Activation of HIV-Specific CD8 + T-cells from HIV+ Donors by Vesatolimod

Abstract

Background

Vesatolimod (VES; GS-9620) is a Toll-like receptor 7 (TLR7) agonist that directly activates human plasmacytoid dendritic cells (pDCs) and B lymphocytes resulting in direct and indirect production of cytokines and immune activation. VES is being evaluated in HIV-1-infected people as part of an HIV remission strategy. Here we investigated the potential of VES to trigger indirect activation of HIV-specific CD8⁺ T-cells using immune cell cultures derived from HIV+ donors.

Methods

Peripheral blood mononuclear cell (PBMC) cultures derived from HIV+ donors virologically suppressed on stable antiretroviral therapy (n=31) were isolated and treated with VES or vehicle for 24 h. Cells were stained with surface and intracellular fluorescent conjugated antibodies and HIV-specific pentamers, and analysed by flow cytometry.

Results

Treatment of PBMCs with VES resulted in all 31 donors demonstrating a concentration dependent increase in CD8⁺ T-cell activation (CD69⁺) of up to 88%. Of these donors, 20 of 31 donors displayed a concentration-dependent increase in HIV-specific CD8⁺ T-cell activation due to VES with a maximum of 20.8%. Intracellular staining was performed in a subset of donors (n=14), 5 of which displayed VES-induced activation of functional HIV-specific CD8⁺ T-cells as assessed by CD107a and/or tumour necrosis factor (TNF)-α upregulation.

Conclusions

This study demonstrates that VES treatment can induce the activation of functional HIV-specific CD8⁺ T-cells in donor derived PBMCs. These data support the potential use of VES to activate functional HIV-specific CD8⁺ T-cells as part of an HIV remission strategy.

Introduction

People living with HIV (PLWH) can live prolonged lives with antiretroviral therapy (ART) but require lifelong treatment. Although ART can suppress HIV replication, the integration of the HIV genome in CD4⁺ T-cells early in infection has led to the creation of latently infected CD4⁺ T-cells (HIV reservoirs), which presents a hurdle to the elimination of HIV-1 [1,2]. Studies have shown that stimulation of the latent reservoirs using agents such as histone deacetylase (HDAC) inhibitors can induce viral production and may lead to recognition and clearance of infected cells by the immune system [3–5].

Toll-like receptors (TLRs) are pattern recognition receptors which can recognize pathogens and lead to the activation of the innate and adaptive immune response [6–9]. Toll-like receptor 7 (TLR7), primarily found in plasmacytoid dendritic cells (pDCs) and B lymphocytes, can recognize patterns associated with viral RNA and subsequently lead to type I interferon (IFN) secretion [10–12]. Vesatolimod (VES; GS-9620) is a potent TLR7 agonist that induces the production of antiviral cytokines such as type I IFN [10,11,13–15]. VES has been observed to modestly increase HIV expression in CD4⁺ T-cells from HIV-infected ART suppressed donors in vitro [14]. In one animal study, VES induced the expression of SIV RNA in SIV-infected rhesus macaques on suppressive ART [16]. In other subsequent monkey studies there were no discernible effects of VES on plasma viral loads [17–20]. A recent clinical study conducted with VES in HIV+ participants suppressed with ART was consistent with these latter monkey studies in showing no effect of the TLR7 agonist on plasma viraemia [21]. In contrast, these studies found that VES was able to consistently induce the activation of immune cells such as CD4⁺ T-cells, CD8⁺ T-cells and natural killer (NK) cells both in vitro and in vivo [14,16–21]. The VES-induced activation of CD8⁺ T-cells may help enhance the clearance of HIV-infected cells if these CD8⁺ T-cells are functional and can recognize HIV-infected CD4⁺ T-cells. Here we investigated the potential for VES to induce the activation of functional cytotoxic HIV-specific CD8⁺ T-cells from HIV-infected donor peripheral blood mononuclear cell (PBMC) cultures.

