Session F11: Physics of Cytoskeleton Across Scales III

Live

Actin and microtubule filaments make a complex interacting network in the cell. This composite network has an essential role in various cell functions such as cell mobility and cell division. Here, we use fluorescent microscopy, fluctuations, and correlation analysis to explore the effect of the actin and microtubule crosslinkers in the cytoskeletal composite network. We systematically vary the amount of actin and microtubule crosslinkers, biotin–NeutrAvidin and MAP65 respectively, and measure the organization and fluctuation of the filaments. We find that the microtubule crosslinker plays the principle role in the organization of the system, while actin crosslinking tunes the mobility of the filaments. We previously showed that the fluctuations of filaments are related to the mechanics, implying that actin crosslinking controls the mechanical properties of the composite network.   

Live

Tubulin is a ubiquitous biomolecule in eukaryotes, known to assemble in a variety of 2D shapes. The most studied of such are microtubules (MTs), slender but rigid cylindrical aggregates of tubulin chains, which are found with some structural variability. Microtubules undergo constant switching between growth and shrink phases, in a process known as dynamic instability.
When suddenly switching to a shrinking state during a catastrophe, the tubulin chains, known as protofilaments, splay apart and peel off from the microtubule tip, curving outwards. A spontaneous curvature of the protofilaments is therefore hypothesized to play a part in the dynamic instability of MTs. As their mechanical properties are influenced by the environmental conditions, such curvature terms are likely also responsible for the fascinating shape variability observed for tubulin assemblies beyond cylindrical MTs: flat sheets, ribbons, macrotubules and hoops.
By modelling tubulin aggregates as continuous elastic materials, we investigate how spontaneous curvature and anisotropic elastic properties influence the resulting equilibrium shape. Interestingly, this provides an insight on the role of mechanics in microtubule catastrophe.   

Live

Co-entangled actin-microtubule networks can serve as tunable substrates for self-driven motor proteins and facilitate mesoscale biological behavior such as mitosis, meiosis, cytoplasmic streaming, and regeneration in the in-vivo, non-equilibrium environment. In-vitro characterization of actin-microtubule networks have revealed tunable emergent mechanical properties differing from those of networks containing either component alone. However, how non-equilibrium, motor-driven activity impacts the mechanical response of actin-microtubule networks remains unknown. To address this gap, we perform nonlinear optical tweezers microrheology measurements on active actin-microtubule networks driven by the molecular motor myosin II. We characterize the mesoscale mechanical response of the non-equilibrium networks as a function of activity time and in varying locations throughout the networks to determine the spatial distribution of responses. Our methods are applicable to a wide range of active matter systems and our results establish guiding principles towards the creation of tractable bio-inspired active materials.   

Live

The cytoskeleton, composed of actin, microtubules, and associated motor and binding proteins, is an actively rearranging network with tunable structure and mechanics. For example, myosin motors can induce contraction, extension and flow of actin networks. While actomyosin networks have been well-characterized, the impact of microtubules on actomyosin dynamics remains poorly understood. Here, we create active composite networks of actin and microtubules that exhibit contractile dynamics driven by myosin II. We tune the dynamics, activity and structure of composites by varying the relative concentrations of actin, microtubules, and myosin. Using multi-spectral confocal microscopy, along with differential dynamic microscopy (DDM), particle image velocimetry (PIV), and spatial image autocorrelation (SIA) analyses, we characterize how network composition affects the non-equilibrium dynamics and structure over time. Specifically, we use DDM and PIV to quantify the rate and directionality of network motion, and we use SIA to quantify time-varying network correlation length scales. These results provide new insight into the diverse interactions cells can use to tune activity, as well as the design of tunable active materials.   

