma-Aldrich, Cat. No. SML0569) at a final concentration of 10 \u0026mu;M for 20 h at 37\u0026deg;C. To release the cells from G2/M, RO-3306 was removed from the cells by 5 washes in warm media.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSynchronization at mitosis (prometaphase or metaphase)\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSTLC ((+)-S-trityl-L-cysteine) (Tocris Bioscience, Cat. No. 2191), which is an Eg5 (Kinesin-5) inhibitor, was used to arrest cells at metaphase. Cells were initially synchronized at G2/M-border by treating with 10 \u0026mu;M RO-3306 for 20 h at 37\u0026deg;C. Upon washout of RO-3306, STLC (50 \u0026mu;M) was added after RO-3306 removal for 1 h to stall at prometaphase or 1.5 h for metaphase. \u003c/p\u003e\n\u003cp\u003e\u003cu\u003e \u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eEndocytosis inhibition by dynasore treatment\u003c/p\u003e\n\u003cp\u003eFor endocytosis inhibition, MDCK II cells were treated with 80\u0026thinsp;\u0026mu;M Dynasore (Selleck Chemicals, Cat. No. S8047) in growth medium for 24 h at 37\u0026deg;C. After washing with PBS, cells were fixed and then performed immunofluorescence staining. To examine mitotic rescue upon Dynasore washout, MDCK II cells were initially synchronized at G2/M by treating with 10 \u0026mu;M RO-3306 as described above. Dynasore (80 \u0026mu;M) was added to the cells 1 h before RO-3306 release by 5 washes with warm medium. Cells were allowed to enter prometaphase for 1.5 h in the presence of Dynasore. After dynasore washout, the cells were stalled at prometaphase using 50 \u0026mu;M STLC and endocytosis was allowed to resume for 30 min before image acquisition.\u003c/p\u003e\n\n\u003cp\u003ePharmacological inhibition \u003c/p\u003e\n\u003cp\u003eTo inhibit Arp2/3 complex-driven actin assembly, cells were treated by adding a 100 \u0026mu;M CK-666 (MedChemExpress, Cat. No. HY-16926) prepared in warm medium and incubated for 1 h at 37\u0026deg;C before fixation. To inhibit formin-driven actin assembly, we treated the cells by adding a 25 \u0026mu;M SMIFH2 (Sigma-Aldrich, Cat. No. 344092) prepared in warm medium and incubated for 1 h at 37\u0026deg;C. Cells were fixed and subsequently performed immunostaining. To inhibit Cdc42 activity, we treated the cells by adding 25 \u0026mu;M ML141 (MedChemExpress, Cat. No.: HY-12755) or 10 \u0026mu;M CASIN (MedChemExpress, Cat. No.: HY-12874) prepared in warm medium and incubated for 2 h at 37\u0026deg;C. Cells were fixed and subsequently performed immunostaining. To inhibit MRCKb activity, confluent monolayer of MDCK II WT was treated by 5 \u0026mu;M BDP9066 (MedChemExpress, Cat. No.: HY-111424) and incubated for 2 h at 37\u0026deg;C. Cells were fixed and subsequently performed immunostaining.\u003c/p\u003e\n\n\u003cp\u003eImmunofluorescence staining\u003c/p\u003e\n\u003cp\u003eCells grown on 12 mm or 27 mm glass-bottom dish (IWAKI) were fixed with fresh 4% (v/v) paraformaldehyde (PFA, Electron Microscopy Sciences) in DPBS containing calcium and magnesium (PBS+, Gibco\u0026trade;, Cat. No. 14040117) to preserve cell adhesion for 15 min at room temperature. After 3 washes with PBS, cells were sequentially permeabilized with 0.1% (v/v) Triton X-100 in PBS for 10 min and blocked non-specific binding with 2.5% (w/v) BSA in PBS for 30 min. After rinsing, cells were incubated with primary antibody (1:100) in 1% (w/v) BSA in PBS for overnight at 4\u0026deg;C on shaker and then incubated with respective secondary antibody conjugated to Alexa Fluor 488, 568 or 647 at a dilution of 1:200 (Invitrogen) in 1% (w/v) BSA in PBS for 1 h at room temperature. When necessary, DAPI (1:400, Thermo Fisher) and Phalloidin\u0026ndash;Atto 647N (1:120, Sigma-Aldrich, Cat. No. 65906) were used to stain nuclei and F-actin, respectively. Samples were rinsed and performed image acquisition in PBS.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003e \u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eOptimized fixation for preserving astral actin networks\u003c/p\u003e\n\u003cp\u003eFor interior actin structure visualization, great care must be taken to minimize mechanical disturbance during medium exchange. Culture medium was carefully aspirated from cells by leaving a small volume (~100 \u0026micro;l) sufficient to cover the cells. Cells were then fixed with fresh pre-warmed 4% (v/v) PFA in PBS+ by continuously dropping fixative solution slowly along the walls of the dish to avoid direct contact with the cells. Cells were incubated for 15 min at 37\u0026deg;C to ensure structural preservation. After three washes with PBS, cells were permeabilized with 0.1% (v/v) Triton X-100 in PBS for 7-8 min. Subsequent antibody staining steps were done as described above. Actin images were taken by confocal microscope on the same day of staining. \u003c/p\u003e\n\n\u003cp\u003eAntibodies\u003c/p\u003e\n\u003cp\u003eRabbit polyclonal anti-Rab5 (ab218624), anti-clathrin (ab21679), anti-\u0026alpha;-tubulin (ab18251), anti-\u0026gamma;-tubulin (ab179503), anti-\u0026alpha;-catenin (ab51032), anti-\u0026beta;-catenin (ab2365), mouse monoclonal anti-\u0026alpha;-tubulin (ab7290), mouse polyclonal anti-DNMBP (TUBA, ab88534) and rabbit polyclonal anti-Intersectin-2 (ab237509) primary antibodies were purchased from Abcam. Rabbit polyclonal anti-FMNL2 (PA5-52148), anti-FMNL3 (PA5-23317) and mouse monoclonal anti-CDC42BPB (MRCK\u0026beta;, H00009578-M03) primary antibodies were purchased from Thermo Fisher. Rabbit polyclonal anti-Rab11 (71-5300), anti-Rab35 (PA5-31674), anti-ECT2 (PA5-65769) and mouse monoclonal anti-P-cadherin (32-4000) primary antibodies were obtained from Invitrogen. Mouse monoclonal antibody against P-cadherin extracellular domain (sc-74545) and GAPDH (sc-365062) were purchased from Santa Cruz. Rat monoclonal anti-E-cadherin (MAB726) antibody was purchased from EMD Millipore. Mouse monoclonal anti-p120 catenin (610133) antibody was obtained from BD Biosciences. Rabbit polyclonal antibody to Intersectin-2 was purchased from Antibodies.com. Mouse monoclonal anti-\u0026beta;-actin (MCA5775GA) antibody was obtained from Bio-Rad. \u003c/p\u003e\n\n\u003cp\u003eMitochondria, microtubule, and DNA labelling for live-cell imaging\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMitochondria labelling\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eMitoTracker\u0026trade; Orange CMTMRos (Thermo Fisher, Cat. No. M7512) and MitoTracker\u0026trade; Green FM (Thermo Fisher, Cat. No. M7514) were prepared in warm cell culture medium at a final concentration of 100 nM, and then added to cells for 15 min at 37\u0026deg;C. The cells were washed twice with warm medium before subsequently incubating with other probes or imaging in FluoroBrite medium (Thermo Fisher). For mitochondrial membrane potential imaging, the indicator Image-iT\u0026trade; TMRM (Thermo Fisher, Cat. No. I3436), was added to cells at a concentration of 100 nM and incubated for 30 min at 37\u0026deg;C prior to imaging. For the detection of mitochondrial superoxide in cells, MitoSOX\u0026trade;Red Mitochondrial Superoxide Indicator (Thermo Fisher, Cat. No. M36007) was incubated with live cells at a concentration of 5 \u0026micro;M at 37\u0026deg;C prior to imaging.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMicrotubule labelling\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSiR-Tubulin (Cytoskeleton Inc., Cat. No. CY-SC002) was freshly prepared at a final concentration of 1 \u0026micro;M in warm FluoroBrite (Thermo Fisher) medium. The probes were added to the cells followed by incubating for 1-3 h at 37\u0026deg;C.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDNA labelling\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eNucBlue\u0026trade; (Hoechst 33342) reagent was dropped into the cells according to the manufacturer\u0026apos;s instructions and incubated for at least 15 min for nuclear visualization before performing live cell-imaging without washing step.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003e \u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eFluorescence microscopy\u003c/p\u003e\n\u003cp\u003eFixed cells were imaged with a Nikon CFI Plan Apochromat \u0026times;100/1.45 NA oil objective using a spinning disk confocal unit (CSU-W1, Yokogawa) on a Nikon Eclipse Ti-E inverted microscope (Nikon Instruments), equipped with 405 nm (100 mW), 488 nm (150 mW), 561 nm (100 mW) and 642 nm (100 mW) diode lasers, and corresponding dichroic (Chroma Technology) on a Prime 95B sCMOS camera (1200x1200 pixels, Teledyne Photometrics), with acquisition control by MetaMorph Version 7.10.1.161 software. The 405 nm, 488 nm, 561 nm, and 642 nm laser powers were set at ~40% with exposure times of ~500 ms. Super-resolution imaging was performed by activating the Live-SR (Roper Scientific) module with pixel resolution at 100X magnification equal to 0.06587 \u0026micro;m/pixel. Z-stacks were taken using a Piezo stage (Plano\u0026reg; Z, Physik Instrument) in which the step size was 0.2 or 0.5 \u0026micro;m for a total thickness of ~10-15 \u0026micro;m.\u003c/p\u003e\n\u003cp\u003eFor time-lapse or live-cell imaging, the on-stage incubator (LCI CU-501 Temperature Controller) was stabilized to 37\u0026deg;C with 5% CO2 (LCI FC-5N CO2 mixer) in humidified chamber before image acquisition. The 405 nm, 488 nm, 561 nm, and 642 nm laser powers for live-cell imaging were set at 6-8%, 10%, 9.5% and 13% with exposure times of 400 ms, 400 ms, 400 ms and 800 ms, respectively.\u003c/p\u003e\n\u003cp\u003e\u003cu\u003e \u003c/u\u003e\u003c/p\u003e\n\u003cp\u003eImage processing and analyses\u003c/p\u003e\n\u003cp\u003eConfocal images and movies were processed and analyzed using Imaris (Andor Inc.), ImageJ, MATLAB software, or custom-written python code for astral density quantification. All fluorescence images shown in Figures are from a single z-slice unless stated otherwise.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMitotic defect quantification\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eQuantifications of mitotic abnormality phenotypes were performed from metaphase cells imaged under confocal microscope after fixation and staining for \u0026alpha;-tubulin and DAPI to visualize mitotic spindle morphology and chromosomes, respectively. Metaphase cells displayed at least one uncongressed/unaligned chromosomes were counted as chromosome misalignment mitotic cells. Metaphase cells in which the spindle axis is bended instead of straight were counted as spindle distortion mitotic cells. Spindle misorientation was counted when spindle poles were not oriented parallel to the cell-substrate adhesion plane. Multipolar spindles were counted when more than two spindle poles are present.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eColocalization analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe JACoP ImageJ plugin was used to calculate the Pearson\u0026apos;s correlation coefficient (PCC, r) to analyze colocalization. Single z slice images were background subtracted and denoised (1 pixel) in ImageJ. Individual cells were cropped and specified cytoplasm as the region to use for colocalization analysis and applied threshold to create a mask for each channel before running JACoP.