(M1030-04-23) Using Molecular Probes to Dissect and Control the Energetics of Liquid-Liquid Phase Separation in a Model Intrinsically Disordered Protein System

Purpose: Biomolecular condensates are essential in all eukaryotic cells.1–3 Examples of condensates include nuclear stress bodies that regulate gene expression and P-bodies that are responsible for mRNA processing and decay.1–3 These condensates can establish stable boundaries despite the lack of membrane encapsulation. Certain condensates have been linked to disease, as some can convert into insoluble aggregates present in neurodegenerative diseases.4 The formation of many biomolecular condensates results from a reversible process known as liquid-liquid phase separation (LLPS). LLPS describes the demixing of a homogenous liquid into two distinct liquid phases: a protein-rich dense phase that appears as liquid droplets in coexistence with a protein-depleted dilute phase.1–4 Although LLPS is an essential cellular process, the specific energetic contributions of the interactions that establish such phase-separating protein systems have not been fully determined. Interactions contributing to LLPS include multivalent noncovalent interactions such as hydrophobic, charge-charge, cation-𝜋, and 𝜋-𝜋 interactions.5 The lack of understanding of the relative contributions of these driving forces and the inability to reliably separate the energetic contributions make it challenging to explain the specificity we see in complex condensate assemblies. A common feature of a protein system that undergoes LLPS is an intrinsically disordered region: a stretch of sequence that exists as an ensemble of rapidly fluctuating conformations that lack stable secondary or tertiary structure.1–4 Here, we study a model physiological LLPS system, the intrinsically disordered Laf-1 RGG protein (residues 1-170), with molecular probes to define the energetic contribution of specific intermolecular forces controlling the formation of LLPS in disordered protein systems.

Methods: The N-terminal RGG domain of Laf-1 (residues 1-170) is sufficient and necessary for LLPS under physiological conditions.6 In the Laf-1 RGG system, electrostatic interactions are important for phase separation, however, we also believe that 𝜋-interactions, more specifically 𝜋-𝜋 and CH-𝜋 interactions, are also important for the specific energetics of Laf-1 RGG LLPS. Laf-1 RGG protein was overexpressed and purified from E. coli. Methanol (15% MeOH) and acetonitrile (15% MeCN) were used in experiments to assess the importance of hydrophobic vs 𝜋-interactions in Laf-1 RGG phase separation using temperature-dependent UV-vis turbidity measurements. Additionally, as many proteins that undergo LLPS also bind nucleic acids, we aimed to deconvolute the energetic contributions of electrostatics (via the phosphate backbone) from interactions via the 𝜋-system of nucleobases in our LLPS system. Fluorescence imaging of labeled DNA oligonucleotides (25-mers) was compared to sequence-matched polymers where the polymer backbone phosphate groups are replaced with nonionic morpholino groups to examine the contributions of electrostatic vs 𝜋-interactions in the Laf-1 RGG LLPS system.

Results: Methanol co-solvent did not destabilize assembly via disruption of hydrophobic interactions. Surprisingly, it instead increased the stability of the phase-separated system with an observed turbidity onset 7 ºC higher in 15% MeOH relative to 0% co-solvent (Fig.1). Despite the presence of 11 tyrosine (Y) residues in Laf-1 RGG, it appears that hydrophobic interactions are not important for Laf-1 RGG LLPS. The stabilization may be due to increased charge/charge interactions in the reduced dielectric (pure MeOH dielectric constant = 32.7). However, we observed attenuation of LLPS in acetonitrile, despite also having a reduced dielectric constant (pure MeCN dielectric constant = 37.5) (Fig.1). Acetonitrile has a strongly polarized C-H bond that may compete for CH-𝜋 interactions. The 𝜋-character of the cyano group may interfere with critical 𝜋-interactions and the MeCN polarized C-H bonds potentially interfere with CH-𝜋 interactions. We observed a turbidity onset 5 ºC lower in 15% MeCN relative to 0% co-solvent (Fig.1). Additionally, we observed no partitioning of the 25-mer nonionic morpholino oligomer into dense phase droplets, suggesting that the strongly anionic phosphate backbone is necessary for nucleic acid partitioning in Laf-1 LLPS (Fig.2). This is in clear contrast to the DNA oligomer, which showed preferential partitioning into the dense phase (Fig.2). These results suggest that while aromatic 𝜋-interactions can tune interaction strength, they alone are insufficient for preferential partitioning of nucleic acid into Laf-1 LLPS dense phase.

Conclusion: We observed a destabilizing effect of MeCN co-solvent which suggests that 𝜋-interactions are potential drivers of LLPS. Furthermore, MeOH co-solvent increased the stability of Laf-1 LLPS, suggesting that hydrophobic interactions are not important. The phosphate backbone appears to be necessary for the partitioning of nucleic acids into the Laf-1 LLPS system, however, aromatic-𝜋 interactions alone are insufficient. Understanding the energetics of molecular probes and their specific partitioning will help in designing small molecules that are directed into or out of particular biomolecular condensates, elucidating the specificity of LLPS systems, and explaining how both natural and synthetic molecules can be targeted to them.

References: (1) Uversky, V. N. Intrinsically Disordered Proteins in Overcrowded Milieu: Membrane-Less Organelles, Phase Separation, and Intrinsic Disorder. Curr. Opin. Struct. Biol. 2017, 44, 18–30
(2) Alberti, S.; Gladfelter, A.; Mittag, T. Considerations and Challenges in Studying Liquid-Liquid Phase Separation and Biomolecular Condensates. Cell 2019, 176 (3), 419–434
(3) Schuster, B. S.; Reed, E. H.; Parthasarathy, R.; Jahnke, C. N.; Caldwell, R. M.; Bermudez, J. G.; Ramage, H.; Good, M. C.; Hammer, D. A. Controllable Protein Phase Separation and Modular Recruitment to Form Responsive Membraneless Organelles. Nat. Commun. 2018, 9 (1), 2985
(4) Shin, Y.; Brangwynne, C. P. Liquid Phase Condensation in Cell Physiology and Disease. Science 2017, 357 (6357), eaaf4382
(5) Das, S.; Lin, Y.-H.; Vernon, R. M.; Forman-Kay, J. D.; Chan, H. S. Comparative Roles of Charge, π, and Hydrophobic Interactions in Sequence-Dependent Phase Separation of Intrinsically Disordered Proteins. Proc. Natl. Acad. Sci. 2020, 117 (46), 28795–28805
(6) Elbaum-Garfinkle, S.; Kim, Y.; Szczepaniak, K.; Chen, C. C.-H.; Eckmann, C. R.; Myong, S.; Brangwynne, C. P. The Disordered P Granule Protein LAF-1 Drives Phase Separation into Droplets with Tunable Viscosity and Dynamics. Proc. Natl. Acad. Sci. 2015, 112 (23), 7189–7194

Acknowledgements:

Funding: Institute for Soft Matter Synthesis and Metrology

Figure 1: Average turbidity @ 600 nm of 25 µM Laf-1 RGG phase separating under physiological conditions (Buffer conditions: 125 mM NaCl, 20 mM Tris, pH 7.4). Turbidity onset at 18 ℃ was observed for RGG in 0% cosolvent, 25 ℃ for RGG in 15% MeOH co-solvent, and 13 ℃ for RGG in MeCN co-solvent.

Figure 2: Micrographs of 25 µM RGG protein in the presence of 500 nM DNA and 500 nM morpholino molecular probes (Buffer conditions = 125 mM NaCl, 20 mM Tris, pH 7.4) Temperature = 4 ºC. Scale bar = 50 µm.