A cocultivation assay system consisting of uninfected human T cells and cells chronically infected with human immunodeficiency virus type 1 has been used to investigate syncytium formation in short-term assays. Continuous treatment or short-term pretreatment of uninfected CD4-expressing human T-cell lines with 3'-azido-3'-deoxythymidine (AZT) reduces the ability of these cells to participate in syncytium formation when mixed with chronically infected cells. The effect of AZT on syncytium formation is observed both as a reduction in the number of syncytia and as a reduction in the size of the syncytia that are detected. This syncytiumreducing effect of AZT is dose and time dependent and does not result from a modulation of CD4 antigen expression on the cell surface of uninfected, treated cells. Maximum syncytium reduction is observed with the continuous presence of AZT; however, pretreatment for times as short as 15 min results in a significant reduction in syncytium formation. Since reverse transcription is not required for efficient syncytium formation, the syncytium-reducing effect of AZT on uninfected human cells may represent an antiviral property of AZT with important therapeutic potential.sections from AIDS patients (22). The ability to efficiently form syncytia may be a major contribution to the virulence of HIV and may play a role in disease progression. Several reports have suggested that an increased ability to form syncytia in cell culture is associated with increasing virulence of HIV (23)(24)(25).Few studies have addressed the effects of AZT on syncytium formation (20,26). We have determined that AZT has a significant effect on syncytium formation and cell killing in an in vitro cocultivation assay. The results of experiments presented here indicate that exposure to AZT exerts a syncytium-reducing effect on the uninfected CD4-expressing cell population that is both time and dose dependent. Pretreatment of the chronically infected cell population with AZT has no effect on the ability of these cells to participate in syncytium formation. It has been postulated that cell fusion and syncytium formation are not dependent on virus replication and reverse transcription (27), and the results of our experiments support this conclusion. Thus, we believe that the ability to reduce syncytium formation is an additional antiviral property of AZT.