HIV-1 derived oligonucleotides induce a type I IFN/STING dependent immune suppression reversible by targeting IFNARI

TL;DR

HIV-1 causes immune dysfunction by triggering type I interferon signaling, which can be reversed by blocking IFNAR1 with anifrolumab.

Contribution

Identifies a novel mechanism of HIV-induced immune suppression via STING pathway activation and demonstrates reversal using IFNAR1 targeting.

Findings

HIV-derived ssDNA activates STING in dendritic cells, leading to type I IFN production and suppressive T cell expansion.

Blocking IFNAR1 with anifrolumab restores immune cell function and cytokine secretion in co-cultures.

Persistent type I IFN signaling contributes to immune dysfunction in HIV-1 infection.

Abstract

The hallmark of HIV-1 (HIV) infection is the progressive development of multicellular and systemic immune dysfunction, culminating in AIDS. Dendritic cells (DCs) play a pivotal role in HIV dissemination to CD4 + T cells, which are subsequently depleted by the virus leading to HIV disease progression. Type I interferons (IFNs) are critical for host defense during acute infection but contribute to chronic immune activation during the later stages of HIV disease. This persistent activation leads to immune cell exhaustion. HIV can activate type I IFN responses via several pathways, including the STING pathway, which is activated by, e.g., virus-derived oligonucleotides. Here, we investigated the underlying mechanisms creating HIV-mediated immune dysfunction and role of type I IFNs using a DC and T cell co-culture model. HIV exposure in the DC-T cell co-culture promoted the expansion of suppressive T cells with diminished proliferation and effector functions. The impairment required type I IFNs and subsequent IFNα/β receptor signaling, and our data indicate that this was initiated by HIV-derived ssDNA activation of IFI16/cGAS followed by STING signaling in the DCs. Targeting IFNAR1 with anifrolumab restored the immune functions of both DCs and T cells, as well as T cell proliferation and T cell effector functions, including their secretion of IL-2, IFNγ, and granzyme B. Our findings support that the immune impairments existing in untreated or antiretroviral therapy (ART) treated HIV-infected individuals are mediated, if not fully in part by type I IFN’s negative effect on DC and T cells. Therapeutics targeting IFNα/β receptors, such as anifrolumab, hold potential as combination treatment alongside ART, to achieve a more complete immune restoration and contribute to improved quality of life among people living with HIV.

HIV-1 infection causes progressive immune dysfunction that leads to AIDS. Dendritic cells (DCs) play a central role in spreading HIV to CD4 ⁺ T cells, which are subsequently depleted. Type I interferons (IFNs) are vital for antiviral defense during acute infection, but chronic activation of this pathway contributes to immune exhaustion and dysfunction. In this study, we investigated how HIV-1 triggers type I IFN signaling and impairs immune responses using a DC–T cell co-culture model. We found that HIV and HIV derived ss DNA activates the STING pathway in DCs, inducing type I IFN production that drives the expansion of suppressive T cells with reduced proliferation and effector function. Blocking IFN-α/β receptor (IFNAR1) signaling with the monoclonal antibody Anifrolumab restored DC and T cell functions, enhancing proliferation and cytokine secretion. These findings reveal that persistent type I IFN signaling contributes to immune dysfunction in HIV-1 infection and highlight IFNAR-targeting therapies as promising adjuncts to antiretroviral treatment to promote immune restoration.

Figures

Both chronic and founder HIV-1 strains suppressed dendritic cells ability to prime naïve T cells.HIV BaL a late-stage derived lab adapted strain or HIV THRO a transmitted/founder strain exposed mature DCs were co-cultured with naïve T cells in the ratio of 1:10. The DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation was assessed via 3H thymidine incorporation (A) and functionality by enumerating lL2-producing cells by ELISPOT assays (B). mRNA transcript of PDL1, CD86, PD1, and TIM3, and (C) IFNα4, MXA, IFI13, and IFI16 (D) by qPCR, N = 4. Whiskers: Min to Max. Statistical significance was determined using the ANOVA. * = p-value <0.05,  = p-value <0.01, * = p-value <0.001 = p-value <0.0001.

Type I IFN inducing pathways were enriched and predicted to be activated when HIV was present in DC-T cell co-culture and exogenous type I IFN-induced suppressed T cell responses similar to HIV.Transcriptomic data from our published RNA seq [20], where DCs purified from the DC-T cell co-culture after one day of re-stimulation, were assessed for top regulator network (A). HIV BaL exposed mature DCs were co-cultured with naïve T cells in the presence or absence of exogenous human recombinant IFNα4. The DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation and the IFNα2 replenished. One day after re-stimulation, the T proliferation was assessed via 3H thymidine incorporation (B). mRNA transcript of PDL1, and PD1 (C) by qPCR, N = 9. Statistical significance was determined using the ANOVA. Whiskers: Min to Max. * = p-value <0.05,  = p-value <0.01, * = p-value <0.001,  = p-value <0.0001.

HIV derived replication intermediate ssDNA participated in the induction of the suppressive effects seen by HIV in the DC-T cell co-culture.Mature DCs were exposed to HIV BaL or transfected with the ssDNA, dsDNA, DNA:RNA hybrid, many loops ssDNA, or p24 using DOTAP to deliver these HIV components into the cytosol. The different DC groups were co-cultured with naïve T cells (1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation was assessed via 3H thymidine incorporation (A-D), N = 3-10. Statistical significance was determined using the ANOVA. Whiskers: Min to Max. * = p-value <0.05,  = p-value <0.01, * = p-value <0.001,  = p-value <0.0001.

HIV and ssDNA induce type I IFN production and ISG in DC-T cell co-cultures.Mature DCs were exposed to HIV BaL or transfected with the ssDNA, dsDNA, DNA:RNA hybrid, many loops ssDNA, or p24 using DOTAP to deliver these HIV components into the cytosol. The different DC groups were co-cultured with naïve T cells (1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as at the initial stimulation. mRNA transcript of (A) IFNα4, MXA, IFI13, and IFI16 by qPCR and (B) Protein levels of type I IFNs, IFNα4 and IFNβ were measured by ELISA, N = 4-7. ANOVA. Whiskers: Min to Max. * = p-value <0.05,  = p-value <0.01, * = p-value <0.001,  = p-value <0.0001.

HIV and ssDNA exposure of DCs induce an immune suppressive phenotype of both DCs and T cells after co-culture crosstalk.Mature DCs were exposed to HIV BaL or transfected with the ssDNA using DOTAP to deliver oligonucleotide into the cytosol. The different DC groups were co-cultured with naïve T cells (in a ratio of 1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as at the initial stimulation. Spectral flow cytometric analysis of (A) PDL1 and CD86 on CD1c+ cells, (B) PD1 and CTLA4 on CD3 + T cells and (C) CD28 on CD4 and CD8 + T cells. Spectral flow cytometric analysis of (D) HLADR+CD38+ on CD4+ and CD8 + T cells (E) PD1 expression on HLADR+CD38 + CD4+ and CD8 + T cells, (F) CD69 on CD4+ and CD8 + T cells, and (G) PD1 expression on CD69 + CD4+ and CD8 + T cells. ANOVA, N = 12-16. Whiskers: Min to Max. * = p-value <0.05,  = p-value <0.01, * = p-value <0.001,  = p-value <0.0001.

