NK cells have been reported to respond against EBV-infected B cells in the lytic cycle and to control the viral infection involving IFN-γ secretion. Early reports proposed a role for NK cell Ab-dependent cellular cytotoxicity (ADCC) triggered via FcγR-IIIA (CD16) in the response to EBV. In the current study, we revisited this issue, showing that serum from EBV+ individuals triggered vigorous NK cell degranulation and cytokine production (i.e., TNF-α and IFN-γ) against EBV-infected cells, enhancing NK cell activation. The effect was preferentially directed against cells in the lytic phase and was associated with surface expression of the gp350/220 envelope Ag. In contrast, binding of gp350+ particles, released by EBV-infected cells, to B cell lines or autologous primary B lymphocytes also promoted specific Ab-dependent NK cell degranulation and TNF-α production but induced minimal IFN-γ secretion. In that case, target cell damage appeared marginal compared with the effect of a control anti-CD20 Ab (rituximab) at concentrations that triggered similar NK cell activation, indicating that cell-associated gp350+ particles may divert the cytolytic machinery, impairing its direct action on the plasma membrane. These observations support that Ab-dependent NK cell activation plays an important role in the control of EBV, enhancing NK cell effector functions against infected B cells in the lytic cycle. In contrast, the data reveal that gp350+ particles bound to bystander B cells trigger Ab-dependent NK cell degranulation and TNF-α but not cytotoxicity or IFN-γ production, potentially favoring the progression of viral infection.

Epstein-Barr virus is a complex human γ herpesvirus (HHV-4) that causes a highly prevalent life-long infection. Among different cell types permissive to EBV infection, B lymphocytes constitute the main reservoir in which the virus establishes latency, and activation of the lytic phase allows its transmission through secretions (1). Uncontrolled EBV replication becomes a serious threat in immunocompromised patients (e.g., immunosuppressed transplant recipients). EBV may cause lymphoproliferative disorders, trigger hemophagocytic lympohistiocytosis, and has oncogenic potential, contributing to the development of hematopoietic (e.g., Burkitt and Hodgkin lymphomas) and epithelial (e.g., nasopharyngeal and gastric carcinomas) neoplasms (2). In contrast, EBV infection has been associated with the development of some autoimmune disorders (e.g., multiple sclerosis) (3, 4).

Conventionally, EBV infection is considered to be primarily controlled by T cells, with specific Abs contributing to viral neutralization (5, 6), and the virus displays a variety of immune-evasion mechanisms (7, 8); yet the relevance of NK cells in the response to EBV is gaining attention (913). Human NK cell functions may be triggered by an array of activating/costimulatory NK cell receptors (NKRs), under the control of different inhibitory NKRs mainly specific for HLA-I molecules (i.e., KIR and CD94/NKG2A) (14, 15). Identification of SAP deficiency as the cause of X-linked lymphoproliferative disease triggered by EBV infection drove attention on the role of the SLAM receptor family in the NK and T cell response to this pathogen (16). EBV-infected B lymphoblastoid cell lines (B-LCLs), mostly harboring the virus in latency, are known to be rather resistant to direct NK cell–mediated cytotoxicity. Induction of the lytic phase in an EBV+ lymphoma cell line (Akata) downregulated HLA-I surface expression and rendered it susceptible to NK cell cytotoxicity assessed in 51Cr-release assays, which involved NKG2D and DNAM-1 NKRs (9). Recently, expression of the BZLF1 early lytic gene was shown to sensitize EBV-infected cells to NK cell killing, whereas the BHRF1 vBcl-2 protein conferred resistance in the late lytic cycle (13). Based on experimental studies in vitro and in mice with a human immune system, circulating and tonsil CD56bright NK cells were reported to play a role in the control of EBV infection involving IFN-γ production (10, 17, 18).

Compared with the cooperative functional pattern observed for most activating NKRs, FcγR-IIIA (CD16) was reported to trigger, in a relatively autonomous manner, IgG-dependent NK cell functions, including Ab-dependent cellular cytotoxicity (ADCC) (19). CD16 was the first characterized NKR, and the importance of ADCC in the control of different viral infections is being increasingly appreciated (2023), yet few studies have addressed its role in the defense against EBV. Some pioneer reports described NK cell–mediated ADCC against EBV-infected cells, identifying as a major target Ag the late lytic cycle gp350/220 molecule, which is involved in the pathogen interaction with CD21 (CR2) expressed on B cells (2426). This initial step in the infection process was reported to be inhibited by CD21 binding of gp350+ exosomes that are released by EBV-infected cells (27). In this article, we provide evidence supporting that Ab-dependent NK cell activation plays an important role in the control of EBV, enhancing NK cell activity against infected B cells. Moreover, we show that B cells coated by gp350+ particles, released by EBV-infected cells, triggered Ab-dependent NK cell degranulation and TNF-α production dissociated from cell damage and IFN-γ production, thus potentially favoring viral immune evasion and dissemination.

Heparinized blood and serum samples were obtained from volunteer healthy individuals. Written informed consent was obtained from every donor, and the study protocol was approved by the institutional Ethics Committee (Parc de Salut Mar number 2010/3766/I). For some experiments, buffy coat samples were obtained from anonymized blood donors (Blood and Tissue Bank, Barcelona, Spain). PBMCs were separated on Ficoll-Hypaque gradient (Lymphoprep; Axis-Shield, Oslo, Norway). Serum samples were collected, heat inactivated (56°C, 30 min), and stored at −20°C. Standard clinical diagnostic tests were used to analyze serum samples for circulating IgG Abs against EBV.

AKBM cells were generated by stable transfection of the EBV+ Burkitt lymphoma cell line (Akata) with the pHEBO-BMRF1p-rCD2/GFP reporter plasmid, expressing GFP when the virus enters the lytic cycle (8). The lytic cycle was induced by incubating 2 × 106 AKBM cells per milliliter with 10 μg/ml F(ab′)2 goat anti-human IgG (Cappel, Malvern, PA) for 24 h. Raji is an EBV+ Burkitt lymphoma–derived cell line. EBV-transformed B-LCLs were generated upon PBMC incubation with the B95.8 virus strain. The use of a 721.221 HLA-I–deficient B-LCL as a target for rituximab-induced ADCC was described previously (20). All cell lines were maintained in RPMI 1640 GlutaMAX (Thermo Fisher Scientific, Waltham, MA) supplemented with penicillin (100 U/ml), streptomycin (100 μg/ml), sodium pyruvate (1 mM), and 10% FBS.