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMitochondria morphological quantification\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo quantify mitochondria morphology, the raw images of mitochondria (MitoTracker channel) and nuclei (DAPI channel) were denoised (1 pixel) and performed Gaussian filter with sigma radius of 1.5. Single slice of nucleus channel was applied thresholding to create binary mask. Then find maxima algorithm with the segmented particles option (prominence \u0026gt;100) was done to generate individual cell mask. Mitochondria diameter, length, area and number of each cell were quantified using ImageJ.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMitochondria area coverage quantification\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo quantify mitochondria area coverage, maximum projection of MitoTracker (mitochondria) and DAPI (nuclei) channels were denoised (1 pixel) and subjected to Gaussian filtering (sigma radius = 1), followed by thresholding to generate mitochondria and individual cell masks, respectively. The total mitochondria area per cell was measured using the Analyze particles function in ImageJ. Individual cell area was measured to calculate the percentage of cell area occupied by mitochondria. \u003c/p\u003e\n\u003cp\u003eTo quantify mitochondria area coverage within the spindle region, maximum projection of a-tubulin (mitotic spindle) channel in metaphase cells was denoised (1 pixel) and Gaussian filtered (sigma radius = 1), followed by thresholding to generate a spindle mask. The mitochondria area within the spindle region was measured and expressed as the percentage of spindle area occupied by mitochondria.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMitochondria dynamics quantification\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo quantify mitochondrial dynamics, raw images from the MitoTracker channel acquired by live-imaging (2 s per frame) were cropped to isolate individual cells, denoised (1 pixel), Gaussian filtered (sigma radius = 1) and followed by background subtraction. Mitochondria were automatically thresholded to generate binary masks and tracked using the TrackMate plugin in ImageJ. Mean mitochondrial track speed (mm/s) per cell was calculated for comparison.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMitoSOX and TMRM intensity quantitation\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eImages of MitoSOX or TMRM were denoised (1 pixel) and Gaussian filtered (sigma radius = 1), followed by background subtraction. MitoTracker images were used to generate mitochondria masks, which were applied via the ROI Manager in ImageJ. Mitochondria MitoSOX or TMRM intensity was measured per cells and used for comparison.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eActin cable length quantification\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eActin cable length within the clustered chromosome region was measured manually using the line tool in ImageJ. The DAPI channel was used to define the clustered chromosome area.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eChromosome congression analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo quantify chromosome dynamics, individual mitotic cells were cropped and DAPI channel from spinning disc confocal images were used. Gaussian blurring with 1 pixel radius was used for noise reduction, followed by background subtraction with 50 pixel radius in ImageJ. Otsu thresholding was then performed with threshold calculated for each frame to create binary mask. Chromosome cluster area from each frame was extracted in FIJI.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSpindle pole migration quantification\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo quantify spindle pole migration, individual mitotic cells were cropped and Sir-tubulin channel images were used. Gaussian blurring with 0.5-1 pixel radius was used for noise reduction in ImageJ. MTrackJ plugin was used to manually locate each spindle pole in every frame of time-lapse image series and recorded the distance between them frame by frame over time.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEndo-MRCK\u003c/em\u003e\u003cem\u003eb\u003c/em\u003e\u003cem\u003e density quantification\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSingle z-slices of MRCKb images were denoised (1 pixel) and smoothed before detecting endosomal-associated MRCKb puncta in the cytoplasm\u003cem\u003e \u003c/em\u003eusing the Find maxima with the point selection option in ImageJ. DAPI channel images were used to generate nuclear and cell masks to measure nuclear and total cell areas. Cytoplasmic area was calculated as cell area minus nucleus area and used to determine endosomal MRCKb density.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eParticle image velocimetry (PIV) analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo analyze motion of cytoplasmic actin, chromosomes and mitochondria, PIV analysis was performed using a MATLAB toolbox, MatPIV 1.6.1 (http://urn.nb.no/URN:NBN:no-27806). For each movie, single cell was cropped from the raw data and background region was cleared so that only region of interest was analyzed. Single-pass PIV with window size of 32x32 pixels and 50% overlapping was applied. For each frame (interval), average speed (\u0026micro;m/min) was computed by averaging all the local instantaneous speeds.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNormalization by actin cortex intensity\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo facilitate comparison of actin density, the F-actin intensity in confocal image is normalized by the average cortical actin intensity of each cell of interest. To determine the average cortical actin intensity, the cell boundary was interactively traced in custom-written python code, guided by the super-resolution confocal slice. A margin of \u0026plusmn;10 pixels (.65 \u0026micro;m) was defined around the cell boundary to define the cortical zone. A histogram of pixel intensity within the cortical zone is calculated, and the Otsu threshold was calculated to define high-density cortical region. Average intensity of thresholded pixels were then calculated and used to normalize the F-actin intensity in each cell.