Session K35: Bio-Inspired Phase Separation

Rational design of multicomponent condensates

Biology provides numerous examples of phase-separated protein and nucleic acid condensates, which establish distinct compartments for spatially organizing biomolecules within living cells. This mechanism of spatial organization relies on the ability of biomolecular systems to produce complex phase diagrams by tuning the interactions among molecules in a multicomponent mixture. To reproduce this behavior in synthetic systems, it is essential that we identify design rules that map the sequences of individual biomolecules to the emergent phase behavior of multicomponent systems. In this talk, I will describe recent theoretical and computational advances towards the goal of designing fully programmable, synthetic multiphase condensates. These results take the form of scaling relations that bound the complexity of phase diagrams that can be achieved in various biomolecular systems, as well as optimization algorithms for designing biomolecular mixtures that can spontaneously assemble into prescribed multiphase condensates. I will also discuss how nucleation pathways for biomolecular condensate assembly can be rationally designed, providing a mechanism for achieving precise spatiotemporal control of multicomponent, multiphase systems. Taken together, these design rules provide a deeper understanding of the limits of phase-separation-mediated spatial organization in biological systems and establish practical strategies for engineering fully programmable multiphase condensates.   

Coacervate vesicles and double emulsions

Coacervates are liquid-like droplets that form by phase separation of macromolecules such as charged polymers. They serve as a model for understanding intracellular phase separation. They have also been proposed as models for protocells, the precursors of living cells, because of their membraneless behavior that allows for easy transport and concentration of biomolecules such as nucleic acids based on natural partitioning. This study investigates the formation of vesicles and double emulsions of polyelectrolyte coacervates. We show that the type of polyelectrolyte, that is, their charge density and their size, can control and influence the morphology of the droplet vs. vesicles, and the morphology further emerges with time due to the coarsening of the vesicles. We found that these coacervate vesicles inflate over time, mimicking inflating balloons. We found that this growth is driven by the osmosis of water from the surroundings into the lumen of vesicles, and the rate of this osmosis can be controlled using osmolytes. Overall, our work shows how changes in the molecular properties of polyelectrolytes can give rise to different morphologies of condensates that might have been present and probably influential in the prebiotic world.   

Phase Separation of a Heterogeneous DNA Nanostar System

DNA nanostars (NSs) have, in the last decade, been established as a model system for the study of biomolecular liquid-liquid phase separation. A major advantage of the NS system is that interactions between NS particles can easily be tuned by exploiting DNA sequence specificity. Here, we consider mixtures of multiple, distinct NS particle species and investigate the effect on the phase behavior of the system.   

Dynamic swarms regulate the growth and morphology of membrane domains

Our experiments investigate the dynamics of cell membranes attached to the underlying cytoskeleton and molecular motors. We developed a synthetic model of a multiphase lipid bilayer coupled to purified actin and myosin proteins. In the presence of gliding actin filaments, membrane domains evolve into unique structures, a distinct deviation from circular shapes observed in coarsening. These structures frequently merge and break up, leading to arrested domain growth. In this work, we combine the Toner-Tu equation for flocking dynamics with a phase field model to study such systems of activity-mediated phase separation. Our model predicts the growth and structure of domains in the presence of swarming flows. We find that the domain structure reaches a dynamic steady state. We extend our analysis to different swarms and show that this steady-state morphology is correlated with the peak of the velocity spectra. We further analyze the domain structure by deriving an evolution equation of interfacial fluctuations and show that the nature of active stresses controls the growth of initial perturbations at the interface.   

Directed Movement of Condensates by Reaction-driven Assembly

The cellular environment, characterized by its intricate composition and spatial organization, hosts a variety of organelles, ranging from membrane-bound ones to membraneless structures that are formed through liquid-liquid phase separation. Cells show precise control over the position of such condensates. We demonstrate that organelle movement within cellular domains can naturally arise from biochemical reactions, which are driven by a chemical fuel and produce waste. Simulations and analytical arguments within a minimal model of phase separation and reaction cycles reveal that the directed movement stems from two contributions: fuel and waste are refilled or extracted locally, resulting in concentration gradients, which (i) induce product fluxes via incompressibility and (ii) results in an asymmetric production in the organelle's surroundings and thereby shifts its position. We show that the former contribution dominates and sets the direction of the movement, away from or towards fuel source and waste sink, depending on the product molecules' affinity towards fuel and waste, respectively. The mechanism thus provides a simple means to organize condensates with different compositions. Particle-based simulations as a control and systems with more complex reaction cycles underline the universality of this mechanism.   

Modeling the Effects of HU Proteomic Interactions in the Genome Configuration of a Minimal Cell

The genetic material inside cells is encoded in DNA molecules, which contain the necessary information to produce proteins and other key molecules. Currently, new models have been developed from a physico-chemical perspective where the dynamics of cell constituents is governed by physical and thermodynamic forces. To this end, small-genome cells, such as the synthetic JCVI-Syn3A minimal cell, which only contains 452 protein-coding genes, provides a good framework to investigate the physics underlying key cellular processes. To properly characterize the kinetics of such processes, it is important to study the physical organization of the genome. In this work, we investigate the effect of the HU proteomic interactions on chromosomal configuration. To model the circular bacterial chromosome, we develop a mesoscale bead-spring model where medium-sized gene blocks are represented by coarse-grained beads. The physical-chemical properties of specific beads depend on the genome sequence of JCVI-Syn3A. The presence of HU proteins affects the chromosomal distribution, including coiling and the compactification of the genetic material, leading to nucleoid formation, which can potentially affect transcription kinetics and gene expression.   

