Dietary Inhibitors of CYP3A4 Are Revealed Using Virtual Screening by Using a New Deep-Learning Classifier

Yelena Guttman, Zohar Kerem*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

10 Scopus citations

Abstract

CYP3A4 is the main human enzyme responsible for phase I metabolism of dietary compounds, prescribed drugs and xenobiotics, steroid hormones, and bile acids. The inhibition of CYP3A4 activity might impair physiological mechanisms, including the endocrine system and response to drug admission. Here, we aimed to discover new CYP3A4 inhibitors from food and dietary supplements. A deep-learning model was built that classifies compounds as either an inhibitor or noninhibitor, with a high specificity of 0.997. We used this classifier to virtually screen ∼60,000 dietary compounds. Of the 115 identified potential inhibitors, only 31 were previously suggested. Many herbals, as predicted here, might cause impaired metabolism of drugs, and endogenous hormones and bile acids. Additionally, by applying Lipinski’s rules of five, 17 compounds were also classified as potential intestine local inhibitors. New CYP3A4 inhibitors predicted by the model, bilobetin and picropodophyllin, were assayed in vitro.

Original languageAmerican English
Pages (from-to)2752-2761
Number of pages10
JournalJournal of Agricultural and Food Chemistry
Volume70
Issue number8
DOIs
StatePublished - 2 Mar 2022

Bibliographical note

Publisher Copyright:
© 2022 The Authors. Published by American Chemical Society

Keywords

  • cytochrome P450 3A4 (CYP3A4)
  • deep learning
  • dietary compounds
  • food−drug interactions
  • intestine

Fingerprint

Dive into the research topics of 'Dietary Inhibitors of CYP3A4 Are Revealed Using Virtual Screening by Using a New Deep-Learning Classifier'. Together they form a unique fingerprint.

Cite this