How cells feel: stochastic model for a molecular mechanosensor.

Understanding mechanosensitivity (i.e., how cells sense the stiffness of their environment) is very important, yet there is a fundamental difficulty in understanding its mechanism: to measure an elastic modulus one requires two points of application of force-a measuring and a reference point. The cell in contact with substrate has only one (adhesion) point to work with, and thus a new method of measurement needs to be invented. The aim of this theoretical work is to develop a self-consistent physical model for mechanosensitivity, a process by which a cell detects the mechanical stiffness of its environment (e.g., a substrate it is attached to via adhesion points) and generates an appropriate chemical signaling to remodel itself in response to this environment. The model uses the molecular mechanosensing complex of latent TGF-β attached to the adhesion point as the biomarker. We show that the underlying Brownian motion in the substrate is the reference element in the measuring process. The model produces a closed expression for the rate of release of active TGF-β, which depends on the substrate stiffness and the pulling force coming from the cell in a subtle and nontrivial way. It is consistent with basic experimental data showing an increase in signal for stiffer substrates and higher pulling forces. In addition, we find that for each cell there is a range of stiffness where a homeostatic configuration of the cell can be achieved, outside of which the cell either relaxes its cytoskeletal forces and detaches from the very weak substrate, or generates an increasingly strong pulling force through stress fibers with a positive feedback loop on very stiff substrates. In this way, the theory offers the underlying mechanism for the myofibroblast conversion in wound healing and smooth muscle cell dysfunction in cardiac disease.