Determining Mechanisms Underlying the Goldenseal‐Midazolam Interaction in Humans Using an In Vitro to In Vivo Extrapolation Approach
Open Access
- 1 April 2019
- journal article
- research article
- Published by Wiley in The FASEB Journal
- Vol. 33 (S1) , 814.14
Abstract
Background: ldenseal is a botanical natural product commonly used to self‐treat the common cold and various digestive ailments. Major constituents of goldenseal include the alkaloids berberine, (−)‐hydrastine, and hydrastinine; each constituent contains a methylenedioxyphenyl ring. This structural feature can give rise to time‐dependent inhibition (TDI) of cytochromes P450 (CYPs), which may contribute to the mechanism(s) underlying reported clinical goldenseal‐drug interactions involving the CYP3A probe substrate midazolam. The aims of this work were to 1) determine the TDI kinetics of each constituent towards CYP3A activity (midazolam 1’‐hydroxylation) using human liver and intestinal microsomes (HLM, HIM), 2) apply the kinetic parameters to a mechanistic static model to predict the change in systemic exposure to oral midazolam, and 3) assess the accuracy of the prediction in healthy volunteers. Methods: IC 50 shift experiments were conducted to identify potential time‐dependent inhibitors. K I and k inact were determined using a 6 x 6 matrix of inhibitor concentrations (0.5–90 μM) and primary incubation times (0–20 min). These parameters were applied to a mechanistic static model to predict the change in midazolam area under the plasma concentration‐time curve (AUC). The pharmacokinetics of midazolam and hydroxylated metabolites were determined for 6 healthy volunteers administered a single oral dose of midazolam (2.5 mg) with a well‐characterized goldenseal product (3 g). Results: Only berberine and (−)‐hydrastine showed TDI of CYP3A activity in HLM (K I , 13–42 μM; k inact , 0.015–0.061 min −1 ) and HIM (K I , 5.4–8.5 μM; k inact , 0.040 min −1 ). Berberine additionally showed activation of CYP3A activity (up to 2‐fold compared to vehicle) at low midazolam concentrations (≤5 μM). The increase in the 1’‐ to 4‐hydroxymidazolam (1’‐OH/4‐OH) ratio with increasing berberine concentrations in both HLM and HIM (~6 to ~30) further supported activation of CYP3A activity by berberine. Berberine and (−)‐hydrastine were predicted to increase midazolam AUC by at least 6‐fold. Goldenseal increased midazolam AUC in all of the participants (1.3 to 2.2‐fold). Analysis of 1’‐OH/4‐OH partial AUC ratios did not indicate activation of CYP3A in vivo. Conclusions: The average k inact /K I ratios for berberine (1.3 mL•min −1 •μmol −1 ) and (−)‐hydrastine (4.0 mL•min −1 •μmol −1 ) were near those reported for the clinically relevant time‐dependent CYP3A inhibitors diltiazem and erythromycin (1.8–4.0 mL•min −1 •μmol −1 ), suggesting berberine and (−)‐hydrastine as candidate precipitants of the goldenseal‐midazolam interaction. Activation of CYP3A activity by berberine in vitro may have contributed to the over‐predicted midazolam AUC ratio by the static model, which only considers inhibition. Given that the half‐life for CYP3A4 is ~1–3 days, TDI of CYP3A activity in vivo by goldenseal could have persistent effects beyond cessation of consuming the natural product. Support or Funding Information National Center for Complementary and Integrative Health (U54AT008909) This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .Keywords
Funding Information
- National Center for Complementary and Integrative Health (U54AT008909)