Aggregation and vesiculation of membrane proteins by curvature-mediated interactions
Entry by Sandeep Koshy, AP 225, Fall 2010
Title: Aggregation and vesiculation of membrane proteins by curvature-mediated interactions.
Authors: Benedict J. Reynwar, Gregoria Illya, Vagelis A. Harmandaris, Martin M. Muller, Kurt Kremer & Markus Deserno
This work uses computational simulation to explore the role of curvature-induced interactions on driving the vesiculation process on lipid membranes. Simulations reveal that both cap-like membrane proteins and sphere-like particles can cause vesicles to bud from the membrane which requires an energy greater than thermal fluctuations. These results may be important to those which study biological membranes and viruses as well as researchers who are engineering nanoparticles to cross the cell membrane.
Soft Matter Keywords: membrane, curvature-mediated interactions, elasticity, thermal energy
Cells use membrane remodeling to both ingest and eject materials for their survival and replication as well as maintain transport between their internal organelles. They perform this task by using specialized proteins which impart curvature on the membrane, cause tabulation and eventually cause budding off of a piece of membrane. The cost of creating such a vesicle (about 500 times the thermal energy) is much greater than the attractive energy between surface proteins. Thus, it is thought that the curvature induced by the presence of membrane proteins as they approach eachother supplies an elastic membrane energy which could be sufficient to cause budding of a vesicle. The authors of this work sought to explore this phenomena using computational methods.
The authors calculated that a membrane of 100 nm would need to be simulated on the time scale of milliseconds to see if such curvature induced budding would occur. They used a course-grained model, which does not account for atomistic events, for efficient simulation of this process. This model was sufficient to reproduce the bending elasticity of the membrane.
The authors used a membrane consisting of lipids consisting of three balls which assemble due to tail attractions (Fig. 1). A time-scale of τ=15 ns was used which is on the order of the lipid diffusion time scale. Two sizes of curvature inducing cap-like proteins were used. A capsid or sphere shape was also used in later simulations to model viruses or nanoparticles. The caps and spheres contained both membrane attracting (purple) and repulsive (pink) elements.
Effect of membrane bending proteins
The effect of placing 36 cap proteins on the membrane was first explored (Fig. 2 A). Small caps did not readily cluster during the length of simulation. Large caps however clustered on the timescale of 40,000τ (Fig 2. B,C) that vesiculated after an additional 30,000τ (Fig 2. D-F). The larger the cap size, the smaller the vesicles that formed since a fewer number of caps were required to generate sufficient curvature for vesicle budding.
The authors comment that this result ignores much of known cell biology, such as the presence of cytoskeletal elements and specific proteins contacts in the classic clathrin-dependent exocytosis pathway.
Vesicle formation using spherical particles
The next simulation scenario involved the use of 16 equally sized, spherical particles on the membrane surface (Fig. 3 A). These spheres contained 75% membrane-attractive regions which were embedded in the membrane (Fig. 3 B). Upon commencing simulation, pairs of particles were formed (Fig. 3 C) which was followed by budding of vesicles (Fig 3 D-F). The authors note that this budding resembles that observed with Mason-Pfizer monkey viruses which need cooperative binding to exit cells.
Attraction of spheres within the membrane
The force between the two particles was quantified (Fig. 4). At relatively long distances, an attractive force as the curvatures of the two particles overlap. At short distances, there is a repulsion effect due to contact. They calculate an interaction energy of about 10 times the thermal energy which suggests that the interactions are not due to fluctuations but truly due to curvature effects. They note that as the capsids approach eachother, there is an interplay of two different curvatures which results in the spheres being tilted towards each other, allowing further attraction.
The authors suggest that this effect is universal and should be present in biology. They hypothesize that cells must have a mechanism for preventing these events from occurring frequently.