In This Issue

1146 Racemic crystallography of synthetic protein enantiomers used to determine the X-ray structure of plectasin by direct methods

Kalyaneswar Mandal, Brad L. Pentelute, Valentina Tereshko, Vilasak Thammavongsa, Olaf Schneewind, Anthony A. Kossiakoff, and Stephen B. H. Kent

In the report by Mandal, et al., mirror image forms of the antibiotic fungal protein plectasin were prepared by total chemical synthesis. Only L-plectasin was active: D-plectasin was devoid of antimicrobial activity. Co-crystallization of the racemic mixture (L-plectasin + D-plectasin) gave highly ordered crystals with unusually low solvent content. Synchrotron X-ray diffraction data was collected to high resolution from a racemic plectasin crystal. The proteins crystallized in the achiral spacegroup P-1, with one L-plectasin molecule and one D-plectasin molecule in the unit cell. The structure of the plectasin protein molecule was solved using direct methods, more commonly used in small molecule crystallography, to calculate the electron density map.

1321 On the occurrence of linear groups in proteins

Scott A. Hollingsworth, Donald S. Berkholz, and P. Andrew Karplus

Standard protein structures based on a repeating backbone conformation, such as the α-helix and β-strand, are a fundamental aspect of protein structure. Yet, how this topic is handled in biochemistry texts is still largely based on theoretical concepts from the 1960s, possibly because there has been no single easily accessible literature study that directly addresses the relevant questions. The study by Hollingsworth, et al. provides a simple empirical analysis of such groups in real proteins and presents a set of concrete insights. It should move the field forward, impacting both how we think about and how we teach the fundamentals of protein structure.

1293 Motif-directed flexible backbone design of functional interactions

James J. Havranek and David Baker

One goal of computational protein design is to modify existing proteins to perform alternate, even novel, functions. A fundamental challenge in the field is how to reconcile the need for versatility in the placement of functional side chains with the commonly imposed simplification of a fixed backbone. One difficulty is that the optimal backbone for a given design problem depends on the identity of the side chains, which are of course the unknowns of the problem. Using DNA-binding proteins as an example, Havranek and Baker present a method for constructing backbone conformations capable of making functional interactions similar to those observed in experimentally determined structures.

1306 Evolutionary constraints on structural similarity in orthologs and paralogs

Mark E. Peterson, Feng Chen, Jeffery G. Saven, David S. Roos, Patricia C. Babbitt, and Andrej Sali

Evolutionary biologists and bioinformaticians have devoted considerable effort to identifying and distinguishing orthologous and paralogous proteins. Whereas orthologs have diverged as a consequence of speciation and tend to retain function, paralogs are derived by gene duplication and are more prone to functional divergence. The report by Peterson, et al. assesses the impact of these evolutionary relationships on the fundamental relationship between protein sequence and structure. Orthologous pairs of proteins can share the same level of structural similarity as paralogous pairs with higher sequence identity (up to 17 percentage points higher). These findings provide a means of improving protein structure prediction and better understanding protein evolution.