Conformational analysis guided by NMR: an analytical strategy for structure elucidation of pharmaceutical compounds

Nuclear Magnetic Resonance (NMR) spectroscopy is one of the most powerful techniques for structure elucidation and characterization. Moreover, NMR, coupled with computational chemistry, can be useful for the conformational analysis of free ligands in solution to investigate target-ligand interactions within Structure-Based Drug Design (SBDD) approach. Structure-based drug design, using X-ray crystallography, computational molecular modelling, molecular docking and NMR methods, is a powerful and straightforward approach in drug design process. Conformational analysis by NMR is based upon the idea of ligand conformational selection, in which the free ligand in solution can be fully pre-organized into the bioactive conformation. Thus, the free ligand can adopt multiple three-dimensional conformations due to rotatable bonds, including the one adopted in the binding pocket of the ligand-target complex. This method includes a detailed analysis of multiple NMR constraints, comprising chemical shifts, through-space distances between pairs of atoms derived from rotating-frame Overhauser spectroscopy (ROESY) or quantitative nuclear Overhauser effect (qNOE), torsion angles derived from J-couplings and dynamic average of orientation of internuclear vectors derived from residual dipolar couplings (RDCs). In addition, quantum mechanical (QM) calculations to predict some NMR parameters, such as isotropic shielding constants, within a reasonable timeframe can be useful. Chemical shift predictions, which are conformation-dependent, provide important information for the conformational analysis as far as some diagnostic protons are detected. Calculated chemical shifts can also be used to confirm the proposed conformational ensemble derived from NMR-based analysis. The objective of this project is to set up and validate a protocol with high accuracy and precision, which exploits NMR constraints to obtain structural information from free ligands in solution, combining experimental information with computational chemistry simulations. For this purpose, three active macrocyclic compounds, i.e. Lorlatinib, Pacritinib and Simeprevir were tested. Several approaches such as single or multi-parametric fitting, different solvents and various alignment media were investigated.