Sixteen alkyl-substituted dispiro-1,2,4,5-tetraoxanes (7,8,15, 16-tetraoxadispiro[5.2.5.2]hexadecanes) were synthesized to explore dispiro-1,2,4,5-tetraoxane SAR and to identify tetraoxanes with better oral antimalarial activity than prototype tetraoxane 1 (WR 148999). The tetraoxanes were prepared either by peroxidation of the corresponding cyclohexanone derivatives in H(2)SO(4)/CH(3)CN or by ozonolysis of the corresponding cyclohexanone methyl oximes. Those tetraoxanes with alkyl substituents at the 1 and 10 positions were formed as single stereoisomers, whereas the five tetraoxanes formed without the stereochemical control provided by alkyl groups at the 1 and 10 positions were isolated as mixtures of diastereomers. Three of the sixteen tetraoxanes were inactive (IC(50)'s > 1000 nM), but five (2, 6, 10, 11, 12) had IC(50)'s between 10 and 30 nM against the chloroquine-sensitive D6 and chloroquine-resistant W2 clones of Plasmodium falciparum compared to corresponding IC(50)'s of 55 and 32 nM for 1 and 8.4 and 7.3 nM for artemisinin. We suggest that tetraoxanes 13, 16, and 17 were inactive and tetraoxanes 4 and 7 were weakly active due to steric effects preventing or hindering peroxide bond access to parasite heme. Tetraoxanes 1, 10, 11, and 14, along with artemisinin and arteether as controls, were administered po b.i.d. (128 mg/kg/day) to P. berghei-infected mice on days 3, 4, and 5 post-infection. At this dose, tetraoxanes 10, 11, and 14 cured between 40% and 60% of the infected animals. In comparison, artemisinin and tetraoxane 1 produced no cures, whereas arteether cured 100% of the infected animals. There was no apparent relationship between tetraoxane structure and in vitro neurotoxicity, nor was there any correlation between antimalarial activity and neurotoxicity for these seventeen tetraoxanes.
