New born mice were used to co-transfect 0.2 µg of plasmid DNA and 0.05 pmol to 0.5 pmol of siRNA into a total volume of 2 µl by stereotaxic intra-cerebroventricular injections. The optimal ratio of jetSI 10 mM : DOPE was 1:2.

BACKGROUND: Efficient in vivo vectors are needed to exploit the enormous potential of RNA interference (RNAi). Such methods require optimisation for specific delivery routes, tissues and usages. We tested the capacity of different non-viral vectors and formulation methods for inhibition of exogenous (luciferase) gene expression when used to introduce small interfering RNA (siRNA) into the mouse brain in vivo. METHODS: Polyethylenimine (PEI)-based polyplexes and JetSI (a mixture of cationic lipids)-based lipoplexes were used to vectorise plasmid DNA encoding the firefly Photinus pyralis luciferase gene and picomolar amounts of siRNA directed against this gene. Two controls were used, DNA encoding an unrelated luciferase from Renilla reniformis and a mutated siRNA sequence. RESULTS: First, we found that linear PEI, although efficient for delivering nucleic acids to cells, did not permit development of siRNA activity within the dose range tested (<0.5 pmol). Second, various combinations of cationic lipids were tried and the best formulation was found to be a combination of JetSI with the fusogenic lipid dioleoylphosphatidylethanolamine (DOPE). Efficient inhibition of target, firefly luciferase was obtained with exceedingly low amounts of siRNA: 78 +/- 6% inhibition at 24 h post-transfection with 0.2 pmol siRNA. This inhibition was dose-dependent and specific. No effect was seen on the control gene, co-transfected Renilla luciferase, and the control mutated siRNA sequence had no effect on the targeted firefly luciferase. CONCLUSIONS: We have optimised an efficient cationic lipoplex method for delivery of siRNA into the newborn mouse brain. Specific inhibition of exogenous target gene expression is obtained with picomolar amounts of siRNA.