Session Y40: Focus Session: Cytoskeleton and Biomechanics - Role of Motors (Including DBIO Best Thesis Award Lecture)

Forces driving three-dimensional tissue patterning during morphogenesis

Dorsal closure is an essential stage of \textit{Drosophila} embryogenesis and is a model system for \textit{in-vivo} investigations of cell sheet morphogenesis. During closure a system of four biological processes work collectively to close a gap in the epithelium, which initially is filled with a transient tissue. The geometry of the dorsal opening is similar to that of two intersecting circular arcs being pulled apart at a nearly constant rate. Substantial progress in understanding the dynamics has been made in the past by largely viewing closure as a two-dimensional process. However, tissue and cell dynamics are not confined to the embryo surface. We have been investigating the three-dimensional kinematics of dorsal closure by imaging the actomyosin purse strings at the periphery of the dorsal opening and by imaging the apical belts of DE-cadherin in each cell within the opening. We have analyzed the results with the methods of analytic geometry. In addition, in the past we have determined the relative magnitudes of the forces that drive dorsal closure. We have been using magnetic tweezers, time-resolved \textit{in-vivo} microscopy, and biophysical modeling to measure the net force and to determine the magnitude of each force.   

Cell Autonomous Shape Changes in Germband Retraction

Germband retraction involves the cohesive movement and regulated cellular mechanics of two tissues on the surface of fruit fly embryos, the germband and the amnioserosa. The germband initially forms a `U' shape, curling from the ventral surface, around the posterior of the embryo, and onto the dorsal surface; the amnioserosa lies between the arms of this `U'. Retraction straightens the germband and leaves it only on the ventral side. During retraction, the germband becomes clearly segmented with deep furrows between segments, and its cells elongate towards the amnioserosa, along what becomes the dorsal-ventral axis. To determine the importance of these changes for the overall movement of the tissues, we observed embryos that did not complete germband retraction due to targeted laser ablation of half the amnioserosa. Without the chemical and mechanical influence of the amnioserosa, germband furrows still formed and germband cells still elongated; however, this elongation was misaligned compared to unablated embryos. Thus, furrow formation and cell elongation in the germband are autonomous, but insufficient to drive proper tissue motion. These results suggest that part of the necessary role of the amnioserosa is proper orientation of germband cell elongation.   

Wavefronts and mechanical signaling in early Drosophila embryos

Mitosis in the early syncytial Drosophila embryo has a high degree of spatial and temporal correlations, visible as mitotic wavefronts that travel across the embryo. This mitosis wavefront is preceded by another wavefront which corresponds to chromosome condensation. The two wavefronts are separated by a time interval that is independent of cell cycle and propagate at the same speed for a given embryo in a given cycle. We study the wavefronts in the context of excitable medium theory, using two different models, one with biochemical signaling and one with mechanical signaling. We find that the dependence of wavefront speed on cell cycle number is most naturally explained via a mechanical signaling, and that the entire process suggests a scenario in which biochemical and mechanical signaling are coupled.   

Passive cellular microrheology in developing fruit fly embryos

The development of fruit fly (\textit{Drosophila)} embryos involves spatial and temporal regulation of cellular mechanical properties. These properties can be probed \textit{in vivo }using laser hole drilling experiments; however, this technique only infers relative forces. Conversion to absolute forces requires measurement of cellular viscoelastic properties. Here, we use passive microrheology of fluorescently labeled cell membranes to measure the viscoelastic properties of amnioserosa cells. These dynamic epithelial cells play an important mechanical role during two developmental stages: germ band retraction and dorsal closure. Passive microrheology in this system is confounded by active contractions in the cytoskeleton. Thus, the fruit fly embryos are transiently anesthetized with CO$_{2}$, halting active cellular movements, leaving only passive Brownian motion. The power spectra of these fluctuations are well fit by a Lorentzian -- as expected for Brownian motion -- and allow us to extract cellular viscoelastic parameters at different developmental stages. These measured parameters inform previous hole-drilling experiments and provide inputs for quantitative computational models of fruit fly embryonic development.   

Remodeling of cellular cytoskeleton drives tissue level morphogenesis

Mechanical stresses are central to morphogenesis, both as a cause that generates geometric and topological change, and as regulatory signals that couple cells. Live imaging of fluorescently tagged tissues gives us insight into the cellular processes underlying tissue dynamics during morphogenesis. Amongst these is the remodeling of the cytoskeleton and cellular adhesion. Here, following observations from \textit{drosophila} germ band extension and ventral furrow formation, we a) investigate the mechanical state of the tissue b) perform a quantitative analysis and verification of the cell and tissue level stresses and c) determine how conserved cellular processes are regulated to generate tissue level stresses that drive morphogenesis.   

