Phenytoin-initiated hydroxyl radical formation: characterization by enhanced salicylate hydroxylation

Bioactivation of phenytoin and related teratogens by peroxidases such as prostaglandin H synthase (PHS) may initiate hydroxyl radical (.OH) formation that is teratogenic. Salicylate is hydroxylated by .OH at the third and fifth carbon atoms, forming 2,3- and 2,5-dihydroxybenzoic acids (DHBA). In vivo salicylate metabolism produces only the 2,5-isomer, so 2,3-DHBA formation may reflect .OH production. In the present study, we validated the salicylate assay using the known .OH generator paraquat and evaluated .OH production by phenytoin. Female CD-1 mice were treated with paraquat (30 mg/kg, intraperitoneally) given 30 min after acetylsalicylic acid (ASA) (200 mg/kg, intraperitoneally). Blood was collected at 5, 15, and 30 min and 1 and 2 hr after paraquat, and plasma was analyzed for DHBA isomers and glucuronide conjugates by high performance liquid chromatography with electrochemical detection. Paraquat increased 2,3-DHBA formation 19.2-fold, with substantial inter-individual variability in the time of maximal formation (p = 0.0001). The 2,3-DHBA glucuronide conjugates in vivo and in hepatic microsomal studies amounted to approximately 11% and 0.43%, respectively, of total 2,3-DHBA equivalents. To investigate putative .OH production initiated via PHS-catalyzed phenytoin bioactivation, ASA was given 30 min before phenytoin (65 or 100 mg/kg, intraperitoneally), resulting in respective 7.6-fold (p = 0.02) and 14.2-fold (p = 0.003) increases in phenytoin-initiated maximal 2,3-DHBA formation. Maximal 2,3-DHBA formation was 2.1-fold higher when ASA was administered after rather than before the same dose (65 mg/kg) of phenytoin (p = 0.03), indicating ASA inhibition of PHS-catalyzed phenytoin bioactivation. Urinary analysis was much less sensitive, and the 2,5-isomer reflected enzymatic rather than .OH-mediated hydroxylation. The paraquat studies demonstrate the importance of timing in accurately quantifying 2,3-DHBA formation and suggest that glucuronidation does not interfere. The substantial, dose-dependent initiation of 2,3-DHBA formation by phenytoin, and its inhibition by ASA, provide the first in vivo evidence that PHS-dependent .OH formation could contribute to the molecular mechanism of phenytoin teratogenesis.