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eInterior actin density quantification \u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo enable comparison of interior actin network density, custom-written python code was used to generate a standardized map for ease of registration between cells. From the microtubule channel of a multi-color super-resolution confocal stacks, the position of each centrosome of a metaphase cell is interactively defined. The spindle axis is then calculated as the vector linking the 2 metaphase centrosomes. Polar grids were then defined with 60 azimuthal bins, and 10 radial bins (6˚ per azimuthal bin and 10 pixel or .65 \u0026micro;m per radial bin). Polar grids within \u0026plusmn; 90˚ relative to the spindle axis are considered to encompass the interior actin. Subsequently, from the cortex-normalized F-actin density described above, a binary mask of the cortex region is applied to reject pixels outside of the cell. The density map is then calculated as the average of all pixels within each bin. \u003c/p\u003e\n\n\u003cp\u003eStatistical analysis\u003c/p\u003e\n\u003cp\u003eStatistical analyses were performed in GraphPad Prism 10.0 software using two-tailed unpaired Student\u0026rsquo;s t-tests and one-way ANOVA for multiple comparison. All data were shown as means \u0026plusmn; SD, unless stated otherwise. A value of p \u0026lt; 0.05 was considered statistically significant. \u003c/p\u003e\n"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eACKNOWLEDGEMENTS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Timothy Mitchison (Harvard Medical School), Gregg Gundersen (Columbia University), and Jin Zhu (Mechanobiology Institute, Singapore) for helpful discussion. We acknowledge funding support from Ministry of Education Singapore Academic Research Fund Tier 2 (MOE-T2-EP3-0124-0012, to P.K.), Ministry of Education Singapore Academic Research Fund Tier 3 (MOE-T3-2020-01, to P.K.), National Research Foundation Singapore (NRF-MSG-2023-0001, to P.K.), National Research Foundation Singapore-Israel Science Foundation joint grant (NRF2019-NRF-ISF003-2952 to P.K.), Israel Science Foundation grant 2952/19 to R.Z-B, Chan Zuckerberg Initiative DAF CZI Imaging Scientist Award DOI:10.37921/694870itnyzk (UM), the Goeddel Family Technology Sandbox, and National Science Foundation NSF NeuroNex Award 2014862 (UM). We thank the microscopy, IT, high-throughput molecular genetics, and wet lab core facilities of the Mechanobiology Institute (MBI) for helpful discussion and infrastructure support.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u003c/strong\u003e \u003c/p\u003e\n\u003cp\u003eConceptualization: PK, RZB, KF\u003c/p\u003e\n\u003cp\u003eMethodology: KF, AK, SG, RZB, PK, UM, CS, HTO, PR, RN, KYEL, SSC, AV\u003c/p\u003e\n\u003cp\u003eInvestigation: KF, AK, SG\u003c/p\u003e\n\u003cp\u003eVisualization: KF, PK\u003c/p\u003e\n\u003cp\u003eFunding acquisition: PK, RZB\u003c/p\u003e\n\u003cp\u003eProject administration: PK, RZB\u003c/p\u003e\n\u003cp\u003eSupervision: PK, RZB\u003c/p\u003e\n\u003cp\u003eWriting \u0026ndash; original draft: KF, PK, RZB\u003c/p\u003e\n\u003cp\u003eWriting \u0026ndash; review \u0026amp; editing: KF, PK, RZB, UM\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eData and Materials Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData, materials, and code that support the plots and figures within this paper and other findings of study are available from the corresponding author upon reasonable request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAuthors declare that they have no competing interests.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSandquist, J. C., Kita, A. M. \u0026amp; Bement, W. M. And the dead shall rise: actin and myosin return to the spindle. \u003cem\u003eDevelopmental cell\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 410-419 (2011). \u003c/li\u003e\n\u003cli\u003eLechler, T. \u0026amp; Mapelli, M. Spindle positioning and its impact on vertebrate tissue architecture and cell fate. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 691-708 (2021). \u003c/li\u003e\n\u003cli\u003eHernandez, B.\u003cem\u003e et al.\u003c/em\u003e Actin organizes chromosomes and microtubules to ensure mitotic fidelity in the preimplantation embryo. \u003cem\u003eScience (New York, N.Y\u003c/em\u003e \u003cstrong\u003e388\u003c/strong\u003e, eads1234 (2025). \u003c/li\u003e\n\u003cli\u003eMogessie, B. \u0026amp; Schuh, M. Actin protects mammalian eggs against chromosome segregation errors. \u003cem\u003eScience (New York, N.Y\u003c/em\u003e \u003cstrong\u003e357\u003c/strong\u003e, eaal1647 (2017). \u003c/li\u003e\n\u003cli\u003eSchulz, O.\u003cem\u003e et al.\u003c/em\u003e Resolution doubling in fluorescence microscopy with confocal spinning-disk image scanning microscopy. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e110\u003c/strong\u003e, 21000-21005 (2013). \u003c/li\u003e\n\u003cli\u003eVieira, A. F. \u0026amp; Paredes, J. P-cadherin and the journey to cancer metastasis. \u003cem\u003eMolecular cancer\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 178 (2015). \u003c/li\u003e\n\u003cli\u003eMoore, A. S., Wong, Y. C., Simpson, C. L. \u0026amp; Holzbaur, E. L. Dynamic actin cycling through mitochondrial subpopulations locally regulates the fission\u0026ndash;fusion balance within mitochondrial networks. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 12886 (2016). \u003c/li\u003e\n\u003cli\u003eFung, T. S., Chakrabarti, R. \u0026amp; Higgs, H. N. The multiple links between actin and mitochondria. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 651-667 (2023). \u003c/li\u003e\n\u003cli\u003eValdez, V. A., Neahring, L., Petry, S. \u0026amp; Dumont, S. Mechanisms underlying spindle assembly and robustness. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 523-542 (2023). \u003c/li\u003e\n\u003cli\u003eProsser, S. L. \u0026amp; Pelletier, L. Mitotic spindle assembly in animal cells: a fine balancing act. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 187-201 (2017). \u003c/li\u003e\n\u003cli\u003eBakhoum, S. F. \u0026amp; Cantley, L. C. The multifaceted role of chromosomal instability in cancer and its microenvironment. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e174\u003c/strong\u003e, 1347-1360 (2018). \u003c/li\u003e\n\u003cli\u003eLi, R. \u0026amp; Zhu, J. Effects of aneuploidy on cell behaviour and function. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 250-265 (2022). \u003c/li\u003e\n\u003cli\u003eBharadwaj, R. \u0026amp; Yu, H. The spindle checkpoint, aneuploidy, and cancer. \u003cem\u003eOncogene\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 2016-2027 (2004). \u003c/li\u003e\n\u003cli\u003eDumont, J. \u0026amp; Desai, A. Acentrosomal spindle assembly and chromosome segregation during oocyte meiosis. \u003cem\u003eTrends in cell biology\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 241-249 (2012). \u003c/li\u003e\n\u003cli\u003eSchuh, M. \u0026amp; Ellenberg, J. Self-organization of MTOCs replaces centrosome function during acentrosomal spindle assembly in live mouse oocytes. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e130\u003c/strong\u003e, 484-498 (2007). \u003c/li\u003e\n\u003cli\u003eRoeles, J. \u0026amp; Tsiavaliaris, G. Actin-microtubule interplay coordinates spindle assembly in human oocytes. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 4651 (2019). \u003c/li\u003e\n\u003cli\u003ePlessner, M., Knerr, J. \u0026amp; Grosse, R. Centrosomal actin assembly is required for proper mitotic spindle formation and chromosome congression. \u003cem\u003eIScience\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 274-281 (2019). \u003c/li\u003e\n\u003cli\u003eFarina, F.\u003cem\u003e et al.\u003c/em\u003e Local actin nucleation tunes centrosomal microtubule nucleation during passage through mitosis. \u003cem\u003eEmbo J\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, e99843 (2019). \u003c/li\u003e\n\u003cli\u003eLancaster, O. M. \u0026amp; Baum, B. in \u003cem\u003eSemin Cell Dev Biol.\u003c/em\u003e 109-115 (Elsevier).\u003c/li\u003e\n\u003cli\u003eStewart, M. P., Helenius, J., Toyoda, Y., Ramanathan, S. P., Muller, D. J. \u0026amp; Hyman, A. A. Hydrostatic pressure and the actomyosin cortex drive mitotic cell rounding. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e469\u003c/strong\u003e, 226-230 (2011). \u003c/li\u003e\n\u003cli\u003eSalbreux, G., Charras, G. \u0026amp; Paluch, E. Actin cortex mechanics and cellular morphogenesis. \u003cem\u003eTrends in Cell Biology\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 536-545 (2012). https://doi.org:10.1016/j.tcb.2012.07.001\u003c/li\u003e\n\u003cli\u003eKunda, P. \u0026amp; Baum, B. The actin cytoskeleton in spindle assembly and positioning. \u003cem\u003eTrends in cell biology\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 174-179 (2009). \u003c/li\u003e\n\u003cli\u003eGreen, R. A., Paluch, E. \u0026amp; Oegema, K. Cytokinesis in animal cells. \u003cem\u003eAnnual review of cell and developmental biology\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 29-58 (2012). \u003c/li\u003e\n\u003cli\u003eCoscia, S. M., Moore, A. S., Thompson, C. P., Tirrito, C. F., Ostap, E. M. \u0026amp; Holzbaur, E. L. An interphase actin wave promotes mitochondrial content mixing and organelle homeostasis. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 3793 (2024). \u003c/li\u003e\n\u003cli\u003eMoore, A. S.\u003cem\u003e et al.\u003c/em\u003e Actin cables and comet tails organize mitochondrial networks in mitosis. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e591\u003c/strong\u003e, 659-664 (2021). \u003c/li\u003e\n\u003cli\u003eRagkousi, K. \u0026amp; Gibson, M. C. Cell division and the maintenance of epithelial order. \u003cem\u003eJournal of Cell Biology\u003c/em\u003e \u003cstrong\u003e207\u003c/strong\u003e, 181-188 (2014). \u003c/li\u003e\n\u003cli\u003eCharras, G. \u0026amp; Yap, A. S. Tensile forces and mechanotransduction at cell\u0026ndash;cell junctions. \u003cem\u003eCurrent Biology\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, R445-R457 (2018). \u003c/li\u003e\n\u003cli\u003eCamp\u0026agrave;s, O., Noordstra, I. \u0026amp; Yap, A. S. Adherens junctions as molecular regulators of emergent tissue mechanics. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e, 1-18 (2023). \u003c/li\u003e\n\u003cli\u003eNelson, W. J. Regulation of cell\u0026ndash;cell adhesion by the cadherin\u0026ndash;catenin complex. \u003cem\u003eBiochem Soc Trans\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 149 (2008). \u003c/li\u003e\n\u003cli\u003eTakeichi, M. Dynamic contacts: rearranging adherens junctions to drive epithelial remodelling. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 397-410 (2014). \u003c/li\u003e\n\u003cli\u003eGloerich, M., Bianchini, J. M., Siemers, K. A., Cohen, D. J. \u0026amp; Nelson, W. J. Cell division orientation is coupled to cell-cell adhesion by the E-cadherin/LGN complex. