Poster X8 - Evaluating methods of formulating peptide-targeted lipid nanoparticles

Introduction: Lipid nanoparticles (LNPs) are currently the most clinically advanced non-viral gene therapy vector. Current technology is efficient at targeting the liver after systemic administration, however there is interest in developing LNPs for cell specific delivery in extrahepatic organs. Targeting ligands include antibodies, aptamers, peptides, and small molecules. Peptides are potentially attractive ligands because they can represent a large diversity of binding motifs, can be synthesized inexpensively, and they are relatively shelf stable. The low unimolecular affinity of peptides can be overcome through multivalent presentation, for example on the surface of an LNP. Several studies have demonstrated the potential for peptides to improve LNP targeting to specific cell populations, including hematopoietic progenitor cells in the bone marrow, photoreceptors in the eye, and hepatic stellate cells in the liver. Considering the importance of peptides as a class of molecules for LNP targeting, in this work, we compared two different synthesis methods of peptide targeted LNPs: in-line formulation, where peptide-lipid conjugates are synthesized and directly used for LNP formulation, and post-conjugation, where the LNPs are formulated with chemical handles that are subsequently modified with peptides. We compared the physicochemical properties of these LNPs, their binding and efficacy in vitro, and their activity in vivo after systemic administration.

Materials and

Methods: LNPs were formulated in a staggered herringbone microfluidic mixer using a 3:1 ratio of lipids in ethanol and nucleic acids in aqueous solution. In the in-line method, we synthesized polyethylene glycol (PEG) lipids modified with peptides. These peptide-PEG lipids were added to the ethanol stream of the microfluidic device to formulate LNPs. In the post-conjugation method, LNPs were formulated with PEG lipids with maleimide groups. After formulation, cysteine-labeled peptides were subsequently added for conjugation. All LNP formulations were purified using centrifugal filters. The physicochemical properties of these LNPs—hydrodynamic diameter, polydispersity index, and zeta potential—were measured using dynamic light scattering. In order to evaluate binding, internalization and activity, we used cyclic RGD as a model peptide ligand. We compared the cellular interaction of the in-line and post-conjugation LNPs in integrin-expressing MDA-MB-435 cells at 4°C to evaluate binding and at 37°C to evaluate internalization. Binding and internalization were evaluated via flow cytometry and microscopy. Transfection efficiencies of in-line vs. post-conjugation LNPs were quantified by formulating LNPs with mRNA encoding luciferase and measuring luciferase activity. Finally, the biodistribution and transfection efficiency in vivo was measured via DiR labeling and luciferase measurement, respectively.

Results, Conclusions, and Discussions: Our studies demonstrated that the in-line and post-conjugation formulations resulted in LNPs of similar physicochemical properties for small peptides. LNP hydrodynamic diameters ranged from 60–100 nm and polydispersity indices from 0.05 to 0.2. Only RVG, our largest peptide at 29 residues in length, resulted in the aggregation of particles when formulated by in-line but not post-conjugation. Zeta potentials were near neutral for all peptides except RGD and cRGD, which were slightly negative. In-line and post-conjugation cRGD LNPs demonstrated similar degrees of binding to and uptake in cells, measured by microscopy and flow cytometry. However, post-conjugation cRGD LNPs had a two-fold greater transfection efficiency when compared to in-line LNPs at matched mRNA dose. We conclude that in-line formulations may lead to disruption of LNP structure that results in changed endosomal escape efficiency. These results provide a methodology to formulate peptide targeted LNPs, and suggest that the method of incorporating peptide ligands in LNPs has a potential impact on transfection efficiency. Future studies are focused on evaluating differences in endosomal escape of in-line and post-conjugation LNPs in vitro and impacts on in vivo transfection efficiency.

Acknowledgements (Optional): The authors acknowledge the use of facilities and instrumentation supported by the National Science Foundation through the University of California San Diego Materials Research Science and Engineering Center DMR-2011924.

This work was performed in part at the San Diego Nanotechnology Infrastructure (SDNI) of UCSD, a member of the National Nanotechnology Coordinated Infrastructure, which is supported by the National Science Foundation (Grant ECCS-2025752).

Katelyn Miyasaki acknowledges support from: NHLBI training program Integrative Bioengineering of Heart Vessels Blood (T32 HL105373).