Crosstalk between R848 and abortive HIV-1 RNA-induced signaling enhances antiviral immunity.

Pathogens trigger multiple pattern recognition receptors (PRRs) that together dictate innate and adaptive immune responses. Understanding the crosstalk between PRRs is important to enhance vaccine efficacy. Abortive HIV-1 RNA transcripts are produced during acute and chronic HIV-1 infection and are known ligands for different PRRs, leading to antiviral and proinflammatory responses. Here, we have investigated the crosstalk between responses induced by these 58 nucleotide-long HIV-1 RNA transcripts and different TLR ligands. Costimulation of dendritic cells (DCs) with abortive HIV-1 RNA and TLR7/8 agonist R848, but not other TLR agonists, resulted in enhanced antiviral type I IFN responses as well as adaptive immune responses via the induction of DC-mediated T helper 1 (TH 1) responses and IFNγ+ CD8+ T cells. Our data underscore the importance of crosstalk between abortive HIV-1 RNA and R848-induced signaling for the induction of effective antiviral immunity.

dendritic cells

DEAD‐box RNA helicase 3

IFN‐stimulated genes

lipoteichoic acid

Pam3CysSerLys4

pathogen‐associated molecular patterns

protein kinase RNA‐activated

polyinosinic:polycytidylic acid

pattern recognition receptors

RIG‐I‐like receptors

single‐stranded ribonucleic acid

trans‐activation response element

T helper

INTRODUCTION

The ability of dendritic cells (DCs) to induce potent innate and adaptive immune responses is paramount to combating invading pathogens. Host pattern recognition receptors (PRRs) present on the cell surface, in the cytoplasm or in endosomes, detect pathogen‐associated molecular patterns (PAMPs), resulting in responses directed against the encountered pathogen. , , ,

TLRs are a well‐described family of PRRs that induce DC activation and specific cytokine production, leading to innate and adaptive immune responses. , Triggering of multiple TLRs together can result in synergistic immune responses that are more potent than the responses induced by triggering of one single TLR, but more importantly also lead to tailored T cell responses. , , Besides crosstalk between TLRs, crosstalk between TLRs and members of other PRR families, such as C‐type lectin receptors, NOD‐like receptors, and RIG‐I‐like receptors (RLRs), has also been reported, which dictates cytokine responses based on the specific crosstalk induced by these receptors. , , However, the effect of crosstalk between TLRs and other PRRs on antiviral immunity is less well established.

Human immunodeficiency virus 1 (HIV‐1) triggers a variety of PRRs, including cGAS, , TLR7/8, , , RIG‐I, DEAD‐box RNA helicase 3 (DDX3), , , and protein kinase RNA‐activated (PKR). , During HIV‐1 infection, endosomal TLR7 and TLR8 recognize single‐stranded (ss) HIV‐1 RNA and induce immune responses upon infection. , , , Furthermore, the cytosolic PRRs DDX3 and PKR recognize abortive HIV‐1 transcripts that are generated during both acute and chronic HIV‐1 infection as well as in latently infected cells. , , , These abortive HIV‐1 transcripts are formed after transcription initiation, when transcription elongation is blocked or not induced. Abortive HIV‐1 transcripts consist of the first 58 nucleotides common to all HIV‐1 transcripts and contain the trans‐activation response (TAR) element that adopts a hairpin structure. , DDX3 detects abortive HIV‐1 RNAs and induces antiviral type I IFN responses and adaptive responses via DC‐mediated T cell polarization, , while detection by PKR induces proinflammatory responses. Interestingly, abortive HIV‐1 transcripts are secreted from cells since they are detected in exosomes, both in vitro as well as in vivo. , , , , Stimulation of macrophages with isolated short TAR RNA‐containing exosomes results in proinflammatory responses, confirming their stimulatory properties and suggesting that TAR RNA affects bystander cells.

