Lymphocyte activation is regulated by costimulatory and inhibitory receptors, of which both B and T lymphocyte attenuator (BTLA) and CD160 engage herpesvirus entry mediator (HVEM). Notably, it remains unclear how HVEM functions with each of its ligands during immune responses. In this study, we show that HVEM specifically activates CD160 on effector NK cells challenged with virus-infected cells. Human CD56dim NK cells were costimulated specifically by HVEM but not by other receptors that share the HVEM ligands LIGHT, Lymphotoxin-α, or BTLA. HVEM enhanced human NK cell activation by type I IFN and IL-2, resulting in increased IFN-γ and TNF-α secretion, and tumor cell–expressed HVEM activated CD160 in a human NK cell line, causing rapid hyperphosphorylation of serine kinases ERK1/2 and AKT and enhanced cytolysis of target cells. In contrast, HVEM activation of BTLA reduced cytolysis of target cells. Together, our results demonstrate that HVEM functions as a regulator of immune function that activates NK cells via CD160 and limits lymphocyte-induced inflammation via association with BTLA.

Natural killer cells are an essential component of the innate immune system that protect against a wide range of pathogens, particularly against herpesviruses. During the early stages of immune responses to viruses, NK cells are primed by cytokines expressed by pathogen-sensing cells, such as macrophages and dendritic cells (1, 2). Upon maturation, NK cells express a diverse array of receptors that activate cytolysis and cytokine release (35). NK cell activation is restrained by a variety of inhibitory receptors that prevent uncontrolled cytolysis and inflammation through the recognition of self-MHC molecules expressed in healthy, uninfected cells (6). Although many herpesviruses have manipulated the balance between inhibitory and activating signaling to prevent clearance of infected cells, allowing for viral evasion and replication (7, 8), many of the host and pathogen factors that regulate NK cell activation remain unidentified.

The β-herpesvirus, CMV, expresses numerous genes that modulate host-immune responses and, specifically, NK cell activation (9). In human CMV, many of these genes are encoded within the unique long genomic subregion (UL)/b′ that is not essential for in vitro replication (10). The UL144 open reading frame contained within the (UL)/b′ locus was first identified as an expressed transcript encoding a type 1 transmembrane protein and as an ortholog to cellular herpesvirus entry mediator (HVEM; TNFRSF14), a member of the TNFR superfamily (11). HVEM binds the TNF-related ligands LIGHT (TNFSF14) and LT-α (12), and the Ig domain–containing receptors, B and T lymphocyte attenuator (BTLA) (13, 14) and CD160 (15, 16). Although UL144 does not bind LIGHT or LT-α, presumably because it lacks the third and fourth cysteine-rich domains contained in HVEM, it does bind and activate BTLA via CRD1 to attenuate T cell proliferation (17). BTLA activation results in phosphorylation of its cytoplasmic tyrosines and recruitment of the tyrosine phosphatases Src homology domain 2 containing phosphatase-1 and -2, resulting in diminished AgR signaling in T cells and B cells (13, 14, 18). BTLA-expressing T cells are inhibited by HVEM expressed by APCs, regulatory T cells, or mucosal epithelium (16, 19, 20).

The role of CD160 in lymphocyte activation remains unclear. CD160 functions as an inhibitory receptor in a subset of CD4+ T cells (15), whereas increased CD160 expression with reduced BTLA expression in CD8+ T cells is associated with T cell exhaustion in hosts with chronic viral infections (2123). In contrast, CD160 cross-linking by MHC ligands (HLA-C) costimulates CD8+ T cells and activates NK cell cytotoxicity and cytokine production (2427). Activation of HVEM signaling by LIGHT, BTLA, or CD160 enhances Ag-induced T cell proliferation and cytokine production (2831), as well as epithelial cell expression of host defense genes in response to bacterial infection (32). Thus, the HVEM–LIGHT–BTLA–CD160 signaling axis may result in productive or aborted lymphocyte signaling, depending upon which receptor is activated and the cellular context of activation. Furthermore, the nature of the selective pressures mitigated by UL144 as CMV coevolved with primate hosts remains elusive.

In this study, we used HVEM and UL144 as molecular probes to elucidate differences in human NK cell–signaling pathways triggered by viral infection. We observed greater activation of NK cells by HVEM compared with viral UL144, which reflects the inability of UL144 to bind CD160. The uniquely high expression of CD160 by primary CD56dim NK cells in the absence of other HVEM ligands efficiently costimulates NK cell effector functions in response to HVEM binding. In contrast, HVEM binding to NK cells coexpressing BTLA and CD160 inhibits effector functions, such as cytolysis. Thus, CD160 and BTLA regulate NK cell activation through costimulatory and inhibitory pathways activated by HVEM-expressing cells in the microenvironment. These findings reconcile the contextual activity of HVEM through BTLA and CD160 and provide a framework by which this network can be manipulated to control inflammatory responses in human infection and chronic disease, such as cancer.

Fresh human blood was collected from healthy donors giving written informed consent at The Scripps Research Institute Normal Blood Donor Service, and all handling was approved by the Sanford|Burnham Medical Research Institute Internal Review Board. Samples were mixed 1:1 with PBS and overlaid onto Ficoll (GE Healthcare, Uppsala, SE) for density gradient centrifugation. PBMCs were isolated from buffy coats and washed twice with PBS. NK cells were further purified using an EasySep Human NK Cell Enrichment Kit (STEMCELL Technologies, Vancouver, CA) and confirmed to be >95% pure by CD56 staining. Cells resuspended to 1–2 × 106 cells/ml in RPMI 1640 supplemented with 10% heat-inactivated FBS, antibiotics, l-glutamine, and 50 μM 2-ME were first incubated on ice for 15–30 min with Fc fusion proteins or human IgG1 control. For infectious coculture experiments, normal human dermal fibroblast (NHDF) cells were infected with CMV (laboratory strain AD169) at a multiplicity of infection of 1 for 6 h, washed with PBS, and mixed with pretreated PBMCs at a ratio of 100:1 (PBMCs/NHDFs). Productive infection was validated by RT-PCR (Supplemental Fig. 1B). Alternatively, pretreated cells were activated at 37°C in flat-bottom plates with recombinant human IFN-β (R&D Systems, Minneapolis, MN), recombinant human IL-2 (Biogen, Cambridge, MA), or anti-NKG2D (eBioscience, San Diego, CA).

Purified fusion proteins of the extracellular domains of human BTLA, HVEM, human CMV UL144 and variants, and rhesus CMV UL144 with human IgG1 Fc were produced as previously described (17).

Abs used to identify human PBMC populations include CD3 eFluor 450, CD4 PE-Cy7, CD8a allophycocyanin eFluor 780, CD14 FITC, CD19 FITC, BTLA PE, HVEM PE, LIGHT allophycocyanin (eBioscience), CD69 PerCP-Cy5.5, CD107a Alexa Fluor 647, CD160 Alexa Fluor 647, IFN-γ PE-Cy7 (BioLegend, San Diego, CA), CD25 PE, CD56 Alexa Fluor 700 (BD Biosciences, San Diego, CA), and NKG2C PE (R&D Systems).

PBMCs within the live gate of flow cytometric analysis were defined as CD19+ B cells (CD14+/CD19+/SSClow), CD14+ monocytes (CD14+/CD19+/SSChigh), CD4+ T cells (CD14/CD19/CD56/CD3+/CD4+/CD8), CD8+ T cells (CD14/CD19/CD56/CD3+/CD8+/CD4), CD56dim NK cells (CD14/CD19/CD3/CD56dim), and CD56bright NK cells (CD14/CD19/CD3/CD56bright).

CD107a expression was tested by first incubating anti-CD107a and GolgiStop (BD Biosciences) during the final 4 h of PBMC or NK cell activation cultures at a final dilution of 1:1000. Cells were washed and resuspended in buffer for extracellular staining and then washed, fixed in 2% paraformaldehyde, and analyzed.

Phosphatidylinositol-specific phospholipase C (PI-PLC; Invitrogen, Carlsbad, CA) was used to distinguish between the glycophosphoinositide (GPI)-linked and transmembrane forms of CD160.