STING agonists suppressed T cell proliferation and increased PDL1 expression of the cells in the co-culture.Mature DCs were unexposed or exposed to HIV BaL, or transfected with STING agonists cGAMP, and co-cultured with naïve T cells. Additionally, during co-culture the cells were exposed to the stabilized 2’3’cyclic-di-AMP analog ADU100, MSAD-2, and STING antagonist SN-011. Co-cultures were re-stimulated on day 7 by the addition of the same DCs as during setup. One day after re-stimulation, the T proliferation was measured by 3H thymidine incorporation (A-D). mRNA transcript levels of IFNβ, MXA, and PDL1 in ADU100 (E), MSA2 (F), and SN-011 (G) treated DC-T cell co-culture assessed by qPCR N = 7. Protein levels of IFNα and IFNβ in ADU100 (E), MSAD2 (F), and SN-011 (G) treated DC-T cell co-culture assessed by ELISA, N = 4-5. Statistical significance was determined using the ANOVA. Whiskers: Min to Max. * = p-value <0.05,  = p-value <0.01, * = p-value <0.001,  = p-value <0.0001.

RIG activation played no role in HIV-induced immune suppression in the DC-T cell co-culture.Mature DCs were exposed to HIV BaL or DOTAP transfected with RIG agonist RIG012 to achieve delivery into the cytosol. The different DC groups were co-cultured with naïve T cells (1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation measured via 3H thymidine incorporation (A), mRNA transcript of CD86, PD1, PDL1, and TIM3, N = 4-7 (B), and IFNα4, MXA, IFI13, and IFI16 (C) by qPCR, N = 4. Statistical significance was determined using the ANOVA. Whiskers: Min to Max. * = p-value <0.05,  = p-value <0.01, * = p-value <0.001,  = p-value <0.0001.

IFI16 activation suppresses T cell proliferation but is not enough to induce co-inhibitory molecules.Mature DCs were exposed to HIV BaL or DOTAP transfected with IFI16 agonist dsDNA VACV-60 to deliver the VACV-60 into the cytosol. The different DC groups were co-cultured with naïve T cells (1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation measured by 3H thymidine incorporation (A). mRNA transcripts of PDL1, Galectin 9, TRAIL, and TRIM3 (B), and of IFNα4, IFI16, and MXA (C) were assessed by qPCR. Statistical significance was determined using the ANOVA, N = 4. Whiskers: Min to Max. * = p-value <0.05,  = p-value <0.01, * = p-value <0.001,  = p-value <0.0001.

Targeting IFNaR1 by anifrolumab restored immune function in HIV-induced suppressed DCs and T cells.Mature DC were exposed to HIV BaL and incubated for 24h. The untreated (mock) and HIV exposed DCs were harvested and exposed to 20µg/ml anifrolumab for 30 min before the DCs were added to naïve T cells and co-cultured (1:10). The DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation was assessed via H3 thymidine incorporation and IFNγ, IL2 and granzyme B producing T cells by ELISPOT (A). The effects of anifrolumab on mRNA transcripts levels for IFNα4, IFI16, and MXA, and IFI16 (B) and PD1L, Galectin-9, TIM3, and TRAIL (C) were assessed by qPCR, N = 7. Whiskers: Min to Max. Statistical significance was determined using the ANOVA. * = p-value <0.05,  = p-value <0.01, * = p-value <0.001,  = p-value <0.0001.

Full text

Introduction

Over three decades since its discovery, human immunodeficiency virus type 1 (HIV) infection continues to be a major global public health challenge. As of 2022, more than 38 million individuals live with HIV, with ~1.3 million new infections reported (UNAIDS, 2022). Encouragingly, the incidence of new infections is declining, attributed to that over 75% of the people living with HIV-1 have access to antiretroviral therapy (ART). The therapy provides effective control of viral replication and limits the negative effect HIV exerts on the immune system and general health (WHO, 2022).

Dendritic cells (DCs) are pivotal players in the initiation and regulation of immune responses, as the functionality of these cells in the lymphoid tissues determines the quality of the ensuing response [1–3]. DCs are among the first immune cells targeted by HIV upon infection of the genital mucosa, where they reside within the mucosal epithelium [3]. The mucosal DCs are less permissive to infection than CD4 + T cells, but relatively more permissive than blood DCs [4,5]. Still, the DCs exhibit remarkable efficiency in facilitating HIV infection while in cellular contact by transferring virions to the interacting T cells even in the presence of ART [6,7]. Of note, in HIV-infected individuals, DCs in lymphoid tissues can serve as a source of latently infected cells despite years of suppressive ART. These lymphoid located myeloid DCs also express programmed death-1 ligand 1 (PD-L1) in combination with an activated phenotype, which might impair the activation of HIV-specific T cells [8,9].

Chronic HIV infection is characterized by profound immune dysfunction and systemic immune activation. The immune activation is driven by several factors, and prolonged and elevated type I IFN levels are detrimental to immune responses during chronic HIV infection, as previously demonstrated both in HIV and simian immunodeficiency virus (SIV) infections, where type I IFN signaling contributes to disease progression [10,11]. Many immune cells, including T cell and DC subsets, are functionally impaired in HIV infected individuals, with altered cytokine profiles, activation status, and elevated levels of immune checkpoint molecules, caused by direct HIV effects or by virus-induced bystander effects [12–15]. Importantly, the impaired T cell function in HIV-infected individuals with or without ART can be partially restored through ex vivo blockade of immune checkpoint molecules [16]. Anifrolumab, a fully human immunoglobulin G1κ monoclonal antibody that binds to the type I interferon receptor subunit 1 and inhibits signaling by all type I interferons, is approved for moderate-to-severe non-renal systemic lupus erythematosus (SLE) since 2021 [17].

We have demonstrated that during the interaction between mature DCs and T cells in the presence of HIV, the virus induces impaired responses with expression of, e.g., PDL1 and Galectin-9 on DCs, and programmed death-1 (PD-1) and CTLA4 on T cells [18–20], which is in line with findings in chronic HIV infection [14,21]. In addition, we demonstrated that the transcriptome profile of HIV-exposed DCs engaged in crosstalk with T cells exhibit a robust prolonged type I IFN signature [20].

During retroviral and DNA virus infections, oligonucleotides are generated at various stages of the viral life cycle and in the context of HIV an array of replication intermediates such as ssDNA, dsDNA, RNA:DNA hybrids be found in the cytosol [22,23]. These viral nucleic acids are recognized by innate immune sensors such as retinoic acid-inducible gene I (RIG-I), which detects RNA:DNA hybrids [24], and cyclic GMP-AMP synthase (cGAS) and IFN-inducible protein 16 (IFI16), that trigger type I IFN responses via the stimulator of interferon genes (STING) pathway [22,23,25]. In DCs, the intracellular detection of HIV has been deemed to be dependent on cGAS [26,27]. In addition to cytosolic sensing, in endosomal compartments containing TLR8 in myeloid DCs and TLR7 and TLR9 in plasmacytoid DCs, the degraded virus is detected even in the absence of productive viral replication [28,29]. The activation of the STING pathway and subsequent expression of type I IFNs has been associated with the induction of immunoregulatory factors such as indoleamine 2, 3 dioxygenase (IDO), PD1, PDL1, and PDL2, contributing to the suppression of T cell proliferation and a general immunosuppressive microenvironment [30,31].

