Chitosan structure related to transfection efficiency and colloidal stability

We systematically examined the influence of the molecular weight (MW) and level of deacetylation (DDA) of chitosan on transfection efficiency of chitosan-plasmid DNA polyplexes. A library of chitosans was produced with different DDAs (72% to 98% DDA) and MWs (10 to 150 kDa). Gene transfection of HEK 293 cells revealed an important coupling between DDA and MW of chitosan in determining transgene expression. Similar expression levels could be achieved by simultaneously lowering MW and increasing DDA or lowering DDA and increasing MW, suggesting a predominant role of particle stability, through co-operative electrostatic binding, in determining transfection efficiency. These results were confirmed by isothermal titration calorimetry and indicate than an intermediate stability of the complex is optimal, high enough to condense and protect DNA outside the cell and inside endosomo-lysosomal vesicles, but low enough to allow intracellular DNA release from the vector for gene expression. We found that chitosan can achieve levels of transgene expression as high as commercial phospholipid-based formulations and requires specific choice of DDA and MW. For instance, the most efficient chitosan formulation, 92–10-5 [DDA–MW–N:P ratio], resulted in transgene level of expression equivalent to LipofectamineTM and more than 10x higher than for 92–150-5 or 72–10-5. The formulation 92–10-5 was used to study the role of pH and serum on the transfection process. The cellular uptake of polyplexes into HEK 293 cells was found to increase at lower pH. Cell uptake was also found to increase in the presence of serum. One possible reason for this observation is that positively charged chitosan forms small complexes with the negatively charged serum proteins which can then be efficiently internalized. Optimal transfection was achieved when cells were incubated with polyplexes for a period of 8–24 hours in complete media (10% serum) at pH 6.5 and then replenished with complete media at pH 7.4 for an additional 24–40 hours. Finally, the influence of chitosan molecular weight on polyplexe colloidal stability in presence of bovine serum albumin (BSA) was studied by comparing formulations 92–10-5 and 92–150-5. Dynamic light scattering results show that low MW chitosan polyplexes are more prone to aggregation as compared to high MW formulation, the latter being stable even when BSA is present in amount equivalent to 10% serum. Interestingly, the formulation showing colloidal instability results in transgene expression levels in vitro that are at least one order of magnitude higher than for its colloidally stable counterpart. These results suggest that colloidal stability, even if it appears to be a desirable property of any polyplexe formulation, is not an absolute prerequisite for efficient transfection in vitro. The lower colloidal stability of 92–10 could possibly be related to its lower affinity to DNA that facilitates polyplexe destabilization/dissociation by competing polyanion that is believed to be involved in the DNA release required for transfection to occur. Taken together our results suggest that careful evaluation of physicochemical and cell biological behavior of chitosan-based gene transfer systems can provide valuable insights for therapeutically effective gene therapy systems.