We report that human peripheral NK cells expressing high CD56 levels (CD56+high) are terminally differentiated cells indistinguishable from mature NK cells recently activated in the presence of IL-12, and not a functionally distinct NK-cell subset or progenitors to mature CD56+low NK cells. CD56+high NK cells coexpress all differentiation Ags constitutive or inducible in mature (CD56+) NK cells, except CD16, present at lower level than on most mature NK cells. Also, activation markers, activating receptors and adhesion molecules, and most inducible receptors are expressed exclusively and constitutively and are inducible at higher levels on CD56+high than on CD56+low NK cells. Consistent with their activated phenotype, many CD56+high NK cells are cycling and mediate heightened effector functions (proliferation, IFN-γ and IL-10 but not IL-13 production) in response to IL-12 and other NK cell-specific stimuli. Conversely, IL-12 induces on CD56+low NK cells all markers constitutively expressed on the CD56+high NK cells, concomitantly preventing the IL-2 (and IL-15)-inducible expression of NKp44 and CD16 re-expression after immune complex-induced down-modulation, and CD56−/+low NK cells acquire a CD56+high NK cell phenotype in short term in vitro culture with IL-12. The significance of these findings to the NK cell-mediated regulation of immune responses and NK cell development is discussed.

The major role of IL-12 in NK-cell biology possibly rests on the regulation of their terminal development from phenotypically and functionally immature CD161+CD56 type 2 cytokine+ to mature CD56+IFN-γ+ cells expressing all NK cell markers (1). However, its primary role in the NK cell-mediated effects as innate effectors and regulators of adaptive immune responses depends on its concomitant direct effects on functions and phenotype of peripheral CD56+ NK cells. These include: synergy with IL-2 to enhance IFN-γ production regulating, in a transcription-independent fashion, IFN-γ mRNA stability (2); induction of proliferation in CD56+ NK cells reverted to a resting state after cytokine-induced activation (3); enhancement of cytotoxicity (4) and granulogenesis (5); and modulated expression of NK cell differentiation Ags via transcriptional regulation of genes for selected molecules (e.g., CD161) specialized in binding target cell ligands, antagonizing, in this context, the effects of other cytokines like IL-2 (6).

Two NK cell populations, CD56bright3 (7) and CD56+CD16 (8), detectable in low proportions in freshly separated PBL, have been proposed to represent functionally unique or possibly more immature cells (7, 8). However, in light of the direct effects of IL-12 and other cytokines and ligands on phenotype and functions of mature NK cells and the fact that immature NK cells are CD56 (1, 9, 10), the possibility should be considered that these populations contain mature NK cells, the functions or phenotype of which have been recently modulated in vivo. This is relevant to understanding NK cell biology and important on a clinical standpoint to define any possible significance of NK cell populations that, like the CD56bright NK cells, are detected at increased proportions in vivo during IL-2 therapy (11) or in affected tissues in pathological inflammatory conditions like rheumatoid arthritis (12). Additionally, in the mouse, CD56 is not expressed (13), and reagents specific for CD16 (and not the FcγRII exclusively expressed on myeloid and B cells; Refs. 14 and 15) are lacking, and NK cell detection can rely only on expression of CD161 (expressed on both immature and mature NK cells), CD94 (16), and the Ly49 family members (reviewed in Ref. 17), functionally equivalent to, but structurally distinct from human killer Ig-like receptors (KIR) 4 (reviewed in Ref. 18). Thus, conclusive definition of functions and phenotype of human NK cells expressing CD56 and CD16 levels different from those on most mature NK cells, as addressed here, is also essential to guide interpretation of preclinical studies manipulating NK cells in animal models.

