Toxicity of cationic lipids and cationic polymers in gene delivery

Purpose: Cationic lipids and polymers are promising non-viral gene delivery vectors because they are non-immunogenic, easy to produce, and not oncogenic, but toxicity remains a major obstacle to their clinical application. The review evaluates structural features of cationic compounds and summarizes toxicity–structure relationships to guide the development of safer vectors.

Hypothesis: As a review, no formal experimental hypothesis is tested. The central premise is that the toxicity of cationic lipids and polymers is largely determined by their cationic nature and specific structural domains—hydrophilic headgroup, hydrophobic chain, and linker—and that rational modifications such as heterocyclic headgroups, biodegradable linkages, low-molecular-weight or acid-labile polymers, PEGylation, and co-lipids can reduce toxicity while maintaining transfection efficiency.

Aims: Review the toxicity of cationic lipids and cationic polymers used in gene delivery. - Evaluate structural features of cationic compounds and summarize toxicity–structure relationships. - Discuss strategies to reduce toxicity, including biodegradable linkers, heterocyclic headgroups, low-molecular-weight polymers, acid-labile linkages, PEGylation, and co-lipids/co-polymers. - Provide suggestions for developing new cationic compounds with high transfection efficiency and low toxicity.

Delivery system: Cationic lipids: DOTMA, DOTAP, CTAB, LHON, pyridinium lipids (e.g., 2Oc), imidazolium lipids, guanidinium-containing lipids, cholesterol derivatives, galactosylated cholesterol derivatives, carbamate-linked lipids. Neutral co-lipids: DOPE, DOPC, cholesterol. Cationic polymers: PEI (linear and branched), PLL and dendritic PLL (KG6), chitosan, dextran-spermine (D-SPM), pDMAEMA, pDAMA, pHPMA-DMAE, reducible polycations (RPC), poly[α-(4-aminobutyl)-L-glycolic acid] (PAGA), PEI-Chol lipopolymer. Payloads: plasmid DNA, oligonucleotides, ribozymes. Modifications discussed: PEGylation, targeting ligands (e.g., transferrin), acid-labile imine linkers, β-cyclodextrin-linked low-molecular-weight PEI, amphiphilic PLL, water-soluble lipopolymers.

Approach: Review article; no original experimental groups, n, doses, or controls. It synthesizes published in vitro and in vivo studies on toxicity and transfection, including cell lines such as Hep G2, mouse fibroblasts, cancer cell lines, cultured neurons; in vivo models include mouse lung, central nervous system, and CT-26 tumor-bearing BALB/c mice.

Key methods: Summarizes techniques from cited studies: cell viability/cytotoxicity assays, transfection efficiency measurements, DNase I sensitivity assays, protein kinase C (PKC) inhibition assays, histopathological assays, and in vivo toxicity/transfection evaluations.

Key results: Quaternary ammonium amphiphiles are more toxic than tertiary amine counterparts. - Single-tailed cationic lipids are generally more toxic and less efficient than double-tailed lipids, but LHON (single-tailed) was more efficient and less cytotoxic than DOTAP. - Cholesterol derivatives with steroid backbones are more potent PKC inhibitors and more toxic than straight-chain analogues. - Heterocyclic headgroups (pyridinium, imidazolium) spread the positive charge and reduce cytotoxicity; pyridinium lipid 2Oc transfected cancer cell lines with similar or better efficiency than DOTAP and lower cytotoxicity. - Ether-linked lipids are stable, poorly biodegradable, and toxic; ester- and carbamate-linked lipids are more biodegradable and less cytotoxic. Carbamate-linked lipids are stable at neutral pH but acid-hydrolyzed in endosomes. - Free liposomes contribute to toxicity; removing free liposomes from high-charge-ratio lipoplexes reduced toxicity without decreasing transfection efficiency. - DOPE-containing liposomes showed low toxicity and high transfection efficiency in Hep G2 cells; co-lipids (DOPE, DOPC, cholesterol) reduce toxicity and improve transfection. - PEI: high transfection efficiency but high cytotoxicity. Low-molecular-weight (10 kDa) moderately branched PEI gave efficient delivery with low toxicity. Free PEI caused immediate toxicity via serum protein/RBC interaction; PEI/DNA complexes caused delayed toxicity from released free PEI. Acid-labile PEI was much less toxic than PEI 25 kDa. PEGylation and transferrin targeting increased transfection 5-fold with lower toxicity. Branched PEI 25 kDa at N/P = 10 caused death of all injected mice, whereas fully deacetylated PEI 25 kDa showed high transfection and low toxicity. PEI-Chol lipopolymer was non-toxic to several cell types. - PLL: higher molecular weight increases DNA condensation and transfection but also toxicity. Dendritic PLL KG6 showed high transfection without significant toxicity. Amphiphilic PLL with PEG and palmitoyl groups reduced toxicity without compromising gene delivery. - Cationic polysaccharides: chitosan and dextran-spermine showed low toxicity. Chitosan and trimethylated chitosan oligomers were non-toxic, whereas DOTAP decreased viability to 50%. D-SPM showed mild muscle toxicity, no liver/lung abnormalities, and no systemic toxicity in mice. - Other polymers: pDMAEMA had similar cytotoxicity to PEI; pHPMA-DMAE exhibited very low toxicity; reducible polycations facilitated intracellular nucleic acid release and reduced cytotoxicity; PAGA had significantly reduced cytotoxicity but low transfection efficiency.

Interpretation: Toxicity of cationic vectors is mainly determined by their cationic nature, but structural modifications can lower toxicity while maintaining or improving transfection. Heterocyclic cationic lipids, biodegradable linkages (ester, carbamate), low-molecular-weight or acid-labile polymers, PEGylation, co-lipids, and natural polysaccharides are promising strategies. Water-soluble lipopolymers combining polymer and liposome advantages are considered an encouraging design direction. Cationic polysaccharides may be widely used in the future due to their high biodegradability and low toxicity.

Limitations: Review only; no new primary data. Many studies are in vitro, and in vivo toxicity data remain limited. Structure–toxicity relationships are not fully understood, especially for the hydrophobic chain. Toxicity is dose-, charge-ratio-, cell-type-, and formulation-dependent. Long-term toxicity, immunogenicity, and clinical translation are unresolved. No large-animal or clinical validation is reported.