EL4 and 293T cells were maintained in DMEM supplemented with 10% heat-inactivated FBS, antibiotics, l-glutamine, and 50 μM 2-ME. NK92 cells were maintained in RPMI 1640 supplemented with 8% heat-inactivated FBS, 8% equine serum, antibiotics, l-glutamine, 50 μM 2-ME, and 100 U/ml IL-2.

EL4 cells were transduced with human BTLA-IRES-GFP or human CD160 (Open Biosystems, Huntsville, AL) cloned into IRES-GFP retroviral plasmid by PCR amplification (hCD1605BglII: 5′-AGTCAGATCTGCGTGCAGGATGCTGTTG-3′; hCD1603XhoI: 5′-AGTCCTCGAGGGCTTACAAAGCTTGAAGGG-3′) (33). Pseudotyped single-infection retrovirus was produced by cotransfection of retroviral plasmid, pCG-VSVg envelope protein, and Hit60 gag-pol, as previously described, or by transfection of retroviral plasmid into Phoenix-A cells. EL4 cells were sorted for GFP expression (13).

293T cells were transduced with UL144 derived from human CMV strains cloned in pND vector by calcium phosphate transfection (17). The UL144 G46K mutant was produced by site-directed mutagenesis (FUL144-G46K-5′: 5′-AAACAAGGATATCGTGTTACAAAACAATGTACGCAATATACGAGT-3′; FUL144-G46K-3′: 5′-ACTCGTATATTGCGTACATTGTTTTGTAACACGATATCCTTGTTT-3′). 293T cells were transduced with human BTLA or CD160, as described above, or with de novo synthesized rhesus BTLA or CD160 (Mr. Gene, Regensburg, DE) cloned into IRES-GFP retroviral plasmid by PCR amplification (RhBTLABglII: 5′-AGTCAGATCTGTGCAGGAAATGAAGACATTG-3′; RhBTLAXhoI: 5′-AGTCCTCGAGTCAGAAACAGACTTAACTCCTCACAC-3′; RhCD160BglII: 5′-AGCTAGATCTGCGTGCAGGATGCTGATG-3′; RhCD160XhoI: 5′-AGTCCTCGAGAAGGCTTACAAAGCTTGAAGGACC-3′).

K562 cells were transduced with human HVEM-IRES-GFP or control IRES-GFP retroviral vector and sorted for GFP expression (17).

NK92 cells were transduced with human BTLA-IRES-GFP, human BTLA containing a cytoplasmic domain truncation (BTLAΔCyt)-IRES-GFP, or control IRES-GFP retroviral vector and sorted for GFP expression. BTLAΔCyt mutant was produced by round-the-world PCR (hBTLAR179stop-forward: 5′-TGCTGCCTGTGAAGGCACCAAGGAAAGC-3′; hBTLAR179stop-reverse: 5′-GCCTTCACAGGCAGCAGAACAGGC-3′).

Flow cytometric binding assays were performed as previously described (17). Cells were incubated with Fc ligands for 30 min at 4°C in buffer (PBS with 2% FBS), washed twice, incubated with donkey anti-human Fc allophycocyanin (Jackson ImmunoResearch, West Grove, PA) for 15 min at 4°C in buffer, washed twice, and analyzed. Specific mean fluorescence intensity (MFI) was calculated by subtracting experimental cellular MFI from control cellular MFI.

Supernatants from PBMC and NK activation cultures were analyzed by FlowCytomix and Procarta multiplexing kits (eBioscience), according to the manufacturer’s instructions.

RNA was harvested from PBMC/NHDF mixtures using an RNeasy Mini kit (QIAGEN). cDNA was transcribed using the iScript cDNA Synthesis kit (Bio-Rad, Hercules, CA), and transcripts for HCMV IE1 (HCMV-IE1[+]: 5′-CATCCACACTAGGAGAGCAGACTC-3′; HCMV-IE1[−]: 5′-GCATGAAGGTCTTTGCCCAG-3′), HCMV gB (HCMV-gB[+]: 5′-AACACCCACAGTACCCGTTACG-3′; HCMV-gB[−]: 5′-ATAGAGCCAGGTGCTGCCG), and L32 (L32F: 5′-GGATCTGGCCCTTGAACCTT-3′; L32R: GAAACTGGCGGAAACCCA-3′) were amplified using Power SYBR Green PCR Master Mix (Life Technologies).

Biochemical activation of the NK92 model cell line was triggered by K562 target cells. To block Bcr-Abl–induced signaling, K562 cells were treated with 10 μM imatinib for 60–90 min and washed prior to incubation with NK92 cells. In some experiments, to block PI3K-induced signaling, NK92 cells were treated with 1 μM wortmannin for 60 min and washed prior to incubation with K562 cells. In experiments using fusion proteins, NK92 cells were coated with control human IgG1, LTβR-Fc, HVEM-Fc, UL144-Fc, or HVEM Y61A-Fc for ≥15 min on ice prior to activation. NK92 cells were aliquoted to 2 × 106 cells/condition in 100 μl and mixed with an equal volume of K562 target cells aliquoted to 2 × 105 cells/condition. Cell mixtures were activated at 37°C for the indicated times, quenched with ice-cold PBS, lysed in RIPA buffer at 4°C for 20 min, and centrifuged at 14,000 rpm at 4°C. Extracts were boiled in SDS loading buffer containing 1% 2-ME for 5 min and resolved by SDS-PAGE on 10% Bis-Tris gels (Bio-Rad). Proteins were transferred using the tank method to polyvinylidene difluoride membrane, blocked with 1% OVA in TBST buffer, and blotted with phospho-AKT (S473), phospho-ERK1/2, total AKT (Cell Signaling, Danvers, MA), and total ERK2 (Santa Cruz Biotechnology, Santa Cruz, CA), followed by anti-rabbit HRP (GE Healthcare), and visualized by ECL (Thermo Scientific, Rockford, IL).

NK92 cytotoxicity was assayed using a flow cytometry–based killing assay (34). K562 target cells transduced with HVEM or GFP vector control were labeled with Cell Proliferation Dye e450 (eBioscience), according to the manufacturer’s instructions. NK92 cells transduced with BTLA, BTLAΔCyt, or GFP vector control were cocultured with labeled K562 cells at various E:T ratios for 3 h prior to staining with 7-aminoactinomycin D. Specific lysis was calculated as described. Primary NK cell cytotoxicity was assayed using the JAM protocol (35). K562 target cells were radiolabeled with [3H]thymidine, washed, and incubated with IL-2–activated purified human NK cells treated with fusion proteins or control Ig at various E:T ratios for 4 h. Cultures were harvested onto filtermats, and [3H]thymidine incorporation was counted. Specific lysis was calculated as described.

To test how HVEM and its viral ortholog, UL144, function to regulate immune responses during a viral infection, we monitored the expression of activation markers in cells from human PBMCs mixed with CMV-infected fibroblasts (Fig. 1). As expected, these cocultures were marked by early expression of inflammatory cytokines, including IFN-γ, LT-α, TNF-α, IL-6, IL-8, and IL-17A, which drive innate cellular activation (Supplemental Fig. 1A). Expression of the S1P1-dependent regulator of lymphocyte egress, CD69, was induced on all PBMC subsets and steadily increased throughout the duration of the cocultures. Notably, CD69 expression was uniquely upregulated on CD56dim NK cells through day 3 in cells treated with HVEM-Fc, a bivalent soluble fusion protein of the HVEM ectodomain and Fc region of IgG1 (Fig. 1A). We observed a similar upregulation of CD107a expression on CD56dim NK cells by HVEM-Fc after 1 d of culture (Fig. 1B). Importantly, HVEM induction of CD69 expression was not associated with demographic factors, such as age, sex, or CMV-seropositive status of donors (Supplemental Fig. 1B–D). However, we do note that within CMV-seropositive donors, the ability of HVEM-Fc to costimulate responses to CMV-infected cells was inversely correlated with anti-CMV titers and NKG2C expression in CD56dim NK cells (Fig. 1C–E), both hallmarks of adaptive and innate memory responses to CMV, respectively (3638). Nevertheless, CD56dim NK cells from all but one donor were responsive to HVEM-Fc costimulation. Thus, HVEM-Fc specifically enhances early activation of CD56dim NK cells during responses to CMV.

