Gene therapy promises to be an effective treatment for inherited disease but despite significant research effort, a clinically-viable gene transfer method has yet to be developed. The various methods under development-naked plasmid DNA, liposomes, retroviruses, adenoviruses, herpes viruses and, most recently, adenoassociated viruses-all suffer from various problems (Herweijer et al., 1995). The critical issues are the efficiency of gene transfer, expression level, and stability of expression. Requirements for efficiency vary greatly depending on the type and function of the protein to be expressed. If it is secreted, gene transfer efficiency (fraction of cells transfected) can be lower since the protein produced will be secreted from the transfected cells and distributed throughout the body for use by other cells. In the case of structural proteins, high transfection efficiency is required to ensure that enough cells express the protein to effect structural integrity. Long term stability of expression of the transferred gene is desirable in order to avoid the process of repeated transfections or gene administrations.