Macromolecular Regulators Have Matching Effects on the Phase Equilibrium and Interfacial Tension of Biomolecular Condensates

Abstract

The interfacial tension of phase-separated biomolecular condensates affects their fusion and multiphase organization, and yet how this important property depends on the composition and interactions of the constituent macromolecules is poorly understood. Here we use molecular dynamics simulations to determine the interfacial tension and phase equilibrium of model condensate-forming systems. The model systems consist of binary mixtures of Lennard-Jones particles or chains of such particles. We refer to the two components as drivers and regulators; the former has stronger self-interactions and hence a higher critical temperature (T_c) for phase separation. In previous work, we have shown that, depending on the relative strengths of driver-regulator interactions and driver-driver interactions, regulators can either promote or suppress phase separation (i.e., increase or decrease T_c). Here we find that the effects of regulators on T_c quantitatively match the effects on interfacial tension (γ). This important finding means that, when a condensate-forming system experiences a change in macromolecular composition or a change in intermolecular interactions (e.g., by mutation or posttranslational modification, or by variation in solvent conditions such as temperature, pH, or salt), the resulting change in T_c can be used to predict the change in γ and vice versa. We also report initial results showing that disparity in intermolecular interactions drives multiphase coexistence. These findings provide much needed guidance for understanding how biomolecular condensates mediate cellular functions.

Multiphase Organization Is a Second Phase Transition Within Multi-Component Biomolecular Condensates

“…1,[3][4][5][6][7][8][9][10][11] Multiphase coexistence has also been observed in coarse-grained molecular simulations. 8,[11][12][13][14][15][16] Yet the theoretical underpinning is still unclear. Here we present a simple theoretical model to show that the multiphase organization of multi-component condensates is a second phase transition.…”

“…The strengths, ε DD and ε RR , of self-attraction of the two components, are different. Specifically, ε DD > ε RR , so that the critical temperature for phase separation of pure D is higher than the counterpart of pure R. The phase separation of a mixture is driven by the self-attraction of D (for “driver”) but regulated by the self-attraction of R (for “regulator”) and the cross-species attraction between D and R. 14, 18-19 We work in temperatures ( T ) below the critical temperature, T c , for the phase separation of the mixture, where the first phase transition has resulted in the coexistence of a dense phase and a bulk phase. Our interest is the dense phase, specifically its multiphase organization.…”

“…All these predictions are consistent with our initial results of demixing from molecular dynamics simulations of binary mixtures of Lennard-Jones particles or chains of such particles ( L =10). 14 For these Lennard-Jones systems, we used ε αβ to denote the well-depth for the interaction potential energy between particles of species α and β (= D or R), and fixed ε DD and ε RR at 1 and 0.9, respectively. Complete intermixing between D and R was observed at ε DR = 1.2 and 1.0, but demixing was observed at ε DR = 0.8 for chain mixtures at lower temperatures.…”

“…Demixing slows down the equilibration process and also can produce multiple metastable states. We thus extended the simulations here from 10 million to 100 million steps (ref 14 ; Fig. S1) and carried out five replicate runs when there were signs of demixing.…”

“…3). 14 The entire T vs. x R plane can be divided into three distinct zones. Above the T c curve, there is no phase separation; between the T c curve and the T vs. x R binodal, the dense phase is a homogenous mixture of the components; below the T vs. x R binodal, the dense phase is in a demixed state.…”