Current approaches to macromolecular crystallization.

Given our current expertise, and the certain future developments in genetically altering organisms to produce proteins of modified structure and function, the concept of protein engineering is nearing reality. Similarly, our ability to describe and utilize protein structure and to define interactions with ligands has made possible the rational design of new drugs and pharmacological agents. Even in the absence of any intention toward applied use or value, the correlation of regulation, mechanism, and function of proteins with their detailed molecular structure has now become a primary concern of modern biochemistry and molecular biology. At the present time, there are numerous physical-chemical approaches that yield information regarding macromolecular structure. Some of these methods, such as NMR and molecular dynamics, are becoming increasingly valuable in defining detailed protein structure, particularly for lower-molecularmass proteins. There is, however, only one general technique that yields a detailed and precise description, in useful mathematical terms, of a macromolecule’s structure, a description that can serve as a basis for drug design, and an intelligent guide for protein engineering. The method is X-ray diffraction analysis of single crystals of proteins, nucleic acids, and their complexes with one another and with conventional small molecules. Some inspirational examples of representative crystals are shown in Fig. 1 - 3. In the past 20 years, the practice of X-ray crystallography has made enormous strides. Nearly all of the critical and timeconsuming components of the technique have been improved, accelerated, and refined. X-ray crystallography today is not simply an awesome method used by physical chemists to reveal the vast beauty of macromolecular architecture; it is a practical, reliable, and relatively rapid means to obtain straightforward answers to perplexing questions. X-ray diffraction data that once required years to obtain, can now be collected in a matter of weeks, even days in some cases. Computers of extraordinary speed and capacity are now common tools as are computer graphics systems of a versatility and cleverness that would have been unimaginable only a few years ago. Software, too, exists that is sophisticated yet friendly, flexible yet reliable, and readily available to anyone in need of it. The question, then, is where does the problem lie? What prevents the full utilization and exploitation of this enormously powerful approach. The answer, of course, is that for application of the method to a particular macromolecule, the protein or nucleic acid