IL-18 responsiveness defines limitations in immune help for specialized FcRγ^– NK cells

Abstract

Despite being prolific innate killers, NK cells are also key helper cells in antiviral defense, influencing adaptive immune responses via interactions with dendritic cells (DCs). In addition to causing NK cell dysfunction, HIV-1 infection contributes to the expansion of a rare population of NK cells deficient in FcRγ (FcRγ^–), an intracellular adaptor protein that associates with CD16. The implications of this inflated NK cell subset in treated HIV-1 infection remain unclear. In this study, we explored the helper function of human NK cells in chronic HIV-1 infection, with a particular focus on characterizing FcRγ^– NK cells. Exposure of NK cells to innate DC-derived co-stimulatory factors triggered their helper activity, defined by their ability to produce IFNγ and to drive the maturation of high IL-12-producing DCs. In this setting, however, FcRγ^– NK cells were defective at producing the dominant DC-polarizing agent IFNγ. The reduced responsiveness of FcRγ^– NK cells to IL-18 in particular, which was attributable to impaired inducible expression of IL18Rα, extended beyond an inability to produce IFNγ, as FcRγ^– NK cells showed limited potential to differentiate into CD16^–/CD25⁺/CD83⁺ helper cells. Notwithstanding their deficiencies in responsiveness to innate environmental cues, FcRγ^– NK cells responded robustly to adaptive antibody-mediated signaling through CD16. The presence of an expanded population of FcRγ^– NK cells with a diminished capacity to respond to IL-18 and to effectively modulate DC function may contribute to disturbances in proper immune homeostasis associated with HIV-1 infection and to defects in the initiation of optimal adaptive antiviral responses.

Introduction

NK cells are specialized lymphoid cells with an innate capacity to recognize and kill transformed and virally infected cells (1–4). A powerful first line of defense, they play a critical role in the early control of viral infections, and genetic aberrations related to defects in NK cell frequency or functionality are associated with an increased susceptibility to intracellular pathogens (4, 5), including chronic herpesvirus infections (6). NK cells are divided into two functional subsets based on their surface expression of CD56 and CD16 (7–9). CD56^dimCD16⁺ NK cells, which account for approximately 90% of peripheral blood human NK cells, are comprised of mature effector cells and demonstrate superior cytotoxic activity, including CD16-mediated antibody-dependent cellular cytotoxicity (ADCC) (7, 10–12). By contrast, CD56^brightCD16^– NK cells produce abundant immunomodulatory cytokines and display longer telomeres than their CD56^dim counterparts, lending credence to the linear model of NK cell differentiation, in which CD56^bright NK cells precede the CD56^dim population (13, 14).

While NK cells are widely recognized for their cytolytic abilities, they also play a pivotal role as immune helper cells, providing innate alarm signals that shape and regulate the adaptive immune response. Their interactions with dendritic cells (DCs) result in the important exchange of reciprocal activation signals (15–21). For example, pro-inflammatory cytokines produced by activated DCs, including IL-18 and IL-12, stimulate NK cell production of TNFα and IFNγ. In turn, these NK cell-derived factors provide critical feedback signals to drive both DC maturation and polarization, supporting their continued capacity to mediate type 1-biased T cell responses (18–21). Of note, IL-18 has been shown to have a unique ability to induce the differentiation of a subset of CD56^dim NK cells into CD83⁺/CCR7⁺ helper cells, primed to immediately produce IFNγ following migration to lymphoid tissues and exposure to secondary stimuli, including IL-12, IL-2, or IFNα (19). The ability of NK cells to properly respond to innate signaling with agility and flexibility is crucial for the induction and maintenance of effective adaptive T cell responses to viral infections (22). Not surprisingly, viruses such as HIV-1 have developed a targeted means to cripple NK cell helper functions by interrupting NK-DC crosstalk to consequently curb the capacity of DCs to promote effective antiviral T cell responses (23–27).

As effectors of the innate immune response, NK cells are restricted by the variegated expression of germline-encoded receptors, and their activation depends on the integration of activating and inhibitory signals. Interactions of NKG2A and inhibitory killer immunoglobulin-like receptors with their cognate self-MHC class I ligands shape and fine-tune the functional responses of NK cells in a process termed licensing, which promotes self-tolerance through functional competence (28–31). The unique combinatorial expression patterns of activating and inhibitory receptors give rise to a heterogenous NK cell population marked by diverse phenotypes and response potentials (32–34). Mounting evidence also indicates that cumulative pathogen exposures elicit dynamic shifts in the repertoire of NK cell receptors, prompting further increases in phenotypic and functional diversity that potentially impact the quality of NK cell responses to subsequent infections (33).

In line with this notion, exposure to specific inflammatory cytokine milieus or human cytomegalovirus (HCMV) can confer a lasting imprint on NK cells, inducing the preferential expansion of NK cell populations with features of immunological memory (35–45). While the term ‘adaptive’ refers to HCMV-specific NK cells that uniformly express NKG2C (35, 37), the term ‘memory-like’ describes an NKG2C-independent NK cell subset notably deficient in the intracellular protein FcRγ (FcRγ^–), otherwise known as FcεRIγ (44–46). FcRγ is a signaling adaptor protein that can associate intracellularly with CD16, NKp46, and NKp30 (44). CD56^dim ‘memory-like’ NK cells display enhanced effector functions upon stimulation through CD16 and mediate augmented responses to HCMV-infected cells in the presence of HCMV-specific antibodies (44, 45). However, FcRγ^– NK cells from HCMV seropositive individuals are characterized by an impaired innate ability to produce IFNγ (47). An inflated population of FcRγ^– NK cells has also been identified in HIV-1-infected individuals (43, 48, 49). Although the underlying mechanism and implications of this phenomenon in chronic HIV-1 infection remain to be elucidated, the reduced IFNγ-producing capability of FcRγ^– NK cells suggests potential abnormalities in their interactions with DCs.

Given the importance of the bidirectional crosstalk between NK cells and DCs in choreographing the nature of the adaptive immune response, our study sought to explore the helper function of NK cells in the setting of treated chronic HIV-1 infection and to highlight the unique phenotypic and functional attributes adopted by FcRγ^– NK cells in response to both innate and adaptive stimuli. Our data demonstrate that NK cells are capable of providing immune help during antiretroviral therapy (ART)-mediated viral suppression. However, by specializing in antibody-specific responses, FcRγ^– NK cells compromise their responsiveness to DC-derived innate stimuli, leading to qualitative differences in their helper function and crosstalk with DCs.

Materials and Methods

Study participants

HIV-1-infected participants (n = 14), who self-identify as men who have sex with men (MSM), were randomly selected from the Pittsburgh clinical site of the Multicenter AIDS Cohort Study (MACS). All HIV-1⁺ participants recorded a plasma HIV-1 load less than 20 copies/ml at the time of their study visit, with a median virally controlled treatment duration of 12.08 years (range: 1.83–22.67) (table I). Age-matched HIV-1 seronegative MSM (n = 14) were also selected from the MACS (median age of 58 and 57 years for HIV-1^– and HIV-1⁺ men, respectively). All participants provided written informed consent prior to inclusion in this study, which was approved by the Institutional Review Board at the University of Pittsburgh.

