Human NK cells can respond rapidly to Plasmodium falciparum-infected RBC (iRBC) to produce IFN-γ. In this study, we have examined the heterogeneity of this response among malaria-naive blood donors. Cells from all donors become partially activated (up-regulating CD69, perforin, and granzyme) upon exposure to iRBC but cells from only a subset of donors become fully activated (additionally up-regulating CD25, IFN-γ, and surface expression of lysosomal-associated membrane protein 1 (LAMP-1)). Although both CD56dim and CD56bright NK cell populations can express IFN-γ in response to iRBC, CD25 and LAMP-1 are up-regulated only by CD56dim NK cells and CD69 is up-regulated to a greater extent in this subset; by contrast, perforin and granzyme A are preferentially up-regulated by CD56bright NK cells. NK cells expressing IFN-γ in response to iRBC always coexpress CD69 and CD25 but rarely LAMP-1, suggesting that individual NK cells respond to iRBC either by IFN-γ production or cytotoxicity. Furthermore, physical contact with iRBC can, in a proportion of donors, lead to NK cell cytoskeletal reorganization suggestive of functional interactions between the cells. These observations imply that individuals may vary in their ability to mount an innate immune response to malaria infection with obvious implications for disease resistance or susceptibility.

Human malaria, specifically infection with Plasmodium falciparum, causes a spectrum of disease ranging from self-limiting infection to an acute, febrile illness which is fatal in a significant minority of cases. Current estimates indicate that there are up to 600 million clinical cases and 2 million deaths from P. falciparum infection each year (1). Variation in outcome of malaria infection between individuals is multifactorial but includes both immunological and genetic components (2, 3). Emerging data suggest that innate responses play a crucial role in controlling malaria infection (4), raising the possibility that interindividual variation in the innate immune response might influence disease susceptibility.

Rapid induction of IFN-γ is absolutely required to control malarial parasitemia in mice (5, 6, 7), and NK cells play an important role in this early response (5, 8, 9). Similarly, in humans, the ability to mount a robust IFN-γ response is associated with reduced susceptibility to infection (10, 11) and we have previously shown that human NK cells can be activated to produce IFN-γ within 24 h in response to P. falciparum-infected RBC (iRBC)4 (12), indicating that they may play a key role in the initial control of blood-stage malaria infections. In contrast, as in mice, excessive production of IFN-γ has been linked to immune-mediated pathology (reviewed in Ref.13), suggesting that unregulated activation of innate sources of IFN-γ may contribute to severe malarial morbidity. Importantly in this regard, we have shown that there is considerable heterogeneity in the individual NK cell response to P. falciparum iRBC among malaria-naive donors (12, 14). Clinical epidemiological studies are required to determine whether a strong NK cell IFN-γ response is ultimately beneficial or detrimental to the infected individual, but these marked differences in the NK response also raise fundamental questions about the way in which NK cells recognize iRBC and how this response is regulated at the cellular level.

We have previously shown that the NK cell IFN-γ response to iRBC is dependent on both cytokine-mediated (IL-12 and IL-18) and contact-mediated signals (12, 14). The requisite interactions may be direct, i.e., between iRBC and NK cells, or indirect, involving a three-way interaction between iRBC, APCs, and NK cells. To begin to unravel the cellular and molecular interactions that regulate NK cell responses to iRBC, we have conducted a systematic analysis of the NK cell response to P. falciparum iRBC among malaria-naive donors. We found evidence for variation in the degree of NK cell activation; partial activation is universal among NK cell donors whereas full activation is restricted to the subgroup of donors able to produce significant amounts of IFN-γ. Furthermore, we present evidence for direct interactions between NK cells and iRBC, as determined by laser scanning confocal microscopy.

Adult blood donors were recruited at the London School of Hygiene and Tropical Medicine through an anonymous blood donation system. All donors were malaria naive and healthy and gave fully informed consent for their blood to be used in this study. Ethical approval was given by the London School of Hygiene and Tropical Medicine Ethics Committee, application 805.

P. falciparum parasites of the strain 3D7 were grown in ORh− human erythrocytes (National Blood Service) in RPMI 1640 (Invitrogen Life Technologies) supplemented with 25 mM HEPES (Sigma-Aldrich), 28 mM sodium bicarbonate (BDH), 20 μg/L hypoxanthine (Sigma-Aldrich), and 10% normal human AB serum (National Blood Service). Cultures were gassed with 3% O2, 4% CO2, and 93% N2 and incubated at 37°C. The culture medium was changed daily and the parasitemia was determined by examination of Giemsa-stained thin blood smears. Parasite cultures were routinely shown to be free from Mycoplasma contamination by PCR (Stratagene). Mature schizonts were harvested from cultures of 5–8% parasitemia by centrifugation through a 60% Percoll gradient (Sigma-Aldrich).

Venous blood was collected into sodium heparin (10 IU/ml blood; CP Pharmaceuticals) and diluted 1/1 in RPMI 1640 (Invitrogen Life Technologies). PBMC were isolated by density centrifugation through a Histopaque 1077 gradient (Sigma-Aldrich) and washed twice in RPMI 1640. Cells were resuspended in growth medium (RPMI 1640, 5% autologous serum, 1% 100× penicillin/streptomycin (Invitrogen Life Technologies), 2 mM l-glutamine (Invitrogen Life Technologies)), transferred to 24-well flat-bottom tissue culture plates (Nunc), and incubated at 37°C in an atmosphere of 5% CO2. Schizont-infected (iRBC) or uninfected (uRBC) erythrocytes were added at a ratio of 3 RBC per mononuclear cell. A mixture of recombinant human IL-12 (100 ng/ml; PeproTech) and recombinant human IL-18 (100 ng/ml; MBI Fermentas) was used as a positive control.

Brefeldin A (10 μg/ml; Sigma-Aldrich) was added to the cells 3 h before the end of the incubation; for time course experiments, brefeldin A was not used. Cells were washed twice in FACS buffer (1× PBS, 0.1% sodium azide (Sigma-Aldrich), and 1% FBS (Invitrogen Life Technologies)) and centrifuged for 10 min at 470 × g. The cells were resuspended in FACS buffer, and fluorochrome-conjugated Abs to cell surface markers were added, in the dark, at 4°C and at concentrations previously determined by titration. The cells were incubated in the dark at 4°C for 30 min, washed twice in FACS buffer, and fixed for 15 min at room temperature in fixation buffer (1× PBS and 2% paraformaldehyde (Sigma-Aldrich)). After an additional washing step, the cells were resuspended in permeabilization buffer (1× PBS, 1% saponin, and 0.1% sodium azide (Sigma-Aldrich)) along with fluorochrome-conjugated Abs to IFN-γ, perforin, or granzyme A, incubated for 30 min in the dark at 4°C, washed in FACS buffer, and resuspended in FACS buffer. Isotype control Abs were used for each staining combination. Flow cytometry analysis was performed using a BD Biosciences FACSCalibur flow cytometer and CellQuest analysis software. The following Abs were used: anti-CD56 allophycocyanin and IgG1 allophycocyanin (Beckman Coulter); anti-IFN-γ FITC, IgG1 FITC, anti-CD3 PerCP, IgG1 PerCP, anti-granzyme A PE, IgG1 PE, anti-perforin PE, IgG2b PE, and anti-CD107a PE (all BD Biosciences); and anti-CD69 PE, anti-CD25 PE, and IgG2a PE (all Caltag Medsystems).

