Adeno-associated virus (AAV) is a small non-enveloped virus and as a member of the parvovirus family, it contains a single stranded DNA genome. AAV's life cycle in vitro is strongly dependent not only on cellular factors but also on helper functions provided by other viruses, for example adenovirus, herpes virus and, to a certain extent, papillomavirus (HPV). Although AAVs are wide spread throughout the human population, no disease is yet associated with them. Accordingly, AAV became attractive as a vector for gene therapy studies, but even though the virus is well studied in vitro and now used in clinical trials as a vector, not much is known about AAV infection in vivo. AAV was initially isolated from humans together with its helper adenovirus. However, adenovirus-independent replication of AAV has been reported for a number of cellular situations, including differentiating keratinocytes with or without the presence of HPV, and in cells with an activated DNA damage response due to genotoxic agents. In the work reported here we have undertaken a systematic study of these alternative helper activities with a view to evaluating their contribution to natural infections by AAV. We used three different systems of cultured keratinocytes and showed that in each case in vitro differentiation was achieved. In two of these, the cells were from HPV-induced lesions. We found that AAV replicates efficiently in differentiating monolayer foreskin keratinocytes in the presence but not in the absence of adenovirus. Likewise naturally HPV-infected cervical keratinocytes only support the full life cycle of AAV if adenovirus is present. Even though HPV can provide certain helper functions, it is less efficient as a helper than adenovirus. We also exploited osteosarcoma cells, which are particularly well infected by AAV. Our results indicate that U2OS osteosarcoma cells are semi-permissive for AAV growth. Induction of a DNA damage response in these cells, or in HPV-containing keratinocytes, stimulates AAV genome amplification. Although AAV replication in the absence of helper virus can readily be demonstrated, this is either at a relatively low level or limited to part of the AAV growth cycle. We conclude that although these alternative helper systems can contribute to the natural history of AAV, the presence of an efficient helper virus seems likely to be needed for AAV to persist in the human population.