Harnessing simulations to uncover molecular mechanisms of mechanosensing

New York University, New York NY

Investigators

Linked publications & trials

Abstract

Project Summary Mechanical forces play a fundamental role in the behavior of many proteins within cells. Specific proteins have evolved to sense and alter their behavior in response to a stretching or compressive stress. This can happen in environments that are naturally under tension, such as cellular membranes, or in environments that are constantly rearranging due to active forces generated by molecular motors. In such environments, there is a close coupling between macroscopic forces at the level of entire cellular systems, and how those forces are sensed and generated by individual macromolecules. Predicting how these systems interoperate using computational techniques remains a challenge due to the need to span a wide range of length and time scales. Moreover, at the molecular level, the mechanical forces involved are quite small, and hence it is difficult to predict how proteins undergo significant changes to their conformational ensemble in response to these tiny perturbations. In our group, we have pioneered approaches to overcome this challenge, resulting in new computational methods that allow us to robustly predict both changes in conformational ensembles under tension as well as the force-dependence of unbinding rates using atomistic molecular dynamics simulations. We have also worked to make all our methodological advances available as open-source software tools. Currently, we are combining these new approaches to tackle mechanosensing problems of unprecedented complexity. We also work closely with experimental collaborators, including through joint mentees, and that work inspires our future to study load-bearing structures in both mammalian and bacterial systems. In this proposal, we describe planned efforts to extend our approaches to large biomolecular complexes and proteins in complex cellular environments such as bacterial membranes, which will necessitate further advances in simulation and machine learning approaches to treat these assemblies. Our studies will result in new and improved tools and techniques to share with the community, as well as insight into the functioning of crucial mechanosensitive motifs in living systems. Ultimately, our aim is to connect molecular responses to force to with emergent large scale mechanical properties of living systems.