Live

A large molecular machine called the mitotic spindle segregates chromosomes during eukaryotic cell division. The spindle of the fission yeast Schizosaccharomyces pombe consists of a bundle of microtubules (protein filaments) within an intact nuclear envelope. During spindle elongation, motor proteins slide antiparallel microtubules apart while resisting the nucleus's compressive forces. We probe this force balance by severing the spindle via laser ablation. Similar to previous studies [1,2], we find that after cutting the spindle in half, the fragments rapidly collapse towards each other, often reattach, and resume elongation. While this behavior has been previously observed, many questions remain about its dynamics, mechanics, and molecular requirements. In this work, we find that previously hypothesized viscoelastic relaxation of the nucleus cannot fully explain spindle shortening in response to laser ablation. Instead, spindle collapse requires microtubule dynamics and is powered at least partly by the motor protein dynein. These results suggest a role for dynein in redundantly supporting force balance and bipolarity in the S. pombe spindle.
[1] Khodjakov et al. Curr Biol 2004. [2] Tolic-Nørrelykke et al Curr Biol 2004.   

Live

How cells control the size and morphology of their internal structures is an open and central question in biology. While the size of most intracellular structures scales with cell size - larger cells tend to have a larger nucleus and mitotic spindles - it is unclear what aspect of cell size sets the scaling and how. Here, we provide a mechanistic explanation for scaling the first mitotic spindle with cell size in C. elegans. We used a combination of quantitative microscopy and biophysical perturbations to establish the importance of a balance of cortical pulling forces in the regulation of spindle length and dynamics. This led us to construct a model for which the stoichiometric interactions of microtubules and cortical force generators (each force generator can bind only one microtubule) is key to explaining the dynamics of spindle positioning, its elongation, and length scaling with cell length. This model accounts for variations in all the spindle traits we studied here, both within species and across nematode species spanning over 100 million years of evolution.   

Live

Actin networks at the leading edge of cells generate protrusion forces against cell membranes to facilitate cell migration and cytoskeletal deformation. There is an emerging paradigm for regulation in the actin networks whereby components are passively sorted to their correct locations due to network growth. The tradeoff between actin growth speeds and network patterns under non-equilibrium driving is still a subject of active investigation. Here we use the framework of stochastic thermodynamics to investigate how the actin polymerization rates tune the sorting of actin binding proteins(ABPs) in a growing actin bundle. Our main results show that thermodynamics can be used to elucidate the role of actin polymerization in the sorting of ABPs in the growing actin bundle. In particular, we derive a thermodynamic uncertainty relation between non-equilibrium driving, emergent actin structure, and growth speed. Our work can be viewed as a starting point that reveals how the energy dissipation accompanying polymerization dynamics governs the emerging structure of networks formed by force transmitting biological agents.   

Live

Katanin is a microtubule (MT) severing motor from the AAA+ family that forms asymmetric spiral hexamers, which undergo a conformational change to a ring structure through ATP hydrolysis. The presence of ATP and of the carboxy terminal tails (CTTs) of tubulin dimers in a MT is required for oligomerization of severing proteins prior to severing. The CTT binds to loops found in the central pore of the hexamer. Studies showed that the hexamer dissociates into lower order oligomers in the absence of at least one binding partner, but there is no consensus on the order(s) of the oligomers, beyond being up to three, nor on the dissociation dynamics. To address these questions, we carried out molecular dynamics simulations of oligomeric states of katanin, starting from recently solved cryo-EM structures of the hexamers. A central finding is that for hexameric katanin the absence of the ATP and CTT leads to the dissociation of the central pore loops into trimers. We use the combination of structural, kinetic, and Machine Learning approaches to shed light on the stability of lower order oligomers, and the influence of the nucleotide and the substrate on the networks of interactions responsible for the dynamic stability of oligomers.   

