Peptide-conjugated lipid nanoparticles for targeted mRNA delivery to brain endothelial and neuronal cells

Introduction: Neurological disorders such as neurodegenerative disease, brain cancer, traumatic brain injury, and stroke affect millions of people in the US, yet lack sufficient treatments. Nucleic acid therapies, including messenger RNA (mRNA) protein replacement and gene editing therapies, have great potential in treating these disorders. Lipid nanoparticles (LNPs) are currently the most clinically advanced delivery vehicle for nucleic acid therapeutics. However, a major hurdle in the systemic delivery of mRNA and their delivery vehicles to the brain is the blood-brain barrier (BBB), a highly selective diffusion barrier made of endothelial cells, basement membrane, pericytes, and astrocytes. Thus, there is an urgent need to design novel LNPs that cross the BBB and transfect target cells of interest in the brain parenchyma. In this work, we investigated the potential of targeting peptides to improve mRNA LNP delivery to the brain. We utilized simple click chemistry to conjugate various peptides that target receptors overexpressed on brain endothelial cells (BECs) and neurons onto LNPs (Fig. 1A). We then explore the effect of targeting on transfection of BECs and neuronal cells in multiple in vitro assay systems, including serum and serum-free media, co-culture transwell systems, and exosome isolation. Finally, we show that targeted LNPs improve mRNA transfection in the brain in vivo after systemic administration.

Materials and

Methods: We selected four peptides for their brain targeting ability: RVG29 derived from the rabies virus glycoprotein which targets the nicotinic acetylcholine receptor (nAChR), T7 derived from transferrin which targets the transferrin receptor, and angiopep-2 and mApoE which target low density lipoprotein receptor-related protein 1 (LRP-1). Untargeted LNPs containing ionizable lipid, phospholipid, cholesterol, and lipid-PEG were formulated via microfluidic mixing with luciferase or mCherry mRNA. Targeted LNPs were formulated by adding lipid-PEG-maleimide as a fifth component, and then incubated with peptides synthesized with a terminal cysteine to allow for a thiol-maleimide click chemistry reaction. LNP size was characterized by dynamic light scattering, charge by zeta potential measurements, and encapsulation efficiency by RiboGreen assay. In vitro transfection assays were conducted on hCMEC/D3 brain endothelial cells and SH-SY5Y neuronal cells. In vivo luciferase mRNA delivery via systemic administration of LNPs was assessed in healthy adult mice.

Results, Conclusions, and Discussions: We conjugated the peptides to LNPs made with various PEG-lipid-maleimide substitutions, under the hypothesis that different peptides would have optimal substitution percentages given their varying physicochemical properties. Indeed, we saw that varying lipid-PEG-maleimide substitutions resulted in varying LNP sizes (Fig. 1B). Next, we screened the targeted LNPs on BECs and neuronal cells in vitro for luciferase transfection. We conducted this in both regular serum-containing growth media and serum-free media to study the effect of serum proteins on transfection. We then tested several lead targeted LNPs in vivo via systemic administration, and saw improved transfection in the brain compared to untargeted controls (Fig. 1C).
We also wanted to conduct more advanced in vitro screening to further study the mechanism of delivery. First, we conducted a co-culture transwell study with BECs grown on the transwell insert and neuronal cells in the basolateral compartment and saw that LNPs which crossed the BEC monolayer were able to further transfect the neuronal cells (Fig. 1D). Second, we treated BECs with targeted LNPs, and later collected the media and isolated exosomes under the hypothesis that neuronal transfection after systemic administration may be mediated by BEC-secreted exosomes. We then treated neuronal cells with the collected exosomes and saw transfection (Fig. 1E). Transfection was especially high from the targeted LNPs made with a 10% lipid-PEG-maleimide substitution, further supporting our finding that the substitution percentage is a critical parameter in designing targeted LNPs.
Future work should include deeper investigation into the mechanism for targeted LNPs crossing the BBB, such as whether the LNPs themselves or exosomes still retain the peptides on the surface. We are also currently investigating which specific brain cell types the LNPs are reaching in vivo. This work lays the groundwork for improving targeted mRNA delivery to the brain, which would be useful in therapeutic applications aiming to increase the absolute amount of delivery to the brain, as well as reducing off-target delivery to other organs.

Acknowledgements (Optional):