CD56+CD3 NK cells were enriched from PBMC by magnetic bead separation (MACS) using the NK Cell Isolation Kit II (Miltenyi Biotec) according to the manufacturer’s instructions. Cells were counted, tested for viability using trypan blue, and purity confirmed by flow cytometry. Briefly, 106 NK cells and 3 × 106 iRBC were resuspended in prewarmed cell growth medium and transferred to a 96-well V-bottom plate. The cells were incubated at 37°C in 5% CO2 for 15 min and fixed for 30 min at room temperature with 1.5% electron microscopy grade formaldehyde (Sigma-Aldrich). The cells were spun for 10 min (400 × g) and washed once with washing buffer. After 1 h of incubation in blocking solution (Dulbecco’s PBS, 5% horse serum (Sigma-Aldrich), and 3% FBS) and washing, the cells were incubated with phalloidin conjugated to Alexa Fluor 546 (1/40 in blocking solution; Molecular Probes) for 1 h at 4°C. After two washes, the cells were incubated for 10 min at 4°C with 0.1 mg/ml 4′,6′-diaminidino-2-phenylindole (DAPI), washed once, and placed between a microscope slide and a 24 × 24-mm coverslip (thickness no. 1).

Confocal microscopic examination of erythrocyte/NK cell conjugates was performed under a 100× Plan Apochromat oil-immersion objective (numerical aperture, 1.4) with a Zeiss LSM 510/Axiovert 200 M inverted microscope (Zeiss) using excitation wavelengths of 405 and 543 nm. Cell-cell contacts were scanned in the xy direction every 0.4 μm through the z-plane. Maximum intensity projection-based three-dimensional reconstructions of the cell-cell contact were performed using the LSM Image Examiner software (Zeiss).

All statistical analyses were performed using Prism 4 software (GraphPad Software).

As we have previously reported (12, 14), the human NK cell IFN-γ response to iRBC is variable, with a significant proportion of the peripheral NK cells from some donors consistently producing large amounts of IFN-γ while NK cells from other donors make no appreciable IFN-γ response. Examples of the NK cell IFN-γ response to iRBC for four individual donors used in this current study are shown (Fig. 1,A). IFN-γ production from CD56bright and CD56dim cells appears to be differentially regulated. For donor 181F, neither the CD56bright nor the CD56dim NK cells produced IFN-γ, whereas for donor 178M, ∼30% of each cell population stained positively for IFN-γ after 24 h. By contrast, for donor 056, the IFN-γ+ cells were predominantly among the CD56dim subset while for donor EX03, the very few IFN-γ+ cells seen were CD56bright. Taking together, the data for 28 donors, cells from 11 donors clearly up-regulated IFN-γ in response to iRBC (>5% of NK cells being IFN-γ+) (Fig. 1,B). Importantly, the phenotype of each donor (percentage of NK cells producing IFN-γ in response to iRBC) was highly reproducible over time (Fig. 1 C).

FIGURE 1.

The NK cell IFN-γ response to P. falciparum-infected erythrocytes is heterogeneous between donors but stable over time. PBMC from malaria-naive donors were stimulated with live P. falciparum iRBC or uRBC for 24 h. The production of IFN-γ by NK cells was analyzed by intracellular staining and flow cytometric analysis. FACS plots are gated on CD3 lymphocytes. Percentages indicate the proportion of CD3CD56dim or CD3CD56high cells that were positive for IFN-γ; data are based on the collection of 50,000 total events. A, Representative examples of the IFN-γ response from four donors. B, The percentage of IFN-γ+ NK cells after 24 h of coculture with uRBC or iRBC for all NK cells (upper panel), CD56bright (middle panel), and CD56dim NK cell subsets (lower panel) is shown for 37 donors. C, NK cell IFN-γ responses to uRBC and iRBC for four individual donors tested on four different days, several weeks apart. GM, growth medium (i.e., no stimulation).

FIGURE 1.

The NK cell IFN-γ response to P. falciparum-infected erythrocytes is heterogeneous between donors but stable over time. PBMC from malaria-naive donors were stimulated with live P. falciparum iRBC or uRBC for 24 h. The production of IFN-γ by NK cells was analyzed by intracellular staining and flow cytometric analysis. FACS plots are gated on CD3 lymphocytes. Percentages indicate the proportion of CD3CD56dim or CD3CD56high cells that were positive for IFN-γ; data are based on the collection of 50,000 total events. A, Representative examples of the IFN-γ response from four donors. B, The percentage of IFN-γ+ NK cells after 24 h of coculture with uRBC or iRBC for all NK cells (upper panel), CD56bright (middle panel), and CD56dim NK cell subsets (lower panel) is shown for 37 donors. C, NK cell IFN-γ responses to uRBC and iRBC for four individual donors tested on four different days, several weeks apart. GM, growth medium (i.e., no stimulation).

Close modal

Since both CD56dim and CD56bright cells appeared to be able to respond to iRBC by making IFN-γ, but previous studies had suggested that CD56dim and CD56bright cells varied in their ability to make this cytokine, we compared the IFN-γ response among CD3CD56bright and CD3CD56dim NK cells within individual donors. In accordance with previous studies (15), a significantly higher proportion of CD56bright cells than CD56dim cells made IFN-γ in response to stimulation with high doses of IL-12 and IL-18 (mean 63.6 vs 42.9%, t = 6.90, df = 27, p < 0.0001). By contrast, there was no significant difference in the proportion of CD56bright and CD56dim cells responding to iRBC (Fig. 1 B) but, since CD56dim cells outnumber CD56bright cells by at least 15:1 in peripheral blood, significant numbers of IFN-γ+ NK cells were seen only in donors in whom the CD56dim population was able to respond.

To determine whether or not NK cells from IFN-γ low responders were completely nonresponsive to iRBC, we examined expression of the early activation marker CD69 after coculture with iRBC. The C-type lectin-like glycoprotein CD69 is known to be a sensitive and very early marker of leukocyte activation (16, 17). It is encoded within the NK gene complex (18), is rapidly expressed on the cell surface from preformed intracellular stores in a RNA and peptide synthesis-independent manner (19), and its cross-linking by Ab triggers exocytosis of cytotoxic granules (20).