mAbs used for flow cytometry include anti-NKG2C–PE (clone 134591; R&D Systems, Minneapolis, MN), anti-CD3–PerCP-Cy5.5 (clone SK7), anti-CD56–allophycocyanin (clone CMSSB), anti-IFN-γ–PE (clone 4S.B3), anti-CD107a–FITC (clone H4A3) (BD Pharmingen, San Diego, CA), anti-CD19–PE-Cy7 (clone SJ25C1) (eBioscience, San Diego, CA), anti-KIR2D–PE (clone NKVFS1), anti-KIR2DL2/2DL3–PE (clone DX27), anti-KIR2DL1/2DS1–PE (clone 11PB6), anti-KIR3DL1/DL2–PE (clone 5.133) (Miltenyi Biotec, Bergisch Gladbach, Germany), and anti-KIR2DL4–PE (clone 181703) (R&D Systems, Minneapolis, MN). Anti–TNF-α (infliximab; REMICADE) was directly labeled with CF-Blue by Immunostep (Salamanca, Spain). Anti–HLA-I (clone W6/32), produced in our laboratory from the original hybridoma, was labeled with allophycocyanin. EBV-specific mAbs included anti-BZLF1 (clone BZ.1; Santa Cruz Biotechnology, Heidelberg, Germany) and anti-gp350/220 (clone 72A1, mouse IgG1; Millipore, Solna, Sweden). Cells were pretreated with human aggregated IgG (10 μg/ml) and subsequently labeled with specific mAbs. For indirect immunostaining, samples were incubated with unlabeled Abs, followed by PE-conjugated F(ab′)2 polyclonal rabbit anti-mouse IgG+IgM (Jackson ImmunoResearch, West Grove, PA) or PE-Cy7–conjugated F(ab′)2 polyclonal goat anti-mouse IgG (BioLegend, San Diego, CA). Mouse anti-human CD21 (clone B-E5, mouse IgG2a; Abcam, Cambridge, U.K.) was used for blocking experiments, using PE-conjugated rat anti-mouse IgG1 (clone A85-1; BD Pharmingen) for indirect detection of anti-gp350/220 mAb. For intracellular staining, cells were treated using a fixation/permeabilization kit (BD Biosciences). Samples were acquired on a FACSCalibur or LSR II flow cytometer (BD Biosciences), and data were analyzed with FlowJo software (TreeStar).

As described (28), fresh PBMC samples were incubated overnight with recombinant human IL-2 (200 U/ml) prior to NK cell isolation by negative selection using an EasySep Human NK cell enrichment kit (STEMCELL Technologies, Grenoble, France), according to the manufacturer’s recommendations. NK cell degranulation (CD107a) and cytokine production (TNF-α and IFN-γ) in response to target cells (E:T = 2.5:1) were monitored in the presence or absence of serum (3%) from EBV-seropositive (EBV+) or seronegative (EBV) individuals, after 4 h (CD107a, TNF-α) or 18 h (TNF-α, IFN-γ) of coculture using standard flow cytometry protocols, as described (20). B cells were purified from PBMCs by negative immunomagnetic selection with Human B Cell Enrichment Kit (STEMCELL Technologies). IgG from EBV+ and EBV sera was purified by affinity chromatography with protein G Sepharose (GE Healthcare, Little Chalfont, U.K.). In some experiments, rituximab (anti-CD20) was used as a positive control for NK cell–mediated ADCC against B cells. IFN-γ in cell-free supernatants was analyzed by ELISA (eBioscience).

For cytotoxicity assays, primary B cells were labeled with 0.3 μM CFSE CellTrace reagent (Thermo Fisher Scientific) prior to coculture with NK cells in the presence or absence of EBV+ sera. DAPI (Sigma-Aldrich, St Louis, MO) was added at the end of the assay, and target cell death was assessed by detecting CFSE+ DAPI+ cells. In cytotoxicity assays against AKBM cells, targets were labeled with 3 μM CellTrace Violet reagent (Thermo Fisher Scientific), and dead cells were detected at the end of the coculture by the addition of 7-aminoactinomycin D (7AAD; BD Biosciences).

For Calcein-AM release assays (29), target cells were preloaded with ultrapure-grade Calcein-AM (Anaspec, Fremont, CA) for 30 min at 37°C (4 μg/ml), washed in complete medium, and plated in triplicate with isolated NK cells in the presence or absence of serum from EBV+ donors (3%) for 2 h at 37°C in 5% CO2 in V-bottom 96-well microtiter plates (Nunc). Supernatants (75 μl) were harvested and analyzed using an Infiniti M200 Plate reader (Tecan, Männedorf, Switzerland) (excitation filter: 485; band-pass filter: 530). Supernatants from target cells that were incubated alone or treated with 3% Triton X-100 were assayed for spontaneous and maximum calcein release, respectively. Specific lysis was calculated according to the formula [(test release − spontaneous release)/(maximum release − spontaneous release)] × 100.

The EBV lytic phase was induced in AKBM cells by cross-linking surface IgG with 10 μg/ml of F(ab′)2-specific goat anti-human IgG for 24 h. Cells were stained with anti-gp350/220, followed by PE-Cy7–conjugated F(ab′)2 polyclonal goat anti-mouse IgG. Four cell fractions corresponding to gp350/220hi GFP+, gp350/220hi GFP, gp350/220dull GFP, and gp350/220 GFP were separated using a FACSAria (BD Biosciences) and harvested for functional assays.

Supernatants were harvested from 24-h cultures of AKBM cells (2 × 106 cells per milliliter) that were stimulated or not with 10 μg/ml goat anti-human IgG. The Raji B cell line or primary B lymphocytes were incubated with AKBM cell-free supernatant for 18 h at 37°C; acquisition of EBV Ags by target cells was analyzed by anti-gp350 staining. The possibility of lytic cycle induction in target cells was monitored by intracellular staining of BZLF1. For blocking experiments, Raji B cells were pretreated with 20 μg/ml anti-CD21 mAb (clone B-E5; Abcam) or anti–HLA-I mAb (clone W6/32) for 1 h at room temperature, prior to incubation with AKBM supernatant, as previously described (27). B cells treated with AKBM supernatant were used as targets in Ab-dependent NK cell activation and cytotoxicity assays. In some experiments, AKBM supernatants were passed through 0.22- and 0.1-μm filters (Merck Millipore, Billerica, MA) prior to incubation with B cells. Whole and filtered supernatant samples were digested with DNase from a High Pure RNA Isolation Kit (Roche, Basel, Switzerland) for 15 min at room temperature. DNase was inactivated by the addition of 2 mM EGTA and 30 mM EDTA, and samples were analyzed by quantitative PCR (qPCR) for EBV DNA with a RealStar EBV PCR Kit 1.2 (Altona Diagnostics, Hamburg, Germany).

Values of continuous variables were compared between groups using the Mann–Whitney U test, and between NK cell subsets with the Wilcoxon test.

A first set of experiments was designed to assess Ab-dependent NK cell activation against AKBM EBV-infected B cells, originally derived from a Burkitt lymphoma cell line (Akata), which express GFP upon lytic cycle activation (8). Under basal conditions, AKBM displayed high levels of HLA-I molecules (Fig. 1A), including small proportions of GFP+ and BZLF1+ cells (Fig. 1B), and poorly stimulated NK cell degranulation, as assessed by CD107a expression (Fig. 1C). As previously reported, induction of the lytic cycle by cross-linking the BCR with anti-IgG was associated with GFP and BZLF1 expression and promoted a partial downregulation of HLA-I (Fig. 1A, 1B). Incubation in the presence of EBV+ serum triggered vigorous NK cell degranulation against induced AKBM cells, increasing the NK cell response above the levels achieved by direct AKBM cell recognition (Fig. 1C). This was accompanied by an overall reduction in GFP+ cells (Fig. 1D) and an increase in dead 7AAD+ GFP+ cells (Fig. 1E), reflecting the elimination of AKBM cells in the lytic cycle. Serum from EBV donors did not induce NK cell activation, indicating that the response was mediated by EBV-specific Abs (Fig. 1C); this was confirmed by assessing the effect of purified IgG (Supplemental Fig. 1). Of note, EBV+ serum also promoted NK cell degranulation against noninduced AKBM cells, of which a substantial proportion displayed surface gp350 but no detectable GFP and BZLF1 expression (Fig. 1B, 1C).

