Identification of selective dual-specificity tyrosine phosphorylation-regulated kinase 1A (DYRK1A) inhibitors and their effects on tau and microtubule.

The overexpression of dual-specificity tyrosine phosphorylation-regulated kinase 1A (DYRK1A), commonly observed in neurodegenerative diseases like Alzheimer's disease (AD) and Down syndrome (DS), can induce the formation of neurofibrillary tangles (NFTs) and amyloid plaques. Hence, designing a selective DYRK1A inhibitor would result in a promising small molecule for treating neurodegenerative diseases. Developing selective inhibitors for DYRK1A has been a difficult challenge due to the highly preserved ATP-binding site of protein kinases. In this study, we employed a structure-based virtual screening (SBVS) campaign targeting DYRK1A from a database containing 1.6 million compounds. Enzymatic assays were utilized to verify inhibitory properties, confirming that Y020-3945 and Y020-3957 showed inhibitory activity towards DYRK1A. In particular, the compounds exhibited high selectivity for DYRK1A over a panel of 120 kinases, reduced the phosphorylation of tau, and reversed the tubulin polymerization for microtubule stability. Additionally, treatment with the compounds significantly reduced the secretion of inflammatory cytokines IL-6 and TNF-α activated by DYRK1A-assisted NFTs and Aß oligomers. These identified inhibitors possess promising therapeutic potential for conditions associated with DYRK1A in neurodegenerative diseases. The results showed that Y020-3945 and Y020-3957 demonstrated structural novelty compared to known DYRK1A inhibitors, making them a valuable addition to developing potential treatments for neurodegenerative diseases.

An ensemble machine learning model generates a focused screening library for the identification of CDK8 inhibitors.

The identification of an effective inhibitor is an important starting step in drug development. Unfortunately, many issues such as the characterization of protein binding sites, the screening library, materials for assays, etc., make drug screening a difficult proposition. As the size of screening libraries increases, more resources will be inefficiently consumed. Thus, new strategies are needed to preprocess and focus a screening library towards a targeted protein. Herein, we report an ensemble machine learning (ML) model to generate a CDK8-focused screening library. The ensemble model consists of six different algorithms optimized for CDK8 inhibitor classification. The models were trained using a CDK8-specific fragment library along with molecules containing CDK8 activity. The optimized ensemble model processed a commercial library containing 1.6 million molecules. This resulted in a CDK8-focused screening library containing 1,672 molecules, a reduction of more than 99.90%. The CDK8-focused library was then subjected to molecular docking, and 25 candidate compounds were selected. Enzymatic assays confirmed six CDK8 inhibitors, with one compound producing an IC50 value of ≤100 nM. Analysis of the ensemble ML model reveals the role of the CDK8 fragment library during training. Structural analysis of molecules reveals the hit compounds to be structurally novel CDK8 inhibitors. Together, the results highlight a pipeline for curating a focused library for a specific protein target, such as CDK8.