Representative examples of changes in CD69 expression on NK cells after 24 h of coculture with iRBC, uRBC, or IL-12/18 for one IFN-γ low-responding donor (Fig. 2,A) and one high-responding donor (Fig. 2,B) are shown. Following incubation of PBMC for 24 h in growth medium alone, the proportion of NK cells expressing CD69 varied between 10 and 50%, presumably reflecting differing levels of ex vivo activation (Fig. 2,C). Stimulation with high doses of IL-12 and IL-18 induced CD69 expression on virtually all NK cells (examples are shown in Fig. 2, A and B). CD69 expression levels did not increase after incubation with uninfected erythrocytes; indeed taking the data from all 30 donors tested the percentage of CD69+ NK cells showed a tendency to be lower in uRBC cocultures than among NK cells cultured in growth medium alone (paired t test, t = 1.964, p = 0.059).

FIGURE 2.

The early activation marker CD69 is universally up-regulated on NK cells in response to live intact P. falciparum-infected erythrocytes. PBMC from IFN-γ low-responding and IFN-γ high-responding donors were stimulated with IL-12 and IL-18 (IL-12/18), uRBC, or live P. falciparum iRBC for 24 h. (Donors were designated as high responders if >5% of their NK cells expressed IFN-γ after 24-h PBMC incubation with iRBC.) CD69 expression on NK cells was analyzed by surface staining and flow cytometric analysis. FACS plots are gated on CD3 lymphocytes. Percentages indicate the proportion of CD3CD56dim or CD3CD56high cells that were positive for CD69; data are based on the collection of 50,000 total events. Representative plots for one IFN-γ low responder (A) and one IFN-γ high responder (B) are shown. C, The percentage of NK cells expressing CD69 in response to growth medium (GM) alone, uRBC, or iRBC was calculated for 30 donors. Each dot represents a single donor. CD69 expression is shown for all CD3CD56+ NK cells (left panel), CD3CD56dim cells (middle panel), and CD3CD56bright cells (right panel). D, Comparison of CD69 expression between IFN-γ high-responding and low-responding donors in response to iRBC. The percentage of NK cells expressing CD69 (left panel) and the fold increase in CD69 MFI (right panel) are shown. E, Comparison of CD69 expression in CD56dim and CD56bright NK cells shown as percentage CD69+ (left panel) and fold increase in CD69 MFI (right panel). GM, growth medium (i.e., no stimulation). ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001. F, Correlation between CD69 up-regulation (MFI) and IFN-γ response in CD56dim NK cells from individual donors.

FIGURE 2.

The early activation marker CD69 is universally up-regulated on NK cells in response to live intact P. falciparum-infected erythrocytes. PBMC from IFN-γ low-responding and IFN-γ high-responding donors were stimulated with IL-12 and IL-18 (IL-12/18), uRBC, or live P. falciparum iRBC for 24 h. (Donors were designated as high responders if >5% of their NK cells expressed IFN-γ after 24-h PBMC incubation with iRBC.) CD69 expression on NK cells was analyzed by surface staining and flow cytometric analysis. FACS plots are gated on CD3 lymphocytes. Percentages indicate the proportion of CD3CD56dim or CD3CD56high cells that were positive for CD69; data are based on the collection of 50,000 total events. Representative plots for one IFN-γ low responder (A) and one IFN-γ high responder (B) are shown. C, The percentage of NK cells expressing CD69 in response to growth medium (GM) alone, uRBC, or iRBC was calculated for 30 donors. Each dot represents a single donor. CD69 expression is shown for all CD3CD56+ NK cells (left panel), CD3CD56dim cells (middle panel), and CD3CD56bright cells (right panel). D, Comparison of CD69 expression between IFN-γ high-responding and low-responding donors in response to iRBC. The percentage of NK cells expressing CD69 (left panel) and the fold increase in CD69 MFI (right panel) are shown. E, Comparison of CD69 expression in CD56dim and CD56bright NK cells shown as percentage CD69+ (left panel) and fold increase in CD69 MFI (right panel). GM, growth medium (i.e., no stimulation). ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001. F, Correlation between CD69 up-regulation (MFI) and IFN-γ response in CD56dim NK cells from individual donors.

Close modal

CD69 was up-regulated in NK cells after incubation with iRBC. Interestingly, CD69 up-regulation was universal, being expressed on >75% of all NK cells in all donors tested (Fig. 2,C, left panel). However, CD69 was up-regulated to a significantly greater extent in CD56dim cells than in CD56bright cells, both in terms of the percentage of CD69+ NK cells and the fold increase in median fluorescence intensity (MFI) (Fig. 2, C and E). Importantly, there were also significant differences in the degree of CD69 up-regulation between donors. Both the percentage of NK cells expressing CD69 and the fold increase in MFI for CD69 expression were significantly higher in individuals classified as IFN-γ high responders (>5% of NK cells able to make IFN-γ after coculture with iRBC) than in low responders (Fig. 2,D). This higher expression of CD69 in high IFN-γ responders than in low IFN-γ responders was seen in both the CD56dim and CD56bright NK cell subsets (Fig. 2,E). Consequently, the greatest degree of up-regulation of CD69 expression was seen in the CD56dim NK cells of IFN-γ high responders (Fig. 2 F).

The observation that cells from all donors up-regulate CD69 to a certain extent, but that maximal CD69 expression occurs only in donors whose cells go on to produce IFN-γ, suggests that of the two signals required for NK activation by iRBC (21), one signal may be universal but the other may be restricted to the subgroup of IFN-γ-producing donors.

To explore further these differences in NK responses between donors and within the CD56dim and CD56bright NK populations within donors, we examined the expression of an alternative marker of activation, the high-affinity IL-2Rα/CD25, which reflects the proliferative capacity of NK cells (22). PBMC were incubated for 24 h with iRBC, uRBC, and IL-12/18 or no stimulation and the levels of CD25 were assessed by flow cytometry; representative examples of responses to iRBC in an IFN-γ low (Fig. 3,A)- and high (Fig. 3,B)-responding donor are shown. For resting NK cells, cultured in growth medium alone, the proportion of CD56bright cells expressing CD25 was significantly higher than the proportion of CD25+CD56dim cells (mean = 19 and 6%, respectively, n = 10, p = 0.01; Fig. 3 C). After stimulation with IL-12/18, the proportion of NK cells expressing CD25 reached >95% in all donors tested while no up-regulation was observed in response to uRBC (data not shown).

FIGURE 3.

Expression of CD25 on NK cells in response to live P. falciparum-infected erythrocytes. PBMC were cultured with or without iRBC for 24 h and expression of CD25 was assessed by surface staining and flow cytometric analysis. Representative plots for an IFN-γ low-responding (A) and an IFN-γ high-responding donor (B) are shown. (Donors were designated as high responders if >5% of their NK cells expressed IFN-γ after 24-h PBMC incubation with iRBC.) FACS plots are gated on CD3 lymphocytes. Percentages indicate the proportion of CD3CD56dim or CD3CD56high cells that were positive for CD25; data are based on the collection of 50,000 total events. C, Expression of CD25 on the CD56dim and CD56bright NK cell subsets for 10 donors after 24-h incubation with growth medium (GM; i.e., no stimulation) or iRBC. Each dot represents a single donor. D, Correlation between the increase in CD25 expression on NK cells with the proportion of IFN-γ+ NK cells for 10 donors after incubation with iRBC. ∗, p < 0.05; ∗∗, p < 0.01.