TLR7 and 8 have been shown to sense HIV‐1 ssRNA , and are therefore triggered during HIV‐1 infection. Thus, during HIV‐1 infection, both TLR7/8 ligands and abortive HIV‐1 RNA are present. , Triggering TLR7/8 by R848 induces antiviral immune responses. , , Similarly, abortive HIV‐1 RNA also induces antiviral and proinflammatory responses by triggering DDX3 and PKR. , , However, it is unknown whether there is crosstalk between abortive HIV‐1 RNA and HIV‐1 ssRNA‐specific TLR7/8‐induced responses and how this affects antiviral immunity.

Here, we demonstrate that crosstalk between signaling by abortive HIV‐1 RNA and TLR7/8 agonist R848 enhanced antiviral innate responses, which was not observed with other TLR agonists. Moreover, costimulation of DCs with both R848 and abortive HIV‐1 RNA enhanced specific adaptive responses by inducing TH1 polarization and IFNγ+CD8+ T cell responses. Our data strongly suggest that crosstalk between R848 and abortive HIV‐1 RNA‐induced signaling is paramount for an effective antiviral immune response. Considering that HIV‐1 triggers TLR7/8 signaling and the production of abortive HIV‐1 RNAs, these effective antiviral immune responses could be especially important to limit HIV‐1 infection.

MATERIALS AND METHODS

Cells and stimulations

This study was performed in accordance with the Declaration of Helsinki and the Amsterdam University Medical Centers (location AMC) Medical Ethics Committee guidelines. Human buffy coats (Sanquin) were processed using a lymphoprep density separation step to obtain peripheral blood mononuclear cells (PBMCs). PBMCs were used to isolate naïve CD4+ T cells or a percoll gradient was obtained to isolate monocytes or peripheral blood lymphocytes (PBLs). Naïve CD4+ T cells were isolated using a CD4+ T cell isolation kit II (Miltenyi Biotec) and subsequent depletion of memory CD4+ T cells (CD45RO) was performed using PE‐conjugated anti‐CD45RO (Dako) and anti PE‐beads (Miltenyi Biotec), according to manufacturer's instructions. Monocytes were cultured in RPMI medium supplemented with 10% FCS (Biological Industries), l‐glutamine (2 mM; Lonza), penicillin and streptomycin (100 U/ml and 100 μg/ml, respectively; Thermo Fisher) and either stimulated directly or cultured in the presence of GM‐CSF and IL‐4 (800 and 500 U/ml, respectively; Invitrogen) at 37°C 5% CO2 for 6 days to generate monocyte‐derived DCs. On day 6, DCs were stimulated with optimal concentrations of TLR 1/2 agonist Pam3CysSerLys4 (PAM3CSK4, 10 μg/ml; Invivogen), TLR2/6 agonist lipoteichoic acid (LTA, 10 μg/ml; Invivogen), TLR3 agonist polyinosinic:polycytidylic acid (poly‐I:C, 10 μg/ml; Invivogen), TLR4 agonist LPS Salmonella enterica serotype typhimurium (10 ng/ml; Sigma), TLR5 agonist flagellin from Bacillus subtilis (10 μg/ml; Invivogen), TLR7/8 agonist R848 (10 μg/ml; Invivogen), and/or stimulated with synthetic abortive HIV‐1 RNA (HIV‐1 Cap‐RNA58 Biosynthesis) or a similar RNA sequence without the 5′cap (HIV‐1 Cap‐RNA58) complexed with transfection reagent lyovec (Invivogen). Both HIV‐1 RNA constructs were used at a concentration of 1 nM as this induced a potent IFN response in DCs. HIV‐1 Cap‐RNA58 mimics the first 58 nucleotides of the HIV‐1 genome and contains a 5′m7GTP cap, while lacking a poly A tail. Synthetic abortive HIV‐1 RNA was synthesized using an in vitro transcription (IVT) method including cocapping techniques, as previously described. ,

Quantitative real‐time PCR

mRNA extraction of DCs was performed using lysis buffer obtained from the mRNA capture kit (Roche) and mRNA was reversely transcribed to cDNA using the reverse transcriptase kit (Promega). Quantification of gene expression levels was performed with SYBR Green (Thermo Fisher) and primer sets (Sigma) using an ABI 7500 Fast Real‐Time PCR detection system (Applied Biosystems). Primer sets were designed with Primer Express 2.0 (Applied Biosystems; Table S1). Expression levels of target genes were normalized to expression levels of household gene GAPDH, with the formula: Nt = 2Ct(GAPDH)−Ct(target).