FIGURE 1.

HVEM-Fc enhances activation of human CD56dim NK cells in response to virus-infected cells. (A and B) Freshly isolated PBMCs cultured with mock- or CMV-infected NHDF cells were left untreated or were treated with HVEM-Fc, UL144-Fc, or human Ig control. Graphs show the percentage of cells expressing CD69 within CD3+CD8+, CD3+CD4+, CD19+, CD56dim, CD56bright, or CD14+ gates over 1 wk of culture (A) or the percentage of cells expressing surface CD107a within CD56dim, CD56bright, and CD3+CD8+ cells following overnight culture (B). Results are representative of at least two separate experiments with at least four donors each. Graphs show mean ± SEM. Significant p values are shown in (B). Costimulation of CD56dim NK cells with HVEM-Fc was plotted against CMV IgG titers (C) and the frequency of NKG2C expression within CD56dim cells in CMV-seropositive donors (CMV IgG > 0.5 IU/ml) (D). Costimulation of CD56dim NK cells is calculated as the percentage expression of CD69 with HVEM-Fc treatment minus the percentage expression of CD69 with control Ig treatment (HVEM-Fc–induced CD69 [%]). (E) CMV IgG titers in CMV-seropositive donors are plotted against the frequency of NKG2C expression within CD56dim cells. *p < 0.05, **p < 0.01, between Ig and HVEM-Fc.

FIGURE 1.

HVEM-Fc enhances activation of human CD56dim NK cells in response to virus-infected cells. (A and B) Freshly isolated PBMCs cultured with mock- or CMV-infected NHDF cells were left untreated or were treated with HVEM-Fc, UL144-Fc, or human Ig control. Graphs show the percentage of cells expressing CD69 within CD3+CD8+, CD3+CD4+, CD19+, CD56dim, CD56bright, or CD14+ gates over 1 wk of culture (A) or the percentage of cells expressing surface CD107a within CD56dim, CD56bright, and CD3+CD8+ cells following overnight culture (B). Results are representative of at least two separate experiments with at least four donors each. Graphs show mean ± SEM. Significant p values are shown in (B). Costimulation of CD56dim NK cells with HVEM-Fc was plotted against CMV IgG titers (C) and the frequency of NKG2C expression within CD56dim cells in CMV-seropositive donors (CMV IgG > 0.5 IU/ml) (D). Costimulation of CD56dim NK cells is calculated as the percentage expression of CD69 with HVEM-Fc treatment minus the percentage expression of CD69 with control Ig treatment (HVEM-Fc–induced CD69 [%]). (E) CMV IgG titers in CMV-seropositive donors are plotted against the frequency of NKG2C expression within CD56dim cells. *p < 0.05, **p < 0.01, between Ig and HVEM-Fc.

Close modal

To identify which HVEM ligands were expressed on lymphocytes, we examined the expression of BTLA, CD160, LIGHT, and HVEM in human peripheral blood (Fig. 2A–D). B cells showed high BTLA expression and low CD160 expression, whereas T cells and monocytes expressed intermediate levels of both BTLA and CD160. Importantly, CD56dim NK cells showed the highest surface expression of CD160, confirming previously reported data (39), and the lowest BTLA expression among all PBMCs, whereas CD56bright NK cells expressed low levels of both BTLA and CD160. LIGHT was specifically expressed by monocytes and CD8+ T cells and weakly by NK cells. In contrast, HVEM was broadly expressed by all PBMCs.

FIGURE 2.

Expression of HVEM and its ligands in human CD56dim NK cells. Box-and-whisker plots show MFI of CD160 (A), BTLA (B), LIGHT (C), or HVEMs (D) in PBMCs gated on CD19+ B cells, CD14+ monocytes, CD3+CD4+ T cells, CD3+CD8+ T cells, CD56dim NK cells, and CD56bright NK cells. (E and F) NK92 cells or PBMCs were left untreated or treated with the indicated doses of PI-PLC. Overlaid graphs show CD160 expression in NK92 cells (E) or in NKG2C+/− CD56dim and CD56bright NK cells (F), with and without PI-PLC treatment. (G) CD56+ cells purified from whole blood were left untreated or activated with anti-NKG2D for 2 d and then analyzed for surface marker expression. Dot plots show CD56 plotted against BTLA, CD160, HVEM, and LIGHT, with receptor-positive CD56dim and CD56bright fractions indicated. Results are representative of three independent donors.

FIGURE 2.

Expression of HVEM and its ligands in human CD56dim NK cells. Box-and-whisker plots show MFI of CD160 (A), BTLA (B), LIGHT (C), or HVEMs (D) in PBMCs gated on CD19+ B cells, CD14+ monocytes, CD3+CD4+ T cells, CD3+CD8+ T cells, CD56dim NK cells, and CD56bright NK cells. (E and F) NK92 cells or PBMCs were left untreated or treated with the indicated doses of PI-PLC. Overlaid graphs show CD160 expression in NK92 cells (E) or in NKG2C+/− CD56dim and CD56bright NK cells (F), with and without PI-PLC treatment. (G) CD56+ cells purified from whole blood were left untreated or activated with anti-NKG2D for 2 d and then analyzed for surface marker expression. Dot plots show CD56 plotted against BTLA, CD160, HVEM, and LIGHT, with receptor-positive CD56dim and CD56bright fractions indicated. Results are representative of three independent donors.

Close modal

We next sought to distinguish the expression of CD160 splice variants encoding GPI or transmembrane cellular linkages using PI-PLC treatment, which removes GPI-linked proteins (40). We found that, although most primary NK cells retained some uncleavable fraction of CD160, similar to levels observed on the NK92 cell line, nearly all surface-expressed CD160 was cleaved from primary NKG2C+ NK cells (Fig. 2E, 2F). Finally, human NK cells upregulated the expression of BTLA in response to stimulation through the activating receptor NKG2D, similar to BTLA expression on T cells stimulated through the TCR (Fig. 2G) (41). Thus, in resting human NK cells, CD160 is the predominant HVEM receptor, a portion of which is not GPI linked, whereas in activated NK cells BTLA and CD160 are coexpressed and potentially compete for HVEM binding.

UL144 is a structural ortholog of cellular HVEM; however, the engagement of CD160 by UL144 has not been determined in NK cells. We measured UL144 binding to cells expressing human BTLA or CD160 using purified HVEM-Fc or UL144-Fc proteins (Fig. 3A, 3B). We found that UL144-Fc only bound cells expressing BTLA and not CD160, whereas HVEM-Fc bound both BTLA- and CD160-expressing cells with similar disassociation constants and required overlapping surfaces of HVEM to bind these receptors, because binding to both receptors was abrogated by the Y61A HVEM mutant (Fig. 3C) (42). We sought to determine whether the failure to detect UL144–CD160 binding was due to a low affinity of UL144 for CD160 using a UL144 mutant (G46K) identified while mapping the binding surface of UL144 that bound BTLA with high affinity (J.R. Šedý, W. Smith, I. Nemčovičova, P.S. Norris, D. Zajonc, C.F. Ware, manuscript in preparation). In this regard, CD160 failed to show any binding to UL144 (Fig. 3D), although BTLA showed robust binding to the UL144 G46K mutant. The ectodomain of UL144 is highly polymorphic across primate CMV, with five distinct human CMV isoforms diverging up to 36% in their amino acid sequences (43). We examined UL144 selectivity for BTLA throughout these diverse sequences using representative UL144 variants derived from clinical human CMV strains (Fig. 3E). Despite the extensive sequence divergence, BTLA-Fc bound all UL144 variants (17), whereas CD160-Fc failed to bind any. We note one exception: UL144 from rhesus CMV bound human and rhesus CD160 with low affinity (Fig. 3D, 3F). This interaction likely represents a divergence between viral species and not host species, because primate BTLA and CD160 are highly homologous (J.R. Šedý and C.F. Ware, unpublished observations). Together, these data show that UL144 is a highly selective molecule that mimics HVEM binding, yet discriminates between BTLA and CD160 and selectively uses the inhibitory BTLA pathway. These results suggest that HVEM engagement and activation of CD160 may serve as a critical regulatory pathway for human NK cells.