FIGURE 3.

Expression of CD25 on NK cells in response to live P. falciparum-infected erythrocytes. PBMC were cultured with or without iRBC for 24 h and expression of CD25 was assessed by surface staining and flow cytometric analysis. Representative plots for an IFN-γ low-responding (A) and an IFN-γ high-responding donor (B) are shown. (Donors were designated as high responders if >5% of their NK cells expressed IFN-γ after 24-h PBMC incubation with iRBC.) FACS plots are gated on CD3 lymphocytes. Percentages indicate the proportion of CD3CD56dim or CD3CD56high cells that were positive for CD25; data are based on the collection of 50,000 total events. C, Expression of CD25 on the CD56dim and CD56bright NK cell subsets for 10 donors after 24-h incubation with growth medium (GM; i.e., no stimulation) or iRBC. Each dot represents a single donor. D, Correlation between the increase in CD25 expression on NK cells with the proportion of IFN-γ+ NK cells for 10 donors after incubation with iRBC. ∗, p < 0.05; ∗∗, p < 0.01.

Close modal

Expression of CD25 was up-regulated in response to iRBC in NK cells from all donors; however, the degree of up-regulation differed between donors and significant up-regulation of CD25 expression was restricted to the CD56dim population with no significant change in expression of CD25 in the CD56bright population (Fig. 3,C). Interestingly, the degree of up-regulation of CD25 was significantly correlated with the ability to produce IFN-γ (r2 = 0.86, n = 10, p = 0.0001; Fig. 3 D).

Thus, the heterogeneity of the NK response to iRBC can be seen at the level of IFN-γ production and expression of both CD69 and CD25, and induction of these markers of NK activation appears to be linked.

Since it has previously been reported that human NK cells can be cytotoxic for iRBC (23) and since CD69, which has been linked to the cytotoxic potential of NK cells (22), was heterogeneously expressed, we wanted to determine whether expression of other markers of cytotoxic activity were also differentially expressed among donors. Two major components of NK cell cytotoxic granules are the membrane-disrupting protein perforin and the serine protease granzyme A, which mediates caspase-independent apoptosis (24). A more recently described marker of cytotoxic activity is the lysosomal-associated membrane protein 1 (LAMP-1 or CD107a) which is expressed on the NK cell surface following cytotoxic granule exocytosis induced by exposure to MHC-deficient target cells (25).

Thus, PBMC were incubated for 24 h with iRBC, uRBC, and IL-12/18 or no stimulation, and NK cell expression of LAMP-1 or perforin and granzyme A was assessed by surface and intracellular staining, respectively (Fig. 4).

FIGURE 4.

Expression of perforin, granzyme A, and LAMP-1 on NK cells in response to P. falciparum iRBC. PBMC were cultured with or without iRBC for 24 h and expression of perforin (A), granzyme A (B), and LAMP-1 (C, CD107a) was assessed by intracellular staining followed by flow cytometric analysis. All plots are gated on CD3 lymphocytes; data are based on the collection of 50,000 total events. Numbers indicate the MFI of staining of the CD56bright and CD56dim subsets of NK cells, respectively. Data from a representative experiment is shown in the first three panels. A and B, Right-hand panels show the fold increase in NK cell MFI for perforin and granzyme A or the CD56bright and CD56dim subsets of NK cells from eight donors. Donors are ranked in order of increasing IFN-γ response from low (left) to high (right). C, Right-hand panel, correlation between the fold increase in LAMP-1 expression on NK cells and percentage of IFN-γ+ NK cells, following incubation with iRBC as compared with incubation with no stimulation, for 13 donors.

FIGURE 4.

Expression of perforin, granzyme A, and LAMP-1 on NK cells in response to P. falciparum iRBC. PBMC were cultured with or without iRBC for 24 h and expression of perforin (A), granzyme A (B), and LAMP-1 (C, CD107a) was assessed by intracellular staining followed by flow cytometric analysis. All plots are gated on CD3 lymphocytes; data are based on the collection of 50,000 total events. Numbers indicate the MFI of staining of the CD56bright and CD56dim subsets of NK cells, respectively. Data from a representative experiment is shown in the first three panels. A and B, Right-hand panels show the fold increase in NK cell MFI for perforin and granzyme A or the CD56bright and CD56dim subsets of NK cells from eight donors. Donors are ranked in order of increasing IFN-γ response from low (left) to high (right). C, Right-hand panel, correlation between the fold increase in LAMP-1 expression on NK cells and percentage of IFN-γ+ NK cells, following incubation with iRBC as compared with incubation with no stimulation, for 13 donors.

Close modal

In contrast to a previous report (26), freshly isolated NK were not found to be positive for perforin (Fig. 4,A) and granzyme A was expressed only at low levels (Fig. 4,B). However, in agreement with the previous report (26), both perforin and granzyme A were spontaneously up-regulated during in vitro culture; after 24 h in culture in the absence of any exogenous stimuli, close to 100% of NK cells stained positively for both granzyme and perforin and both markers were especially highly expressed in CD56dim cells (Fig. 4, A and B). Interestingly, this spontaneous up-regulation of perforin was slightly, but statistically significantly, inhibited in both CD56bright and CD56dim NK cells cultured with uRBC (paired t test, MFI for uRBC vs growth medium alone, t >2.43, df = 7, p < 0.05 in both cases).

After coculture with iRBC for 24 h, the levels of NK expression of both perforin and granzyme (expressed as fold increase in MFI) were modestly increased when compared with PBMC cultured alone; this increase was particularly pronounced for the CD56bright population (Fig. 4, A and B, right panels). The degree of up-regulation of intracellular perforin was similar in all donors tested while up-regulation of granzyme A was more variable between donors. However, variation in granzyme A up-regulation was not correlated with IFN-γ responses in either CD56bright or CD56dim cells (r2 < 0.25, p > 0.2 in each case).

By contrast, there was no spontaneous increase in surface expression of LAMP-1 during in vitro culture (Fig. 4,C) and, accordingly, incubation with uRBC did not affect levels of LAMP-1 expression (data not shown). LAMP-1 expression was not significantly enhanced by stimulation with IL-12/18 (data not shown). Surface expression of LAMP-1 was however up-regulated in response to iRBC in NK cells from some donors and, where LAMP-1 up-regulation was observed, it was confined exclusively to the CD56dim subset of NK cells (Fig. 4 C). Moreover, the degree of up-regulation of LAMP-1 was highly correlated with the frequency of IFN-γ-producing cells (r2 = 0.90, n = 13, p < 0.0001).

