Pii: S1010-7940(01)00714-x

Objective: Transient expression of therapeutic genes within lung allografts may modulate the pathological processes following allotransplantation. Whilst ef®cient gene transfer to lungs has been reported with viral vectors, their usefulness is limited on the grounds of safety. Since non-viral systems overcome many of these safety issues, our studies were designed to evaluate the ef®ciency of several non-viral gene delivery vectors for in vivo transfer of plasmid DNA to rat lungs via the airways. Methods: Fischer rats (230±260 g) underwent a thoracotomy, right main bronchus occlusion and instillation of 300 mg naked or complexed DNA (pCIluci, luciferase gene/CMV promoter) to the left lung followed by ventilation for 10 min. Rats were divided into ®ve treatment groups (n ˆ 5): (1) Glucose, (2) Naked DNA, (3) Linear polyethylenimine (PEI), (4) Branched PEI, (5) Lipid GL-67/DOPE and (6) DOTAP/cholesterol. Animals were sacri®ced 24 h after gene delivery for measurement of reporter gene activity and gas exchange of the left lung. Results: Linear PEI was the most ef®cient gene delivery vector and was signi®cantly better than DOTAP/cholesterol (P ˆ 0:00002) and naked DNA (P ˆ 0:004). All gene delivery vectors impaired function of the transfected left lung compared with DNA alone. Of all the gene delivery vectors tested, lipid GL-67/DOPE exerted the least effect on lung function whilst DOTAP/cholesterol mediated the most adverse effect. Conclusion: Linear PEI was the most ef®cient vector for gene delivery to rat lungs in our experimental setting although it mediated a moderate impairment in lung function. Further studies are needed to evaluate whether this effect is transient. q 2001 Elsevier Science B.V. All rights reserved.