FIGURE 3.

HVEM, but not its ortholog UL144, binds CD160. (A and B) Human BTLA- or CD160-expressing EL4 cells were stained with the indicated concentrations of HVEM-Fc or human CMV UL144-Fc. EC50 values were calculated using four-parameter (variable slope) analysis. (C) Cells used above were stained with 20 μg/ml of wild-type, Y61A, or K64A HVEM-Fc. (D) Cells used above were stained with 20 μg/ml of Fiala strain human CMV, G46K, or rhesus CMV UL144-Fc. (E) Representative human CMV group UL144- or HVEM-expressing 293T cells were stained with 50 μg/ml of BTLA-Fc (white bars) or CD160-Fc (black bars). (F) Human or rhesus BTLA- or CD160- expressing 293T cells were stained with 20 μg/ml of HVEM-Fc, human CMV UL144-Fc, or rhesus CMV UL144-Fc. Dot plots of GFP plotted against anti-human Fc show selective loss of interaction between CD160 and human CMV UL144. *No staining.

FIGURE 3.

HVEM, but not its ortholog UL144, binds CD160. (A and B) Human BTLA- or CD160-expressing EL4 cells were stained with the indicated concentrations of HVEM-Fc or human CMV UL144-Fc. EC50 values were calculated using four-parameter (variable slope) analysis. (C) Cells used above were stained with 20 μg/ml of wild-type, Y61A, or K64A HVEM-Fc. (D) Cells used above were stained with 20 μg/ml of Fiala strain human CMV, G46K, or rhesus CMV UL144-Fc. (E) Representative human CMV group UL144- or HVEM-expressing 293T cells were stained with 50 μg/ml of BTLA-Fc (white bars) or CD160-Fc (black bars). (F) Human or rhesus BTLA- or CD160- expressing 293T cells were stained with 20 μg/ml of HVEM-Fc, human CMV UL144-Fc, or rhesus CMV UL144-Fc. Dot plots of GFP plotted against anti-human Fc show selective loss of interaction between CD160 and human CMV UL144. *No staining.

Close modal

The production of cytokines, particularly type I IFN, during early virus infection promotes NK cell differentiation into active effector cells to help limit infection (2, 44, 45). To test how HVEM and its CMV viral ortholog UL144 regulate cytokine-mediated activation of NK cells, we monitored the expression of activation markers in lymphoid cells from human peripheral blood stimulated with IFN-β or IL-2 (46). Notably, HVEM-Fc consistently enhanced the number of CD56dim NK cells expressing CD69, with or without IFN-β stimulation, in 18-h cultures (Fig. 4A, 4B), as well as in 8-h cultures stimulated with IL-2 (Fig. 4C, 4D). Additionally, in purified NK cells stimulated with IFN-β or IL-2, HVEM-Fc costimulated expression of the high-affinity IL-2Rα subunit, CD25, and the cytolytic granule marker, CD107a, in the CD56dim subset (Fig. 5A–D). Furthermore, HVEM-Fc costimulation specifically increased the levels of secreted IFN-γ and TNF-α in IFN-β–treated NK cells (Fig. 5E). Together, these data indicated that accessory cells were not required for the activity of HVEM-Fc in NK cells. In contrast, UL144-Fc did not promote NK cell activation. Thus, HVEM-Fc directly costimulates cytokine-induced activation and expression of inflammatory cytokines by CD56dim NK cells.

FIGURE 4.

HVEM-Fc costimulates type I IFN and IL-2 activation of CD56dim NK cells within PBMCs. Freshly isolated PBMCs were treated with HVEM-Fc or human Ig control and stimulated with 20 U/ml of IFN-β for 18 h or with 10 or 100 U/ml of IL-2 for 8 h. The percentage of CD56dim cells that are CD69+ within gated CD14/CD19/CD3 cells are shown in dot plots of CD56 versus CD69 with the percentage of CD56dim/CD69+ cells indicated for representative donors in (A) and (C) and graphed in (B) and (D). Results are representative of at least three separate experiments with at least four donors each. Graphs show mean + SEM. Specific p values are shown.

FIGURE 4.

HVEM-Fc costimulates type I IFN and IL-2 activation of CD56dim NK cells within PBMCs. Freshly isolated PBMCs were treated with HVEM-Fc or human Ig control and stimulated with 20 U/ml of IFN-β for 18 h or with 10 or 100 U/ml of IL-2 for 8 h. The percentage of CD56dim cells that are CD69+ within gated CD14/CD19/CD3 cells are shown in dot plots of CD56 versus CD69 with the percentage of CD56dim/CD69+ cells indicated for representative donors in (A) and (C) and graphed in (B) and (D). Results are representative of at least three separate experiments with at least four donors each. Graphs show mean + SEM. Specific p values are shown.

Close modal
FIGURE 5.

HVEM-Fc coactivates type I IFN and IL-2 induction of inflammatory effectors by CD56dim NK cells. Purified CD56+ cells from whole blood were untreated or treated with HVEM-Fc, UL144-Fc, or human Ig control and stimulated overnight with 20 U/ml of IFN-β or 10 U/ml IL-2. (A and C) Overlaid graphs of cells from representative donors show expression of CD69 (top row), CD25 (middle row), or CD107a (bottom row) in CD56dim and CD56bright NK cells. (B and D) Graphs show the percentage of CD56dim cells expressing CD69 (top panels), CD25 (middle panels), and CD107a (bottom panels). Results are representative of two separate experiments with at least four donors each, mean + SEM and specific p values are shown. (E) Culture supernatants were collected and assayed for the presence of IFN-γ and TNF-α. Graphs show mean + SEM from two experiments. Significant p values are shown. *None detected.

FIGURE 5.

HVEM-Fc coactivates type I IFN and IL-2 induction of inflammatory effectors by CD56dim NK cells. Purified CD56+ cells from whole blood were untreated or treated with HVEM-Fc, UL144-Fc, or human Ig control and stimulated overnight with 20 U/ml of IFN-β or 10 U/ml IL-2. (A and C) Overlaid graphs of cells from representative donors show expression of CD69 (top row), CD25 (middle row), or CD107a (bottom row) in CD56dim and CD56bright NK cells. (B and D) Graphs show the percentage of CD56dim cells expressing CD69 (top panels), CD25 (middle panels), and CD107a (bottom panels). Results are representative of two separate experiments with at least four donors each, mean + SEM and specific p values are shown. (E) Culture supernatants were collected and assayed for the presence of IFN-γ and TNF-α. Graphs show mean + SEM from two experiments. Significant p values are shown. *None detected.

Close modal

We next tested whether activation of NK cells by target cells was affected by HVEM ligation by CD160 using the NK92 cell line as a model of activated NK cells. Similar to the results obtained using peripheral blood CD56dim NK cells, NK92 cells expressed abundant CD160 and low BTLA (Fig. 2E, Supplemental Fig. 2A). We observed rapid ERK1/2 phosphorylation, followed by AKT phosphorylation in NK92 cells mixed with K562 target cells (Fig. 6A). Notably, ERK1/2 and AKT were hyperphosphorylated in NK92 cells coated with HVEM-Fc but not with control Ig, LTβR-Fc, or the Y61A mutant HVEM-Fc. In contrast, ERK1/2 and AKT phosphorylation in NK92 cells was reduced upon UL144 treatment compared with control Ig (Fig. 6A, 6B). Because LIGHT ligation does not coactivate ERK1/2 and AKT signaling, and BTLA ligation inhibits ERK1/2 and AKT signaling, we reasoned that HVEM-induced ERK1/2 and AKT activation occurs via CD160, consistent with previous reports of CD160 signaling (40, 47). To rule out a role for FcR binding in HVEM–CD160 costimulation, we used target cells expressing high levels of HVEM (Supplemental Fig. 2B). Similar to NK92 cells coated with HVEM-Fc, K562 cells expressing high levels of HVEM stimulated robust ERK1/2 and AKT phosphorylation in NK92 cells (Fig. 6C). Importantly, we confirmed that AKT phosphorylation was occurring in the NK92 cells and not in the target cells, because pretreatment of NK92 cells with the PI3K inhibitor wortmannin blocked target cell–induced AKT activation (Fig. 6D). Together, these results demonstrate that HVEM engagement of CD160, and not other receptors, promotes signaling downstream of target cell recognition in NK cells.

