Predicting the redox state and secondary structure of cysteine residues in proteins using NMR chemical shifts

Abstract

We report 2D cluster analyses of 1Halpha, 1HN, 13Calpha, and 13C' versus 13Cbeta NMR chemical shifts (CSs) that can be used to predict the redox state and secondary structure of cysteine residues in proteins. A database of cysteine 1Halpha, 1Hbeta2, 1Hbeta3, 1HN, 13Calpha, 13Cbeta, 13C', and 15NH CSs as a function of secondary structure and redox state was constructed from BioMagResBank entries. One-dimensional statistical analysis showed that cysteine 1Halpha, 1HN, 13Calpha, 13C', and 15NH CSs reflected the secondary structure, and that cysteine Cbeta CS is extremely sensitive to the redox state. In contrast, cysteine 1Hbeta CS was not correlated with its redox state or secondary structure. Two-dimensional cluster analysis revealed that 2D Calpha/Cbeta, C'/Cbeta, HNu/Cbeta, and Halpha/Cbeta clusters were helpful in distinguishing both the redox state and secondary structure of cysteine residues. Based on these results, we derived rules using a score matrix to predict the redox state and secondary structure of cysteines using their CSs. The score matrix predicts the redox state and secondary structure of cysteine residues in proteins with approximately 90% accuracy. This suggests that the redox state and secondary structure of cysteine residues in peptides and proteins can be obtained from their CSs without recourse to nuclear Overhauser effect measurements.