ELISA

The secretion of TNF, IL‐6, IL‐10, IL‐12p70, IL‐23, and IL‐27 in DC supernatants was analyzed 24 h after cell stimulation and quantified using ELISA according to manufacturer's instructions (eBiosciences). Secretion of IFNβ in the DC supernatant was analyzed 6 h after cell stimulations (R&D systems). OD450 nm values were read using BioTek Synergy HT.

T cell differentiation

DCs were stimulated with HIV‐1 Cap‐RNA58, R848, or a combination. As a positive control for T helper (TH) 1 skewing, DCs were stimulated with LPS (10 ng/ml; Sigma) and IFNγ (1000 U/ml; u‐CyTech) and with LPS and PGE2 (1 μM; Sigma) to induce TH2 skewing. After 48 h, DCs were collected, washed, and cocultured with isolated naïve CD4+ T cells from an allogeneic donor in a 1:4 ratio. DC‐T cell cocultures were performed in IMDM medium (Invitrogen) supplemented with 10% FCS, penicillin, and streptomycin (100 U/ml and 100 μg/ml, respectively; Thermo Fisher, IMDM complete) and in the presence of Staphylococcus aureus enterotoxin B (10 pg/ml; Sigma). On day 5, IL‐2 (10 U/ml; Chiron) was added and on day 11–13, resting T cells were restimulated using PMA (10 ng/ml; Sigma) and ionomycin (1 mg/ml; Sigma) for 6 h and brefeldin A (10 μg/ml; Sigma) for the final 4 h. Cells were fixed with 4% PFA, permeabilized with 0.1% saponin in PBS with 0.5% BSA, and stained with FITC‐conjugated anti‐IFNγ (1:5, 340449; BD Biosciences) and APC‐conjugated anti‐IL‐4 (1:25, 554486; BD Biosciences), indicative of TH1 and TH2 skewing, respectively. Flow cytometry was performed using FACSCanto II (BD Biosciences) and analyzed with FlowJo software v10.7.

Cytotoxic T lymphocyte responses

DCs were stimulated with HIV‐1 Cap‐RNA58, R848 or a combination, for 48 h and subsequently cocultured with allogeneic PBLs in IMDM complete, using a 1:8 ratio. As a positive control, T cells were stimulated with anti‐CD3 (1:10,000; Sanquin) and anti‐CD28 (1 μg/ml; Sanquin). After 3 days, IL‐2 was added to the cocultures. After 6 days, cells were restimulated using PMA (10 ng/ml; Sigma) and ionomycin (1 mg/ml; Sigma) for 6 h, and brefeldin A (10 μg/ml; Sigma) for the final 4 h, followed by fixation and permeabilization using the fixation/permeabilization solution kit (BD Biosciences) according to manufacturer's instructions. Cells were then stained using APC‐Cy7‐conjugated anti‐CD3 (1:100, 300317; Biolegend), PE‐Cy7‐conjugated anti‐CD8 (1:100, 344711; Biolegend), FITC‐conjugated anti‐IFNγ (1:5, 340449; BD Biosciences), PE‐conjugated anti‐Perforin (1:10, 12‐9994‐42; Thermo Fisher) and AF700‐conjugated anti‐Granzyme B (1:200, 560213; BD Biosciences) and flow cytometry was performed using FACS Canto II (BD Biosciences) and analyzed with FlowJo software v10.7.

