Development of novel cyclodextrins as non-viral gene delivery vectors and assessment of differentiated cell culture models to predict in vivo gene expression

Abstract

The aims of this thesis were to examine the use of differentiated cell culture systems in the in vitro assessment of non-viral gene delivery systems and to investigate the use of commercial and novel cyclodextrins as vectors for gene delivery. The transfection efficiency of a range of the most commonly used non-viral gene delivery vectors was tested using the pCMVluc reporter plasmid in the non-differentiated COS-7 cell line and in fully differentiated CaCo-2, Ht29GlucH and co-culture CaCo-2:Ht29GlucH monolayers. The cationic lipids, DOTAP and Lipofectin™, and the cationic polymers, PEI HMW and Superfect™, when complexed with pCMV/wc, produced high levels of transgene expression in all the models. While the most efficient individual GDVs were the same in all models, the rank order of their efficiency did change. While microgram quantities of luciferase per well were produced by the best vector systems in the COS-7 model, only nanograms were produced per filter in the differentiated systems. From confocal studies it was clear that this was due to decreased levels of transgene expression in each cell rather than a decreased number of cells being transfected. The binding and internalisation of pDNA by differentiated cells appeared to be less dependent on the presence of sulphated proteoglycans on the cell surface and either less dependent on active uptake processes or slower than in non-differentiated cells. Investigations into the effect of serum on transfection efficiency showed that its presence has less of an attenuating effect on the transfection of differentiated cells, than non-differentiated cells. The presence of mucus in the co-culture model, rather than acting as a barrier, was surprisingly found to enhance transfection efficiency, in order to determine the predictive ability of these differentiated cell culture models, an in vivo gut model for gene delivery was established and the best of the non-viral GDVs tested. PEI HMW proved superior to DOTAP complexes in transfecting in vivo but transfection levels achieved by PEI HMW/ pCMV/wc were low and erratic. From these results the CaCo-2 model appeared to better predict in vivo gene delivery to the gut lumen but both differentiated models, given their increased resistance to transfection, appeared to have potential as models for in vivo gene delivery to other organs of the body, e.g. the lung.