Motor proteins and cytoskeletal dynamics in T cells, B cells and mesenchymal cells

National Heart, Lung, And Blood Institute

Investigators

Linked publications, trials & patents

Abstract

Mechanisms underlying Myosin 10s contribution to the maintenance of mitotic spindle bipolarity. Myosin 10 (Myo10) has the ability to link actin filaments to integrin-based adhesions and to microtubules by virtue of its integrin-binding FERM domain and microtubule-binding MyTH4 domain, respectively. We used Myo10 knockout cells to define its contribution to the maintenance of spindle bipolarity, Halo tag knock-in cells to reveal its localization during mitosis, and complementation to determine the relative contributions of its MyTH4 and FERM domains to maintaining spindle bipolarity. Myo10 knockout HeLa cells and mouse embryo fibroblasts (MEFs) both exhibit a pronounced increase in the frequency of multipolar spindles. Staining of unsynchronized metaphase cells showed that the primary driver of spindle multipolarity in knockout MEFs and knockout HeLa cells lacking supernumerary centrosomes is pericentriolar material (PCM) fragmentation, which creates y-tubulin-positive acentriolar foci that serve as additional spindle poles. For HeLa cells possessing supernumerary centrosomes, Myo10 depletion further accentuates spindle multipolarity by impairing the clustering of the extra spindle poles. Complementation experiments show that Myo10 must interact with both integrins and microtubules to promote PCM/pole integrity. Conversely, Myo10s ability to promote the clustering of supernumerary centrosomes only requires that it interact with integrins. Importantly, images of Halo-Myo10 knock-in cells show that the myosin localizes exclusively within adhesive retraction fibers during mitosis. Based on these and other results, we conclude that Myo10 promotes PCM/pole integrity at a distance, and that it facilitates supernumerary centrosome clustering by promoting retraction fiber-based cell adhesion, which likely provides an anchor for the microtubule-based forces driving pole focusing. Myosin 10 at the tips of filopodia-derived retraction fibers supports adhesion during mitosis when conventional focal adhesions disassemble. Historically, adhesion during mitosis for cells grown in 2D has been attributed to retraction fibers, which are thought to arise from a combination of the cell rounding that occurs upon mitotic entry and the persistence of interphase focal adhesions. A recent study showed, however, that Talin, the main clutch component connecting actin to integrin, largely disappears from sites of adhesion in mitotic cells (Dix et al Dev Cell 2018). What then connects actin to integrin during mitosis? Here we show that endogenously tagged Myo10, an integrin-binding MyTH4/FERM domain myosin commonly referred to as the filopodial myosin, localizes together with active integrin and IRM signals along the shaft and at the tips of metaphase retraction fibers. Consistent with the results of Dix et al, and with the idea that retraction fibers are in fact filopodia, time lapse imaging shows that Talin-rich focal adhesions at the cell perimeter vanish upon mitotic entry while pre-existing, Myo10-positive, interphase filopodia persist, such that 95% of them become retraction fibers. In support of this, metaphase retraction fibers stain for the filopodial crosslinker fascin and the filopodial tip marker VASP, and endogenous Myo10 moves out retraction fibers at 0.7 um/s, consistent with the bundled, barbed end-out organization of actin found within filopodia. These results, together with the fact that fluorescence intensity measurements within the TIRF field as proxies for adhesion support show that Myo10 increases and Talin decreases between mitotic entry and metaphase, suggest that Myo10 at the tips of filopodia-derived retraction fibers is replacing Talin as the main clutch component connecting actin to integrins during mitosis. Consistent with this idea, measurements of retraction fiber failure frequency and the content of active integrin within retraction fibers indicate that adhesion is attenuated in dividing cells lacking Myo10. Moreover, these defects are rescued by wild type Myo10 but not by a version that cannot bind integrin. Together, these data reveal a self-organizing property of mitotic retraction fibers: their ability to support adhesion during mitosis is hardwired by the fact that they pre-exist as Myo10-dependent adhesive filopodia, and their barbed end-out organization licenses Myo10-dependent adhesion reinforcement during mitosis. Myosin 2A-driven planar cell division ensures lumen integrity in intestinal organoids. Mouse intestinal organoids retain the crypt-villus topology of the intestine in the form of a three-dimensional, stem cell-driven tissue model that is amenable to structure: function analysis. Here we investigated the role that myosin 2A (M2A) plays in maintaining a single lumen in growing organoids. Airyscan images of organoids made from a GFP-M2A knockin mouse show that the myosin localizes along the lateral surface of organoids, consistent with its known role in supporting cadherin-based cell: cell adhesion, and within stress fibers at the base, consistent with its known role in supporting integrin-based cell: ECM adhesion. In dividing cells, M2A is seen not only in the cleavage furrow, but also in evenly-spaced strings of cortical mini-filaments that span the long axis of the cell prior to contractile ring formation. shRNA-mediated knockdown (KD) of M2A in GFP-M2A KI organoids leads to the appearance of a multilobed phenotype that scales with the degree of KD (the dimmer the GFP-M2A signal, the more multilobed the organoid appears). Imaging with a membrane marker shows that each lobe possesses its own, separate lumen. Lattice Light Sheet imaging of stem cell-marked organoids treated with Blebbistatin (BB), which phenocopies M2A KD, shows that individual lobes arise from stem cells undergoing cell division. In the absence of M2A-dependent force, interkinetic nuclear migration in these cells is inhibited and their axis of cell division changes from planer to orthogonal. As a result, the daughter cells are no longer side-by-side in the plane of the epithelium, but rather stacked one on top of the other. This leads to epithelial stratification (the formation of multiple cell layers). Staining of the daughter cells arising from orthogonal divisions with ZO-1 to mark their apical domains shows that the nascent lumen of the forming lobe is created at the interface between the two stacked cells where their apical domains abut. Interestingly, restoration of M2A contractility by BB washout, if done before the multilobed phenotype progress too far, results in lobe merger to restore the single lumen. We conclude that M2A-dependent contractility serves to maintain a single lumen in growing organoids by maintaining a planar axis of cell division. How each of the several places within enterocytes where M2A functions contribute to this is under investigation.