Data analysis

Statistical analysis was performed using Student's t‐test for paired observations using GraphPad version 8. Statistical significance was set at p < 0.05.

RESULTS

R848 enhances IFNB transcription induced by abortive HIV‐1 RNA

To assess the crosstalk between HIV‐1‐specific TLR‐ and abortive HIV‐1 RNA‐induced signaling, we stimulated monocyte‐derived DCs with TLR7/8 agonist R848 together with HIV‐1 Cap‐RNA58, a synthetic abortive HIV‐1 RNA that triggers DDX3 and PKR. , , Stimulation with HIV‐1 Cap‐RNA58 alone induced antiviral IFNB transcription , (Figure 1(A)). Strikingly, costimulation of HIV‐1 Cap‐RNA58 with R848 significantly enhanced IFNB transcription (Figure 1(A)). In contrast, costimulation with control HIV‐1 RNA58, lacking the 5′ cap structure, did not enhance IFNB transcription, suggesting that the crosstalk is mediated by DDX3 as PKR recognizes both uncapped control HIV‐1 RNA58 and HIV‐1 Cap‐RNA58 . To address whether costimulation with HIV‐1 Cap‐RNA58 and other TLR ligands resulted in a similar enhancement of IFNB, DCs were stimulated with HIV‐1 Cap‐RNA58 in combination with TLR agonists PAM3CSK4 (TLR1/2), LTA (TLR2/6), poly‐I:C (TLR3), LPS (TLR4), or flagellin (TLR5). Similar to R848, stimulations with PAM3CSK4 and flagellin induced low levels of IFNB transcription, while LTA, poly‐I:C, and LPS induced high levels of IFNB transcription similar to HIV‐1 Cap‐RNA58 (Figure 1(A)). However, none of these TLR agonists enhanced HIV‐1 Cap‐RNA58‐induced IFNB transcription (Figure 1(A)), suggesting that crosstalk between abortive HIV‐1 RNA‐ and specifically TLR7/8‐induced responses leads to enhanced IFNB transcription. In addition to enhanced IFNB transcription levels, costimulation with R848 and HIV‐1 Cap‐RNA58 resulted in enhanced IFNβ protein secretion (Figures 1(B) and S1). Furthermore, we have stimulated monocytes with HIV‐1 Cap‐RNA58, R848, or a combination (Figure 1(C)). Interestingly, a similar enhancement of IFNB transcription was observed (Figure 1(C)), suggesting that R848 enhances abortive HIV‐1 RNA‐induced IFNB transcription in primary myeloid subsets.

FIGURE 1

R848 enhances . Monocyte‐derived dendritic cells (DCs) were (co)stimulated with HIV‐1 Cap‐RNA58 or control HIV‐1 RNA58 and different TLR agonists (TLR7/8 agonist R848, TLR1/2 agonist PAM3CSK4, TLR2/6 agonist LTA, TLR3 agonist poly‐I:C, TLR4 agonist LPS, and TLR5 agonist flagellin) (A–C). Every 2 h (and only 8 h after costimulation with control HIV‐1 RNA58), mRNA was extracted to analyze IFNB transcription by quantitative real‐time PCR. mRNA expression was relative to GAPDH. Responses induced by HIV‐1 Cap‐RNA58 at 8 h were set at 1 (A). IFNβ protein secretion was detected in the supernatant of DCs 6 h after stimulation and quantified using ELISA (B). IFNB transcription in primary monocytes was assessed 8 h after stimulations with R848 and HIV‐1 Cap‐RNA58. mRNA expression was relative to GAPDH and responses induced by HIV‐1 Cap‐RNA58 and R848 were set at 1 (B and C). Data are representative of collated data of 4 (A) or 3 (B and C) donors (mean ± sd). Significance was determined between responses induced by cotreatment with HIV‐1 Cap‐RNA58 and R848, and HIV‐1 Cap‐RNA58‐induced responses (A). *p < 0.05, **p < 0.01, Student's t‐test