FIGURE 6.

Target cell–expressed HVEM costimulates AKT signaling in NK cells. NK92 cells were treated with human IgG1, HVEM-Fc, and UL144-Fc (A) or with human IgG1, LTβR-Fc, HVEM-Fc, and HVEM Y61A-Fc (B) and stimulated with imatinib-treated K562 cells for the indicated times. Untreated NK92 cells or wortmannin-treated NK92 cells were stimulated with imatinib-treated K562 cells transduced with GFP control– or HVEM-expressing vector for the indicated times (C) or for 15 min (D). Western blots of whole-cell extracts show phospho-ERK1/2 and phospho-AKT (S473) to monitor activation and total AKT and total ERK2 to control for total protein levels. K562 cells alone are shown to show target cell-specific signals. Untreated or imatinib-treated K562 cells are shown where indicated to show target cell contribution to phosphorylation signals. Im, imatinib.

FIGURE 6.

Target cell–expressed HVEM costimulates AKT signaling in NK cells. NK92 cells were treated with human IgG1, HVEM-Fc, and UL144-Fc (A) or with human IgG1, LTβR-Fc, HVEM-Fc, and HVEM Y61A-Fc (B) and stimulated with imatinib-treated K562 cells for the indicated times. Untreated NK92 cells or wortmannin-treated NK92 cells were stimulated with imatinib-treated K562 cells transduced with GFP control– or HVEM-expressing vector for the indicated times (C) or for 15 min (D). Western blots of whole-cell extracts show phospho-ERK1/2 and phospho-AKT (S473) to monitor activation and total AKT and total ERK2 to control for total protein levels. K562 cells alone are shown to show target cell-specific signals. Untreated or imatinib-treated K562 cells are shown where indicated to show target cell contribution to phosphorylation signals. Im, imatinib.

Close modal

We next tested whether enhanced biochemical activation of NK cells by HVEM–CD160 engagement was correlated with enhanced NK cell lytic function. Additionally, we determined whether the presence of BTLA alters lytic activity using NK92 cells expressing high levels of BTLA or a BTLA mutant lacking the BTLA-signaling domain (BTLAΔCyt) (Supplemental Fig. 2C). In control NK92 cells, we observed increased lysis of target cells expressing HVEM compared with control K562 cells, consistent with HVEM costimulation through CD160 (Fig. 7A). Interestingly, NK92 cells expressing high levels of BTLA showed reduced lysis of target cells expressing HVEM compared with control K562 cells, consistent with HVEM inhibition through BTLA. Importantly, in NK92 cells expressing high levels of BTLAΔCyt, we again observed increased lysis of target cells expressing HVEM compared with control K562 cells, demonstrating that BTLA signaling could inhibit lytic activity of NK cells. We also measured lytic function of IL-2–activated primary NK cells treated with fusion proteins targeting HVEM ligands (Fig. 7B). In this regard, NK cells treated with HVEM-Fc showed greater lytic activity compared with NK cells treated with LTβR-Fc, UL144-Fc, or the Y61A HVEM-Fc mutant, consistent with HVEM costimulation of lytic activity through CD160 (26). Thus, HVEM costimulates or inhibits NK cytolysis, depending on whether CD160 or BTLA is activated (Fig. 8).

FIGURE 7.

HVEM enhances lytic activity of NK cells. (A) Titrated GFP-, BTLA-, or BTLAΔCyt-expressing NK92 cells were incubated with labeled GFP- or HVEM-expressing K562 cells. Graphs show the specific lysis of K562 target cell lines following incubation with effector cells. Curves are mean with replicates plotted and are representative of at least two experiments. (B) IL-2–activated purified human NK cells were incubated with the indicated fusion proteins and titrated onto K562 target cells. Graph shows specific lysis of target cells following incubation with effector cells. Curves are means ± SEM and are representative of at least two experiments.

FIGURE 7.

HVEM enhances lytic activity of NK cells. (A) Titrated GFP-, BTLA-, or BTLAΔCyt-expressing NK92 cells were incubated with labeled GFP- or HVEM-expressing K562 cells. Graphs show the specific lysis of K562 target cell lines following incubation with effector cells. Curves are mean with replicates plotted and are representative of at least two experiments. (B) IL-2–activated purified human NK cells were incubated with the indicated fusion proteins and titrated onto K562 target cells. Graph shows specific lysis of target cells following incubation with effector cells. Curves are means ± SEM and are representative of at least two experiments.

Close modal
FIGURE 8.

Model of HVEM and human CMV UL144 regulation of NK cell activation. HVEM binding to CD160 expressed by NK cells costimulates activation signals from cytokines and target cells. HVEM and human CMV UL144 binding to BTLA on NK cells inhibits NK activation, resulting in attenuated effector function.

FIGURE 8.

Model of HVEM and human CMV UL144 regulation of NK cell activation. HVEM binding to CD160 expressed by NK cells costimulates activation signals from cytokines and target cells. HVEM and human CMV UL144 binding to BTLA on NK cells inhibits NK activation, resulting in attenuated effector function.

Close modal

We report that HVEM interaction with CD160 on NK cells results in costimulation of NK cell effector function (Fig. 8). Specifically, HVEM binding to CD160 enhances CD69 expression, inflammatory cytokine expression, degranulation (as measured by CD107a expression), and cytolysis by NK cells. Importantly, we show that HVEM activity is not the result of HVEM association with other ligands, such as LIGHT or BTLA, because these proteins are poorly expressed on NK cells, and because LTβR, the Y61A mutant HVEM, and the viral protein UL144 all fail to induce costimulatory activity on NK cells. NK cells express abundant levels of the HVEM ligand CD160, which was shown to activate NK cells when engaged with MHC-related ligands (39). Interestingly, we find that the viral protein UL144 selectively binds BTLA and not CD160, thus exclusively activating inhibitory signaling through BTLA (Fig. 8).

Our results are consistent with the idea that CD160 is an activating receptor. Recent work demonstrated the presence of an alternative splice variant of CD160 coding for a tyrosine-containing cytoplasmic tail that may recruit Fyn, SHC1, or the p85 subunit of PI3K (48), which is required for induction of ERK1/2 phosphorylation and proliferation in Jurkat cells (40, 47). We identified PI-PLC–resistant CD160 in NKG2C NK cells, whereas the majority of CD160 was cleaved in NKG2C+ NK cells. Notably, it is difficult to estimate the proportion of GPI- or transmembrane-linked CD160, because Abs do not recognize these forms equivalently (J.R. Šedý and C.F. Ware, unpublished observations) (40). Nevertheless, the increased proportion of transmembrane-linked CD160 suggests that NKG2C cells are poised to respond to HVEM coupled with inflammatory stimuli. Together with the extensive reports that there is an expansion of the NKG2C+ NK cell subset in human CMV–infected individuals (36), we propose that, as the NK compartment adapts to human CMV, these cells become less reliant on nonspecific cues from the environment, such as HVEM. Thus, HVEM–CD160 may be a pathway to activate Ag in experienced NK cells.

The ubiquitous expression of the CD160 ligand, HVEM (15), and MHC proteins (24, 26) in lymphoid and nonlymphoid cells, including mucosal epithelia cells, raises the possibility of constitutive NK cell activation. It is important to note that HVEM induces robust NK cell effector activity only in conjunction with cytokines (IFN-β or IL-2) or direct target cell contact. Thus, the HVEM–CD160 interaction functions to costimulate NK cell activity in the context of inflammation. Additionally, BTLA upregulation competes with CD160 to provide negative feedback for NK cell activation, although it is unclear whether BTLA antagonizes CD160 signaling directly. In our model, UL144 selectivity for BTLA prevents costimulation of NK cells and, thus, may attenuate antiviral functions to promote viral replication and spread.

