The aim of this study was to establish a nonviral method for gene deli
very to the rat kidney. To this purpose, a panel of reagents was teste
d, including a monocationic lipid, DOTAP, a polycationic lipid, DOGS (
or Transfectam), and three different forms of the cationic polymer pol
yethyleninine (PEI). A comparison among these compounds was performed
in vivo, using luciferase as reporter gene. Complexes containing 10 mu
g of DNA were injected into the left renal artery of rats and allowed
to remain in contact with the kidney for 10 min. Forty-eight hours la
ter, luciferase expression levels in kidney extracts were measured, Ki
dneys injected with DNA complexed to the branched, 25-kD PEI polymer (
PEI 25k) yielded activity levels significantly higher than control, sh
am-operated kidneys (2.7 X 10(4) vs, 5.2 X 10(3) RLU/kidney, respectiv
ely), whereas the other transfecting agents did not yield significant
activity over controls, PEI 25k was therefore chosen for further optim
ization of transfection conditions. Dose-dependent luciferase expressi
on was shown for 10, 50, and 100 mu g of PEI-complexed DNA, reaching m
aximal levels of 2.4 X 10(5) RLU/kidney at 100 mu g DNA, The optimal P
EI nitrogen/DNA phosphate molar ratio was 10 equivalents. Luciferase a
ctivity peaked at 2 days, was still significantly higher than controls
at 7 days, and was undetectable at 14 days post-injection. Using beta
-galactosidase (beta-Gal) as a reporter, transgene expression was loca
lized almost exclusively in proximal tubular cells.