Live

Many cellular functions such as mitosis, mechanics and intracellular transport rely on the organization and interaction of actin filaments, microtubules (MTs) and intermediate filaments (IFs), which are the main constituents of the eukaryotic cytoskeleton. Here, we study the interaction between vimentin IFs and MTs in a minimal in vitro system and show that MTs are stabilized against depolymerization by the IFs. To explore the nature of this interaction and probe for electrostatic and hydrophobic contributions, we directly measure forces between individual MTs and vimentin IFs using optical tweezers in different buffer conditions. Theoretical modeling results in the corresponding energy landscape. Feeding back the physical parameters describing the interaction into a Monte Carlo simulation that mimicks dynamic MTs indeed confirms that the additional interaction with IFs stabilizes MTs. We suggest that within cells, the interactions we observe might be a mechanism for cells to fine-tune cytoskeletal crosstalk and MT stability.
doi.org/10.1101/2020.05.20.106179   

Live

The mitotic spindle, by exerting forces, segregates chromosomes into two daughter cells during cell division. During metaphase, chromosomes are positioned in the spindle equatorial plane, which is necessary to prevent lagging chromosomes and abnormal nuclear envelope reformation. It has been proposed that two centering mechanisms play a key role here, microtubule catastrophe promoted by kinesin-8 motors and pushing forces exerted by chromokinesins. Here we show, by combining a theoretical model and quantitative experiments, that kinetochore microtubules cross-linked by bridging microtubules exert length-dependent centering pulling forces. The signature of this centering mechanism is larger poleward flux velocities of bridging fiber as compared to the velocities of kinetochore fiber, which we confirmed in a preliminary experiment. In conclusion, we propose that antiparallel overlaps exert length-dependent forces on kinetochores to navigate their positioning in the center of the metaphase plate.   

Live

Active biological systems such as the cytoskeleton are known to exhibit a wide range of interaction types and are subject to a variety of external stimuli. Here we aim to find the metrics that govern the formation and stability of structures in the cytoskeleton. Previous work on non-equilibrium liquids [Tociu et al. PRX 2018] demonstrated that structure and dynamics can be altered by changing the rate at which the system dissipates energy. Biasing trajectory probabilities with a function related to energy dissipation had the same effect as increasing the interaction strength between individual particles or introducing an external driving force. We extrapolate this concept to systems of cross-linked filaments by considering a similar connection between structure and thermodynamic quantities such as energy dissipation. Through molecular dynamics simulations, we compare the effects of independently tuning component properties and thermodynamic quantities on system phase behavior. Quantifying the relationship between system-wide properties and those of individual components may lead to improved understanding of underlying principles of cytoskeletal structure and dynamics.   

Live

For many cell functions, a concerted effort of several membrane proteins is needed, for example in signaling, division, and endocytosis. Besides specific protein-protein interactions and interactions with the cytoskeleton, protein organization in membranes is thought to be driven by a universal interaction force arising from membrane deformations. However, the small size and the inherent complexity of the proteins’ shape and interactions make it difficult to experimentally measure this interaction. We recently developed an experimental model system consisting of Giant Unilamellar Vesicles and adhesive colloids which allows us to quantitatively study these interactions. Here, we use this setup to investigate the many-body interactions that arise from three membrane-deforming spheres on a GUV. We quantify their interactions and arrangements and conclude that there exist two favorable configurations on the membrane: (1) a linear, and (2) a triangular configuration. These observations demonstrate the non-additive nature of membrane-deformation induced interactions.   

Live

Actomyosin network contractility is a crucial driver of force generation in eukaryotic cells. These networks are driven out of equilibrium in part by the molecular motor myosin which crosslinks and exerts forces on actin filaments. While myosin's role in force generation is well studied, the mechanism by which these forces drive cellular-scale deformation is still debated. As a step towards addressing this, we here consider the actomyosin-driven surface contraction wave of meiotic starfish oocytes. Using pharmacological inhibitions targeting actin polymerization, we find that cellular deformation during the contraction wave is not a monotonic function of cortical actin density but is instead peaked near the wild-type density. This surprising observation provides a window to test the mechanisms underlying actomyosin contractility. We’re using a combination of targeted biochemical perturbations, agent-based simulations, and theoretical modeling to understand this phenomenon.