DNA–liposome complexes, as discussed in Chapters 7 to 9, have been shown to be effective in in vitro transfection of various cells. It was estimated that in the case of transfection by Ca-precipitation, approximately one out of a million plasmids reaches the cell nucleus and initiates protein synthesis. In the case of liposome-aided transfection, the fraction may be 1 in 10,000 or even less (Feigner and Ringold, 1989). Because the lipid-delivered DNA in general does not incorporate into the chromosome, the longevity of gene expression (i.e., the coded protein synthesis), is, in the absence of special “self-replicating” sequences, at best, a function of the half-life of a particular cell. Despite significant improvements, such efficiencies still cannot be compared with viruses where one can have virtually a 100% efficiency of transfection. A possible improvement might be to include more information, such as nuclear targeting and membrane fusion, from the virus into the plasmid and/or to add a nonimmunogenic, lipid- or polymer-based fusion function on the complex. The ultimate goal — in vivo delivery — involves many other hurdles but the options of complex targeting and the use of tissue-specific promoters for gene expression can further improve transfer of DNA into target cells and subsequent protein synthesis.