Clare Waterman - US grants

Project 1;Myosin II mediates local cortical tension to guide endothelial cell branching[unreadable] morphogenesis and migration in 3D.[unreadable] Personell: Robert Fischer[unreadable] [unreadable] Angiogenesis is a critical process in both development and disease, where it represents[unreadable] a prime target for therapeutic intervention. Since new blood vessels are directed[unreadable] to tissues with metabolic demands, a key feature of angiogenesis is directional control of[unreadable] endothelial cell (EC) morphogenesis and movement. During angiogenic sprouting,[unreadable] single endothelial tip cells directionally branch from existing vessels in response to[unreadable] biochemical cues such as VEGF and migrate and invade the surrounding extracellular[unreadable] matrix (ECM) in a process that requires ECM remodeling by matrix metalloproteases[unreadable] (MMPs). This is followed by EC proliferation and tubulation to establish a new[unreadable] vessel. Tip cell branching is mediated by directional protrusion at the subcellular level. Here we sought to understand how directional protrusion is locally regulated to guide[unreadable] EC branching morphogenesis. We developed an in vitro 3D EC model system that[unreadable] faithfully mimicks EC morphodynamics in angiogenesis in living zebrafish. We[unreadable] demonstrate that ECM stiffness and ROCK-mediated myosin II activity inhibit EC branch[unreadable] initiation. Myosin II is localized at the EC cortex, from which it is partially released under[unreadable] conditions that promote EC branching. Local depletion of cortical myosin II precedes[unreadable] branch initiation, and initiation can be induced by local inhibition of myosin II activity.[unreadable] Thus, local downregulation of myosin II cortical contraction allows pseudopod initiation[unreadable] to regulate EC branching and hence guide directional migration and angiogenesis.[unreadable] [unreadable] This work was performed in collaboration with Bob Adelstein and Xuefei Ma (NHLBI) and Margaret Gardel (U Chicago) and has been submitted for publication.[unreadable] [unreadable] Project 2: Development of algorithms for tracking cell morphodynamics in three dimensions.[unreadable] [unreadable] Personell: Bob Fischer[unreadable] In collaboration with Gaudenz Danuser and Sam Ching at Scripps and performed at MBL at Woods Hole. This is ongoing.

Project 1: Vinculin stabilizes nascent adhesions and establishes lamellipodium-lamella border in migrating cells Ingo Thievessen, R. Ross The actin cytoskeleton at the leading edge of migrating cells consists of two actin networks, the lamellipodium (LP), characterized by fast polymerization-driven retrograde actin flow and the lamellum (LM) with slow myosin II (myoII) mediated actin flow. The engagement of LP actin to the ECM via nascent integrin-mediated focal adhesions (FA) establishes the flow velocity gradient between LP and LM. Nascent adhesions then elongate and mature via myoII LM actin flow. How integrins are connected to the retrograde actin flow is not known. Using primary murine embryonic fibroblasts (MEF) deficient for the vinculin gene (Vcl), we sought to test the hypothesis that vinculin mediates the coupling of actin retrograde flow to the ECM in FA. Our results suggest that vinculin stabilizes nascent FA by coupling to lamellipodial actin flow, thus establishing the flow velocity gradient between LP and LM and promoting the maturation of nascent adhesions. This implicates vinculin as an essential component in linking the dynamic actin cytoskeleton to the ECM during cell migration. This work was done in collaboration with Bob Ross, and has been presented at Cell Biology Society meeting and Gordon Conference and a manuscript describing these results was published in the JCB this past yearProject 2: Nanoscale architecture of integrin-based cell adhesions Project 2:Specific protein-interactions regulate the three-dimensional nanoscale organization of vinculin within focal adhesions L.B. Case, G. Shtengel, H.F. Hess, S. Campbell C.M. Waterman Specific protein-interactions regulate the three-dimensional nanoscale organization and conformation of vinculin within focal adhesions. L.B. Case, G. Shtengel, M. Baird, M. Davidson, S. Campbell, H.F. Hess, C.M. Waterman Previous studies have shown that FAs have a conserved stratified nanoscale structure, with an integrin signaling layer (ISL) 10-20 nm from the plasma membrane, an actin regulatory layer (ARL) 100 nm from the membrane that extends into the stress fiber, and a force transduction layer (FTL) that spans these two layers. Vinculin is an FA protein that functions in signaling, force transduction, and regulation of the actin cytoskeleton. Vinculin has at least 10 binding partners distributed throughout the FA including paxillin