New in vitro Affinity‐Based Method of Studying Herb‐Drug Interactions for Direct Identification of CYP1A2, 3A4, and 2C9 Inhibitors from Herbal Extract Using Ultrafiltration‐High Performance Liquid Chromatography

Abstract

In the study of herb‐drug interaction, the complex composition of phytochemicals in herbal medicines makes it difficult to identify which component is responsible for any specific enzyme effect using traditional in vitro or in vivo methods. In the present study, a novel in vitro method of affinity‐based ultrafiltration‐high performance liquid chromatography (HPLC) was developed for studying herb‐drug interactions. This method uses human liver microsome (HLM) as the source of cytochrome P450 (CYP450) in which the ligands of specific enzymes can be directly identified. Danshen (Salvia miltiorrhiza), a well‐studied medicinal plant for the treatment of cardiovascular diseases, was selected to validate the new method. In our method, the herbal extract was incubated without HLM, with HLM, or with the HLM active site blocked before ultrafiltration was performed to isolate ligand‐enzyme complexes from the unbound compounds. This was followed by the addition of 50% methanol to release the ligands from the complexes. The 7,8‐benzoflavone, ketoconazole, and sulfaphenazole were used as the active site blockers of CYP1A2, 3A4, and 2C9, respectively. HPLC analysis was then performed on the released ligands. According to the obtained peak areas, the signal‐noise ratio (S/N) and specific‐signal‐noise ratio (S‐S/N) of each compound can be calculated. S/N is the factor ranking the binding affinities of the compounds from Danshen towards HLM, whereas S‐S/N is the factor indicating the competitive binding ability of specific enzymes between the active site blocker and ligands. Finally, two criteria were applied to select putative targets of each specific CYP450 enzyme: 1) S/N > 1; 2) S‐S/N > 0. Using the method mentioned above, Dihydrotanshinone was identified as the specific ligands of CYP1A2; tanshinone I, cryptotanshino, and tanshinone IIA were identified as specific ligands of CYP1A2, 2C9, and 3A4. These results were verified by the traditional in vitro method. It is demonstrated that the newly developed method has the advantage that the specific ligands (competitive inhibitors) of each CYP450 enzyme in HLM can be identified rapidly and directly from herbal extract. Moreover, this is the first report on multi‐target identification using affinity‐based ultrafiltration‐HPLC.Support or Funding InformationThis work was supported by Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries (IPET) through High Value‐added Food Technology Development Program, funded by Ministry of Agriculture, Food and Rural Affairs (MAFRA)(115001‐3) S/N, S‐S/N, IC50, Ki and inhibition mode of tanshinones on human CYP1A2, 3A4, and 2C9. CYP isoform Danshen components1 S/N2 S‐S/N3 IC50 (μM)4 Ki (μM)4 Mode of inhibition4 1A2 Dihydrotanshinone (1)* 1.16 0.09 0.50 0.53 Competitive Tanshinone I (2)* 1.88 0.04 1.70 2.16 Competitive Cryptotanshinone (3)* 1.33 0.51 3.06 1.88 Competitive Tanshinone IIA (4)* 2.34 0.62 2.01 1.45 Competitive 3A4 Dihydrotanshinone (1) 1.16 −0.21 3.22 2.11 Non‐competitive Tanshinone I (2)* 1.88 0.13 >100 86.9 Competitive Cryptotanshinone (3)* 1.33 0.54 >100 120.4 Competitive Tanshinone IIA (4)* 2.34 0.39 >100 218.7 Competitive 2C9 Dihydrotanshinone (1) 1.16 −0.07 7.48 1.92 Competitive Tanshinone I (2)* 1.88 0.16 >100 51.2 Competitive Cryptotanshinone (3)* 1.33 0.47 23.86 22.9 Competitive Tanshinone IIA (4)* 2.34 0.62 >100 61.6 Competitive Asterisk indicated the targets identified by ultrafiltration. S/N is the signal‐noise ratio. S‐S/N is the specific signal‐noise ratio. IC50, Ki and the mode of inhibition were cited from reference [1], which Ki is the inhibitory constant.