Inhibitory effects of herbal constituents on P-glycoprotein in vitro and in vivo: Herb–drug interactions mediated via P-gp

• Emodin, 18β-GA, DAG, and 20(S)-GF1 significantly inhibited P-gp in vitro.
• P-gp ATPase activity was stimulated by emodin and DAG.
• 18β-GA and 20(S)-GF1 exhibited significant inhibition on P-gp ATPase activity.
• Molecular docking analysis elucidated the SAR of herbal constituents with P-gp.
• Pretreatment with emodin or 18β-GA increased the AUC and Cmax of digoxin in vivo.

Modulation of drug transporters via herbal medicines which have been widely used in combination with conventional prescription drugs may result in herb–drug interactions in clinical practice. The present study was designed to investigate the inhibitory effects of 50 major herbal constituents on P-glycoprotein (P-gp) in vitro and in vivo as well as related inhibitory mechanisms. Among these herbal medicines, four constituents, including emodin, 18β-glycyrrhetic acid (18β-GA), dehydroandrographolide (DAG), and 20(S)-ginsenoside F1 [20(S)-GF1] exhibited significant inhibition (> 50%) on P-gp in MDR1-MDCKII and Caco-2 cells. Emodin was the strongest inhibitor of P-gp (IC50 = 9.42 μM), followed by 18β-GA (IC50 = 21.78 μM), 20(S)-GF1 (IC50 = 76.08 μM) and DAG (IC50 = 77.80 μM). P-gp ATPase activity, which was used to evaluate the affinity of substrates to P-gp, was stimulated by emodin and DAG with Km and Vmax values of 48.61, 29.09 μM and 71.29, 38.45 nmol/min/mg protein, respectively. However, 18β-GA and 20(S)-GF1 exhibited significant inhibition on both basal and verapamil-stimulated P-gp ATPase activities at high concentration. Molecular docking analysis (CDOCKER) further elucidated the mechanism for structure–inhibition relationships of herbal constituents with P-gp. When digoxin was co-administered to male SD rats with emodin or 18β-GA, the AUC0 − t and Cmax of digoxin were increased by approximately 51% and 58%, respectively. Furthermore, 18β-GA, DAG, 20(S)-GF1 and Rh1 at 10 μM significantly inhibited CYP3A4/5 activity, while emodin activated the metabolism of midazolam in human liver microsomes. In conclusion, four herbal constituents demonstrated inhibition of P-gp to specific extents in vitro and in vivo. Taken together, our findings provided the basis for the reliable assessment of the potential risks of herb–drug interactions in humans.