Costimulation with R848 and abortive HIV‐1 RNA modulates cytokine gene expression

Next, we examined whether crosstalk between R848 and HIV‐1 Cap‐RNA58 modulates expression of cytokine genes. Treatment of DCs with R848 resulted in transcription of all measured cytokine genes. Treatment with HIV‐1 Cap‐RNA58 alone induced low levels of TNF and IL6 transcription and higher levels of IL12A and IL27A (IL27p28) transcription, without inducing IL23A, IL10, IL12B, or IL27B (IL27EBI3) transcription (Figure 2). Strikingly, costimulations with R848 and HIV‐1 Cap‐RNA58, but not the uncapped HIV‐1 RNA58, synergistically enhanced transcription of IL6, IL10, IL12A, IL12B, IL27A, and IL27B compared with either stimulus alone, while R848‐induced TNF and IL23A transcription was unaffected (Figure 2). Thus, crosstalk between HIV‐1 Cap‐RNA58 and R848 modulates the expression of cytokine genes and enhanced antiviral type I IFN responses via enhanced transcription of IFNB and IFN‐stimulated gene (ISG) IL27A.

FIGURE 2

Costimulation with R848 and abortive HIV‐1 RNA modulates cytokine gene expression. DCs were (co)stimulated with R848 and HIV‐1 Cap‐RNA58 or control HIV‐1 RNA58 Every 2 h, (and only 8 h after costimulation with control HIV‐1 RNA58) mRNA was extracted to analyze TNF, IL6, IL10, IL12A, IL12B, IL23A, IL27A, and IL27B transcription by quantitative real‐time PCR. mRNA expression was relative to GAPDH. Responses induced by HIV‐1 Cap‐RNA58 with R848 at 8 h were set at 1. Data are representative of collated data of 3 (IL10, IL12A) or 4 donors (mean ± sd). Significance was determined between responses induced by cotreatment with HIV‐1 Cap‐RNA58 and R848, and R848‐induced responses. *p < 0.05, **p < 0.01, ***p < 0.001, Student's t‐test

Crosstalk between abortive HIV‐1 RNA and R848 enhances IL‐12p70 and IL‐27 expression

We next examined the protein expression levels upon costimulation with R848 and HIV‐1 Cap‐RNA58. Of note, due to high levels of donor variation and to perform statistical analysis, data are shown as fold change in protein expression (Figures 3(A)–3(C)) and as individual concentrations in pg/ml in Figure S2 (Figures S2(A)–S2(C)). TNF and IL‐23 were secreted upon stimulation with R848 and costimulation with HIV‐1 Cap‐RNA58 did not enhance TNF and IL‐23 levels (Figures 3(A) and S2(A)), similar to our mRNA results (Figure 2). Although mRNA transcripts of IL6 and IL10 were enhanced by costimulation (Figure 2), this was not observed for IL‐6 and IL‐10 protein expression (Figures 3(A) and S2(A)). In contrast, costimulation with R848 and HIV‐1 Cap‐RNA58 significantly enhanced IL‐12p70 and IL‐27 (Figures 3(A) and S2(A)), following transcriptional patterns (Figure 2). Both IL‐12p70 and IL‐27 are important cytokines in directing adaptive responses. , , , Next, we further assessed whether the enhanced IL‐12p70 and IL‐27 levels upon costimulation with HIV‐1 Cap‐RNA58 and R848 are unique for TLR7/8 triggering, or that other TLRs lead to a similar enhancement of cytokine production. We observed that costimulation with HIV‐1 Cap‐RNA58 and PAM3CSK4 or LTA significantly enhanced IL‐12p70 secretion, while poly‐I:C‐, LPS‐, or flagellin‐induced IL‐12p70 levels were only marginally enhanced by costimulation with HIV‐1 Cap‐RNA58 (Figures 3(B) and S2(B)). Notably, although the different TLR agonists induced IL‐27 expression, these levels were not significantly affected by costimulation with HIV‐1 Cap‐RNA58 (Figures 3(C) and S2(C)). These data suggest that crosstalk between TLR‐ and abortive HIV‐1 RNA‐induced signaling synergistically enhanced IL‐12p70 expression, whereas only signaling induced by abortive HIV‐1 RNA and R848 enhanced IL‐27 expression.

