| |
CD32a is a marker of a CD4 T-cell HIV reservoir
harbouring replication-competent proviruses
|
| |
| |
Download the PDF here
Nature March 15 2017
"Altogether, our experiments show that CD32a fulfils the necessary criteria of a biomarker for HIV-persistent CD4 T cells. Although CD32a does not stain the entire CD4 T-cell reservoir, qVOA and viral induction assays demonstrate that CD32a+ CD4 T cells make up a substantial part of the pool of CD4 lymphocytes containing inducible replication-competent proviruses. Accordingly, we observed correlation between IUPM of total CD4 T cells and HIV DNA copy per cell in CD32a+ CD4 T cells, as well as between IUPM and infection frequencies of CD32a+ CD4 T-cells (data not shown). As sorting of CD32a CD4 T cells from ART-treated participants gives a nearly pure population of infected cells (average of 0.9 HIV DNA copy per cell with median of 0.5), it would be possible to study viral and cellular diversity of the HIV reservoir at the single-cell level. An exciting future direction is to determine the distribution of defective, replication-competent and inducible viruses16, 17 within this population of reservoir cells. Our findings also raise important questions about the functional significance of CD32a expression on reservoir cells. The Fc region of HIV-1 Env-specific broadly neutralizing antibodies is required for their antiviral activity18 and antibody-dependent cellular cytotoxicity has a role in this process19. It will thus be interesting to determine whether CD32a expression helps to explain the recent findings that combination of ART and broadly neutralizing antibodies reduces cell-associated HIV-1 DNA and prolonged viral control20. It will also be of interest to determine whether CD32a expression is part of a cellular response to viral infection or is induced by the virus. Most importantly, developing strategies aimed at targeting and eliminating the CD32a+ CD4 T cell HIV reservoir may establish the basis for a 'direct kill' as an alternative and or complementary approach toward a HIV cure."
-------------------------------
Discovery of an HIV reservoir marker: New avenue toward eliminating the virus
Date:
March 15, 2017
Source:
CNRS (Delegation Paris Michel-Ange)
Summary: A marker has been identified that makes it possible to differentiate "dormant" HIV-infected cells from healthy cells. This discovery will make it possible to isolate and analyze reservoir cells which, by silently hosting the virus, are responsible for its persistence even among patients receiving antiviral treatment, whose viral load is undetectable, say scientists.
https://www.sciencedaily.com/releases/2017/03/170315144033.htm
French researchers have identified a marker that makes it possible to differentiate "dormant" HIV-infected cells from healthy cells. This discovery will make it possible to isolate and analyze reservoir cells which, by silently hosting the virus, are responsible for its persistence even among patients receiving antiviral treatment, whose viral load is undetectable. It offers new therapeutic strategies for targeting infected cells. This research is part of the ANRS strategic program "Reservoirs du VIH." It is the result of a collaboration between the CNRS, Montpellier University, Inserm, the Institut Pasteur, the Henri-Mondor AP-HP hospital in Creteil, the Gui de Chauliac hospital (CHU de Montpellier) and the VRI (Vaccine Research Institute), and is published in the journal Nature on March 15, 2017. A patent owned by the CNRS has been filed for the diagnostic and therapeutic use of the identified marker.
Since 1996, there has been consensus among the scientific community that a cure for HIV will involve targeting "reservoir cells" that host the virus in the organisms of patients undergoing triple therapy. HIV can remain hidden in these reservoirs, in latent form, for several decades, eluding the immune system's response and antiviral treatments, without any viral protein being expressed. But if treatment ceases, the virus massively proliferates and the disease progresses again. Patients must therefore receive treatment for life. To envisage eliminating this dormant virus, a first stage consists in distinguishing the HIV-infected reservoir cells from their healthy counterpart cells, which resemble them to a very large degree. This is what has been achieved by a team of researchers, who have identified a marker of reservoir cells: a protein present only on the surface of infected cells.
