Location: Montreal, Quebec
Conference Dates: August 5-12
Exhibition Booth: #231 – shared with CCG
Poster Number: MS96.P01.B534
Title: Accurate water site determination in macromolecular refinement using 3D-RSIM
Authors: Oleg Y. Borbulevycha, Jean-François Truchonb, Paul Labutec, and Lance M. Westerhoffa
aQuantumBio Inc., bVertex Pharmaceuticals Inc., cChemical Computing Group, Inc.
Abstract: Structure Based Drug Discovery (SBDD) is employed by virtually all pharmaceutical R&D organizations, and understanding the protein:ligand complex structure along with explicit solvent effects is necessary to obtain meaningful results from docking, thermodynamic calculations, and active site exploration. Phenix/DivCon is able to accurately elucidate the protein:ligand complex structure through in situ treatment of the structure using quantum mechanics; however, at standard SBDD resolutions, the crystallographic data unambiguously reveals only a small fraction of water molecules in protein crystals – even within the first hydrogen shell of the protein molecule. Further, the implicit solvent correction in conventional methods does not take into account non-linear effects of hydrogen bonding and dispersion interactions introduced by the nearest hydration shells.
To address this deficiency, we have used the 3D Reference Interaction Site Model (3D-RISM) method as implemented in MOE to filter crystallographic map data and create a more complete first solvation shell of the biomolecular complex. The combination, implemented within the Phenix/DivCon refinement workflow, allows us to capture weaker difference density peaks and thus “rescue” water sites that are normally undetectable using conventional crystallographic protocols. This workflow has been applied to “standard” resolution structures, and the results have been compared to corresponding higher-resolution structures. We have observed consistent improvements in R-factors and in water site determination. For example, the lysozyme structure PDBid:2EPE (2.5 Å) has 48 crystallographic waters while PDBid:193L (1.33 Å) has 142 waters. When considering overlapping sites, 2EPE captures 30% of the waters found in 193L. When our method is applied to 2EPE however, it is able to find almost 60% of the waters observed in the higher-resolution, and it is able to predict sites that may have been missed at the higher resolution.
Poster Number: MS96.P09.B542
Title: High-performance, quantum mechanics-based macromolecular x-ray refinement
Authors: Oleg Y. Borbulevych and Lance M. Westerhoff
QuantumBio Inc., State College, PA 16801, USA
Abstract: Modern, structure based drug discovery (SBDD) is dependent upon accurate protein:ligand structure determination and characterization. In conventional x-ray refinement, the geometry of the ligand within the active site is modeled according to the practitioner’s beliefs as expressed in the form of stereochemical restraints provided by the ligand library or CIF file. Further, metal centers, bound species, and so on can be difficult to refine correctly without significant human intervention. Our work has addressed this problem through the integration of DivCon6 – a linear scaling, semiempirical, quantum mechanics (SE-QM) functional – with the Phenix refinement package. With Phenix/DivCon, SE-QM is used in “real-time” during each microcycle over the course of the refinement. With its inclusion of electrostatics, charge transfer, polarization, dispersion, hydrogen bonds, etcetera, this method is a much more rigorous, robust alternative to conventional stereochemical restraints and is better able to accurately model protein:ligand structures without “tweaking” any restraints.
We report PM6 refinement results for several key examples including structures with metal coordination spheres, covalent bonds, and other exotic protein:ligand chemistry situations. When compared with the originally deposited PDB structures, we found in all cases that QM refinement leads to ligand structures with much lower strain, and in some cases, the improvement is dramatic and as much 10+ fold. At the same time, SE-QM methods are better able to capture the influence of the surrounding structure (e.g. active site) on the ligand. These interactions are particularly interesting in SBDD as they are often the targets for lead design and optimization, and examples that illustrate how these interactions are captured with SE-QM will also be discussed.