CD160 was shown to act as an inhibitory receptor in a fraction of CD4+ T cells notably lacking a transmembrane variant of CD160, although it remains unclear how GPI-linked proteins initiate inhibitory signaling (15). BTLA activation reduces CD3ζ phosphorylation in T cells and Syk, BLNK, and PLCγ2 phosphorylation in B cells (18, 49). Thus, human CMV has evolved UL144 as a BTLA-specific ligand to inhibit lymphocyte activation, and to prevent NK cell activation of effector functions associated with HVEM and CD160. Of note, activation of CD160 by HVEM may be primate specific, because CD160 alternative transcripts have not been identified, and CD160 loci do not encode tyrosine-containing cytoplasmic domains in nonprimates (J.R. Šedý and C.F. Ware, unpublished observations). Additionally, mouse NK cells may not be activated through CD160 (50). Thus, we suggest that primate CMV may have coevolved the expression of UL144 in response to the evolution of CD160 as an activating receptor in primates.

The association between different strains of CMV and disease outcome in congenital or postnatal infection is controversial (51, 52). Nevertheless, there continue to be reports that specific CMV variants encoding unique UL144 sequences may be associated with termination of pregnancy, newborn viremia, symptomatic infection, and developmental sequelae (53, 54). The uniform BTLA selectivity among all UL144 variants implies that UL144 has a particularly forgiving structure; however, the factors that drive hypervariability of the UL144 ectodomain remain elusive (43). In addition, UL144 can regulate NF-κB–dependent–signaling pathways (55), and it was recently revealed to be expressed in human myeloid cells latently infected with specific isolates of CMV (56), strongly suggesting multiple functions for this viral HVEM ortholog. HVEM may also be a proinflammatory factor in tumors. In this regard, in follicular lymphoma, the most common secondary karyotypic alteration at 1p36 is due to deletions or mutations in TNFRSF14, which are associated with poor prognosis (57). In accordance with the cancer immunoediting model (58), HVEM-deficient tumors may escape immunosurveillance by CD160-expressing NK cells to acquire additional mutations. Thus, regulation of HVEM–BTLA–CD160 may represent a common immune-evasion mechanism used by both viruses and tumors and suggests that manipulation of this regulatory network may serve as a potential therapeutic target to control inflammatory responses.

We thank the personnel of the Flow Cytometry Shared Resources facility at Sanford|Burnham Medical Research Institute for technical assistance and Ana V. Miletic Šedý and Daniel L. Popkin for critical reading of the manuscript.

This work was supported by grants from the National Institutes of Health (AI-033068, AI48073, and CA164679 to C.F.W. and AI069298 to C.A.B.), postdoctoral training fellowships from the National Institutes of Health (T32A1060536 to J.R.S. and T32CA121949 to R.L.B.), a Diabetes and Immune Disease National Research Institute Fellowship (DIDNRI/10 to J.R.S.), a Type 1 Diabetes Clinical Research Fellowship (T1D-CRF to J.R.S.), and a gift from the Perkins Family Foundation (to C.F.W.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

BTLA

B and T lymphocyte attenuator

GPI

glycophosphoinositide

HVEM

herpesvirus entry mediator

MFI

mean fluorescence intensity

NHDF

normal human dermal fibroblast

PI-PLC

phosphatidylinositol-specific phospholipase C.

