Poster O4 - Aromatic ionizable lipid nanoparticle platform for mRNA vaccine delivery with reduced off-target liver delivery

Introduction: Lipid nanoparticles (LNPs) are a clinically advanced mRNA vaccine platform, as demonstrated by the COVID-19 mRNA vaccines, as they can robustly traffic into lymph nodes and elicit a strong immune response. However, these LNPs also highly distribute into the liver, leading to hepatic toxicity.1 Furthermore, lipid tail length and regiochemistry can functionally change how the lipids interact with mRNA, yet these aspects of lipid design have not yet been widely explored, and more broadly, the relationship between lipid design and LNP performance is poorly understood.

Here, we synthesized a library of 14 novel ionizable lipids (ILs) with bioreducible disulfide bonds and varied tail lengths and regiochemistries. We hypothesize that these moieties will reduce lipid toxicity while enhancing transfection, as LNPs with aromatic rings have demonstrated significantly improved transfection in vivo when compared to LNPs without aromatic rings.2 These ILs further enable us to investigate the structure-function relationship between IL tail length, regiochemistry, and transfection in lymph nodes and the liver.

We administered LNPs utilizing our ILs and gold standards C12-200 and SM-102 to mice intramuscularly and found that ILs with a 6-10 carbon tail length in the meta and para positions outperformed C12-200 and SM-102 in achieving a high lymph node to liver delivery ratio. Further, LNPs containing ILs with disulfide bonds and aromatic rings outperformed similar ILs without these moieties in achieving high lymph node and decreased liver delivery, indicating that aromatic rings and disulfide bonds may play a central role in retaining LNPs in the endothelium.

Materials and

Methods: Lipid purification and characterization: Synthesized lipids were purified via flash chromatography and characterized via NMR and LCMS.

LNP formulation: All LNP organic phases except for SM-102 were formulated using an IL, 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), cholesterol, and 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (PEG 2000) in ethanol at a molar ratio of 35:16:46.5:2.5, respectively. SM-102 LNPs were formulated using 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), cholesterol, and 1,2-dimyristoyl-rac-glycero-3-methoxypolyethylene glycol-2000 (DMG-PEG 2000) in ethanol at a molar ratio of 50:10:38.5:1.5, respectively. The aqueous phase was prepared by diluting firefly luciferase (N1pseudoU) mRNA to 0.074 mg/mL in 10 mM citrate buffer (pH 3). Aqueous and organic phases were combined using a microfluidic mixing device with a staggered herringbone pattern at a volume ratio of 3:1, respectively, to form LNPs. LNPs were then dialyzed in 20 kDa MWCO dialysis cassettes for 2 h in 1x PBS. They were then sterile filtered using a 0.22 µm syringe filter.

LNP characterization: LNP size and PDI were measured via a DynaPro Plate Reader III and zeta potential was measured via a Zetasizer Nano. mRNA encapsulation efficiency was measured using a RiboGreen kit according to the manufacturer’s protocol.

Animal work: C57BL/6J mice were injected intramuscularly in triplicate into their right thigh muscle with LNPs at 0.05 mg/kg body weight. At 6 h post-injection, mice were intraperitoneally injected with D-luciferin potassium salt at 150 mg/kg body weight. The heart, lungs, liver, spleen, kidneys, and iliac and inguinal lymph nodes were then harvested, and bioluminescence imaging was performed using IVIS. Total bioluminescent flux was quantified using Living Image software.

Results, Conclusions, and Discussions: IL synthesis: Novel ILs with an aromatic ring and differing carbon tail lengths and regiochemistries were successfully synthesized and purified (Fig 1A). ILs without an aromatic moiety (and resulting distinct regiochemistries) were synthesized and purified as controls (Fig. 1A). All LNPs are named according to their structure as follows “C[number of carbons on tail] [regiochemistry, if applicable].”

LNP formulation: Most LNPs had a generally high encapsulation efficiency (EE) between 80-95%, although some LNPs (C6, C12, C12 para) had slightly lower encapsulation efficiencies, potentially indicating that ILs that are too hydrophilic or hydrophobic are unideal candidates for formulating LNPs (Fig. 1C).

LNP delivery of mRNA in vivo: LNPs with ILs with shorter carbon tail lengths (6-8 C) increased transfection in the lymph nodes and liver when an aromatic ring and disulfide bond was added, indicating that these moieties may aid LNPs in successfully transfecting cells overall (Fig 1D, E). LNPs with shorter tail lengths (6-8 C) perform 10-100 fold better than LNPs with lipids with longer tail lengths (10-12 C), indicating that increasing hydrophobicity of these lipids may be detrimental to LNP transfection efficacy. Similarly, lipids with 12 C tails tended to have a lower encapsulation efficiency of 65-80%. There are no significant trends across different tail lengths in transfection efficacy for differing regiochemistry, indicating that regiochemistry may not be as strong of a driving factor in LNP transfection in vivo than the hydrophobicity induced by increasing carbon tail lengths.

All groups had significantly decreased liver transfection and no significant difference in lymph node transfection when compared to SM-102. Furthermore, the ratios of combined iliac and inguinal lymph node luminescent flux to liver luminescent total flux of lipids C6 para and C10 meta were significantly greater (between 10-100 fold) than that of C12-200 and SM-102. These results indicate that these lipids are better candidates for targeting lymph nodes, and subsequently, may be stronger LNP vaccine candidates. More lipids are currently being synthesized, as the project is ongoing. Moving forward, we plan to conduct a COVID-19 vaccine study with top performers to further explore these lipids in a more translational model.

Acknowledgements (Optional):