Table I.

Characteristics of HIV-1⁺ MACS Participants.

age	time to treatment (yrs)	time on suppressive ART (yrs)	CD4+ T cell count* (cells/mm3)	viral load* (copies/ml)	highest viral load (copies/ml)	nadir CD4+ T cell count (cells/mm3)	ART regimen	CMV status
65	14.17	20.33	529	< 20	60,909	58	TUM	SP
56	1.42	18.92	1265	< 20	68,067	392	TUM	SP
56	6.83	10.17	373	< 20	87,040	132	ATV, RAL, TUM	SP
59	4.75	8.42	486	< 20	38,580	282	ODS	SP
55	2.17	11.83	434	< 20	71,964	184	ATP	SP
83	2.75	22.67	542	< 20	94	129	DCV, DTG	SP
60	NA	16.50	1275	< 20	499,797	356	CBV, NFV	SP
53	1.42	14.75	589	< 20	400	275	RTV, DTG, DRV	SN
58	NA	16.83	1037	< 20	309,132	272	TUM	SP
69	7.67	8.83	651	< 20	460,400	210	TUM	SP
35	0.50	1.83	788	< 20	17,497	544	GNV	SP
55	3.08	12.33	421	< 20	3,524,100	148	TUM	SP
64	4.25	11.33	792	< 20	32,892	315	TUM	SP
55	1.00	5.75	707	< 20	769,565	216	TUM	SP

^(*)Asterisk, at study visit; NA, not available due to unknown date of seroconversion; ATP, Atripla; ATV, atazanavir; CBV, combivir; DCV, Descovy; DRV, darunavir; DTG, dolutegravir; GNV, Genvoya; NFV, nelfinavir; ODS, Odefsey; RAL, raltegravir; RTV, ritonavir; TUM, Triumeq; SP, seropositive; SN, seronegative

HCMV status

The HCMV status of study participants was determined by quantitative Cytomegaly IgG Antibody ELISA (GenWay) following the manufacturer’s protocol. Serum samples were diluted 1:101 with the provided sample diluent, and absorbance was read at 450 nm using the BioTek ELx800™.

Isolation of monocytes and PBLs

PBMCs were isolated from whole blood of MACS participants, or from buffy coat blood products (purchased from the Central Blood Bank of Pittsburgh), by standard density gradient separation utilizing lymphocyte separation medium (Corning). PBMCs were further separated into monocytes and PBLs using human CD14 MicroBeads (Miltenyi Biotec).

Generation of monocyte-derived immature DCs (iDCs)

To generate iDCs, isolated monocytes were cultured for four days in 24-well plates (Costar^®) at a density of 6.25×10⁵ cells/well in Iscove’s Modified Dulbecco’s Medium (Gibco^®) containing 10% fetal bovine serum (Atlanta Biologicals) and 0.5% gentamicin (Gibco^®) (cIMDM). Cultures were supplemented with GM-CSF (1000 IU/ml; Sanofi-Aventis) and IL-4 (1000 IU/ml; R&D Systems^®).

Activation and polarization of DCs by NK cells

Autologous NK cells, purified from fresh or cryopreserved PBLs by magnetic bead negative selection using a human NK cell enrichment kit (EasySep™), were added directly to day-4 iDC cultures at a 1:1 ratio. Recombinant human (rh) IL-18 (hereinafter referred to as IL-18; 1 μg/ml; MBL^® International) and rhIL-12p70 (hereinafter referred to as IL-12; 100 ng/ml; BioLegend^®) were also added directly to the co-cultures (NK-DC). Where indicated, DCs were cultured only with IL-18 and IL-12 (DC0) or in the absence of NK cells and additional cytokines (iDC). After 48h, DCs were harvested, thoroughly washed, and analyzed for their expression of maturation-associated surface markers (see Flow cytometry) and ability to produce IL-12. To test their IL-12-producing capacity, DCs were plated in 96-well flat bottom plates (Costar^®) at 2.5×10⁴ cells/well. To mimic the interaction with CD40L-expressing Th cells, CD40L-transfected J558 cells (a gift from Dr. P. Lane, University of Birmingham, Birmingham, U.K.) were added at 5×10⁴ cells/well (18). Supernatants were collected after 24h and tested for the presence of IL-12p70 by ELISA.

To compare the ability of NK cells from HIV-1⁺ MSM to impact DC production of IL-12, isolated NK cells were cultured in serum free AIM V^® medium (Gibco^®) in 96-well round bottom plates (Costar^®) at a density of 1×10⁶ cells/ml in the presence of IL-18 (1 μg/ml; MBL^® International) and rhIL-2 (hereinafter referred to as IL-2; 1000 IU/ml; Proleukin^®, Prometheus Laboratories, Inc.). Supernatants from these NK cell and mock (cytokines only) cultures were collected after 48h. To control for high donor-to-donor variability in DC IL-12 production, iDC cultures were generated from a single HIV-1⁺ participant and treated for 48h with rhTNFα (20 ng/ml; R&D Systems^®). The DCs were harvested, thoroughly washed, and plated in 96-well flat bottom plates (Costar^®) at 2.5×10⁴ cells/well and used as assay responder cells. Supernatants from the NK cell and mock cultures were added to the responder DCs, which were subsequently stimulated with CD40L-transfected J558 cells (5×10⁴ cells/well). Supernatants were collected after 24h and tested for IL-12p70 by ELISA.

NK cell cultures

NK cells were purified from fresh or cryopreserved PBLs by magnetic bead negative selection using a human NK cell enrichment kit (EasySep™). The baseline phenotypic profile and frequencies of FcRγ^– NK cells were determined by flow cytometry (see Flow cytometry). Isolated NK cells from the selected participants having an FcRγ^– baseline frequency greater than 10% (n = 8) were cultured in serum free AIM V^® medium (Gibco^®) in 96-well round bottom plates (Costar^®) at a density of 1×10⁵ cells/well. The phenotypic and functional impact of cytokine treatment was assessed on NK cell subsets following a 24h or 48h culture in media alone or in the presence of IL-18 (1 μg/ml; MBL^® International) and/or IL-12 (20 ng/ml; BioLegend^®). For priming experiments, NK cell cultures were washed after a 48h exposure to a primary signal and cultured for an additional 24h in the presence of a secondary signal of either IL-18, IL-12, or IL-2 (1000 IU/ml; Proleukin^®, Prometheus Laboratories, Inc.).