Methods

PBMC isolation and treatment

Leukopaks were collected from HIV-infected donors virologically suppressed (<50 HIV RNA copies/ml) receiving antiretroviral therapy (ART). PBMCs were isolated from the leukopaks using Ficoll (GE Healthcare, Marlborough, MA, USA) centrifugation techniques and red blood cells were removed using red blood cell lysis buffer (Biolegend, San Diego, CA, USA). The isolated PBMCs were plated at a concentration of 1 million cells/200 μl cell culture media in 96-well tissue culture plates and treated for 24 h with either DMSO (background control), PMA/ionomycin (2.5 and 0.5 μg/ml, respectively; positive control), or VES at varying concentrations of 0.17 nM–10 μM (threefold serial dilutions) or 20 nM, 1 μM, 10 μM. Treatment with 2.5 mg/ml ProMix HIV peptide pool and 2.5 mg/ml ProMix CEF peptide pool (ProImmune, Sarasota, FL, USA) was also performed. Each condition was set up in 4 replicate wells and then pulled together before staining. Cell culture media used was RPMI (Gibco, Waltham, MA, USA) supplemented with 10% heat inactivated Hyclone FBS (GE Healthcare) and 1x Penicillin/Streptomycin (Gibco).

Sample staining

Pentamers were used to detect antigen-specific CD8⁺ T-cells. ProImmune Pro5 MHC Class I HIV Pentamers are composed of 5 MHC class I peptide complexes which display HIV peptides with the complexes conjugated to the fluorophore APC. These are used in this assay to detect HIV peptide (antigen) specific CD8⁺ T-cells by flow cytometry. Pentamers were chosen for use in a given experiment based on the donors’ HLA type(s). HIV+ donors used here were serotyped and had at least 1 matching MHC allele from the available pentamers.

At the end of cell culture incubation, PBMCs were labelled with pentamers in one of two different staining methods, either with cell surface marker antibody staining or with both cell surface and intracellular marker antibody staining. Staining was optimized and performed according to ProImmune's product specifications as follows.

Sample staining with pentamers and cell surface antibodies

After 24 h of treatment with VES and controls, PBMCs ere washed to remove test agents and were incubated with FcR Blocking Reagent (Miltenyi Biotec, Bergisch Gladbach, North Rhine-Westphalia, Germany), to minimize non-specific antibody binding. To assess the level of activated HIV-specific CD8⁺ T-cells, PBMCs were stained with the chosen allele matched pentamers or a non-allele matched pentamer (irrelevant MHC allele) for a mismatched pentamer negative control (this was done for section: Sample staining with pentamers and cell surface antibodies, not for section: Sample staining with pentamers, cell surface antibodies, and intra-cellular antibodies). Activated T-cells were identified by staining with FITC-anti-CD8, PE-Cy7-anti-CD3 and APC-Cy7-anti-CD69 (BD Biosciences, San Jose, CA, USA). To gate out NK cells (CD56⁺) and B-cells (CD19⁺) from the analysis, cells were also stained with PerCP-Cy5.5-anti-CD56 and V450-anti-CD19 (BD Biosciences). An isotype-negative control was also used to assess background staining. Cells were treated with BD Cytofix/Cytoperm fixation buffer (BD Biosciences), data was acquired on an LSR Fortessa (BD Biosciences) and data analysis was done using Flow Jo software (BD Biosciences, Ashland, OR, USA). Statistical significance (P<0.05, P<0.01, P<0.001) of the values at each concentration was calculated using Excel (Student's t-test).

Sample staining with pentamers, cell surface antibodies and intracellular antibodies

After 18 h of treatment with VES and controls, PBMCs were washed to remove activating agents and stained with allele matched pentamers. After pentamer staining, the cells were placed back into treatment using fresh drug for 6 additional h (total treatment incubation time 24 h). To detect the accumulation of CD107a (a degranulation marker), a BV421-anti-CD107a antibody (BD Biosciences) was added to the 6-h treatment along with the 2 protein transport inhibitors, Brefeldin A (BD Golgi Plug) and Monensin (BD Golgi Stop; BD Biosciences). At the end of the incubation, cells were washed and stained with LIVE/DEAD Fixable Aqua Dead Cell viability stain (Thermo Fisher Scientific, Waltham, MA, USA) and FcR Blocking Reagent (Miltenyi Biotec) to minimize non-specific antibody binding. Activated T-cells were identified by staining with FITC-anti-CD8, PE-Cy7-anti-CD3 and APC-Cy7-anti-CD69 (BD Biosciences). For gating out NK cells (CD56⁺) and B-cells (CD19⁺) from the analysis, cells were stained with PerCP-Cy5.5-anti-CD56 and PerCP-Cy5.5-anti-CD19 (BD Biosciences). An isotype negative control was also used to assess background staining of the antibodies. Cells were then treated with BD Cytofix/Cytoperm fixation buffer (BD Biosciences) and stained for intracellular cytokines using BD Perm/Wash Buffer (BD Biosciences) with PE-anti-IFN-γ and Alexa Fluor 700-anti-TNF-α (BD Biosciences). Cells were again treated with BD Cytofix/Cytoperm fixation buffer (BD Biosciences) and data was acquired on an LSR Fortessa (BD Biosciences) and analysed by Flow Jo software (BD Biosciences, Ashland, OR, USA). Statistical significance (P<0.05, P<0.01, P<0.001) of the values at each concentration was calculated using Excel (Student's t-test).

