Comparison of 2,2′-azobis(2-amidinopropane) hydrochloride (AAPH) and 2,2′-azobis(2,4-dimethylvaleronitrile)(AMVN) as free radical initiators: a spin-trapping study

Abstract

Spin trapping with 5,5-dimethyl-1-pyrroline 1-oxide (DMPO) and its hydrophobic analogue 2,2-dimethyl-4-phenyl-2H-imidazole 1-oxide (DMPIO) was used to identify and to monitor the concentration of participating radical species in oxidation reactions initiated by azo compounds: water-soluble 2,2′-azobis(2-amidinopropane) hydrochloride (AAPH) and lipophilic 2,2′-azobis(2,4-dimethylvaleronitrile)(AMVN). Incubation of AAPH with spin traps in aqueous media produced alkoxyl radical spin adducts with hyperfine splitting constants being a_N= 14.62 G, ^βa_H= 15.29 G, ^γa_H= 0.72 G and a_N= 13.46 G, ^βa_H= 12.53 G, for DMPO and DMPIO, respectively. In contrast, formation of AMVN-derived peroxyl radicals was detected by both direct EPR and spin trapping in DMSO (dimethylsulfoxide) solutions. In the presence of either rabbit skeletal sarcoplasmic reticulum (SR) membranes (10 g dm^–3 of SR protein) or egg phosphatidylcholine liposomes (10 g dm^–3 of lipid) preloaded with AMVN no spin adduct formation was observed, for both DMPO and DMPIO spin traps, indicating that AMVN-derived radical species do not escape the lipid environment. Only a small portion of AAPH-derived alkoxyl radicals was trapped by DMPIO in the presence of SR membranes. Spectral characteristics of the DMPIO spin adduct indicate its location at the lipid–water interface. At the same time, there was virtually no effect of SR on the rate of formation and steady-state level of the DMPO-spin adduct formed in the aqueous phase. From these data we suggest that the bulky cytosolic domains of the SR Ca²⁺-ATPase protect the membrane surface from radicals generated in the bulk (aqueous) solvent. Other evidence also demonstrates different mechanisms for free radical formation by AAPH and AMVN azo-initiators.