Label-Free, AI-Driven Evanescent Microscopy Decodes Single-Cell Membrane Protein Binding Kinetics and Adhesion...

Area of Science:

• Biophysics
• Cell Biology
• Artificial Intelligence

Background:

• Quantifying ligand interactions with membrane proteins is vital for cell biology and drug discovery.
• Existing methods face challenges in real-time, label-free analysis within native cellular environments.

Purpose of the Study:

• To introduce an AI-driven evanescent microscopy platform for label-free quantification of ligand-membrane protein binding kinetics.
• To enable automated, subcellular analysis of binding on individual cells in real time.

Main Methods:

• Utilized AI-driven evanescent microscopy for label-free imaging.
• Developed an AI-based multifeature recognition framework to address noise in label-free imaging.
• Employed explainable machine learning to analyze cellular adhesion biomechanics and their impact on binding kinetics.

Main Results:

• Achieved label-free quantification of binding kinetics for both large and small molecules to membrane proteins on multiple individual cells.
• Demonstrated automated, subcellular quantification of ligand binding kinetics, overcoming high noise levels.
• Showcased how cellular biomechanics influence heterogeneous membrane protein binding kinetics, varying with specific ligands and proteins.

Conclusions:

• Presents a scalable solution for label-free single-cell analysis of ligand-membrane protein interactions.
• Highlights the role of cellular biomechanics in regulating binding kinetics.
• Offers a new avenue for understanding drug resistance mechanisms driven by cellular heterogeneity.