FIGURE 1.

Ab-dependent NK cell activation against the EBV+ AKBM cell line. (A and B) HLA-I expression and EBV lytic cycle markers (BZLF1, gp350, GFP) were monitored by flow cytometry in AKBM cells before and 24 h after treatment with anti-IgG (α-IgG). (A) Line graph of surface HLA-I in nontreated (Nt) and anti-IgG–induced AKBM cells from a representative experiment. The shaded graph represents isotype control. (B) AKBM cells prior to and after anti-IgG treatment. Data are mean ± SEM from five independent experiments. Asterisks indicate significant differences compared with noninduced AKBM cells. (C) Anti-IgG–induced and noninduced AKBM cells were cocultured with purified NK cells in the absence (N.S) or presence of 3% serum from EBV+ [S. (+)] or EBV [S. (-)] individuals; NK cell degranulation was analyzed at 4 h by flow cytometry. Proportions of CD56dim CD107+ NK cells are shown (mean ± SEM, n = 4). (D) Proportions of GFP+ AKBM cells upon 4 h coculture with purified NK cells in the presence or absence of EBV+ sera (mean ± SEM, n = 5). (E) Proportions of 7AAD+ cells among GFP+ AKBM cells after 4 h of coculture with purified NK cells in the presence of EBV+ sera (mean ± SEM, n = 5). *p < 0.05, **p < 0.01, ****p < 0.0001, Mann–Whitney U test.

FIGURE 1.

Ab-dependent NK cell activation against the EBV+ AKBM cell line. (A and B) HLA-I expression and EBV lytic cycle markers (BZLF1, gp350, GFP) were monitored by flow cytometry in AKBM cells before and 24 h after treatment with anti-IgG (α-IgG). (A) Line graph of surface HLA-I in nontreated (Nt) and anti-IgG–induced AKBM cells from a representative experiment. The shaded graph represents isotype control. (B) AKBM cells prior to and after anti-IgG treatment. Data are mean ± SEM from five independent experiments. Asterisks indicate significant differences compared with noninduced AKBM cells. (C) Anti-IgG–induced and noninduced AKBM cells were cocultured with purified NK cells in the absence (N.S) or presence of 3% serum from EBV+ [S. (+)] or EBV [S. (-)] individuals; NK cell degranulation was analyzed at 4 h by flow cytometry. Proportions of CD56dim CD107+ NK cells are shown (mean ± SEM, n = 4). (D) Proportions of GFP+ AKBM cells upon 4 h coculture with purified NK cells in the presence or absence of EBV+ sera (mean ± SEM, n = 5). (E) Proportions of 7AAD+ cells among GFP+ AKBM cells after 4 h of coculture with purified NK cells in the presence of EBV+ sera (mean ± SEM, n = 5). *p < 0.05, **p < 0.01, ****p < 0.0001, Mann–Whitney U test.

Close modal

Additional experiments were carried out to compare the direct and Ab-mediated responses of different NK cell subsets against EBV lytically infected AKBM cells (Supplemental Fig. 2). In the absence of specific Abs, CD56bright (mostly CD16) and CD56dim KIR NK cell subsets (Supplemental Fig. 2A, 2B) degranulated in response to direct recognition of anti-IgG–induced AKBM more effectively than did CD56dim KIR+ cells (Supplemental Fig. 2C). As predicted, the response of CD56dim NK cells, which express CD16, was stimulated significantly by EBV-reactive Abs (Supplemental Fig. 2B, 2C).

The previous observations indicated that Ab-dependent NK cell activation enhanced the response not only against EBV-infected B cells in lytic cycle but also against latently infected AKBM cells displaying the gp350 viral Ag. To explore that issue in more detail, anti-IgG–induced AKBM cells were sorted based on GFP and gp350 expression levels (Fig. 2A). NK cell activation against sorted AKBM populations was assessed in the absence or presence of EBV+ serum, measuring CD107a surface expression and TNF-α and IFN-γ production (Fig. 2B–E). Direct NK cell activation was detected in response to AKBM cells in the lytic cycle expressing high surface levels of gp350 and GFP (G1), as well as against the gp350hi GFP fraction (G2), presumably corresponding to revertants missing the GFP plasmid (Fig. 2C–E). Ab-dependent NK cell activation was also predominantly observed against the G1 and G2 fractions, whereas only a marginal response was detected against GFP gp350 cells in the latent phase (G4). Of note, AKBM GFP cells expressing low levels of gp350 (G3) also triggered Ab-dependent NK cell degranulation and TNF-α secretion (Fig. 2B–E); yet, in that case, minimal IFN-γ production was detectable, as assessed by flow cytometry and ELISA (Fig. 2E, 2F). These results were in line with the reported ability of AKBM cells in the lytic cycle to trigger direct NK cell activation, as assessed by 51Cr-release assays (9). Moreover, the data confirmed the contribution of specific serum Abs to the NK cell response against EBV lytically infected cells, which was also detected to a lesser degree against gp350+ infected cells in the latent phase.

FIGURE 2.

Direct and Ab-dependent NK cell response to anti-IgG-induced AKBM cells sorted according to GFP and surface gp350 expression. (A) AKBM cells treated with anti-IgG mAb for 24 h were sorted into four fractions (G1–G4) based on gp350 and GFP expression. NK cell activation (CD107a and TNF-α at 4 h and IFN-γ at 24 h) against sorted AKBM cell subpopulations in the presence or absence of EBV+ [S. (+)] or EBV [S. (-)] serum was analyzed by flow cytometry and IFN-γ–specific ELISA. (B) Representative dot plots of the NK cell response (CD107a, TNF-α and IFN-γ) to distinct AKBM cell subpopulations in the presence of EBV+ sera. (CE) Proportions of CD107a+ cells (mean ± SEM, n = 5), TNF-α+ cells (mean ± SEM, n = 5), and IFN-γ+ CD56dim NK cells (mean ± SEM, n = 3) in the different conditions tested. (F) IFN-γ production detected by ELISA in cell-free supernatants from AKBM and NK cell cocultures in the presence of EBV+ sera (mean ± SEM, n = 2). *p < 0.05, **p < 0.01, Mann–Whitney U test.

FIGURE 2.

