Molecular Structure Determination by High Resolution Electron Microscopy

Abstract

The application of electron microscopy in the determination of molecular and macromolecular structure at high resolution has been a tantalizing prospect for many years. The advent of commercial instruments offering a 5-A resolving power just less than two decades ago should have ushered in an era of high resolution structure determination in biology. But it did not. Although metallurgists and materials scientists soon obtained images of lattice planes and defects in crystals commensurate with the potential of the instruments, images of biological specimens were interpret­ able at best at a level of 20 A, and heated arguments concerning structure even at the 50or 80-A. level were the order of the day. Nevertheless, even with such resolutions, knowledge of biological structure increased tremendously, and it still does today. In fact, the attainment of 20-A resolution in a biological specimen often is considered high resolution work. However, I restrict the definition of high resolution for the purposes of this article to mean the size of interpretable structural detail in the image that approaches the potential resolution of the electron microscope, which today is 3 to 5 A. The visualization of biological materials in the electron microscope suf­ fers from the fact that electrons interact very inefficiently with the light atoms constituting such a specimen, compared to the interaction with heavier atoms that make up most inorganic materials. Therefore the con129