Search

The kinetic basis for the in vitro hypoxia-selective cytotoxicity (HSC) of the Co(III)-nitrogen mustard complex SN 24771 (NSC 675352) has been investigated using pulse radiolysis. The rate constants for the one-electron reduction of SN 24771 by model reductants exhibited a marked dependence on the one-electron reduction potential of the reductant, with values up to several orders of magnitude slower than for the nitroimidazole drug misonidazole. Following one-electron reduction to form the Co(II) complex (species I) consecutive conversion to further transient species (II and III) occurs with first order rate constants of 120 +/- 10 s-1 and 10 +/- 2 s-1 and are associated with release of ligands. Neither of these subsequent processes are inhibited by the addition of O2 up to a concentration of 0.5 mmol l-1 suggesting that the HSC action of SN 24771 most likely arises from a mechanism other than simple redox cycling between the Co(III) and Co(II) forms by O2. If the measured low rate constants of one-electron reduction by model reductants of SN 24771 (as compared to the reduction of nitroaromatics), is mirrored by biological reductants, then it is proposed that HSC may occur through competition between SN 24771 and O2 for these reductants.

Efficient extravascular diffusion is a critical requirement for hypoxic cell radiosensitisers, bioreductive drugs and hypoxic cell markers that must reach cells distant from functional blood vessels in tumours. The diffusion of simple nitroimidazoles with neutral (misonidazole, miso) and basic (pimonidazole, pimo) side chains, as well as 2-nitroimidazoles with acridine (NLA-1) and phenanthridine (2-NLP-3) DNA intercalating moieties was investigated using multicellular membranes (MMs), a new in vitro model for the extravascular compartment of tumours. The diffusion of miso through V79 MMs was concentration independent over the range 0.1-10 mM. Mathematical modelling of the flux kinetics provided a diffusion coefficient in MMs (DMM) of 5.5 x 10(-7) cm2 s-1 which was approximately 13-fold lower than in culture medium. Flux was little affected by the extent of hypoxia in MMs, indicating that hypoxic metabolism does not compromise diffusion of miso over distances in the order of 200 microns. The DMM for pimo was similar to miso, while those for 2-NLP-3 and NLA-1 were both lower. The results demonstrate compromised extravascular diffusion for DNA-intercalating nitroimidazoles, but indicate that this problem is more severe for the basic acridine derivative, NLA-1, than for the phenanthridine, 2-NLP-3. The MM model appears to be well suited to quantitative determination of drug diffusion in a multicellular environment.