Direct and Ab-dependent NK cell response to anti-IgG-induced AKBM cells sorted according to GFP and surface gp350 expression. (A) AKBM cells treated with anti-IgG mAb for 24 h were sorted into four fractions (G1–G4) based on gp350 and GFP expression. NK cell activation (CD107a and TNF-α at 4 h and IFN-γ at 24 h) against sorted AKBM cell subpopulations in the presence or absence of EBV+ [S. (+)] or EBV [S. (-)] serum was analyzed by flow cytometry and IFN-γ–specific ELISA. (B) Representative dot plots of the NK cell response (CD107a, TNF-α and IFN-γ) to distinct AKBM cell subpopulations in the presence of EBV+ sera. (CE) Proportions of CD107a+ cells (mean ± SEM, n = 5), TNF-α+ cells (mean ± SEM, n = 5), and IFN-γ+ CD56dim NK cells (mean ± SEM, n = 3) in the different conditions tested. (F) IFN-γ production detected by ELISA in cell-free supernatants from AKBM and NK cell cocultures in the presence of EBV+ sera (mean ± SEM, n = 2). *p < 0.05, **p < 0.01, Mann–Whitney U test.

Close modal

The previous data suggested that AKBM cells entering the lytic cycle, either spontaneously or induced by anti-IgG treatment, release gp350+ particles capable of binding to cells in the latent phase, rendering them sensitive to Ab-dependent effector mechanisms. To directly test this hypothesis, the EBV+ Burkitt lymphoma Raji cell line, which does not express lytic cycle genes, was incubated for 18 h with AKBM supernatants harvested under basal culture conditions or after induction of the lytic cycle with anti-IgG treatment; subsequently, expression of BZLF1 and gp350 was assessed by flow cytometry. Surface gp350 was detected in Raji cells incubated with AKBM-derived supernatants, in the absence of BZLF1 expression (Fig. 3A). Most cell surface–bound gp350 Ag remained detectable upon incubation with AKBM supernatants passed through 0.22-μm-pore filters, which contained DNase-resistant EBV DNA detected by qPCR, but it was markedly reduced after passing through 0.1-μm-pore filters (data not shown). These observations supported that gp350+ particles (0.1–0.22 μm), which likely include virions and extracellular vesicles, interacted with the Raji B cell line. Cell attachment of gp350+ particles was only partially inhibited in some samples by an anti-CD21 (CR2) mAb (Fig. 3B), indirectly supporting the involvement of additional molecular interactions. Of note, adsorption of gp350+ particles to the B cell line did not significantly alter its baseline susceptibility to direct NK cell recognition, but it triggered Ab-mediated NK cell degranulation in the presence of EBV+ serum (Fig. 3C). Similar experiments were conducted analyzing degranulation and cytotoxicity against EBV+ B-LCL in the absence or presence of EBV+ serum (Fig. 4). As shown for noninduced AKBM cells, B-LCL cells displayed high surface HLA-I levels concomitant with significant proportions of gp350+ cells, exceeding the detection of BZLF1+ cells in the lytic cycle (Fig. 4A, 4B). In this experimental system, EBV+ serum triggered NK cell degranulation, yet minimal B-LCL cytotoxicity was detected in Calcein-release assays. In contrast, similar levels of rituximab-induced NK cell degranulation were associated with a clear cytotoxic effect on B-LCLs (Fig. 4C).

FIGURE 3.

Binding of gp350+ particles present in AKBM supernatants sensitize the Raji B cell line to Ab-mediated NK cell activation. Raji cells incubated with cell-free supernatant from noninduced or anti-IgG–induced AKBM cells were cocultured with purified NK cells in the absence or presence of EBV+ or EBV serum. NK cell degranulation (CD107a mobilization) was assessed by flow cytometry. (A) Detection of BZLF1 and gp350 expression on Raji cells incubated with AKBM supernatants by flow cytometry. (B) Immunostaining of surface gp350 in Raji cells incubated with anti-CD21 or anti–HLA-I mAb for 1 h prior to treatment with AKBM supernatants. (C) Proportions of CD107a+ CD56dim NK cells in the different coculture conditions (mean ± SEM, n = 11). Background in the absence of Raji cells was subtracted. *p < 0.05, ***p < 0.001, Mann–Whitney U test.

FIGURE 3.

Binding of gp350+ particles present in AKBM supernatants sensitize the Raji B cell line to Ab-mediated NK cell activation. Raji cells incubated with cell-free supernatant from noninduced or anti-IgG–induced AKBM cells were cocultured with purified NK cells in the absence or presence of EBV+ or EBV serum. NK cell degranulation (CD107a mobilization) was assessed by flow cytometry. (A) Detection of BZLF1 and gp350 expression on Raji cells incubated with AKBM supernatants by flow cytometry. (B) Immunostaining of surface gp350 in Raji cells incubated with anti-CD21 or anti–HLA-I mAb for 1 h prior to treatment with AKBM supernatants. (C) Proportions of CD107a+ CD56dim NK cells in the different coculture conditions (mean ± SEM, n = 11). Background in the absence of Raji cells was subtracted. *p < 0.05, ***p < 0.001, Mann–Whitney U test.

Close modal
FIGURE 4.

EBV-specific Ab-dependent NK cell activation against B-LCLs is uncoupled from target cell damage. (A) HLA-I expression on B-LCLs (open graph). Isotype control is represent by the shaded graph. (B) Proportions of BZLF1+ and gp350+ B-LCLs (mean ± SEM, n = 6). (C) Calcein AM–labeled B-LCLs were cocultured with purified NK cells in the absence or presence of EBV+ sera. CD56dim NK degranulation was monitored by CD107a mobilization by flow cytometry in 4-h assays (open bars). In parallel, B-LCL specific lysis was analyzed by a 2-h calcein-release assay (shaded bars). The effect of rituximab was assessed as a control (mean ± SEM, n = 3).

FIGURE 4.

EBV-specific Ab-dependent NK cell activation against B-LCLs is uncoupled from target cell damage. (A) HLA-I expression on B-LCLs (open graph). Isotype control is represent by the shaded graph. (B) Proportions of BZLF1+ and gp350+ B-LCLs (mean ± SEM, n = 6). (C) Calcein AM–labeled B-LCLs were cocultured with purified NK cells in the absence or presence of EBV+ sera. CD56dim NK degranulation was monitored by CD107a mobilization by flow cytometry in 4-h assays (open bars). In parallel, B-LCL specific lysis was analyzed by a 2-h calcein-release assay (shaded bars). The effect of rituximab was assessed as a control (mean ± SEM, n = 3).

Close modal

Based on the expression of BZLF1, a minor fraction of EBV+ B-LCL cells were in the lytic cycle, likely contributing to direct NK cell activation. Moreover, given that allogeneic B-LCLs were used in these assays, the response of NK cells could be modulated to a variable extent under the influence of inhibitory NK cell receptor–HLA-I interactions.

To circumvent these pitfalls, the Ab-dependent response of NK cells was assessed against autologous B cells pretreated with AKBM supernatants, which acquired surface gp350 but did not display BZLF1 (Fig. 5A). In these experimental conditions, gp350+ B cells did not induce direct NK cell activation, but they specifically triggered NK cell degranulation and TNF-α production in the presence of EBV+ serum (Fig. 5B, 5C). However, NK cell activation appeared clearly uncoupled from B cell damage, as evaluated by DAPI staining (Fig. 5D), compared with the cytotoxic effect of rituximab at concentrations adjusted to induce comparable degranulation. Of note, cell-bound gp350+ particles did not significantly interfere with rituximab-induced cytotoxicity (Fig. 5D). Altogether, these results supported that gp350+ particles attached to the cell surface triggered Ab-dependent NK cell activation, but they appeared to selectively divert the cytolytic machinery, exerting a barrier effect.

