Computational Mapping of Proteins for Binding of Ligands

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Computational mapping moves molecular probes - small organic molecules containing various functional groups - around the protein surface, finds favorable positions using empirical free energy functions, clusters the conformations, and ranks the clusters on the basis of the average free energy. Mapping is an important step in fragment-based drug design, which starts with the identification and characterization of druggable binding sites, i.e., regions of the protein surface that are major contributors to the binding free energy. Based on NMR and X-ray screening experiments with fragment-sized compounds, such "hot spots" also bind a variety of small organic molecules, and hence the "hit rate" in protein mapping is a good predictor of druggability. We have developed the CSMAP algorithm that reproduces the available experimental mapping results. In applications to enzymes, the probes always cluster in major subsites of the active site, and the amino acid residues that interact with the probes also bind the specific ligands (primarily substrate and transition state analogues, inhibitors, and products). The method also applies to non-enzyme proteins, and provides detailed information on the binding sites. We have recently developed a novel mapping algorithm based on the Fast Fourier Transform (FFT) correlation approach that works with pairwise interaction potentials. The method retains the accuracy of the CSMAP algorithm, but reduces computing times by two orders of magnitude. This will give us the opportunity to map very large sets of proteins, and study the validity of the hypothesis that the method can identify the most important regions of protein binding sites. In addition to method development, the general goals of this proposal are the analysis and dissemination of the information obtained on protein binding sites, and a rigorous assessment of its value for fragment-based drug design, including the identification of druggable "hot spots" and the ability to predict correct positions for probes that are homologous to a functional group in the native ligand. We will consider a curated set of proteins with both high affinity and less potent ligands, map the proteins, and compare the probe-residue interactions to the protein-ligand interactions in the x-ray structures. It is expected that the largest consensus sites will be at "hot spots" that are critical for the binding of high affinity ligands. To study the relationship between the bound positions of specific probes and the positions of similar functional groups in the native ligand we will use both fragments of these ligands and compounds from an extended fragment library as probes. Results will help to understand both the principles that govern the weakly specific binding of small molecules in functional sites of proteins and the potential limitations of the mapping method.