Comparison of the endocytic properties of linear and branched PEIs, and cationic PAMAM dendrimers in B16f10 melanoma cells

Many different polymers and architectures are now being developed as polymer therapeutics and non-viral vectors for cytosolic delivery, and cationic dendrimers, and linear and branched poly(ethylenimine)s (PEIs) have been widely used. For rational design and safe transfer into the clinic, it is important to better understand the cellular pharmacokinetics of the carrier, even if this will likely change when it is conjugated to, or complexed with, a targeting residue or therapeutic payload. The aim of these studies was to compare binding, endocytic capture and intracellular trafficking of linear and branched PEIs (Mw 25,000 g/mol) and cationic PAMAM dendrimers (generations (gen) 2– 4) using B16F10 murine melanoma cells. FITC-dextran was used as a control for comparison. All polymers were first conjugated to Oregon Green (OG) and carefully characterised in respect of pH- and concentration-dependence of fluorescence. Throughout, non-toxic concentrations of polymer were used. Flow cytometry showed that all the cationic polymers were internalised by “adsorptive” endocytosis, with maximum uptake seen for PAMAM gen 4 >> branched PEI > linear PEI > PAMAM gen 3 > PAMAM gen 2. The PAMAM gen 4 uptake rate was 130 fold greater than seen for FITC-dextran. Branched PEI had the highest extracellular binding (accounting for > 50% of total cell-associated fluorescence) whereas for the linear PEI, binding was only 13% of the cell-associated fluorescence. Unlike FITC-dextran, all cationic polymers lacked significant exocytosis over the time period studied. Whereas PAMAM gen 4 and the branched PEI were predominately internalised by cholesterol-dependent pathways, internalisation of linear PEI appeared to be independent of clathrin and cholesterol. A perception of the rate and mechanisms of cellular uptake of these vectors will be important in the context of their proposed use as drug delivery systems.