Investigation of autonomous cell dynamics using holographic laser microsurgery

Laser-microsurgery has emerged as a powerful technique for evaluating \textit{in vivo} tissue mechanics. We extend this technique by using a spatial light modulator (SLM) to diffract a single UV laser pulse to simultaneously ablate multiple points in living tissue. Using this method, we can quickly and cleanly isolate a single cell by destroying all its nearest neighbors. The post-ablation dynamics of such an isolated cell are then largely dependent on autonomous intracellular forces. Here, we use this technique to investigate cell shape pulsations in amnioserosa cells in \textit{Drosophila} embryos during dorsal closure -- specifically to address the degree to which these pulsations are cell autonomous or driven by the contractions of neighboring cells. When cells are isolated at different points in the pulsation cycle, we find that the post-isolation dynamics are strongly dependent on the pre-isolation pulsation phase: cells in a contractile phase collapse immediately, but cells in an expansionary phase continue to expand -- at least for 20-60 s before collapsing. These results are in conflict with previous pulsation models that place expanding cells under large extensional strain, and instead suggest that even the expansion phase has a significant cell autonomous component.   

DBIO Best Thesis Award: Mechanics, Dynamics, and Organization of the Bacterial Cytoskeleton and Cell Wall

Bacteria come in a variety of shapes. While the peptidoglycan (PG) cell wall serves as an exoskeleton that defines the static cell shape, the internal bacterial cytoskeleton mediates cell shape by recruiting PG synthesis machinery and thus defining the pattern of cell-wall synthesis. While much is known about the chemistry and biology of the cytoskeleton and cell wall, much of their biophysics, including essential aspects of the functionality, dynamics, and organization, remain unknown. This dissertation aims to elucidate the detailed biophysical mechanisms of cytoskeleton guided wall synthesis. First, I find that the bacterial cytoskeleton MreB contributes nearly as much to the rigidity of an \textit{Escherichia coli} cell as the cell wall. This conclusion implies that the cytoskeletal polymer MreB applies meaningful force to the cell wall, an idea favored by theoretical modeling of wall growth, and suggests an evolutionary origin of cytoskeleton-governed cell rigidity. Second, I observe that MreB rotates around the long axis of \textit{E. coli}, and the motion depends on wall synthesis. This is the first discovery of a cell-wall assembly driven molecular motor in bacteria. Third, I prove that both cell-wall synthesis and the PG network have chiral ordering, which is established by the spatial pattern of MreB. This work links the molecular structure of the cytoskeleton and of the cell wall with organismal-scale behavior. Finally, I develop a mathematical model of cytoskeleton-cell membrane interactions, which explains the preferential orientation of different cytoskeleton components in bacteria.   

Coupling of Active Motion and Advection Shapes Intracellular Cargo Transport

Three different mechanisms can contribute to intracellular cargo transport: (1) passive diffusion, (2) active motor-driven transport along cytoskeletal filament networks and (3) passive advection by fluid flows. Active and advective transport are coupled because cytoplasmic flows can arise through entrainment of the fluid that surrounds actively moving cargo on cytoskeletal networks. Here, we report a reaction-advection-diffusion model for transport of a passive mass-conserved scalar that can cycle between a bound state, where advection represents active transport on a cytoskeletal network, and an unbound state, where the advecting fluid flow field is driven by forces from the cytoskeletal network. Cargo transport and localization patterns are explored for different cytoskeletal network topologies and for varying reaction kinetics. We find that for sufficiently low diffusion, localization of cargo to a target area is optimized either by low reaction kinetics and decoupling of bound and unbound state, or by a mostly disordered cytoskeletal network with only weak directional bias. The principles exemplified by this model likely have implications for our understanding and interpretation of transport patterns and cytoskeletal network structures.   

Motion in partially and fully cross-linked F-actin networks

Single molecule experiments have measured stall forces and procession rates of molecular motors on isolated cytoskeletal fibers in Newtonian fluids. But in the cell, these motors are transporting cargo through a highly complex cytoskeletal network. To compare these single molecule results to the forces exerted by motors within the cell, an evaluation of the response of the cytoskeletal network is needed. Using magnetic tweezers and fluorescence confocal microscopy we observe and quantify the relationship between bead motion and filament response in F-actin networks both partially and fully cross-linked with filamin We find that when the transition from full to partial cross-linking is brought about by a decrease in cross-linker concentration there is a simultaneous decline in the elasticity of the network, but the response of the bead remains qualitatively similar. However, when the cross-linking is reduced through a shortening of the F-actin filaments the bead response is completely altered. The characteristics of the altered bead response will be discussed here.   

