Simulation source code for "Myosin and α-actinin regulation of stress fiber contractility under tensile stress"

MEDYAN v5.4.0 was used to generate simulation trajectories of bundled actin networks.Stress fibers are actomyosin bundles that regulate cellular mechanosensation and force transduction. Connecting to extracellular matrix through focal adhesion complexes, stress fibers actively generate contractile forces with myosin motors and crosslinking proteins. Under external mechanical stimuli such as tensile forces, the stress fiber remodels its architectures to adapt to the external cues, displaying properties of viscoelastic materials. How the structural remodeling of stress fibers is related to the generation of contractile force is not well understood. In this work, we simulate mechanochemical dynamics and force generation of stress fibers using the molecular simulation platform MEDYAN. We model stress fiber as two connecting bipolar bundles attached at the ends to focal adhesion complexes. The simulated stress fibers generate contractile force that is regulated by myosin motors and α-actinin crosslinkers. We find that stress fibers are able to enhance contractility by reducing the distance between actin filaments to increase crosslinker binding, while this structural remodeling ability depends on the crosslinker turnover rate. Under tensile pulling, the stress fiber shows an instantaneous increase of the contractile forces followed by a slow relaxation into a new steady state. While the new steady state contractility after pulling only depends on the overlap between actin bundles, the short term contractility enhancement is sensitive to the tensile pulling distance. We further show that this mechanical response is sensitive to the crosslinker turnover rate. Our results provide insights into the stress fiber mechanics that have significant implications for understanding the cellular adaptation to mechanical signaling.This work was supported by Public Health Service grants R01CA201340 from NIH/NCI and 1R01EY028450 from NIH/NEI, and the endowment for Frady Whipple Chair in Mathematics at Georgia State University (to YJ), and the grant CHE-2102684 from the National Science Foundation (to GAP). The authors acknowledge the University of Maryland supercomputing resources (http://hpcc.umd.edu) made available for conducting the research reported in this paper