In Silico Study of Ionizable Lipid Nanoparticles Using the SPICA Force Field.

Abstract

Lipid nanoparticles (LNPs), composed of ionizable amino lipids, phosphatidylcholines (PC) lipids, and cholesterol, have shown promise as delivery vehicles for therapeutic oligonucleotides in various applications, including cancer immunotherapies, cellular reprogramming, genome editing, and viral vaccines (e.g., COVID-19 vaccines). However, the molecular characterization of ionizable amino lipids and their assemblies, such as LNPs, both in silico and in vitro, remains in its early stages. In particular, in silico studies on LNPs to understand their nanostructure have been limited due to the need for accurate coarse-grained (CG) models. In this study, we expand the SPICA force field to develop a more reliable and accurate explicit CG model for investigating the structure and properties of model LNPs through in silico experiments. Using this CG model, we performed molecular dynamics simulations on LNP systems with varying helper lipids and pH conditions. Our results reveal bilayer structures with double-stranded DNA (dsDNA) sandwiched between closely apposed monolayers in LNPs at pH 4, while at pH 7, dsDNA molecules are embedded within amorphous domains inside the LNPs. These in silico-optimized microstructures align well with the experimental observations obtained from small-angle X-ray scattering and cryogenic transmission electron microscopy (cryo-TEM). Additionally, a detailed analysis of LNPs containing different helper lipids explains why replacing saturated PC lipids with unsaturated PC lipids enhances the DNA transfection activity. Overall, this study provides a robust CG model for in silico studies of LNPs and offers in-depth molecular-level insights to advance their design for improved stability and efficacy.