Flow cytometry

Cells were pre-exposed for 15 minutes to 50.0 μg/ml of unfractionated murine IgG (Sigma-Aldrich) to block nonspecific Fc receptor binding before immunostaining. The LIVE/DEAD™ Fixable Aqua Dead Cell Stain (Life Technologies) was used for viability exclusion, and the following antibodies were used for immunostaining: CD3-APC-H7 (clone SK7, BD Pharmingen™), CD56-PE-Cy7 (clone N901, Beckman Coulter), CD57-BV421 (clone NK-1, BD Horizon™), CD16-PerCP-Cy™5.5 (clone 3G8, BD Pharmingen™), CD25-BV605 (clone M-A251, BD OptiBuild™), CD83-PE (clone HB15a, Beckman Coulter), CD86-PE (clone HA5.2B7, Beckman Coulter), CCR7-FITC (clone 150503, R&D Systems^®), Siglec-1/CD169-PE (clone 7–239, BioLegend^®), NKG2A-APC (clone Z199, Beckman Coulter), NKG2C-PE (clone 134591, R&D Systems^®), NKp46-PE (clone BAB281, Beckman Coulter), IL18Rα-Alexa Flour^® 700 (clone 70625, R&D Systems^®), and IL12Rβ2-APC (clone REA333, Miltenyi Biotec). Staining was done in FACS buffer consisting of 1x PBS solution (GE Life Sciences), 0.5% bovine serum albumin, and 0.1% sodium azide (Sigma). For intracellular protein expression, NK cells were fixed with BD Cytofix/Cytoperm™ (BD Biosciences), permeabilized using BD Perm/Wash™ (BD Biosciences), and labeled with anti-FcRγ-FITC (Milli-Mark^®) and/or anti-IFNγ-Alexa Flour^® 700 (clone B27, BD Pharmingen™). For CD107a mobilization assays, the cells were exposed to anti-CD107a-APC (clone H4A3, BD Pharmingen™) in the presence of 0.1% monensin by volume (BD GolgiStop™), followed by the viability exclusion, surface marker, and intracellular immunostaining procedures described above. Samples were stored in FACS buffer until data acquisition using a BD LSRFortessa™ flow cytometer. Data were analyzed using FlowJo version 10.5.3 (Tree Star), with expression levels based on comparison to fluorescence minus one (FMO) samples or unstimulated controls for CD107a and intracellular cytokine staining (50). Gating strategies are provided in the supplementary figures. To generate the heatmap, mean fluorescence intensities (MFIs) for each marker were obtained by flow cytometry and subsequently normalized, whereby the lowest and highest MFIs of each dataset were assigned values of 0 and 100, respectively. The heatmap, with two-dimensional unbiased hierarchical clustering, was created utilizing R version 3.5.2 and the RColorBrewer package.

t-distributed Stochastic Neighbor Embedding (t-SNE)

FCS files were imported into FlowJo version 10.5.3 (Tree Star) and manual gates were applied to exclude debris, doublets, and dead cells from each sample as defined in Supplementary Fig. 2. The number of events on each live gate population was reduced using the DownSample gate tool, followed by merging of downsampled gates with the Export/Concatenate Populations tool, during which keyword-based derived parameters were created representing the different stimulation conditions. t-SNE was then performed on the concatenated file. Different groups of samples (i.e., unstimulated vs IL-18 treatment) were gated on according to the keyword parameter. Manual gating was used to overlay color on the FcRγ^– population.

Cytokine Detection

Supernatants from the DC and cytokine-stimulated NK cell cultures were collected at the end of the 24h or 48h incubation and stored at −80°C until tested. Concentrations of IL-12p70 and IFNγ were measured by a sandwich ELISA using recombinant human protein for standards (R&D Systems^®) and matched capture and detection antibody pairs (Invitrogen™) following the manufacturer’s protocol. Absorbance was read at 450 nm using the BioTek ELx800™.

Flow cytometry-based ADCC assay

NK cells were plated in 96-well round bottom plates at a concentration of 1.6×10⁶ cells/ml and serially diluted two-fold until achieving a concentration of 2×10⁵ cells/ml. Raji cells were incubated for 30 minutes at 37°C with 10 μg/ml rituximab and washed two times with 1x PBS. Unopsonized Raji cells were stained with 1 μM CellTrace™ Violet dye (Thermo Fisher) for 10 minutes at 37°C. After quenching the reaction with 10 ml of cIMDM, the cells were incubated for an additional 10 minutes at 37°C and washed two times with cIMDM. The CellTrace™ Violet dye-labeled Raji cells, used both as ‘cold target’ inhibitors and as control reference cells, were mixed with an equal number of unstained opsonized Raji target cells. The mixture of Raji cells was subsequently resuspended in serum free AIM V^® medium at a concentration of 4×10⁵ cells/ml. 100 μl of the Raji cell suspension was plated per well of NK cells, resulting in effector (NK) to target (Raji) cell ratios of 8:1, 4:1, 2:1, and 1:1. After incubating the co-cultures at 37°C for 6h, cells were stained for viability (LIVE/DEAD™ Fixable Aqua Dead Cell Stain, Life Technologies) and surface expression of CD3-APC-H7 (clone SK7, BD Pharmingen™), CD19-PerCP-Cy™5.5 (clone HIB19, BD Pharmingen™), and CD56-PE-Cy7 (clone N901, Beckman Coulter), and were subsequently fixed in 2% paraformaldehyde. The percentage of ADCC-mediated killing was calculated based on the relative proportion of viable opsonized targets to unopsonized dye-labeled Raji cells.

Statistical analyses

Data were analyzed using GraphPad Prism version 8.0.2. Normality was determined by the Shapiro-Wilk test, and data not following a normal distribution were analyzed by the non-parametric Mann-Whitney U test (Fig. 2G–2J). The paired Student’s t-test was used to determine statistical significance for direct comparisons of the FcRγ^– and FcRγ⁺ NK cell subsets within HIV-1⁺ participants with an FcRγ^– NK cell frequency greater than 10%. The unpaired Student’s t-test was used for Fig. 1B, 5E, 5G, 5H. Where the mean of each condition was compared to a single control mean of 1.00, statistical significance was assessed by one-way ANOVA, followed by Dunnett’s test for multiple comparisons (Fig. 1E, 4E, 6I). A one-way ANOVA was used to measure the significance of differences between the means of three or more groups, with correction for multiple comparisons by Tukey’s HSD post-hoc test (Fig. 1A, 1D, 2A-2C, 4D, 7A). A two-way ANOVA was used to establish the statistical significance of differences between the means of multiple groups (e.g., secondary signal) of two factors (e.g., primary signal) (Fig. 6H). The linear relationship between FcRγ^– NK cell frequency and IFNγ expression was determined by Pearson correlation analysis (Fig. 5F), and we used multiple linear regression analysis to investigate the effect of FcRγ^– NK cell frequency and HIV-1 status on IFNγ expression. In each figure, community controls are denoted by triangles, HIV-1^– MSM by squares, and HIV-1⁺ MSM by circles.

