Differential effect of Shenmai injection, a herbal preparation, on the cytochrome P450 3A-mediated 1′-hydroxylation and 4-hydroxylation of midazolam

Shenmai injection (SMI), one of the most popular herbal preparations, is widely used for the treatment of coronary atherosclerotic cardiopathy and viral myocarditis. The purpose of this study was to investigate the effect of Shenmai injection (SMI) on the CYP3A-mediated metabolism of midazolam (MDZ). The present study demonstrated that SMI could significantly inhibit MDZ 4-hydroxylation but activate its 1′-hydroxylation in human liver microsomes (HLMs), rat liver microsomes (RLM) and recombinant human CYP3A4 and CYP3A5. The opposing effect of SMI was characterized by the kinetic change of increasing Vmax/Km for MDZ 1′-hydroxylation and decreasing Vmax/Km for MDZ 4-hydroxylation in HLM and RLM. The presence of SMI enhanced the inhibition of ketoconazole on MDZ 4-hydroxylation but weakened or reversed its inhibition on MDZ 1′-hydroxylation in HLM. After single or multiple pretreatment with SMI, the ratios of AUC4-OH MDZ/AUCMDZ in rats were significantly decreased, while the ratios of AUC1′-OH MDZ/AUCMDZ were increased. Among the major components in SMI, total ginsenoside (TG), ophiopogon total saponins (OTS), ophiopogon total flavone (OTF), ginsenoside Rd, ophiopogonin D and ophiopogonone A exhibited significant inhibition on both 4-hydroxylation and 1′-hydroxylation of MDZ in HLM and RLM, while no activation on MDZ metabolism was observed in the presence of these major constituents alone or together. To further explore the responsible components, 3 mL of SMI was loaded on a solid phase extraction (SPE) C18 cartridge and then separated by different concentrations of methanol. The fractions eluted with 60% and 90% methanol both showed significant activation on MDZ 1′-hydroxylation in HLM, but the fraction eluted with 30% methanol had no such effect. The results indicated that the activation of SMI on MDZ 1′-hydroxylation might be mainly resulted from the lipid-soluble components in SMI.