A Combinatorial Polymer Library Approach Yields Insight into Nonviral Gene Delivery

Purpose: Gene therapy has enormous therapeutic potential, but delivery remains the central bottleneck. Viral vectors are efficient but carry serious safety, immunogenicity, manufacturing, and cargo-capacity limitations. Nonviral vectors are safer and more scalable but typically orders of magnitude less effective. This Account describes a combinatorial polymer library strategy to discover biodegradable synthetic polymers that deliver DNA with viral-like efficacy.

Hypothesis: Systematically varying the structure of biodegradable poly(β-amino esters) (PBAEs) — backbone, side chains, molecular weight, and terminal groups — will reveal structure/function relationships that enable DNA condensation, cellular uptake, endosomal escape, DNA unpacking, and high gene expression with low cytotoxicity.

Aims: Synthesize large, structurally diverse libraries of PBAEs using high-throughput combinatorial chemistry. - Screen polymers for DNA binding, nanoparticle formation, transfection efficacy, and cytotoxicity. - Identify key polymer structural features that govern nonviral gene delivery. - Iteratively optimize lead polymers via end-modification and nanoparticle coating. - Evaluate lead PBAE/DNA nanoparticles in vitro and in vivo, including cancer models and genetic vaccines.

Delivery system: Polymer class: Biodegradable cationic poly(β-amino esters) (PBAEs), synthesized by conjugate addition of amines to diacrylates. - Lead polymer: C32 = 1,4-butanediol diacrylate-co-5-amino-1-pentanol; linear, ~10 kDa, hydroxyl side chains, tertiary amines, primary amine end groups after diamine end-modification. - Payload: Plasmid DNA (e.g., GFP, luciferase, diphtheria toxin suicide gene). - Nanoparticle type: Polymer/DNA polyplexes, typically <200 nm, stable in serum, near-neutral ζ potential in serum-containing media. - Targeting/coating: Electrostatic coating with anionic ligand-containing peptides, e.g., poly(glutamic acid)-polyglycine-RGD, for ligand-specific delivery. - Other formulation: PBAE-containing PLGA microparticles for genetic vaccines.

Approach: In vitro: COS-7 cells for high-throughput screening; human umbilical vein endothelial cells (HUVECs) as hard-to-transfect primary cells; serum-free and serum-containing conditions. - In vivo: Prostate cancer xenografts, ovarian cancer mouse models, intramuscular/intratumoral/intraprostate/intraperitoneal/intravenous routes; genetic vaccine models. - Controls/comparators: 25 kDa PEI, jet-PEI, Lipofectamine 2000, adenovirus/lentivirus, naked DNA, buffer, scrambled RGD peptide. - Iterative design: First-generation library → second-generation library → end-modified PBAEs → coated nanoparticles.

Key methods: High-throughput parallel synthesis and screening. - Electrophoretic DNA-binding assay. - Dynamic light scattering and zeta potential for particle size/surface charge. - TEM for nanoparticle morphology and serum stability. - Transfection assays: GFP fluorescence, luciferase expression. - Cellular uptake measurements. - Dual-labeled pH-sensitive/pH-insensitive DNA assay to measure endosomal pH environment. - In vivo gene expression, biodistribution, tumor growth, and apoptosis assays. - Cytotoxicity assays.

Key results: Library scale: >2000 structurally unique PBAEs; 2350-member library synthesized; 46 polymers transfected as well as or better than PEI; second-generation library of 486 polymers. - Lead structural features: Best PBAEs were linear, ~10 kDa, made from amino alcohols with 3–5 linear carbons and diacrylates with 4–6 interior carbons, at amine/acrylate ratio ~1.2:1; hydroxyl side chains and primary/secondary amine end groups improved delivery. - Nanoparticle properties: Lead PBAEs formed small (~200 nm), stable nanoparticles with near-neutral ζ potential in serum; diamine end-modified C32 bound DNA more tightly, formed ~30% smaller particles, and increased cellular DNA uptake up to 5-fold. - In vitro efficacy: C32-103 and C32-117 achieved gene delivery comparable to adenovirus/lentivirus in HUVECs and were ~2 orders of magnitude more effective than 25 kDa PEI, with ~2 orders of magnitude lower toxicity. - In vivo efficacy: Intratumoral C32 was 4-fold better than jet-PEI; C32/diphtheria toxin prevented tumor growth and caused 40% tumor regression; intraprostate injection caused apoptosis in 80% of tumor cells. Intraperitoneal end-modified C32 gave 175–500-fold higher expression than buffer, 4–12-fold higher than regular C32, and 15–42-fold higher than jet-PEI. C32-117 expression was detected at 2 months. - Biodistribution: End-modification altered tissue distribution; in an ovarian cancer model, C32-117 produced >100-fold higher tumor expression than unmodified C32. - Vaccines: PBAE/PLGA microparticles increased macrophage gene delivery by 3–5 orders of magnitude and caused antigen-specific rejection of transplanted syngeneic tumor cells.

Interpretation: The combinatorial polymer library approach successfully elucidated PBAE structure/function relationships and yielded biodegradable polymeric gene delivery nanoparticles with viral-like efficacy in vitro and promising in vivo activity. Lead PBAEs are effective, low-toxicity nonviral vectors and may set a new benchmark for polymeric transfection. The strategy is generalizable to other biomaterial design problems.

10. Limitations (optional; not formally stated as such): - The paper is an Account/summary rather than a single primary study, so detailed experimental protocols and statistical rigor are not fully presented. - In vivo efficacy is strongest for local administration; systemic intravenous delivery showed lower overall expression and primarily spleen/liver localization. - Serum protein interactions, complement activation, targeting specificity, and long-term safety remain challenges for clinical translation. - No clinical data or large-animal validation are reported.