P178 THE BENZYLIC ALCOHOL METABOLITE OF TRIMETHOPRIM COVALENTLY BINDS TO PROTEINS IN VITRO

Jennifer L Goldman , Division of Clinical Pharmacology, Toxicology & Therapeutic Innovation, Children's Mercy Hospital, Kansas City, MO
J. Steven Leeder , Division of Clinical Pharmacology, Toxicology & Therapeutic Innovation, Children's Mercy Hospital, Kansas City, MO
Leon van Haandel , Division of Clinical Pharmacology, Toxicology & Therapeutic Innovation, Children's Mercy Hospital, Kansas City, MO
Background: The combination antibiotic trimethoprim-sulfamethoxazole (TMP-SMX) is effective, inexpensive, and widely prescribed, yet it also causes a high rate of idiosyncratic adverse drug reactions (IADRs). These IADRS have been exclusively attributed to SMX in the past, but recent clinical and in vitro laboratory data suggests that TMP may be more than an innocent bystander. In-vitro data suggests that TMP undergoes CYP3A4 mediated bioactivation in the liver to an imminoquinone methide that is reactive towards thiols on the methylene bridge carbon and the Michael site on the pyrimidine ring. Although this mechanism for covalent binding is a plausible explanation for the liver toxicity observed with this drug, the short half live (millisecond range) of the reactive intermediate does not adequately explain the commonly observed skin reactions, since it is unlikely that the reactive intermediate will escape the liver and reach the skin within this time. Additionally, our previous study aimed at detecting detoxified mercapturic acid conjugates of TMP in the urine of children taking TMP-SMX found significant adduct formation on the methylene bridge of TMP but not on the pyrimidine ring. We also detected a strong correlation between the concentration of the benzylic alcohol metabolite of TMP (Cα-OH-TMP) and the formation of TMP adducts, leading to the formation of our hypothesis that Cα-OH-TMP, a circulating liver generated TMP metabolite, can undergo covalent binding with N-acetyl-cysteine (NAC) and human serum albumin (HSA) based on its intrinsic reactivity. Methods: To investigate the reactivity of Cα-OH-TMP towards a thiol nucleophile we incubated Cα-OH-TMP at 1.0 µM with a 10 fold excess of NAC for 3 to 14 days at 37°C in PBS. We analyzed the mixture every couple of days with a calibrated UPLC-MS/MS method for product formation on the methylene bridge (Cα-NAC-TMP) and on the pyrimidine ring (Pyr-NAC-TMP). To investigate the reactivity of Cα-OH-TMP towards a protein, we incubated 50mg/mL HSA with 10 mM (~14 fold excess) of Cα-OH-TMP. The protein was precipitated, washed, digested with trypsin, and analyzed for TMP-peptide adducts with high resolution mass spectrometry (HR). Results: Cα-OH-TMP is capable of covalently binding with NAC based on intrinsic reactivity at physiologically relevant concentrations. After two weeks ~2% of the Cα-OH-TMP pool had reacted with NAC. We estimate the rate to be higher in-vivo, since most of the NAC in this experiment was lost due to autoxidation. We only observed formation of the Cα-NAC-TMP product which is consistent with a SN2 nucleophilic displacement reaction. Incubation of the Cα-OH-TMP metabolite with HSA resulted in covalent modification of HSA on the Cys34 residue, which was detected by HR-MS/MS of the tryptic peptide. Conclusions: Cα-OH-TMP is a known circulating plasma metabolite of TMP generated in the liver. These data show that Cα-OH-TMP is a reactive secondary benzyl alcohol that undergoes covalent binding with cysteine residues in proteins, without the need for drug metabolizing enzymes. Since patients are exposed to TMP-SMX for days to weeks, the constant exposure to Cα-OH-TMP, may result in cumulative formation of protein adducts throughout the body.