Human papillomavirus (HPV) infection is the most important risk factor for the development of cervical cancer.The oncogene E7 from high-risk HPV strains has the ability to immortalize epithelial cells and increase cellular transformation in culture. In this study, we explored the possibility of preventing cervical cancer growth by inhibiting HPV16 E7 expression through gene transfer of an antisense construct. A recombinant adeno-associated virus (rAAV) vector was chosen for the transfer, based on its transfection efficiency, in vivo stability, and lack of detectable pathology. In vitro transfer of an rAAV vector expressing antisense HPV16 E7 (AAV-HPV16E7AS) inhibited cell proliferation, induced apoptosis, reduced cell migration, and restrained in vivo proliferation of HPV16/HPV18^positive cervical cancer CaSki cells. These results indicate that down-regulation of HPV16 E7 with antisense RNA is beneficial in reducing the tumorigenicity of CaSki cells, and rAAV vectors ought to be a new efficient approach for delivering the expression of therapeutic genes.Most acquired and inherited diseases are rooted at the genetic level, and the ability to correct genetic defects that cause disease remains the ultimate goal of gene therapy. For some diseases, gene therapy presents the only hope for patients, but truly effective gene therapy continues to be elusive. Although a great variety of genes show potential for the treatment of human disease, delivering genetic material with therapeutic potential to specific target sites remains a challenge (1, 2).Gene delivery systems can be categorized as viral or nonviral systems. The more commonly used viral gene delivery systems are retrovirus (3), adenovirus (4), and adeno-associated virus (AAV; ref. 5) vectors. Commonly used nonviral delivery systems include cationic liposome (6), HVJ-liposome (7), and mechanical approaches, such as ''gene gun' ' (8), DNA infusion, and DNA injection (9). Many gene delivery systems have shown some degree of success in vitro, but all have fallen short in in vivo trials. The major problems encountered have been low efficiency of gene delivery and/or inability to achieve targeted gene placement.Antisense technology has been used mainly to knock out or down-regulate the expression of certain genes associated with disease. Antisense constructs are small and generally do not code for any known biological activities in the host, and in early trials, they have proven to be well tolerated (10, 11). However, the effectiveness of antisense therapy has generally been short lived, and problems with the efficiency and specificity of gene delivery have also limited its use. The side effects that result from nonspecific gene delivery can be circumvented if the chosen target for antisense attack is an acquired genetic material, such as a viral sequence, and not a somatic gene. Targeting acquired genes has been proposed for diseases associated with infectious agents, including some cancers, such as cervical cancer associated with human papillomavirus 16 (HP...