The data presented so far indicate that some markers of NK activation (notably CD69, perforin, and granzyme A) are up-regulated in cells from all donors and in both CD56bright and CD56dim NK subsets, whereas other markers (IFN-γ, CD25, and LAMP-1) are up-regulated only in a subset of donors and, in the case of CD25 and LAMP, only in CD56dim cells. To determine whether up-regulation of IFN-γ occurs in the same cells that express CD25, LAMP-1, and very high levels of CD69, cells were incubated with iRBC for 24 h and stained for CD3, CD56, intracellular IFN-γ, and either CD69, CD25, or LAMP-1. Representative plots are shown in Fig. 5.

FIGURE 5.

Coexpression of CD69 and CD25 but not LAMP-1 with IFN-γ in iRBC-stimulated NK cells. Four-color FACS staining was conducted to allow comparison of the expression of CD69, CD25, or LAMP-1 with IFN-γ in NK cells following 24 h of PBMC culture with iRBC. All plots are gated on CD3CD56+ NK cells. Cells were costained for intracellular IFN-γ and surface expression of either CD69 (A), CD25 (B), or LAMP-1 (C). One representative experiment of 33 (CD69) or 4 (CD25 and LAMP-1) is shown.

FIGURE 5.

Coexpression of CD69 and CD25 but not LAMP-1 with IFN-γ in iRBC-stimulated NK cells. Four-color FACS staining was conducted to allow comparison of the expression of CD69, CD25, or LAMP-1 with IFN-γ in NK cells following 24 h of PBMC culture with iRBC. All plots are gated on CD3CD56+ NK cells. Cells were costained for intracellular IFN-γ and surface expression of either CD69 (A), CD25 (B), or LAMP-1 (C). One representative experiment of 33 (CD69) or 4 (CD25 and LAMP-1) is shown.

Close modal

IFN-γ+ cells were all CD69high and the CD69 MFI of the IFN-γ+ cells was higher than the CD69 MFI of the IFN-γ cells (Fig. 5 A). The vast majority of IFN-γ+ were also CD25+ and the MFI was higher for IFN-γ+ cells than for IFN-γ cells; however, not all CD25+ cells expressed IFN-γ, at least at the 24-h time point examined. By contrast, although a very few cells were found to be double positive, IFN-γ and LAMP-1 tended to be expressed in a mutually exclusive manner.

To determine the kinetics of NK cell activation by iRBC and to check that data obtained after 24 h of coculture are a true representation of the potential responsiveness of individual donors, 24-h time-course experiments were conducted with cells from both IFN-γ high and low responders. Replicate wells containing PBMC plus iRBC were cultured for up to 24 h and, at each time point, cells were collected and stained for CD3, CD56, and either IFN-γ, CD69, LAMP-1, or CD25 (Fig. 6). As we had found previously (12), IFN-γ is not detectable in iRBC-activated NK cells at 6 h but increases rapidly to maximal levels between 12 and 24 h (Fig. 6,A). The kinetics of CD25 expression were indistinguishable from those of IFN-γ (Fig. 6,B), suggesting that these two markers are indeed coinduced. By contrast, CD69 up-regulation begins as early as 2 h after initiation of the coculture, reaches 50% of the maximal level within 3 h, and is maximal at 6–12 h (Fig. 6,C). LAMP-1 up-regulation shows an intermediate profile, being significantly above baseline at 6 h but reaching maximal levels between 12 and 24 h (Fig. 6 D).

FIGURE 6.

Kinetics of NK cell activation in response to iRBC. The expression of IFN-γ (A), CD25 (B), CD69 (C), and LAMP-1 (D) in response to iRBC was assessed in replicate wells of PBMC from three or four individual donors at 0, 1, 2, 3, 6, 12, 18, or 24 h. The proportion of NK cells staining for each marker was calculated as described previously. Closed circles indicate IFN-γ high-responding donors, and open circles IFN-γ low-responding donors. Donors were designated as high responders if >5% of their NK cells expressed IFN-γ after 24-h PBMC incubation with iRBC.

FIGURE 6.

Kinetics of NK cell activation in response to iRBC. The expression of IFN-γ (A), CD25 (B), CD69 (C), and LAMP-1 (D) in response to iRBC was assessed in replicate wells of PBMC from three or four individual donors at 0, 1, 2, 3, 6, 12, 18, or 24 h. The proportion of NK cells staining for each marker was calculated as described previously. Closed circles indicate IFN-γ high-responding donors, and open circles IFN-γ low-responding donors. Donors were designated as high responders if >5% of their NK cells expressed IFN-γ after 24-h PBMC incubation with iRBC.

Close modal

We have previously shown that direct contact between NK cells and iRBC and/or accessory cells present in the PBMC population is essential for optimal induction of IFN-γ production (14). We have also shown that, during coculture of iRBC and PBMC, a significant fraction of CD56+CD3 NK cells acquire surface expression of the erythrocyte-specific marker glycophorin A (14), indicating either the formation of stable conjugates between NK cells and iRBC or transfer of iRBC membrane components to NK cells.

Functional interaction (either activating or inhibitory) between NK cells and their targets requires the formation of an immunological synapse (IS); one of the first events in the assembly of the IS is rearrangement of the cytoskeleton to facilitate clustering of lipid rafts, surface receptors, and signaling molecules (reviewed in Ref.27). Thus, to determine whether NK cells make direct functional contacts with iRBC, NK cells and iRBC were cocultured for 15 min, fixed, stained with DAPI (DNA) and fluorochrome-conjugated phalloidin (F-actin), and visualized by laser scanning confocal microscopy to look for alterations in the NK cell cytoskeleton that might be indicative of IS formation.

Cells from eight donors were examined. Representative images are shown of NK cells from two donors (Fig. 7, A and B). We observed that NK cells from all donors tested could form conjugates with iRBC and that iRBC:NK cell conjugate formation was a very frequent event, with 40–50% of NK cells making stable contacts with iRBC at any one time (Fig. 7,A). However, clustering of F-actin at the contact site (which is indicative of cytoskeletal reorganization) was observed in only four donors, three IFN-γ high responders and one low responder (compare Fig. 7,A, no clustering, with the example of F-actin clustering shown in Fig. 7,B), and, in those donors, clustering of F-actin was seen in only 10–20% of contacts. Clustering of F-actin was confirmed by scanning the contact sites in the xy direction throughout the z-plane (Fig. 7,C) and by projected three-dimensional reconstruction (Fig. 7 D).

FIGURE 7.

