Date of Award
Doctor of Philosophy (PhD)
S-Nitrosation of cysteine (Cys) residues, a covalent modification of its S atom by NO group, is a major post-translational modification of proteins. Despite the importance of S-nitrosoproteins in numerous physiological processes, lability of the S-nitrosothiol (-SNO) group hinders the research progress. In this work, computational chemistry methods were applied to S-nitrosated cysteine (CysNO) models to gain a deeper insight into its structure and properties. First, we obtained the most accurate at the moment computational estimation of the molecular structure and properties of CH3SNO model molecule using Feller-Peterson-Dixon (FPD) ab initio protocol. The S–N bond length in cis- CH3SNO is calculated as 1.814 Å, and its dissociation energy (BDE) is 32.4 kcal/mol. We found that although the vibrational frequency of the S–N stretch is unusually low for a covalent bond (398 cm-1), the S–N bond has a remarkably harmonic character. After the benchmarking of the density functional theory (DFT) methods against the FPD reference, we recommend MPW2PLYP and MPW2PLYPD double hybrid functionals for calculation of the geometric properties, vibrational frequencies and isomerization barriers of S-nitrosothiols, and PBE0 (PBE0-GD3) hybrid functional for the S–N BDEs. Further, we evaluated the influence of charged amino acid residues and steric constraints on the conformational dynamics of CysNO, using an a-helix model and a DJ-1 (PARK7) protein using hybrid quantum mechanics/molecular mechanics (QM/MM) approach. We found that while the rotational barrier around the S–N bond is ca. 13 kcal/mol in free CysNO, it can vary between 10 and 24 kcal/mol in the protein environment due to the effect of neighboring charged amino acids. Finally, to address a long-standing problem of the CysNO identification in proteins, we computationally designed a novel CysNO labeling reaction, (3+2) dipolar cycloaddition between the -SNO group activated by N-coordinated Lewis acid and a strain-activated alkene. We show that the most effective labeling reagent should covalently link both crucial reaction components—the Lewis acid and the dipolarophile—into one molecule, to lower the entropic penalty and corresponding reaction barrier.