The foamy virus (FV) genome contains two promoters, the canonical long terminal repeat (LTR) promoter, containing three consensus AP-1 binding sites, and an internal promoter (IP) within the env gene. We investigated the regulation of the two promoters in lytic and persistent infections and found that in the presence of a constitutive source of the viral transactivator protein Tas, transactivation of the LTR promoter and that of the IP differ. In lytic infections, both the LTR promoter and the IP are efficiently transactivated by Tas, while in persistent infections, the IP is efficiently transactivated by Tas, but the LTR promoter is not. Analysis of proteins expressed from the LTR promoter and the IP during infection indicated that IP transcription is more robust than that of the LTR promoter in persistently infected cells, while the opposite is true for lytically infected cells. Coculture experiments also showed that LTR promoter transcription is greatest in cells which support lytic replication. Replacement of much of the LTR promoter with the IP leads to increased viral replication in persistent but not lytic infections. We also found that the induction of persistently infected cells with phorbol 12-myristate 13-acetate (PMA) greatly enhanced viral replication and transcription from the
SFVcpz(hu) (new name for human FV) LTR promoter. However, mutation of three consensus AP-1 binding sites in the FV LTR promoter did not affect viral replication in lytically or persistently infected cells, nor did the same mutations affect LTR promoter transactivation by Tas in PMA-treated cells. Our data indicate that differential regulation of transcription is important in the outcome of FV infection but is unlikely to depend on AP-1.Foamy viruses (FVs) are unique among retroviruses in their establishment of life-long persistent infections without any accompanying pathologies. Infection is characterized by the presence of viral DNA in a large number of organs (9, 42), without detectable levels of viral RNA or protein expression (6,9,42,44). Indeed, viral transcription has been detected only in the oral mucosa of a single infected animal (9). However, virus can be recovered readily by coculturing of infected tissues, peripheral blood, or throat swab specimens with susceptible cell lines (6,18,42,44,46,49). Thus, in most locations in vivo, FV replication is latent; however, when the virus is removed from such a context, replication can proceed. In contrast to the in vivo situation, FV replication in vitro can result in either lytic or persistent infection (13,41,53). Infection of many cell types in vitro is often accompanied by cytopathic effects (CPE) and rapid cell killing. Since such infections do not mimic the in vivo situation, we sought to develop a tissue culture system in which there is little viral replication. For these studies we used the prototypic human FV (HFV) clone HFV13 (29). HFV has recently been renamed SFVcpz(hu) to more clearly indicate that the original HFV isolate is a chimpanzee FV isolated from a huma...