Machine Learning-Guided Virtual Screening Identifies a Novel MELK-Targeted Antiproliferative Lead Compound with G2/M Arrest Activity in Triple-Negative Breast Cancer and Lung Adenocarcinoma Cells
4 affiliations
- 1Beijing Key Laboratory of Innovative Drug Discovery and Polymorphic Druggability Research for Cerebrovascular Diseases, Institute of Materia Medica, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100050, China
- 2School of Clinical Pharmacy, Shenyang Pharmaceutical University, Shenyang 117004, China
- 3School of Pharmacy, Henan University, Kaifeng 475004, China
- 4Modern Research Center for Traditional Chinese Medicine, Shanxi University, Taiyuan 030006, China
Abstract
Background: Maternal embryonic leucine zipper kinase (MELK) is a cell cycle-regulated serine/threonine kinase frequently overexpressed in human malignancies, and represents a promising anticancer drug target. Discovery of structurally novel, highly selective MELK inhibitors remains an unmet need. Methods: In the study, we implemented an integrated computational pipeline combining four machine learning regressors (Random Forest, Gradient Boosting, XGBoost, LightGBM), hierarchical Glide docking (HTVS to SP to XP), MM/GBSA rescoring, ADMET prediction, and 100 ns molecular dynamics simulations to screen a ~15.58 million-compound commercial library, yielding 12 prioritized candidates. Three commercially available compounds (CPD4, CPD6, CPD10) were evaluated in an antiproliferative panel. Results: CPD6 was identified as the most active hit, with IC50 values of 1.44 μM in MDA-MB-468 and 1.66 μM in H1299, modestly exceeding the reference compound MELK-8a under the same assay conditions (2.48 and 2.51 μM, respectively). Cellular thermal shift assay in cell lysates provided biochemical evidence of MELK engagement, and 100 ns molecular dynamics simulations supported a computationally predicted binding mode in the MELK kinase domain. Additional substrate phosphorylation and MELK overexpression rescue assays further supported MELK-dependent inhibitory activity. Colony formation, EdU incorporation, and cell cycle analyses demonstrated dose-dependent antiproliferative effects and G2/M arrest, with concomitant downregulation of CDC25C, CDK1, and Cyclin B1. Discussion: CPD6 is a structurally novel MELK lead distinct from reported inhibitors. While our data support MELK target engagement, kinome selectivity profiling remains required, and this ML pipeline demonstrates the utility of AI-driven large-scale anticancer lead screening. Conclusions: Using the integrated machine-learning virtual screening pipeline established in this work, we identified CPD6 as a structurally novel MELK-targeted lead that engages MELK, induces G2/M arrest, and exerts low-micromolar antiproliferative activity in triple-negative breast cancer and lung adenocarcinoma models.
Related
- Long-Term Drug Screening Identifies PARP Inhibitors as Irinotecan Combination Partners in BRCA1/2-Wild-Type Pancreatic Ductal Adenocarcinoma.
- 1098 A 3D tumor-stroma-immune co-culture platform for bispecific antibody screening in pancreatic ductal adenocarcinoma
- Single-cell RNA sequencing identifies stem-like subpopulations and their gene signature in lung adenocarcinoma progression
- A single-cell transcriptomics-based prognostic model identifies TM4SF1 as a key promoter of malignant progression in lung adenocarcinoma
- Comprehensive multi-omics analysis identifies BTG2 as a potential regulator of radiotherapy response in lung adenocarcinoma.
BibTeX
@article{cao2026machine,
title = {{Machine Learning-Guided Virtual Screening Identifies a Novel MELK-Targeted Antiproliferative Lead Compound with G2/M Arrest Activity in Triple-Negative Breast Cancer and Lung Adenocarcinoma Cells}},
author = {Wanxin Cao and Bin Jiang and Xiangyin Chi and Wan Li and Hong Yang and Xiaoxue Li and Yuxuan Zhao and Xu Zhang and Rui Meng and Yanyu Cang and Guanhua Du and Jinhua Wang},
journal = {Cancers},
year = {2026},
month = oct,
doi = {10.3390/cancers18203336},
url = {https://www.mdpi.com/2072-6694/18/20/3336},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Machine Learning-Guided Virtual Screening Identifies a Novel MELK-Targeted Antiproliferative Lead Compound with G2/M Arrest Activity in Triple-Negative Breast Cancer and Lung Adenocarcinoma Cells
AU - Wanxin Cao
AU - Bin Jiang
AU - Xiangyin Chi
AU - Wan Li
AU - Hong Yang
AU - Xiaoxue Li
AU - Yuxuan Zhao
AU - Xu Zhang
AU - Rui Meng
AU - Yanyu Cang
AU - Guanhua Du
AU - Jinhua Wang
PY - 2026
DA - 2026/10/01
JO - Cancers
DO - 10.3390/cancers18203336
UR - https://www.mdpi.com/2072-6694/18/20/3336
AB - Background: Maternal embryonic leucine zipper kinase (MELK) is a cell cycle-regulated serine/threonine kinase frequently overexpressed in human malignancies, and represents a promising anticancer drug target. Discovery of structurally novel, highly selective MELK inhibitors remains an unmet need. Methods: In the study, we implemented an integrated computational pipeline combining four machine learning regressors (Random Forest, Gradient Boosting, XGBoost, LightGBM), hierarchical Glide docking (HTVS to SP to XP), MM/GBSA rescoring, ADMET prediction, and 100 ns molecular dynamics simulations to screen a ~15.58 million-compound commercial library, yielding 12 prioritized candidates. Three commercially available compounds (CPD4, CPD6, CPD10) were evaluated in an antiproliferative panel. Results: CPD6 was identified as the most active hit, with IC50 values of 1.44 μM in MDA-MB-468 and 1.66 μM in H1299, modestly exceeding the reference compound MELK-8a under the same assay conditions (2.48 and 2.51 μM, respectively). Cellular thermal shift assay in cell lysates provided biochemical evidence of MELK engagement, and 100 ns molecular dynamics simulations supported a computationally predicted binding mode in the MELK kinase domain. Additional substrate phosphorylation and MELK overexpression rescue assays further supported MELK-dependent inhibitory activity. Colony formation, EdU incorporation, and cell cycle analyses demonstrated dose-dependent antiproliferative effects and G2/M arrest, with concomitant downregulation of CDC25C, CDK1, and Cyclin B1. Discussion: CPD6 is a structurally novel MELK lead distinct from reported inhibitors. While our data support MELK target engagement, kinome selectivity profiling remains required, and this ML pipeline demonstrates the utility of AI-driven large-scale anticancer lead screening. Conclusions: Using the integrated machine-learning virtual screening pipeline established in this work, we identified CPD6 as a structurally novel MELK-targeted lead that engages MELK, induces G2/M arrest, and exerts low-micromolar antiproliferative activity in triple-negative breast cancer and lung adenocarcinoma models.
SN - 2072-6694
ER -