On the mechanism of reaction of radicals with tirapazamine.

Ketyl radicals produced by photolysis of ketones or di-tert-butyl peroxide (DTBP) in alcohol solvents react rapidly with tirapazamine (TPZ). The acetone ketyl radical (ACOH) reacts with TPZ with an absolute second-order rate constant of (9.7 +/- 0.4) x 108 M-1 s-1. The reaction kinetics can be followed by monitoring the bleaching of TPZ absorption at 475 nm or the formation of a reaction product which absorbs at 320 and 410 nm. The ACOD radical reacts with TPZ in 2-propanol-OD with an absolute rate constant of (6.7 +/- 0.5) x 108 M-1 s-1, corresponding to a kinetic isotope effect (KIE) of 1.4. Deuteration of the radical on carbon (ACOH-d6) retards the reaction of the radical with TPZ even further (absolute rate constant = (4.8 +/- 0.04) x 108 M-1 s-1). This result corresponds to a KIE of 2.0. Radicals derived from dioxane and diisopropyl ether by flash photolysis of DTBP in ethereal solvent react with TPZ more slowly than do ketyl radicals. It is concluded that ketyl radicals react, in part, with TPZ in organic solvents by transfer of a hydrogen atom from the OH and CH3 groups of the ketyl radical to the oxygen atom at the N4 position of TPZ to form acetone or acetone enol and a radical derivative of TPZ (TPZH). The latter species absorbs at 320 and 405 nm, has a lifetime of hundreds of microseconds in alcohol solvents, and decays by disproportionation to form TPZ and a reduced heterocycle. The reduced heterocycle eventually forms a desoxytirapazamine by a polar mechanism. The results are supported by density functional theory calculations. It is proposed that dioxanyl radical will also react, in part, with TPZ by transfer of a hydrogen atom from the carbon adjacent to the radical center to the oxygen atom at the N4 position of TPZ. This produces the enol ether and the previously mentioned TPZH radical. It is further posited that ether radicals react a bit more slowly than ketyl radicals because they lack the second mode of hydrogen transfer (from the OH group) that is present in the ACOH radical. Our data are permissive of the possibility that ether radicals add to TPZ at a rate that is competitive with beta-hydrogen atom transfer.