Chemistry of Lipid Nanoparticles for RNA Delivery

Purpose: LNPs enabled the COVID-19 mRNA vaccines and the first siRNA-LNP drug, but there is still no one-size-fits-all LNP for every RNA therapeutic. A better chemistry-level understanding of LNP self-assembly, component roles, and whole-particle properties is needed to rationally design safer and more efficient RNA delivery systems.

3. Hypothesis: LNP performance is governed by the coordinated chemistry of ionizable lipids, sterols, phospholipids, and PEG-lipids, together with whole-particle properties such as size, charge, internal structure, lipid packing, membrane hydration, stability, and biomolecular corona. Rational tuning of these parameters can improve RNA delivery and expand therapeutic applications.

Hypothesis: LNP performance is governed by the coordinated chemistry of ionizable lipids, sterols, phospholipids, and PEG-lipids, together with whole-particle properties such as size, charge, internal structure, lipid packing, membrane hydration, stability, and biomolecular corona. Rational tuning of these parameters can improve RNA delivery and expand therapeutic applications.

Aims: Introduce LNPs, advantages/disadvantages of nucleic acid therapeutics, and rationale for LNP molecular makeup. - Describe theory and methods of LNP self-assembly and manufacturing. - Break down LNP composition: ionizable/cationic lipids, sterols, phospholipids, and PEG-anchored lipids, including design criteria and intracellular roles. - Review whole-LNP attributes: size, charge, internal structure, lipid packing, membrane hydration, stability, and affinity for biomacromolecules. - Discuss characterization techniques and remaining questions for LNP formulation and optimization.

Delivery system: Platform: Lipid nanoparticles (LNPs) with a homogeneous lipid core for RNA delivery. - Key components: Ionizable/cationic lipids (30–50%), sterols/cholesterol (20–50%), phospholipids (10–20%), PEG-anchored lipids (<2% typical for IV). - Payload: mRNA and siRNA; adaptable to other nucleic acids. - Representative lipids: DLin-MC3-DMA, SM-102, ALC-0315 (ionizable); DSPC, DOPE (phospholipids); cholesterol and analogues such as β-sitosterol; PEG-lipids. - Targeting: No intrinsic targeting ligand; ApoE-mediated hepatic uptake is common. Functionalized PEG-lipids can attach ligands; selective organ targeting (SORT) can be achieved with charged lipids.

Approach: This is a review/Account, not a primary experimental study. It synthesizes the authors’ work and the broader literature on LNP chemistry, self-assembly, formulation, characterization, and RNA delivery, with examples from in vitro and in vivo studies.

Key methods: Dynamic light scattering (DLS) for size and polydispersity. - Zeta potential for surface charge. - Cryo-TEM, SAXS/SANS, and NMR (e.g., ³¹P) for internal structure, lipid packing, and phase behavior. - Differential scanning calorimetry (DSC) for stability/thermal behavior. - In vitro transfection, gene silencing, cytotoxicity, and Gal8-GFP endosomal escape assays. - Biodistribution, cellular uptake, and ApoE/LDLR-dependence studies.

8. Key Results: - Ionizable lipids typically have tertiary amine headgroups and biodegradable ester linkers; cpKa ~9–10.5, LNP pKa ~6–7, and cLogD ~10–14 are associated with potency. - Cholesterol analogues can strongly affect delivery; one cited study showed up to 200-fold improvement in transfection with modified sterols. ApoE exposure can relocate cholesterol from the core to the shell. - DOPE vs DSPC changes LNP biodistribution and transfection: DOPE can favor hepatic accumulation, while DSPC can favor splenic accumulation in some systems. - PEG-lipids: <2% is common for IV; higher PEG can improve tumor accumulation but inhibit ApoE binding and cellular uptake. Anchor length controls PEG shedding. - Optimal LNP size is generally 20–200 nm; 45 nm was potent for subcutaneous delivery and 80 nm for IV delivery in mice, while nonhuman primates were less size-sensitive after intramuscular delivery. - Surface charge is considered weak between −20 and +20 mV. SORT: cationic lipids can direct transfection to lungs; anionic lipids to spleen. - mRNA-LNPs contain more water than siRNA-LNPs; low N/P ratios are more hydrated and can improve transfection.

Key results: - Ionizable lipids typically have tertiary amine headgroups and biodegradable ester linkers; cpKa ~9–10.5, LNP pKa ~6–7, and cLogD ~10–14 are associated with potency. - Cholesterol analogues can strongly affect delivery; one cited study showed up to 200-fold improvement in transfection with modified sterols. ApoE exposure can relocate cholesterol from the core to the shell. - DOPE vs DSPC changes LNP biodistribution and transfection: DOPE can favor hepatic accumulation, while DSPC can favor splenic accumulation in some systems. - PEG-lipids: <2% is common for IV; higher PEG can improve tumor accumulation but inhibit ApoE binding and cellular uptake. Anchor length controls PEG shedding. - Optimal LNP size is generally 20–200 nm; 45 nm was potent for subcutaneous delivery and 80 nm for IV delivery in mice, while nonhuman primates were less size-sensitive after intramuscular delivery. - Surface charge is considered weak between −20 and +20 mV. SORT: cationic lipids can direct transfection to lungs; anionic lipids to spleen. - mRNA-LNPs contain more water than siRNA-LNPs; low N/P ratios are more hydrated and can improve transfection.

Interpretation: LNPs are highly customizable, clinically validated RNA delivery vectors. Chemistry-guided design can extend RNA therapeutics beyond vaccines to cancer vaccines, protein replacement, and gene editing. The authors emphasize that LNP science is still emerging, with many unresolved structural and biological questions.

Limitations: Review article; no new primary experimental data. - Many mechanistic questions remain: internal LNP organization, stereochemistry effects, biomolecular corona control, and predictive stability testing. - Clinical success is still concentrated in hepatic delivery and intramuscular vaccines; extrahepatic and non-liver targeting remain challenging. - Cholesterol analogues may be limited by supply, cost, and batch variability. - No universal LNP formulation exists for all RNA cargoes, diseases, or administration routes.