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 13996 (2017). https://doi.org:10.1038/ncomms13996\u003c/li\u003e\n\u003cli\u003eBenham-Pyle, B. W., Pruitt, B. L. \u0026amp; Nelson, W. J. Mechanical strain induces E-cadherin\u0026ndash;dependent Yap1 and \u0026beta;-catenin activation to drive cell cycle entry. \u003cem\u003eScience (New York, N.Y\u003c/em\u003e \u003cstrong\u003e348\u003c/strong\u003e, 1024-1027 (2015). \u003c/li\u003e\n\u003cli\u003eMendonsa, A. M., Na, T. Y. \u0026amp; Gumbiner, B. M. E-cadherin in contact inhibition and cancer. \u003cem\u003eOncogene\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 4769-4780 (2018). https://doi.org:10.1038/s41388-018-0304-2\u003c/li\u003e\n\u003cli\u003eWheelock, M. J., Shintani, Y., Maeda, M., Fukumoto, Y. \u0026amp; Johnson, K. R. Cadherin switching. \u003cem\u003eJournal of cell science\u003c/em\u003e \u003cstrong\u003e121\u003c/strong\u003e, 727-735 (2008). \u003c/li\u003e\n\u003cli\u003eYates, A. D.\u003cem\u003e et al.\u003c/em\u003e Ensembl 2026. \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, D1053-D1060 (2026). \u003c/li\u003e\n\u003cli\u003eVan Roy, F. Beyond E-cadherin: roles of other cadherin superfamily members in cancer. \u003cem\u003eNature Reviews Cancer\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 121-134 (2014). \u003c/li\u003e\n\u003cli\u003eAndrews, J. L., Kim, A. C. \u0026amp; Hens, J. R. The role and function of cadherins in the mammary gland. \u003cem\u003eBreast cancer research\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 203 (2012). \u003c/li\u003e\n\u003cli\u003eSamuelov, L., Sprecher, E. \u0026amp; Paus, R. The role of P-cadherin in skin biology and skin pathology: lessons from the hair follicle. \u003cem\u003eCell and tissue research\u003c/em\u003e \u003cstrong\u003e360\u003c/strong\u003e, 761-771 (2015). \u003c/li\u003e\n\u003cli\u003eNose, A. \u0026amp; Takeichi, M. A novel cadherin cell adhesion molecule: its expression patterns associated with implantation and organogenesis of mouse embryos. \u003cem\u003eThe Journal of cell biology\u003c/em\u003e \u003cstrong\u003e103\u003c/strong\u003e, 2649-2658 (1986). \u003c/li\u003e\n\u003cli\u003eWu, J.-C., Gregory, C. W. \u0026amp; DePhilip, R. M. P-cadherin and E-cadherin are co-expressed in MDCK cells. \u003cem\u003eBiochem Biophys Res Commun\u003c/em\u003e \u003cstrong\u003e195\u003c/strong\u003e, 1329-1335 (1993). \u003c/li\u003e\n\u003cli\u003eTran, S. K.\u003cem\u003e et al.\u003c/em\u003e P-cadherin dependent adhesions are required for single lumen formation and HGF-mediated cell protrusions during epithelial morphogenesis. \u003cem\u003eiScience\u003c/em\u003e (2025). \u003c/li\u003e\n\u003cli\u003eGavard, J., Marthiens, V., Monnet, C., Lambert, M. \u0026amp; Mege, R. M. N-cadherin activation substitutes for the cell contact control in cell cycle arrest and myogenic differentiation: involvement of p120 and beta-catenin. \u003cem\u003eJ Biol Chem\u003c/em\u003e \u003cstrong\u003e279\u003c/strong\u003e, 36795-36802 (2004). https://doi.org:10.1074/jbc.M401705200\u003c/li\u003e\n\u003cli\u003eBauer, D. E., Canver, M. C. \u0026amp; Orkin, S. H. Generation of genomic deletions in mammalian cell lines via CRISPR/Cas9. \u003cem\u003eJoVE (Journal of Visualized Experiments)\u003c/em\u003e, e52118 (2015). \u003c/li\u003e\n\u003cli\u003eBalasubramaniam, L.\u003cem\u003e et al.\u003c/em\u003e Investigating the nature of active forces in tissues reveals how contractile cells can form extensile monolayers. \u003cem\u003eNat Mater\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 1156-1166 (2021). \u003c/li\u003e\n\u003cli\u003eMaupin, P. \u0026amp; Pollard, T. D. Arrangement of actin filaments and myosin-like filaments in the contractile ring and of actin-like filaments in the mitotic spindle of dividing HeLa cells. \u003cem\u003eJournal of ultrastructure and molecular structure research\u003c/em\u003e \u003cstrong\u003e94\u003c/strong\u003e, 92-103 (1986). \u003c/li\u003e\n\u003cli\u003eAzuma, T. \u0026amp; Kei, T. Super-resolution spinning-disk confocal microscopy using optical photon reassignment. \u003cem\u003eOptics Express\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 15003-15011 (2015). \u003c/li\u003e\n\u003cli\u003eKita, A. M., Swider, Z. T., Erofeev, I., Halloran, M. C., Goryachev, A. B. \u0026amp; Bement, W. M. Spindle\u0026ndash;F-actin interactions in mitotic spindles in an intact vertebrate epithelium. \u003cem\u003eMolecular biology of the cell\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 1645-1654 (2019). \u003c/li\u003e\n\u003cli\u003eHarasimov, K.\u003cem\u003e et al.\u003c/em\u003e Actin-driven chromosome clustering facilitates fast and complete chromosome capture in mammalian oocytes. \u003cem\u003eNature cell biology\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 439-452 (2023). \u003c/li\u003e\n\u003cli\u003eBooth, A. J.\u003cem\u003e et al.\u003c/em\u003e Contractile acto-myosin network on nuclear envelope remnants positions human chromosomes for mitosis. \u003cem\u003eElife\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, e46902 (2019). \u003c/li\u003e\n\u003cli\u003eZheng, Y. A membranous spindle matrix orchestrates cell division. \u003cem\u003eNature reviews Molecular cell biology\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 529-535 (2010). \u003c/li\u003e\n\u003cli\u003eStamhuis, E. \u0026amp; Thielicke, W. PIVlab\u0026ndash;towards user-friendly, affordable and accurate digital particle image velocimetry in MATLAB. \u003cem\u003eJournal of open research software\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 30 (2014). \u003c/li\u003e\n\u003cli\u003eSchiavon, C. R.\u003cem\u003e et al.