Enzyme-driven phase separation

TL;DR

The paper addresses how polarized membrane domains emerge and persist through energy-consuming enzymatic cycles rather than passive equilibrium phase separation. It derives a minimal reaction–diffusion model with two-state scaffold molecules φ^± and catalytic enzymes E^± exchanging with a reservoir, yielding a reduced order parameter φ governed by a nonconserved dynamics ∂tφ = D∇^2φ + A + √B ξ under a global constraint; this places the system in the class of active Model A with multiplicative noise. Mean-field analysis provides an explicit phase diagram in a reduced parameter space, identifying conditions for phase coexistence and giving a closed-form steady-state order parameter ⟨φ⟩∞ and interfacial properties; fluctuations are treated with large-deviation theory to obtain a critical nucleus radius Rc and to quantify interface tension σ and width w, with numerical simulations validating the analytic results. The work shows that non-equilibrium parameters, such as catalytic rates and enzyme asymmetry, robustly control phase separation on membranes, predict energy dissipation concentrated at interfaces, and align with experimental observations on kinase–phosphatase–driven lipid domains and Rab5 membrane patterning, offering a unified framework for active membrane organization and extensions to multispecies modules.

Abstract

The formation of polarized signaling domains on cell membranes is a fundamental example of biological pattern formation. While such patterns resemble structures from equilibrium phase separation, they are intrinsically non-equilibrium, driven by energy-consuming enzymatic cycles that switch molecules like phosphoinositides or small GTPases between distinct states. Here, we develop a minimal model of this enzyme-driven phase ordering process. Starting from microscopic reaction kinetics, we derive a mesoscopic theory that belongs to the class of active Model A with a global constraint. This framework yields an explicit mean-field phase diagram and closed-form expressions for key observables, such as interfacial tension, domain fractions, and phase coexistence boundaries, in terms of kinetic rates. In this context, phase coexistence is controlled by non-equilibrium parameters like catalytic rates and enzymatic asymmetry, rather than equilibrium parameters such as saturation concentrations. The resulting phase-separated domains rapidly exchange material with their surroundings. Their maintenance requires a continuous power input determined by enzymatic kinetics. The predicted phenomenology is consistent with experimental observations on reconstituted systems of phosphoinositide and Rab5 membrane patterning. We further study how metastable uniform states decay via nucleation of minority-phase domains and subsequent coarsening, driven by an effective interfacial tension. Using large deviation theory, we derive the critical nucleation radius under the action of the intrinsic, multiplicative chemical noise. The analytical results are quantitatively confirmed by stochastic simulations of the process. Our work provides a theoretical framework identifying key biochemical parameters controlling active phase separation on membrane scaffolds, offering testable predictions for experiments.

Figures (9)

• Figure 1: Schematic representation of the model as a chemical reaction network.
• Figure 2: Adiabatic relaxation in the $(k_\mathrm{e}^+/k_\mathrm{e}^-,K_\mathrm{d}/C)$ plane for $K_\mathrm{m}/C=10^{-2}$ and $\rho_0=3$. The ratio $\rho=k_\mathrm{e}^+/k_\mathrm{e}^-$ is restricted to the physical region comprised between the solid lines [Eq. (\ref{['eq:phys_strip']})] and centered at $\rho=\rho_0$. The $+$ phase is favored for $\rho>1$ (yellow), and the $-$ phase for $\rho<1$ (purple). The dotted lines enclose the bistable region [Eq. (\ref{['eq:bistability']})]. The arrows show the slow adiabatic drift of the system as $\langle\phi\rangle$ relaxes, driving it toward the phase coexistence line at $\rho=1$ (solid vertical line). A uniform state with $\langle\phi\rangle=-c$ corresponds to a point on the right boundary of the physical region [rightmost white dots, cf. Eq. (\ref{['eq:alphas']})], where the competing $+$ phase is favored. Random fluctuations within this uniform state will then generate regions of the $-$ phase that will grow, inducing a slow adiabatic drift of the system's representative point towards the phase coexistence line (leftward arrows). Decay towards the phase-coexistence state may take place either via homogeneous nucleation (upper arrow) or via a linear instability (lower arrow). A symmetric picture (rightward arrows) is observed for uniform states with $\langle\phi\rangle=+c$.
• Figure 3: (a,b) Steady-state phase diagram in the reduced parameter space $(\rho_0, K_\mathrm{d}/C,K_\mathrm{m}/C)$. Panel (a) shows a projection on the $(\rho_0,K_\mathrm{d}/C)$ plane with $K_\mathrm{m}/C$ fixed at 0.1. Panel (b) shows a projection on the $(\rho_0,K_\mathrm{m}/C)$ plane with $K_\mathrm{d}/C$ fixed at 0.1. We considered here $K^+_\mathrm{d}=K^-_\mathrm{d}=K_\mathrm{d}$ and $K_\mathrm{m}^+=K_\mathrm{m}^-=K_\mathrm{m}$. The solid white lines, defined implicitly by the conditions $\rho_0=\rho_\pm$ [Eqs. (\ref{['eq:def_rhos']}),(\ref{['eq:phase_coex']})] from the mean-field theory, separate regions of pure phases from regions of phase coexistence. The dashed white lines distinguish metastable from unstable regions. The background color represents the theoretical prediction for the steady-state order parameter $\langle\phi\rangle_\infty/c$ from Eq. (\ref{['eq:phi_eq']}). The color inside each circle denotes the measured $\langle \phi\rangle/c$ in the final time $t=10^5 k_D^{-1}$ of a numerical simulation with the corresponding parameters. (c) Time evolution of field configurations from simulations with different $\rho_0$ values, tuned via the ratio $E^+_\mathrm{tot}/E^-_\mathrm{tot}$. Here, $K_\mathrm{d}/C=0.1$. The mechanism of phase separation depends on $\rho_0$, and can proceed either via nucleation or linear instability, as predicted in panels (a,b). The colorscale is the same as in the phase diagram above. Simulations were performed with $k_\mathrm{b}\sim10^{-3} k_\mathrm{e}^\mathrm{max}$ (see App. \ref{['app:nucleation']}). (d) Domain coalescence driven by interface minimization. The simulation shows the merging of two initially separate domains, driven by minimization of the effective energy $\mathcal{F}$ concentrated at phase interfaces.
• Figure 4: Numerical measurements (symbols) of the steady-state average $\langle \phi \rangle$ compared with the theoretical prediction (solid line). The parameter $\rho_0$ is varied by independently tuning either the catalytic rates $k_\mathrm{c}^\pm$ (circles) or the total enzyme concentrations $E_\mathrm{tot}^\pm$ (diamonds). Simulations were performed with $K_\mathrm{d}/C=0.1$ and $K_\mathrm{m}/C=0.1$.
• Figure 5: Power consumption per unit interface length as a function of the catalytic rate $k_\mathrm{c}$ for both enzyme species ($+$: diamonds, $-$: circles). Numerical measurements of the frequency of catalytic events are compared with the predicted $\sim k_\mathrm{c}^{1/2}$ scaling (dashed lines). Left inset: A system configuration showing a domain of the $+$ phase (yellow) within the $-$ phase (purple). Right inset: The corresponding spatial distribution of power consumption, where yellow indicates the maximum value and purple indicates zero. Energy dissipation is concentrated at the phase interface, where futile cycles of antagonist reactions take place.