A Comparison of Potential Azide Antidotes in a Mouse Model

Three cobalt-containing macrocyclic compounds previously shown to antagonize cyanide toxicity have been comparatively evaluated for the amelioration of sublethal azide toxicity in juvenile (7–8 weeks) Swiss-Webster mice. The lowest effective doses were determined for hydroxocobalamin, a cobalt porphyrin, and a cobalt-Schiff base macrocycle by giving the antidotes 5 min prior to the toxicant, 27 mg (415 μmol) /kg sodium azide. Both male and female mice were evaluated for their response to the toxicant as well as the antidotes, and no significant differences were noted once weight differences were taken into account. Two of the three compounds significantly decreased the recovery time of azide-intoxicated mice at 10 min after the administration of sodium azide, as determined by a behavioral test (pole climbing). Additionally, azide was determined to cause a several degree drop (∼3 °C) in measured tail temperature, and warming the mice led to a more rapid recovery. The mice were also shown to recover more rapidly when given sodium nitrite, 24 mg (350 μmol)/kg, 5 min after the toxicant; this treatment also suppressed the azide-induced tail temperature decrease. Electron paramagnetic resonance (EPR) measurements of mouse blood treated with sodium azide demonstrated the presence of nitrosylhemoglobin at levels of 10–20 μM which persisted for ∼300 min. The presence of the methemoglobin azide adduct was also detected by EPR at a maximum level of ∼300 μM, but these signals disappeared around 200 min after the administration of azide. The treatment of mice with 15 N sodium azide proved that the nitrosylhemoglobin was a product of the administered azide by the appearance of a two-line hyperfine (due to the 15 N) in the EPR spectrum of mouse blood.