\u003c/em\u003e Actin chromobody imaging reveals sub-organellar actin dynamics. \u003cem\u003eNature methods\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 917-921 (2020). \u003c/li\u003e\n\u003cli\u003eGatti, P., Schiavon, C., Cicero, J., Manor, U. \u0026amp; Germain, M. Mitochondria-and ER-associated actin are required for mitochondrial fusion. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 451 (2025). \u003c/li\u003e\n\u003cli\u003eRiedl, J.\u003cem\u003e et al.\u003c/em\u003e Lifeact: a versatile marker to visualize F-actin. \u003cem\u003eNature methods\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 605-607 (2008). \u003c/li\u003e\n\u003cli\u003eSchweizer, N., Pawar, N., Weiss, M. \u0026amp; Maiato, H. An organelle-exclusion envelope assists mitosis and underlies distinct molecular crowding in the spindle region. \u003cem\u003eJournal of Cell Biology\u003c/em\u003e \u003cstrong\u003e210\u003c/strong\u003e, 695-704 (2015). \u003c/li\u003e\n\u003cli\u003eSuarez, C. \u0026amp; Kovar, D. R. Internetwork competition for monomers governs actin cytoskeleton organization. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 799-810 (2016). \u003c/li\u003e\n\u003cli\u003eMiserey-Lenkei, S. \u0026amp; Colombo, M. I. Small RAB GTPases regulate multiple steps of mitosis. \u003cem\u003eFrontiers in Cell and Developmental Biology\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 2 (2016). \u003c/li\u003e\n\u003cli\u003eCadwell, C. M., Su, W. \u0026amp; Kowalczyk, A. P. Cadherin tales: Regulation of cadherin function by endocytic membrane trafficking. \u003cem\u003eTraffic\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 1262-1271 (2016). \u003c/li\u003e\n\u003cli\u003eMacia, E., Ehrlich, M., Massol, R., Boucrot, E., Brunner, C. \u0026amp; Kirchhausen, T. Dynasore, a cell-permeable inhibitor of dynamin. \u003cem\u003eDevelopmental cell\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 839-850 (2006). \u003c/li\u003e\n\u003cli\u003eChircop, M. Rho GTPases as regulators of mitosis and cytokinesis in mammalian cells. \u003cem\u003eSmall GTPases\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, e29770 (2014). \u003c/li\u003e\n\u003cli\u003eMitsushima, M., Toyoshima, F. \u0026amp; Nishida, E. Dual role of Cdc42 in spindle orientation control of adherent cells. \u003cem\u003eMolecular and cellular biology\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 2816-2827 (2009). \u003c/li\u003e\n\u003cli\u003eBenink, H. A. \u0026amp; Bement, W. M. Concentric zones of active RhoA and Cdc42 around single cell wounds. \u003cem\u003eThe Journal of cell biology\u003c/em\u003e \u003cstrong\u003e168\u003c/strong\u003e, 429-439 (2005). \u003c/li\u003e\n\u003cli\u003eSurviladze, Z.\u003cem\u003e et al.\u003c/em\u003e A potent and selective inhibitor of Cdc42 GTPase. \u003cem\u003eProbe Reports from the NIH Molecular Libraries Program [Internet]\u003c/em\u003e (2010). \u003c/li\u003e\n\u003cli\u003eLiu, W.\u003cem\u003e et al.\u003c/em\u003e Rational identification of a Cdc42 inhibitor presents a new regimen for long-term hematopoietic stem cell mobilization. \u003cem\u003eLeukemia\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 749-761 (2019). \u003c/li\u003e\n\u003cli\u003eWatson, J. R., Owen, D. \u0026amp; Mott, H. R. Cdc42 in actin dynamics: An ordered pathway governed by complex equilibria and directional effector handover. \u003cem\u003eSmall GTPases\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 237-244 (2017). \u003c/li\u003e\n\u003cli\u003eK\u0026uuml;hn, S. \u0026amp; Geyer, M. Formins as effector proteins of Rho GTPases. \u003cem\u003eSmall GTPases\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, e983876 (2014). \u003c/li\u003e\n\u003cli\u003eNishimura, Y.\u003cem\u003e et al.\u003c/em\u003e The formin inhibitor SMIFH2 inhibits members of the myosin superfamily. \u003cem\u003eJournal of cell science\u003c/em\u003e \u003cstrong\u003e134\u003c/strong\u003e, jcs253708 (2021). \u003c/li\u003e\n\u003cli\u003eK\u0026uuml;hn, S.\u003cem\u003e et al.\u003c/em\u003e The structure of FMNL2\u0026ndash;Cdc42 yields insights into the mechanism of lamellipodia and filopodia formation. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 7088 (2015). \u003c/li\u003e\n\u003cli\u003ePfisterer, K.\u003cem\u003e et al.\u003c/em\u003e FMNL2 regulates dynamics of fascin in filopodia. \u003cem\u003eJournal of Cell Biology\u003c/em\u003e \u003cstrong\u003e219\u003c/strong\u003e (2020). \u003c/li\u003e\n\u003cli\u003eCho, K. F., Branon, T. C., Udeshi, N. D., Myers, S. A., Carr, S. A. \u0026amp; Ting, A. Y. Proximity labeling in mammalian cells with TurboID and split-TurboID. \u003cem\u003eNat Protoc\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 3971-3999 (2020). \u003c/li\u003e\n\u003cli\u003eOllech, D.\u003cem\u003e et al.\u003c/em\u003e An optochemical tool for light-induced dissociation of adherens junctions to control mechanical coupling between cells. \u003cstrong\u003e11\u003c/strong\u003e, 472 (2020). \u003c/li\u003e\n\u003cli\u003eSzklarczyk, D.\u003cem\u003e et al.\u003c/em\u003e The STRING database in 2023: protein\u0026ndash;protein association networks and functional enrichment analyses for any sequenced genome of interest. \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, D638-D646 (2023). \u003c/li\u003e\n\u003cli\u003eFox, S.\u003cem\u003e et al.\u003c/em\u003e Identification of an FMNL2 Interactome by Quantitative Mass Spectrometry. \u003cem\u003eInternational Journal of Molecular Sciences\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 5686 (2024). \u003c/li\u003e\n\u003cli\u003eGadella Jr, T. W.\u003cem\u003e et al.\u003c/em\u003e mScarlet3: a brilliant and fast-maturing red fluorescent protein. \u003cem\u003eNature methods\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 541-545 (2023). \u003c/li\u003e\n\u003cli\u003eBirch, J. L.