NK cells form stable contacts with P. falciparum-infected erythrocytes. A total of 3 × 106 iRBC and 106 MACS-purified NK cells from an IFN-γ low-responding (A) and an IFN-γ high-responding donor (B) were cocultured for 15 min at 37°C, fixed at room temperature, and stained with phalloidin conjugated to the fluorochrome Alexa Fluor 546 as a probe for F-actin (green); nuclei were stained with DAPI (blue). The cells were visualized by laser scanning confocal microscopy. The transmitted light image is shown in the left panel and the fluorescence image at 405 and 543 nm is shown in the right panel. Scale bars in A and B represent 5 μm. C, Representative scans through the z-plane of the cell-cell contact shown in B. F-actin fluorescence intensity is false color-coded, the highest intensity being cyan. D, Three-dimensional reconstruction of the cell-cell contact shown in B using a maximum intensity projection based on scans through the z-plane every 0.4 μm. F-actin fluorescence intensity is false color-coded and nuclei were stained with DAPI (blue). DIC, differential interference contrast.

FIGURE 7.

NK cells form stable contacts with P. falciparum-infected erythrocytes. A total of 3 × 106 iRBC and 106 MACS-purified NK cells from an IFN-γ low-responding (A) and an IFN-γ high-responding donor (B) were cocultured for 15 min at 37°C, fixed at room temperature, and stained with phalloidin conjugated to the fluorochrome Alexa Fluor 546 as a probe for F-actin (green); nuclei were stained with DAPI (blue). The cells were visualized by laser scanning confocal microscopy. The transmitted light image is shown in the left panel and the fluorescence image at 405 and 543 nm is shown in the right panel. Scale bars in A and B represent 5 μm. C, Representative scans through the z-plane of the cell-cell contact shown in B. F-actin fluorescence intensity is false color-coded, the highest intensity being cyan. D, Three-dimensional reconstruction of the cell-cell contact shown in B using a maximum intensity projection based on scans through the z-plane every 0.4 μm. F-actin fluorescence intensity is false color-coded and nuclei were stained with DAPI (blue). DIC, differential interference contrast.

Close modal

The data presented here firmly establish that P. falciparum-infected erythrocytes can induce IFN-γ production in human NK cells from malaria-naive donors but, while being highly reproducible within donors, the proportion of peripheral NK cells that are able to respond in this way differs between donors. Among IFN-γ low responders, NK cell activation appears to be only partial and is characterized by moderate up-regulation of CD69, perforin, and granzyme in both the CD56bright and CD56dim NK subsets. By contrast, among high responders, IFN-γ production is accompanied by maximal up-regulation of CD69 and selective up-regulation of CD25 and LAMP-1 in the CD56dim NK cell subset. That the response to iRBC can be such a stable phenotype within an individual suggests that it is hardwired and possibly genetically regulated; that the response varies so greatly within a population suggests that the genes involved may be polymorphic. Furthermore, the fact that only a subset of the NK repertoire is able to respond to iRBC even in high responding donors suggests that the relevant gene (or genes) may be variably expressed among different NK cell clones.

Our previous studies have shown that iRBC-mediated induction of IFN-γ production by NK cells is dependent on both accessory cell-dependent, cytokine-mediated signals and a signal mediated by direct contact between NK cells and iRBC and/or accessory cells within PBMC (12, 14), but the exact nature of the interactions between these three cell types remained unclear (21). We have shown that IL-12 and IL-18 are required for optimal NK activation by iRBC ((12, 14); preliminary data suggest that IFN-α, but not IL-15, is also required and that TGF-β can down-modulate the NK IFN-γ response but, to date, we have no evidence for systematic variation in these cytokine-mediated effects between IFN-γ high-responder or low-responder individuals (K. C. Newman, D. S. Korbel, K. Cowgill, and E. M. Riley, unpublished data). Experiments are ongoing to determine the primary source of these cytokines and to identify the key accessory cells required for optimal NK activation by iRBC.

Our direct imaging data demonstrate that stable contacts occur between iRBC and NK cells and suggest that NK activation may be related to cytoskeletal reorganization at the contact sites, as has been reported for activating IS formation at the contact site between NK cells and EBV-transformed B cells (28). Stable contacts between NK cells and iRBC were abundantly seen in cultures from both IFN-γ high- and low-responding donors, suggesting that NK cells from all donors are able to recognize and bind iRBC. However, cytoskeletal reorganization (clustering of F-actin) was observed only in some NK cells from some donors, suggesting that iRBC recognition does not always lead to functional consequences. More extensive imaging studies are underway to determine whether this cytoskeletal reorganization reflects formation of a true activating immune synapse, to identify the signals required for its induction, and to determine the downstream consequences in terms of IFN-γ production and/or cytotoxicity.

Currently available data indicate that the contact-mediated signal may be required for full activation of NK cells. Thus, freeze-thaw lysates of iRBC, which are very unlikely to form immune synapses with either NK cells or accessory cells but are efficient inducers of proinflammatory cytokines from monocyte/macrophages (29), induce significantly less IFN-γ production (12) and CD69 up-regulation (K. C. Newman, D. S. Korbel, and E. M. Riley, manuscript in preparation) than do intact iRBC. In Transwell experiments in which NK cells receive soluble but not contact-mediated signals, CD69 is up-regulated (K. C. Newman, D. S. Korbel, and E. M. Riley, manuscript in preparation) but IFN-γ is not (14). Thus, it is conceivable that contact between NK cells and iRBC results in immune synapse formation only in a subset of donors and that this leads to full NK cell activation, including up-regulation of CD25 and either IFN-γ production or exocytosis of cytotoxic granules. In this model, partial activation could be explained by bystander effects mediated by cytokines. In support of this, CD69 up-regulation on NK cells occurs within the first 6 h of coculture of PBMC with iRBC, which corresponds with the secretion of proinflammatory cytokines such as IL-1β, IL-6, TNF-α, and IL-12 (Ref.29 and M. Walther, et al., manuscript in preparation). Alternatively, given that contact appears to occur between NK cells and iRBC in all donors, it may be that contact mediates partial NK cell activation while the accessory cell-dependent signal may differ between donors and underlie the heterogeneity of response. Further studies are ongoing to dissect the contributions of different signals to the observed differences in response.

A consistent finding in this study was that partial activation was seen in both CD56bright and CD56dim NK cells while full activation was preferentially seen in CD56dim cells, and indeed LAMP-1 and CD25 were only up-regulated in the CD56dim population. CD56dim NK cells are the major population found in the peripheral blood while CD56bright NK cells are the major population in the secondary lymphoid tissue (30). CD56bright cells have been reported to lack perforin and to be superior to CD56dim cells in their ability to secrete IFN-γ (31), and, consequently, it has been suggested that the major effector function of CD56dim NK cells is cytotoxicity while CD56bright NK cells act mainly by secretion of cytokines (26, 31). Our results do not entirely support this notion. Thus, although CD56bright cells could produce IFN-γ in response to iRBC and CD56dim cells were the only ones to express LAMP-1 on their surface, CD56dim cells were just as likely as CD56bright cells to produce IFN-γ while perforin and granzyme A were up-regulated to a greater extent in CD56bright than CD56dim cells. Thus, our data suggest that fully iRBC-activated CD56dim cells sequentially up-regulate CD69 and CD25 and at the same time either differentiate into IFN-γ-producing cells or begin to express LAMP-1. This dichotomy of effector function within the CD56dim subset is in line with data indicating distinct intracellular signaling pathways leading either to cytokine secretion or cytotoxicity (32, 33).

