Current antiretroviral therapies (ART) do not allow for the eradication of the human immunodeficiency virus (HIV) due to the presence of latent proviruses in rare, long-lived resting CD4+ T cells. The main research efforts to eliminate the viral reservoir are focused on the use of latency reversing agents (LRAs) to force the reactivation of the latent provirus, while maintaining ART to prevent de novo infections. Subsequently, reactivation of HIV expression would kill reservoir cells via viral cytopathic effects and/or immune clearance (“shock and kill” strategy). So far, no LRAs tested in clinical trials have succeeded in reducing the reservoir. Furthermore, there is limited understanding as to why some latent proviruses are being induced, while others are not.
We used an improved dual-fluorescent HIV reporter (GKO), which distinguishes productively infected cells from the latent population, to investigate the efficacy of the “shock and kill” strategy. In addition, GKO provides a unique opportunity to (1) explore the impact of HIV integration site specificity on the fate of the infection, and to (2) characterize the inducible subpopulation of latently infected cells, since it allows the isolation of inducible latent and non-inducible latent populations from the productively infected majority.
We first showed that our patients' data was consistent with previously published studies. However, we found that at most 5% of GKO latent proviruses were reactivated. Moreover, the analysis of HIV-1 integration sites from productively, non-inducible and inducible latently infected populations reveals heterogeneity within the latent infections. In contrast to non-inducible latent infections, the integration sites of inducible latent proviruses have similar features to those of productive proviruses, thus demonstrating a prominent role for the site of integration and its chromatin context for the fate of the initial infection as well as for latency reversal.
Our study shows an important roadblock for the “shock and kill” approach to reservoir eradication. Differences between inducible and permanently latent reservoir cells suggest that complete reservoir reactivation and eradication with LRAs may prove impossible, and that a multipronged “functional cure” approach may be necessary.