Results

Two-signal activated NK cells promote type-1 polarized DCs

As the most efficient antigen presenting cells, DCs are particularly revered for their ability to translate environmental cues received in the periphery into polarizing signals for naïve T cells in the lymph node, thus directing the nature of the adaptive immune response (51–54). NK cells are important not only for the elimination of infected cells but also for shaping the character and quality of antigen-specific immunity through their local interaction with DCs in infected tissues. In fact, NK cells have the capacity to either limit DC function, through a process referred to as DC editing (25, 55–57), or to promote their maturation and subsequent lymph node-homing and T cell-priming functions (18, 21). Previous studies have demonstrated that active HIV-1 infection in particular leads to dysfunctional interactions between NK cells and DCs (23–27). Therefore, we investigated the quality of the crosstalk between NK cells and DCs from virally suppressed HIV-1-infected MACS participants. We began by assessing the basic functional status of the NK cells, confirming their requirements for the production of IFNγ, an important effector cytokine and potent DC-modulating agent (17). Consistent with previous reports (18, 21), the induction of NK cell helper activity required a second NK cell-activating signal. That is, only stimulation with two activating signals triggered production of IFNγ by NK cells isolated from virally suppressed MACS participants with chronic HIV-1 infection (Fig. 1A; Supplementary Fig. 1A), and this two-signal requirement remained true for HIV-1-uninfected blood bank donors (community controls; Supplementary Fig. 1B). NK helper cells were capable of responding to a variety of secondary signals, including IFNα and IL-2, but the combination of IL-18 and IL-12 resulted in the highest levels of IFNγ secretion (Supplementary Fig. 1C). Based on these data, we utilized IL-18 with IL-12, two important DC-derived cytokines for triggering NK cell activation (15, 17–20), in subsequent experiments. Although NK cells from HIV-1-infected participants produced IFNγ in response to co-stimulation with IL-18 and IL-12, this production differed significantly between HIV-1-infected and uninfected individuals (Fig. 1B). To directly evaluate NK-DC crosstalk in the setting of chronic HIV-1 infection, iDCs were cultured with IL-18 and IL-12 in the presence or absence of autologous NK cells. The cells were harvested after 48h and assessed for DC maturation and polarization. Two-signal activated NK cells induced DC maturation, demonstrated by surface expression of CD83 and CCR7, as well as upregulation of CD86 and Siglec-1 (Fig. 1C). Additionally, the presence of NK cells promoted the programming of type-1 polarized DCs with an enhanced ability to produce IL-12 (Fig. 1D, 1E), indicating the combination of IL-18 and IL-12 activates the helper function of NK cells from participants with chronic HIV-1 infection.

Figure 1. Two-signal activated NK cells promote type-1 polarized DCs.

(A) Exposure to one signal was insufficient for activating the helper function of NK cells. Only two-signal activated NK cells produced high levels of IFNγ as determined by intracellular staining; n = 4 HIV-1⁺ participants. (B) In response to a 48h exposure to IL-18+IL-12, IFNγ expression was significantly lower among HIV-1⁺ individuals (n = 14) compared to the HIV-1^– group (n = 20), comprised of n = 6 HIV-1^– community controls (black triangles) and n = 14 HIV-1^– MSM (steel squares). (C) Day-4 iDCs were cultured for 48h either with IL-18+IL-12 in the presence of autologous NK cells (1:1), only with IL-18+IL-12 (DC0), or in the absence of NK cells and additional cytokines (iDC). Two-signal activated NK cells induced the development of type-1 polarized mature DCs (NK-DC), demonstrated by expression of CD83 and CCR7 and upregulation of CD86 and Siglec-1; data are representative of n = 5 HIV-1⁺ participants. (D, E) Importantly, the presence of NK cells promoted programming of type-1 polarized DCs with enhanced IL-12p70 production following a 24h stimulation with J558-CD40L; n = 3 replicates from 3 HIV-1⁺ participants. Statistical significance was determined by one-way ANOVA (A, D, E) or the unpaired Student’s t-test (B) (****p < 0.0001; *p < 0.05).

FcRγ^– NK cells expand during chronic HIV-1 infection

While there appeared to not be a defect in NK-DC crosstalk in the setting of treated chronic HIV-1 infection, we did note lower IFNγ production among HIV-1⁺ participants (Fig. 1B), suggesting potential qualitative differences in the interactions between NK cells and DCs. Since impaired NK-DC crosstalk in HIV-1 infection has previously been attributed to the preferential expansion of CD56^–CD16⁺ NK cells (25), we determined the relative distribution of NK cell subsets in HIV-1-infected and uninfected individuals. NK cells were enriched from PBLs of HIV-1-uninfected community controls (n = 8), as well as randomly selected HIV-1-infected and age-matched HIV-1-uninfected MACS participants (n = 14), who self-identify as MSM. The observed variability in IFNγ production could not be explained by an unusual CD56^–CD16⁺ population as no differences in the relative proportions of CD56^brightCD16^–, CD56^dimCD16^±, or CD56^–CD16⁺ NK cells were detected between the three groups (Fig. 2A–2C). However, baseline frequencies of peripheral blood-derived FcRγ^– NK cells, measured by flow cytometry (Supplementary Fig. 2), varied widely, ranging from 1 to 45% (Fig. 2D, 2E). While FcRγ^– NK cells comprised at least 10% of the total NK cell population in approximately 30% and 60% of HIV-1^– and HIV-1⁺ MACS participants, respectively, they accounted for less than 10% of the total NK cell population in all of the uninfected community controls (Fig. 2F). A modest increase in FcRγ^– NK cell frequencies was observed in HIV-1^– MSM compared to HIV-1^– community controls, but this trend did not reach statistical significance (p = 0.2119; Fig. 2F). By contrast, chronic HIV-1 infection was characterized by a significantly larger population of FcRγ^– NK cells (Fig. 2G–2I). Since the induction of FcRγ^– NK cells is strongly associated with HCMV infection (45), we compared FcRγ^– NK cell frequencies only among HCMV seropositive individuals. Both the HIV-1⁺ and HIV-1^– MSM experienced an HCMV seropositivity rate of 93% (n = 13 of 14; table I). Only 25% of the HIV-1^– community controls were seropositive for HCMV (n = 2 of 8), for a combined HCMV seropositivity rate among the HIV-1^– individuals of 68%. Importantly, higher proportions of FcRγ^– NK cells were still present in HIV-1⁺ participants relative to the HIV-1^– group (Fig. 2J). FcRγ^– NK cells, indeed, expand in the setting of chronic HIV-1 infection, but their prominence in both HIV-1^– and HIV-1⁺ MACS participants relative to community controls underscores the influence of MSM status on FcRγ^– NK cell proportions, supporting the findings of Hearps et al. (48).

Figure 2. A notable population of NK cells deficient in FcRγ is present in chronic HIV-1 infection.

