Abstract
As cryogenic electron microscopy (cryoEM) gains traction in the structural biology community as a method of choice for determining atomic structures of biological complexes, it has been increasingly recognized that many complexes that behave well under conventional negative-stain electron microscopy tend to have preferential orientation, aggregate or simply mysteriously "disappear" on cryoEM grids. However, the reasons for such misbehavior are not well understood, which limits systematic approaches to solving the problem. Here, we have developed a theoretical formulation that explains these observations. Our formulation predicts that all particles migrate to the air-water interface (AWI) to lower the total potential surface energy-rationalizing the use of surfactant, which is a direct solution to reduce the surface tension of the aqueous solution. By performing cryogenic electron tomography (cryoET) on the widely-tested sample, GroEL, we demonstrate that, in a standard buffer solution, nearly all particles migrate to the AWI. Gradually reducing the surface tension by introducing surfactants decreased the percentage of particles exposed to the surface. By conducting single-particle cryoEM, we confirm that suitable surfactants do not damage the biological complex, thus suggesting that they might provide a practical, simple, and general solution to the problem for high-resolution cryoEM. Applying this solution to a real-world AWI adsorption problem involving a more challenging membrane protein, namely, the ClC-1 channel, has resulted in its near-atomic structure determination using cryoEM.