Measurements of the constituent contributions to the physical properties of fibroblast populated collagen microtissues with magnetic micro-tissue stretchers

The mechanical properties of fibroblast populated collagen matrix (FPCM) provide important physical cues to regulate physiological and pathological processes of encapsulated cells. The mechanical strength of FPCM is arises from both of its constituents: the collagen matrix and the fibroblasts. Existing methods to separate the contribution of individual constituents by treating cm-scale tissue samples with decellularization drugs for prolonged periods have been shown to adversely affect the properties of the collagen matrix. To minimize such matrix damage, we have developed a magnetic microtissue stretching system that allows us to grow arrays of sub-mm scale microtissues that can be rapidly decellularized. This consists of arrays of paired micro-cantilevers that support the 3D FPCM and can be driven by incorporated magnetic material via externally applied magnetic fields. By measuring the tensile force applied to the FPCM and the tissue strain, we found the stiffness of the matured FPCM is 28.1 +- 1.8 kPa and that of the decellularized collagen matrix is 23.1 +- 3.1 kPa. These measurements of the stiffness of the intact collagen matrix in a remodeled FPCM can provide important clues on the mechanical environment that regulates the biological function of encapsulated cells.   

Crowding of molecular motors determines microtubule depolymerization

Assembly and disassembly dynamics of microtubules (MTs) is tightly controlled by MT associated proteins. Here, it is investigated how plus-end-directed depolymerases of the kinesin-8 family regulate MT depolymerization dynamics. We reproduce experimental findings within the framework of a totally asymmetric simple exclusion process with Langmuir kinetics (TASEP/LK). Thereby, crowding is identified as the key regulatory mechanism of depolymerization dynamics. The analysis gives two qualitatively distinct phases. For motor densities above a particular threshold, a macroscopic traffic jam emerges at the plus-end and the MT dynamics become independent of the motor concentration. Below this threshold, microscopic traffic jams at the tip arise which cancel out the effect of the depolymerization kinetics such that the depolymerization speed is determined by the motor density. Because this density varies over the MT length, length-dependent regulation is possible. The critical length at which MTs start to depolymerize in a length-dependent way is discussed. Reference: Louis Reese, Anna Melbinger and Erwin Frey. Biophys. J. 101, 2190 (2011)   

Physics of the actin cortex in shape oscillations of dividing cells

During cytokinesis, a contractile actomyosin cortex is present at the poles of the dividing cell. For a large enough tension of the polar cortex, a physical symmetry-breaking instability of the cell shape can occur where one pole transfers its volume to the other. Such a shape instability can indeed be observed in control and treated dividing cultured cells where it results in cell shape oscillations and in some cases leads to cytokinesis failure. The cell oscillation properties can be accurately described with a theoretical model based on a competition between cortex turnover and contraction dynamics. Interestingly, our results indicate that a sufficiently large cell elasticity is needed to ensure successful cytokinesis.   

Geometrical Reorganisation of the Cytoskeleton and Changes in Cellular Stiffness Following Stretch

Cells in the lung and the vasculature are under a highly dynamic mechanical environment where they are constantly exposed to stretch. Cells adapt to these fluctuations in stretch by remodeling their cytoskeleton. However, the influence of these geometrical changes on cell stiffness is not well understood. We developed a computational model to simulate the geometrical reorganization of the actin by non-muscle myosin-II under conditions of monotonous cyclic stretch, where amplitude and frequency is constant from cycle to cycle and variable stretch, where the amplitude is varied from cycle-to-cycle and the frequency is inversely proportional to amplitude. With the monotonous cyclic-stretch, the network exhibited significant hysteresis in geometry, it reorganized itself into a more stable configuration and the internal prestress decreased after each cycle. In contrast, the more realistic variability in stretch amplitude prevented these stable configurations from forming and preserved the prestress. This behavior was dependent on the variability in stretch amplitude and the timing of the large amplitude stretches. We conclude that prestress is a consequence of cytoskeletal reorganization which exhibits structural hysteresis and is dependent on the nature of the stretch pattern.