FIGURE 3

Crosstalk between abortive HIV‐1 RNA and R848 enhances IL‐12p70 and IL‐27 expression. (A–C) DCs were (co)stimulated with R848 (A–C) or TLR agonists (TLR1/2 agonist PAM3CSK4, TLR2/6 agonist LTA, TLR3 agonist poly‐I:C, TLR4 agonist LPS, TLR5 agonist flagellin) (B and C) and HIV‐1 Cap‐RNA58. TNF, IL‐6, IL‐10, and IL‐23 (A), IL‐12p70 (A, B) or IL‐27 (A, C) expression levels in the DC supernatant were measured after 24 h by ELISA (A–C). Responses induced by HIV‐1 Cap‐RNA58 with TLR agonists were set at 1. Data are shown as fold change in protein expression due to donor variation in cytokine expression levels. Data are representative of collated data of 4 donors and R848 stimulations are used in A–C. HIV‐1 Cap‐RNA58‐induced IL‐27 expression was not detected (n.d.) (C). (mean ± sd). *p < 0.05, Student's t‐test

Costimulation with R848 and abortive HIV‐1 RNA leads to TH1 differentiation

Next, we set out to assess the effect of crosstalk between R848‐ and abortive HIV‐1 RNA‐induced signaling on DC‐mediated T helper (TH) differentiation, which was assessed by production of intracellular cytokines IFNγ and IL‐4, indicative of a TH1 or TH2 response, respectively. Treatment with HIV‐1 Cap‐RNA58 alone resulted in DC‐mediated T helper 2 (TH2) differentiation (Figure 4(A)). R848‐primed DCs also induced IL‐4+ CD4+ T cells, while decreasing IFNγ+CD4+ T cells, indicative of TH2 skewing (Figure 4(A)). Remarkably, costimulation of R848‐primed DCs with HIV‐1 Cap‐RNA58 redirected TH differentiation towards TH1 polarization, as shown by high levels of intracellular IFNγ and low levels of IL‐4, comparable to the TH1 priming induced by the LPS + IFNγ control (Figures 4(A) and 4(B)). These results imply that crosstalk between abortive HIV‐1 RNA‐ and TLR7/8‐induced signaling specifically redirects abortive HIV‐1 RNA‐induced TH2 differentiation, resulting in strong TH1 polarization.

FIGURE 4

Costimulation with abortive HIV‐1 RNA and R848 leads to T. (A and B) DCs were (co)stimulated with R848 and HIV‐1 Cap‐RNA58. DCs were treated with either LPS + IFNγ or LPS + PGE2 to induce TH1 and TH2 skewing, respectively (A and B). After 48 h, DCs were cocultured with naïve CD4+ T cells for 11–13 days. Resting T cells were re‐stimulated and intracellular expression levels of IFNγ and IL‐4 were assessed by flow cytometry to indicate TH1 and TH2 differentiation, respectively (A and B). The percentage of cells in each quadrant is indicated by numbers in the plots (A) and only the percentage of single IL‐4+ or IFNγ+ cells were plotted in B. Data are representative of 3 donors of different experiments (A) or are collated data of 3 donors with every symbol representing a different donor (B), (mean ± sd). *p < 0.05, **p < 0.01, Student's t‐test

Abortive HIV‐1 RNA and R848‐induced crosstalk enhances CD8+ T cell IFNγ responses

