Integration in or Near Oncogenes Plays Only a Minor Role in Determining the in Vivo Distribution of HIV Integration Sites Before or During Suppressive Antiretroviral Therapy

John M. Coffin, Michael J. Bale, Daria Wells, Shuang Guo, Brian T. Luke, Jennifer M. Zerbato, Michele D. Sobolewskii, Twan Sia, Wei Shao, Xiaolin Wu, Frank Maldarelli, Mary F. Kearney, John W. Mellors, Stephen H. HughesbioRxiv 2020 · DOI

TL;DR

This study reveals that while HIV integration preferentially targets highly expressed genes, the location of the integrated provirus plays a surprisingly minor role in determining which infected cells persist long-term on antiretroviral therapy (ART). Instead, factors independent of integration site, such as clonal expansion driven by non-viral mechanisms, are the primary drivers of the persistent HIV reservoir.

What Problem Does This Paper Address?

HIV's ability to persist in the body, even under suppressive antiretroviral therapy (ART), is a major hurdle to achieving a cure. This persistence is largely due to integrated HIV DNA (proviruses) residing in long-lived cells. Understanding what controls the survival and expansion of these infected cells is crucial for developing effective eradication strategies. Previous research suggested that HIV integration into specific genomic locations, particularly near oncogenes or highly expressed genes, might influence cell survival and proliferation. This paper aims to clarify the role of integration site preference and its impact on the long-term fate of HIV-infected cells, both before and during ART.

Methods

The researchers employed a multi-pronged approach to analyze HIV integration sites (IS):

  • In vitro infection of PBMCs: Peripheral Blood Mononuclear Cells (PBMCs) from healthy donors were infected with HIV in vitro. This served as a model for the initial, unbiased distribution of integration sites before any selective pressures could act.
  • Linker-mediated PCR and Next-Generation Sequencing: This standard technique was used to identify and map the precise locations of integrated proviruses in both in vitro and ex vivo samples.
  • RNA-sequencing (RNA-seq): This was used to quantify gene expression levels in the infected cells, allowing researchers to correlate integration site preference with gene activity.
  • Comparative analysis of datasets: Integration site data from in vitro infected cells were compared with data from HIV-infected individuals sampled both before ART (pre-ART) and during suppressive ART (on-ART). This allowed for the identification of selective forces that shape the integration landscape over time.
  • Bioinformatic analysis: Specialized pipelines were used to process raw sequencing data, remove artifacts, and map integration sites to the human genome.

Datasets

  • PBMC dataset: Approximately 385,000 integration sites from in vitro HIV-infected PBMCs. This dataset serves as a baseline for initial integration patterns.
  • Pre-ART dataset: Approximately 16,000 integration sites from HIV-infected individuals sampled before initiating ART.
  • On-ART dataset: Approximately 54,000 integration sites from HIV-infected individuals sampled during suppressive ART.
  • In vivo datasets: Combined data from multiple studies, totaling 15,780 unique IS from pre-ART samples and 53,945 unique IS from on-ART samples.

Key Insights and Ideas

  • Integration site is not destiny: The location where HIV integrates into the host genome has a much smaller impact on long-term cell survival than previously thought.
  • Oncogenes are not the main drivers of persistence: While integration near some specific genes (like STAT5B, BACH2, MKL1, MKL2, MYB, IL2RB) can confer a survival advantage, this effect is limited to a small fraction of infected cells and is not the primary mechanism for the vast HIV reservoir.
  • Clonal expansion is key, but not necessarily driven by integration site: The majority of persistent infected cells are found in expanded clones. However, the study suggests that this expansion is largely driven by factors independent of the provirus's location, such as homeostatic or antigen-driven expansion of T cells.
  • Expressed genes are generally selected against: Proviruses in highly expressed genes tend to be lost over time, suggesting a negative selection pressure.
  • In vitro data is a good proxy for initial integration: The distribution of integration sites in in vitro infected PBMCs closely mirrors the initial distribution in vivo, validating its use as a baseline.

