Angewandte Chemie International Edition
© WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
For full article and contact information, see Angew. Chem. Int. Ed. 2001, 40 (17), 3141 - 3144
Peeking into Proteins' Pockets
Recognizing functional similarities between proteins by comparing their binding cavities
The human genome has been unlocked. However, most questions about biological processes - and thus about the origins of diseases - cannot be answered by analysis of the genome alone. There are millions of proteins that actually translate the information contained in the genome into biological functions. Only an understanding of all of the proteins, their spatial structures, their function, and their state within the cell, the proteome, can bring some enlightenment. But how does one recognize the functionality of a previously unknown protein?
Gerhard Klebe, Stefan Schmitt, and Manfred Hendlich explain why this is not a trivial task, "In the near future, we will be able to quickly determine the structures of many proteins through modern techniques such as high-throughput X-ray crystallography. However, this doesn’t come close to revealing their biochemical functionality, which is not necessarily determined by the sequence of amino acids and the folding pattern of a protein."
In contrast to previous strategies, which compare the amino acid sequence and folding pattern with known proteins, the Marburg researchers base their method on the form and composition of the protein's "binding cavity". In order to do its job, the protein must recognize a substrate or its ligands. For this, the protein has pocket-like depressions in its surface, into which these molecules fit snugly. An enzymatic reaction additionally requires a specific spatial orientation of the two reactants to each other. The enzyme must therefore have a very precise arrangement of individual binding sites. Proteins with related functions should thus have similarities in their "pockets".
Klebe and his co-workers have developed a method by which the binding cavities can be easily identified and their binding sites represented on a computer. The full data set containing well over 30,000 cavities was incorporated into a data bank, which can be searched for similarities in the pockets' substructures. In this way, functional relatives of unknown proteins can be found.
During their test runs the researchers made an interesting discovery quite by coincidence. A synthetic inhibitor that binds to the active center of an HIV protease leaves one binding site, an adenine binding site, open. Klebe says, " a valuable tip for the design of pharmaceuticals: inclusion of an adenine appendage in the inhibitor could make it more effective."