An Ounce of Tat Prevention Is Worth a Pound of Functional Cure

Infection by HIV-1 can effectively be suppressed by antiretroviral therapy (ART). However, the persistence of a small, latent virus reservoir impedes viral eradication. Thus, most infected individuals must be treated with ART for life. While efforts to find a sterilizing cure that eradicates the virus are ongoing, an alternative approach deemed a functional cure has remained less explored. A functional cure for HIV is envisioned as long-term control of virus replication in the absence of ART (reviewed in references 1 and 2). Such a functional cure might encompass effective immune interventions, reduction of the reservoir to a highly “controllable” size, and/or induction of a long-lasting suppression of viral gene expression. The Tat viral protein is a strong transactivator of viral gene expression, and viral replication is severely restricted in its absence; thus, it makes a logical antiviral target. A recent article by Mousseau et al. (3) reports on a novel Tat inhibitor that the authors show has the unusual ability to induce a “permanent state of latency.” Didehydro-cortistatin A (dCA) is a departure from current FDA-approved antiretroviral drugs in that it inhibits viral gene expression from integrated viruses. For the above reasons, dCA represents a novel inhibitor class. Cortistatin A (CA) is a steroidal alkaloid from Corticium simplex, a marine sponge, which has raised interest based on its antiproliferative effects (4). dCA is a derivative of CA that was reported to be equipotent (5) to the parent compound as well as relatively easy to obtain by chemical synthesis. One of the salient features of CA is its ability to bind to and inhibit cyclin-dependent kinase 11 (CDK11) with nanomolar affinity (6). CDK11 was recently shown by Valente and colleagues to be essential for HIV-1 gene expression by phosphorylating factors required for proper 3= processing of viral mRNAs (7). Based on these findings, Mousseau and colleagues in the Valente laboratory predicted that dCA, by blocking the activity of CDK11, would in turn hinder HIV-1 transcription (8). dCA proved to be a potent inhibitor of HIV replication (8). Intriguingly, Mousseau et al. were unable to reproduce CAmediated CDK11 inhibition when testing dCA and instead demonstrated a different mechanism for dCA’s antiviral effects. Mousseau and colleagues suggest that dCA binds to Tat’s basic domain, the same domain that mediates localization to the nucleolus and binding to the transactivation response element (TAR) (8). dCA thus prevents recruitment of Tat to the TAR stem-loop structure, presumably hindering the recruitment of P-TEFb (positive transcription elongation factor-b) and transcription elongation (see Fig. 1). The latest study by the Valente group further explores the effects of dCA (3) by asking whether the inhibitor has an impact on viral latency. This study begins by testing ex vivo the effect of dCA on peripheral-blood-derived resting T cells from HIV-infected, aviremic patients. Cells from nine selected patients were activated with anti-CD3 and -CD28 antibodies, in the presence or absence of dCA, and viruses released to the supernatant were measured with an ultrasensitive, quantitative reverse transcription-PCR (RT-qPCR) assay. Addition of dCA along with anti-CD3 and antiCD28 diminished the release of virus in all patient samples by 92.3% on average. In the face of the powerful stimulation used, this represents a significant blockade of virus reactivation. Efforts to further understand the dCA effects on latency led Mousseau and colleagues to experiment with several laboratory models of latency. First, the authors used an unpublished model consisting of HeLa CD4 cells latently infected with HIV-1NL4-3, where residual viral gene expression produces about 500 pg/ml of p24 in the culture supernatant. Treatment of these cells with dCA suppressed residual p24 to undetectable levels ( 3.1 pg/ml). This was consistent with a simultaneous reduction in viral mRNA production of 97% imposed by dCA. In contrast, conventional ART (lamivudine, raltegravir, efavirenz) failed to reduce the residual p24 levels. Cessation of ART, both in vivo and in vitro, leads to a rebound in viral replication. When treatment with dCA of the latently infected HeLa CD4 cells was interrupted at day 100, viral rebound was not observed for the remainder of the experiment (150 days). This finding was quite surprising and, possibly, reveals an additional activity of dCA, which is to induce a “permanent state of latency.” Interestingly, an earlier treatment cessation at day 24 did not lead to suppression of residual virus, which suggests that successful induction of the long-term suppressive effect of dCA may require longer exposure to the drug. Similar effects were observed in other cell culture models of latency, demonstrating that the findings are not unique to one model. To further establish that the inhibitory effect of dCA was at the level of transcription, the authors performed chromatin immunoprecipitation analysis with an antibody to RNA polymerase II (Pol II), which transcribes HIV-1 genes. This analysis revealed that dCA treatment modestly impaired recruitment of the polymerase to promoter-proximal regions (indicative of transcription initiation) and severely impaired recruitment to a distal open reading frame, vpr (transcription elongation). Recruitment of Pol II to these viral sequences in response to tumor necrosis factor alpha (TNF) stimulation, an NFB-mediated event that does not require Tat, was unaffected. The authors report that dCA, at the concentrations tested, was neither toxic nor cytostatic. Transcription inhibitors, in particular flavopiridol and DRB (5,6-dichloro-1-D-ribofuranosylbenzimidazole), which, like dCA block elongation, are known to be cytotoxic (reviewed in reference 9). A critical difference between the mode of action of these compounds and that of dCA is that