Shinji Matsushita

Effect of tensile force on the mechanical behavior of actin filaments

Actin filaments are the most abundant components of the cellular cytoskeleton, and play critical roles in various cellular functions such as migration, division and shape control. In these activities, mechanical tension causes structural changes in the double-helical structure of the actin filament, which is a key modulator of cytoskeletal reorganization. This study performed large-scale molecular dynamics (MD) and steered MD simulations to quantitatively analyze the effects of tensile force on the mechanical behavior of actin filaments. The results revealed that when a tensile force of 200pN was applied to a filament consisting of 14 actin subunits, the twist angle of the filament decreased by approximately 20°, corresponding to a rotation of approximately -2° per subunit, representing a critical structural change in actin filaments. Based on these structural changes, the variance in filament length and twist angle was found to decrease, leading to increases in extensional and torsional stiffness. Torsional stiffness increased significantly under the tensile condition, and the ratio of filament stiffness under tensile force to that under no external force increased significantly on longer temporal scales. The results obtained from this study contribute to the understanding of mechano-chemical interactions concerning actin dynamics, showing that increased tensile force in the filament prevents actin regulatory proteins from binding to the filament.

Evaluation of extensional and torsional stiffness of single actin filaments by molecular dynamics analysis

It is essential to investigate the mechanical behaviour of cytoskeletal actin filaments in order to understand their critical role as mechanical components in various cellular functional activities. These actin filaments consisting of monomeric molecules function in the thermal fluctuations. Hence, it is important to understand their mechanical behaviour on the microscopic scale by comparing the stiffness based on thermal fluctuations with the one experimentally measured on the macroscopic scale. In this study, we perform a large-scale molecular dynamics (MD) simulation for a half-turn structure of an actin filament. We analyse its longitudinal and twisting Brownian motions in equilibrium and evaluated its apparent extensional and torsional stiffness on the nanosecond scale. Upon increasing the sampling-window durations for analysis, the apparent stiffness gradually decreases and exhibits a trend to converge to a value that is close to the experimental value. This suggests that by extrapolating the data obtained in the MD analysis, we can estimate the experimentally determined stiffness on the microsecond to millisecond scales. For shorter temporal scales, the apparent stiffness is larger than experimental values, indicating that fast, local motions of the molecular structure are dominant. To quantify the local structural changes within the filament on the nanosecond scale and investigate the molecular mechanisms, such as the binding of the actin-regulatory proteins to the filaments, it is preferable to analyse the mechanical behaviour on the nanometre and nanosecond scales using MD simulation.

Multiscale modeling and mechanics of filamentous actin cytoskeleton

The adaptive structure and functional changes of the actin cytoskeleton are induced by its mechanical behavior at various temporal and spatial scales. In particular, the mechanical behaviors at different scales play important roles in the mechanical functions of various cells, and these multiscale phenomena require clarification. To establish a milestone toward achieving multiscale modeling and simulation, this paper reviews mathematical analyses and simulation methods applied to the mechanics of the filamentous actin cytoskeleton. The actin cytoskeleton demonstrates characteristic behaviors at every temporal and spatial scale, and mathematical models and simulation methods can be applied to each level of actin cytoskeletal structure ranging from the molecular to the network level. This paper considers studies on mathematical models and simulation methods based on the molecular dynamics, coarse-graining, and continuum dynamics approaches. Every temporal and spatial scale of actin cytoskeletal structure is considered, and it is expected that discrete and continuum dynamics ranging from functional expression at the molecular level to macroscopic functional expression at the whole cell level will be developed and applied to multiscale modeling and simulation.

Quantitative analysis of extension–torsion coupling of actin filaments

Actin filaments have a double-helix structure consisting of globular actin molecules. In many mechanical cellular activities, such as cell movement, division, and shape control, modulation of the extensional and torsional dynamics of the filament has been linked to regulatory actin-binding protein functions. Therefore, it is important to quantitatively evaluate extension-torsion coupling of filament to better understand the actin filament dynamics. In the present study, the extension-torsion coupling was investigated using molecular dynamics simulations. We constructed a model for the actin filament consisting of 14 actin subunits in an ionic solvent as a minimal functional unit, and analyzed longitudinal and twisting Brownian motions of the filament. We then derived the expected value of energy associated with extension and torsion at equilibrium, and evaluated the extension-torsion stiffness of the filament from the thermal fluctuations obtained from the MD simulations. The results demonstrated that as the analyzed sampling-window duration was increased, the extension-torsion coupling stiffness evaluated on a nanosecond scale tended to converge to a value of 7.6×10(-11) N. The results obtained from this study will contribute to the understanding of biomechanical events, under mechanical tension and torque, involving extension-torsion coupling of filaments.

Atomistic mechano-chemical modeling of kinesins

This work is concerned with the dynamics of motor proteins. In particular, we discuss the development of computational analysis tools for predicting the dynamics of molecular motors such as certain types of kinesin. The ability to model and predict how these biomolecular machines work forms the critical link to biotechnological device development, including lab-on-a-chip applications and many others. The focus of this research is on the identification and modeling of nonlinear dynamic phenomena caused by coupled thermal, chemical, and mechanical fields. A mechanistic model of kinesin has been developed recently at the University of Michigan. This model accounts for transient dynamics and uses parameters which have to be identified from experimental data and/or from first principles. In this work, accurate atomistic simulations using a monomeric human kinesin structure (PDB ID: 1MKJ, 2.70 Angstroms resolution) is used instead of experimental data to obtain key nano-scale properties of the motor protein. The approach allows an accurate bridging between nano-scale processes occurring over pico seconds and micron- or millimeter-scale processes occurring over seconds.