Session L10: Principles of Cellular Remodeling

Disordered Actomyosin Is Sufficient to Promote Cooperative and Telescopic Contractility

While the molecular interactions between myosin motors and F-actin are well known, the relationship between F-actin organization and myosin-mediated force generation remains poorly understood. Here, we explore the accumulation of myosin-induced stresses within a 2D biomimetic model of the actomyosin cortex, where myosin activity is controlled spatially and temporally using light. By controlling the geometry and the duration of myosin activation, we show that contraction of disordered actomyosin is highly cooperative, telescopic with the activation area and generates a pattern of mechanical stresses consistent with those observed in contractile cells. We quantitatively reproduce these properties using an in vitro isotropic model of the actomyosin cytoskeleton, and explore the physical origins of telescopic contractility in disordered networks using agent-based simulations.   

3D Model of Cytokinetic Contractile Ring Assembly: Node-Mediated and Backup Pathways

Cytokinetic ring assembly in model organism fission yeast is a dynamic process, involving condensation of a network of actin filaments and myosin motors bound to the cell membrane through cortical nodes. A 3D computational model of ring assembly illustrates how the combined activities of myosin motors, filament crosslinkers and actin turnover lead to robust ring formation [Bidone et al. Biophys. J, 2014]. We modeled the importance of the physical properties of node movement along the cell membrane and of myosin recruitment to nodes. Experiments by D. Zhang (Temasek Life Sciences) show that tethering of the cortical endoplasmic reticulum (ER) to the plasma membrane modulates the speed of node condensation and the degree of node clumping. We captured the trend observed in these experiments by changes in the node drag coefficient and initial node distribution in simulations PM [Zhang, Bidone, and Vavylonis, Curr. Biol. (2016)]. The model predicted that reducing crosslinking activities in ER tethering mutants with faster node speed enhances actomyosin clumping. We developed a model of how tilted and/or misplaced rings assemble in cells that lack the node structural component anillin-like Mid1 and thus fail to recruit myosin II to nodes independently of actin. If actin-dependent binding of diffusive myosin to the cortex is incorporated into the model, it generates progressively elongating cortical actomyosin strands with fluctuating actin bundles at the tails. These stands often close into a ring, similar to observations by the group of J.Q. Wu (The Ohio State University).   

Patterned Cell Alignment in Response to Macroscale Curvature

The formation of spatial behavior patterns in tissues is a long-standing problem in biology. Decades of research have focused on understanding how biochemical signaling and morphogen gradients establish cell patterns during development and tissue morphogenesis. Here, we show that geometry and physical cues can drive organization and pattern formation. We find that mouse embryonic fibroblasts and human vascular smooth muscle cells sense curvature differently when in monolayers than when isolated on surfaces with various amounts of Gaussian curvature. While the long, apical stress fibers within these cells align in the direction of minimum curvature on cylindrical substrates, a subpopulation of stress fibers beneath the nucleus aligns in the circumferential direction and is bent maximally. We find dramatic reorganization of the actin cytoskeleton upon activation of RhoA, which is associated with increased contractility of the fibers. Thus, stress fiber alignment is likely a result of a complex balance between energy penalties associated with stress fiber bending, contractility, and the dynamics of F-actin assembly.   

Mechanical Coordination of Single-Cell and Collective-Cell Amoeboid Migration

Amoeboid migration consists of the sequential repetition of pseudopod extensions and retractions driven by actin polymerization and actomyosin contraction, and requires cells to apply mechanical forces on their surroundings. We measure the three-dimensional forces exerted by chemotaxing Dictyostelium cells, and examine wild-type cells as well as mutants with defects in contractility, F-actin polymerization, internal F-actin crosslinking, and cortical integrity. We find that cells pull on their substrate adhesions using two distinct, yet interconnected mechanisms: axial actomyosin contractility and cortical tension. The 3D pulling forces generated by both mechanisms are internally balanced by an increase in cytoplasmic pressure that allows cells to push on their substrate, and we show that these pushing forces are relevant for cell invasion and migration in three-dimensional environments. We observe that cells migrate mainly by forming two stationary adhesion sites at the front and back of the cell, over which the cell body moves forward in a step-wise fashion. During this process, the traction forces at each adhesion site are switched off and subsequently their direction is reversed. The cell migration speed is found to be proportional to the rate at which cells are able regulate these forces to produce the cell shape changes needed for locomotion, which is increased when axial contractility overcomes the stabilizing effect of cortical tension. This spatiotemporal coordination is conserved in streams of multiple migratory cells connected head to tail, which also migrate by exerting traction forces on stationary sites. Furthermore, we observe that trailing cells reuse the adhesion sites of the leading cells. Finally, we provide evidence that the above modes of migration may be conserved in a range of other amoeboid-type moving cells such as neutrophils.   

