. STEPHAN , ! ZHI - YONG YANG , » ' 2 HONG SAN , i ROBERT D . SIMARI , ' CARL J . WHEELER , 3 PHILIP L . FELGNER

An important goal of gene therapy for cardiovascular diseases and cancer is the development of effective vectors for catheter-based gene delivery. Although adenoviral vectors have proven effective for this purpose in animal models, the ability to achieve comparable gene transfer with nonviral vectors would provide potentially desirable safety and toxicity features for clinical studies. In this report, w e describe the use of a new cationic DNA-liposome complex using an improved expression vector and lipid, A^-(3-aminopropyl)A^^-dimethyl-2,3-bis(dodecyloxy)-l-propaniminium bromide/dioleoyl phosphatidylethanolamine (GAP-DLR I E / D O P E ) to optimize catheter-mediated gene transfer in porcine arteries. The efficiency of this vector was compared to D N A alone, D N A with a previously described cationic liposome complex, ( ± )-A^-(2-hydroxyethyl)-A^^-dimethyl-2,3-bis(tetradecyloxy)-l-propanaminium bromide ( D M R I E / D O P E ) , and a replicationdefective adenoviral vector in a porcine artery gene transfer model. W h e n used in optimal ratios, G A P D L R I E / D O P E liposomes provided a 15-fold higher level of gene expression in arteries compared to D N A alone or D M R I E / D O P E . Gene expression was observed in intimal and medial cells. However, when compared to adenoviral vectors (10^" pfu/ml), gene expression following G A P D L R I E / D O P E transfection was ~20-fold lower. Following intravenous injection of G A P D L R I E / D O P E in mice, biochemical, hematological, and histopathological abnormalities were not observed. Significant improvements in the efficacy of arterial gene expression can be achieved by optimization of transfection conditions with DNA-liposome complexes in vivo that m a y prove useful for arterial gene delivery in cardiovascular diseases and cancer. OVERVIEW SUMMARY INTRODUCTION GAP-DLRIE/DOPE, a new cationic liposome preparation, /^ ene transfer into the arterial wall has the potential to alter is an efficient liposomal vector that increases gene expresV J t h e biological response of vascular cells to growth stimusion in arteries compared to naked D N A or previously delation, lipid accumulation, coagulation, and inflammation, all scribed cationic DNA-liposome complexes hy more than 15of which may contribute to the development of cardiovascular fold. Although less efficient than adenoviral gene transfer, diseases. Large arteries also provide access to microvessels of these levels of gene expression represent a significant imorgans and malignant neoplasms for recombinant gene transprovement in liposome transfection in vivo and approach fer. An important objective of arterial gene transfer is the delevels observed vnth cUnically acceptable doses of adenovivelopment of effective vectors that optunize gene expression ral vectors. The improvement in gene expression, together and minimize toxicity. Previous studies have demonstrated the with the relative safety associated with liposomal gene transfeasibility of arterial gene delivery in vivo with viral and nonfer, suggests that such nonviral vectors may he appropriate viral vectors (for review, see Nabel, 1995). Side-by-side comfor' human gene therapy protocols which utilize catheterparisons of nonviral and viral vectors have not been made, and based sene delivery. " '̂ currently accepted fliat adenoviral vectors result in higher Departments of 'Internal Medicine, ^Pathology, 'Biological Chemistry, and ^Physiology and ^Howard Hughes Medical Institute, University of Michigan, Ann Arbor, MI 48109-0644. 3Vical Inc., San Diego, CA 92121.