Loading...

Optimization of Hybrid-Lipid Nanoparticles Delivering mRNA using Machine Learning Algorithms for Immunotherapeutic Applications

Rafiei, Mehrnoosh | 2025

0 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 58072 (48)
  4. University: Sharif University of Technology
  5. Department: Institute for Nanoscience and Nanotechnology
  6. Advisor(s): Shojaei, Akbar; Mashayekhan, Shohreh; Chau, Ying
  7. Abstract:
  8. The human’s immune system consists of two main components: innate immunity and adaptive immunity. The innate immune system serves as the first line of defense against infections and responds to threats through inflammation. Acute inflammation ceases rapidly after pathogen clearance; however, chronic inflammation can arise, leading to significant tissue damage. Microglia cells, play a crucial role in neurodegeneration and are promising therapeutic targets for chronic inflammation-related diseases. This study focuses on designing and developing lipid nanoparticle (LNP) systems to modulate immune responses by delivering mRNA to hyperactive microglia, particularly those in inflammatory states, using machine learning (ML) algorithms. A library of 216 LNP formulations with varying lipid compositions, N/P ratios, and hyaluronic acid (HA) modifications was developed and screened using high-throughput methods. The transfection efficiency of eGFP mRNA was evaluated on BV-2 microglial cell lines under different immune conditions, including resting and activated states (induced by LPS and IL4/IL13). ML-guided morphometric analysis revealed distinct microglial subphenotypes before and after transfection. Four supervised ML algorithms were assessed to predict transfection efficiency and phenotypic changes based on LNP design parameters. A multilayer perceptron (MLP) neural network emerged as the best model with a weighted F1-score metrics ≥ 0.8, effectively predicting responses in LPS-activated cells. The MLP model was validated by accurately predicting the performance of four new LNP formulations not included in the initial library, successfully delivering eGFP mRNA to LPS-activated cells. Furthermore, the delivery of IL10 mRNA using optimized HA-LNP2 formulations demonstrated significant suppression of inflammatory phenotypes. These formulations increased IL10 expression to levels exceeding 10,000 pg/ml while reducing TNF-α expression to 32 pg/ml. This research highlights the potential of using ML techniques to optimize lipid-based nanostructures for delivering anti-inflammatory mRNA (e.g., IL10) to reprogram microglial cells from pro-inflammatory to anti-inflammatory states, offering promising therapeutic strategies for neuroinflammatory disorders
  9. Keywords:
  10. Machine Learning ; mRNA Delivery ; Immunomodulation ; Hybrid-Lipid Nanoparticle ; Inherent Safety

 Digital Object List

 Bookmark

No TOC