Demixing on curved surfaces

Like Comment
Read the Paper

From the helical coiling in plants to patterns of bird flocks in the sky above us, biological systems can be elegantly described by their curvature, the measure of how much a shape deviates from flatness. In cellular biology, curvature is omnipresent. For instance, membranes can be curved by the underlying cytoskeleton or by membrane proteins, in order to achieve specific functions. The interaction between membrane curvature and the membrane’s physical processes is a fascinating and yet elusive subject. This work, led by Daniela Kraft and Luca Giomi with first authors Melissa Rinaldin and Piermarco Fonda, sheds some light on the interplay between curvature and phase separation.

In artificial lipid membranes, lipids can segregate into coexisting domains, like oil droplets in vinaigrette. Oil molecules tend to neighbor with other oil molecules to minimize the oil-water interface. However, in phase-separated membranes, interface minimization is not the only factor at play. The liquid phases have different stiffness because of their molecular structures. To minimize the elastic energy, the softer phase prefers to localize in high curvature regions and the stiffer phase prefers to occupy flatter areas. However, this is not the whole story, because a membrane can change its shape during the phase separation process. This makes it difficult to understand whether a membrane shape is induced by a phase separation pattern, or whether a phase separation pattern induces a specific membrane shape.

To solve this chicken-and-egg-problem, the authors have fixed the membrane geometry to specific shapes. In this way, they can understand how geometry influences phase separation patterns. In the experiments, they use micrometer-sized particles that are spherical, cubic, dumbbell, and snowman-shaped, which are covered with a multicomponent lipid membrane that fully envelops the particles.

The authors find that the softer phase is attracted to regions of high curvature, as expected. However, surprisingly, this only happens for a small amount of soft phase and high membrane curvature. Even more unexpectedly, the authors observe that geometry affects the internal lipid composition of phase-separated domains, a process that the authors term antimixing. For a long time, the authors were puzzled by this phenomenon, and in particular, they asked themselves: “Is antimixing a mixed or phase-separated state?” Only a continuous discussion between the experimental and theoretical groups led to the solution of this mystery: it is a mixed state because the lipid compositions of the two lobes of the dumbbells lie on antipodal sides of the miscibility gap, or in other words, outside the phase separation region.

Overall, this work shows that geometry can affect a phase-separated system dramatically. The authors hope that this study can provide important insights into the broader field of biology, as the concept of liquid-liquid phase separation is understood as an organizational pathway. The paper can be found on Nature Communications under the title "Geometric pinning and antimixing in scaffolded lipid vesicles".

Melissa Rinaldin

HFSP postdoctoral fellow, Brandeis University

I am a curiosity-driven experimental physicist interested in the organization of biological systems. My research is highly interdisciplinary and the work described in this article is part of my doctoral dissertation performed in the groups of Daniela Kraft and Luca Giomi at Leiden University.

No comments yet.