An important aspect of antiviral adaptive immunity is the induction of cytotoxic CD8+ T cells. , Thus, we next investigated the effect of abortive HIV‐1 RNA‐ and R848‐induced crosstalk on induction of DC‐mediated antiviral CD8+ T cells. DCs (co)stimulated with R848 and/or HIV‐1 Cap‐RNA58 were cocultured with PBLs, and the percentages of cells positive for intracellular expression of cytolytic molecules granzyme B and perforin, and IFNγ in CD8+ T cells were assessed (Figure S3). DCs treated with R848 but not HIV‐1 Cap‐RNA58 induced IFNγ+CD3+CD8+ T cells, while, strikingly, crosstalk between R848 and HIV‐1 Cap‐RNA58 significantly enhanced the IFNγ+CD3+CD8+ T cell responses (Figures 5(A) and 5(B)). Crosstalk did not affect the levels of granzyme B+ and perforin+CD3+CD8+ T cells compared with the levels induced by R848 and HIV‐1 Cap‐RNA58 stimulations alone (Figures 5(A) and 5(B)). Our data indicate that crosstalk between R848 and abortive HIV‐1 RNA enhances antiviral CD8+ T cells, albeit without affecting the expression of cytolytic molecules.

FIGURE 5

Synergy between abortive HIV‐1 RNA and R848‐induced signaling enhances CD8 DCs were (co)stimulated with R848 and HIV‐1 Cap‐RNA58 for 48 h and cocultured with PBLs. PBLs were directly treated with anti‐CD3 and anti‐CD28 as a positive control. After 6 days, cells were restimulated and the percentages of CD3+CD8+ cells with intracellular expression levels of granzyme B, perforin, and IFNγ were assessed by flow cytometry. The percentage of cells in each gate is indicated by numbers in the plots (A). Data are representative of 3 donors of different experiments (A) or are collated data of 3 donors with each symbol representing a different donor (B), (mean ± sd). *p < 0.05, **p < 0.01, Student's t‐test

DISCUSSION

Triggering of specific PRRs induces innate and adaptive responses. However, when innate immune cells encounter pathogens, distinct PAMPs are exposed, resulting in the triggering of multiple PRRs and leading to a tailored immune response to that pathogen. , , , , Here, we show that crosstalk between abortive HIV‐1 RNA and TLR7/8 agonist R848 leads to antiviral innate and adaptive immunity.

Abortive HIV‐1 RNA is a potent inducer of antiviral type I IFNs as it induces high levels of IFNB and ISG transcription. , Most TLR agonists did not modulate the abortive HIV‐1 RNA‐induced IFNB transcription; however, crosstalk with TLR7/8 agonist R848 enhanced transcription of IFNB and the ISG IL27A, and also enhanced IFNβ protein secretion. Moreover, IFNB transcription was also observed in primary blood monocytes in response to R848 and abortive HIV‐1 RNA, suggesting that the effects described apply to primary myeloid subsets. Furthermore, synergy between R848‐ and abortive HIV‐1 RNA‐induced signaling enhanced the transcription of IL6, IL10, IL12A, IL12B, IL27A, and IL27B. Interestingly, this synergy was detected for bioactive IL‐12p70 and IL‐27 protein as well, which relies on transcription of IL12A/IL12B and IL27A/IL27B, respectively. In contrast to the enhanced transcription of IL6 and IL10 by abortive HIV‐1 RNA and R848 costimulation, no difference in IL‐6 and IL‐10 protein levels was observed, possibly due to a crosstalk‐specific negative control of translation. These data indicate that crosstalk between abortive HIV‐1 RNA and R848 modulates cytokine gene transcription and protein secretion.

We have previously reported that abortive HIV‐1 RNA is sensed by the cytosolic PRR DDX3 and result in antiviral type I IFN responses. Additionally, abortive HIV‐1 RNA is also recognized by PKR, which upon triggering induces inflammasome activation and IL‐1β secretion. In addition to type I IFN, we have shown that sensing of abortive HIV‐1 RNA by DDX3 also induced the transcription of various cytokine genes such as TNF, IL6, and IL12A, to protein production and to TH2 responses. Here, similar responses were detected and enhanced by the presence of R848 and were not observed when we used a capless HIV‐1 RNA58 control, which is also sensed by PKR, suggesting that the responses induced by R848 and abortive HIV‐1 RNA‐induced crosstalk are mediated by DDX3‐dependent sensing of abortive HIV‐1 RNA.