in the ISL, talin in the FTL, and actin in the ARL. We hypothesize that vinculin interacts with distinct binding partners within distinct FA layers to regulate its activation and mediate its functional specificity. To test this, we utilized point mutants to perturb specific protein interactions and assayed their nanoscale localization, activation state, and binding stability in FAs. Our results suggest a model in which inactive vinculin is recruited to the ISL near the plasma membrane in FA by a weak interaction with paxillin. We speculate that this localization puts vinculin in proximity of multiple ligands that promote activation, which allows a shift to the FTL and ARL where interactions of activated vinculin with talin and actin promote FA stabilization. We are assaying additional vinculin mutants to provide further evidence for this model. A Manuscript describing this work is being prepared for publication Project 3: Active organization of integrins in focal adhesions Vinay Swaminathan, Pontus Norenfeldt, Joseph Matthew, Timothy Springer, Satyajit Mayor Integrins are transmembrane ECM receptors that link the extra-cellular matrix (ECM) and the cytoskeleton and play a crucial role in the immune response, cell migration and tissue morphogenesis. ECM-engaged integrins cluster together with proteins that mediate their signaling functions and linkage to the cytoskeleton to form focal adhesions that grow and turn over in an actin and myosin II dependent manner. How actin and myosin mediate the clustering and organization of integrins during activation, focal adhesion growth and turnover is not known. We utilized fluorescence emission anisotropy imaging of cells expressing GFP-tagged integrins to analyze the evolution of integrin organization during focal adhesion dynamics in migrating fibroblasts and to test the role of integrin activation and actomyosin contractility in this process. By employing specific perturbations we are attempting to understand how the association of focal adhesion components and the acto-myosin machinery affect the organization of integrins during the formation of mature adhesions. Project 4: A novel actin-adhesion structure involved in nuclear positioning requires the formin FMN2 Colleen T. Skau and Clare M. Waterman Active asymmetric positioning of the nucleus in nondividing cells is critical to a variety of cell functions, including cell migration, particularly in complex 3D environments. Based on the importance of integrin-FAK signaling in controlling nuclear orientation in migrating fibroblasts, we hypothesize that adhesion of the cell body to the extracellular matrix is likely critical in nuclear positioning. Furthermore, involvement of LPA/Rho signaling in nuclear orientation suggests a role for the actin cytoskeleton in nuclear positioning. We therefore hypothesize that the actin cytoskeleton, working with adhesions, exerts force on the nucleus through its connections to the nuclear lamina via proteins spanning the nuclear envelope. Although proteins that bind actin at the nuclear envelope have been identified, it is unclear specifically which adhesion/ actin structures mechanically couple the nucleus to the extracellular matrix to control active nuclear positioning. To address this question, we examined the interplay between actin, adhesions and the nucleus in live and fixed mouse embryonic fibroblasts using fluorescence microscopy. We identified novel adhesive structures located underneath the nucleus termed subnuclear adhesions that are compositionally and dynamically distinct from canonical focal adhesions found near the leading edge of migrating cells. Unlike focal adhesions at the leading edge, subnuclear adhesions assemble in bursts toward the front of the cell and translocate along the ventral surface of the cell. Additionally, we find that subnuclear adhesions contain a distinct set of proteins from leading edge adhesions, and that these proteins assemble with distinct kinetics. Subnuclear adhesions, like leading edge adhesions, have a specific set of actin filaments associated with them. These subnuclear actin fibers connect to subnuclear adhesions not leading edge adhesions, and we have found that they physically impinge upon the nucleus and help control nuclear shape. We have previously identified the actin nucleation factor formin FMN2 in a screen of adhesion components. We now find that the formin FMN2 localizes underneath the nucleus and is important for these subnuclear actin fibers, which are likely involved in known active positioning mechanisms of the nucleus in migrating cells. Together, our data reveal a previously unidentified mechanism for control of nuclear position via a novel structure connecting the actin cytoskeleton, which impinges upon the nucleus, to the extracellular matrix underneath the cell body.