To investigate the role of type I IFN in suppressive responses induced by HIV-exposed DCs, we examined the HIV replication intermediates driving type I IFN activation and the signaling pathways involved in immune modulation/impairment. To achieve this, we utilized a HIV exposed or unexposed mature DC-T cell model of immune activation, mirroring immune response initiation occurring in lymphoid tissue in an HIV-infected individual.

Our results show that HIV suppressed the DCs and T cells, independent of the HIV-1 M strains used (BaL vs THRO), and that there was a strong and prolonged type I IFN response in HIV-exposed co-cultures. HIV replication intermediates/oligonucleotides participated in the HIV-induced suppression, with ssDNA being the most potent and the IFI16/cGAS/STING signaling pathways indicated as the main signaling pathways contributing to the induction of suppressive responses. The negative effects on DCs and T cells in the co-culture were reverted by addition of the anti-IFNAR1 antibody anifrolumab, blocking the IFNaR1 signaling pathway. Therapies targeting IFNα/β receptors, such as anifrolumab, hold potential as combination treatment alongside ART to accomplish complete immune restoration and thereby improve health in people living with HIV.

Materials and methods

Ethics statement

Buffy coats and leukocyte concentrates were purchased from the local blood bank/transfusion medicine unit. The blood donors filled out and signed the blood donation form, which states that the blood products from the whole blood donation can be used for research.. Ethical permission for the use of human blood products has been reviewed and approved by the Swedish Ethical Review Authority with reference number: M173-07.

Cell culture media and reagents

Cell culture was conducted in RPMI1640 medium (ThermoFisher, Stockholm, Sweden) supplemented with gentamicin (ThermoFisher), HEPES (ThermoFisher), and either 5% pooled human serum (PHS) (Innovative Research, Novi, USA) or 1% single donor human plasma. The ssDNA, dsDNA, various DNA loops, and DNA:RNA hybrids (Invivogen, Toulouse, France), derived from sequences described by Jakobsen et al. [23] (S1 Table), were diluted in pH adjusted potassium chloride buffer and added to the cells in amounts ranging from 2 to 16µg. Recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) (Prepotech, London, UK) at 100IU/mL and recombinant human interleukin-4 (rhIL-4) (Prepotech) at 300U/mL were utilized to differentiate monocytes into DCs in vitro. To investigate the STING signaling pathway, the following reagents and concentrations were employed: cGAMP (10µg/mL, Invivogen), MSAD2 (33µM, MedChemExpress Europe, Sollentuna, Sweden), ADS100 (10µM, MedChemExpress), SN011 (1 µM, MedChemExpress), 5′pppRNA (1µg/mL, Invivogen), RIG012 (1.25 µM, Axon Medchem, Groningen, Netherlands), and VACV70 (10µg/mL, Invivogen). Additional chemicals and compounds used included CL075 (1µg/mL, Invivogen), recombinant p24 (1µg/mL, Invivogen), and IFNAR blockade using anifrolumab at 20µg/mL, as a kind gift from AstraZeneca.

Dendritic cell propagation

Whole blood or leukocyte preparations from healthy volunteers were processed using a Ficoll-Paque (Amersham Pharmacia, Piscataway, NJ, USA) density gradient. The PBMCs were subjected to sequential centrifugation at 1800, 1500, 1100, and 900 RPM for 10 minutes each at 4°C. After centrifugation, the cells were counted and plated at a density of 40 million cells per plate in 10 mL of 5% PHS. The plates were incubated at 37°C for 1–2h to allow monocyte adherence. Subsequently, non-adherent cells were removed, and the plates were washed. The remaining adherent monocytes were cultured in 1% plasma medium and differentiated into monocyte-derived DCs (moDCs) by adding GM-CSF and IL-4 every other day for 5 days. On day 5, phenotype was investigated by flow cytometry to exclude undifferentiated cells by CD14 expression and spontaneously activated cells by CD83 expression. The maximum cut-off for both markers was set to ~15%.

Maturation of dendritic cells

Following the cut-off selection for viable cells, moDCs were harvested, counted, and re-seeded in 10 mL of the same conditioned culture medium used during differentiation, at a density of 4 million cells per plate. The TLR3 ligand poly I:C (3ng/ml: Invivogen) was then added to the cultures, and the moDCs were incubated at 37°C for 24h to promote cell maturation.

HIV strains and preparations

HIV-1BaL/Sup-T1 (Lot #4238 and #4369) and HIV-1 THRO-1 A66-R5 (Lot# 4380) were provided by the Biological Products Core Laboratory (BPC)/AIDS and Cancer Virus Program/Leidos Biomedical Research, Inc./Frederick National Laboratory for Cancer Research. The different HIV-1 strains were produced using the chronically infected cell lines. Prior to virus purification cells and large cell debris were removed by 5.0 µm capsule filtration and additional non-viral particulates removed from the filtrate by sequential 1.0 µm followed by 0.45 µm filtration. Virus was purified from the filtrate by continuous flow sucrose density gradient centrifugation as previously [32] described (1). The pellets were resuspend in buffer at a final concentration factor of 1000X.

HIV infection and oligo transfection of moDCs

After 24h of maturation CCR5-tropic HIV BaL 4238 or HIV THRO A66 4393 was added in a concentration of 750 ng/mL p24 (4 virions/cell) and DCs were incubated another 24h. For intracellular addition of nucleotides, proteins, and antagonists/agonists the N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium methylsulfate (DOTAP) (Merck, Darmstadt, Germany) liposomal transfection system was used. An amount of 30µl of DOTAP was mixed with 70µl of HBSS buffer (0.9% NaCl + 10mM HEPES) and thereafter the oligos or proteins was added in indicated concentrations diluted in 50µl of HBSS. The combined mixture was then incubated for 30 min at room temperature. The whole volume was added dropwise to 0.5-2 million resuspended mature DCs in 100 µl of culture media. Thereafter the cells were incubated for 2h at 37°C, spun down and washed to remove the DOTAP mixture and resuspended in conditioned media at a concentration of 1 million cells/mL.

Naïve T cell enrichment

Nonadherent cells from the DC propagation were counted and incubated with 20µl of buffer and 10µl of each MACS antibody (anti-CD14, CD56, CD19, and CD45R0) (Miltenyi, Stockholm, Sweden) per 10 million cells for 15 min at 4°C. The labelled cells were washed and then passed through an LD-column attached to a magnet for the negative selection of bulk naïve T cells. After collection of total effluent, the naïve T cells were counted and resuspended to 1 million cells/mL in 5% PHS.

