INPHARMA: an efficient NMR-based methodology for structure-based drug design
Final Report Abstract
A key component to success in structure based drug design is reliable information on protein-ligand interactions. High quality structural information is commonly obtained by X-ray crystallography, which is often limited by high costs and compatibility with crystallization – both issues potentially forcing the dismissal of the SBDD route for the target of interest. Conversely, computational docking protocols are rapid, inexpensive and dependent only on the availability of an atomic resolution structure of the target protein. Despite being a driving force in SBDD, docking protocols suffer from a limited reliability of the prediction of the interaction mode for arbitrary proteinligand pairs, partly due to the inability to account for protein flexibility and entropic effects in an accurate manner. INPHARMA (Interligand NOEs for PHARmacophore MApping) exploits NOEs occurring between two ligands binding competitively to the same target protein measured in a “ligand-detecting” NOESY experiment. These interligand NOE crosspeaks result from magnetization transfer mediated by receptor protons and are not limited by the protein size or availability of isotope-labeled protein. The technique requires low affinity ligands, making it attractive for lead generation. Interligand INPHARMA-NOEs can be theoretically estimated for pairs of protein-ligand complexes (generated e.g. by protein-ligand docking), and subsequently compared to the experimental ones to discriminate the correct binding modes. However, the solutions attained from fitting experimental and theoretical interligand NOEs are often not unique, and might lead to false positive hits or ambiguous answers. Most of the NMR-based techniques in SBDD have been benchmarked using the receptor structures in which the protein-ligand was crystallized. Thus, their performance in more realistic docking experiments, where the exact conformation of the receptor structure is not known a priori, is still unclear. During this project we demonstrated that the performance of INPHARMA deteriorates when the starting structure for the docking experiments is inaccurate. We analysed the performance in a range of scenarios, encompassing traditionally difficult cases such as docking to homology models and ligand dependent structural rearrangements. In this project we addressed the problems of (1) false positives in the INPHARMA-guided selection of docking modes and (2) poor performance of both the docking experiments and the INPHARMA selection in the presence of protein rearrangements. For (1), we developed a new algorithm, INPHARMA-STRING, which exploits the criterion of consensus between the docking modes selected by multiple sets of ligand-ligand INPHARMA-NOEs measured for the pairwise combination of ligands in a > 2 set. INPHARMA-STRING shows a remarkable performance of filtering false positives, even in the presence of sparse NMR data. For (2), we developed an “ensemble docking” protocol, which, when coupled with the INPHARMA-STRING selection, shows excellent results also in the case of induced-fit or when using an inaccurate protein structure, obtained for example by homology modeling. This protocol should be widely applicable for protein-ligand docking and highlights the important role of NMR-based approaches in SBDD. Furthermore, we validated INPHARMA-STRING on two additional systems consisting of the proteins CDK-2 and PDE10a in complex with 5 ligands each.
Publications
-
“An NMR- based scoring function improves the accuracy of binding pose predictions by docking by two orders of magnitude” Journal of Biomolecular NMR 2012 52, 23-30
J. Orts, S. Bartoschek, C. Griesinger, P. Monecke, T. Carlomagno
-
“The description of protein internal motions aids selection of ligand binding poses by the INPHARMA method” Journal of Biomolecular NMR 2012 54, 245-256
B. Stauch, J. Orts, T. Carlomagno
-
“Accounting for Conformational Variability in Protein–Ligand Docking with NMR-Guided Rescoring” Journal of the American Chemical Society 2013 135, 5819-5827
L. Skjærven, L. Codutti, A. Angelini, M. Grimaldi, D. Latek, P. Monecke, M. Dreyer, T. Carlomagno
-
“Identification of new hit scaffolds by INPHARMA guided virtual screening” Medicinal Chemistry Communications, 2015 6, 1501-1507
J. Sikorska, L. Codutti, L. Skjærven, B. Elshorst, R. Saez-Ameneiro, A. Angelini, P. Monecke, T. Carlomagno
