Virtual screening of drugs targeting PD-L1 protein

Abstract

Monoclonal antibody inhibitors targeting PD1/PD-L1 immune checkpoints are gradually entering the market and have achieved certain positive effects in various types of tumor treatments. However, with the expansion of applications, the limitations of antibody drugs have gradually emerged, and small molecule compound inhibitors have become a new focus of attention for researchers. This study aims to use ligand-based and structure-based binding activity prediction methods to achieve virtual screening of small molecule compounds targeting PD-L1, thereby helping to accelerate the development of small molecule drugs. A dataset of PD-L1 small molecule inhibitory activity from relevant research literatures and patents was collected and machine learning activity judgment classification models with activity intensity prediction models were constructed based on different molecular characterization methods and algorithms. The two types of models filtered 68 candidate compounds with high PD-L1 inhibitory activity from a large drug-like small molecule screening pool. Ten of these compounds not only had good drug similarity and pharmacokinetics, but also showed the same level of binding strength and similar mechanism of action with previous hot compounds in molecule docking. This phenomenon was further verified in subsequent molecular dynamics simulation and binding free energy estimation. In this study, a virtual screening workflow integrating ligand-based method and structure-based method was developed, which effectively screened potential PD-L1 small molecule inhibitors in large compound databases, and is expected to help accelerate the application and expansion of tumor immunotherapy.