Investigate the potential inhibitors of sphingosine kinase 1 (SphK1) with molecular dynamics and artificial intelligence drug design methods

Abstract

Context

Sphingosine kinase 1 (SphK1) is a sphingosine kinase that can catalyze the phosphorylation of sphingosine to generate sphingosine-1-phosphate. The J-type channel of SPHK1 plays an important role in processes such as cell signaling. Therefore, this study aims to investigate the interaction mechanism between Epidanshenspiroketallactone, PF-543, and SPHK1 in the J-type channel, and to design new small molecules using AI Drug Design (AIDD). Molecular dynamics (MD) simulations reveal that hydrophobic interactions and π-π stacking are of critical significance in stabilizing the J-channel conformation of Sphk1. With MD and AIDD methods, our research provides a novel potential approach for the exploration and design of SphK1 inhibitors.

Methods

The binding mechanism of Epidanshenspiroketallactone and PF-543 with SphK1 was predicted by the molecular dynamics (MD) method using Gromacs-2022–2. Molecular docking was carried out with MolAICal, and the structures were visualized with the Pymol software. The MD simulation force field was selected as the AMBER99SB force field, the temperature was set at 310 K, and the total MD simulation time was 7.2 μs. A recurrent neural network-long short-term memory (RNN-LSTM) machine model was employed for the design of novel inhibitors.