\u003cem\u003e et al.\u003c/em\u003e A novel small-molecule inhibitor of MRCK prevents radiation-driven invasion in glioblastoma. \u003cem\u003eCancer research\u003c/em\u003e \u003cstrong\u003e78\u003c/strong\u003e, 6509-6522 (2018). \u003c/li\u003e\n\u003cli\u003eFujita, Y.\u003cem\u003e et al.\u003c/em\u003e Hakai, a c-Cbl-like protein, ubiquitinates and induces endocytosis of the E-cadherin complex. \u003cem\u003eNature cell biology\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 222-231 (2002). \u003c/li\u003e\n\u003cli\u003ePan, M.-H.\u003cem\u003e et al.\u003c/em\u003e FMNL2 regulates actin for endoplasmic reticulum and mitochondria distribution in oocyte meiosis. \u003cem\u003eElife\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, RP92732 (2024). \u003c/li\u003e\n\u003cli\u003eRufas, O., Fisch, B., Ziv, S. \u0026amp; Shalgi, R. Expression of cadherin adhesion molecules on human gametes. \u003cem\u003eMol Hum Reprod\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 163-169 (2000). \u003c/li\u003e\n\u003cli\u003eT\u0026aacute;bara, L.-C., Segawa, M. \u0026amp; Prudent, J. Molecular mechanisms of mitochondrial dynamics. \u003cem\u003eNature Reviews Molecular Cell Biology\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 123-146 (2025). \u003c/li\u003e\n\u003cli\u003eIvaska, J., Heino, J. J. A. r. o. c. \u0026amp; biology, d. Cooperation between integrins and growth factor receptors in signaling and endocytosis. \u003cstrong\u003e27\u003c/strong\u003e, 291-320 (2011). \u003c/li\u003e\n\u003cli\u003eAlanko, J.\u003cem\u003e et al.\u003c/em\u003e Integrin endosomal signalling suppresses anoikis. \u003cem\u003eNature cell biology\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 1412-1421 (2015). \u003c/li\u003e\n\u003cli\u003eVassilev, V., Platek, A., Hiver, S., Enomoto, H. \u0026amp; Takeichi, M. Catenins steer cell migration via stabilization of front-rear polarity. \u003cem\u003eDevelopmental cell\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, 463-479. e465 (2017). \u003c/li\u003e\n\u003cli\u003eNollet, F., Kools, P. \u0026amp; Van Roy, F. Phylogenetic analysis of the cadherin superfamily allows identification of six major subfamilies besides several solitary members. \u003cem\u003eJ Mol Biol\u003c/em\u003e \u003cstrong\u003e299\u003c/strong\u003e, 551-572 (2000). \u003c/li\u003e\n\u003cli\u003eZhang, C. C.\u003cem\u003e et al.\u003c/em\u003e PF-03732010: a fully human monoclonal antibody against P-cadherin with antitumor and antimetastatic activity. \u003cem\u003eClinical Cancer Research\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 5177-5188 (2010). \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"P-cadherin, mitotic spindle, mitotic fidelity, mitosis, endosomes, actin, mitochondria, endosomal signaling","lastPublishedDoi":"10.21203/rs.3.rs-9256156/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9256156/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Accurate chromosome segregation during somatic cell division depends on precise assembly of the mitotic spindle, a process traditionally regarded as microtubule-centric1,2. Recent works in acentrosomal systems have implicated actin in spindle function3,4, but whether spindle-associated actin is a broader feature of centrosomal mitosis remains unclear1. Here we show that the mitotic spindle is cradled by a surprisingly delicate cytoplasmic interior actin meshwork that requires optimized preservation and super-resolution imaging5 for its visualization. We show that this mitotic interior actin network contributes to mitotic fidelity and is generated by early endosomes containing P-cadherin, a cell–cell adhesion receptor frequently dysregulated in cancer6. Mechanistically, P-cadherin undergoes cell-cycle-coupled trafficking through Rab5 endosomes and, together with Cdc42, MRCKβ, and FMNL2, drives assembly of dynamic actin cables. Loss of P-cadherin causes widespread spindle defects and impaired chromosome congression. Strikingly, P-cadherin depletion simultaneously collapses this endosome-derived actin network and enhances mitochondria-associated actin7,8, giving rise to peri-mitochondrial actin shells and mitochondrial dysfunctions, consistent with reciprocal regulation between distinct interior actin pools. Our findings reveal a cadherin-dependent, endosome-generated actin system that dynamically organizes the mitotic cell interior to ensure faithful chromosome segregation, with implications for epithelial homeostasis and cancer.","manuscriptTitle":"Endosome-generated interior actin network regulates mitotic fidelity and organelle homeostasis downstream of P-cadherin","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-27 11:00:18","doi":"10.21203/rs.3.rs-9256156/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"f0133cdf-e4fd-462a-b6d7-2085199d8de3","owner":[],"postedDate":"April 27th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":66893746,"name":"Biological sciences/Cell biology/Cell division/Mitosis"},{"id":66893747,"name":"Biological sciences/Cell biology/Cell division/Mitotic spindle"},{"id":66893748,"name":"Biological sciences/Cell biology/Cytoskeleton/Actin"},{"id":66893749,"name":"Biological sciences/Cell biology/Cell adhesion/Cadherins"},{"id":66893750,"name":"Biological sciences/Cell biology/Cellular imaging/Super-resolution microscopy"}],"tags":[],"updatedAt":"2026-04-27T11:00:18+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-27 11:00:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9256156","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9256156","identity":"rs-9256156","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}