Modular optimization of lipid nanoparticles improves brain transfection after traumatic brain injury

Introduction: : Traumatic brain injury (TBI) affects over 50 million people annually and is a leading cause of death and disability. Although TBI is a significant clinical problem, leading to neurological impairment that can persist for months to years, treatments are palliative, with no FDA-approved therapeutics that address the underlying pathophysiology. The delivery of nucleic acid cargo presents an opportunity to address some of the pathological pathways of the secondary injury that lead to neurodegeneration. Lipid nanoparticles (LNPs) are the most clinically advanced non-viral gene delivery platform, with benefits over their viral counterparts including their low immunogenicity and ease of modification. FDA-approved formulations include Onpattro, systemic siRNA delivery to the liver, and the SARS-CoV-2 vaccine, intramuscular mRNA delivery. An outstanding challenge for the application of LNPs to treat neurological disorders is delivery to the brain. In TBI there is evidence of transient BBB breakdown due to increased inflammation that can be leveraged to deliver nanoparticles, which passively accumulate in the injured tissue as seen with the enhanced permeation and retention (EPR) effect in tumors. This EPR-like effect has been reported for nanoparticles after intravenous administration up to 24 hours after TBI, and our group has shown that systemically administered LNPs can access the injured brain and mediate transgene expression. However, little is known about how timing of LNP dosing after injury can impact transfection in the brain. The goal of this work was to optimize LNPs for systemic mRNA delivery to the brain after TBI by optimizing PEG-lipid composition and delivery timing.

Materials and

Methods: : All protocols were carried out in accordance with the IACUC at UCSD. LNPs encoding for luciferase were formulated via microfluidic mixing with a lipid composition similar to those used in the Onpattro and COVID vaccine formulations. LNPs were made with 1.5% of either DSPE- or DMG PEG, and then were evaluated for their hydrodynamic size and polydispersity index using dynamic light scattering. mRNA encapsulation and yield were subsequently measured using a RiboGreen assay. Protein expression was assessed in HEK293T and Neuro-2a cells at multiple doses 24h post LNP-transfection. A controlled cortical impact (CCI) model of TBI was used in which a CCI was applied on the right hemisphere of the brain at 3m/s with a 2mm diameter probe tip under isoflurane anesthetic and post-operative buprenorphine analgesia. LNPs were delivered intravenously into mice (C57/BL6J, n=3 female) at a dose of 0.5mg/kg 1 hour after CCI and organs, including the brain, heart, lungs, liver, kidney, and spleen, were harvested 24 hours after injection. The brain was split into the injured hemisphere, contralateral uninjured hemisphere and hindbrain prior to analysis. To quantify the absolute amount of luciferase expression, organs were homogenized and subsequently measured for luciferase activity against a standard of recombinant luciferase protein. To then determine the spatial distribution after injury, LNPs were formulated with Cre mRNA and delivered in Ai9 mice (n=2 male, n=2 female) 24 hours after CCI. After 1 week, mice were perfused with formalin, and collected organs were sectioned, counterstained for Hoechst, and imaged for TdTomato+ cells.

Results, Conclusions, and Discussions:: Using luciferase as a reporter protein, we were able to successfully transfect cells in the injured brain microenvironment with LNPs formulated with either DSPE- or DMG-PEG delivered 1 hour after injury. Although the difference in PEG lipid did not significantly impact the formation and stability of LNPs (Fig 1A), choice of PEG-lipid impacted organ transfection. DMG-PEG LNPs had a significant increase in activity in the liver, and modest increases in the injured brain and spleen, while DSPE-PEG showed increases in heart, lung and kidney transfection (Fig 1B). DMG-PEG LNPs had high liver transfection, similar to what is known for the Onpattro formulation, which facilitates hepatocyte uptake of LNPs via ApoE binding in the blood. Increasing the length and concentration of PEG-lipids is known to improve accumulation in extrahepatic organs, as they reduce adsorption of blood proteins and extend circular half-life. This is reflected in our observations, as we see significant liver detargeting with the longer length DSPE-PEG LNPs, and therefore improved brain to liver ratio (Fig 1C). We evaluated cell types and distribution of transfection through analysis of tdTomato expression after delivery of Cre mRNA in Ai9 mice (Fig. 1D). When LNPs were delivered 24 hours after injury, we observed broad transfection of cells throughout the perilesional injury when LNPs were formulated with DMG-PEG vs. DSPE-PEG. Similar to what we observed for bulk luciferase analysis, there was significant liver detargeting with LNPs formulated with DSPE-PEG. In summary, we elucidated the transient time window during which systemically administered LNPs can traverse the damaged BBB to infiltrate into the injury site and provides insight on the role of PEG-lipids in delivery of LNPs to the brain. Future directions include adding targeting ligands to improve transfection of specific cells of interest after TBI. Our ultimate goal would be to extend this to a therapeutic cargo, such as BDNF mRNA to address neurodegeneration after TBI.

Acknowledgements and/or References (Optional)::