The solution structure of the Ga(III)-bleomycin A2 complex resolved by NMR and molecular modeling; interaction with d(CCAGGCCTGG)

Direct Hydrogen-Atom Abstraction by Activated Bleomycin: An Experimental and Computational Study

Bleomycin (BLM), a glycopeptide antibiotic chemotherapy agent, is capable of single- and double-strand DNA damage. Activated bleomycin (ABLM), a low-spin Fe(III)-OOH complex, is the last intermediate detected prior to DNA cleavage following hydrogen-atom abstraction from the C-4' of a deoxyribose sugar moiety. The mechanism of this C-H bond cleavage reaction and the nature of the active oxidizing species are still open issues. We have used kinetic measurements in combination with density functional calculations to study the reactivity of ABLM and the mechanism of the initial attack on DNA. Circular dichroism spectroscopy was used to directly monitor the kinetics of the ABLM reaction. These experiments yield a deuterium isotope effect, kH/kD approximately 3 for ABLM decay, indicating the involvement of a hydrogen atom in the rate-determining step. H-atom donors with relatively weak X-H bonds accelerate the reaction rate, establishing that ABLM is capable of hydrogen-atom abstraction. Density functional calculations were used to evaluate the two-dimensional potential energy surface for the direct hydrogen-atom abstraction reaction of the deoxyribose 4'-H by ABLM. The calculations confirm that ABLM is thermodynamically and kinetically competent for H-atom abstraction. The activation and reaction energies for this pathway are favored over both homolytic and heterolytic O-O bond cleavage. Direct H-atom abstraction by ABLM would generate a reactive Fe(IV)=O species, which would be capable of a second DNA strand cleavage, as observed in vivo. This study provides experimental and theoretical evidence for direct H-atom abstraction by ABLM and proposes an attractive mechanism for the role of ABLM in double-strand cleavage.

Structural study of copper(I)-bleomycin

Previous NMR studies on Cu(I)-bleomycin have suggested that this adduct has a geometry distinct from Fe(II)BLM. The coordination chemistry of this bleomycin derivative has been investigated through the extension of the NMR data reported previously, and the use of molecular dynamics calculations. The data collected from the NMR experiments support the coordination to the metal center of the primary and secondary amines in beta-aminoalanine and the pyrimidine ring. The detection in the NMR spectra of the signal derived from the amide hydrogen in beta-hydroxyhistidine indicates that this amide is protonated in Cu(I)-bleomycin, precluding participation of the pyrimidinyl carboxamide nitrogen in the coordination of Cu(I), as previously reported. Three-dimensional solution structures compatible with the NMR data have been assayed for Cu(I)-bleomycin for the first time by way of molecular dynamics calculations, and two models showing four and five coordination have been found to be those that better fit the experimental data. In both models the primary amine in beta-aminoalanine is coordinated such that it is located on the same side, with respect to the coordination cage, as the peptide linker fragment. This result seems important for the favored models to be compatible with either their possible oxidation to become one of the reported structures for Cu(II)BLM, or their transformation into Fe(II) adducts able to cause DNA damage.