Hypothesizing that HIV might leave a mark on the surface of its host cell, researchers from the Institut de genetique humaine (CNRS/Montpellier University) first worked in vitro on an infection model developed in their laboratory. After comparing infected cells and healthy cells, they noticed one particular protein, coded by a gene among the hundred of those expressed in a specific way by infected cells. Present only on the surface of the infected cells, the CD32a protein thus met, in vitro, the criteria of a reservoir cell marker. This was then confirmed by experiments on clinical samples. By studying blood samples from 12 patients living with HIV and receiving treatment, the researchers isolated the cells expressing the marker and observed that almost all were HIV carriers. In vitro, the activation of these cells induced a production of viruses capable of reinfecting healthy cells whereas their elimination entailed a significant delay in viral production.
In the fight against HIV, this discovery paves the way to a better fundamental understanding of viral reservoirs, which it will now be possible to isolate more easily and analyze directly. In the longer term, it should lead to therapeutic strategies aiming to eliminate the latent virus from the organism and make remission -- at least temporary -- possible in the absence of antiviral treatments.
----------------------------------
Nature 15 March 2017
CD32a is a marker of a CD4 T-cell HIV reservoir harbouring replication-competent proviruses
The persistence of the HIV reservoir in infected individuals is a major obstacle to the development of a cure for HIV1, 2, 3. Here, using an in vitro model of HIV-infected quiescent CD4 T cells, we reveal a gene expression signature of 103 upregulated genes that are specific for latently infected cells, including genes for 16 transmembrane proteins. In vitro screening for surface expression in HIV-infected quiescent CD4 T cells shows that the low-affinity receptor for the immunoglobulin G Fc fragment, CD32a, is the most highly induced, with no detectable expression in bystander cells. Notably, productive HIV-1 infection of T-cell-receptor-stimulated CD4 T cells is not associated with CD32a expression, suggesting that a quiescence-dependent mechanism is required for its induction. Using blood samples from HIV-1-positive participants receiving suppressive antiretroviral therapy, we identify a subpopulation of 0.012% of CD4 T cells that express CD32a and host up to three copies of HIV DNA per cell. This CD32a+ reservoir was highly enriched in inducible replication-competent proviruses and can be predominant in some participants. Our discovery that CD32a+ lymphocytes represent the elusive HIV-1 reservoir may lead to insights that will facilitate the specific targeting and elimination of this reservoir.
Resting CD4 T cells are paradoxically non-permissive to HIV-1, but also predominantly support viral persistence under antiretroviral therapy in HIV-1-infected participants1, 4, 5, 6. We took advantage of our recent discovery that SAMHD1 is responsible for HIV-1 restriction in resting CD4 T cells to generate latently infected cells without delivering an activation signal7, 8, 9, 10, 11, 12. We could thus test the hypothesis that these latently infected non-stimulated T cells might have a gene expression signature that distinguishes them from their non-infected counterparts. Peripheral blood mononuclear cells (PBMCs) from four HIV-negative controls were isolated and exposed to viral-like particles (VLP) containing the SIVmac251 accessory protein Vpx (VLP-Vpx), which induce SAMHD1 degradation. The Vpx-treated cells were then infected with an HIV-1-derived vector expressing green fluorescent protein (GFP) from a cytomegalovirus promoter (HIV-cmv-GFP). Four days later, treated GFP+ and GFP− quiescent CD4 T cells (denoted hereafter as XH+ and XH−, respectively) were sorted by fluorescence-activated cell sorting (FACS) and total RNA was extracted. We took total RNA from non-infected, untreated CD4 T cells and non-infected, VLP-Vpx-treated (denoted as X) samples as controls. We applied ultra-deep RNA sequencing (RNA-seq) and bioinformatics analyses to determine the effect of viral infection on the transcriptional program of quiescent CD4 T cells (Fig. 1a, Supplementary Table 1). Hierarchical clustering (Fig. 1b) and principal component analyses (Fig. 1c) performed on sequencing data revealed, first, that the gene expression profiles of the XH+ subset from four donors clustered together and away from the XH−, X and non-infected CD4 T-cell profiles. This result indicates that XH+ subsets from different donors share a signature that distinguishes them from the other subsets and that HIV infection shapes the transcriptional landscape of its resting host. We identified differentially expressed genes, particularly those specifically regulated in the XH+ subset, but not in bystander cells (Fig. 1d). Applying a stringent statistical false discovery rate (FDR) cutoff of 10−8 revealed 111 differentially expressed genes that are significantly induced in latently infected cells as compared to bystanders (Fig. 1d); of note, no down-modulated genes were identified. Among the 111 genes, 103 were exclusively induced in XH+ compared to XH− and non-infected controls (Fig. 1e; Supplementary Table 2). Functional network analyses revealed an enrichment of genes involved in regulating cytokine production and inflammatory responses (Extended Data Fig. 1). Among the 103 genes, 16 encode cell surface transmembrane proteins, which we selected for in vitro validation as cell surface markers of latent infection (Fig. 2a). For this purpose, PBMCs from 2–6 individuals not infected with HIV were isolated and infected as described in Fig. 1a. Four days later, infected total CD4 T lymphocytes were screened by flow cytometry for the expression of the 16 candidate markers. FCGR2A was the most specifically and potently induced in GFP+ CD4 T cells, by approximately 50-fold (Fig. 2a, upper panel), but not in the GFP– counterparts (Fig. 2a, lower panel). FCGR2A is one of three genes of the FcγRII subfamily of IgG Fc receptors. FCGR2A is present in humans but not in mice, whereas FCGR2B is also found in mice. FCGR2A encodes CD32a, an activating receptor with low affinity for the IgG Fc fragment, which is expressed exclusively by effector cells of the innate immune system but not by lymphoid cells13, 14. RNA-seq analyses showed that CD32b and CD32c were not significantly induced following HIV infection of resting CD4 T cells (data not shown).
We therefore next investigated CD32a expression in vitro, as described in Fig. 1a, using PBMCs from 2–6 additional individuals not infected with HIV. The results were analysed in the subset of quiescent cells that did not express the marker of activation HLA-DR. In each donor, we confirmed the selective induction of CD32a after HIV infection of resting CD4 T cells (Fig. 2b, left panel). Furthermore, when resting CD4 T cells expressing CD32a were analysed for their infection level, we found that 90% were GFP+ and thus were infected with recombinant HIV virions (Fig. 2b, right panel). Interestingly, treatment with the integrase inhibitor raltegravir (RAL) before infection prevented CD32a expression by infected resting CD4 T cells (Fig. 2c), which highlights the importance of the integration step in CD32a induction. We next determined whether productive HIV-1 infection recapitulates the CD32a induction phenotype. PBMCs from 2–3 HIV-uninfected individuals either received no stimulation or were stimulated with phytohaemagglutinin (PHA) and interleukin 2 (IL-2) and then mock- or VLP-Vpx-treated prior to infection with HIV. FACS analyses revealed that productive infection of stimulated CD4 T cells was not associated with a significant CD32a induction, as compared to resting CD4 T cells (Fig. 2d). These experiments demonstrate that CD32a is a highly specific marker of HIV-1-infected resting CD4 T cells in vitro.
Efforts towards the development of therapies aimed at directly targeting and eliminating the reservoir of HIV-1-infected T cells have been impaired by the lack of a specific marker to distinguish the cells from their non-infected counterparts. On the basis of our in vitro results, we reasoned that the cell surface marker CD32a would allow us to sort reservoir cells in anti-retroviral therapy (ART)-treated participants. CD32a immunostaining of CD4 T cells isolated from 12 participants (Supplementary Table 3) shows a continuum of expression. We therefore separated CD4 T lymphocyte subsets with negative (CD32a−), intermediate (CD32aint) and high (CD32ahi) expression levels, as compared to staining with control isotype (Fig. 3a, Extended Data Fig. 2). FACS-sorted total, CD32a−, CD32aint and CD32ahi CD4 T cells were assessed for infection frequency, as measured by total HIV-1 DNA and absolute cell number (β-globin normalization) (Fig. 3b). Notably, infection frequencies correlated with CD32a expression, as increased viral DNA content was observed in total, CD32a−, CD32aint- and CD32ahi-expressing CD4 T cells (median of 0.003, 0.001, 0.17 and 0.56 HIV-1 DNA copies per cell, respectively) (Fig. 3b; Supplementary Table 4). Averages of 633-fold and 1,024-fold enrichment of HIV DNA copies per cell was observed in CD32ahi CD4 T cells, as compared to total and CD32a− CD4 T cells, respectively (Fig. 3b; Supplementary Table 4). The observed enrichment is independent of the time of infection and duration of treatment (Supplementary Tables 3, 4). Depending on the participants, the contribution of CD32a+ population to the total CD4 T-cell reservoir ranges from 26.8% to 86.3%, as measured by total viral DNA (Fig. 3c). These results validate CD32a as a cell surface marker of the CD4 T cell HIV reservoir in HIV-infected virally suppressed participants.