(A-C) Baseline flow cytometry analyses of purified NK cells from age-matched HIV-1^– and HIV-1⁺ MSM of the MACS cohort (n = 14), as well as uninfected community controls (n = 8). The relative distribution of (A) CD56^brightCD16^–, (B) CD56^dimCD16^±, and (C) CD56^–CD16⁺ NK cell subsets was comparable between the three groups. (D) Representative staining at baseline for FcRγ on NK cells enriched from peripheral blood, showcasing the scarcity of FcRγ^– NK cells in uninfected community controls, (E) compared to HIV-1-infected MSM, where the frequencies varied, accounting for up to 45% of total NK cells in select participants tested. (F) While the proportions of FcRγ^– NK cells increased among HIV-1^– MSM, the highest percentages of FcRγ^– NK cells emerged in HIV-1⁺ MACS participants, with 8 of 14 HIV-1⁺ men having an FcRγ^– NK cell frequency >10%, compared with 4 of 14 HIV-1^– MSM and 0 of 8 HIV-1^– community controls. (G-I) Chronic HIV-1 infection was characterized by an expanded population of FcRγ^– NK cells, with comparisons between (G) HIV-1⁺ MSM (n = 14; circles) and HIV-1^– individuals (n = 22), comprised of n = 8 HIV-1^– community controls (triangles) and n = 14 HIV-1^– MSM (squares); (H) HIV-1^– community controls (n = 8; triangles) and HIV-1⁺ MSM (n = 14; circles); and (I) HIV-1^– MSM (n = 14; squares) and HIV-1⁺ MSM (n = 14; circles). (J) Among the HCMV seropositive individuals only, higher frequencies of FcRγ^– NK cells were present in HIV-1⁺ individuals (n = 13; circles) relative to the HIV-1^– group (n = 15), comprised of n = 2 HIV-1^– community controls (triangles) and n = 13 HIV-1^– MSM (squares). Normally distributed data were measured by one-way ANOVA, with the solid black line representing the mean (A-C). For data not following a normal distribution, statistical significance was determined by the Mann-Whitney U test, with the solid red line representing the median (G-J) (**p < 0.01; *p < 0.05).

A mature phenotype distinguishes NK cells deficient in FcRγ

As we were particularly interested in analyzing FcRγ^– NK cells in the setting of chronic HIV-1 infection, we selected the HIV-1⁺ MACS participants with a baseline FcRγ^– NK cell frequency greater than 10% for comprehensive analyses (n = 8). An evaluation of the normalized MFIs of phenotypic and functional markers analyzed by flow cytometry illustrated the differential expression patterns of FcRγ^– and conventional (FcRγ⁺) NK cells (Fig. 3A), corroborating that FcRγ^– NK cells from HIV-1⁺ individuals adopt a unique expression profile in comparison to their FcRγ⁺ counterparts (43).

Figure 3. A differential expression signature distinguishes FcRγ^– NK cells from conventional NK cells.

(A) An evaluation of the normalized MFIs of phenotypic and functional markers analyzed by flow cytometry emphasized the differential expression patterns of FcRγ^– and conventional NK cells. (B) Representative staining at baseline for the indicated surface markers on NK cells gated based on the expression of FcRγ. (C) Compared with FcRγ⁺ NK cells from the same donor, FcRγ^– NK cells were characterized by decreased levels of NKp46 and NKG2A but elevated expression of CD57 and NKG2C, (D) resulting in an inverse of the NKG2A to NKG2C ratio in the FcRγ^– population; n = 8 HIV-1⁺ participants with an FcRγ^– NK cell frequency >10%. Statistical significance was determined using the paired Student’s t-test (****p < 0.0001; ***p < 0.001; **p < 0.01; *p < 0.05).

We subsequently probed the phenotypic properties of FcRγ^– NK cells from HIV-1⁺ participants as they related to the descriptions of ‘memory-like’ NK cell subsets delineated in the current literature (38, 39, 44, 45). FcRγ^– NK cells from HCMV seropositive individuals are marked by increased expression of NKG2C and CD57 but a lack of expression of NKG2A (44, 45). They also express lower levels of NKp46 and NKp30. This speaks to their reduced capacity for direct cytotoxicity of target cells (44, 45). Similarly, FcRγ^– NK cells from HIV-1⁺ MACS participants presented less NKp46 on the cell surface, both in terms of percent positivity and MFI (Fig. 3B, 3C). Compared with conventional NK cells, FcRγ^– NK cells featured an enrichment of CD57 (Fig. 3B, 3C; Supplementary Fig. 3), a marker of replicative senescence (58–60), suggesting the FcRγ^– subset is highly differentiated (61). Additionally, the levels of NKG2A, an inhibitory receptor, and NKG2C, an activating receptor, were reduced and elevated, respectively (Fig. 3B, 3C; Supplementary Fig. 3), resulting in an inverse of the NKG2A to NKG2C ratio typically found with conventional NK cells (Fig. 3D). These data corroborate the reported inverse correlation of CD57 expression with NKG2A and NKp46 levels on CD56^dim NK cells (61), and also confirm that FcRγ^– NK cells from HIV-1⁺ men resemble those reported in association with HCMV (44, 45). Notably, though, the phenotypic hallmarks we observed for FcRγ^– NK cells in our study starkly contrast the characteristic qualities of the previously described cytokine-induced ‘memory-like’ NK cells, which are marked by increased levels of NKG2A and NKp46, decreased expression of CD57, and a weak association with NKG2C (39). Therefore, the FcRγ^– ‘memory-like’ NK cells that expand during chronic HIV-1 infection differ phenotypically at baseline from cytokine-induced ‘memory-like’ NK cells.

IL-18 triggers CD16 downregulation exclusively in FcRγ⁺ NK cells

As IL-18 plays a prominent role in programming the differentiation of CD56^dim NK helper cells, we were next interested in exploring the responsiveness of FcRγ^– NK cells to this DC-derived cytokine. NK cells were, once again, enriched from PBLs of the selected HIV-1⁺ participants having an FcRγ^– NK cell frequency greater than 10%, followed by a 24h culture in media alone, in the presence of IL-18, or with a combination of IL-18 and IL-12. We then assessed the phenotypic and functional impact of these cytokine treatments on the FcRγ^– and FcRγ⁺ subsets via flow cytometry. IL-18 alone remarkably influences NK cell differentiation (19), which was highlighted by the shift in overall protein expression patterns seen when performing t-SNE analysis of purified NK cells following a 24h exposure to the cytokine (Fig. 4A). This effect was particularly apparent within the major CD56^dim population of NK cells, which exclusively expressed high levels of CD16 at baseline (7–9) (Fig. 4B). The differentiation of CD56^dim NK cells into a helper cell phenotype is characterized by the downregulation of CD16 expression in response to IL-18, which can be enhanced with exposure to secondary or co-activation signals (19). Indeed, short-term exposure of NK cells isolated from HIV-1-infected participants to IL-18 resulted in a decrease in CD16 expression, with the combination of IL-18 and IL-12 causing a dramatic loss of CD16 in the CD56^dim population (Fig. 4B). Conversely, exposure to IL-12 alone elicited enhanced surface levels of both CD56 and CD16 among CD56^dim NK cells (Fig. 4B). Interestingly, the majority of NK cells downregulating CD16 in response to cytokine co-stimulation were positive for FcRγ expression, but FcRγ^– NK cells predominantly resided within a distinct population that retained expression of CD16 (Fig. 4C). While cytokine co-stimulated FcRγ^– NK cells showed only a slight reduction in CD16 expression intensity relative to unstimulated controls, conventional NK cells exhibited a pronounced decrease in cell surface expression of CD16 (Fig. 4D, 4E), indicating that FcRγ^– NK cells are minimally responsive to IL-18.

