Machine learning to determine optimal conditions for controlling the size of elastin-based particles

Area of Science:

• Biomaterials Science
• Polymer Chemistry
• Nanotechnology

Background:

• Elastin-like polypeptides (ELP) exhibit tunable phase transitions but lack methods for controlled aggregate size. Branched polyethyleneimine (PEI) is a cationic polymer used in delivery systems.
• Controlling aggregate size and stability is crucial for effective drug and gene delivery applications.

Purpose of the Study:

• To evaluate the aggregation behavior of PEI-ELP conjugates for controlled particle formation.
• To investigate the influence of pH, polymer concentration, and salt concentration on aggregate size and stability.
• To demonstrate the potential of PEI-ELP conjugates as stable biological delivery vehicles.

Main Methods:

• Conjugation of ELP with PEI to create hybrid polymers.
• Systematic variation of solution conditions: pH (3, 7, 10), polymer concentration (0.1–0.3 mg/mL), and salt concentration (0–1 M).
• K-means cluster analysis to determine aggregate hydrodynamic radius and characterize aggregation behavior.

Main Results:

• Salt concentration was identified as the most critical factor influencing hydrodynamic radius and lower critical solution temperature (LCST).
• Controlled aggregate radii were achieved by optimizing solution conditions (pH, polymer concentration, salt concentration).
• Crosslinking of PEI-ELP conjugates resulted in stable particles that maintained size below LCST, even after removal of inducing conditions.

Conclusions:

• PEI-ELP conjugation enables precise control over aggregate size and stability for potential drug and gene delivery.
• Optimized solution conditions, particularly salt concentration, are key to tuning the aggregation behavior.
• The ability to form stable, size-controlled particles offers significant promise for advanced biological delivery systems.