Quantum chemical 13Ca chemical shift calculations for protein NMR structure determination, refinement, and validation

Abstract

A recently determined set of 20 NMR-derived conformations of a 48-residue all-α-helical protein, (PDB ID code 2JVD), is validated here by comparing the observed 13Cα chemical shifts with those computed at the density functional level of theory. In addition, a recently introduced physics-based method, aimed at determining protein structures by using NOE-derived distance constraints together with observed and computed 13Cα chemical shifts, was applied to determine a new set of 10 conformations, (Set-bt), as a blind test for the same protein. A cross-validation of these two sets of conformations in terms of the agreement between computed and observed 13Cα chemical shifts, several stereochemical quality factors, and some NMR quality assessment scores reveals the good quality of both sets of structures. We also carried out an analysis of the agreement between the observed and computed 13Cα chemical shifts for a slightly longer construct of the protein solved by x-ray crystallography at 2.0-Å resolution (PDB ID code 3BHP) with an identical amino acid residue sequence to the 2JVD structure for the first 46 residues. Our results reveal that both of the NMR-derived sets, namely 2JVD and Set-bt, are somewhat better representations of the observed 13Cα chemical shifts in solution than the 3BHP crystal structure. In addition, the 13Cα-based validation analysis appears to be more sensitive to subtle structural differences across the three sets of structures than any other NMR quality-assessment scores used here, and, although it is computationally intensive, this analysis has potential value as a standard procedure to determine, refine, and validate protein structures.