Membrane fusion is an essential and ubiquitous cellular process. It is involved in a variety of functions and stages of a cell’s life, including import of foodstuffs and export of waste, signaling between nerve cells, fertilization and virus infection.Fusion occurs when membranes come into contact and local perturbations of the bilayers lead to the formation and subsequent expansion of fusion pores, or necks. Previous studies with both electron and atomic force microscopies have suggested the formation of single fusion necks with a diameter of 50 to 100 nm, but direct imaging of the formation dynamics with the necessary temporal resolution has yet to be reported. Optical video microscopy holds promise for such a purpose, but until recently, the method offered only millisecond resolution.The investigators explored the dynamics of membrane fusion, which occurs on a very fast time scale, by triggering fusion in a controlled manner and recording it with a fast digital camera. They used two protocols to trigger membrane fusion. In the first, shown here, they brought two functionalized lipid vesicles into contact (a) and injected multivalent ions into the area of contact, resulting in formation of intermembrane complexes (b). This led to the opening of the fusion neck (c), the dynamics of which were previously unreported.In the Oct. 24 issue of PNAS, researchers with Max Planck Institute of Colloids and Interfaces in Potsdam, Germany, and with Collège de France in Paris reported a study in which they performed direct imaging of membrane fusion with a temporal resolution of 50 μs. To achieve this, they used a very fast digital camera — the type used by the automobile industry to check air-bag deployment during crash tests. Even then, direct imaging of membrane fusion was possible only when they triggered fusion in a controlled manner. To this end, they used two protocols: fusion, for functionalized membranes, and electrofusion.Researchers have reported a method for direct imaging of membrane fusion, an essential cellular process. Shown here are confocalmicroscopy images of two lipid vesicles, labeled with different fluorescent dyes, prior to fusion (a) and after fusion induced by an electrical pulse (b). Panel (c) contains a three-dimensional image of a two-domain vesicle resulting from fusion of membranes with different composition.The setup consisted of a microscope made by Carl Zeiss of Jena, Germany, outfitted with 20× and 40× phase-contrast objectives. The researchers recorded the fusion events with an HG-100K fast digital camera made by Redlake MASD LLC of Tucson, Ariz., that was mounted on the microscope. They acquired image sequences at a frequency of up to 20,000 fps. The onboard memory of the camera limited acquisition time to 2 s at this frequency.In the other protocol used for controlled membrane fusion, researchers brought two lipid vesicles into contact using an AC field (a) and created a pore by applying a short DC pulse (b). As a result, the lipids from the opposing bilayers mixed, leading to the opening of the fusion neck (c).Imaging fusionThe fusion protocols were either new or applied in a novel way. The researchers developed the fusion protocol for functionalized membranes. The electrofusion protocol is already widely used with cells and vesicles 100 nm in size, which cannot be seen with an optical microscope. In the present study, they applied it to giant vesicles, which they could observe directly.Both protocols allowed them to induce fusion in a controlled manner. In the fusion protocol for functionalized membranes, they placed a solution with vesicles functionalized with β-diketone ligand molecules in a specially designed chamber with three micropipettes: two for vesicle manipulation and one for ion injection. They isolated two vesicles, manipulated them so that they were adjacent to one another, and injected EuCl3 to induce adhesion and fusion. They examined ~50 vesicle couples in this protocol, observing fusion in five of them. They recorded fusion events with a CCD camera in three of the five, at a frequency of 28 fps, and with the fast digital camera in the other two.The fusion events they observed consisted of a prefusion stage, formation of the fusion neck and, finally, opening of the neck. The prefusion stage saw vesicles adhering to one another and forming an extended contact zone. The contact area slowly decreased as the prefusion stage progressed, until it was no longer discernible with optical microscopy.In the electrofusion protocol, they placed a vesicle solution in a chamber with two electrodes, positioned at a distance of 475 ±5 μm. They applied an alternating electric field for about 10 s and then a DC pulse. The AC field served to line up the vesicles along the direction of the field, much like the micropipettes were used to manipulate the vesicles in the other protocol, bringing them into contact with one another. The DC field perturbed the vesicles, inducing membrane fusion. In vesicle solutions with no salt, fusion took place at several contact points. In solutions with salt, the perturbation produced a flat contact zone between the vesicles, followed by the formation of a single fusion neck. The investigators recorded fusion events in 10 vesicle couples in this protocol.The direct imaging afforded by the two fusion protocols and the fast digital camera confirmed that the fusion process is extremely fast, and it offered insight into the dynamics of the process. The scientists found that the opening of the fusion pore occurs with an expansion velocity of a few centimeters per second, said Rumiana Dimova, a researcher at the institute and principal investigator of the study. “From this observation we could conclude that, for the formation of a fusion neck, the cell needs only some few hundred nanoseconds.”Having demonstrated the potential of the method for controlling and imaging membrane fusion, the researchers hope to apply it to more sophisticated membrane systems — the system used in the present study was very simple, composed of a single lipid. Dimova noted several possible directions. The first is to test the method in real cells. The second involves a current hypothesis about lipid rafts in cells that are important for cell signaling or for the functionality of various proteins in the cell membrane. “With our method, we should be able to fuse such a raftlike vesicle with a real cell and test whether the produced domain [or raft] would survive.”Finally, the scientists plan to use vesicle fusion as a tool to create microreactors with very small volumes. The vesicles used in the present study were only tens of microns in size. “Fusing two of these vesicles of different content would be equivalent to performing a reaction in a tiny volume of some picoliters,” Dimova said. “This would be advantageous for synthesis of nanomaterials.”