Amino Acid Radicals in Rhenium-Modified Copper Proteins

Citation

Wehbi, William Amine (2003) Amino Acid Radicals in Rhenium-Modified Copper Proteins. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/HF8D-1W16. https://resolver.caltech.edu/CaltechETD:etd-05302003-153624

Abstract

Oxidative flash quench of [Re(CO)3(phen)(His)]⁺ generates a high-potential [Re]²⁺ oxidant (E° (Re^2+/+) = 2 eV v. NHE), which has been used to obtain rates of electron transfer of Cu(I) oxidation in rhenium-modified azurins. These rates are enhanced over the [Ru(bpy)₂(im)(His)]²⁺ analogues (E° Ru3^+/2+ ~ 1 eV), suggesting an alternate mechanism from driving force optimized, singe-step electron tunneling. To test whether other intermediates can be involved, oxidative freeze flash quench of the zinc(II) derivatives were undertaken. These experiments reveal that [Re]²⁺ can produce the amino acid radicals of tyrosine and cysteine, as detected by EPR. The properties of these radicals in structurally well-defined protein microenvironments in Pseudomonas aeruginosa azurin mutants have focused, in particular, on the g₁ component of the g-tensor, which is sensitive to the strength of the hydrogen bond to the radical. The g₁ for Tyr48 radical, which resides in a completely hydrophobic pocket and is inaccessible to solvent, is found to be greater than the g1 for the solvent exposed Tyr108 radical. This comparison could not be made for the cysteine radicals as Cys108 formed a sulfenyl radical upon oxidation; the Cys48 radical has been demonstrated to be a thiyl radical species and provides the EPR spectroscopic benchmark for a non-hydrogen bonded thiyl radical.

In azurin mutants without any tyrosine, tryptophan, or cysteine residues, oxidative flash quench results in another organic based radical. This radical is located on the histidine imidazole ring that is coordinated to the rhenium atom. DFT calculations suggest that the spin density resides mainly on the imidazole ring when it is deprotonated. Corrected distances in the tunneling timetable to the imidazole ring from the copper atom predict an identical exponential decay in the electron transfer rates as for the ruthenium-labeled azurins. The rate enhancement is explained in terms of a "trivial hop" whereby Re²⁺ rapidly oxidizes the non-innocent histidine ligand in a proton dependent process; the histidine radical in turn oxidizes the copper atom or tyrosine, cysteine, or tryptophan when zinc is present. This model explains all of the enhanced Cu(I) oxidation rates by [Re]²⁺ and suggests that for Cu(I) oxidation in azurin, multistep electron tunneling through other amino acid radicals does not occur and that the observed radicals are generated in off-path processes.

Thesis Files