Project 1: Rac1 induces PCK-dependent myosinIIA phosphorylation to regulate association with focal adhesions and cell migration. Pasapera AM1, Fischer RS1,Plotnikov SV1,Egelhoff T2, Waterman CM1. Cell Biology and Physiology Center, NHLBI, NIH1;Department of Cell Biology, Lerner Research Institute NC-10, Cleveland Clinic.2 Cell migration requires coordinated assembly of focal adhesions and contraction in the actomyosin cytoskeleton. The small GTPase Rac1 is critical to cell migration through its known functions in regulation of focal adhesion and actin cytoskeletal assembly dynamics, but its role in regulation of myosin II is not known. Myosin II dynamically assembles into minifilaments at the leading edge of migrating cells, and PKC-mediated phosphorylation in Ser 1916 in the non-helical tail is one of the main regulators. We hypothesized that Rac1 may regulate myosin II minifilament assembly dynamics during cell migration via downstream regulation of PKC and Ser 1916 phosphorylation. To test this, we analyzed the effects of Rac1 activation on the phosphorylation and dynamics of myosin IIA in U2OS cells. We found that transfection of active Rac1 (Rac1V12) induced PKC- and integrin-dependent myosin IIA phosphoryation on Ser 1916. Live cell imaging of GFP-myosin IIA revealed that Rac1 activation promotes rapid assembly, motion, and turnover of myosin IIA minifilaments, as well as perpendicular orientation to the leading edge, resulting in its accumulation specifically in focal adhesions. To determine the role of Ser 1916 phosphorylation on myosin IIA dynamics and localization, we expressed phospho-mimetic (S1916D) and non-phosphorylatable mutants (S1916A) of myosin IIA. This showed that phosphorylation is critical to the Rac1-induced rapid assembly and turnover of myosin IIA minifilaments as well as to the focal adhesion association of myosin IIA. Thus, Rac1 acts as a master regulator of cell migration by coordinating actin assembly-mediated protrusion, adhesion, and actomyosin contraction dynamics. This work is being prepared for publication Project 2: PROTEOMIC ANALYSIS OF FOCAL ADHESION MATURATION Proteomic Analysis of Myosin II-mediated Focal Adhesion Maturation Reveals a Role for -Pix in Relaxation-mediated Rac1 Activation J. Kuo, X. Han, C.T Hsiao, J. Yates, C. M. Waterman Focal adhesions (FAs) undergo contraction-mediated maturation wherein they grow and change composition to differentially transduce signals from the extracellular matrix to modulate cell migration, growth and differentiation. To determine how FA protein composition is globally modulated by myosinII contraction, we developed a proteomics approach to isolate native FAs, identify their protein composition, and compare specific protein abundance in FAs from cells with and without myosinII inhibition. We reproducibly identified 905 FA-associated proteins, half (402) of which changed in FA abundance in response to perturbation of myosinII activity, thus defining the myosinII-responsive FA proteome. FA abundance of 75% of proteins in the myosinII-responsive FA proteome were enhanced by contractility, including those involved in Rho-mediated FA maturation, stress fiber formation, and endocytosis- and calpain-dependent FA disassembly. Surprisingly, 25% of the myosinII responsive FA proteome, including proteins involved in Rac-mediated lamellipodial protrusion, were enriched in FA by myosinII inhibition, establishing for the first time negative regulation of FA protein recruitment by contractility. We focused on the role of the Rac guanine nucleotide exchange factor, -PIX, documenting its depletion from FA during myosin-mediated FA maturation and its role in negative regulation of FA maturation to promote rapid FA turnover, lamellipodial protrusion and fast cell migration. A mthods paper describing our method was published this year. Project 3: Analysis of traction stress variation across single focal adhesions. Sergey V. Plotnikov, Benedikt Sabass, Clare M. Waterman The ability of eukaryotic cells to sense mechanical properties of the extracellular matrix (ECM) and to exhibit durotaxis (directed migration toward stiffer environments) is thought to underly many biological processes including angiogenesis, neurogenesis and cancer metastasis. ECM stiffness sensing is achieved by integrin-mediated focal adhesions (FA), protein assemblies that couple contractile actomyosin bundles to the plasma membrane and transmit force generated by the cytoskeleton to the ECM. Although it has been shown that both structural and signaling components of FA are crucial to translate mechanical cues into cell behavior, the molecular mechanism of mechanosensing remains unknown. Here we demonstrate that a molecular clutch, a mechanical link between the actin cytoskeleton and ECM-engaged integrins, acts as a mechanosensor in FAs, and the strength of the clutch determines range of ECM stiffness cells are able to sense to mediate durotaxis. Using high-resolution traction force microscopy on polyacrylamide ECMs of varying stiffnesses, we found that the mechanical behavior of the integrin-actin interface