1
Lodoen
M. B.
,
Lanier
L. L.
.
2006
.
Natural killer cells as an initial defense against pathogens.
Curr. Opin. Immunol.
18
:
391
398
.
2
Baranek
T.
,
Manh
T. P.
,
Alexandre
Y.
,
Maqbool
M. A.
,
Cabeza
J. Z.
,
Tomasello
E.
,
Crozat
K.
,
Bessou
G.
,
Zucchini
N.
,
Robbins
S. H.
, et al
.
2012
.
Differential responses of immune cells to type I interferon contribute to host resistance to viral infection.
Cell Host Microbe
12
:
571
584
.
3
Lanier
L. L.
2009
.
DAP10- and DAP12-associated receptors in innate immunity.
Immunol. Rev.
227
:
150
160
.
4
Vivier
E.
,
Raulet
D. H.
,
Moretta
A.
,
Caligiuri
M. A.
,
Zitvogel
L.
,
Lanier
L. L.
,
Yokoyama
W. M.
,
Ugolini
S.
.
2011
.
Innate or adaptive immunity? The example of natural killer cells.
Science
331
:
44
49
.
5
Elliott
J. M.
,
Yokoyama
W. M.
.
2011
.
Unifying concepts of MHC-dependent natural killer cell education.
Trends Immunol.
32
:
364
372
.
6
Parham
P.
,
Moffett
A.
.
2013
.
Variable NK cell receptors and their MHC class I ligands in immunity, reproduction and human evolution.
Nat. Rev. Immunol.
13
:
133
144
.
7
Reddehase
M. J.
2002
.
Antigens and immunoevasins: opponents in cytomegalovirus immune surveillance.
Nat. Rev. Immunol.
2
:
831
844
.
8
Tortorella
D.
,
Gewurz
B. E.
,
Furman
M. H.
,
Schust
D. J.
,
Ploegh
H. L.
.
2000
.
Viral subversion of the immune system.
Annu. Rev. Immunol.
18
:
861
926
.
9
Lanier
L. L.
2008
.
Evolutionary struggles between NK cells and viruses.
Nat. Rev. Immunol.
8
:
259
268
.
10
Cha
T.-A.
,
Tom
E.
,
Kemble
G. W.
,
Duke
G. M.
,
Mocarski
E. S.
,
Spaete
R. R.
.
1996
.
Human cytomegalovirus clinical isolates carry at least 19 genes not found in laboratory strains.
J. Virol.
70
:
78
83
.
11
Benedict
C. A.
,
Butrovich
K. D.
,
Lurain
N. S.
,
Corbeil
J.
,
Rooney
I.
,
Schneider
P.
,
Tschopp
J.
,
Ware
C. F.
.
1999
.
Cutting edge: a novel viral TNF receptor superfamily member in virulent strains of human cytomegalovirus.
J. Immunol.
162
:
6967
6970
.
12
Mauri
D. N.
,
Ebner
R.
,
Montgomery
R. I.
,
Kochel
K. D.
,
Cheung
T. C.
,
Yu
G. L.
,
Ruben
S.
,
Murphy
M.
,
Eisenberg
R. J.
,
Cohen
G. H.
, et al
.
1998
.
LIGHT, a new member of the TNF superfamily, and lymphotoxin alpha are ligands for herpesvirus entry mediator.
Immunity
8
:
21
30
.
13
Sedy
J. R.
,
Gavrieli
M.
,
Potter
K. G.
,
Hurchla
M. A.
,
Lindsley
R. C.
,
Hildner
K.
,
Scheu
S.
,
Pfeffer
K.
,
Ware
C. F.
,
Murphy
T. L.
,
Murphy
K. M.
.
2005
.
B and T lymphocyte attenuator regulates T cell activation through interaction with herpesvirus entry mediator.
Nat. Immunol.
6
:
90
98
.
14
Gonzalez
L. C.
,
Loyet
K. M.
,
Calemine-Fenaux
J.
,
Chauhan
V.
,
Wranik
B.
,
Ouyang
W.
,
Eaton
D. L.
.
2005
.
A coreceptor interaction between the CD28 and TNF receptor family members B and T lymphocyte attenuator and herpesvirus entry mediator.
Proc. Natl. Acad. Sci. USA
102
:
1116
1121
.
15
Cai, G., A. Anumanthan, J. A. Brown, E. A. Greenfield, B. Zhu, and G. J. Freeman. 2008. CD160 inhibits activation of human CD4(+) T cells through interaction with herpesvirus entry mediator. Nat. Immunol. 9: 176–185.
16
Murphy
T. L.
,
Murphy
K. M.
.
2010
.
Slow down and survive: Enigmatic immunoregulation by BTLA and HVEM.
Annu. Rev. Immunol.
28
:
389
411
.
17
Cheung
T. C.
,
Humphreys
I. R.
,
Potter
K. G.
,
Norris
P. S.
,
Shumway
H. M.
,
Tran
B. R.
,
Patterson
G.
,
Jean-Jacques
R.
,
Yoon
M.
,
Spear
P. G.
, et al
.
2005
.
Evolutionarily divergent herpesviruses modulate T cell activation by targeting the herpesvirus entry mediator cosignaling pathway.
Proc. Natl. Acad. Sci. USA
102
:
13218
13223
.
18
Vendel
A. C.
,
Calemine-Fenaux
J.
,
Izrael-Tomasevic
A.
,
Chauhan
V.
,
Arnott
D.
,
Eaton
D. L.
.
2009
.
B and T lymphocyte attenuator regulates B cell receptor signaling by targeting Syk and BLNK.
J. Immunol.
182
:
1509
1517
.
19
Tao
R.
,
Wang
L.
,
Murphy
K. M.
,
Fraser
C. C.
,
Hancock
W. W.
.
2008
.
Regulatory T cell expression of herpesvirus entry mediator suppresses the function of B and T lymphocyte attenuator-positive effector T cells.
J. Immunol.
180
:
6649
6655
.
20
Steinberg
M. W.
,
Turovskaya
O.
,
Shaikh
R. B.
,
Kim
G.
,
McCole
D. F.
,
Pfeffer
K.
,
Murphy
K. M.
,
Ware
C. F.
,
Kronenberg
M.
.
2008
.
A crucial role for HVEM and BTLA in preventing intestinal inflammation.
J. Exp. Med.
205
:
1463
1476
.
21
Blackburn
S. D.
,
Shin
H.
,
Haining
W. N.
,
Zou
T.
,
Workman
C. J.
,
Polley
A.
,
Betts
M. R.
,
Freeman
G. J.
,
Vignali
D. A.
,
Wherry
E. J.
.
2009
.
Coregulation of CD8+ T cell exhaustion by multiple inhibitory receptors during chronic viral infection.
Nat. Immunol.
10
:
29
37
.
22
Doering
T. A.
,
Crawford
A.
,
Angelosanto
J. M.
,
Paley
M. A.
,
Ziegler
C. G.
,
Wherry
E. J.
.
2012
.
Network analysis reveals centrally connected genes and pathways involved in CD8+ T cell exhaustion versus memory.
Immunity
37
:
1130
1144
.
23
Zhang
Z.
,
Xu
X.
,
Lu
J.
,
Zhang
S.
,
Gu
L.
,
Fu
J.
,
Jin
L.
,
Li
H.
,
Zhao
M.
,
Zhang
J.
, et al
.
2011
.
B and T lymphocyte attenuator down-regulation by HIV-1 depends on type I interferon and contributes to T-cell hyperactivation.
J. Infect. Dis.
203
:
1668
1678
.
24
Agrawal
S.
,
Marquet
J.
,
Freeman
G. J.
,
Tawab
A.
,
Bouteiller
P. L.
,
Roth
P.
,
Bolton
W.
,
Ogg
G.
,
Boumsell
L.
,
Bensussan
A.
.
1999
.
Cutting edge: MHC class I triggering by a novel cell surface ligand costimulates proliferation of activated human T cells.
J. Immunol.
162
:
1223
1226
.
25
Nikolova
M.
,
Marie-Cardine
A.
,
Boumsell
L.
,
Bensussan
A.
.
2002
.
BY55/CD160 acts as a co-receptor in TCR signal transduction of a human circulating cytotoxic effector T lymphocyte subset lacking CD28 expression.
Int. Immunol.
14
:
445
451
.
26
Le Bouteiller
P.
,
Barakonyi
A.
,
Giustiniani
J.
,
Lenfant
F.
,
Marie-Cardine
A.
,
Aguerre-Girr
M.
,
Rabot
M.
,
Hilgert
I.
,
Mami-Chouaib
F.
,
Tabiasco
J.
, et al
.
2002
.
Engagement of CD160 receptor by HLA-C is a triggering mechanism used by circulating natural killer (NK) cells to mediate cytotoxicity.
Proc. Natl. Acad. Sci. USA
99
:
16963
16968
.
27
Barakonyi
A.
,
Rabot
M.
,
Marie-Cardine
A.
,
Aguerre-Girr
M.
,
Polgar
B.
,
Schiavon
V.
,
Bensussan
A.
,
Le Bouteiller
P.
.
2004
.
Cutting edge: engagement of CD160 by its HLA-C physiological ligand triggers a unique cytokine profile secretion in the cytotoxic peripheral blood NK cell subset.
J. Immunol.
173
:
5349
5354
.
28
Harrop
J. A.
,
McDonnell
P. C.
,
Brigham-Burke
M.
,
Lyn
S. D.
,
Minton
J.
,
Tan
K. B.
,
Dede
K.
,
Spampanato
J.
,
Silverman
C.
,
Hensley
P.
, et al
.
1998
.
Herpesvirus entry mediator ligand (HVEM-L), a novel ligand for HVEM/TR2, stimulates proliferation of T cells and inhibits HT29 cell growth.
J. Biol. Chem.
273
:
27548
27556
.
29
Tamada
K.
,
Shimozaki
K.
,
Chapoval
A. I.
,
Zhai
Y.
,
Su
J.
,
Chen
S.-F.
,
Hsieh
S.-L.
,
Nagata
S.
,
Ni
J.
,
Chen
L.
.
2000
.
LIGHT, a TNF-like molecule, costimulates T cell proliferation and is required for dendritic cell-mediated allogeneic T cell response.
J. Immunol.
164
:
4105
4110
.
30
Tamada
K.
,
Shimozaki
K.
,
Chapoval
A. I.
,
Zhu
G.
,
Sica
G.
,
Flies
D.
,
Boone
