A CE kinetic assay was developed to study the stability of the adducts of a novel ruthenium(III)-based anticancer agent with serum proteins under simulated reductive physiological conditions. Formation of the reactive Ru(II) species and their release from the serum proteins are thought to play an important role in the mode-of-action of indazolium. trans-[tetrachlorobis(lH-indazole)ruthenate(III)] (KP1019) which has successfully finished a clinical phase I study. The CE method was adapted, in zone electrophoresis and affinity CE modes, to make obvious that such transformation would take place in the hypoxic tumor tissue rather than in the bloodstream. Indeed, no measurable effect of extracellular concentration levels of glutathione incorporated into the BGE on the UV signals of albumin and transferrin adducts was observed over 30 min of examination. Incubation of the KP1019-albumin adduct with the major blood reducing agent, ascorbic acid, revealed no changes in the continuously monitored peak areas (average corrected responses were 9.56 +/- 0.86 and 9.87 +/- 0.60 mAU for the adduct and its mixtures with ascorbic acid in the physiological range of 1 x 10(-5)- 8 x 10(-5) M, respectively). On the other hand, both the transferrin adduct and transferrin itself accelerated the oxidation of ascorbic acid; however, the oxidation rate constants measured by CE were virtually the same: (19.1 +/- 4.4) x 10(-3) and (18.2 +/- 5.0) x 10(-3) min(-1), respectively. In order to confirm more unambiguously the stability of KP1019-protein adducts in the presence of ascorbic acid (UV absorbance detection does not distinguish the adduct and protein signals), CE with inductively coupled plasma (ICP) MS detection was applied to follow metal-selectively the signal of bound ruthenium, which remained unaffected by this reducing agent. This work appears the first to present the application of CE to the stability studies of the protein-bound metallodrugs.