Introduction and Objective: Cathepsin L (CTSL) is a lysosomal cysteine protease that promotes tubular inflammation, extracellular matrix remodeling, and renal fibrosis in diabetic kidney disease (DKD). Although CTSL is a validated therapeutic target, the lack of clinically actionable inhibitors remains a major translational barrier. This study aimed to establish a deep learning-enabled discovery framework to identify CTSL inhibitors from natural products and evaluate their therapeutic potential in DKD-relevant models. Methods: An integrated pipeline combining deep learning-based activity prediction, structure-informed molecular docking, and multidimensional ranking was used to prioritize CTSL inhibitor candidates from a curated natural product library. Two hundred compounds were selected for experimental validation. CTSL enzymatic inhibition was assessed in vitro, followed by cytotoxicity screening in human renal tubular HK-2 cells. Lead compounds were evaluated in metabolically stressed DKD cellular models. Enzyme kinetic analyses and molecular dynamics simulations were performed to characterize inhibition mechanisms and target engagement. Results: Nine compounds showed reproducible, concentration-dependent CTSL inhibition. Kuwanon G, Iberverin, and Wighteone achieved greater than 50% inhibition at non-cytotoxic concentrations. Kuwanon G emerged as the lead compound, significantly reducing inflammatory signaling and fibrotic marker expression under high-glucose and high-lipid conditions. Kinetic analyses identified Kuwanon G as a competitive CTSL inhibitor, while molecular dynamics simulations demonstrated stable binding within the catalytic pocket. Conclusion: Deep learning-enabled drug discovery effectively accelerates CTSL inhibitor identification and overcomes key translational barriers in DKD. Kuwanon G represents a promising CTSL-targeted therapeutic candidate for further development. Disclosure F. Ma: None. S. Zhou: None. Q. Li: None. Z. Jingyi: None. F. Shen: None. J. Yang: None. Funding National Natural Science Foundation of China (Award No. 81930019), the Scientific Project of Beijing’s Municipal Science & Technology Commission (Award No. D171100002817005), the Beijing Municipal Administration of Hospitals Clinical Medicine Development of Special Funding Support Programme (Award No. ZYLX201823), the Training Fund for Open Projects at Clinical Institutes and Departments of Capital Medical University (Award No. CCMU2024ZKYXY005).
Ma et al. (Fri,) studied this question.