Machine Learning-Engineered Nanozyme System for Synergistic Anti-Tumor Ferroptosis/Apoptosis Therapy.

Area of Science:

• Biomedical Engineering
• Materials Science
• Computational Biology

Background:

• Nanozymes with multi-enzyme activity show promise for anti-tumor therapy by interacting with the tumor microenvironment (TME).
• Protective autophagy induced by nanozymes can significantly reduce their therapeutic effectiveness.

Purpose of the Study:

• To develop a targeted nanozyme system (FZH) that enhances synergistic anti-tumor ferroptosis and apoptosis therapy.
• To utilize machine learning (ML) for optimizing nanozyme design and discovering novel therapeutic strategies.

Main Methods:

• A novel ML model, sequential backward Tree-Classifier for Gaussian Process Regression (TCGPR), was developed for improved data pattern recognition.
• A Bayesian optimization algorithm was employed for candidate selection within an extensive material search space.
• The Fe-Arg-CDs@ZIF-8/HAD (FZH) nanozyme system was designed and evaluated for its anti-tumor effects.

Main Results:

• The FZH nanozyme system effectively exerts anti-tumor effects by sequentially responding to the TME, inducing a cascade reaction to promote ferroptosis.
• Endogenous nitric oxide (NO) elevation by FZH directly kills tumor cells.
• FZH suppresses protective autophagy induced by oxidative stress (OS), thereby enhancing synergistic ferroptosis/apoptosis therapy.

Conclusions:

• A novel strategy integrating ML, experimental validation, and biological applications has been developed for improving nanozyme-based tumor therapy.
• The FZH system demonstrates significant potential for synergistic anti-tumor ferroptosis and apoptosis therapy by overcoming treatment resistance mechanisms.
• This work highlights the power of ML-driven material discovery in advancing cancer therapeutics.