Machine Learning Reveals Amine Type in Polymer Micelles Determines mRNA Binding, In Vitro, and In Vivo Performance...

Area of Science:

• Biomaterials Science
• Nanotechnology
• Molecular Biology

Background:

• Cationic micelles from amphiphilic block copolymers are promising for mRNA delivery.
• Customizable polycationic coronas allow for tailored nanoparticle properties.

Purpose of the Study:

• To systematically investigate the impact of amine chemistry on mRNA delivery using cationic micelles.
• To identify key structure-performance relationships for optimizing mRNA delivery efficacy and specificity.

Main Methods:

• Formulation of 30 cationic micelle nanoparticles (MNPs) with diverse amine functionalities.
• In vitro mRNA delivery assays using GFP+ mRNA across multiple cell lines.
• Machine learning analysis (SHAP) to correlate amine chemistry with performance metrics.
• In vivo delivery studies and correlation with in vitro models (Multitask Gaussian Process).

Main Results:

• Amine side-chain bulk and chemical structure critically influence mRNA delivery performance.
• Amine-specific binding efficiency is a key determinant of efficacy, cell viability, and GFP intensity.
• The A7 amphiphile demonstrated highest GFP expression in vitro and specific lung delivery in vivo.
• A strong correlation between in vitro and in vivo performance was established.

Conclusions:

• Balancing mRNA binding strength is crucial for optimal mRNA delivery.
• Chemical optimization of amine functionalities is pivotal for advancing targeted mRNA delivery.
• In vitro models can effectively predict in vivo outcomes for mRNA delivery systems.