Cellular defense mechanisms that respond to damage from oxidative and electrophilic stress, such as from quinones, represent a target for chemopreventive agents. Drugs bioactivated to quinones have the potential to activate antioxidant/electrophile responsive element (ARE) transcription of genes for cytoprotective phase 2 enzymes such as NAD(P)H-dependent quinone oxidoreductase (NQO1) but can also cause cellular damage. Two isomeric families of compounds were prepared, including the NO-NSAIDs (NO-donating nonsteroidal anti-inflammatory drugs) NCX 4040 and NCX 4016; one family was postulated to release a quinone methide on esterase bioactivation. The study of reactivity and GSH conjugation in model and cell systems confirmed the postulate. The quinone-forming family, including NCX 4040 and conisogenic bromides and mesylate, was rapidly bioactivated to a quinone, which gave activation of ARE and consequent induction of NQO1 in liver cells. Although the control family, including NCX 4016 and conisogenic bromides and mesylates, cannot form a quinone, ARE activation and NQO1 induction were observed, compatible with slower SN2 reactions with thiol sensor proteins, and consequent ARE-luciferase and NQO1 induction. Using a Chemoprevention Index estimate, the quinone-forming compounds suffered because of high cytoxicity and were more compatible with cancer therapy than chemoprevention. In the Comet assay, NCX 4040 was highly genotoxic relative to NCX 4016. There was no evidence that NO contributes to the observed biological activity and no evidence that NCX 4040 is an NO donor, instead, rapidly releasing NO3- and quinone. These results indicate a strategy for studying the quinone biological activity and reinforce the therapeutic attributes of NO-ASA through structural elements other than NO and ASA.