Abortive HIV‐1 RNA induces DC‐mediated TH2 differentiation, a response predominantly important for B cell help and immunity against extracellular pathogens. , In contrast, TH1 responses are crucial against intracellular pathogens such as viruses. Our data show that crosstalk between abortive HIV‐1 RNA and R848, re‐directed TH polarization towards a TH1 response, possibly due to a synergistic effect between enhanced IL‐12 and IL‐27 production, leading to the enhanced IFNγ‐producing TH1 responses. Synergy between IL‐12 and IL‐27 as a result of crosstalk between abortive HIV‐1 RNA and R848 might also be the contributing factor leading to the enhanced percentage of IFNγ+CD8+ T cells, which also play a crucial role in the cellular control of viruses. , , , , , IFNγ is important for enhanced CD8+ T cell function and killing. , , , Therefore, our data suggest that abortive HIV‐1 RNAs contribute to antiviral immunity by promoting the propagation of IFNγ+CD8+ T cells. Of note, granzyme B and perforin play an important role in killing of HIV‐1 infected cells. , , Although crosstalk does not further enhance granzyme B+ and perforin+CD8+ T cell numbers, granzyme B and perforin were abundantly expressed in the CD8+ T cells. Their presence, in combination with the observed propagation of IFNγ+CD8+ T cells, could potentially lead to higher antiviral responses during viral infection.

As genomic HIV‐1 RNA is a ligand for TLR7/8, , , our data reveal that crosstalk between R848 and abortive HIV‐1 RNA‐induced signaling leads to potent antiviral immune responses that might shape responses against HIV‐1. We have previously shown that the triggering of TLR8 plays a role in HIV‐1 transcription initiation steps, thus enabling for the production of abortive HIV‐1 RNA transcripts. Therefore, these linked events demonstrate that multiple different PRRs are triggered during HIV‐1 infection, together determining the final immune response and highlighting the importance of investigating crosstalk between these receptors. Here, we show that both HIV‐1‐specific events have the potency to induce crosstalk, leading to synergistic antiviral responses. Further studies on the effects of crosstalk between different PRRs, specifically those triggered by HIV‐1 products, will provide insight into the immune responses elicited or blocked by this crosstalk and thus would be valuable for vaccination strategies.

Defining the effects of crosstalk between PRRs provides valuable targets needed to enhance vaccine efficacy. TLR8 agonists are currently used as adjuvants to treat viral infections and as antitumor treatment. , , , Here, our data imply that abortive HIV‐1 RNAs have a strong capacity to modulate R848‐induced signaling, leading to enhanced antiviral immunity. Our findings provide a rationale for the use of abortive HIV‐1 RNA as an adjuvant in current TLR8‐based antiviral and possibly antitumor strategies. Overall, our results underscore the importance of crosstalk between different PRRs for the final direction and potency of innate and adaptive immune responses.

AUTHORSHIP

M. S. contributed to study conceptualization, investigation, data analysis and wrote the original draft. J. L. V. H. and E. Z. W. contributed to study investigation. S. I. G. and T. B. H. G. contributed to study conceptualization, and, together with M. S. assisted with reviewing and editing the final version of this manuscript.

DISCLOSURE

The authors declare no conflict of interest.

Table S1. Quantitative Real‐Time PCR primers to analyze mRNA expression

Click here for additional data file.

Supplementary figure 1. Crosstalk between abortive HIV‐1 RNA and R848 enhances IFNβ. protein expression.

Click here for additional data file.

Supplementary figure 2. Crosstalk between abortive HIV‐1 RNA and R848 enhances IL‐12p70 and IL‐27 expression.

Click here for additional data file.

Supplementary figure 3. Gating strategy of CDB+ T cell responses.

Click here for additional data file.