Human immunodeficiency virus type 1 (HIV-1) transactivating Tat protein is pivotal to virus replication. Tat's potential effects on HIV-1 pathogenesis, however, go well beyond its role in the virus's life cycle. Current data indicate that biologically active Tat is released from HIV-1-infected cells and readily endocytosed and targeted to the nucleus of nearby, or perhaps distant, cells, where it may exert a series of pleiotropic effects. This paracrine action has been extensively investigated, and depending on the amounts of exogenously added Tat, its effects may extend from the suppression of immunocompetent cells to transactivation of heterologous genes to the promotion of growth of Kaposi's sarcoma spindle cells. We have already observed that various cell lines, either permanently transfected with an expressive HIV-1 tat gene construct or cultured in the presence of exogenously added Tat protein, are protected from programmed cell death after serum withdrawal or other apoptotic stimuli. The present article shows that various types (lymphoblastoid, epithelial, neuronal) of permanently tat-transfected cell lines actively release fully bioactive Tat protein. The addition of anti-Tat antibody to the culture medium completely abolishes their increased survival/proliferation capacity in serum-free culture. In these conditions, therefore, the enhanced survival/proliferation potential of permanently tat-transfected cells seems entirely dependent on a Tat-protein autocrine loop. The finding that anti-Tat antibody, added to culture medium, exerts a negative influence on the expression of a Tat-responsive HIV-1 long terminal repeat chloramphenicol-acetyltransferase construct, transiently transfected into permanently tat-transfected cells, suggests that the Tat autocrine loop may also be required for optimal HIV-1 long terminal repeat transactivation.