In vitro to in vivo extrapolation of the complex drug-drug interaction of bupropion and its metabolites with CYP2D6; simultaneous reversible inhibition and CYP2D6 downregulation

Abstract

Bupropion is a widely used antidepressant and smoking cessation aid and a strong inhibitor of CYP2D6 in vivo. Bupropion is administered as a racemic mixture of R- and S-bupropion and has stereoselective pharmacokinetics. Four primary metabolites of bupropion, threo- and erythro-hydrobupropion and R,R- and S,S-OH-bupropion, circulate at higher concentrations than the parent drug and are believed to contribute to the efficacy and side effects of bupropion as well as to the CYP2D6 inhibition. However, bupropion and its metabolites are only weak inhibitors of CYP2D6 in vitro, and the magnitude of the in vivo drug-drug interactions (DDI) caused by bupropion cannot be explained by the in vitro data even when CYP2D6 inhibition by the metabolites is accounted for. The aim of this study was to quantitatively explain the in vivo CYP2D6 DDI magnitude by in vitro DDI data. Bupropion and its metabolites were found to inhibit CYP2D6 stereoselectively with up to 10-fold difference in inhibition potency between enantiomers. However, the reversible inhibition or active uptake into hepatocytes did not explain the in vivo DDIs. In HepG2 cells and in plated human hepatocytes bupropion and its metabolites were found to significantly downregulate CYP2D6 mRNA in a concentration dependent manner. The in vivo DDI was quantitatively predicted by significant down-regulation of CYP2D6 mRNA and reversible inhibition of CYP2D6 by bupropion and its metabolites. This study is the first example of a clinical DDI resulting from CYP down-regulation and first demonstration of a CYP2D6 interaction resulting from transcriptional regulation.

Keywords: Bupropion; CYP2D6; Drug-drug interactions; Enzyme regulation; In vitro to in vivo extrapolation.

Figure 1. Stereoselective bupropion metabolism

R- and S-bupropion are metabolized to R,R-OH-, S,S-OH-bupropion, 4′-OH-bupropion, threohydrobupropion and erythrohydrobupropion. Threo- and erythrohydrobupropion are further metabolized to threo- and erythro-4′-OH-hydrobupropion.

Figure 2. Reversible CYP2D6 inhibition by bupropion and its metabolites in human liver microsomes

The percent dextrorphan formation remaining in comparison to control as a function of increasing concentrations of inhibitors is shown. The line shows the fit of equation 1, $\%\text{of}\text{control}\text{activity}=\text{Nonspecific}\text{Activity}+\frac{\text{Total}\text{Activity}-\text{Nonspecific}\text{Activity}}{1+{10}^{([\mathrm{I}]-log({\text{IC}}_{50}))}}$ to the data. Dextromethorphan was incubated in HLMs with (a) S-bupropion, (b) S,S-OH-bupropion, (c) erythrohydrobupropion, (d) R-bupropion, (e) R,R-OH-bupropion, and (f) threohydrobupropion. The incubations and data fitting was conducted as described in material and methods. All experiments were conducted in triplicate and the standard deviation is shown as the error bar. At some concentrations error bars are not visible as the standard deviation was smaller than the size of the symbol.

Figure 3. Reversible CYP2D6 inhibition by bupropion and its metabolites in human suspension hepatocytes

The percent of control dextrorphan formation remaining following incubation of dextromethorphan in the presence of multiple concentrations of (a) bupropion, (b) OH-bupropion, (c) erythrohydrobupropion, and (d) threohydrobupropion is shown. The incubations were conducted as described in material and methods. All experiments were conducted in triplicate and the standard deviation is shown as the error bar. The points show the experimental data. At some concentrations error bars are not visible as the standard deviation was smaller than the size of the symbol. The lines show the fit of equation 1, $\%\text{of}\text{control}\text{activity}=\text{Nonspecific}\text{Activity}+\frac{\text{Total}\text{Activity}-\text{Nonspecific}\text{Activity}}{1+{10}^{([\mathrm{I}]-log({\text{IC}}_{50}))}}$ to the data.

Figure 4. Effects of bupropion and its metabolites on CYP2D6 and SHP mRNA in HepG2 cells

CYP2D6 mRNA levels in HepG2 cells exposed to multiple concentrations of (A) erythrohydrobupropion, (B) threohydrobupropion, (C) R,R-OH-bupropion, (D) S,S-OH-bupropion and (E) bupropion are shown. The points show the measured effect with its standard deviation and the line shows the fit of equation 2, $E=E_0-\frac{E_{\mathit{max}}\times [I]}{{EC}_{50}+[I]}$ to the data. For R,R-OH-bupropion solubility limitations prevent measurement of dose-response relationship and hence the line shows the fit of equation 3, E = E₀ − slope × [I] to the data. Representative CYP2D6 protein expression blot image is shown in panel F and the quantification of the signal intensity in the western blots is shown in panel G. Panel H shows the fold change in SHP mRNA levels in HepG2 cells following 72 hours of treatment with bupropion (bup 0.5 μM, 2.5 μM, 5 μM), R,R-OH-bupropion (R,R-OH, 5 μM, 25 μM, 50 μM), S,S-OH-bupropion (S,S-OH 0.5 μM, 2.5 μM, 5 μM), erythrohydrobupropion (erythro 0.5 μM, 2.5 μM, 5 μM) or threohydrobupropion (threo 2 μM, 10 μM, 20 μM). *P<0.05, compared to vehicle control cells and # P<0.05, compared to all other groups based on one-way ANOVA with post-hoc Tukey test. All experiments were conducted in triplicate and the standard deviation is shown as the error bar.

Figure 5. CYP2D6 specific downregulation of mRNA expression by bupropion and its metabolites in plated human hepatocytes

Fold change in (a) CYP2D6, (b) CYP3A4 and (c) CYP1A2 mRNA expression in comparison to vehicle treated controls following 72 hour treatment of plated human hepatocytes from three donors with erythrohyrobupropion (Erythro), threohydrobupropion (Threo), bupropion (Bup), R,R-OH-bupropion (R,R-OH) or S,S-OH-bupropion (S,S-OH) at concentrations approximately 5 times steady state concentrations as described in methods section. *P<0.05, compared to vehicle control cells. All experiments were conducted in triplicate and the standard deviation is shown as the error bar.