Figure 4. FcRγ^– NK cells resist IL-18-induced downregulation of CD16 expression.

(A) Representative t-SNE plots of total NK cells from n = 4 HIV-1⁺ participants cultured for 24h under the indicated conditions, highlighting the shift in protein expression patterns that occurred following treatment with IL-18. However, the FcRγ^– population, represented by the dark band, clustered independently from conventional NK cells and remained relatively constant post-stimulation with IL-18. (B) Representative staining for CD16 on gated total NK cells cultured for 24h under the indicated conditions, depicting IL-18-mediated downregulation of CD16. The synergy between IL-18 and IL-12 exaggerated this effect. (C) An analysis of the CD16⁺ and CD16^– fractions of the CD56^dim population indicated the majority of FcRγ^– NK cells failed to downregulate CD16 in response to co-stimulation with IL-18+IL-12. (D, E) Unlike the FcRγ⁺ subset, cytokine co-stimulated FcRγ^– NK cells weakly diminished CD16 expression relative to unstimulated controls. The bar graphs depict (D) raw MFI values and (E) relative CD16 expression, in which the CD16 MFI of cytokine-treated cells was divided by the CD16 MFI of unstimulated cells; n = 8 HIV-1⁺ participants with an FcRγ^– NK cell frequency >10%. Statistical significance was determined by a one-way ANOVA test (****p < 0.0001; **p < 0.01).

IL-18 does not fuel the development of FcRγ^– NK helper cells

IL-18 primes NK cells for their helper functions by causing the upregulation of various receptors. As a result, these NK cells become poised to receive and respond to a variety of potential secondary activation signals, such as IL-2, IFNα, or IL-12 (19, 62). IL-18 induced the expression of CD25 and the transient expression of CD83, both of which were further upregulated following secondary exposure to or co-stimulation with IL-12 (19, 63) (Fig. 5A). Although IL-12 synergized with IL-18 to intensify the upregulation of CD25 and CD83, IL-12 in the absence of IL-18 was powerless to incite strong expression of either CD25 or CD83, accentuating the primary role of IL-18 in mediating this effect (Supplementary Fig. 4). Importantly, the induction of both CD25 and CD83 predominated in the NK cells expressing FcRγ (Fig. 5B). Examining the FcRγ⁺ and FcRγ^– subsets separately, likewise, revealed that FcRγ^– NK cells were grossly inferior at upregulating CD25 and CD83 in response to IL-18 (Fig. 5C, 5D). Therefore, the ability of IL-18 to drive the differentiation of CD16^–/CD25⁺/CD83⁺ NK helper cells in the FcRγ^– population is impaired in comparison to conventional NK cells.

Figure 5. IL-18 does not drive the differentiation of NK helper cells in the FcRγ^– population.

(A) Representative histogram plots of total NK cells following a 24h culture, demonstrating the upregulation of CD25 and CD83 initiated by IL-18 and augmented by co-stimulation with IL-18+IL-12. (B) Representative staining for CD25, CD83, and IFNγ on gated co-stimulated total NK cells, with the heatmap overlay portraying relative FcRγ expression, and (C) on NK cells stratified based on the expression of FcRγ. (D) IL-18 drove the differentiation of CD25⁺/CD83⁺ NK helper cells. By contrast, FcRγ^– NK cells faltered in their responsiveness to IL-18, illustrated by limited expression of CD25 and CD83. In comparison to their FcRγ⁺ counterparts, FcRγ^– NK cells showed an impaired ability to produce IFNγ; n = 8 HIV-1⁺ participants with an FcRγ^– NK cell frequency >10%. (E) Total NK cells from individuals with an FcRγ^– NK cell frequency >10% (high; n = 4 HIV-1^– MSM (squares), n = 8 HIV-1⁺ MSM (circles)) produced lower levels of IFNγ following a 48h exposure to IL-18+IL-12 relative to those with an FcRγ^– NK cell frequency <10% (low; n = 6 HIV-1^– community controls (triangles), n = 10 HIV-1^– MSM (squares), n = 6 HIV-1⁺ MSM (circles)). (F) The proportion of FcRγ^– NK cells inversely correlated with the frequency of NK cells positive for IFNγ by intracellular staining; n = 6 HIV-1^– community controls (triangles), n = 14 HIV-1^– MSM (squares), n = 14 HIV-1⁺ MSM (circles). (G) Supernatants from NK cell cultures activated with IL-18+IL-2 were collected after 48h and quantified for IFNγ by ELISA, with inferior production noted in HIV-1⁺ participants with an FcRγ^– NK cell frequency >10%. (H) Responder DCs from a single HIV-1⁺ participant became differentially polarized, producing less IL-12 when exposed to cytokine-stimulated NK cell culture supernatants from HIV-1⁺ MSM with an FcRγ^– NK cell frequency >10%; n = 2 replicates from 4 HIV-1⁺ participants with an FcRγ^– NK cell frequency >10% (high) and from 4 HIV-1⁺ participants with an FcRγ^– NK cell frequency <10% (low) (G, H). Statistical significance was determined using the paired Student’s t-test (D), unpaired Student’s t-test (E, G, H), or Pearson correlation (F) (****p < 0.0001; ***p < 0.001; **p < 0.01; *p < 0.05).

NK helper cells largely promote type-1 immune responses by secreting IFNγ in response to combinations of innate stimulatory factors, including IL-18, IFNα, and IL-12 (18, 19, 21) (Supplementary Fig. 1C). However, FcRγ^– NK cells were defective at producing IFNγ in response to cytokine co-stimulation, further illustrating their lack of receptivity to IL-18 (Fig. 5B–5D). Failure to produce IFNγ upon co-activation with IL-18 and IL-12 serves as confirmation of FcRγ^– NK cells not sharing the same differentiation pathway as NK helper cells. Of note, NK cells from individuals with a low FcRγ^– NK cell frequency were more dynamic producers of IFNγ compared to those with an FcRγ^– NK cell frequency greater than 10% (Fig. 5E), and the proportion of FcRγ^– NK cells inversely correlated with the frequency of NK cells positive for IFNγ by intracellular staining (Fig. 5F). Multiple linear regression analysis of the HIV-1^– and HIV-1⁺ individuals revealed a significant relationship between the frequency of IFNγ-expressing NK cells and the proportion of FcRγ^– NK cells (|t| = 5.975; p < 0.0001) but not HIV-1 status (|t| = 0.7102; p = 0.4824). Moreover, total NK cells from HIV-1⁺ MSM with an FcRγ^– NK cell frequency greater than 10% secreted lower quantities of IFNγ (Fig. 5G). The impaired release of IFNγ by cytokine-stimulated NK cells from those with an FcRγ^– NK cell frequency greater than 10% translated into less effective polarization of responder DCs as determined by their IL-12-producing capacity (Fig. 5H), although NK cells from both groups promoted relative increases in DC production of IL-12 compared to the mock-treated DCs (11-fold vs 7-fold increase for FcRγ^– low and FcRγ^– high participants, respectively). These data support the notion that an expanded population of FcRγ^– NK cells in chronic HIV-1 infection negatively impacts the quality of NK-DC crosstalk.

