Acta Cryst. (1998). D54, 708-709.

Macromolecular crystallography. Part A. Methods in Enzymology (Vol. 276.)

Edited by Charles W. Carter Jr and Robert M. Sweet.

By the end of 1967 there were five protein crystal structures solved by X-ray diffraction methods (myoglobin, lysozyme, chymotrypsin, carboxypeptidase and ribonuclease). Each had been the product of several years of work. By the end of 1997, the Protein Data Bank contained 6291 protein crystal structures. Not all of these are independent structures, some represent known structures in ligand complexes or with site-directed mutations. Nevertheless, the 1000-fold increase in solved structures over the last 30 years indicates the relative ease with which some macromolecular crystal structures can be solved. The outsider might well consider that methods have become routine, as has been achieved by the highly automated and computationally efficient methods for small-molecule crystal structures. However, although some steps of macromolecular crystallography have been automated, there are many areas where the professional practitioner needs to have a thorough understanding of the methods in order to achieve results. The new volumes of Macromolecular Crystallography in the `Methods in Enzymology' series provide the basis for this understanding; they contain information at the forefront of research by some of the major practitioners in the field. The articles elaborate on the original research papers and evaluate methods in terms of successful applications. The overview sections will interest the non-professional, but the remainder of the chapters are not for the faint hearted; they are for graduate students and research workers who wish to master the fundamentals of their subject. Part A, reviewed here, covers crystallization, data collection, phasing and molecular replacement. Part B, reviewed by Professor Adman, covers map interpretation, refinement, display and validation of structures, dynamics and databases.

The recombinant DNA revolution and advances in protein purification techniques have made major contributions to the success of protein crystallography. Cloning, expression and purification methods are outside the scope of this volume and the book begins by assuming that about 1 mg of pure protein is available for crystallization trials. The material on crystallization provides both an overview and more specialist sections. The beginner could well get going by reading the overview and applying the methods of hanging-drop vapour diffusion, with the help of the Hampton Research crystal screen kit. This company supplies prepared cocktails that have proved most productive in crystallization trials with other proteins. The contributions on the use of cosolvents, understanding of physical-chemical principles, especially consideration of temporal factors in approaches to equilibrium, and the use of physical techniques to examine the state of aggregation of the protein are valuable. Sections on membrane proteins and RNA crystallization add specialist diversity. For `desperate situations', the sections on the use of two-dimensional crystals, based on adsorption of proteins to lipid layers, and the chemical modification of proteins, such as reductive methylation that was used to obtain myosin sub-fragment 1 crystals, provide imaginative proposals. Crystallization is not yet routine. These chapters provide a good summary of the present state of the art. Reliable precise structures depend on good intensity data that in turn depend on good crystals. The chapters on data collection cover the topics that have had most impact on the ease with which crystal structures may be solved. Flash freezing of crystals to 100K or lower alleviates radiation damage and crystals become almost immortal in the X-ray beam. Synchrotron radiation provides the brilliance needed so that even small crystals (10µm or less) or very large unit cells may yield accurate data. These topics are covered well, together with a description of the most widely used image-plate area detectors and their likely successors, charge-coupled device based (CCD) detectors. These sections will provide the student with an understanding of the physical processes (e.g. the distinction between flux, brightness and brilliance of synchrotron radiation beams) and their applications. They are followed by descriptions of data-processing software (particularly MADNES and DENZO) and it is valuable to have an up-to-date account of these programs that are all too often used as `black boxes'.

After expression and crystallization, phasing presents the next critical step in a crystal structure determination. The section describing phase determination begins with a long discussion of Bayesian methods. These are methods aimed at reconciling numerical computation and human decision making, and are based on sound probability distributions and resolution of ambiguities by systematic evaluation of multiple hypotheses about missing information. Eventually, the methods hold promise for ab initio phase determination but their most definite contributions to date have been to provide improved maximum-likelihood methods for heavy-atom refinement with the program SHARP (described in this volume) and for macromolecular refinement (described in Part B). I particularly enjoyed the sections on multiple isomorphous replacement, that contain a league table of successful derivatives (suitably qualified with the comment that the quality of the heavy-atom derivative is a more important criterion than the number of times a particular heavy-atom reagent has been used), and the sections on multiwavelength anomalous diffraction (MAD) methods that include the preparation of selenomethionyl proteins for phase determination. Wayne Hendrickson reports that since its first practical demonstration in 1985, the application of MAD phasing has resulted in solution of over 40 structures in the decade to 1995, nearly half of them solved in 1995. The explosion in the use of MAD methods has been made possible by cryocrystallography, by increased number of synchrotron beamlines with facilities for tuning radiation wavelength, and by the increased ease with which anomalous scatterers may be incorporated into macromolecular crystals. The application of MAD phasing is likely to make a most significant impact in the coming years.

Patterson search correlation or molecular replacement methods have proved effective for solution of structures where a similar structure is already known. As the database of protein folds increases, these methods are finding increased application. The accounts of molecular replacement methods provide a sound background and give useful advice, for example for those problems where the starting model may not be an accurate representation of the structure sought. The book ends with `horizon' methods. There is an entertaining description of crystal structures of racemic mixtures, although the methods are restricted to those molecules that can be synthesized and crystallized in both d and l forms. The sections on ab initio phasing using very high resolution data and the computation of very low resolution phases, that will allow us to bridge the gap between electron microscopy and X-ray crystallography images, point the way to the future.

This is an excellent volume. It is recommended to all graduate students and postdoctoral workers in macromolecular crystallography. I shall be purchasing both volumes for my laboratory and I shall expect to refer to my copy frequently.

Louise N. Johnson

University of Oxford
Laboratory of Molecular Biophysics
The Rex Richards Building
South Parks Road
Oxford OX1 3QU
England

Copyright © 1997 International Union of Crystallography