Effect of membrane lipid phosphorylation on protein recruitment at molecular and cellular scales.
The project focuses on the quantitative modeling of protein assemblies at the core of the synaptic exocytic mechanism. In order to reach time and length scale that are biologically relevant, we will use coarse-grained models designed by machine learning, and incorporating experimental data.
Exocytosis is a ubiquitous mechanism for transporting molecules across cell membranes and is crucial for signaling in presynaptic neurons. While the release and recognition of Ca2+ ions can be considered as the molecular switch triggering the exocytosis machinery, the transient states and transition mechanisms are still far from understood in detail. While most of the relevant players, e.g., Synaptotagmin, SNARE proteins, Com- plexin, have been identified, these proteins are partially disordered, individually weak binders, and go through large-scale conformational changes during exocytosis. These properties make the machinery difficult to char- acterize with structural biology or atomistic molecular dynamics (MD) simulation.
In the past few decades the study of collective/organizing motions at long timescales and system sizes inac- cessible to atomistic simulations has been mostly based on coarse-grained models that “renormalize” groups of atoms into “effective” degrees of freedom. By sacrificing the atomistic details coarse-grained models can explore larger time and length scales, providing a good starting point to identify relevant regions on the vast configurational landscape of protein complexes. However, these simplified models are usually built with ap- proximations and assumptions that may strongly affect the accuracy of the results. We have recently shown that Machine Learning methods can be used to design efficient coarse-grained models that are able to accu- rately reproduce the thermodynamics of a macromolecule . This approach can be combined with experi- mental data in order to compensate for errors in the all-atom MD model [2,3]. We propose to use this approach to study transient states and mechanisms of the exocytosis molecular switch consisting of SNARE, Complexin and Synaptotagmin.