Membrane bending by protein phase separation

Giulia Paci and Gautam Dey

Context

How do cells bend membranes? 

The remodelling of membranes into high-curvature shapes is essential for many cellular processes including vesicles formation, virus budding and cytokinesis [1]. These bent surfaces can be oriented both towards and away of the cytoplasm, and their formation entails a substantial energetic cost. Membrane bending can be driven by changes in lipid composition and clustering of transmembrane proteins, but to achieve nanoscopic curvatures (below the micrometre scale) cells often exploit protein scaffolds. These include for example crescent-shaped BAR domains, that bind and impose their intrinsic curvature to membranes and ESCRTs that polymerize into spiral-shaped filaments, driving membrane constriction. 

 Going beyond this “rigid template” paradigm, recent work has started to focus on the intrinsically disordered regions of membrane-binding proteins [2,3], which lack a stable structure. Multivalent binding of intrinsically disordered proteins (IDPs) can give rise to liquid-liquid phase separation, where the proteins condense in a dynamic phase with liquid properties. A large body of research in recent years has shown that this phenomenon is ubiquitous and plays a role in countless biological processes: here, Yuan et al investigate whether it can also act as a novel driver of membrane remodelling.  

 

Key outcomes 

The authors use one of the most well-characterized phase-separating proteins, fused in sarcoma (FUS), as a model and assemble it on the surface of synthetic and cell-derived membranes at different concentrations. 

Above a minimum threshold concentration, they observe phase separation of the protein on the membrane surface and the striking spontaneous formation of inward-pointing tubules. The tubules are lined by proteins and display different conformations, from undulating shapes to “string of pearls” and sub-diffraction structures (Figure 1). In addition to its dependence on protein concentration, tubule formation can be tuned by altering the strength of protein-protein and protein-membrane interactions. In order to understand the mechanism driving membrane bending, the authors develop a continuum mechanical model. Their simulations show that the compressive stress triggered by protein phase separation onto the surface can trigger the formation of tubules with similar morphologies to those observed in the experiments. The model predicts the scaling of tubule diameter with membrane rigidity and its ratio to the rigidity of the protein layer – predictions the authors are able to confirm experimentally. 

 

 

Perspective 

These new exciting findings by Yuan et al reveal a novel membrane-bending mechanism that does not rely on rigid protein scaffolds, but is driven by protein phase-separation. Given how ubiquitous liquid-liquid phase separation appears to be in cells, it will be particularly exciting to focus on the material properties of protein condensates and their potential mechanical interactions with different cellular structures. The preprint also raises the  intriguing possibility that structured protein domains might act cooperatively with phase-separated disordered domains to sculpt membranes. 

Questions for the authors 

1. The idea of this new paradigm for membrane deformation is really intriguing: could the authors speculate about which proteins  might reach sufficient concentrations within cells for phase separation while also binding membranes strongly enough?
2.  If we consider a potential role for this process in the formation of trafficking vesicles, would one expect that the protein-lined vesicles  have different properties in terms of their stability and ability to fuse with other membranes? Do you ever observe fusion of different tubules? 
3. What factors determine the differential enrichment of protein in the tubule vs the rest of the membrane surface? Depending on the condition this seems to vary from uniform intensity of protein in the tubule and membrane, to total enrichment in the tubule. 
4. In a related question, do the authors think that this new membrane bending mechanism could be tuned effectively enough to be exploited in synthetic applications – for example, to deliver proteins into cells? 

 

References

[1] McMahon, H. T. & Boucrot, E. Membrane curvature at a glance. J. Cell Sci. 128, 1065–1070 (2015).

[2] Busch, D. J. et al. Intrinsically disordered proteins drive membrane curvature. Nat. Commun. 6, 1–11 (2015).

[3] Snead, W. T. et al. BAR scaffolds drive membrane fission by crowding disordered domains. J. Cell Biol. 218, 664–682 (2019).

 

Tags: membrane bending, phase separation

Posted on: 1st June 2020 , updated on: 16th June 2020

doi: https://doi.org/10.1242/prelights.21454

Read preprint (No Ratings Yet)