Determinants of the establishment of human immunodeficiency virus type 1 latency.

The ultimate barrier to the eradication of HIV from an infected individual is depletion of the latent reservoir. Strategies aimed at this require an accurate understanding of the mechanisms governing latency. Importantly, this study argues against prevailing models for the establishment and maintenance of latency and provides evidence for novel mechanisms at play.

The prevailing model for the establishment and maintenance of HIV-1 latency suggests that viral gene production is silenced shortly following integration of a provirus. This latent state is then believed to be reinforced by the formation of restrictive chromatin structures, primarily histone modifications. Based on this understanding, depletion of the latent reservoir in patients has been attempted several times, using strategies aimed at activating the latent provirus through cellular stimulation, or by altering histone modifications with histone deacetylase inhibitors. Results have been largely inconclusive. The most important finding of this study is strong evidence for a novel mechanism governing the establishment of latency. Through elegant in vitro experiments involving primary T cells, the authors conclude that the latent reservoir is composed almost entirely (>99.9%) of viruses that were transcriptionally silent immediately following integration, which they show is dependent solely on the availability of the transcription factor, NF-KB, at the time of infection. Furthermore, they find no evidence for the formation of restrictive chromatin structures in this silencing, or in its maintenance. Unexpectedly, they show that inhibitors of restrictive chromatin modification resulted in greater, not fewer, latent infection events, which they propose to be due to the greater availability of 'open' chromatin for successful integration. Additionally, the authors show that prolonged (chemical) inhibition of viral transcription was insufficient for the establishment of latency. These results are in direct conflict with prevailing dogma and at least one recent study {1}. However, the use of primary T cells and fully infectious virus in the current study, compared to transformed T cell lines and non-full-length mutant viral constructs in {1}, lend credence to the current paper. This study also provides confirmation of a previously proposed hypothesis {2} which, through the use of resting T cells from HIV-1-infected patients, suggests that latency is maintained by transcriptional interference. According to this model, active transcription of nearby cellular genes continues through the integrated viral DNA and, thus, prevents the initiation of viral transcription and reinforces the latent state. Using a nested RT-PCR approach, the authors of the present study show directly that all tested latent infection events (n=13) were within active host genes, and that host mRNA transcripts included viral sequences. The formation of restrictive chromatin structures within actively transcribing host genes is unlikely. Overall, this study has important implications for clinical strategies aimed at reducing the size of the latent reservoir in infected individuals. Obviously, any such attempt hinges on an accurate understanding of the mechanisms governing latency. If, as the authors suggest, latency is maintained by transcriptional interference and not by restrictive chromatin structures, any attempt to deplete the latent reservoir based on current models may be doomed to fail, and novel strategies are urgently needed.

References: {1} Pearson et al. J Virol 2008, 82:12291-303 [PMID:18829756]. {2} Han et al. J Virol 2004, 78:6122-33 [PMID:15163705].

Acknowledgements: I would like to thank Aaron Donahue for their assistance in the preparation of this evaluation.