Atomwise Blog

Discovery of Small Molecule Inhibitors Targeting the HEG1-KRIT1 Protein-Protein Interaction with Deep Convolutional Neural Network

Written by Atomwise Inc. | Aug 10, 2021 10:01:52 PM

At the American Chemical Society Fall 2021 National Meeting & Expo, Atomwise members were selected to present their research and work. Learn what our Atoms have been working on below and visit Atomwise at ACS Fall 2021 National Meeting for other presentation sessions. 

 

Ting-Rong Chern, PhD 
Atomwise

Co-Authors: Karol Francisco, Carlo Ballatore, Alexandre Gingras

Title: Discovery of Small Molecule Inhibitors Targeting the HEG1-KRIT1 Protein-Protein Interaction with Deep Convolutional Neural Network

Division: Medicinal Chemistry

View Poster Presentation

 

Abstract

The Kruppel-like factors 4 and 2 (KLF4/2) are transcription factors that play essential roles in endothelial cell phenotype and vascular homeostasis. KLF4/2 regulates the expression of factors responsible for antithrombotic and anti-inflammatory effects in endothelial cells. Therefore, an increased level of endothelial KLF4/2 has a vasoprotective effect. Genetic studies demonstrated that the suppression of endothelial KRIT1 (Krev interaction trapped protein 1) or HEG1 (Heart of glass 1) results in upregulation of KLF4/2. Besides, HEG1 serves as an endothelial cell membrane anchor for KRIT1 and is essential for KRIT1 function. Thus, we hypothesize that pharmacological manipulation of the HEG1-KRIT1 interactions could lead to upregulation of KLF4/2, promoting vasoprotection. We applied the AtomNetĀ® model, a deep convolutional neural network for structure-based drug discovery, to screen 2.5 million commercially available compounds against the HEG1-KRIT1 interface. With an unbiased, AI-guided selection process, we tested 94 compounds in vitro and identified 2 hits. Target engagement experiments demonstrated that both compounds blocked HEG1-KRIT1 interaction in a dose-dependent manner with flow cytometry and fluorescence polarization assay. We continue to harness the power of AtomNetĀ® technology for the hit-to-lead effort. 

 

Join our team

Our team is comprised of over 80 PhD scientists who contribute to a high-performance academic-like culture that fosters robust scientific and technical excellence. We strongly believe that data wins over opinions, and aim for as little dogma as possible in our decision making. Learn more about our team and opportunities at Atomwise.