Results

Activation of CD8⁺ T-cells (n=12) by VES

Initial experiments in PBMC cultures derived from 12 ART-suppressed HIV-infected donors demonstrated concentration dependent increases in activated CD8⁺ T-cells due to VES treatment (Figure 1). Donor to donor variation in maximum percent activated CD8⁺ T-cells was observed (range: 50.6–79.6%). Maximum percent activated CD8⁺ T-cells was reached at VES concentrations of approximately 5–40 nM. Based on these results, subsequent experiments were designed and conducted using VES concentrations ≥20 nM (20 nM, 1 μM and 10 μM) that provided a maximum activation of CD8⁺ T-cells anticipated for further donors to be studied.

Figure 1

VES-induced CD8⁺ T-cell activation (n=12)

Activation of CD8⁺ T-cells (n=31) by VES

PBMCs from 31 ART-suppressed HIV-infected donors were treated with VES at 20 nM, 1 μM and 10 μM (14 of 31 donors treated at 10 μM) and assessed for CD8⁺ T-cell activation by quantifying the level of activation marker CD69. For each donor, CD8⁺ T-cell activation levels were normalized to the DMSO control to display the activation induced by VES or PMA/ionomycin. Raw activation data capturing the basal level of CD8⁺ T-cell activation for each donor are shown in Additional file 1.

Consistent with the first set of experiments, induction of CD8⁺ T-cell activation by VES was seen in all donors at one or more concentrations (Figure 2). An increase in CD8⁺ T-cell activation was observed with increasing VES concentration. The VES response led to high levels of CD8⁺ T-cell activation ranging from 0–48.1% at 20 nM, 9.7–79.5% at 1 μM and 18.6– 88.2% at 10 μM. The mean CD8⁺ T-cell activation for all donors at 20 nM, 1 μM and 10 μM VES was 9.6%, 34.5% and 42.2%, respectively, indicative of a dose response (P<0.001; paired Student's t-test comparing percent values at 20 nM and 1 μM, 20 nM and 10 μM, and 1 μM and 10 μM). The positive control PMA/iono-mycin displayed a high level of CD8⁺ T-cell activation (mean 90.2%). ProMix HIV peptide pool and ProMix CEF peptide pool (ProImmune) showed only low levels of CD8⁺ T-cell activation (range: 0–15.6%), likely due to the short treatment period of 24 h (data not shown).

Figure 2

VES-induced CD8⁺ T-cell activation (n=31)

Activation of HIV-specific CD8⁺ T-cells by VES

To determine if VES treatment also activates HIV-specific CD8⁺ T-cells, we quantified the level of CD8⁺ T-cells that were activated and bound to HIV-specific pentamers. Briefly, PBMCs were treated with VES at concentrations of 20 nM, 1 μM and 10 μM (10 μM tested in 14/31 donors) and then assessed for levels of CD3, CD8, CD69 expression and pentamer binding by flow cytometry. CD8⁺ T-cells were gated for activation and pentamer binding to quantify the level of activated HIV-specific cells within the CD8⁺ T-cell population.

VES treatment led to HIV-specific CD8⁺ T-cell activation in a dose dependent manner (P<0.05; paired Student's t-test comparing percent values at 20 nM and 10 μM, and 1 μM and 10 μM; P<0.01; paired Student's t-test comparing percent values at 20 nM and 1 μM; Figure 3; Table 1). Of the 31 donors, 20 displayed VES-induced activation of HIV-specific CD8⁺ T-cells at a level ≥0.5% at one or more VES concentrations. VES treatment led to HIV-specific CD8⁺ T-cell activation in 5 donors at 20 nM, 17 donors at 1 μM, and 8 donors at 10 μM (14/31 donors were tested at 10 μM). The level of activation of HIV-specific CD8⁺ T-cells ranged from 0.7–6.4% in the 5 donors at 20 nM VES, 0.5–15.3% in the 17 donors at 1 μM VES, and 0.5–20.8% in the 8 donors at 10 μM VES. The mean level of activation of HIV-specific CD8⁺ T-cells for these donors at 20 nM, 1 μM and 10 μM VES was 2.8%, 3.7% and 6.1%, respectively. Most of the 20 donors displayed a dose response showing that an increase in VES concentration was associated with an increase in activation of HIV-specific CD8⁺ T-cells.