FIGURE 5.

Ab-mediated NK cell activation and cytotoxicity against autologous primary B cells coated by gp350+ particles. Purified primary B cells were treated with cell-free supernatant from anti-IgG–induced AKBM cells prior to incubation with purified autologous NK cells in the absence (N.S.) or presence [S. (+)] of EBV+ serum or rituximab (25 ng/ml). (A) Detection of BZLF1 and gp350 in primary B cells preincubated (open histogram) or not (gray filled histogram) with AKBM supernatants. Proportions of CD107a+ (mean ± SEM, n = 10) (B) and TNF-α+ (mean ± SEM, n = 6) (C) NK cells after 4 h of coculture with primary B cells in the absence (white bars) or the presence of EBV+ serum (gray bars) or rituximab (black bars); the background response in the absence of B cells was subtracted. (D) CFSE-labeled B cells were cultured under the indicated conditions. B cell death was analyzed by DAPI staining; basal values in the absence of NK cells were subtracted (mean ± SEM, n = 11). (E and F) TNF-α+ and IFN-γ+ NK cells detected after 18 h of coculture. A coculture of NK cells with 721.221 cells plus rituximab was performed as positive control. Data correspond to samples from a representative donor of three analyzed. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, Mann–Whitney U test.

FIGURE 5.

Ab-mediated NK cell activation and cytotoxicity against autologous primary B cells coated by gp350+ particles. Purified primary B cells were treated with cell-free supernatant from anti-IgG–induced AKBM cells prior to incubation with purified autologous NK cells in the absence (N.S.) or presence [S. (+)] of EBV+ serum or rituximab (25 ng/ml). (A) Detection of BZLF1 and gp350 in primary B cells preincubated (open histogram) or not (gray filled histogram) with AKBM supernatants. Proportions of CD107a+ (mean ± SEM, n = 10) (B) and TNF-α+ (mean ± SEM, n = 6) (C) NK cells after 4 h of coculture with primary B cells in the absence (white bars) or the presence of EBV+ serum (gray bars) or rituximab (black bars); the background response in the absence of B cells was subtracted. (D) CFSE-labeled B cells were cultured under the indicated conditions. B cell death was analyzed by DAPI staining; basal values in the absence of NK cells were subtracted (mean ± SEM, n = 11). (E and F) TNF-α+ and IFN-γ+ NK cells detected after 18 h of coculture. A coculture of NK cells with 721.221 cells plus rituximab was performed as positive control. Data correspond to samples from a representative donor of three analyzed. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, Mann–Whitney U test.

Close modal

Considering that Ab-dependent activation of IFN-γ production was only detected against AKBM fractions in lytic cycle (Fig. 2), NK cell secretion of the cytokine was also assessed in response to primary B cells bound to gp350+ vesicles in the presence of EBV+ serum. In these experiments, low levels of Ab-triggered IFN-γ production (Fig. 5E) were also observed to be uncoupled from degranulation (Fig. 5B) and TNF-α production (Fig. 5C). Of note, despite their ability to trigger NK cell cytotoxicity, rituximab-treated autologous B cells also failed to induce IFN-γ production, irrespectively of the presence of gp350+ particles (Fig. 5F). These results suggested that accessory signals required for optimally triggering Ab-dependent activation of IFN-γ production by NK cells were not provided by primary B lymphocytes, even after their interaction with gp350+ particles.

Adaptive NKG2C+ NK cells that develop in response to human CMV (HCMV) have been shown to efficiently undergo Ab-dependent activation in response to HCMV- and HSV-1–infected fibroblasts, as well as rituximab-coated B cells (20, 23). The Ab-mediated response of NKG2C+ and NKG2C CD56dim NK cell subsets, derived from HCMV+ donors, against EBV-infected cells was assessed (Fig. 6). Direct NKG2C+ NK cell degranulation against EBV+ B-LCLs appeared less efficient than that mediated by NKG2C NK cells; a similar trend was observed for Ab-triggered CD107a detection (Fig. 6A, 6B). In contrast, in the presence of EBV+ serum, greater proportions of TNF-α+ NKG2C+ NK cells were detected compared with NKG2C NK cell populations. To determine whether this pattern of response was maintained against EBV+ cells in the lytic cycle, similar experiments were conducted with anti-IgG–induced AKBM cells (Fig. 6C, 6D). In that case, Ab-dependent degranulation of NKG2C+ and NKG2C subsets was similar, whereas greater proportions of TNF-α+ NKG2C+ cells were detectable compared with TNF-α+ NKG2C NK cells (Fig. 6D). These results suggested that the magnitude of adaptive NKG2C+ NK cell differentiation and expansion in response to HCMV may indirectly influence the specific Ab-mediated NK cell response to EBV.

FIGURE 6.

Ab-dependent activation of HCMV-induced adaptive NK cells against EBV+ cell lines. NK cells were isolated from HCMV-seropositive individuals displaying >30% of NKG2C+ NKG2A NK cells. CD107a and TNF-α production in response to EBV+ B-LCL and AKBM cell lines in the presence [S. (+)] or absence (N. S) of EBV+ serum were analyzed in CD56dim NK cells, gated according to NKG2C expression. (A and B) CD107a+ and TNF-α+ CD56dim NK cells in response to B-LCL. (A) Data from a representative experiment; numbers indicate the proportions of NKG2C+ or NKG2C gated cells. (B) Results of samples from five donors (mean ± SEM). (C and D) CD107a+ and TNF-α+ CD56dim NK cells in response to anti-IgG–pretreated AKBM cells. (C) Data from a representative experiment. (D) Proportions of CD107a+ (mean ± SEM, n = 4) and of TNF-α+ cells (mean ± SEM, n = 3). *p < 0.05, **p < 0.01, Mann–Whitney U test.

FIGURE 6.

Ab-dependent activation of HCMV-induced adaptive NK cells against EBV+ cell lines. NK cells were isolated from HCMV-seropositive individuals displaying >30% of NKG2C+ NKG2A NK cells. CD107a and TNF-α production in response to EBV+ B-LCL and AKBM cell lines in the presence [S. (+)] or absence (N. S) of EBV+ serum were analyzed in CD56dim NK cells, gated according to NKG2C expression. (A and B) CD107a+ and TNF-α+ CD56dim NK cells in response to B-LCL. (A) Data from a representative experiment; numbers indicate the proportions of NKG2C+ or NKG2C gated cells. (B) Results of samples from five donors (mean ± SEM). (C and D) CD107a+ and TNF-α+ CD56dim NK cells in response to anti-IgG–pretreated AKBM cells. (C) Data from a representative experiment. (D) Proportions of CD107a+ (mean ± SEM, n = 4) and of TNF-α+ cells (mean ± SEM, n = 3). *p < 0.05, **p < 0.01, Mann–Whitney U test.