Two recent studies have suggested that both intact iRBC and schizont extracts can induce cytokine secretion from human and mouse accessory cells in a TLR9-dependent manner (34, 35), but the nature of the receptors involved in direct iRBC recognition is unclear. Despite their lack of MHC class 1 expression, uninfected RBC clearly do not activate NK cells even when, as here, they derive from unrelated donors. Indeed, our observation that they significantly reduce baseline levels of NK activation suggests that uRBC may express alternative ligands for inhibitory NK cell receptors. If so, then reduced expression of this ligand on iRBC, or modification of its structure such that it no longer binds effectively to its receptor, might explain the ability of iRBC to activate NK cells. Alternatively, iRBC may express on their surface a novel ligand for an activating receptor. In either case, that only a subset of NK cells from a subset of individuals respond to iRBC might be explained by polymorphism or variable expression of the receptor. Our previously published data showing an association between NK cell IFN-γ responsiveness and killer cell Ig-like receptor (KIR) genotype (14), combined with our finding here that full activation of NK cells is essentially restricted to the CD56dim subset, which, unlike the CD56bright subset, expresses KIR receptors at high levels (26), suggests that KIR gene expression may influence the NK response to iRBC. The balance of expression of activating and inhibitory KIRs and the affinity of binding of individual KIR variants to polymorphic MHC class I ligands influence the ease of activation of NK cells (36), and heterogeneous KIR gene expression among NK clones has been shown to affect their activation and effector function (37). Since erythrocytes express few, if any, surface MHC molecules, any moderation of NK activation would likely be moderated through “trans” KIR-MHC interactions with other cells, e.g., cytokine-producing accessory cells.

The impact of differential NK cell activation on susceptibility or resistance to clinical malaria is not yet known. Expression of CD25 reflects the proliferative capacity of NK cells (22) and up-regulation of this marker by iRBC may lead to rapid expansion of the NK cell population early in a malaria infection, affecting the magnitude of the subsequent NK cell response. The notion that the rapid availability of IFN-γ from NK cells may limit the initial expansion of the parasite population (and thus be host protective, as in murine malaria (8, 9)) is strongly supported by our observation that, during experimentally induced human malaria infections, growth of blood-stage parasites is directly correlated with the extent of the early proinflammatory cytokine response (38). Whether NK cells might also contribute to killing of iRBC via direct cytotoxicity is less apparent, although up-regulation of markers such as LAMP-1 in high-responding cells is highly suggestive of this. Clearly, clinical studies of the relationship among NK cell phenotype, NK receptor genotype, and outcome of malaria infection are required to determine how NK cell responses to iRBC contribute to host resistance to malaria in humans.

We thank the London School of Hygiene and Tropical Medicine blood donors for their participation in this study, Carolynne Stanley, Elizabeth King, and Claire Swales for excellent technical assistance, and Peter Parham for advice and support.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This study was funded by the U.K. Medical Research Council, the Biotechnology and Biological Sciences Research Council, the Department of Trade and Industry, the Human Frontier Science Program, and the Lister Research Prize. D.S.K. is a Boehringer Ingelheim Fonds Predoctoral Fellow and C.R.A. is funded by the Fundação para a Ciência e a Tecnologia, Portugal.

4

Abbreviations used in this paper: iRBC, infected RBC; uRBC, uninfected RBC; DAPI, 4′,6′-diamidino-2-phenylindole; MFI, mean fluorescence intensity; LAMP-1, lysosomal-associated membrane protein 1; IS, immunological synapse; KIR, killer cell Ig-like receptor.