PBL were separated from healthy adult and neonatal (umbilical cord) peripheral blood (19). When indicated, sheep E treated with 2-aminoethylisothiouronium bromide (EAET) were used to enrich most mature CD2+ T and NK cells (20). Immature CD56 NK cells were purified from the EAET cell fraction after depletion of leukocytes expressing FcγR upon adherence to immune complex monolayers (EA) (20). To induce CD16 downmodulation, PBL were depleted of B cells (EAET rosetting) and let adhere on EA monolayers (EA+), from which, after a 2-h incubation at 37°C, nonadherent cells were discarded, and the adherent ones recovered. Cultures of the different cell populations were for the indicated time periods in the presence of the indicated cytokines and anti-TNF-α mAb to prevent the otherwise massive NK cell death in cultures with IL-12 (10) (conditions and concentrations as detailed in Ref. 1). Culture of sorted cells included autologous PBMC as feeder cells (CFSE-labeled (1), to distinguish them from the cells of interest, and 50-Gy irradiated), and anti-IL-12 mAb (1).

The fluorochrome- or biotin-labeled CD2, CD3, CD5, CD16, CD56, CD161, NKp46, KIR2DL1, HLA-DR mAb used for surface phenotyping or cell sorting (immunofluorescence (IFL), flow cytometry, according to our published protocols) were as previously reported (1, 9). mAb to NKp44 (3.43.13) and CD244 (HD2) (2B4, a member of the CD2 family) (21) were gifts of M. Colonna; CD45RO and CD62L mAb were from Caltag Laboratories (Burlingame, CA), and the CD162R mAb (PEN5, clone 5H10.21.5) was from Beckman Coulter (Fullerton, CA). Viable cells, gated based on light scatter characteristics, were analyzed on an XL-MCL automated analytical flow cytometer or sorted on an EPICS Elite flow cytometer (Beckman Coulter). Listmode data were analyzed with the WinMDI Flow Cytometry Application (J. Trotter). When sorting Ag−/low and Ag+ populations, ∼10% of the cells with, respectively, the highest and lowest fluorescence intensity were excluded. Stringency of purification was confirmed on restaining and analysis of cell aliquots immediately after sorting (not shown).

Cytokine accumulation in cells stimulated maximally with PMA (2 × 10−9 M) plus A23187 (0.2 μg/ml) plus IL-2 (102 U/ml), or specifically with the indicated cytokines and target cells, was detected by intracellular IFL using previously described anti-IL-13 (PE), -IFN-γ (PE), -TNF-α (PE), -IL-10 (PE) mAb. Conditions (22) and mAb (23) were according to our established protocols. For IL-13 detection in CD3CD161+CD56 immature NK cells, (CD3+CD5+CD16+CD56)-biotin and CD161 (nonlabeled) mAb detected with CyC-labeled streptavidin and FITC-labeled goat anti-mouse Ig were used to stain EAETEA PBL and analysis was performed on gated CyCFITC+ cells. Autologous EAET+ CD56+low and CD56+high NK cells were analyzed in parallel. ELISA for IL-13 detection was as previously reported (10).

Lymphocytes were cultured for the indicated times in the different conditions after addition of BrdU (60 μM) (24) at the start of culture. BrdU+ cells were detected with FITC-anti-BrdU mAb (BD PharMingen, San Diego, CA; IFL) on cells fixed and permeabilized as for intracellular cytokine detection (22), except that they were incubated (30 min, 37°C) with 300 μg/ml DNase I after the permeabilization step (modified from Ref. 25) and washed before adding the anti-BrdU mAb. When indicated, the G1-S phase transition cell cycle inhibitor mimosine was added to the cultures as reported (22).

This was performed, using THP-1 target cells and the indicated mAb in 4-h 51Cr release redirected cytotoxicity according to published protocols (26).

To determine whether IL-12 and IL-15 modulate phenotype and functions of peripheral mature CD161+CD56+ NK cells, the expression of functionally relevant differentiation Ags was analyzed on NK cells from 5-day cultures of freshly isolated PBL with these monokines, single or combined (Fig. 1). As expected, CD56 expression was higher on CD3CD161+ total NK cells from cultures with IL-15 or IL-12 than without monokines; the CD161 levels were highest after culture with IL-12; and CD16 expression was decreased on the CD3CD56+ mature NK cells after culture with IL-12, and almost undetectable when IL-15 was also added. The activating receptor NKp46 and the two costimulatory/adhesion molecules CD2 and CD244 (2B4) had highest expression on cells from cultures with IL-12, both IL-15 and IL-12, and IL-12 alone, respectively. KIR expression (KIR3DL1 reported) was moderately increased compared with that on cells freshly isolated or cultured without exogenous monokines, only after culture with IL-15. The proportion of NK cells expressing CD62L (L-selectin), an adhesion molecule involved in migration through high endothelial venules, was increased in cultures with IL-12, whereas CD162R (PEN5), a potential ligand for CD62L (27), was lost by most NK cells from cultures with both IL-15 and IL-12, but not significantly modulated during culture with IL-12 alone. Identical results were obtained using IL-2 instead of IL-15, and cultures for shorter times gave similar results, especially CD16 down-regulation, although less apparent (not shown). Longer culture times were not analyzed, given the uncontrollable intervention of complicating factors like cell death, especially in cultures with IL-12 (1) and proliferation in response to IL-15, maximal at days 5–7 (not shown).