Allogenic DC-Naïve T cell co-culture setup and re-stimulation

The mature DCs exposed to mock, HIV, transfected, or exposed to agonist/antagonists were harvested and resuspended in 5% PHS. The different DC groups were divided, with one portion frozen down in freezing media and stored at -80°C and the other portion plated in a flat bottomed 96-well cell culture plate. Naïve T cells were added to the DCs with the different treatments at a ratio of 1:10 and the DC-T cell co-culture incubated in a CO_2_ incubator at 37°C for 7 days. After 7 days, the DC-T cell co-cultures were re-stimulated with 10000 DCs derived from the same donor as used during the co-culture setup and left overnight until harvest.

ELISPOT and ELISA measurements

For ELISA measurements of type I IFN released into the supernatant a PAN IFNα ELISA kit (Mabtech, Stockholm, Sweden) and an IFNβ ELISA construction kit (Antigenix America inc, New York, NY, USA) was used. Supernatants harvested on day 8 were either diluted 1:5 or used neat. For the ELISA, plates were coated with (1µg/mL for IFNβ and 4µg/mL for IFNα) capture antibody overnight, washed and blocked in 0.05% Tween 20 and 0.1% BSA for 60 min. Samples and standard were added to the plate after washing and incubated for 2h at room temperature. The biotin tracer antibody was thereafter added in a concentration of 0.5-1µg/mL and incubated for 1h at RT. Streptavidin-HRP was added in 1:1000 dilution for 30 min at room temperature before addition of TMB substrate. The reaction was terminated after 15 min by the addition of 0.2M H_2_SO_4_ and the plates read at 450 nm using a SpectraMax iD3 microplate reader (Molecular devices, San Jose, CA, USA). For ELISPOT assays cells from the co-culture were counted on day 7 and replated onto pre-coated plates for human granzyme-B, and IFNγ (Mabtech). For IL2 PVDF plates were coated in house by the addition of 10µg/mL of capture antibodies (Mabtech). Thereafter DCs were added at 1:10 ratio to each well and plates were incubated at 37°C overnight. Spots were detected using biotin antibodies (0.25-0.5µg/mL) and streptavidin-HRP (1:1000 dilution). TMB was used to develop color and the reaction stopped by washing with deionized H_2_O. Spots were after development counted manually using a compound microscope.

Spectral flow cytometry

Spectral flow cytometric analysis was performed at co-culture take down (16-18h after re-stimulation), and 1–2 million cells were stained in two different panels with CD3-BV750, CD4-cFlour v450, CD8-cFlour v547, CD38-cFlour R720, HLADR-BV570, CD28-BV480, CD69-APC-H7, PD1-PE-Cy5.5, CTLA4-BB515, CD1c-BV421, PDL1-BB515, CD86-BV650 and ViaDye Red Fixable Viability Dye Kit. For supplier and dilutions used see S2 Table. After surface staining cells were fixed using 4%PFA for 10 min at 4°C, washed and acquired on a Cytek Aurora (Cytek Biosciences, France). Flow cytometry data was unmixed using SpectroFlo (Cytek Biosciences) and further analyzed using FlowJo V.10.10.0 software (BD Life Sciences), for gating strategy see S2 Fig. The viability of DC- T cells in the co-culture on day 8 were above 80% in all conditions and the levels of DCs (ca 10%) similar between the different conditions (S2 Fig).

3H-thymidine incorporation proliferation assay

To measure T cell proliferation in the DC- T cell co-culture during re-stimulation, 2µCi/µl of ^3^H-thymidine (PerkinElmer, Waltham, MA, USA) was added following the supplement of the DCs. Wells containing RPMI and no cells were used for background counts. The amount of incorporated ^3^H-thymidine was measured using liquid scintillation and counted using a micro-β-counter (PerkinElmer).

Sex as a biological variable

Stratification on age and on biological sex has been performed and showed minimal effects on the data outcome and was therefore not taken into consideration in the analysis. Interindividual differences such as HLA-mismatch play a larger role in our DC-T cell co-culture system.

Statistics

For data normalized to mock equals 1, a Kruskall-Walli’s test followed by Dunn’s multiple comparisons test was performed. For most of the data the sum of values from each sample was set to 100% to reduce inter-donor variation, whereby a One-way ANOVA with Tukey’s post-hoc test was performed to yield significance. All statistical computing was performed using GraphPad PRISM Version 10.

Results

HIV BaL and HIV THRO had similar suppressive effects on the DC priming of naïve T cells

To investigate whether the presence of HIV during the DC priming and re-stimulation of T cells affected the ensuing immune response we co-cultured naïve T cells with DCs exposed to different strains of CCR5-tropic HIV-1, either HIV BaL a late-stage derived lab-adapted strain or HIV THRO a transmitted/founder strain. Our analysis revealed a significant suppression of T cell proliferation (Fig 1A), and IL-2 production (Fig 1B) for both HIV strains but no significant differences in outcomes between the two strains. There were elevated gene transcripts of DC and T cell immunomodulatory factors (Fig 1C) for both HIV-1 strains, but no significant differences observed in outcomes besides significantly higher PDL1 in HIV THRO compared to HIV BaL. Next, we explored induction of type I IFN responses and ISGs in the DC-T cell co-cultures (Fig 1D) and discovered an upregulation of IFNα4 and several ISGs mRNAs by both HIV strains.

Both chronic and founder HIV-1 strains suppressed dendritic cells ability to prime naïve T cells.HIV BaL a late-stage derived lab adapted strain or HIV THRO a transmitted/founder strain exposed mature DCs were co-cultured with naïve T cells in the ratio of 1:10. The DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation was assessed via 3H thymidine incorporation (A) and functionality by enumerating lL2-producing cells by ELISPOT assays (B). mRNA transcript of PDL1, CD86, PD1, and TIM3, and (C) IFNα4, MXA, IFI13, and IFI16 (D) by qPCR, N = 4. Whiskers: Min to Max. Statistical significance was determined using the ANOVA. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001*** = p-value <0.0001.*

Pathways that induce type I IFNs were enriched and predicted to be activated when HIV was present in DC-T cell co-culture, and exogenous type I IFN resulted in suppressed T cell responses similar to HIV

To establish pathways involved in HIV suppression, we examined top regulator networks in an RNA sequencing data set from a previous study on HIV-exposed DCs from DC-T cell co-culture [20]. Therein, we found a strong activation of intracellular sensors involved in type I IFN signaling and responses, (e.g., sGAS, STING, and MAVS), IFN regulatory transcription factors (e.g., IRF3, IRF9, and STAT1), type I IFNs, and IFNAR1 and IFNAR2 (Fig 2A). Therefore, to explore if type I IFN played a key role in the induced T cell suppression in our system, we added recombinant type I IFNα2 protein to the DC-T cell co-cultures. The IFNα2 induced suppressed T cell responses were similar to those induced by HIV in the DC-T cell co-culture (Fig 2B), even at the low concentrations measured in the DC-T cell co-culture exposed to HIV (Figs 2B and 4B). The combination of IFNα2 with HIV in the DC- T cell coculture gave additive effects (Fig 2B).

