- The viral transcriptional transactivator Tat binds the transactivation response (TAR) element RNA stem-loop to activate the expression of genes from and replication of human immunodeficiency virus type 1 (HIV-1) (16, 24). TAR forms a stable hairpin at the 5′ end of all viral transcripts. Although Tat alone can bind to the 5′ bulge (from positions 23 to 25) of TAR in vitro, the central loop (from positions 31 to 36) of TAR is required for Tat transactivation in vivo (8, 16). A recent study suggested that the central loop binds the cellular cofactor of Tat, human cyclin T1 (hCycT1) (28). Thus, hCycT1 and Tat form a high-affinity complex with TAR (28, 29). This interaction results in the recruitment of Cdk9, the kinase partner of hCycT1. Cdk9 and hCycT1 or the related cyclins T2a, T2b, and K form different positive transcription elongation factor b (P-TEFb) complexes (9, 15, 20, 21). P-TEFb phosphorylates the C-terminal domain of RNA polymerase II (RNAPII), Spt5 from the DRB sensitivity-inducing factor, and possibly other proteins (14, 19, 20, 27). This enzymatic activity is required for the conversion of unphosphorylated (RNAPIIa) to phosphorylated (RNAPIIo) forms of RNAPII and the transition from initiation to elongation of eukaryotic transcription (21). hCycT1 contains 726 residues (20, 28). From positions 1 to 250, 251 to 271, 370 to 430, 506 to 530, and 709 to 726, two conserved cyclin boxes, a Tat-TAR recognition motif (TRM), a coiled-coil region, a histidine-rich stretch, and the C-terminal PEST sequence, respectively, are found (28). In the TRM of hCycT1, the cysteine at position 261 forms a Zn2+-dependent interaction with other cysteines and/or histidines from positions 1 to 48 of Tat. Since the cysteine at position 261 is changed to tyrosine in the murine CycT1 (mCycT1), mCycT1 cannot support Tat transactivation (4, 6, 11, 13, 18). Although mCycT1 binds Tat with equal to slightly lower affinity than does hCycT1, its ability to coordinate TAR binding is reduced drastically (11, 13). As with mCycT1, hCycT1 from positions 1 to 250, which lacks the TRM, binds Tat more weakly and does not support Tat transactivation in cells (11). These observations indicate that cyclin boxes also bind Tat and that the TRM primarily positions the arginine-rich motifs (ARM) in Tat and hCycT1 for optimal TAR binding. Finally, hCycT1 cannot bind TAR in the absence of Tat (28). The first 272 residues of hCycT1 are sufficient to bind Tat and TAR in vitro and to support Tat transactivation in vivo (28). However, the C-terminal region of hCycT1 plays an inhibitory role in the interaction among Tat, hCycT1, and TAR (7, 12). This inhibition can be alleviated by Tat-SF1, which is one of several cellular cofactors of Tat (7). Additionally, the autophosphorylation of Cdk9 is required for the binding of P-TEFb, Tat, and TAR (12). These observations indicate that P-TEFb, Tat, and TAR form a dynamic complex and suggest that detailed structural studies of this RNA-protein assembly could be very difficult. Indeed, to date, the complete structures of hCycT1, Cdk9, Tat, and/or TAR have not been delineated. In the present study, we wanted to find the smallest complex consisting of hCycT1 and Tat that could be used for further structural studies of TAR. First, we determined that besides the cysteine at position 261, no other cysteines or histidines in hCycT1 are required for its Zn2+-dependent interaction with Tat. Second, we demonstrated that the fusion protein consisting of hCycT1 from positions 1 to 280 and Tat not only binds TAR in vitro but also activates the HIV-1 long terminal repeat (LTR) in murine cells. Surprisingly, in this chimera, N-terminal deletions to the TRM in hCycT1 supported Tat transactivation, which indicated that this fusion protein can interact with TAR in vivo. Finally, we demonstrated that this mutant chimera binds TAR in vitro. We conclude that the TRM and Tat form a minimal unit for high-affinity binding to TAR.