Design and gene delivery activity of modified polyethylenimines

Purpose: PEI is one of the most effective non-viral gene delivery polymers, but its efficacy, toxicity, and in vivo behavior depend strongly on molecular weight, branching, particle size, surface charge, and modifications. This review addresses how PEI/DNA complexes can be modified with targeting ligands, shielding agents, and functional groups to improve transfection, reduce toxicity, and enable targeted gene delivery in vivo.

Hypothesis: This is a review article and does not test a single formal hypothesis. Its central premise is that PEI possesses intrinsic DNA condensation and endosomal escape functions, and that additional modifications — especially ligand coupling and surface shielding — can add target specificity, improve biocompatibility, and enable systemic in vivo gene delivery.

Aims: Describe PEI synthesis, structure, and molecular species (linear and branched, various molecular weights). - Review DNA condensation, particle size, and surface charge of PEI/DNA complexes. - Discuss cellular uptake mechanisms: non-specific electrostatic interactions and receptor-mediated uptake using ligand-PEI/DNA complexes. - Review endosomal release, nuclear transport, and complex disassembly. - Summarize in vivo gene delivery by topical, systemic, and targeted applications. - Discuss future directions for modified PEI-based vectors.

Delivery system: Vector type: PEI/DNA polyplexes formed by electrostatic interaction between cationic PEI and anionic DNA. - Polymer forms: Linear and branched PEI; molecular weights ranging from 700 Da to 800 kDa; commonly 22 kDa linear and 25 kDa branched PEI. - Payload: Plasmid DNA; potentially oligonucleotides. - Targeting ligands: Transferrin, anti-CD3 antibody, galactose, mannose, EGF, RGD/integrin-binding peptides, asialoorosomucoid. - Shielding agents: PEG, poly-N-(2-hydroxypropyl)methacrylamide (pHPMA), poloxamer/Pluronic, and high ligand density (e.g., transferrin). - Functional mechanisms: Proton sponge for endosomal escape; ligand-receptor mediated endocytosis; shielding to reduce non-specific interactions and toxicity.

Approach: Review of in vitro and in vivo literature. In vitro studies include cell lines such as K562, melanoma cells, and others. In vivo studies include topical application (brain, kidney, lung, subcutaneous tumors) and systemic application in mice, including tail-vein injection and tumor-targeting models. No new primary experiments are reported.

Key methods: Particle size by dynamic laser light scattering and atomic force microscopy. - Surface charge by zeta potential. - DNA condensation and complex stability. - Transfection/gene expression assays. - Receptor competition assays and receptor up-regulation studies. - In vivo reporter gene expression, biodistribution, and toxicity. - Erythrocyte aggregation and complement activation assays.

Key results: PEI protonation: Only every fifth or sixth amino nitrogen is protonated at physiological pH; binding to nucleic acid shifts this to every second to third nitrogen, giving high buffer capacity over a broad pH range. - Particle size: PEI/DNA complexes form toroids of ~40–80 nm at low salt; branched PEI forms small-to-intermediate complexes at physiological ionic strength, while linear PEI22 forms large aggregates at physiological salt but small complexes in 5% glucose. - Transfection vs. size: For branched PEI800 and PEI25, larger particles correlated with higher transfection efficacy; small particles had significantly lower efficiency. - Surface charge: At N/P > 4, PEI/DNA complexes typically have zeta potential of +30 to +35 mV. - Ligand targeting: Transferrin-PEI(800) increased transfection up to several hundred-fold depending on cell type. Anti-CD3-PEI mediated specific delivery to CD3-expressing cells. RGD-PEI gave up to 100-fold enhancement, lost when RGD was replaced by RGE. - Shielding: PEGylation or high transferrin ligand density reduced zeta potential to near neutral, reduced erythrocyte aggregation and toxicity, increased circulation time, and enabled tumor-targeted gene expression after systemic application. - In vivo: Linear PEI22/DNA complexes gave high lung expression after systemic administration; branched PEI25 or PEI800 were less efficient and more toxic. Intratumoral linear PEI transfection reached up to 1% of cells, with expression persisting 15 days. PEGylated transferrin-PEI/DNA complexes were the first demonstration of tumor-specific targeted gene expression after systemic application.

Interpretation: A therapeutically applicable non-viral vector should combine DNA condensation/protection, specific delivery, cellular uptake, endosomal release, nuclear localization, and shielding from non-specific interactions. Modified PEIs can incorporate these functions as exchangeable functional blocks, enabling targeted gene expression with reduced toxicity. Combining passive and active targeting with transcriptional targeting using tissue-specific promoters should improve target specificity and therapeutic potential.

Limitations: This is a review, not a primary study; no new experimental data or meta-analysis. - Nuclear transport and complex disassembly mechanisms remain poorly understood. - Transfection efficiency is cell-cycle dependent and inefficient in non-dividing cells. - In vivo barriers include serum interactions, RES clearance, erythrocyte aggregation, complement activation, and extracellular matrix binding. - Toxicity remains a concern, especially for high-molecular-weight and highly positive complexes. - Clinical translation is limited; most discussed systems are preclinical.