Close modal

The role of NK cells in the response to EBV infection is attracting substantial attention (913). In addition to their direct interaction with target cells, which involves an array of inhibitory, activating, and costimulatory receptors, NK cells are the main effectors of Ab-dependent effector mechanisms triggered by FcγR-IIIA (CD16). Early studies identified ADCC as a mechanism of immune response to EBV-infected cells (2426), and this issue has been re-explored in detail in this study.

Our data support that Ab-dependent activation enhances the NK cell response to EBV-infected B cells and may have an important role in the control of this herpesvirus infection following the development of the T and B cell adaptive response. In agreement with previous reports (26), the presence of the gp350 envelope Ag on the surface of infected cells was associated with the ability of serum Abs to trigger NK cell degranulation and cytokine production. Ab-dependent activation was preferentially directed against AKBM cells in the lytic cycle, as previously reported for the direct NK cell response (9). In contrast, our observations revealed that gp350+ particles (0.1–0.22 μm) released by EBV-infected cells, containing virions and likely extracellular vesicles, could attach to B cell lines and primary B cells, triggering Ab-dependent NK cell degranulation and proinflammatory cytokine production. Under these conditions, NK cell degranulation and TNF-α production triggered by EBV+ serum were comparable to those observed with adjusted concentrations of rituximab, but they induced disproportionately lower target cell killing; of note, cell-bound gp350+ particles did not impair rituximab-activated cytotoxicity. Such dissociation between Ab-mediated NK cell degranulation and target cell killing suggested that cell-bound gp350+ particles might selectively interfere with the impact of the cytotoxic machinery on the plasma membrane.

Ab-dependent IFN-γ production by NK cells was induced in response to AKBM cells in the lytic cycle but not against gp350+ particle–coated primary B cells. Uncoupling of IFN-γ secretion from degranulation and TNF-α production was also observed upon activation with rituximab-treated B cells, regardless of the presence of gp350+ particles. These data suggested that efficient IFN-γ production may require accessory/costimulatory signals that are absent in normal B cells, which may be provided by infected cells in the lytic cycle and are enhanced by HLA-I downregulation. Because IFN-γ is known to play a role in the control of EBV infection (17), its reduced production, in combination with the limited cytotoxicity triggered by gp350+ particles, may hamper the antiviral activity in this scenario while preserving the TNF-α–dependent proinflammatory response. gp350-containing exosomes purified by differential centrifugation from the supernatant of an EBV+ cell line were reported to reduce EBV infection interfering with CD21–viral interaction (27). In our study, attempts to purify gp350+ exosomes from IgG-induced AKBM supernatants were unsuccessful, because DNase-resistant viral DNA remained detectable by qPCR in the fractions obtained by differential centrifugation (27). Although gp350+ particles contain virions and likely vesicles, concurrently released by EBV-infected cells in the lytic cycle, their interaction with B cells may differ. It is plausible that, even in the absence of detectable lytic cycle activation (i.e., BZLF1 expression), internalization of gp350+ particles triggers pathogen-recognition receptors. In this regard, despite the fact that direct NK cell activation in response to gp350-coated B cells was not altered, the possibility that these effects might modulate ADCC is not ruled out. The establishment of reliable experimental conditions that allow a comparison of the influence exerted by virions and gp350+ extracellular vesicles on Ab-dependent NK cell effector functions remains a challenging issue.

Beyond the role of gp350-specific Abs, the participation of IgG specific for other EBV Ags expressed in the plasma membrane and/or present in particles deserves attention (5); in this regard, ADCC triggered by gp110-specific Abs was reported (30). In contrast, considering their poor immunogenicity, it is unlikely that latent-phase surface Ags (i.e., LMP1 and LMP2) represent operational targets for Ab-dependent NK cell activation, in line with the lack of response observed against gp350 GFP AKBM cells. Yet anti-LMP1 Abs were reported to be detected in some individuals (31); thus, IgG-dependent activation of NK cell effector mechanisms might contribute, together with specific cytolytic T cells, to control latently EBV-infected and tumor cells (e.g., nasopharyngeal carcinoma) expressing this oncogenic molecule, becoming potentially important in the context of vaccination strategies against EBV (32).

The involvement of different NK cell subsets in the defense against EBV infection depends on their functional maturation and differentiation state, as well as on their tissue distribution. Compared with the major circulating CD16+ CD56dim NK cell subset, most CD16 CD56bright NK cells have a limited cytolytic activity but efficiently produce cytokines and have been shown to control EBV replication involving IFN-γ secretion, thus playing a role in the innate response against the viral infection. Moreover, NK cells with a CD16 CD56bright phenotype located in mucosal-associated lymphoid tissues (e.g., tonsils) may control EBV reactivation (10, 17, 18). In contrast, cytotoxic CD56dim NK cells constitute the major circulating subset expressing CD16 and include KIR and KIR+ subpopulations. Our observations confirmed the direct response of CD56bright NK cells against EBV lytically infected cells, whereas Ab-dependent activation was observed in CD16+ CD56dim NK cells, which may contribute to the adaptive control of systemic viral dissemination. A subset of functionally mature CD56dim NKG2C+ KIR+ NK cells has been reported to differentiate and expand in response to HCMV infection in some individuals (33). These adaptive NK cells efficiently mediate Ab-dependent responses against HCMV- and HSV-infected cells (20, 23), overcoming the control by their inhibitory KIRs. Recently, CD2 has been shown to play an important accessory function (34, 35). Our data support that development of adaptive NK cells in response to HCMV might indirectly enhance the Ab-dependent response to EBV.

The role of Ab-dependent activation in the response to EBV infection mediated by CD16+ TCRαβ+ and TCRγδ+ cytolytic T lymphocyte subsets (36, 37), as well as of myelomonocytic cells bearing CD16 and other FcγRs, deserves attention. Activation of the EBV lytic cycle in epithelial cells is essential for transmission, and the presence of subsets of CD16+ cytolytic cells in these tissues might be particularly relevant for controlling the pathogen. In contrast, CD16 and IgG allotypes determine the affinity of the Ig–FcγR interaction and have been shown to influence Ab-dependent NK cell activation in response to HSV-1 (23). Thus, it is plausible that these immunogenetic traits might also quantitatively modulate the Ab-dependent response to EBV. In this context, the incidence of EBV-related disorders (e.g., Castleman disease) associated with genetically based CD16 dysfunction is remarkable (38, 39).

We thank Esther Menoyo for valuable collaboration in obtaining blood samples, blood donors for participation in the study, Gemma Heredia and Andrea Vera for technical support, the CRG-UPF Flow Cytometry facility staff for advice, and Laboratori de Referència de Catalunya for analysis of EBV DNA by qPCR.

This work was supported by grants from the Plan Estatal I+D Retos (SAF2013-49063-C2-1-R), the Spanish Ministry of Economy and Competitiveness (MINECO); the EU FP7-MINECO Infect-ERA Program (PCIN-2015-191-C02-01); FIS/PI14/00177 and Red Española de Esclerosis Múltiple (RD16/0015/0011) from the Instituto de Salud Carlos III and the European Regional Development Fund (FEDER). A.M. is supported by the Asociación Española Contra el Cáncer.

