Inhibition of Caffeine Metabolism in Humans by Furanocoumarin-Containing Plant Extracts: In Vivo and In Vitro Studies

Caffeine is found at high concentrations in tea, coffee, soft drinks and energy drinks. Daily consumption of caffeinated beverages is considered to be safe but adverse health effects and deaths have been reported in sensitive or overdosed individuals. A variety of furanocoumarin bioactive has been identified in fruits, spices and herbs from plants in the Apiaceous and Rutaceae families. Since caffeinated beverages are often consumed with food, and both are metabolized by the same CYP1A2 enzyme, we hypothesized metabolic inhibition between caffeine and furanocoumarin-containing food or herbs are common in humans. The goals of this thesis were: (a) to compare the pharmacokinetics of caffeine in humans before and after pre-treatment with a furanocoumarin-containing herb, (b) to elucidate the mechanism(s) of caffeine-herb interaction using in vitro incubations containing pure furanocoumarins and human liver microsomes (HLMs), and (c) to predict in vivo herb-caffeine interactions for humans based on in vitro caffeine metabolism data and in vivo furanocoumarin inhibitor concentrations in the liver.Chapter 1 of this thesis is a brief introduction of caffeine and furanocoumarin-containing food and herbs. In chapter 2, major furanocoumarin bioactives in 29 food and herbs are identified and quantified using gas chromatography mass spectrometry and high performance liquid chromatography. Chapter 3 describes the pharmacokinetics of caffeine in humans after administering 200 mg of caffeine orally before and after pre-treatment with an herbal extract. Caffeine clearance in the volunteers decreased 33.7-77.3% with concomitant increases in area-under-the-concentration-time curve after oral consumption of Ammi majus L., Angelica archangelica L., Angelica pubescens Maxim, Cnidium monnieri (L.) Cusson, or Ruta graveolens L. Chapter 4 provides the experimental evidence for irreversible adduct formation between 14C-labeled 8-methoxypsoralen and HLMs. Moreover, the observed caffeine-herb interaction in humans is best explained by mechanism-based inhibition of CYP1A2 enzyme. Chapter 5 demonstrated the use of in vitro-in vivo drug-drug interaction models and an integrated furanocoumarin dose/concentration to predict in vivo furanocoumarin-caffeine herb interaction in humans. Chapter 6 summarizes the conclusions of this thesis. The experimental and modeling approaches described in this study are also useful in predicting in vivo interactions between prescription drugs and dietary supplements or functional food.