Computational Fragment-Based Binding Site Identification by Ligand Competitive Saturation

Olgun Guvench and Alexander D MacKerell

PLOS Computational Biology, 2009, vol. 5, issue 7, 1-10

Abstract: Fragment-based drug discovery using NMR and x-ray crystallographic methods has proven utility but also non-trivial time, materials, and labor costs. Current computational fragment-based approaches circumvent these issues but suffer from limited representations of protein flexibility and solvation effects, leading to difficulties with rigorous ranking of fragment affinities. To overcome these limitations we describe an explicit solvent all-atom molecular dynamics methodology (SILCS: Site Identification by Ligand Competitive Saturation) that uses small aliphatic and aromatic molecules plus water molecules to map the affinity pattern of a protein for hydrophobic groups, aromatic groups, hydrogen bond donors, and hydrogen bond acceptors. By simultaneously incorporating ligands representative of all these functionalities, the method is an in silico free energy-based competition assay that generates three-dimensional probability maps of fragment binding (FragMaps) indicating favorable fragment∶protein interactions. Applied to the two-fold symmetric oncoprotein BCL-6, the SILCS method yields two-fold symmetric FragMaps that recapitulate the crystallographic binding modes of the SMRT and BCOR peptides. These FragMaps account both for important sequence and structure differences in the C-terminal halves of the two peptides and also the high mobility of the BCL-6 His116 sidechain in the peptide-binding groove. Such SILCS FragMaps can be used to qualitatively inform the design of small-molecule inhibitors or as scoring grids for high-throughput in silico docking that incorporate both an atomic-level description of solvation and protein flexibility.Author Summary: Fragment-based drug discovery is based on a simple yet powerful principle: instead of trying to screen through the vast number of possible drug-like compounds during the drug discovery process, screen representative drug-like fragments, which are far fewer in number. Once a suitable fragment is discovered, it can then be built up or linked with other fragments to give a drug-like molecule. Because such fragments are small, even “good” fragments bind weakly to their targets, therefore requiring significant time, labor, and materials costs for experimental detection and characterization of binding. In the present work, we describe a computational approach to the problem of detecting and characterizing fragment binding. Importantly, the method provides atomic-resolution results and also explicitly takes into account the effect that molecular water has on binding and the inherent flexibility of protein targets. The methodology is demonstrated by application to the BCL-6 protein, which is implicated in a variety of cancers, is conceptually easy to understand, and can yield results in a matter of days using present-day commodity computers.

Date: 2009
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1000435 (text/html)
https://journals.plos.org/ploscompbiol/article/fil ... 00435&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pcbi00:1000435

DOI: 10.1371/journal.pcbi.1000435

Access Statistics for this article

More articles in PLOS Computational Biology from Public Library of Science
Bibliographic data for series maintained by ploscompbiol ().