Force Dynamics During T Cell Activation

T cell activation is an essential step in the adaptive immune response. The binding of the T cell receptor (TCR) with antigen triggers signaling cascades and cell spreading. Physical forces exerted on the TCR by the cytoskeleton have been shown to induce signaling events. While cellular forces are known to depend on the mechanical properties of the cytoskeleton, the biophysical mechanisms underlying force induced activation of TCR-antigen interactions unknown. Here, we use traction force microscopy to measure the force dynamics of activated Jurkat T cells. The movements of beads embedded in an elastic gel serve as a non-invasive reporter of cytoskeletal and molecular motor dynamics. We examined the statistical structure of the force profiles throughout the cell during signaling activation. We found two spatially distinct active regimes of force generation characterized by different time scales. Typically, the interior of the cells was found to be more active than the periphery. Inhibition of myosin motor activity altered the correlation time of the bead displacements indicating additional sources of stochastic force generation. Our results indicate a complex interaction between myosin activity and actin polymerization dynamics in producing cellular forces in immune cells.   

Contractile forces originating from Cancer Diskiod regulated by geometrical ECM properties.

Cancer cell migration in three-dimensional extracellular matrix is a major cause of death for cancer patients. Although extensive studies have enlightened detailed mechanism of single cell 3D invasion and cell-ECM interaction, 3D collective cancer invasion is still poorly understood. To capture collective cancer invasion with more realistic, we developed a novel 3D invasion assay, Diskiod In Geometrically Micropatterned ECM (DIGME). DIGME allows us to independently controlling the shape the shape of tumor organoids, microstructure and spatial heterogeneity of the extracellular matrix all at the same time. Here we study the affect of contractile forces originating from different geometrical cancer diskiods. We show that cancer invasion could be regulated by geometrical ECM properties.   

Matrix remodeling between cells and cellular interactions with collagen bundle

When cells are surrounded by complex environment, they continuously probe and interact with it by applying cellular traction forces. As cells apply traction forces, they can sense rigidity of their local environment and remodel the matrix microstructure simultaneously. Previous study shows that single human carcinoma cell (MDA-MB-231) remodeled its surrounding extracellular matrix (ECM) and the matrix remodeling was reversible. In this study we examined the matrix microstructure between cells and cellular interaction between them using quantitative confocal microscopy. The result shows that the matrix microstructure is the most significantly remodeled between cells consisting of aligned, and densified collagen fibers (collagen bundle)., the result shows that collagen bundle is irreversible and significantly change micromechanics of ECM around the bundle. We further examined cellular interaction with collagen bundle by analyzing dynamics of actin and talin formation along with the direction of bundle. Lastly, we analyzed dynamics of cellular protrusion and migrating direction of cells along the bundle.   

Role of cell division and self-propulsion in self-organization of 2D cell co-cultures

Self-organization of cells is a key process in developmental and cancer biology. The differential adhesion hypothesis (DAH), which assumes cells as equilibrium liquid droplets and relates the self-assembly of cells to differences in inter-cellular adhesiveness, has been very successful in explaining cellular organization during morphogenesis where neighboring cells have the same non-equilibrium properties (motility, proliferation rate). However, recently it has been experimentally shown that for a co-culture of two different cell types proliferating at different rates, the resulting spatial morphologies cannot be explained using the DAH alone. Motivated by this, we develop and study a two-dimensional model of a cell co-culture that includes cell division and self-propulsion in addition to cell-cell adhesion, and systemically study how cells with significantly different adhesion, motility, and proliferation rate dynamically organize themselves in a spatiotemporal and context-dependent manner. Our results may help to understand how differential equilibrium and non-equilibrium properties cooperate and compete leading to different morphologies during tumor development, with important consequences for invasion and metastasis   

Vertex stability and topological transitions in vertex models of foams and epithelia

Vertex models are widely used to computationally simulate dry foams and epithelial tissues. This class of models describes the shape and motion of cells as a function of the forces on vertices where 3 or more cells meet.~ Despite the widespread use of these models, relatively little is known about their basic theoretical properties.~ One outstanding issue is the stability of fourfold vertices. In real foams, fourfold vertices are always unstable, but it has been unclear whether vertex models necessarily reflect this behavior. In biological tissues, fourfold vertices arise as an intermediate in T1 transitions, which are one of the fundamental processes by which tissues change topology, and stable fourfold vertices have recently been observed in several different epithelia. We show that, when all edges have the same tension, stationary fourfold vertices in vertex models must always break up.~ However, when tensions depend on edge orientation, as they might in a planar-polarized tissue, fourfold vertices can become stable.~ These findings pave the way for studies of more biologically realistic models that couple topological transitions to the dynamics of regulatory proteins.   