Type I IFN inducing pathways were enriched and predicted to be activated when HIV was present in DC-T cell co-culture and exogenous type I IFN-induced suppressed T cell responses similar to HIV.Transcriptomic data from our published RNA seq [20], where DCs purified from the DC-T cell co-culture after one day of re-stimulation, were assessed for top regulator network (A). HIV BaL exposed mature DCs were co-cultured with naïve T cells in the presence or absence of exogenous human recombinant IFNα4. The DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation and the IFNα2 replenished. One day after re-stimulation, the T proliferation was assessed via 3H thymidine incorporation (B). mRNA transcript of PDL1, and PD1 (C) by qPCR, N = 9. Statistical significance was determined using the ANOVA. Whiskers: Min to Max. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001, **** = p-value <0.0001.

HIV derived replication intermediate ssDNA participated in the induction of the suppressive effects seen by HIV in the DC-T cell co-culture.Mature DCs were exposed to HIV BaL or transfected with the ssDNA, dsDNA, DNA:RNA hybrid, many loops ssDNA, or p24 using DOTAP to deliver these HIV components into the cytosol. The different DC groups were co-cultured with naïve T cells (1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation was assessed via 3H thymidine incorporation (A-D), N = 3-10. Statistical significance was determined using the ANOVA. Whiskers: Min to Max. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001, **** = p-value <0.0001.

HIV-derived oligonucleotide ssDNA participated in the induction of the suppressive effects mediated by HIV in the DC-T cell co-culture

We set out to explore which of the HIV-1-derived oligonucleotides or proteins produced during the viral life cycle that could be involved in HIV-induced immune impairment of the DCs and T cells during priming. The ssDNA, dsDNA, DNA:RNA hybrid, many loops ssDNA, and p24 were delivered intracellularly into the DCs by DOTAP transfection (Fig 3). We found that ssDNA had the most potent suppressive effects on T cell proliferation and that dsDNA also had negative effects on T cell activation, which nonetheless was more variable (Fig 3A). The RNA:DNA hybrid had significant suppressive effect although much less than that induced by ssDNA (Fig 3A, 3B). To explore the effect of the HIV genome molecular composition, i.e., ssRNA, we delivered a customized ssRNA (S1 Table) and synthetic ssRNA40 intracellularly to the DCs and found that they had no or negligible effect on T cell proliferation (Fig 3C). The HIV-1 p24, and many loops ssDNA, had no effects on T cell proliferation (Fig 3B and 3D). To ensure that the transfection did not affect the DC-T cell priming/activation, we treated HIV-exposed or unexposed DCs with DOTAP and noticed no significant effects (S1 Fig). The HIV derived oligonucleotides known to induce STING signaling [23] had a negative effect on T cell proliferation and among them ssDNA had the most potent and consistent effects and was selected for further studies of the mechanism behind the HIV-induced impairment.

HIV and ssDNA induced type I IFN production and ISG in DC-T cell co-cultures

HIV, ssDNA, and dsDNA, induced mRNA transcript of IFNβ, IFNα and ISGs such as IFI16 (Fig 4A, 4B). Of note, although HIV, ssDNA, and dsDNA gave rise to type I IFN and ISGs on mRNA levels, only the complete virion induced detectable amounts of IFNα and IFNβ on protein level measured by ELISA (Fig 4C). HIV induced IFNα and IFNβ on day 3 after co-culture setup and the production persisted throughout the DC-T cell co-culture (Fig 4C).

HIV and ssDNA induced a regulatory/suppressive phenotype of DCs and T cells mediated by increased PDL1 and PD1 expression

HIV, ssDNA and Mock treated co-cultures were investigated for protein expression of activation markers and immune regulatory molecules by spectral flow cytometry. Treatment with HIV increased the percentage of PDL1 and CD86 levels expressed on DCs, whereas ssDNA only led to significantly increased percentage of PDL1 + DCs (Fig 5A). On the T cells, there were significantly elevated levels of CTLA4 in the co-culture after crosstalk with HIV and ssDNA exposed DCs, whereas significantly elevated levels of PD1 were only detected in the co-culture after crosstalk with ssDNA exposed DCs (Fig 5B). Furthermore, the expression of CD28 on CD4 + T cells but not CD8 + T cells, were downregulated by HIV and ssDNA exposure. (Fig 5C). The percentage of activated HLADR+CD38 + CD4 + T cells were lower in co-cultures treated with HIV (significant) and ssDNA (trend) compared to mock (Fig 5D). CD69-expressing T cells significantly increased after treatment with HIV across both subsets, but not with the ssDNA treatment (Fig 5E). The PD1 levels were significantly increased on both HLADR+CD38 + CD4+ and CD8 + T cells and on CD69 + CD4+ and CD8 + T cells after crosstalk with DCs exposed to HIV or ssDNA (Fig 5F and 5G).

STING agonists suppressed T cell proliferation and increased PD-L1 expression on the cells in the co-culture

Activation of STING by DNA oligonucleotides is an indirect event involving the activation of DNA sensors e.g. cGAS, and/or IFI16, followed by downstream STING activation [30]. To explore the involvement of the STING pathway in the HIV-induced suppression, we targeted the STING protein directly by agonists, i.e., cyclic guanosine monophosphate adenosine monophosphate (cGAMP) produced when cGAS becomes bound to a DNA ligand [25], the stabilized 2’3’cyclic-di-AMP analog ADUS100 [33], and MSAD2 [33]. Additionally, we tested the antagonist SN-011 which competes with cyclic dinucleotide for the binding pocket present on STING [34]. When activating STING using extracellular or intracellular delivered cGAMP we found some decreased proliferation with the intracellular but not extracellular delivered cGAMP (Fig 6A). For the agonists ADU100 and MSAD2 we found an even stronger suppressive effect on T cell proliferation and this effect was even higher when combined with HIV (Fig 6B, 6C). The STING antagonist SN-011 treated cells displayed similar profile as untreated, i.e., it failed to stop/rescue the negative effect exerted by HIV on T cell proliferation in the DC-T cell co-culture (Fig 6D), which could be due to the blocking of cGAMP binding but not IFI16 to STING by SN-011. While exploring the effect on type I IFNs and ISGs, we found induction of IFNβ, MXA, and PDL1 by STING agonists, especially MSAD2 (Fig 6E–6G). STING activation in T cells has been shown to lead to impaired T cell proliferation, and in our hands STING activation gave rise to T cell suppression, which could be attributed to type I IFN conditioning of T cells following their interaction with the DCs. In addition, the failure of the STING antagonist targeting the cGAMP site might indicate that the IFI16 -STING signaling still was active.

RIG activation played no role in the HIV induced immune suppression in the DC-T cell co-culture

To further investigate the sensors upstream of STING that could be involved in the activation of type I IFNs and in the HIV-induced immune suppression, we explored the effect of the RIG antagonist RIG012 [35], seeing that RIG-I activation, can both initiates MAVS-dependent IFN signaling [36] and be involved in enhancing STING expression and function by sustaining production of type I IFNs {Ishikawa, 2008 #95} [37]. Intracellular inhibition of RIG in the DCs in the DC-T cell co-culture did not affect the T cell suppression induced by HIV (Fig 7A). Furthermore, RIG012 inhibition of RIG had no effect on type I IFN or ISGs (Fig 7B), and no or insignificant effect on co-inhibitory molecules (Fig 7C). These findings help to exclude the involvement of RIG in HIV-induced immune impairment.

