AI-Assisted chemical probe discovery for the understudied Calcium-Calmodulin Dependent Kinase, PNCK

Abstract

PNCK, or CAMK1b, is an understudied kinase of the calcium-calmodulin dependent kinase family which recently has been identified as a marker of cancer progression and survival in several large-scale multi-omics studies. The biology of PNCK and its relation to oncogenesis has also begun to be elucidated, with data suggesting various roles in DNA damage response, cell cycle control, apoptosis and HIF-1-alpha related pathways. To further explore PNCK as a clinical target, potent small-molecule molecular probes must be developed. Currently, there are no targeted small molecule inhibitors in pre-clinical or clinical studies for the CAMK family. Additionally, there exists no experimentally derived crystal structure for PNCK. We herein report a three-pronged chemical probe discovery campaign which utilized homology modeling, machine learning, virtual screening and molecular dynamics to identify small molecules with low-micromolar potency against PNCK activity from commercially available compound libraries. We report the discovery of a hit-series for the first targeted effort towards discovering PNCK inhibitors that will serve as the starting point for future medicinal chemistry efforts for hit-to-lead optimization of potent chemical probes. Machine learning and virtual screening are powerful tools in the pharmacologist’s arsenal for accelerating the process of drug discovery. When targeted lesser-known proteins, however, it is important to first develop a potent, selective chemical probe. The chemical probe allows for pharmacological inhibition of protein activity to be used in addition to genetic knock-down or knock-out assays for studying the biological function of your protein in interest. In a previous multi-omics study of patient tumors, we had identified PNCK as a target of interest in kidney cancer. However, the function of PNCK is largely unknown as it has been designated as an understudied kinase. As such, we utilized kinase activity data for compounds known to target structurally similar kinases to develop machine learning models to predict small molecule binding to PNCK. Additionally, we simply used shape screening to find compound similar in shape and electronics to ATP when bound to the active site of PNCK in our multiple homology models. Using a combination of virtual methods, we were able to identify several hit compounds with favorable, tractable scaffolds, to move forward in a hit-to-lead campaign to develop the first in class, selective PNCK chemical probe. This work will lead to the elucidation of PNCKs function in several cancer models and subsequently lead to the development of a novel drug.