Studying RNA Modulated Protein Liquid-Liquid Phase Separation using Coarse-Grained Models

Abstract

Disordered proteins phase separate in biological cells creating distinct membraneless organelles coexisting with the surrounding cytosol. However, living cells contain a multitude of proteins, nucleic acids, and lipids with remarkable spatiotemporal organization controlling their interactions for specific functions. Experiments have found that membraneless organelles such as P granules are mixtures of RNA and proteins that form via phase separation. Hence, it is essential to study biomolecular phase separation, taking into account the presence of RNA for a deeper understanding of the in vivo formation and function of membraneless organelles. We have successfully studied protein phase separation via molecular dynamics, using a coarse-grained (CG) modeling approach where we effectively capture protein-protein interactions at the amino acid level. In this work, we extend our CG approach to incorporate RNA and determine how it modulates protein phase separation. We directly simulate the co-phase separation of both the disordered domain of DEAD-box helicase protein LAF-1 and polyA RNA co-localized in a slab geometry using standard molecular dynamics simulations conducted over a range of temperatures. We find that favorable RNA-protein interactions lead to the formation of condensed phases containing both RNA and protein, which is consistent with the experiment. The presence of RNA lowers the concentration of protein in the dense phase, and RNA-protein interactions also change the dynamics of proteins inside the condensed phase.