HIV and ssDNA induce type I IFN production and ISG in DC-T cell co-cultures.Mature DCs were exposed to HIV BaL or transfected with the ssDNA, dsDNA, DNA:RNA hybrid, many loops ssDNA, or p24 using DOTAP to deliver these HIV components into the cytosol. The different DC groups were co-cultured with naïve T cells (1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as at the initial stimulation. mRNA transcript of (A) IFNα4, MXA, IFI13, and IFI16 by qPCR and (B) Protein levels of type I IFNs, IFNα4 and IFNβ were measured by ELISA, N = 4-7. ANOVA. Whiskers: Min to Max. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001, **** = p-value <0.0001.

HIV and ssDNA exposure of DCs induce an immune suppressive phenotype of both DCs and T cells after co-culture crosstalk.Mature DCs were exposed to HIV BaL or transfected with the ssDNA using DOTAP to deliver oligonucleotide into the cytosol. The different DC groups were co-cultured with naïve T cells (in a ratio of 1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as at the initial stimulation. Spectral flow cytometric analysis of (A) PDL1 and CD86 on CD1c+ cells, (B) PD1 and CTLA4 on CD3 + T cells and (C) CD28 on CD4 and CD8 + T cells. Spectral flow cytometric analysis of (D) HLADR+CD38+ on CD4+ and CD8 + T cells (E) PD1 expression on HLADR+CD38 + CD4+ and CD8 + T cells, (F) CD69 on CD4+ and CD8 + T cells, and (G) PD1 expression on CD69 + CD4+ and CD8 + T cells. ANOVA, N = 12-16. Whiskers: Min to Max. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001, **** = p-value <0.0001.

STING agonists suppressed T cell proliferation and increased PDL1 expression of the cells in the co-culture.Mature DCs were unexposed or exposed to HIV BaL, or transfected with STING agonists cGAMP, and co-cultured with naïve T cells. Additionally, during co-culture the cells were exposed to the stabilized 2’3’cyclic-di-AMP analog ADU100, MSAD-2, and STING antagonist SN-011. Co-cultures were re-stimulated on day 7 by the addition of the same DCs as during setup. One day after re-stimulation, the T proliferation was measured by 3H thymidine incorporation (A-D). mRNA transcript levels of IFNβ, MXA, and PDL1 in ADU100 (E), MSA2 (F), and SN-011 (G) treated DC-T cell co-culture assessed by qPCR N = 7. Protein levels of IFNα and IFNβ in ADU100 (E), MSAD2 (F), and SN-011 (G) treated DC-T cell co-culture assessed by ELISA, N = 4-5. Statistical significance was determined using the ANOVA. Whiskers: Min to Max. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001, **** = p-value <0.0001.

RIG activation played no role in HIV-induced immune suppression in the DC-T cell co-culture.Mature DCs were exposed to HIV BaL or DOTAP transfected with RIG agonist RIG012 to achieve delivery into the cytosol. The different DC groups were co-cultured with naïve T cells (1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation measured via 3H thymidine incorporation (A), mRNA transcript of CD86, PD1, PDL1, and TIM3, N = 4-7 (B), and IFNα4, MXA, IFI13, and IFI16 (C) by qPCR, N = 4. Statistical significance was determined using the ANOVA. Whiskers: Min to Max. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001, **** = p-value <0.0001.

IFI16 activation suppresses T cell proliferation but is not sufficient to induce expression of immune regulatory molecules

To further investigate the upstream sensors of STING involved in HIV-induced immune suppression, we added the IFI16 agonist VACV-70, a dsDNA derived from vaccina virus [23]. The activation of IFI16 sensor suppressed proliferation to similar levels as HIV (Fig 8A). Surprisingly, the IFI16 agonist had no or little enhancing effects on the levels of coinhibitory molecules or MXA, compared to HIV in DCs or T cells in the co-cultures (Fig 8B, 8C), whereas the effects on the IFNα, IFI16, and STING were similar to HIV (Fig 8C). This suggests that some of the negative effects induced by HIV can be replicated by the VACV-70 IFI16 activation, which indicates that this pathway could be one of the major contributors to HIV-induced impairment.

IFI16 activation suppresses T cell proliferation but is not enough to induce co-inhibitory molecules.Mature DCs were exposed to HIV BaL or DOTAP transfected with IFI16 agonist dsDNA VACV-60 to deliver the VACV-60 into the cytosol. The different DC groups were co-cultured with naïve T cells (1:10) and the DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation measured by 3H thymidine incorporation (A). mRNA transcripts of PDL1, Galectin 9, TRAIL, and TRIM3 (B), and of IFNα4, IFI16, and MXA (C) were assessed by qPCR. Statistical significance was determined using the ANOVA, N = 4. Whiskers: Min to Max. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001, **** = p-value <0.0001.

Blocking of the type I IFN receptor restored DC-T cell functions and phenotypes

Given the association of type I IFN response with HIV-induced immune suppression, blockade of the IFNAR could yield deeper understanding on how this occurs. Blocking of the receptor using the anifrolumab, an anti-IFNAR1 antibody, restored all suppressive events induced by HIV, the proliferation and functionality as measured by IFNα, IL-2 and granzyme-B production by T cells (Fig 9A), the expression levels of immunomodulatory molecules (Fig 9B) and abrogation of the type I IFN response (Fig 9C). This clearly demonstrated the immune suppressive autocrine effect that type I IFNs exert via activation of the IFNAR signaling cascade.

Targeting IFNaR1 by anifrolumab restored immune function in HIV-induced suppressed DCs and T cells.Mature DC were exposed to HIV BaL and incubated for 24h. The untreated (mock) and HIV exposed DCs were harvested and exposed to 20µg/ml anifrolumab for 30 min before the DCs were added to naïve T cells and co-cultured (1:10). The DC-T cell co-culture was re-stimulated on day 7 with the same DCs as the initial stimulation. One day after re-stimulation, the T proliferation was assessed via H3 thymidine incorporation and IFNγ, IL2 and granzyme B producing T cells by ELISPOT (A). The effects of anifrolumab on mRNA transcripts levels for IFNα4, IFI16, and MXA, and IFI16 (B) and PD1L, Galectin-9, TIM3, and TRAIL (C) were assessed by qPCR, N = 7. Whiskers: Min to Max. Statistical significance was determined using the ANOVA. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001, **** = p-value <0.0001.

Discussion

Here, we deciphered the HIV components and mechanisms underlying the immune suppression induced by HIV-exposed DCs when activating T cell responses and found major contribution of the cGAS/IFI16-STING signaling and type I IFN signaling pathways in the ensuing immune suppression.

