Efavirenz primary and secondary metabolism in vitro and in vivo: identification of novel metabolic pathways and cytochrome P450 2A6 as the principal catalyst of efavirenz 7-hydroxylation

Abstract

Efavirenz primary and secondary metabolism was investigated in vitro and in vivo. In human liver microsome (HLM) samples, 7- and 8-hydroxyefavirenz accounted for 22.5 and 77.5% of the overall efavirenz metabolism, respectively. Kinetic, inhibition, and correlation analyses in HLM samples and experiments in expressed cytochrome P450 show that CYP2A6 is the principal catalyst of efavirenz 7-hydroxylation. Although CYP2B6 was the main enzyme catalyzing efavirenz 8-hydroxylation, CYP2A6 also seems to contribute. Both 7- and 8-hydroxyefavirenz were further oxidized to novel dihydroxylated metabolite(s) primarily by CYP2B6. These dihydroxylated metabolite(s) were not the same as 8,14-dihydroxyefavirenz, a metabolite that has been suggested to be directly formed via 14-hydroxylation of 8-hydroxyefavirenz, because 8,14-dihydroxyefavirenz was not detected in vitro when efavirenz, 7-, or 8-hydroxyefavirenz were used as substrates. Efavirenz and its primary and secondary metabolites that were identified in vitro were quantified in plasma samples obtained from subjects taking a single 600-mg oral dose of efavirenz. 8,14-Dihydroxyefavirenz was detected and quantified in these plasma samples, suggesting that the glucuronide or the sulfate of 8-hydroxyefavirenz might undergo 14-hydroxylation in vivo. In conclusion, efavirenz metabolism is complex, involving unique and novel secondary metabolism. Although efavirenz 8-hydroxylation by CYP2B6 remains the major clearance mechanism of efavirenz, CYP2A6-mediated 7-hydroxylation (and to some extent 8-hydroxylation) may also contribute. Efavirenz may be a valuable dual phenotyping tool to study CYP2B6 and CYP2A6, and this should be further tested in vivo.

Fig. 1.

MRM trace chromatograms of efavirenz (EFV) in human liver microsomal incubates. Efavirenz (20 μM) was incubated with HLM samples and cofactors. Subsequent sample processing and LC/MS/MS conditions were as described under Materials and Methods. Metabolites that were consistent with monohydroxylated (7- and 8-hydroxyEFV) and secondary metabolite (dihydroxylated EFV) were identified in the microsomal incubates. Nevirapine was used as an internal standard.

Fig. 2.

Representative kinetics for the metabolism of efavirenz (EFV) to 7-hydroxyEFV and 8-hydroxyEFV in human liver samples (IU42). Increasing concentrations of EFV (5–150 μM) were incubated with HLM samples (0.25 mg/ml) and cofactors for 10 min at 37°C. Formation rates of metabolites (pmol/min/mg protein) versus EFV concentration (μM) were best fit to a one-site hyperbolic Michaelis-Menten equation (A) (see Data Analysis). The corresponding Eadie-Hofstee plots are shown (B). Each point represents the average of duplicate incubations. Kinetic parameters derived from these and another six HLM samples are listed in Table 1.

Fig. 3.

Efavirenz (EFV) metabolism in a panel of 15 characterized HLMs. EFV (10 μM) was incubated with microsomes from different human livers (0.25 mg/ml) and cofactors for 15 min at 37°C. Formation rates (pmol product/min/mg protein) of 7-hydroxyEFV (7-OHEFV) (A) and of 8-hydroxyEFV (8-OHEFV) (B) are shown. Rates represent average of duplicate incubation measurements. Correlations between formation rates of 7-OHEFV or 8-OHEFV and the activity of P450 enzymes are illustrated in Table 2.

Fig. 4.

