Title: Multiscale Modeling to unravel cellular and subcellular process in biological systems: Implication on signaling and exocytosis

Speaker: Dr. Anirban Polley, Columbia University

Abstract: Complex biological process can be understood from the properties of specific molecules that are building blocks of living organism. How do living organism composed of molecules have self-organized to do ‘engineering tasks’ such as efficient processing of information and control? This potentially brings together many fields of research including non-equilibrium statistical physics, classical mechanics and control theory to study biology. To do this, we identify specific molecules involved in biological processes, and determine their molecular structures, organization, and properties using multiscale modeling. We address important questions regarding signaling and exocytosis below.

Early works both theoretically (Gourishankar, K. et al., Cell 2012) and experimentally (Goswami, D. et al, Cell 2008) show that the outer leaflet GPI-anchored proteins (GPI-APs) of the cell surface are organized as monomers and cholesterol sensitive nanocluters, which are regulated by the active remodeling of the underlying cortical actin and myosin. Since, GPI-APs are lipid tethered proteins which reside on the outer leaflet of the plasma membrane, the natural question is how the outer-leaflet GPI-APs couple to the cortical actin that abuts the inner leaflet of the cell membrane. As a variety of high resolution experiments on live cells using FRET found no direct linkage between the GPI-APs and cortical actin (CA), there must be an indirect coupling between the outer leaflet GPI-APs and CA or its immediate interacting partners. We address these important issues using atomistic molecular dynamics (MD) simulations on multicomponent model membrane. We find that long saturated acyl-chains are required for forming GPI-anchor nanoclusters. Simultaneously, at the inner leaflet, long acyl-chain containing phosphatidylserine (PS) is necessary for transbilayer coupling.

Exocytosis and trafficking depend on membrane fusion reactions mediated by a cellular fusion machinery whose core comprises the SNARE proteins. In secretion, from insulin release to release of neurotransmitters (NTs) at neuronal synapses, fusion is catalyzed when vesicle associated v-SNAREs and target membrane-associated t-SNAREs zipper into SNAREpin complexes, pulling membranes into proximity and triggering fusion. The mechanism is poorly understood and the number of SNAREpins required for fusion is controversial. Here we developed a highly coarse-grained MD simulation to expose the cooperative behavior of SNAREs at the fusion site and to compute the waiting times for fusion between a docked vesicle and target membrane. Our results present a picture radically different to the traditional one according to which the SNARE zippering energy is somehow funneled into the membranes to fuse them. We find the ~65 kT of zippering energy is entirely dissipated, and entropic forces impose multiple collisions between the membranes. The more SNAREs, the greater the entropic forces, the higher the collision frequency and the faster fusion.