Tuning Formation of Protein–DNA Coacervates by Sequence and Environment

Abstract

The high concentration of nucleic acids and positively charged proteins in the cell nucleus provides many possibilities for complex coacervation. We consider a prototypical mixture of nucleic acids together with the polycationic C-terminus of histone H1 (CH1). Using a minimal coarse-grained model that captures the shape, flexibility, and charge distributions of the molecules, we find that coacervates are readily formed at physiological ionic strengths, in agreement with experiment, with a progressive increase in local ordering at low ionic strength. Variation of the positions of charged residues in the protein tunes phase separation: for well-mixed or only moderately blocky distributions of charge, there is a modest increase of local ordering with increasing blockiness that is also associated with an increased propensity to phase separate. This ordering is also associated with a slowdown of rotational and translational diffusion in the dense phase. However, for more extreme blockiness (and consequently higher local charge density), we see a qualitative change in the condensed phase to become a segregated structure with a dramatically increased ordering of the DNA. Naturally occurring proteins with these sequence properties, such as protamines in sperm cells, are found to be associated with very dense packing of nucleic acids.

The Effect of Polymer Length in Phase Separation

“…No condensates were observed for polyA5 for peptides as long as [RGRGG]15, also suggesting a minimal RNA length, but the peptide length was not extended beyond 75 amino acids. These simulations are generally in agreement with previous computational works on RNA [21][22][23] and DNA 24 showing LLPS is enhanced with longer nucleic acids lengths.…”

Theoretical and Data-Driven Approaches for Biomolecular Condensates

“… This approach made it possible to determine critical temperatures and dense and dilute phase protein concentrations, which showed agreement with theoretical descriptions of polymer phase transitions from Flory–Huggins theory. − Since the development of this coarse-graining framework, additional experimental data and atomistic simulations have been used to develop alternative residue-resolution coarse-grained models that aim to achieve quantitative agreement with experimental phase diagrams. ,,− The improved coarse-grained models can now quantitatively determine saturation concentrations ,, and account for the effects of a range of temperatures and salt concentrations . These models have been used in a variety of investigations of phase separated systems, including to identify the interactions driving and sustaining phase separation of the FUS low-complexity domain, to understand salt-dependence in interactions driving phase separation, and to look at the effect of condensate aging on material properties and dense phase interactions. , Coarse-grained models of nucleic acids have also made it possible to use simulations to look at protein–DNA interactions, including interactions between highly charged proteins and DNA in condensates, the interplay of H1, HP1 and DNA, to investigate the interactions driving phase separation of chromatin-associated proteins and its role in chromatin organization, and as part of a multiscale model of chromatin (Figure ). …”

“…Panel A shows a coarse-grained model of the C-terminus of histone H1 (in red) condensing with DNA (in blue). Figure reproduced with permission from ref . Copyright 2022 American Chemical Society.…”

“…120 These models have been used in a variety of investigations of phase separated systems, including to identify the interactions driving and sustaining phase separation of the FUS low-complexity domain, 124 to understand salt-dependence in interactions driving phase separation, 99 and to look at the effect of condensate aging on material properties and dense phase interactions. 102,103 Coarse-grained models of nucleic acids have also made it possible to use simulations to look at protein− DNA interactions, including interactions between highly charged proteins and DNA in condensates, 125 the interplay of H1, HP1 and DNA, 126 to investigate the interactions driving phase separation of chromatin-associated proteins and its role in chromatin organization, and as part of a multiscale model of chromatin (Figure 5). 127 Coarse-grained models also can be made more computationally efficient by confining molecular movements to a lattice as implemented in the LASSI model.…”

Molecular interactions underlying the phase separation of HP1α: Role of phosphorylation, ligand and nucleic acid binding

“…27,32 There are ongoing efforts to develop and evaluate a nucleic acid (DNA/RNA) CG model to study protein-nucleic acid interactions and the role DNA plays in LLPS. [57][58][59][60] Here, we use a model that separates the nucleotide into two beads; one bead represents the sugar-phosphate backbone, carrying an overall -1 charge, and the other represents the base but differentiates between bases ADE, THY, CYT, and GUA in their respective stacking and hydrogen bonding interactions. Using our CG nucleic acid model, we studied the effects of DNA addition on the LLPS of HP1α by conducting CG coexistence simulations of HP1α homodimers containing a small mole fraction of 205 bp dsDNA, 0.02 and 0.038, at 320 K. This system can be reasonably compared to an experimental system containing 50µM/100 µM of HP1α with 1.5µM of the same 205 bp dsDNA.…”