Most individuals with untreated chronic HIV experience significant impairment in immune cell populations such as DCs, T cells, NK cells, B cells, and MAIT cells, often due to the sustained immune activation [13–15] and in some HIV infected on ART can some level of impairment persist. Immune activation is driven by multiple factors including prolonged elevated type I IFN levels and this is detrimental to the immune system function, which has been shown to contribute to disease progression in both HIV and SIV infection [10,11]. During infection, HIV skews DC function by altering cytokine profiles, induce partial maturation, and contributes to chronic immune activation and defective DC stimulation of T cells [13,14,38]. The observed dysfunction in T cells is in part attributed to the upregulation of coinhibitory molecules and loss of effector functions, further contributing to immune exhaustion. We have previously demonstrated that HIV impairs T cell functionality through DC-mediated cell to cell contact with T cells [18,19]. DCs do not normally support a strong type I IFN after exposure to HIV, which utilizes the HIV restriction factors SAMHD1, and TREX1 to avoid cytosolic sensing by PRRs [39,40]. Nonetheless, our recent finding demonstrated a strong type I IFN profile in HIV-exposed DCs of DC-T cell co-cultures [20], clearly indicating altered cellular programming supporting type I IFN responses. The prolonged type I IFN response observed in a setting of DC-T cell crosstalk in the presence of HIV appears to result from enhanced expression of the STAT1/STAT2/IRF9 transcription complex, sustaining ISG transcription [41]. Alternatively, it could stem from impaired IFNAR recycling, leading to receptor accumulation at the cell surface or increased endosomal uptake necessary for JAK-STAT signaling activation [42]. Dysregulation of downstream signaling, such as reduced ISG15 expression, may also contribute by destabilizing USP18, a negative regulator that inhibits IFNAR assembly and type I IFN signaling [41]. These observations suggest that HIV-driven alterations in DC-T cell crosstalk not only impair antiviral immunity but also reshape the regulatory landscape of type I IFN signaling, thereby perpetuating immune exhaustion through sustained activation and defective feedback regulation.

The similarities in suppressed immune responses in cocultures exposed to HIV-THR0 and HIV-BaL, as well as CXCR4 tropic HIV-MN [38], suggest a conserved mechanism of immune suppression between different HIV-1 strains, and that HIV BaL is a good model virus to study HIV induced immune suppression. The HIV replication cycle involves an array of nucleotides/replication intermediates, such as RNA genomes, ssDNA and dsDNA that are recognized and targeted by the host intracellular defense including RNA and DNA sensors, which is counteracted by HIV to avoid recognition [43]. Our study reveals that HIV-1-derived oligonucleotides, i.e., ssDNA, dsDNA, and RNA:DNA hybrid transfected into the cytosol of DCs participated in the suppressed T cell proliferation, where ssDNA had a key role. The ssDNA, during HIV replication, folds into structures resembling dsDNA which are recognized by the STING pathway [23]. During HIV infection the viral genome is protected by the viral capsid, protecting it from degradation and recognition by cytosolic sensors. If the capsid is disrupted, this can enhance sensing of virus derived oligonucleotides by e.g. cGAS and IFI16 [44–46]. In our system, it might be due to the creation of an environment that disrupts the viral capsid and exposes the viral genome which thereby likely attributes to the prolonged and strong type I IFN responses observed. Our results imply that the interaction between viral oligonucleotides and innate sensing pathways not only drives type I IFN persistence but also represents a critical determinant of HIV-induced immune dysfunction.

We found a clear upregulation of immune regulatory molecules on both the DCs and T cells in the DC- T cell co-cultures following exposure to either ssDNA or HIV. These factors have been well correlated with the production of type I IFNs in previous studies [47]. CD28 plays an important role in maintaining functionality of T cells following priming and activation during infections. In addition, are CD28 also involved in TH1 and TFH differentiation [48]. However, during chronic inflammatory states such as HIV infection there are lower levels of CD28 positive T cells [49], which is in line with our findings. The presence of CD28 negative CD4 + T cells have been reported in chronic infections such as CMV and in settings with elevated levels of type I IFNs where these cells represent proinflammatory and cytotoxic T cells [50]. CD38 and HLADR are markers of T cell activation and function during HIV infection, and high levels of these markers are associated with HIV progression and hyperactivation of T cells [51]. In addition, CD38 can also be upregulated in response to type I IFNs [52]. We found high levels of CD38 + HLA DR+ expressing CD8 + T cells as well as CD4 + T cells in all the DC-T cell co-culture conditions, but the CD38 + HLA DR + CD4 + T cell levels were lower in the HIV group, as expected seeing the impaired activation. Interestingly, the HIV and ssDNA exposed CD38 + HLA DR + T cells showed higher expression of PD1, and this T cell phenotype have previously been reported in chronic HIV infection and other severe conditions such as H7N9 influenza infection [53–55]. These findings suggest that HIV-driven alterations in DC and T cell activation and costimulatory signaling reflect a shift toward an exhausted and dysregulated T cell phenotype shaped by the extended/chronic type I IFN exposure. High levels of CD69 + CD4+ and CD8 + T cells has previously been found in HIV infected individuals with high viral load [56], which is mirrored in our coculture system. HIV also induces increased CD69 expression on T cells in a type I IFN-dependent manner, which might explain the elevated levels of CD69 + T cells in the HIV exposed co-culture [57]. In relation to the increased PD1 expression on these cells our findings indicate activated but functionally suppressed cells.

The mRNA profile sequenced in DCs derived from the HIV exposed DC-T cell co-culture highlighted the significant role of intracellular sensors such as cGAS and STING, and type I IFN pathways in the HIV induced suppression of DCs. This is in accordance with other studies where cytosolic RNA and DNA sensors such as cGAS, RIG-I like receptors, and IFI16 have been implicated in recognizing HIV and activating STING signaling in HIV infected [23,25,58]. We found that addition of exogenous IFNα2 to the DC-T cell co-culture at concentrations detected in HIV exposed co-cultures, induced similar suppressed T cell responses as HIV exposure, suggesting that type I IFNs play a direct role in mediating the induction of impaired immune responses as seen in vivo in HIV and SIV infections [11,59].

Our data showed that cytosolic HIV ssDNA and dsDNA mirroring the effects observed for HIV-1, which suggests that replication intermediates/nucleotides play a crucial role in HIV-1-induced immune suppression. dsDNA-like structures of ssDNA are recognized as cytosolic DNA by the IFI16 -STING pathway and activates type I IFN secretion [60]. This has been shown in cells such as THP1, and monocytes [60] and now by us in DCs, underscoring the unique immune modulatory properties of structured ssDNA. Notably, targeting STING via intracellular cGAMP and agonists suppressed T cell function and induced type I IFN and PDL1 expression, reinforcing the involvement of STING signaling cascade and subsequent type I IFN in the reprogramming of DCs to induce immune impairment in our co-culture system and in individuals with chronic HIV infection [61]. HIV-1 has evolved several escape mechanisms to counteract the STING-induced type I IFNs, e.g., by glutamylation of HIV p6 that inhibits STING and TRIM32 interactions, and by HIV Vif interaction with SHP-1 that leads to STING inhibition [62,63]. Hence, there is ‘tug-of-war’ between HIV blocking the STING pathway and the subsequent activation induced by its oligonucleotides produced during the viral life cycle.