FIGURE 1.

Cytokine-induced modulation of NK cell differentiation Ag expression. PBL were cultured for 5 days in medium without (none) or with the cytokines indicated at right (anti-TNF-α-mAb added throughout). Expression of the markers indicated at the bottom (mAb labeled with the listed fluorochromes was analyzed (four-color IFL) on gated CD3 (PE-Texas Red) CD56 (PE)+ cells. Fluor, fluorescein; TC, Tricolor. Histogram presentation in this and all figures: x-axis, fluorescence intensity; y-axis, relative cell number; ·····, isotype-matched control; ——, experimental mAb; vertical lines, peak fluorescence intensity of the cells cultured with IL-15. Experiment representative of at least five with each mAb with adult and neonatal PBL.

FIGURE 1.

Cytokine-induced modulation of NK cell differentiation Ag expression. PBL were cultured for 5 days in medium without (none) or with the cytokines indicated at right (anti-TNF-α-mAb added throughout). Expression of the markers indicated at the bottom (mAb labeled with the listed fluorochromes was analyzed (four-color IFL) on gated CD3 (PE-Texas Red) CD56 (PE)+ cells. Fluor, fluorescein; TC, Tricolor. Histogram presentation in this and all figures: x-axis, fluorescence intensity; y-axis, relative cell number; ·····, isotype-matched control; ——, experimental mAb; vertical lines, peak fluorescence intensity of the cells cultured with IL-15. Experiment representative of at least five with each mAb with adult and neonatal PBL.

Close modal

NKp44, reportedly undetectable on freshly ex vivo-derived NK cells and inducible on the same cells after a 10-day culture with IL-2 (28), was expressed at high levels on NK cells from a 5-day culture of PBL with IL-2 and IL-15, alone or combined (Fig. 2,A), with induced expression detectable within 15 h (Fig. 2,B, top). Induction of NKp44 expression by IL-15- and IL-2- was significantly inhibited by IL-12 (Fig. 2,A), and, once induced upon culture with IL-2 (or IL-15) alone, it was down-regulated on further culture in the presence of IL-12 (Fig. 2 B, bottom). This down-modulation was moderate or complete, respectively, depending on the presence or absence of IL-2 during secondary culture. Among other cytokines that affect NK cell functions (IL-4, IL-13, IL-18, IFN-α, IFN-γ, TNF-α), only IL-18 and IFN-γ induced low level NKp44 expression in a small proportion of NK cells, and IL-4, IL-18, and IFN-γ only mildly attenuated its IL-2-induced expression, with effects significantly lesser than those of IL-12 (not shown).

FIGURE 2.

Effect of IL-12 on NKp44 expression. A and B, NKp44 expression was analyzed on gated CD3CD56+ lymphocytes, as in Fig. 1, after culture of PBL with the cytokines indicated at top and right for 6 days (A) or 15 h (B, top), or after 3-day secondary culture after the 5-day culture with IL-2 alone (B, bottom). Here, and in all figures, the numbers in the panels are percent positive cells, background subtracted. C, Redirected cytotoxicity assays were performed with THP-1 target cells and mAb to the indicated Ags, using as effector cells PBL after 5-day culture with the indicated cytokines. Each experiment is representative of at least three.

FIGURE 2.