Figure 3

VES-induced HIV-specific CD8⁺ T-cell activation (n=31)

Table 1

VES-induced activation of HIV-specific CD8⁺ T-cells (n=20): % pentamer⁺ CD69⁺ of CD8⁺ T-cells ^a

	VES treatment
Donor number b	20 nM	1 μM	10 μM	PMA+iono
1	-0.1	1.7	na	na
2	6.4	12.2	na	na
3	2.2	2.1	na	na
4	2.2	3.8	na	na
5	2.6	4.8	na	na
6	0.2	1.9	na	na
7	0.1	0.6	na	na
8	0.0	0.5	na	na
9	0.7	1.7	na	na
10	0.2	0.9	na	na
11	0.1	0.7	na	na
12	0.1	0.8	na	na
13	0.1	5.3	7.4	17.3
14	0.0	15.3	20.8	5.2
15	0.0	0.3	0.6	5.2
16	0.0	1.2	3.0	13.7
17	0.0	0.2	0.5	8.7
18	0.1	4.6	8.4	12.4
19	0.0	0.3	0.7	17.2
20	0.1	4.5	7.3	13.2

a

% Pentamer⁺ CD69⁺ of CD8⁺ T-cells values are calculated from the percent bound pentamers to CD69⁺ CD8⁺ T-cells. This indicates % HIV-specific activated cells within the population of CD8⁺ T-cells.

b

Donors listed here displayed pentamer binding ≥0.5% at one of the concentrations of vesatolimod (VES) tested (n=20). 10 μM VES and PMA/ionomycin treatment (PMA+iono) was done on 14 of 31 donors. Values are normalized to the DMSO control. na, not applicable.

The donor's HIV-specific CD8⁺ T-cell activation levels were normalized to the DMSO control to display activation induced by VES or PMA/ionomycin. The DMSO-negative control values were low to negligible (range: 0.005–2.4% HIV-specific CD8⁺ T-cell activation; data not shown). Out of the 31 donors, 27 had DMSO values below 0.3%, and 4 had DMSO values in the range of 0.9–2.4% HIV-specific CD8⁺ T-cell activation. Treatment with 20 μM VES resulted in > twofold increase in percentage for these four donors. The positive control PMA/ionomycin displayed a range of HIV-specific CD8⁺ T-cell activation of 1.3–17.3% among the donors tested (14/31 donors were tested with PMA/ion-omycin). HIV and CEF ProMix peptide pool (ProImmune) treatment led to low levels of activation of HIV-specific CD8⁺ T-cells ranging from 0–0.9% with CEF peptide pool and 0 to 1% with HIV peptide pool (data not shown). The low levels of activation by the HIV and CEF peptide pools may have been due to the short treatment period of 24 h. When treatment with the peptide pools was done for 7 days instead of 24 h, activation of HIV-specific CD8⁺ T-cells was seen in some donors with activation levels ranging from 0–1.3% with CEF peptide pool and 0–5.3% with HIV peptide pool. The 7-day treatment with peptides was not carried out on all donors due to the observation that VES led to CD8⁺ T-cells activation within 24 h. The combination of HIV or CEF ProMix peptide pool (ProImmune) and VES treatment resulted in minimal difference compared with VES alone, again likely due to the 24-h treatment duration (data not shown).

Activation of functional HIV-specific CD8⁺ T-cells by VES

To determine if VES can induce the activation of functional HIV-specific CD8⁺ T-cells, we measured levels of cytokines, tumour necrosis factor (TNF)-α, IFN-γ and degranulation marker CD107a within this cell subset. Briefly, the donors’ PBMCs were treated with VES and then stained with pentamers, cell surface antibodies and intracellular antibodies, and analysed by flow cytometry. HIV-specific CD8⁺ T-cells were assessed for cytotoxic functionality by gating for TNF-α, IFN-γ or CD107a. The donor TNF-α, IFN-γ and CD107a percentage levels were normalized to the DMSO control to display levels induced by VES or PMA/ionomycin. The DMSO-negative control values were low to negligible (range: 0–0.09%; data not shown).