Key Results

  • The distribution of HIV integration sites in vivo (both pre-ART and on-ART) is remarkably similar to the distribution observed in vitro in infected PBMCs.
  • While integration favors highly expressed genes, there's a decrease in the proportion of proviruses in these genes over time on ART, indicating negative selection.
  • A small number of specific genes (STAT5B, BACH2, MKL1, MKL2, MYB, IL2RB) showed evidence of positive selection, where proviruses in these genes were enriched in vivo and contributed to cell growth or survival. However, these accounted for only about 1.2-1.5% of the total proviruses.
  • Clonal amplification is widespread, with an average amplification ratio of 1.6 for on-ART samples. Crucially, this amplification was found to be largely independent of integration site, gene expression level, or provirus orientation.

Limitations and Future Work

  • Completeness of identified genes: While the study identified six genes where proviruses confer a selective advantage, the authors acknowledge that there may be other, rarer genes with similar effects, but their contribution to the overall reservoir would be negligible.
  • Mechanisms of clonal expansion: The study strongly suggests that factors other than integration site drive clonal expansion. Future work could focus on elucidating these non-viral mechanisms, such as homeostatic proliferation and immune responses.
  • Defective proviruses: The study notes that many proviruses in these selected genes are defective, and the reservoir is defined by intact proviruses. Further investigation into the role of defective proviruses in the broader context of persistence and immune activation could be valuable.
  • Centromeric integration: While some integration in centromeric regions was detected, its role in persistence remains less clear and warrants further investigation.

Summary updated

Abstract

Abstract HIV persists during antiretroviral therapy (ART) as integrated proviruses in cells descended from a small fraction of the CD4+ T cells infected prior to the initiation of ART. To better understand what controls HIV persistence and the distribution of integration sites (IS), we compared about 16,000 and 54,000 IS from individuals pre-ART and on ART, respectively, with approximately 385,000 IS from PBMC infected in vitro. The distribution of IS in vivo is quite similar to the distribution in PBMC, modified by selection against proviruses in expressed genes, by selection for proviruses integrated into one of 6 specific genes, and by clonal expansion. The clones in which a provirus integrated in an oncogene contributed to the survival of the clone comprise only a small fraction of the clones that persist in HIV-infected individuals on ART. Mechanisms that do not involve the provirus, or its location in the host genome, are more important in determining which clones expand and persist. Author Summary In HIV-infected individuals, a small fraction of the infected cells persist and divide. This reservoir persists on ART and can rekindle the infection if ART is discontinued. Because the number of possible sites of HIV DNA integration is very large, each infected cell, and all of its descendants, can be identified by the site where the provirus is integrated (IS). To understand the selective forces that determine the fates of infected cells in vivo, we compared the distribution of HIV IS in freshly-infected cells to cells from HIV-infected donors sampled both before and during ART. We found that, as has been previously reported, integration favors highly-expressed genes. However, over time the fraction of cells with proviruses integrated in highly-expressed genes decreases, implying that they grow less well. There are exceptions to this broad negative selection. When a provirus is integrated in a specific region in one of six genes, the proviruses affect the expression of the target gene, promoting growth and/or survival of the cell. Although this effect is striking, it is only a minor component of the forces that promote the growth and survival of the population of infected cells during ART.
BibTeX
@misc{coffin2020integration,
  author = {John M. Coffin and Michael J. Bale and Daria Wells and Shuang Guo and Brian T. Luke and Jennifer M. Zerbato and Michele D. Sobolewskii and Twan Sia and Wei Shao and Xiaolin Wu and Frank Maldarelli and Mary F. Kearney and John W. Mellors and Stephen H. Hughes},
  title = {{Integration in or Near Oncogenes Plays Only a Minor Role in Determining the in Vivo Distribution of HIV Integration Sites Before or During Suppressive Antiretroviral Therapy}},
  year = {2020},
  doi = {10.1101/2020.11.25.397653},
  howpublished = {bioRxiv},
}
Figure