Spatial Organization of DNA Liquids

Cells operate by compartmentalizing chemical reactions. Much recent work has shown that the spatiotemporal formation and control of membraneless compartments inside cells (liquid-liquid phase separation) is integral to cell function. Here, we investigate the dynamics and long-range structures formed by a model phase-separating DNA system. We use DNA nanostars, a system of finite-valence particles, roughly 10nm in size, whose sequence is designed such that they self-assemble into liquid droplets on the micron scale via a binodal phase transition. We find that the structure is hyperuniform, corresponding to a disordered structure with anomalously small long-range density fluctuations, which is characteristic of a spinodal decomposition process that represents a perturbation that then relaxes to equilibrium via droplet Brownian motion. In addition, we quantify the concentration and temperature dependence of the initial droplet appearance time and find that phase separation dynamics are consistent with a classical nucleation picture where droplet growth is dominated by Brownian motion and coalescence. Finally, we investigate how droplet hyperuniformity might be exploited in chemical reaction schemes, analogous to those present in biomolecular condensates, by coupling phase separation to an in vitro transcription reaction. We hope that our work on near-equilibrium droplet assembly and structure provides a foundation to investigate droplet organizational mechanisms in driven/biological environments, or to implement droplet patterns as efficient biochemical reactors.   

A Background Enzymatic Active Bath Affects Liquid-Liquid Phase Separation (LLPS) of Proteins

The cell interior is an active bath driven by a myriad of enzymes. It is an open problem as to how this background activity can affect physical processes in the cell, including liquid-liquid phase separation. We seek to experimentally reconstitute a model system for an active bath of enzymes to determine the effects on the liquid phase separation of a model condensate protein. We will use urease, an exothermic and kinetically fast enzyme that converts urea to carbon dioxide and ammonia, as the background enzyme. The protein that phase separates at high enough salt concentration is ubiquilin-2(UBQLN2). With the newly developed microfluidic chamber that flushes out carbon dioxide and ammonia while maintaining the concentration of urea, we showed that an active bath can prompt protein condensation. In addition, to provide some theoretical interpretation for our experimental results, we implement polymer-based molecular dynamics simulations via a sticker-spacer polymer model that can recapitulate protein condensation in the presence of active particles.    

Are phases an appropriate description for cells? – Fluctuation dominated regime in finite systems with many components

Phase separation has emerged as an important topic for cellular function. From lipid rafts to liquid-liquid phase separation, our current understanding is that it is crucial for organization. We putatively expect that rules extracted from simple systems, two component mixtures, extend to multicomponent systems. While this is true in the thermodynamic limit, I will discuss here the thermodynamic limit for multicomponent systems. Using a toy model, I will show that what we consider "large systems" is largely subjective and dependent on details in multicomponent systems. For "small" systems, rules are very different, and the system is dominated by fluctuations. Usual assumptions, such as equivalence of thermodynamic ensembles, are broken. Still, the system can be driven to exhibit behavior that is similar to a phase transition, for instance by changing the statistical ensemble. Practically, this means that observed phase behavior may be largely dependent on system preparation. The typical signature of this regime is eerily similar to many observations in cells. This naturally leads to a fundamental question: is the traditional phase behavior an appropriate description for cellular behavior?   

Impact of the substrate on the phase transitions properties in supported lipid bilayers

Bacterial membranes are self-assembled structures, comprising a phospholipid bilayer embedding proteins and sugars. It plays a crucial role in the survival of the organism, from providing a physical barrier with the surroundings, to controlling membrane trafficking,and underpinning drugs resistance. The activity of the proteins and their response to external stimuli depends on the biophysical properties of the bilayer such as its thickness, elasticity, packing and molecular mobility. Often overlooked is the fact that membranes are always in close contact with various structural filaments such as cytoskeleton tubules or peptidoglycan chains which can all influence the properties of the bilayer.

Here, we employed a minimal bi-component model system of E. coli’s inner membrane to study the effects of these external interactions on the nanoscale phase behaviour of the model membrane. Using a combination of atomic force microscopy and differential scanning calorimetry, we comparatively track the phase transition kinetics, comparing situation when the membrane is supported and unsupported. The results show that only part of the kinetics depends on the cooling/heating, with the main influence coming from contact with a support. The presence of a contacting substrates not only shifts the transition temperature, but it can also arrest the transition and induce a global re-arrangement of the lipid species. This results in a phase transition that partially follows classical nucleation on short timescales but moves on to a spinodal decomposition on the longer term.

Our work highlights the importance of external membrane contacts to control the composition and biophysical properties of bacterial membranes.   

Spatio-temporal control of optically responsive compressible isotropic active fluids

In a microtubule-kinesin active isotropic fluid, microtubules form bundles through depletion, leading to their extension and the emergence of chaotic flows. Previous studies have developed optically responsive kinesin molecular motors that can be used for spatio-temporal patterning of active flows. In this work, we explore how compressible isotropic active fluids respond to different light patterns. Furthermore, based on the recent advances describing active liquid-liquid phase separation we create active droplets. By a similar approach, we can apply light patterns to these droplets and study their motion. Our goal is to control the shape and motility of these active droplets.