Abstract
The intrinsically disordered HIV-1 Tat protein binds the viral RNA transactivation response structure (TAR), which recruits transcriptional cofactors, amplifying viral mRNA expression. Limited Tat transactivation correlates with HIV-1 latency. Unfortunately, Tat inhibitors are not clinically available. The small molecule didehydro-cortistatin A (dCA) inhibits Tat, locking HIV-1 in persistent latency, blocking viral rebound. We generated chemical derivatives of dCA that rationalized molecular docking of dCA to an active and specific Tat conformer. These revealed the importance of the cycloheptene ring and the isoquinoline nitrogen's positioning in the interaction with specific residues of Tat's basic domain. These features are distinct from the ones required for inhibition of cyclin-dependent kinase 8 (CDK8), the only other known ligand of dCA. Besides, we demonstrated that dCA activity on HIV-1 transcription is independent of CDK8. The binding of dCA to Tat with nanomolar affinity alters the local protein environment, rendering Tat more resistant to proteolytic digestion. dCA thus locks a transient conformer of Tat, specifically blocking functions dependent of its basic domain, namely the Tat-TAR interaction; while proteins with similar basic patches are unaffected by dCA. Our results improve our knowledge of the mode of action of dCA and support structure-based design strategies targeting Tat, to help advance development of dCA, as well as novel Tat inhibitors.IMPORTANCE Tat activates virus production, and limited Tat transactivation correlates with HIV-1 latency. The Tat inhibitor dCA locks HIV in persistent latency. This drug class enables block-and-lock functional cure approaches, aimed at reducing residual viremia during therapy and limiting viral rebound. dCA may also have additional therapeutic benefits since Tat is also neurotoxic. Unfortunately, Tat inhibitors are not clinically available. We generated chemical derivatives and rationalized binding to an active and specific Tat conformer. dCA features required for Tat inhibition are distinct from features needed for inhibition of cyclin-dependent kinase 8 (CDK8), the only other known target of dCA. Furthermore, knockdown of CDK8 did not impact dCA's activity on HIV-1 transcription. Binding of dCA to Tat's basic domain altered the local protein environment and rendered Tat more resistant to proteolytic digestion. dCA locks a transient conformer of Tat, blocking functions dependent on its basic domain, namely its ability to amplify viral transcription. Our results define dCA's mode of action, support structure-based-design strategies targeting Tat, and provide valuable information for drug development around the dCA pharmacophore.