Related Experiment Video
Updated: Jul 7, 2025
Exploring Caspase Mutations and Post-Translational Modification by Molecular Modeling Approaches
Published on: October 13, 2022
Targeting Cathepsin L in Cancer Management: Leveraging Machine Learning, Structure-Based Virtual Screening, and
Abdulraheem Ali Almalki1, Alaa Shafie1, Ali Hazazi2
1Department of Clinical Laboratory Sciences, College of Applied Medical Sciences, Taif University, Taif 21944, Saudi Arabia.
Abstract:
Cathepsin L (CTSL) expression is dysregulated in a variety of cancers. Extensive empirical evidence indicates their direct participation in cancer growth, angiogenic processes, metastatic dissemination, and the development of treatment resistance. Currently, no natural CTSL inhibitors are approved for clinical use. Consequently, the development of novel CTSL inhibition strategies is an urgent necessity. In this study, a combined machine learning (ML) and structure-based virtual screening strategy was employed to identify potential natural CTSL inhibitors. The random forest ML model was trained on IC50 values. The accuracy of the trained model was over 90%. Furthermore, we used this ML model to screen the Biopurify and Targetmol natural compound libraries, yielding 149 hits with prediction scores >0.6. These hits were subsequently selected for virtual screening using a structure-based approach, yielding 13 hits with higher binding affinity compared to the positive control (AZ12878478). Two of these hits, ZINC4097985 and ZINC4098355, have been shown to strongly bind CTSL proteins. In addition to drug-like properties, both compounds demonstrated high affinity, ligand efficiency, and specificity for the CTSL binding pocket. Furthermore, in molecular dynamics simulations spanning 200 ns, these compounds formed stable protein-ligand complexes. ZINC4097985 and ZINC4098355 can be considered promising candidates for CTSL inhibition after experimental validation, with the potential to provide therapeutic benefits in cancer management.
Insights
Researchers identified natural compounds ZINC4097985 and ZINC4098355 as potent inhibitors of Cathepsin L (CTSL). These compounds show promise for cancer therapy by targeting CTSL
Area of Science:
- Biochemistry
- Computational Chemistry
- Oncology
Background:
- Cathepsin L (CTSL) is implicated in cancer progression, angiogenesis, metastasis, and treatment resistance.
- There is a critical need for novel, clinically approved natural inhibitors of CTSL.
- Current therapeutic strategies lack effective natural compounds targeting CTSL.
Purpose of the Study:
- To identify novel natural compounds that inhibit Cathepsin L (CTSL) using a combined computational approach.
- To screen natural product libraries for potential CTSL inhibitors with high binding affinity and drug-like properties.
- To evaluate the stability and binding characteristics of identified compounds with CTSL.
Main Methods:
- Employed a machine learning (ML) random forest model trained on IC50 values for initial screening.
- Utilized structure-based virtual screening to assess binding affinity of ML-identified hits.
- Conducted 200 ns molecular dynamics simulations to analyze protein-ligand complex stability.
Main Results:
- Developed an ML model with over 90% accuracy for predicting CTSL inhibitors.
- Screened natural compound libraries, identifying 149 potential hits via ML.
- Selected 13 compounds via virtual screening with superior binding affinity to the positive control.
- Identified ZINC4097985 and ZINC4098355 as high-affinity binders, demonstrating stable complex formation with CTSL.
Conclusions:
- ZINC4097985 and ZINC4098355 are promising natural candidates for CTSL inhibition.
- These compounds exhibit favorable drug-like properties, high affinity, and specificity for CTSL.
- Further experimental validation is warranted to explore their therapeutic potential in cancer management.