The online version of this article contains supplemental material.

Abbreviations used in this article:

7AAD

7-aminoactinomycin D

ADCC

Ab-dependent cell cytotoxicity

B-LCL

B lymphoblastoid cell line

HCMV

human CMV

NKR

NK cell receptor

qPCR

quantitative PCR.

1
Longnecker
,
R. M.
,
E.
Kieff
,
J. I.
Cohen
.
2013
.
Epstein-barr virus
. In
Fields Virology
, 6th Ed.
Knipe
,
D. M.
,
P. M.
Howley
, eds.
Lippincott Williams & Wilkins
,
Philadelphia
, p.
1898
1959
.
2
Taylor
,
G. S.
,
H. M.
Long
,
J. M.
Brooks
,
A. B.
Rickinson
,
A. D.
Hislop
.
2015
.
The immunology of Epstein-Barr virus-induced disease.
Annu. Rev. Immunol.
33
:
787
821
.
3
Ascherio
,
A.
,
K. L.
Munger
.
2015
.
EBV and autoimmunity.
Curr. Top. Microbiol. Immunol.
390
:
365
385
.
4
Lünemann
,
J. D.
,
C.
Münz
.
2009
.
EBV in MS: guilty by association?
Trends Immunol.
30
:
243
248
.
5
Middeldorp
,
J. M.
2015
.
Epstein-Barr virus-specific humoral immune responses in health and disease.
Curr. Top. Microbiol. Immunol.
391
:
289
323
.
6
Hislop
,
A. D.
2015
.
Early virological and immunological events in Epstein-Barr virus infection.
Curr. Opin. Virol.
15
:
75
79
.
7
Quinn
,
L. L.
,
J.
Zuo
,
R. J.
Abbott
,
C.
Shannon-Lowe
,
R. J.
Tierney
,
A. D.
Hislop
,
M.
Rowe
.
2014
.
Cooperation between Epstein-Barr virus immune evasion proteins spreads protection from CD8+ T cell recognition across all three phases of the lytic cycle.
PLoS Pathog.
10
:
e1004322
.
8
Ressing
,
M. E.
,
S. E.
Keating
,
D.
van Leeuwen
,
D.
Koppers-Lalic
,
I. Y.
Pappworth
,
E. J.
Wiertz
,
M.
Rowe
.
2005
.
Impaired transporter associated with antigen processing-dependent peptide transport during productive EBV infection.
J. Immunol.
174
:
6829
6838
.
9
Pappworth
,
I. Y.
,
E. C.
Wang
,
M.
Rowe
.
2007
.
The switch from latent to productive infection in Epstein-Barr virus-infected B cells is associated with sensitization to NK cell killing.
J. Virol.
81
:
474
482
.
10
Chijioke
,
O.
,
A.
Müller
,
R.
Feederle
,
M. H.
Barros
,
C.
Krieg
,
V.
Emmel
,
E.
Marcenaro
,
C. S.
Leung
,
O.
Antsiferova
,
V.
Landtwing
, et al
.
2013
.
Human natural killer cells prevent infectious mononucleosis features by targeting lytic Epstein-Barr virus infection. [Published erratum appears in 2013 Cell Rep. 12: 901.]
Cell Rep.
5
:
1489
1498
.
11
Azzi
,
T.
,
A.
Lünemann
,
A.
Murer
,
S.
Ueda
,
V.
Béziat
,
K. J.
Malmberg
,
G.
Staubli
,
C.
Gysin
,
C.
Berger
,
C.
Münz
, et al
.
2014
.
Role for early-differentiated natural killer cells in infectious mononucleosis.
Blood
124
:
2533
2543
.
12
Rickinson
,
A. B.
,
H. M.
Long
,
U.
Palendira
,
C.
Münz
,
A. D.
Hislop
.
2014
.
Cellular immune controls over Epstein-Barr virus infection: new lessons from the clinic and the laboratory.
Trends Immunol.
35
:
159
169
.
13
Williams
,
L. R.
,
L. L.
Quinn
,
M.
Rowe
,
J.
Zuo
.
2015
.
Induction of the lytic cycle sensitizes Epstein-Barr virus–infected B cells to NK cell killing that is counteracted by virus-mediated NK cell evasion mechanisms in the late lytic cycle.
J. Virol.
90
:
947
958
.
14
Lanier
,
L. L.
2005
.
NK cell recognition.
Annu. Rev. Immunol.
23
:
225
274
.
15
Moretta
,
L.
,
A.
Moretta
.
2004
.
Killer immunoglobulin-like receptors.
Curr. Opin. Immunol.
16
:
626
633
.
16
Parolini
,
S.
,
C.
Bottino
,
M.
Falco
,
R.
Augugliaro
,
S.
Giliani
,
R.
Franceschini
,
H. D.
Ochs
,
H.
Wolf
,
J. Y.
Bonnefoy
,
R.
Biassoni
, et al
.
2000
.
X-linked lymphoproliferative disease. 2B4 molecules displaying inhibitory rather than activating function are responsible for the inability of natural killer cells to kill Epstein-Barr virus-infected cells.
J. Exp. Med.
192
:
337
346
.
17
Strowig
,
T.
,
F.
Brilot
,
F.
Arrey
,
G.
Bougras
,
D.
Thomas
,
W. A.
Muller
,
C.
Münz
.
2008
.
Tonsilar NK cells restrict B cell transformation by the Epstein-Barr virus via IFN-gamma.
PLoS Pathog.
4
:
e27
.
18
Lünemann
,
A.
,
L. D.
Vanoaica
,
T.
Azzi
,
D.
Nadal
,
C.
Münz
.
2013
.
A distinct subpopulation of human NK cells restricts B cell transformation by EBV.
J. Immunol.
191
:
4989
4995
.
19
Bryceson
,
Y. T.
,
M. E.
March
,
H. G.
Ljunggren
,
E. O.
Long
.
2006
.
Synergy among receptors on resting NK cells for the activation of natural cytotoxicity and cytokine secretion.
Blood
107
:
159
166
.
20
Costa-García
,
M.
,
A.
Vera
,
M.
Moraru
,
C.
Vilches
,
M.
López-Botet
,
A.
Muntasell
.
2015
.
Antibody-mediated response of NKG2Cbright NK cells against human cytomegalovirus.
J. Immunol.
194
:
2715
2724
.
21
Ackerman
,
M. E.
,
A.
Mikhailova
,
E. P.
Brown
,
K. G.
Dowell
,
B. D.
Walker
,
C.
Bailey-Kellogg
,
T. J.
Suscovich
,
G.
Alter
.
2016
.
Polyfunctional HIV-specific antibody responses are associated with spontaneous HIV control.
PLoS Pathog.
12
:
e1005315
.
22
Jegaskanda
,
S.
,
C.
Luke
,
H. D.
Hickman
,
M. Y.
Sangster
,
W. F.
Wieland-Alter
,
J. M.
McBride
,
J. W.
Yewdell
,
P. F.
Wright
,
J.
Treanor
,
C. M.
Rosenberger
,
K.
Subbarao
.
2016
.