We next investigated whether CD32a+ CD4 T cells contain an inducible replication-competent provirus. We performed a quantitative viral outgrowth assay (qVOA) and calculated the infectious unit per million cells (IUPM) in total and CD32a+ CD4 T cells. As previously reported15, IUPM ranged from 2.2 to 5.7 in total CD4 T cells (Table 1). Notably, IUPM measured in CD32a+ CD4 T cells isolated from four participants ranged from 2,158 to 16,422 (Table 1). Thus, up to 3,000-fold enrichment in inducible replication-competent provirus is observed in CD32a+ as compared to total CD4 T cells (Fig. 3d). To determine further the contribution of CD32a+ CD4 T lymphocytes to the inducible viral reservoir in ART-treated participants, CD32a− CD4 T cells were sorted from three aviraemic participants and viral production was measured using HIV Gag p24 ultrasensitive ELISA detection after in vitro T-cell receptor stimulation (using anti-CD2, anti-CD3 and anti-CD28 beads and IL-2). For all tested participants, activation of total CD4 T cells resulted in virus production and spreading in cultures (Fig. 3e). No virus production was observed when the cells were not activated (data not shown). By contrast, depletion of CD32a+ CD4 T cells (median of 500 cells) resulted in marked delay in virus production and spreading (Fig. 3e). qVOA performed using CD32a− and total CD4 T cells isolated from one participant showed a tenfold reduction of IUPM in the former (Extended Data Fig. 3), reinforcing the contribution of CD32a+ CD4 T cells to the inducible viral reservoir contained in total CD4 T cells. These experiments show that CD32a+ CD4 T cells greatly contribute to the pool of HIV-infected CD4 T cells that harbour inducible replication-competent proviruses.
Altogether, our experiments show that CD32a fulfils the necessary criteria of a biomarker for HIV-persistent CD4 T cells. Although CD32a does not stain the entire CD4 T-cell reservoir, qVOA and viral induction assays demonstrate that CD32a+ CD4 T cells make up a substantial part of the pool of CD4 lymphocytes containing inducible replication-competent proviruses. Accordingly, we observed correlation between IUPM of total CD4 T cells and HIV DNA copy per cell in CD32a+ CD4 T cells, as well as between IUPM and infection frequencies of CD32a+ CD4 T-cells (data not shown). As sorting of CD32a CD4 T cells from ART-treated participants gives a nearly pure population of infected cells (average of 0.9 HIV DNA copy per cell with median of 0.5), it would be possible to study viral and cellular diversity of the HIV reservoir at the single-cell level. An exciting future direction is to determine the distribution of defective, replication-competent and inducible viruses16, 17 within this population of reservoir cells. Our findings also raise important questions about the functional significance of CD32a expression on reservoir cells. The Fc region of HIV-1 Env-specific broadly neutralizing antibodies is required for their antiviral activity18 and antibody-dependent cellular cytotoxicity has a role in this process19. It will thus be interesting to determine whether CD32a expression helps to explain the recent findings that combination of ART and broadly neutralizing antibodies reduces cell-associated HIV-1 DNA and prolonged viral control20. It will also be of interest to determine whether CD32a expression is part of a cellular response to viral infection or is induced by the virus. Most importantly, developing strategies aimed at targeting and eliminating the CD32a+ CD4 T cell HIV reservoir may establish the basis for a 'direct kill' as an alternative and or complementary approach toward a HIV cure.
|
|
| |
| |
|
|
|