at FA exhibited ECM stiffness-dependent switching between a load-and-fail compliance sensing regime and a frictional slippage regime as described in the clutch oscillation model (Chan and Odde, 2008). In the load and fail regime, the position of peak traction within the FA resided on average at the distal FA tip, but oscillated over time towards the FA center and back to the tip. As ECM rigidity was increased, the traction peak did not oscillate and remained in the FA center, signifying the frictional slippage regime. We found that perturbing the gradient of paxillin phosphorylation across FA by expressing Y31/118E- or Y31/118F-paxillin mutants or by inhibiting FAK weakened the molecular clutch and switched FAs from load-and-fail compliance sensing to frictional slippage regime on compliant ECMs. In agreement with the clutch oscillation model, the load-and-fail regime could be rescued by further decreasing either substrate stiffness or myosin II contractility. Since paxillin phosphorylation on tyrosine residues 31 and 118 mediates vinculin recruitment into FAs, we demonstrated that vinculin and the paxillin-vinculin interaction are essential to strengthen the molecular clutch and to enable mechanosensing over a wide range of ECM compliances. We demonstrated the physiological importance of the load-and-fail compliance sensing regime by showing a requirement for this FA behavior in durotaxis, but not in chemotaxis in a boyden chamber assay or in random cell migration. This work resulted in 2 publications

Project 1: An RNAi screen of microtubule-regulatory proteins identifies MARK2/Par1 as an effector of Rac1-mediated microtubule growth. Yukako Nishimura, Kathryn Applegate, Gaudenz Danuser, Clare Waterman Proper regulation of microtubule (MT) assembly dynamics is essential for directed cell migration. Microtubule dynamics in migrating cells are spatially regulated by Rho GTPases. We have previously shown that activated Rac1 induces MT net growth by suppressing catastrophe and increasing growth velocity, and that Rac1 activity is required for polarized MT growth in the leading edge of migrating cells. We identified a necessary, but not sufficient PAK kinase-mediated pathway downstream of Rac1 that promoted MT growth. Therefore, we hypothesized that additional factors promote MT net growth downstream of Rac1. To find these factors, we performed a RNAi screen in human U2OS osteosarcoma cells to determine if known MT-regulatory proteins were required for constitutively activated Rac1 promotion of MT growth. To analyze MT dynamics, we imaged fluorescent-tagged EB3, a MT plus-end binding protein that serves as a probe for the position of MT ends, and tracked the motion of EB3 comets in time-lapse movies using an automated computer program. Our results indicate that depletions of several MT-binding proteins change the growth rate of MT in activated Rac1-expressing cells. We have focused on MARK2, a microtubule affinity-regulating kinase homologous to the C. elegans polarity protein Par1, whose depletion reduces the number of elongated MTs in the leading edge of Rac1-activated cells. We are currently testing how MARK2 is involved in promoting MT growth downstream of Rac1 and it requirement in cell migration. A manuscript describing these results was published in Plos One in 2012.Dr Nishimura moved on to a position at the National University of Singapore Project 2: MCAK Activity Controls Interphase Microtubule Dynamics and Directed Cell Migration. Myers, K.A.; Applegate, K; Danuser G.; and Waterman, C.M. Directional cell migration is initiated through extracellular stimuli that coordinate changes in the cytoskeleton to establish a polarized cellular morphology. Cell polarity can be achieved through regional regulation of microtubule (MT) dynamics, including MT growth toward the leading edge and MT shortening in the cell rear. Mitotic Centromere Associated Kinesin (MCAK) is a MT depolymerase that is down-regulated in mitosis by Aurora kinase phosphorylation. While its mitotic functions have been well-characterized, whether MCAK regulates MT dynamics during cell migration is not known. We hypothesize that MCAK is down-regulated locally via a Rac1/Pak1/Aurora-A kinase signaling pathway to establish preferential MT growth toward the leading edge and to promote MT shortening within the cell rear. To test this hypothesis, we performed time-lapse imaging of fluorescently tagged EB3 as a marker of MT plus end growth in HUVEC cells and analyzed MT dynamics and cell behavior under different manipulations of the proposed signaling cascade. We find that MCAK knockdown (KD) produces expected effects on the MT cytoskeleton, including increased levels of tubulin polymer and decreased MT catastrophe frequency. MCAK-KD cells show a reduction in MT polymerization speeds and exhibit a mal-oriented MT array, as well as a statistically significant reduction in cell migration velocity, directional persistence, and distance to origin, indicating a defect in cell migration and/or polarization. These effects are rescued through expression of exogenous wild-type-MCAK, but