Flocking Transition in Confluent Tissues

The emerging of collective migration in biological tissues plays a pivotal role in embryonic morphogenesis, wound healing and cancer invasion. While many aspects of single cell movements are well established, the mechanisms leading to coherent displacements of cohesive cell groups are still poorly understood. Some of us recently proposed a Self-Propelled Voronoi (SPV) model of dense tissues that combines self-propelled particle models and vertex models of confluent cell layers and exhibits a liquid-solid transition as a function of cell shape and cell motility[1]. We now examine the role of cell polarization on collective cell dynamics by introducing an orientation mechanism that aligns cell polarization with local cell motility. The model predicts a density-independent flocking transition tuned by the strength of the aligning interaction, with both solid and liquid flocking states existing in different regions of parameter space. \\ \\ MP and MCM were supported by the Simons Foundation Targeted Grant in the Mathematical Modeling of Living Systems Number: 342354 and by the Syracuse Soft Matter Program. \\ \\ [1] D. Bi, {\itshape et al.}, Phys. Rev. X {\bf{6}}, 021011 (2015).   

Collective cell behavior on basement membranes floating in space

The basement membrane is an essential part of the polarity of endothelial and epithelial tissues. In tissue culture and organ-on-chip devices, monolayer polarity can be established by coating flat surfaces with extracellular matrix proteins and tuning the trans-substrate permeability. In epithelial 3D culture, spheroids spontaneously establish inside-out polarity, morphing into hollow shell-like structures called acini, generating their own basement membrane on the inner radius of the shell. However, 3D culture approaches generally lack the high degree of control provided by the 2D culture plate or organ-on-chip devices, making it difficult to create more faithful \textit{in vitro} tissue models with complex surface curvature and morphology. Here we present a method for 3D printing complex basement membranes covered in cells. We 3D print collagen-I and Matrigel into a 3D growth medium made from jammed microgels. This soft, yielding material allows extracellular matrix to be formed as complex surfaces and shapes, floating in space. We then distribute MCF10A epithelial cells across the polymerized surface. We envision employing this strategy to study 3D collective cell behavior in numerous model tissue layers, beyond this simple epithelial model.   

Contractile recovery of microtissues after giant shear events

Cells are often dispersed in extracellular matrix (ECM) gels like collagen and Matrigel as minimal tissue models. Generally, large-scale contraction of these constructs is observed, in which the degree of contraction of the entire system correlates with cell density and ECM concentration. The freedom to perform diverse mechanical experiments on these contracting constructs is limited by the challenges of handling and supporting these delicate samples. Here, we present a method to create simple cell-ECM constructs that can be manipulated with significantly reduced experimental limitations. We 3D print mixtures of MCF10A cells and ECM (collagen-I and Matrigel) into a 3D growth medium made from jammed microgels. With this approach, we are able to apply shear stresses to the cell constructs times after printing and observe the collective response. Preliminary results reveal that, following shear deformations that exceed 300{\%} and dramatically smear cells and matrix in space, the cells actively re-contract the construct toward the un-sheared construct. These results suggest that new principles of collective recovery can be employed for tissue engineering applications using jammed microgels as a re-configurable support medium.   

A Reaction-Diffusion Model for Synapse Growth and Long-Term Memory

Memory storage involves strengthening of synaptic transmission known as long-term potentiation (LTP). The late phase of LTP is associated with structural processes that enlarge the synapse. Yet, synapses must be stable, despite continual subunit turnover, over the lifetime of an encoded memory. These considerations suggest that synapses are variable-size stable structure (VSSS), meaning they can switch between multiple metastable structures with different sizes. The mechanisms underlying VSSS are poorly understood. While experiments and theory have suggested that the interplay between diffusion and receptor-scaffold interactions can lead to a preferred stable size for synaptic domains, such a mechanism cannot explain how synapses adopt widely different sizes. Here we develop a minimal reaction-diffusion model of VSSS for synapse growth, incorporating the recent observation from super-resolution microscopy that neural activity can build compositional heterogeneities within synaptic domains. We find that introducing such heterogeneities can change the stable domain size in a controlled manner. We discuss a potential connection between this model and experimental data on synapse sizes, and how it provides a possible mechanism to structurally encode graded long-term memory.