Efavirenz (EFV) metabolism by a panel of P450 enzymes. EFV (10 or 100 μM) was incubated with expressed P450s (13 or 26 pmol of P450) and cofactors for 10 min at 37°C. Rates (pmol product/min/pmol P450) of 7-hydroxyEFV (7-OHEFV) (A) and 8-hydroxyEFV (8-OHEFV) (B) are presented. Each point is an average of duplicate incubation measurements.

Fig. 5.

Inhibition of efavirenz (EFV) metabolism in HLM samples. EFV (10 μM) was incubated with HLMs (0.25 mg/ml) and cofactors for 10 min at 37°C in the absence (control) and presence of P450 isoform-specific inhibitor (see Materials and Methods for details). Data (percentage activity remaining after inhibition relative to the vehicle control) represent mean ± S.D. of three duplicate measurements or three different HLMs. TICL, ticlopidine; QUER, quercetin; TAO, troleandomycin; KETO, ketoconazole; PILO, pilocarpine; DEDTC, diethyldithiocarbamate; SULF, sulfaphenazole; FURAF, furafylline; LTR, letrozole; and QUIN, quinidine. The final concentrations of these inhibitors used are indicated in brackets. A preincubation protocol was used for FURAF, TAO, thioTEPA, and DEDTC (see Materials and Methods).

Fig. 6.

MRM trace chromatograms of 7-hydroxyefavirenz (7-OHEFV) and 8-hydroxyefavirenz (8-OHEFV) in human liver microsomal incubates. Each substrate (5 μM) was incubated in HLM samples and cofactors for 10 min at 37°C. A metabolite peak that was consistent with dihydroxylated efavirenz was formed in microsomal incubations of 7-OHEFV (A) and 8-OHEFV (B). Synthetic 8,14-dihydroxyefavirenz (8,14-diOHEFV) standard was directly injected (C).

Fig. 7.

Secondary metabolism of efavirenz (EFV) by a panel of expressed P450s. Sequential metabolism was tested using 7-hydroxyEFV (7-OHEFV) and 8-hydroxyEFV (8-OHEFV) as substrates. 7-OHEFV (5 μM) and 8-OHEFV (5 μM) were incubated with expressed P450s (13 or 26 pmol of P450) and cofactors for 10 min at 37°C, and the formation of a dihydroxylated efavirenz was monitored (A). Kinetics for the metabolism of 7-OHEFV and 8-OHEFV to dihydroxylated efavirenz was determined by incubating each substrate (1–150 μM) with 13 pmol of CYP2B6 and cofactors for 10 min at 37°C (B). Kinetic analysis was performed by fitting to a one-site hyperbolic Michaelis-Menten equation (7-OHEFV metabolism) or using substrate inhibition equation (8-OHEFV) (see Data Analysis). Each point represents average picomole of product per minute per picomole of CYP2B6 of duplicate incubation measurements.

Fig. 8.

Representative MRM trace chromatograms of efavirenz (EFV) and its monohydroxylated (7- and 8-hydroxyefavirenz) and secondary metabolites (8,14-dihydroxyefavirenz and dihydroxyefavirenz) in plasma samples obtained from subjects administered a single 600-mg oral dose of efavirenz. Extracted blank samples spiked with authentic standards of efavirenz and its metabolites (A), and extracted plasma sample obtained 3 h after administration of a single 600-mg oral dose of efavirenz to healthy volunteers (B). Plasma samples were treated with β-glucuronidase, and the peaks represent total (free + conjugated).

Fig. 9.

Pharmacokinetics of efavirenz (EFV) and its primary and secondary metabolites in healthy volunteer subjects (n = 5) administered a single 600-mg oral dose of EFV. Plasma concentrations versus time curves (A) and area under the concentration-time curve (AUC_0–72) (B) of efavirenz and metabolites are shown. 8-OHEFV, 8-hydroxyEFV; 7-OHEFV, 7-hydroxyEFV; 8,14-diOHEFV, 8,14-dihydroxyefavirenz; diOHEFV, dihydroxylated EFV.