An Integrated Deep Learning and Molecular Dynamics Simulation-Based Screening Pipeline Identifies Inhibitors of a New Cancer Drug Target TIPE2

Abstract

The TIPE2 (tumor necrosis factor-alpha-induced protein 8-like 2) protein is a major regulator of cancer and inflammatory diseases. The availability of its sequence and structure, as well as the critical amino acids involved in its ligand binding, provides insights into its function and helps greatly identify novel drug candidates against TIPE2 protein. With the current advances in deep learning and molecular dynamics simulation-based drug screening, large-scale exploration of inhibitory candidates for TIPE2 becomes possible. In this work, we apply deep learning-based methods to perform a preliminary screening against TIPE2 over several commercially available compound datasets. Then, we carried a fine screening by molecular dynamics simulations, followed by metadynamics simulations. Finally, four compounds were selected for experimental validation from 64 candidates obtained from the screening. With surprising accuracy, three compounds out of four can bind to TIPE2. Among them, UM-164 exhibited the strongest binding affinity of 4.97 µM and was able to interfere with the binding of TIPE2 and PIP2 according to competitive bio-layer interferometry (BLI), which indicates that UM-164 is a potential inhibitor against TIPE2 function. The work demonstrates the feasibility of incorporating deep learning and MD simulation in virtual drug screening and provides high potential inhibitors against TIPE2 for drug development.

Validation of Deep Learning-Based DFCNN in Extremely Large-Scale Virtual Screening and Application in Trypsin I Protease Inhibitor Discovery

“…We adopt the trained model (DFCNN) from our previous work ( Zhang et al, 2019 ; Zhang et al, 2020a ; Zhang et al, 2020b ; Zhang et al, 2021 ). Previously, the model was used for inverse target searching; here, we used it as a core module in extremely large-scale virtual screening.…”

“…Previously, the model was used for inverse target searching; here, we used it as a core module in extremely large-scale virtual screening. In addition, we have used DFCNN as a core model in a hybrid pipeline to help identify inhibitors for protein targets RdRp and TIPE2 in our previous works ( Zhang et al, 2020b ; Zhang et al, 2021 ). We have also applied DFCNN to do a virtual screening over the main protease of SARS-CoV-2 ( Zhang et al, 2020a ).…”

“…The RMSprop optimizer was used to minimize the binary cross entropy of CNN and the Adam optimizer was used to minimize the binary cross entropy of DFCNN. The model architecture, the training data preparation, and the training and testing procedures can be found in our previous work in detail ( Zhang et al, 2019 ; Zhang et al, 2020a ; Zhang et al, 2020b ; Zhang et al, 2021 ).…”

“…DFCNN, our previously proposed model, has an advantage both in terms of both accuracy and speed ( Zhang et al, 2019 ). Moreover, as one of the core components of a hybrid drug screening pipeline, DFCNN has already been successfully applied to discover novel inhibitors for some important targets such as RdRp ( Zhang et al, 2020b ) and TIPE2 ( Zhang et al, 2021 ) in our previous research. However, the performance of DFCNN on different types of proteins is urgently needed to be systematically evaluated.…”

An Efficient Modern Strategy to Screen Drug Candidates Targeting RdRp of SARS-CoV-2 With Potentially High Selectivity and Specificity

“…There are many other types of deep learning-based protein-ligand interaction prediction models, such as DeepDTAF ( Wang et al, 2021 ), DeepBindRG ( Zhang et al, 2019b ), and DeepAffinity ( Karimi et al, 2019 ). We also developed several protein-ligand interaction models, and some of them have successfully been applied to virtual screening applications for targets such as RdRp ( Zhang et al, 2020b ), 3C proteases ( Zhang et al, 2020a ), and TIPE2 ( Zhang et al, 2021 ). For those protein-ligand binding prediction models, some predict binding affinity, whereas some are binary predictions of the binding possibility.…”