In our system RIG-I signaling had no significant role in HIV-induced immune suppression. This aligns with previous findings demonstrating that HIV evades RIG-I recognition through mechanisms such as viral RNA modifications and sequestration of RIG-I signaling components [64]. RIG-I activation by viral RNA can initiate MAVS-dependent type I IFN signaling but also enhance STING expression and function [37,65], highlighting an antiviral response that bridges RNA and DNA sensing pathways. The HIV engagement of both cGAS and RIG-I simultaneously has been indicated in enabling advanced innate recognition of HIV by myeloid DCs in elite controllers [66], but if this occurs also in our system needs to be further elucidated.

IFI16 is an important regulator to control HIV replication, independent of immune sensing via inhibiting the transcription factor SP1 in the nucleus of CD4 + T cells. This regulator function is especially efficient for non-clade C HIV-1 strains as these strains are in general more resistant to the IFI16 pathway [67]. Cytosolic DNA sensors such as IFI16 have been implicated in the recognition of HIV ssDNA in THP1 cells and the subsequent activation of STING signaling [23]. Cells that lack IFI16 have uncontrolled replication, which in turn leads either to exhaustion of the cell or cell death [23]. We found that targeting the IFI16 in the DCs directly with an agonist lead to elevated type IFN and suppressed T cell proliferation to levels comparable to those observed with HIV. These results align with previous studies demonstrating that IFI16 plays a key role in sensing DNA from viruses such as HIV-1 [68,69]. Blocking the cGAMP binding to STING had little restoring effect on the responses in our system and this give some evidence that IFI16 could be involved in the STING activation. Surprisingly, activation of STING via IFI16 did not enhance the expression of co-inhibitory molecules or ISGs. IFI16-STING signaling seems to be a key component of the immune modulation seen in our system, but HIV exposure clearly employs additional mechanisms such as cGAS-STING that give rise to elevated expression of co-inhibitory molecules and further impair immune function that is missing when targeting only IFI16. The IFI16 sensor is interlinked with the cGAS-STING pathway in macrophages, by supporting cGAMP production and its downstream signaling, i.e., recruitment and activation of TBK1 [70], which might explain that individually targeting of cGAS or IFI16 sensor might not give the full effect that HIV induce in the DCs. High viral load drives the elevated expressions of IFI16 and cGAMP in HIV positive individuals. In addition, there was a correlation between CD38 and IFI16 in these patients, indicating that the DNA activation of IFI16 might be part of chronic immune activation [58]. Our data from the DC-T cell co-culture support this with high immune activation of type I IFN responses, and IFI16 leading to suppressed responses in the HIV-exposed setting. In chronic HIV infected individuals ART treated or untreated, there are impaired ISGs responses to DNA stimuli [58], which could depend on the cell status due to the chronic activation found in these patients. The discordant checkpoint biomarker and IFN responses between HIV, ssDNA, and IFI16 agonist likely stem from distinct sensing dynamics and signal integration. Unlike ssDNA or IFI16 stimulation, HIV replication generates multiple nucleic acid intermediates and viral components that concurrently engage cGAS–STING and other sensor pathways, amplifying downstream transcriptional and translational responses. The induction of detectable IFN-α/β protein by HIV, but not ssDNA, further suggests that HIV enables a stronger coordinated activation and signal amplification that isolated nucleic acid ligands cannot replicate. These findings underscore that the complexity of viral replication, rather than individual nucleic acid sensing events, drives the breadth and magnitude of HIV-induced immune modulation.

The elevated ISGs clearly indicated autocrine and signaling via IFNα/β receptors as part of the HIV induced immune activation leading to the suppressed responses. The prolonged type I IFN exposure, and the IFNα/β receptors signaling via JAK/STAT/IRF9 has been linked to upregulation of inhibitory molecules, suppression of T cell function, and impaired antigen-presenting cell activity [71]. In our system, the complete recovery rendered by the blocking I IFNα/β receptor antibody anifrolumab is a promising strategy to restore the immune function in people living with HIV on ART and may also lead to lowered tissue viral DNA and improved general health as has been shown for ART treated SIV+ macaques receiving anti-IFNα therapy using an anti-IFNα antibody [59]. Restoration of immune responses after blocking IFN-I/IFNAR signaling has also been shown in humanized mouse models, where it restores T cell numbers and function, reduces immune activation, and, when combined with ART, diminishes HIV-1 reservoirs and delays viral rebound [72–74].

This study has several limitations inherent to the use of an in vitro DC-priming model to examine the immunological effects exerted by HIV. Intracellular levels of HIV and HIV-derived oligonucleotides (ssDNA and dsDNA) following DOTAP transfection were not quantified and thus may not reflect the amounts produced during physiological HIV infection. Consequently, the magnitude and kinetics of the responses observed here may differ from those in vivo, and these factors should be considered when interpreting the findings. Overall, our results suggest that the IFI16/cGAS-STING pathway and subsequent type I IFN signaling with IFNα/β receptors contributes to HIV-induced immune suppression, but additional factors are likely involved in regulating the expression of co-inhibitory molecules. These findings reinforce the understanding that HIV exploits innate immune signaling to establish an immunosuppressive environment that hinders effective T cell responses and supports the role of DC-driven immune regulation in HIV pathogenesis. Moreover, the restoration of immune function when targeting IFNαR via anifrolumab, a drug already approved and on market for non-renal SLE and in phase 3 trials for lupus nephritis, systemic sclerosis, idiopathic inflammatory myopathies, and cutaneous lupus erythematosus, highlights the potential therapeutic value of targeting type I IFN signaling in HIV infection to alleviate immune dysfunction and improve DC and T cell functionality. Future studies should delineate how HIV utilizes the IFI16/cGAS-STING signaling and other factors involved in immune impairment and explore whether modulating the pathway can serve as a viable strategy to counteract immune suppression in HIV infection, both in untreated and ART treated individuals.

Supporting information

S1 TableOligonucleotide sequences.(DOCX)

S2 TableAntibodies for spectral flow panels.(DOCX)

S1 FigDOTAP transfection had no effect on the induction of the HIV suppressive effects seen in the DC T cell coculture.Mature DCs exposed to mock or HIV-1 BaL were untreated or treated by DOTAP transfection. The different DC groups were cocultured with naïve T cells (1:10) and the DC-T cell coculture was restimulated on day 7 with the same DCs as the initial stimulation. One day after restimulation, the T proliferation was assessed via ^3^H thymidine incorporation. Statistical significance was determined using the ANOVA. * = p-value <0.05, ** = p-value <0.01, *** = p-value <0.001, **** = p-value <0.0001.(TIF)

S2 FigRepresentative Gate:ing strategy for spectral flow cytometric analysis.Mature DCs exposed to mock, HIV-1 or DOTAP transfected with ssDNA were cocultured with naïve T cells (in a ratio of 1:10) and the DC-T cell coculture was restimulated on day 7 with the same DCs as the initial stimulation. After 16-18h of incubation, on day 8, cells were harvested and stained for spectral flow cytometry.(TIF)