Role of Cell Membrane-Vector Interactions in Successful Gene Delivery

Abstract

Cationic polymers have been investigated as nonviral vectors for gene delivery due to their favorable safety profile when compared to viral vectors. However, nonviral vectors are limited by poor efficacy in inducing gene expression. The physicochemical properties of cationic polymers enabling successful gene expression have been investigated in order to improve expression efficiency and safety. Studies over the past several years have focused on five possible rate-limiting processes to explain the differences in gene expression: (1) endosomal release, (2) transport within specific intracellular pathways, (3) protection of DNA from nucleases, (4) transport into the nucleus, and (5) DNA release from vectors. However, determining the relative importance of these processes and the vector properties necessary for optimization remain a challenge to the field. In this Account, we describe over a decade of studies focused on understanding the interaction of cationic polymer and cationic polymer/oligonucleotide (polyplex) interactions with model lipid membranes, cell membranes, and cells in culture. In particular, we have been interested in how the interaction between cationic polymers and the membrane influences the intracellular transport of intact DNA to the nucleus. Recent advances in microfluidic patch clamp techniques enabled us to quantify polyplex cell membrane interactions at the cellular level with precise control over material concentrations and exposure times. In attempting to relate these findings to subsequent intracellular transport of DNA and expression of protein, we needed to develop an approach that could distinguish DNA that was intact and potentially functional for gene expression from the much larger pool of degraded, nonfunctional DNA within the cell. We addressed this need by developing a FRET oligonucleotide molecular beacon (OMB) to monitor intact DNA transport. The research highlighted in this Account builds to the conclusion that polyplex transported DNA is released from endosomes by free cationic polymer intercalated into the endosomal membrane. This cationic polymer initially interacts with the cell plasma membrane and appears to reach the endosome by lipid cycling mechanisms. The fraction of cells displaying release of intact DNA from endosomes quantitatively predicts the fraction of cells displaying gene expression for both linear poly(ethylenimine) (L-PEI; an effective vector) and generation five poly(amidoamine) dendrimer (G5 PAMAM; an ineffective vector). Moreover, intact OMB delivered with G5 PAMAM, which normally is confined to endosomes, was released by the subsequent addition of L-PEI with a corresponding 10-fold increase in transgene expression. These observations are consistent with experiments demonstrating that cationic polymer/membrane partition coefficients, not polyplex/membrane partition coefficients, predict successful gene expression. Interestingly, a similar partitioning of cationic polymers into the mitochondrial membranes has been proposed to explain the cytotoxicity of these materials. Thus, the proposed model indicates the same physicochemical property (partitioning into lipid bilayers) is linked to release from endosomes, giving protein expression, and to cytotoxicity.

Figure 1:

A. Polyplexes are internalized into endosomes. B. Endosomal release is thought to be critical for enabling transport of DNA to the nucleus. C-D. Although polyplexes have been observed in internal organelles such as the Endoplasmic reticulum and Golgi apparatus, it is unclear if the polyplexes in these organelles enable gene expression.

Figure 2:

A. PEI enhanced the size of existing defects in supported lipid bilayers. B. Although many cationic polymers induced LDH release from KB cells, PEI induced significantly more LDH release at much lower concentrations. C. PEI also induced the most LDH release in Rat2 cells. D. Cationic materials could form three types of defects in lipid bilayers (i) free pores (ii) polymer supported pores (iii) a carpet of polymers or (iv) polymer intercalated bilayer.

Figure 3:

A. PEI exposure increased cell membrane currents of HEK 293A cells starting at 2.2 ug/mL and the current level stayed under −50 nA even at 113 ug/mL indicating that cells were still intact. Dissolution of cells using the detergent SDS resulted in currents > −70 nA. B. L-PEI exposure increased membrane currents at lower concentrations than G5 PAMAM. C. Free L-PEI had a higher partition constant into HEK 293A cell membranes than free G5 PAMAM. Partition constants of polyplexes were not different within error. D. Free cationic vectors also induced increased membrane currents in HeLa cells with L-PEI and jetPEI showing the greatest increase and showing the greatest gene expression. Currents induced by polyplexes did not predict gene expression.

Figure 4:

A. OMB employed for FRET studies. B. OMB delivered using jetPEI polyplexes (N:P 10:1) exhibits diffuse distribution at 4h. C. OMB delivered using G5 PAMAM polyplexes (N:P 10:1) were confined to vesicles with few cells showing diffuse OMB distribution. D. Treatment of cells with L-PEI for 1 h following 3 h G5 PAMAM polyplex incubation results in diffuse, cytosolic OMB distribution. E. JetPEI induces release of OMBs as early as 2h after delivery and OMBs persist in the cytosol till at least 8 h. F. The fraction of cells (n = total number of cells counted from 10 images each obtained from 3 biological replicates) showing diffuse OMB distribution in the cytosol (jetPEI = 88 ± 5 %, G5 PAMAM = 9 ± 2 %) corresponds to the fraction of cells expressing GFP (jetPEI = 80 ± 5 %, G5 PAMAM 4 ± 1 %). G. Pre-incubation, co-incubation and post-incubation of L-PEI with G5 PAMAM polyplexes increases both fraction of cells displaying diffuse OMB and fraction of cells expressing GFP by >10-fold.

Figure 5:

A. Free cationic polymer intercalates into the plasma membrane and/or is released from associated polyplexes. B. Cationic polymer is dispersed into internal cellular membranes via lipid recycling pathways. Polyplexes are endocytosed. C. Cationic polymer intercalated into endosomal membranes induces membrane permeability and facilitates release of genetic material into the cytosol. D. We hypothesize that cationic polymer associated with the nuclear membrane via lipid recycling and/or cytosolic pathways induces permeabilization and facilitates transport into the nucleus.