T.
,
Hsu
H.
,
Fu
Y.-X.
,
Nagata
S.
, et al
.
2000
.
Modulation of T-cell-mediated immunity in tumor and graft-versus-host disease models through the LIGHT co-stimulatory pathway.
Nat. Med.
6
:
283
289
.
31
Cheung
T. C.
,
Steinberg
M. W.
,
Oborne
L. M.
,
Macauley
M. G.
,
Fukuyama
S.
,
Sanjo
H.
,
D’Souza
C.
,
Norris
P. S.
,
Pfeffer
K.
,
Murphy
K. M.
, et al
.
2009
.
Unconventional ligand activation of herpesvirus entry mediator signals cell survival.
Proc. Natl. Acad. Sci. USA
106
:
6244
6249
.
32
Shui
J. W.
,
Larange
A.
,
Kim
G.
,
Vela
J. L.
,
Zahner
S.
,
Cheroutre
H.
,
Kronenberg
M.
.
2012
.
HVEM signalling at mucosal barriers provides host defence against pathogenic bacteria.
Nature
488
:
222
225
.
33
Watanabe
N.
,
Gavrieli
M.
,
Sedy
J. R.
,
Yang
J.
,
Fallarino
F.
,
Loftin
S. K.
,
Hurchla
M. A.
,
Zimmerman
N.
,
Sim
J.
,
Zang
X.
, et al
.
2003
.
BTLA is a lymphocyte inhibitory receptor with similarities to CTLA-4 and PD-1.
Nat. Immunol.
4
:
670
679
.
34
Kim
G. G.
,
Donnenberg
V. S.
,
Donnenberg
A. D.
,
Gooding
W.
,
Whiteside
T. L.
.
2007
.
A novel multiparametric flow cytometry-based cytotoxicity assay simultaneously immunophenotypes effector cells: comparisons to a 4 h 51Cr-release assay.
J. Immunol. Methods
325
:
51
66
.
35
Matzinger
P.
1991
.
The JAM test. A simple assay for DNA fragmentation and cell death.
J. Immunol. Methods
145
:
185
192
.
36
Muntasell
A.
,
Vilches
C.
,
Angulo
A.
,
López-Botet
M.
.
2013
.
Adaptive reconfiguration of the human NK-cell compartment in response to cytomegalovirus: A different perspective of the host-pathogen interaction.
Eur. J. Immunol.
43
:
1133
1141
.
37
Lopez-Vergès
S.
,
Milush
J. M.
,
Schwartz
B. S.
,
Pando
M. J.
,
Jarjoura
J.
,
York
V. A.
,
Houchins
J. P.
,
Miller
S.
,
Kang
S. M.
,
Norris
P. J.
, et al
.
2011
.
Expansion of a unique CD57⁺NKG2Chi natural killer cell subset during acute human cytomegalovirus infection.
Proc. Natl. Acad. Sci. USA
108
:
14725
14732
.
38
Foley
B.
,
Cooley
S.
,
Verneris
M. R.
,
Curtsinger
J.
,
Luo
X.
,
Waller
E. K.
,
Anasetti
C.
,
Weisdorf
D.
,
Miller
J. S.
.
2012
.
Human cytomegalovirus (CMV)-induced memory-like NKG2C(+) NK cells are transplantable and expand in vivo in response to recipient CMV antigen.
J. Immunol.
189
:
5082
5088
.
39
Le Bouteiller
P.
,
Tabiasco
J.
,
Polgar
B.
,
Kozma
N.
,
Giustiniani
J.
,
Siewiera
J.
,
Berrebi
A.
,
Aguerre-Girr
M.
,
Bensussan
A.
,
Jabrane-Ferrat
N.
.
2011
.
CD160: a unique activating NK cell receptor.
Immunol. Lett.
138
:
93
96
.
40
Giustiniani
J.
,
Bensussan
A.
,
Marie-Cardine
A.
.
2009
.
Identification and characterization of a transmembrane isoform of CD160 (CD160-TM), a unique activating receptor selectively expressed upon human NK cell activation.
J. Immunol.
182
:
63
71
.
41
Murphy
K. M.
,
Nelson
C. A.
,
Sedý
J. R.
.
2006
.
Balancing co-stimulation and inhibition with BTLA and HVEM.
Nat. Rev. Immunol.
6
:
671
681
.
42
Kojima
R.
,
Kajikawa
M.
,
Shiroishi
M.
,
Kuroki
K.
,
Maenaka
K.
.
2011
.
Molecular basis for herpesvirus entry mediator recognition by the human immune inhibitory receptor CD160 and its relationship to the cosignaling molecules BTLA and LIGHT.
J. Mol. Biol.
413
:
762
772
.
43
Lurain
N. S.
,
Kapell
K. S.
,
Huang
D. D.
,
Short
J. A.
,
Paintsil
J.
,
Winkfield
E.
,
Benedict
C. A.
,
Ware
C. F.
,
Bremer
J. W.
.
1999
.
Human cytomegalovirus UL144 open reading frame: sequence hypervariability in low-passage clinical isolates.
J. Virol.
73
:
10040
10050
.
44
Swann
J. B.
,
Hayakawa
Y.
,
Zerafa
N.
,
Sheehan
K. C.
,
Scott
B.
,
Schreiber
R. D.
,
Hertzog
P.
,
Smyth
M. J.
.
2007
.
Type I IFN contributes to NK cell homeostasis, activation, and antitumor function.
J. Immunol.
178
:
7540
7549
.
45
Martinez
J.
,
Huang
X.
,
Yang
Y.
.
2008
.
Direct action of type I IFN on NK cells is required for their activation in response to vaccinia viral infection in vivo.
J. Immunol.
180
:
1592
1597
.
46
Borrego
F.
,
Peña
J.
,
Solana
R.
.
1993
.
Regulation of CD69 expression on human natural killer cells: differential involvement of protein kinase C and protein tyrosine kinases.
Eur. J. Immunol.
23
:
1039
1043
.
47
Rabot
M.
,
El Costa
H.
,
Polgar
B.
,
Marie-Cardine
A.
,
Aguerre-Girr
M.
,
Barakonyi
A.
,
Valitutti
S.
,
Bensussan
A.
,
Le Bouteiller
P.
.
2007
.
CD160-activating NK cell effector functions depend on the phosphatidylinositol 3-kinase recruitment.
Int. Immunol.
19
:
401
409
.
48
Obenauer
J. C.
,
Cantley
L. C.
,
Yaffe
M. B.
.
2003
.
Scansite 2.0: Proteome-wide prediction of cell signaling interactions using short sequence motifs.
Nucleic Acids Res.
31
:
3635
3641
.
49
Wu
T. H.
,
Zhen
Y.
,
Zeng
C.
,
Yi
H. F.
,
Zhao
Y.
.
2007
.
B and T lymphocyte attenuator interacts with CD3zeta and inhibits tyrosine phosphorylation of TCRzeta complex during T-cell activation.
Immunol. Cell Biol.
85
:
590
595
.
50
Maeda
M.
,
Carpenito
C.
,
Russell
R. C.
,
Dasanjh
J.
,
Veinotte
L. L.
,
Ohta
H.
,
Yamamura
T.
,
Tan
R.
,
Takei
F.
.
2005
.
Murine CD160, Ig-like receptor on NK cells and NKT cells, recognizes classical and nonclassical MHC class I and regulates NK cell activation.
J. Immunol.
175
:
4426
4432
.
51
Bale
J. F.
 Jr.
,
Petheram
S. J.
,
Robertson
M.
,
Murph
J. R.
,
Demmler
G.
.
2001
.
Human cytomegalovirus a sequence and UL144 variability in strains from infected children.
J. Med. Virol.
65
:
90
96
.
52
Picone
O.
,
Costa
J. M.
,
Chaix
M. L.
,
Ville
Y.
,
Rouzioux
C.
,
Leruez-Ville
M.
.
2005
.
Human cytomegalovirus UL144 gene polymorphisms in congenital infections.
J. Clin. Microbiol.
43
:
25
29
.
53
Arav-Boger
R.
,
Battaglia
C. A.
,
Lazzarotto
T.
,
Gabrielli
L.
,
Zong
J. C.
,
Hayward
G. S.
,
Diener-West
M.
,
Landini
M. P.
.
2006
.
Cytomegalovirus (CMV)-encoded UL144 (truncated tumor necrosis factor receptor) and outcome of congenital CMV infection.
J. Infect. Dis.
194
:
464
473
.
54
Waters
A.
,
Hassan
J.
,
De Gascun
C.
,
Kissoon
G.
,
Knowles
S.
,
Molloy
E.
,
Connell
J.
,
Hall
W. W.
.
2010
.
Human cytomegalovirus UL144 is associated with viremia and infant development sequelae in congenital infection.
J. Clin. Microbiol.
48
:
3956
3962
.
55
Poole
E.
,
Groves
I.
,
MacDonald
A.
,
Pang
Y.
,
Alcami
A.
,
Sinclair
J.
.
2009
.
Identification of TRIM23 as a cofactor involved in the regulation of NF-kappaB by human cytomegalovirus.
J. Virol.
83
:
3581
3590
.
56
Poole
E.
,
Walther
A.
,
Raven
K.
,
Benedict
C. A.
,
Mason
G. M.
,
Sinclair
J.
.
2013
.
The myeloid transcription factor GATA-2 regulates the viral UL144 gene during human cytomegalovirus latency in an isolate-specific manner.
J. Virol.
87
:
4261
4271
.
57
Cheung
K. J.
,
Johnson
N. A.
,
Affleck
J. G.
,
Severson
T.
,
Steidl
C.
,
Ben-Neriah
S.
,
Schein
J.
,
Morin
R. D.
,
Moore
R.
,
Shah
S. P.
, et al
.
2010
.
Acquired TNFRSF14 mutations in follicular lymphoma are associated with worse prognosis.
Cancer Res.
70
:
9166
9174
.
58
Vesely
M. D.
,
Kershaw
M. H.
,
Schreiber
R. D.
,
Smyth
M. J.
.
2011
.
Natural innate and adaptive immunity to cancer.
Annu. Rev. Immunol.
29
:
235
271
.

The authors have no financial conflicts of interest.