not by expression of either an inactive (ATPase-dead) MCAK mutant or an MCAK mutant that is incapable of phospho-regulation by Aurora-A kinase. Immunolabeling of cells expressing either constitutively active-Rac1 or constitutively active-Pak1 suggests that Rac1 and Pak1 activities correlate with increased Aurora-A activity, as assayed with a phospho-specific antibody, and also correlate with decreased levels of MCAK expression. These data suggest that interphase regulation of MCAK is achieved downstream of a Rac1/Pak1/Aurora-A signaling pathway in order to locally coordinate MCAK-mediated MT depolymerization as a method to ensure proper cell polarization and motility. A manuscript describing these studies is being prepared for publication. Dr Myers received a job as a tenure-track investigator at the University of the Sciences in Philadelphia and will finis this manuscrips as an NIH special volunteer. Project 3: Regulation of microtubules in migrating endothelial cells in 3D ECMs Ken Myers, Kathryn Applegate, Gaudenz Danuser, Robert Fischer, Clare Waterman Cytoskeletal dynamics driving endothelial cell (EC) branching morphogenesis during angiogenesis are thought to be regulated in part by cellular signals elicited in response to compliance and topology of the extracellular matrix (ECM) via a process termed mechanosensing. We hypothesized that ECM mechanosensing of compliance or topology (2 dimensional vs 3 dimensional ECMs, i.e. ECM dimensionality) could elicit different responses of the microtubule (MT) cytoskeleton to mediate EC branching morphogenesis. To test this, we used novel MT end-tracking software to analyze spatial variations in MT dynamics in ECs in 2D and 3D compliance-controlled ECMs during branching morphogenesis. Pharmacological inhibition showed that MT dynamics negatively regulate EC branching, independent of ECM compliance or dimensionality. Analyzing MT dynamics incompliant 2D and 3D ECMs with or without myosinII inhibition indicated that myosinII down-regulation by compliance mechanosensing promotes fast MT assembly, but we found dimensionality-specific effects on MT growth persistence. Comparing MT dynamics in cell bodies versus cell branches in in 2D and 3D ECMs of varying stiffness revealed faster, more dynamically unstable MT growth in EC bodies and slower, more persistent MT growth in EC branches. Thus, distinct compliance and dimensionality ECM mechanosensing pathways regionally regulate MT dynamics in ECs to guide branching morphogenesis in physically complex ECMs. A manuscript describing these results was published in the JCB.

The NHLBI Light Microscopy Core (LMC) has been in operation for fourteen years. The NHLBI-LMF consists of three people, Dr. Christian A. Combs (Director), Dr. Daniela A. Malide, and Dr. Xufeng Wu (Deputy Director) and twenty-one microscopes located in buildings 10 and 50 on the main NIH campus in Bethesda, MD. To date we have helped researchers publish more than 220 papers and assisted almost every NHLBI-DIR Center and Branch. Over the period of 2013 through July, 2014 we have helped to publish forty papers. Research conducted in the LMC has been on many disease states, basic cell biology, and on development and implementation of new imaging techniques. The general makeup of the NHLBI-LMF reflects the evolving light microscopy needs of NHLBI-DIR researchers and a commitment to provide a maximal level of assistance to the NHLBI mission. The mission of this facility is to provide state-of-the-art equipment, training, and image processing capabilities to assist NHLBI-DIR researchers in experiments involving light microscopy. Researchers that work in this facility can expect support from core personnel to whatever level suits their research. This can include advanced microscopy techniques like intra-vital two-photon microscopy and super-resolution microscopy to more ordinary wide-field imaging with a color camera. Our emphasis is on training users to conduct the experiments themselves, although we are available for collaboration and all manner of assistance (experimental planning, data analysis and image processing, etc.) where required. Over the fourteen years this facility has been in existence we have endeavored to provide a flexible and easy to use facility that meets researchers needs and allows them to conduct their experiments in an efficient manner even if they have had no prior microscopy experience. These goals are met in several ways. First, we have an array of microscopes that offer a wide-range of microscopy techniques. These microscopes and the microscopy techniques available have been chosen and developed in response to the specific needs of researchers in the institute. In addition to the microscopes, we provide a full suite of image processing programs in several image processing workstations. Where image-processing capabilities are lacking in these programs we develop our own image processing programs. In summary this facility operates with state of the art equipment and a dedicated team of imaging professionals to further the research mission of the NHLBI-DIR.