Gene transfer vectors based on the human adeno-associated virus serotype 2 (AAV-2) have been developed and tested in pre-clinical studies for almost 20 years, and are currently being evaluated in clinical trials. So far, all these studies have provided evidence that AAV-2 vectors possess many properties making them very attractive for therapeutic gene delivery to humans, such as a lack of pathogenicity or toxicity, and the ability to confer long-term gene expression. However, there is concern that two restrictions of AAV-2 vectors might limit their clinical use in humans. First, these vectors are rather inefficient at transducing some cells of therapeutic interest, such as liver and muscle cells. Second, gene transfer might be hampered by neutralizing anti-AAV-2 antibodies, which are highly prevalent in the human population. In efforts to overcome both limitations, an increasing number of researchers are now focusing on the seven other naturally occurring serotypes of AAV (AAV-1 and AAV-3 to -8), which are structurally and functionally different from AAV-2. To this end, several strategies have been devised to cross-package an AAV-2 vector genome into the capsids of the other AAV serotypes, resulting in a new generation of "pseudotyped" AAV vectors. In vitro and in vivo, these novel vectors were shown to have a host range different from AAV-2, and to escape the anti-AAV-2 immune response, thus underscoring the great potential of this approach. Here the biology of the eight AAV serotypes is summarized, existing technology for pseudotyped AAV vector production is described, initial results from pre-clinical evaluation of the vectors are reviewed, and finally, the prospects of these promising novel tools for human gene therapy are discussed.