A diminished capacity to express IL18Rα renders FcRγ^– NK cells unresponsive to IL-18

As mentioned previously, IL-18 should increase the sensitivity of NK cells to secondary stimuli, including IL-12, by causing an upregulation of cytokine receptors. This is essential for their subsequent production of IFNγ (19). Based on this, we assessed the expression of IL12Rβ2 and found its induction to be weakened in the FcRγ^– NK cell subset (Fig. 6A–6C). Since IL18R signaling plays a critical role for increased IFNγ production by NK cells (19, 62), we investigated whether the lack of responsiveness of FcRγ^– NK cells expanded in HIV-1-infected individuals to IL-18 was due to a defect in the signaling pathway or a lack of expression of the IL18R itself. While conventional NK cells strongly upregulated IL18Rα in response to co-stimulation with IL-12, FcRγ^– NK cells were characterized by an attenuated capacity to upregulate this receptor (Fig. 6D–6F). Even among conventional NK cells, IL-18 minimally impacted the expression of IL18Rα compared to unstimulated controls (Fig. 6F). In contrast, the enhanced levels imparted by IL-12 alone were comparable to those induced by the combination of IL-18 and IL-12 (64) (Fig. 6G). Additionally, primary exposure of isolated NK cells for 48h to IL-12 triggered robust production of IFNγ upon secondary exposure to IL-18, but not IL-2, highlighting the importance of IL-12 in upregulating IL18Rα for heightened IFNγ responses mediated via IL18R signaling (Fig. 6H, 6I). As co-expression of the alpha and beta subunits of the IL18R is required for IL-18 responsiveness (65, 66), the anergy portrayed by FcRγ^– NK cells to IL-18 is likely explained by a diminished capacity to express the alpha subunit.

Figure 6. Aberrant inducible expression of IL18Rα limits the ability of FcRγ^– NK cells to respond to innate stimuli.

(A, D) Representative staining for the indicated cytokine receptors on gated co-stimulated total NK cells, with the heatmap overlay portraying relative FcRγ expression, and (B, E) on NK cells stratified based on the expression of FcRγ. (C, F) FcRγ deficiency was characterized by an attenuated capacity to express IL12Rβ2 and IL18Rα upon IL-18+IL-12 co-activation; n = 8 HIV-1⁺ participants with an FcRγ^– NK cell frequency >10%. (G) Representative staining for IL18Rα on gated total NK cells. (H, I) Isolated NK cells were cultured in the presence or absence of IL-12 for 48h (primary signal), followed by extensive washing and a 24h culture in the presence of a secondary signal. Pre-exposure of NK cells to IL-12 promoted strong IFNγ responses in total NK cells upon secondary exposure to IL-18 as determined by intracellular staining (H; n = 4 HIV-1⁺ participants) and ELISA (I; n = 2 replicates from 2 HIV-1⁺ participants), where the bar graph represents fold increase in IFNγ relative to IL-12-stimulated controls. Statistical significance was determined using the paired Student’s t-test (C, F), two-way ANOVA (H), or one-way ANOVA (I) (****p < 0.0001; ***p < 0.001; **p < 0.01; *p < 0.05).

FcRγ^– NK cells exhibit a functional bias toward antibody-dependent reactivity

NK cells play a supportive antigen-specific role during the adaptive immune response through engagement of CD16 by the Fc regions of IgG antibodies. These antibody-dependent effector functions include lysis of infected cells by ADCC, which has been implicated in vaccine-induced protective immunity against acquisition of infection, phenotypes of viral control, and slower disease progression (67–70). Thus, we measured the ability of freshly isolated NK cells from participants with chronic HIV-1 infection to perform ADCC by exposing them to rituximab-opsonized and non-opsonized Raji cells, which are resistant to NK cell-mediated killing via natural cytotoxicity receptors, at varying ratios of effector to target cells. Based on the relative proportions of opsonized to non-opsonized target cells, we calculated the specific elimination of antibody-coated targets, with NK cells from virally suppressed HIV-1⁺ MACS participants effectively killing rituximab-opsonized Raji cells, indicated by an NK cell dose-dependent decrease of opsonized target cells (Fig. 7A).

Figure 7. FcRγ^– NK cells specialize in antibody-dependent reactivity.

(A) NK cells from virally suppressed HIV-1⁺ MACS participants effectively mediated ADCC activity against rituximab-opsonized Raji cells, with the highest level of specific lysis observed at the largest effector to target cell ratio. The percentage of ADCC-mediated killing was calculated based on the relative proportion of viable opsonized targets to unopsonized dye-labeled Raji cells; n = 4 HIV-1⁺ participants. (B) FcRγ^– NK cells showed greater degranulation in response to both non-opsonized and rituximab-opsonized Raji cells as determined by cell surface CD107a. (C) Compared to their FcRγ⁺ counterparts, FcRγ^– NK cells produced more IFNγ in response to CD16-mediated signaling. (D) The functional bias of FcRγ^– NK cells toward antibody-dependent signaling was exemplified by their differential production of IFNγ following a 6h culture with opsonized Raji cells (adaptive) or a 48h culture in the presence of IL-18+IL-12 (innate); n = 4 HIV-1⁺ participants with an FcRγ^– NK cell frequency >10% from 6 independent experiments (B-D). Statistical significance was determined by one-way ANOVA (A) or the paired Student’s t-test (B-D) (***p < 0.001; **p < 0.01; *p < 0.05).

In addition to evaluating ADCC activity through true lytic responses, we examined CD107a mobilization and intracellular IFNγ by flow cytometry to elucidate the NK cell subpopulations responding to antibody-mediated signaling through CD16. A comparison of FcRγ⁺ and FcRγ^– NK cells revealed that the latter population exhibited a functional bias toward antibody-dependent reactivity, with increased degranulation, as determined by cell surface CD107a, and IFNγ expression in the presence of antibody-coated target cells (Fig. 7B, 7C). Interestingly, the FcRγ^– subset was also marked by a higher level of spontaneous activity against non-opsonized Raji cells, suggestive of a hyper-reactive state (Fig. 7B, 7C). While conventional NK cells from HIV-1⁺ men responded with flexibility to both innate and adaptive signals, FcRγ^– NK cells were highly restricted to antibody-specific responses (Fig. 7D).

