The Stickers and Spacers Framework for Describing Phase Behavior of Multivalent Intrinsically Disordered Proteins

Accurate spatial and temporal organization of cellular components is essential for viability and functionality of a cell. Eukaryotic cells have invented membrane-bound organelles, but cells also employ membraneless organelles for rapid and efficient operations. In the last decade, it has been shown that liquid-liquid phase separation (LLPS) of biopolymers is the underlying principle behind formation, regulation, and dissociation of membraneless organelles in vivo. The LLPS drivers need to bear multivalency and flexibility, and intrinsically disordered proteins (IDPs) occasionally take the role. Although the amount of experimental data has been rapidly growing, however, our theoretical understanding is far from mature. We recently conceptualized the stickers and spacers framework, where IDP residues are designated into two groups: stickers, which drive chain-chain interactions, and spacers, which modulate the chain properties. Based on the framework, we developed an analytical mean-field model and applied it to the saturation concentration data of FUS family proteins to validate the utility of the model. The model predicts additive contributions from different types of sticker pairs, which explains why a dominant sticker pair type in one system can be insignificant in another system. We also demonstrate the results from graph-based simulations, which complement the analytical model. The simulation engine incorporates the concept of a cluster, which is a set of stickers that belong to the same polymer or to the polymers linked by sticker bonds. Intra-cluster interactions have different entropic contributions from inter-cluster ones, which the mean-field approach only considers. Moreover, three- and four-body interactions are implemented to test the effect of higher-order interactions. The mean-field predictions have been successfully reproduced, and it was shown that intra-cluster interactions are more crucial than multi-body interactions to explain the deviations of FUS variants from the mean-field prediction.