Abstract
Safe and functional delivery of delicate mRNA molecules to target tissues is a crucial step in the development of effective vaccines and therapeutic interventions. Lipid nanoparticles (LNP) are the most clinically advanced delivery vehicles for mRNA drugs and crucially rely on ionizable cationic lipids. However, the structure-function relationships between ionizable lipids and efficient in-vivo mRNA delivery remain poorly understood. Here, we focus on the rational design and sequential structural optimization of our ionizable lipids that performed well in vitro, but not in vivo. Through two distinct iterative optimization cycles-targeting the lipid tail and the headgroup-we studied how fusogenicity and pKa of ionizable lipids contribute to LNP delivery performance, in vivo. By engineering lipids with both unsaturated tails and more hydrophobic amino headgroups, we achieved both significant improvement of protein expression in vitro, reduced hemolysis risk, and more than 200-fold improvement of in vivo mRNA delivery. When compared head-to-head to a market-approved LNP benchmark, the newly developed ionizable lipids/LNP resulted in equally highly efficient in vivo mRNA delivery, with strong liver and spleen tropism upon intravenous injection, while matching the safety of the approved platform. Our findings are pivotal for the development of next-generation mRNA-LNP therapies and vaccines.