Discussion

FcRγ^– NK cells embody a distinct phenotypic and functional signature defined by aberrant IL-18 responsiveness. Our results suggest chronic HIV-1 infection skews the differentiation of NK cells away from the helper pathway, leading to the accumulation of an unyielding population of FcRγ^– NK cells with limited flexibility to respond to innate stimuli. Epigenetic remodeling of the IFNG locus in FcRγ^– NK cells favors increased production of IFNγ (47, 71, 72), but this display of superiority is misleading as FcRγ^– NK cells are marked by limited versatility, with responsiveness highly restricted to adaptive cues mediated via CD16 signaling (43–45) (Fig. 7). Although total NK cells from individuals with chronic HIV-1 infection appear capable of inducing type-1 polarized mature DCs (Fig. 1C–1E), we demonstrate that production of IFNγ, a potent inducer of DC differentiation (17), is highly impaired when FcRγ^– NK cells are exposed to DC-associated innate cytokine signals (Fig. 5D–5G). Furthermore, our data substantiate the claim that persistence of an expanded population of FcRγ^– NK cells in chronic HIV-1 infection dilutes the capacity for NK-DC crosstalk (Fig. 5H). The defective ability of FcRγ^– NK cells to respond to innate cytokine stimulation appears to stem from diminished expression of IL12Rβ2 and IL18Rα (Fig. 6A–6F); however, formal proof would require a genetic rescue experiment. Additionally, the FcRγ^– subset acquires a mature phenotype, including increased expression of CD57 and NKG2C coupled with reduced expression of NKG2A (Fig. 3), supporting the reported reduced expression of IL12Rβ2 and IL18Rα with increasing status of NK cell maturation (47, 61, 73). The dysfunctional responsiveness of FcRγ^– NK cells to IL-18 extends beyond impaired production of IFNγ (47) (Fig. 5B–5D), with this population of NK cells resisting IL-18-mediated downregulation of CD16 (Fig. 4) and expression of CD25 and CD83 (Fig. 5B–5D). In other words, FcRγ^– NK cells poorly differentiate into NK helper cells, providing further evidence of their phenotypic and functional divergence from conventional NK cells. This concept is validated by genome-wide DNA methylation analyses revealing epigenetic commonalities between FcRγ^– NK and CD8⁺ effector T cells. By contrast, the methylation profile of FcRγ^– NK cells deviates substantially from that of mature canonical NK cells (47).

While the induction of FcRγ^– NK cells is strongly associated with HCMV infection (45), this population of cells is further inflated in HIV-1 infection (43, 49) (Fig. 2F–2J). The mechanism by which HIV-1 contributes to this phenomenon, though, is poorly understood. However, reports suggest that this expansion is not a direct consequence of HIV-1 infection but rather due to the impact of potentially greater cumulative HCMV exposure (43), as evidenced by higher HCMV antibody titers in HIV-1-infected individuals (43, 48, 74). Although ART successfully suppresses HIV-1 viremia and reduces AIDS-associated mortality, persistence of inflammation and innate immune activation due to subclinical HCMV reactivation could contribute to immune dysfunction (43, 75–78), a notion supported by increased incidences of solid tumors and Hodgkin’s lymphoma in treated HIV-1⁺ individuals (79), as well as a higher prevalence of comorbidities with an inflammatory etiology. In particular, markers of innate immune activation are associated with cardiovascular disease, non-AIDS cancers, and neurocognitive disorders (80–84).

Whereas CD4⁺ T cell activation induced by HIV-1 viremia is resolved within 24 months of ART (48), activated NK cells continue to persist in those maintaining undetectable levels of plasma viremia (48, 76), suggesting NK cells are more sensitive to immunologic perturbations (48). Based on these data, we surmise that HIV-1 infection accelerates the expansion of FcRγ^– NK cells through immune dysfunction and poor HCMV control. We also posit that the selective expansion of FcRγ^– NK cells may, in part, be due to the impact of simultaneous subclinical reactivations of both of these latent viruses, which could occur frequently if a portion of the HCMV-responding, antigen-specific CD4⁺ T cells themselves harbor latent HIV-1, as our previous study suggests (85).

Numerous questions linger related to FcRγ^– NK cells, including the mechanisms involved in triggering their differentiation, as well as the duration and implications of their expansion in ART-treated HIV-1⁺ individuals. The specialization of FcRγ^– NK cells plausibly corresponds with reduced plasticity, as they appear to have lost a degree of flexibility, hindering their ability to rapidly respond as immune helper cells through the production of IFNγ. This, in turn, could affect immune homeostasis and have important implications for future HIV-1 cure interventions aiming to elicit potent CTL responses and for the establishment of primary adaptive immunity to new pathogens such as the novel SARS-CoV-2. In fact, recent studies suggest that innate immune dysregulation and a similar NK cell immunotype are related to COVID-19 disease severity (86, 87). It remains to be determined whether the reduced responsiveness of FcRγ^– NK cells to innate stimuli, as observed in vitro in aviremic HIV-1⁺ individuals, plays a role in generating functionally impaired CTLs. Moreover, FcRγ^– NK cells could perpetuate inflammation through their hyperactivity and enhanced ability to produce inflammatory cytokines following antibody-mediated stimulation (43–45), thereby contributing to the long-term health consequences linked to persistent innate immune activation (48). On the contrary, the potential exists for FcRγ^– NK cells to be exploited for antibody-based vaccines or immunotherapies due to their superior in vitro ADCC activity. Although we were unable to explore the in vivo implications of the expansion of FcRγ^– NK cells in the present study, our findings highlight the need for longitudinal studies to discern how their presence in treated chronic HIV-1 infection relates to disease progression, control of heterologous infections, and inflammation-associated comorbidities.

NK cells are uniquely positioned to influence the antiviral response and limit viral spread due to the pleiotropic nature of their effector functions, including the potential to respond directly to infected cells and to modulate the adaptive immune response through the establishment of a vast communication network encompassing both the innate and adaptive arms of immunity (27, 88–90). This elaborate series of interactions, with NK cells at the helm, is critical for an optimal immune response. Therefore, improving our understanding of the magnitude of dysfunction suffered by NK cells during HIV-1 infection, including their interactions with DCs, will be imperative for rescuing their function and improving long-term health outcomes through successful immune reconstitution. Strategies for harnessing the natural powers of NK cells to control HIV-1 infection, modulate adaptive immune responses, and promote overall health and immune homeostasis will be integral in maximizing the effectiveness of HIV-1 therapies.

Key Points.

• A rare subset of FcRγ^– NK cells is highly expanded in chronic HIV-1 infection

• IL18Rα impairment renders FcRγ^– NK cells deficient in helper cell differentiation

• Deficits in IFNγ due to FcRγ^– NK cell expansions lead to inferior NK-DC crosstalk

Abbreviations:

ADCC

antibody-dependent cellular cytotoxicity

ART

antiretroviral therapy

DC

dendritic cell

FMO

fluorescence minus one

HCMV

human cytomegalovirus

iDC

immature dendritic cell

MFI

mean fluorescence intensity

MACS

Multicenter AIDS Cohort Study

MSM

men who have sex with men

rh

recombinant human

t-SNE

t-distributed Stochastic Neighbor Embedding