1
Snow, R. W., C. A. Guerra, A. M. Noor, H. Y. Myint, S. I. Hay.
2005
. The global distribution of clinical episodes of Plasmodium falciparum malaria.
Nature
434
:
214
-217.
2
Greenwood, B., K. Marsh, R. Snow.
1991
. Why do some African children develop severe malaria?.
Parasitol. Today
7
:
277
-281.
3
Kwiatkowski, D..
2000
. Genetic susceptibility to malaria getting complex.
Curr. Opin. Genet. Dev.
10
:
320
-324.
4
Stevenson, M., E. Riley.
2004
. Innate immunity to malaria.
Nat. Rev. Immunol.
4
:
169
-180.
5
De Souza, J. B., K. H. Williamson, T. Otani, J. H. Playfair.
1997
. Early γ interferon responses in lethal and nonlethal murine blood-stage malaria.
Infect. Immun.
65
:
1593
-1598.
6
Su, Z., M. M. Stevenson.
2000
. Central role of endogenous γ interferon in protective immunity against blood-stage Plasmodium chabaudi AS infection.
Infect. Immun.
68
:
4399
-4406.
7
Meding, S. J., S. C. Cheng, B. Simon-Haarhaus, J. Langhorne.
1990
. Role of γ interferon during infection with Plasmodium chabaudi chabaudi.
Infect. Immun.
58
:
3671
-3678.
8
Mohan, K., P. Moulin, M. M. Stevenson.
1997
. Natural killer cell cytokine production, not cytotoxicity, contributes to resistance against blood-stage Plasmodium chabaudi AS infection.
J. Immunol.
159
:
4990
-4998.
9
Choudhury, H., N. Sheikh, G. Bancroft, D. Katz, J. De Souza.
2000
. Early nonspecific immune responses and immunity to blood-stage nonlethal Plasmodium yoelii malaria.
Infect. Immun.
68
:
6127
-6132.
10
Luty, A. J., B. Lell, R. Schmidt-Ott, L. G. Lehman, D. Luckner, B. Greve, P. Matousek, K. Herbich, D. Schmid, F. Migoy-Nabias, et al
1999
. Interferon-γ responses are associated with resistance to reinfection with Plasmodium falciparum in young African children.
J. Infect. Dis.
179
:
980
-988.
11
Dodoo, D., F. M. Omer, J. Todd, B. D. Akanmori, K. A. Koram, E. M. Riley.
2002
. Absolute levels and ratios of proinflammatory and anti-inflammatory cytokine production in vitro predict clinical immunity to Plasmodium falciparum malaria.
J. Infect. Dis.
185
:
971
-979.
12
Artavanis-Tsakonas, K., E. M. Riley.
2002
. Innate immune response to malaria: rapid induction of IFN-γ from human NK cells by live Plasmodium falciparum-infected erythrocytes.
J. Immunol.
169
:
2956
-2963.
13
Artavanis-Tsakonas, K., J. E. Tongren, E. M. Riley.
2003
. The war between the malaria parasite and the immune system: immunity, immunoregulation and immunopathology.
Clin. Exp. Immunol.
133
:
145
-152.
14
Artavanis-Tsakonas, K., K. Eleme, K. McQueen, P. Parham, D. Davis, E. Riley.
2003
. Activation of a subset of human natural killer cells upon contact with Plasmodium falciparum-infected erythrocytes.
J. Immunol.
171
:
5396
-5405.
15
Cooper, M. A., T. A. Fehniger, S. C. Turner, K. S. Chen, B. A. Ghaheri, T. Ghayur, W. E. Carson, M. A. Caligiuri.
2001
. Human natural killer cells: a unique innate immunoregulatory role for the CD56bright subset.
Blood
97
:
3146
-3151.
16
Testi, R., D. D’Ambrosio, R. De Maria, A. Santoni.
1994
. The CD69 receptor: a multipurpose cell-surface trigger for hematopoietic cells.
Immunol. Today
15
:
479
-483.
17
Lanier, L. L., D. W. Buck, L. Rhodes, A. Ding, E. Evans, C. Barney, J. H. Phillips.
1988
. Interleukin 2 activation of natural killer cells rapidly induces the expression and phosphorylation of the Leu-23 activation antigen.
J. Exp. Med.
167
:
1572
-1585.
18
Brown, M. G., A. A. Scalzo, K. Matsumoto, W. M. Yokoyama.
1997
. The natural killer gene complex: a genetic basis for understanding natural killer cell function and innate immunity.
Immunol. Rev.
155
:
53
-65.
19
Risso, A., D. Smilovich, M. C. Capra, I. Baldissarro, G. Yan, A. Bargellesi, M. E. Cosulich.
1991
. CD69 in resting and activated T lymphocytes: its association with a GTP binding protein and biochemical requirements for its expression.
J. Immunol.
146
:
4105
-4114.
20
Moretta, A., A. Poggi, D. Pende, G. Tripodi, A. M. Orengo, N. Pella, R. Augugliaro, C. Bottom, E. Ciccone, L. Moretta.
1991
. CD69-mediated pathway of lymphocyte activation: anti-CD69 monoclonal antibodies trigger the cytolytic activity of different lymphoid effector cells with the exception of cytolytic T lymphocytes expressing T cell receptor α/β.
J. Exp. Med.
174
:
1393
-1398.
21
Korbel, D., O. Finney, E. Riley.
2004
. Natural killer cells and innate immunity to protozoan pathogens.
Int. J. Parasitol.
34
:
1517
-1528.
22
Clausen, J., B. Vergeiner, M. Enk, A. L. Petzer, G. Gastl, E. Gunsilius.
2003
. Functional significance of the activation-associated receptors CD25 and CD69 on human NK-cells and NK-like T-cells.
Immunobiology
207
:
85
-93.
23
Orago, A., C. Facer.
1991
. Cytotoxicity of human natural killer (NK) cell subsets for Plasmodium falciparum erythrocytic schizonts: stimulation by cytokines and inhibition by neomycin.
Clin. Exp. Immunol.
86
:
22
-29.
24
Beresford, P. J., Z. Xia, A. H. Greenberg, J. Lieberman.
1999
. Granzyme A loading induces rapid cytolysis and a novel form of DNA damage independently of caspase activation.
Immunity
10
:
585
-594.
25
Alter, G., J. M. Malenfant, M. Altfeld.
2004
. CD107a as a functional marker for the identification of natural killer cell activity.
J. Immunol. Methods
294
:
15
-22.
26
Jacobs, R., G. Hintzen, A. Kemper, K. Beul, S. Kempf, G. Behrens, K. W. Sykora, R. E. Schmidt.
2001
. CD56bright cells differ in their KIR repertoire and cytotoxic features from CD56dim NK cells.
Eur. J. Immunol.
31
:
3121
-3127.
27
Davis, D., M. Dustin.
2004
. What is the importance of the immunological synapse?.
Trends Immunol.
25
:
323
-327.
28
McCann, F. E., B. Vanherberghen, K. Eleme, L. M. Carlin, R. J. Newsam, D. Goulding, D. M. Davis.
2003
. The size of the synaptic cleft and distinct distributions of filamentous actin, ezrin, CD43, and CD45 at activating and inhibitory human NK cell immune synapses.
J. Immunol.
170
:
2862
-2870.
29
Scragg, I. G., M. Hensmann, C. A. Bate, D. Kwiatkowski.
1999
. Early cytokine induction by Plasmodium falciparum is not a classical endotoxin-like process.
Eur. J. Immunol.
29
:
2636
-2644.
30
Fehniger, T. A., M. A. Cooper, G. J. Nuovo, M. Cella, F. Facchetti, M. Colonna, M. A. Caligiuri.
2003
. CD56bright natural killer cells are present in human lymph nodes and are activated by T cell-derived IL-2: a potential new link between adaptive and innate immunity.
Blood
101
:
3052
-3057.
31
Ferlazzo, G., C. Munz.
2004
. NK cell compartments and their activation by dendritic cells.
J. Immunol.
172
:
1333
-1339.
32
Zompi, S., F. Colucci.
2005
. Anatomy of a murder–signal transduction pathways leading to activation of natural killer cells.
Immunol. Lett.
97
:
31
-39.
33
Vivier, E., J. A. Nunes, F. Vely.
2004
. Natural killer cell signaling pathways.
Science
306
:
1517
-1519.
34
Pichyangkul, S., K. Yongvanitchit, U. Kum-arb, H. Hemmi, S. Akira, A. M. Krieg, D. G. Heppner, V. A. Stewart, H. Hasegawa, S. Looareesuwan, et al
2004
. Malaria blood stage parasites activate human plasmacytoid dendritic cells and murine dendritic cells through a Toll-like receptor 9-dependent pathway.
J. Immunol.
172
:
4926
-4933.
35
Coban, C., K. Ishii, T. Kawai, H. Hemmi, S. Sato, S. Uematsu, M. Yamamoto, O. Takeuchi, S. Itagaki, N. Kumar, et al
2005
. Toll-like receptor 9 mediates innate immune activation by the malaria pigment hemozoin.
J. Exp. Med.
201
:
19
-25.
36
McQueen, K., P. Parham.
2002
. Variable receptors controlling activation and inhibition of NK cells.
Curr. Opin. Immunol.
14
:
615
-621.
37
Carr, W., A. Little, E. Mocarski, P. Parham.
2002
. NK cell-mediated lysis of autologous HCMV-infected skin fibroblasts is highly variable among NK cell clones and polyclonal NK cell lines.
Clin. Immunol.
105
:
126
-140.
38
Walther, R. M., J. E. Tongren, L. Andrews, D. Korbel, E. King, H. Fletcher, R. F. Andersen, P. Bejon, F. Thompson, S. J. Dunachie, et alUp-regulation of TGF-β, FOXP3, and CD4+CD25+ regulatory T cells correlates with more rapid parasite growth in human malaria infection.
Immunity
23
:
287
-296.