Machine Learning-Driven Discovery of an Endoplasmic Reticulum-Targeting Gallium(III) Complex that Triggers Immunogenic Paraptosis in Osteosarcoma
Abstract Osteosarcoma is a highly aggressive bone tumor characterized by profound chemoresistance and immune evasion, resulting in poor clinical prognoses. The rational design of metallodrugs to overcome these obstacles is a formidable challenge, as traditional discovery methods are inefficient for predicting the organelle-specific targeting required for specific mechanisms of action. Here, we introduce a machine learning-driven platform to rationally design and identify endoplasmic reticulum (ER)-targeting Ga(III) complexes capable of activating nonapoptotic immunogenic cell death. This strategy identified Ga20, a complex that triggers a dual mechanism of paraptosis and immunogenic cell death (ICD), directly addressing osteosarcoma’s apoptosis resistance. Formulated as a polymeric nanoparticle (Ga NP) for targeted in vivo delivery, this nanomedicine demonstrated profound tumor growth inhibition across aggressive and clinically relevant orthotopic, metastatic, bilateral, and patient-derived xenograft models. Critically, Ga NP treatment generated a durable in situ vaccine effect, conferring long-term immune memory and preventing tumor recurrence and metastasis. This work establishes a strategy for the computational design of organelle-targeted metallodrugs and presents a powerful chemoimmunotherapy that directly addresses the core clinical challenges of osteosarcoma.
Authors
- Nicolás Montesdeoca (ORCID: https://orcid.org/0000-0002-9629-1265)
- Johannes Karges (ORCID: https://orcid.org/0000-0001-5258-0260)
- Haihua Xiao (ORCID: https://orcid.org/0000-0001-8159-8187)
- Minhui Cui
- Tang Dongsheng
- Ganghao Liang
- Yingjie Yu (ORCID: https://orcid.org/0000-0002-2543-9808)
- Delong Huang
- Meifang Shen
- Zisis Papadopoulos (ORCID: https://orcid.org/0000-0002-2536-2134)
- Hanchen Zhang
Institutions
- Chinese Academy of Sciences (CN)
- Fiberhome Technology Group (China) (CN)
- University of Chinese Academy of Sciences (CN)
- Beijing University of Chemical Technology (CN)
- Ruhr University Bochum (DE)
Publication Details
- Journal
- ACS Nano
- Published
- 2026-11-16
- DOI
- https://doi.org/10.1021/acsnano.5c20201
- Primary Topic
- Metal complexes synthesis and properties
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Fonds der Chemischen Industrie
- National Natural Science Foundation of China
- Paul-Ehrlich-Gesellschaft Stiftung
- Aventis Foundation