Development of lipid membrane based assays to accurately predict the transfection efficiency of cell-penetrating peptide-based gene nanoparticles.

Abstract

Successful gene therapy requires the development of vectors that enable efficient delivery of genetic materials (e.g., pDNA or siRNA) to targeted cells, without degradation of the genetic materials. We have shown that nanoparticles formed by combining cell-penetrating peptide and pDNA (CPP-pDNA) into complexes and condensing them with calcium chloride can provide gene nanoparticles with high transfection efficiency and low cytotoxicity. In this work, we compare in situ measurements of the membrane insertion potential of three arginine-based gene nanoparticles (RW9-NPs, R9-NPs, and RH9-NPs) using four lipid compositions and two types of model membrane (Langmuir monolayers vs. supported bilayers) with their transfection efficiency in two human cancer cell lines. Using a Langmuir trough, we measured the membrane insertion potential of our gene nanoparticles to model membrane monolayers. A Quartz Crystal Microbalance with Dissipation (QCM-D) technique was used to monitor the adsorption of these nanoparticles to lipid bilayers of various compositions. Finally, gene expression using these nanoparticles was measured in breast cancer and cervical cancer cell lines. Our cell culture studies indicate that although R9-NPs and RW9-NPs show a significant increase in transfection efficiency compared to free pDNA, RH9-NPs do not show any significant difference. Both the Langmuir monolayer and QCM-D bilayer studies show that these results are best reflected in the in situ measurement assays when lipid systems containing a mixture of phospholipids, cholesterol, and sphingolipids are used. It is important to note that the mechanism of penetration is expected to differ for RW9 vs. R9; however, gene nanoparticles containing either of these CPPs show similar transfection efficiency. Our results therefore demonstrate that the design of predictive assays for gene therapy using CPPs must involve carefully chosen model lipid membrane systems that accurately represent the varying compositions of cell membranes.