Role of oxygen free radicals in cocaine-induced vascular disruption in mice.

To test the hypothesis that cocaine-induced embryonic vascular disruption is mediated by oxygen free radicals, the antioxidants 2-oxothiazolidine-4-carboxylate (OTC) and alpha-phenyl-N-t-butyl nitrone (PBN) were employed. When cocaine (78 mg/kg) was administered on day 8 of gestation to ICR mice and embryos evaluated on day 10 (in vivo), 62.3% of cocaine-treated embryos showed increased vasodilation compared to 4.9% for controls, and 33.1% of the cocaine-exposed embryos showed marked hemorrhage compared to 3.3% for controls. In addition, cocaine increased the incidence of neural defects, in the form of open neural tube, hypoplastic prosencephalon, and microcephaly. Administration of OTC (0.25 and 0.5 mmol/kg) or PBN (300 mg/kg) prior to cocaine significantly reduced cocaine-induced vasodilation and hemorrhage, while not preventing neural defects. When cocaine (78 mg/kg) was administered in vivo on day 8 of gestation and embryos were dissected 15 min later and subsequently cultured for 48 hr in the absence of cocaine (in vivo-in vitro), marked vascular disruption was observed: normal yolk circulation/heartbeat was decreased to 26.6%, while edema/blisters and vasodilation/hemorrhage were increased to 45.6% and 59.6%, respectively. Administration of PBN (300 mg/kg) prior to cocaine completely prevented cocaine-induced vascular disruption. When cocaine was administered in vivo and PBN (300 micrograms/ml) was incubated with cultured embryos in vitro, the antioxidant only partially prevented cocaine-induced cardiovascular defects in this model. Neural defects produced by cocaine were not significantly affected by PBN, administered either in vivo or in vitro. Cocaine (78 mg/kg) administered in vivo stimulated lipid peroxidation maximally after 3 hr in both day 8 and day 9 embryos. When cocaine was incubated in vitro during embryo culture at 33 micrograms/ml, a concentration that produces nonspecific inhibition of growth and development, embryonic lipid peroxidation on day 9 was not affected. Finally, when PBN (300 mg/kg) was administered prior to cocaine (78 mg/kg) on day 8 of gestation, stimulation of lipid peroxidation by cocaine was prevented. These results suggest that cocaine-induced vascular disruption in early development is mediated by maternal production of oxygen free radicals.