This functionality assessment was conducted on a subset of 14 of the 31 original donors. Eight of these 14 donors displayed VES-induced activation of HIV-specific CD8⁺ T-cells. Of the eight donors, five displayed activated HIV-specific CD8⁺ T-cells with some level of cytotoxic functionality (≥0.5% at one or more concentrations; Figure 4). Five donors displayed VES induced activation of HIV-specific CD8⁺ T-cells leading to upregulation of TNF-α (range: 0.6-3.9%) and four donors displayed VES-induced activation of HIV-specific CD8⁺ T-cells associated with upregulation of CD107a (range: 0.5-4.8%). None of these five donors showed IFN-γ induction. VES treatment led to cytokine or degranulation upregulation in a dose dependent manner (P<0.05; paired t-test comparing percent values at 1 μM and 10 μM [TNF-α], and 20 nM and 10 μM [TNF-α]).

Figure 4

Functional activity of activated HIV-specific CD8⁺ T-cells induced by VES (n=14)

HIV and CEF ProMix peptide pool (ProImmune) treatment led to low levels of TNF-α (0-0.11% and 0-0.21%, respectively), IFN-γ (0-0.02% and 0-0.05%, respectively) and CD107a (0-0.12% and 0-0.47%, respectively; data not shown). The low levels of cytokines and degranulation induced by the HIV and CEF peptide pools may be due to the short treatment duration of 24 h. The combination of HIV or CEF ProMix peptide pool (ProImmune) and VES treatment resulted in minimal difference compared with VES alone, due to the 24 h treatment period (data not shown).

Discussion

VES is a TLR7 agonist which stimulates the receptor in pDCs leading to the activation of the innate immune system and production of acute phase cytokines including type I IFN [14]. The proximal cytokine response leads to the activation of the adaptive immune system to fight off viral infection [10]. This adaptive immune response can include CD8⁺ T-cells which can recognize pathogen antigens and in the case of HIV-1, help eliminate infected CD4⁺ T-cells. Here we investigated the ability of VES to indirectly activate functional HIV-specific cytotoxic CD8⁺ T-cells.

We observed that treatment of PBMCs from HIV-infected donors with VES resulted in robust activation of CD8⁺ T-cells as demonstrated by expression of the activation marker CD69 in a concentration-dependent manner, in agreement with prior observations [14]. VES led to the activation of HIV-specific CD8⁺ T-cells in 65% (20/31) of the donors tested, also in a dose-dependent manner, as evidenced by pentamer staining. Since HIV pentamers were engineered to contain known dominant HIV epitopes, they therefore do not cover the complete HIV protein sequence. In addition, all MHC alleles have not been studied to the same extent, leading to the potential underrepresentation of particular alleles [22] due to the absence of pentamers corresponding to all alleles and the entire HIV protein sequence. Due to these limitations, the observed activation of HIV-specific CD8⁺ T-cells by VES in 65% of the donors is likely to be an underestimate. Overall, the activation of HIV-specific CD8⁺ T-cells by VES, suggest that if functional, these cells may be able to target and clear HIV-infected cells.

To test whether VES-induced activated HIV-specific CD8⁺ T-cells showed functionality, cytokine production (TNF-α, IFN-γ) and degranulation (CD107a), which have been shown to be associated with CD8⁺ T-cell cytotoxic activity [23,24], were assessed in a subset of donors (n=14). Eight of the 14 donors showed VES-induced HIV-specific CD8⁺ T-cell activation, with 5 of the donors displaying degranulation upregulation (CD107a) and/or TNF-α production. Therefore, VES was able to activate functional HIV-specific CD8⁺ T-cells in 36% (5/14) of the donors tested, with the potential to enhance the killing of infected CD4⁺ T-cells. In our studies, intracellular IFN-γ was not detected. Other studies have shown IFN-γ was rarely produced in HIV tetramer⁺ CD8⁺ T-cells upon stimulation with other agents such as peptides antigens [22,25].

VES has been observed to increase HIV expression in vitro [14], but further explorations in vivo have led to different outcomes [16,17,21]. Viral expression may not be enough by itself to eliminate HIV-positive cells as exemplified by a study from Shan et al. [26] who looked at CD4⁺ T-cell latency reversal and showed that activated HIV-specific CD8⁺ T-cells are necessary for HIV elimination. Here we showed that VES can induce the activation of functional HIV-specific CD8⁺ T-cells which may help eliminate HIV-infected CD4⁺ cells. These data support the potential use of VES to activate functional HIV-specific CD8⁺ T-cells as part of an HIV remission strategy.