Generation and protective ability of influenza virus-specific antibody-dependent cellular cytotoxicity in humans elicited by vaccination, natural infection, and experimental challenge.
J. Infect. Dis.
214
:
945
952
.
23
Moraru
,
M.
,
L. E.
Black
,
A.
Muntasell
,
F.
Portero
,
M.
López-Botet
,
H. T.
Reyburn
,
J. P.
Pandey
,
C.
Vilches
.
2015
.
NK cell and Ig interplay in defense against herpes simplex virus type 1: epistatic interaction of CD16A and IgG1 allotypes of variable affinities modulates antibody-dependent cellular cytotoxicity and susceptibility to clinical reactivation.
J. Immunol.
195
:
1676
1684
.
24
Jondal
,
M.
1976
.
Antibody-dependent cellular cytotoxicity (ADCC) against Epstein-Barr virus-determined membrane antigens. I. Reactivity in sera from normal persons and from patients with acute infectious mononucleosis.
Clin. Exp. Immunol.
25
:
1
5
.
25
Patarroyo
,
M.
,
B.
Blazar
,
G.
Pearson
,
E.
Klein
,
G.
Klein
.
1980
.
Induction of the EBV cycle in B-lymphocyte-derived lines is accompanied by increased natural killer (NK) sensitivity and the expression of EBV-related antigen(s) detected by the ADCC reaction.
Int. J. Cancer
26
:
365
371
.
26
Khyatti
,
M.
,
P. C.
Patel
,
I.
Stefanescu
,
J.
Menezes
.
1991
.
Epstein-Barr virus (EBV) glycoprotein gp350 expressed on transfected cells resistant to natural killer cell activity serves as a target antigen for EBV-specific antibody-dependent cellular cytotoxicity.
J. Virol.
65
:
996
1001
.
27
Vallhov
,
H.
,
C.
Gutzeit
,
S. M.
Johansson
,
N.
Nagy
,
M.
Paul
,
Q.
Li
,
S.
Friend
,
T. C.
George
,
E.
Klein
,
A.
Scheynius
,
S.
Gabrielsson
.
2011
.
Exosomes containing glycoprotein 350 released by EBV-transformed B cells selectively target B cells through CD21 and block EBV infection in vitro.
J. Immunol.
186
:
73
82
.
28
Magri
,
G.
,
A.
Muntasell
,
N.
Romo
,
A.
Sáez-Borderías
,
D.
Pende
,
D. E.
Geraghty
,
H.
Hengel
,
A.
Angulo
,
A.
Moretta
,
M.
López-Botet
.
2011
.
NKp46 and DNAM-1 NK-cell receptors drive the response to human cytomegalovirus-infected myeloid dendritic cells overcoming viral immune evasion strategies.
Blood
117
:
848
856
.
29
Neri
,
S.
,
E.
Mariani
,
A.
Meneghetti
,
L.
Cattini
,
A.
Facchini
.
2001
.
Calcein-acetyoxymethyl cytotoxicity assay: standardization of a method allowing additional analyses on recovered effector cells and supernatants.
Clin. Diagn. Lab. Immunol.
8
:
1131
1135
.
30
Jilg
,
W.
,
C.
Bogedain
,
H.
Mairhofer
,
S. Y.
Gu
,
H.
Wolf
.
1994
.
The Epstein-Barr virus-encoded glycoprotein gp 110 (BALF 4) can serve as a target for antibody-dependent cell-mediated cytotoxicity (ADCC).
Virology
202
:
974
977
.
31
Xu
,
J.
,
A.
Ahmad
,
M.
D’Addario
,
L.
Knafo
,
J. F.
Jones
,
U.
Prasad
,
R.
Dolcetti
,
E.
Vaccher
,
J.
Menezes
.
2000
.
Analysis and significance of anti-latent membrane protein-1 antibodies in the sera of patients with EBV-associated diseases.
J. Immunol.
164
:
2815
2822
.
32
Cohen
,
J. I.
,
E. S.
Mocarski
,
N.
Raab-Traub
,
L.
Corey
,
G. J.
Nabel
.
2013
.
The need and challenges for development of an Epstein-Barr virus vaccine.
Vaccine
31
(
Suppl. 2
):
B194
B196
.
33
López-Botet
,
M.
,
A.
Muntasell
,
C.
Vilches
.
2014
.
The CD94/NKG2C+ NK-cell subset on the edge of innate and adaptive immunity to human cytomegalovirus infection.
Semin. Immunol.
26
:
145
151
.
34
Rölle
,
A.
,
A.
Halenius
,
E. M.
Ewen
,
A.
Cerwenka
,
H.
Hengel
,
F.
Momburg
.
2016
.
CD2-CD58 interactions are pivotal for the activation and function of adaptive natural killer cells in human cytomegalovirus infection.
Eur. J. Immunol.
46
:
2420
2425
.
35
Liu
,
L. L.
,
J.
Landskron
,
E. H.
Ask
,
M.
Enqvist
,
E.
Sohlberg
,
J. A.
Traherne
,
Q.
Hammer
,
J. P.
Goodridge
,
S.
Larsson
,
J.
Jayaraman
, et al
.
2016
.
Critical role of CD2 co-stimulation in adaptive natural killer cell responses revealed in NKG2C-deficient humans.
Cell Reports
15
:
1088
1099
.
36
Clémenceau
,
B.
,
R.
Vivien
,
M.
Berthomé
,
N.
Robillard
,
R.
Garand
,
G.
Gallot
,
S.
Vollant
,
H.
Vié
.
2008
.
Effector memory alphabeta T lymphocytes can express FcgammaRIIIa and mediate antibody-dependent cellular cytotoxicity.
J. Immunol.
180
:
5327
5334
.
37
Couzi
,
L.
,
V.
Pitard
,
J. F.
Moreau
,
P.
Merville
,
J.
Déchanet-Merville
.
2015
.
Direct and indirect effects of cytomegalovirus-induced γδ T cells after kidney transplantation.
Front. Immunol.
6
:
3
.
38
de Vries
,
E.
,
H. R.
Koene
,
J. M.
Vossen
,
J. W.
Gratama
,
A. E.
von dem Borne
,
J. L.
Waaijer
,
A.
Haraldsson
,
M.
de Haas
,
M. J.
van Tol
.
1996
.
Identification of an unusual Fc gamma receptor IIIa (CD16) on natural killer cells in a patient with recurrent infections.
Blood
88
:
3022
3027
.
39
Grier
,
J. T.
,
L. R.
Forbes
,
L.
Monaco-Shawver
,
J.
Oshinsky
,
T. P.
Atkinson
,
C.
Moody
,
R.
Pandey
,
K. S.
Campbell
,
J. S.
Orange
.
2012
.
Human immunodeficiency-causing mutation defines CD16 in spontaneous NK cell cytotoxicity.
J. Clin. Invest.
122
:
3769
3780
.

The authors have no financial conflicts of interest.

Supplementary data