Effect of IL-12 on NKp44 expression. A and B, NKp44 expression was analyzed on gated CD3CD56+ lymphocytes, as in Fig. 1, after culture of PBL with the cytokines indicated at top and right for 6 days (A) or 15 h (B, top), or after 3-day secondary culture after the 5-day culture with IL-2 alone (B, bottom). Here, and in all figures, the numbers in the panels are percent positive cells, background subtracted. C, Redirected cytotoxicity assays were performed with THP-1 target cells and mAb to the indicated Ags, using as effector cells PBL after 5-day culture with the indicated cytokines. Each experiment is representative of at least three.

Close modal

Although most other activating and costimulatory/adhesion molecules were up-regulated in response to IL-12 (alone or with IL-15), CD16 was expressed at significantly lower levels after culture with IL-12. However, it was capable of triggering the highest cytotoxicity levels in redirected cytotoxicity (Fig. 2 C), confirming that the expression level of activating receptors in I.Fl. does not necessarily correlate with their ability to trigger cytotoxicity when engaged by excess ligand. NKp44, although expressed at lower levels after culture with IL-12 plus IL-15 than with IL-15 alone, triggered in both conditions cytotoxicity at levels similar to those of NKp46, expressed at higher levels in the former condition.

The phenotype of mature CD56+ NK cells after short term culture with IL-12 and IL-15 was similar to that reported for CD56bright NK cells (summarized and referenced in Table I, reporting also the results on IL-12-induced modulation of all NK cell differentiation Ags tested here and, with similar outcome (not shown), also in NK cells from 10-day cocultures of PBL with the RPMI-8866 B-lymphoblastoid cell line; Ref. 29). This included high CD56, CD62L, and CD2 levels, and low-to-null CD16 levels. The CD56+low NK cells in freshly isolated adult PBL (Fig. 3 A) were mostly CD16+, although a minor proportion was consistently CD16 (9 ± 3%, n = 7), similar to that in neonates (14 ± 8%, n = 5, p > 0.13). The proportions of CD56+high cells within the CD56+ NK cells or total PBL were similar in adult (8 ± 4% and 0.8 ± 0.7%, respectively, n = 18) and neonatal (7 ± 4% and 0.9 ± 0.3%, respectively, n = 9) cells. Consistent with previous reports (8, 30), the CD56+high NK cells were mostly CD16 (adult, 72 ± 14%, n = 7; neonate, 79 ± 4%, n = 5) or CD16+low.

Table I.

Expression and modulation of NK cell differentiation Ags

AgModulation of ExpressionaExpressionb
IL-12Reference(s)cCD56+lowCD56+highReference(s)
Activating receptors CD16 ↓ F, (6) −/+low F, (30
 NKp44 ∗ − +low 
 NKp46 ↑ +high 
Inhibitory receptors CD94 ↑ NS, (44+high (44
 KIR −/+ −/+ F, (45
Costimulatory/adhesion molecules CD2 ↑ F, (46+high 
 CD8 NS −/+ −/+ NS 
 CD11b  NS NS 
 CD62L ↑ F, (47−/+ F, (47
 CD162R ↓ −/+ − F, (27
 CD244 ↑ 
AgModulation of ExpressionaExpressionb
IL-12Reference(s)cCD56+lowCD56+highReference(s)
Activating receptors CD16 ↓ F, (6) −/+low F, (30
 NKp44 ∗ − +low 
 NKp46 ↑ +high 
Inhibitory receptors CD94 ↑ NS, (44+high (44
 KIR −/+ −/+ F, (45
Costimulatory/adhesion molecules CD2 ↑ F, (46+high 
 CD8 NS −/+ −/+ NS 
 CD11b  NS NS 
 CD62L ↑ F, (47−/+ F, (47
 CD162R ↓ −/+ − F, (27
 CD244 ↑ 
a

Modulation of expression of the indicated antigens on NK cells after culture with IL-2 (IL-15) + IL-12, including any induced, direct, or indirect effect. Up arrow, increased; down arrow, decreased; equal sign, unchanged.

b

Expression of the indicated antigens on CD3CD56+low and CD56+high NK cells in freshly isolated adult PBL. +, positive; −, negative; +low and +high, relatively low and high compared with the other population; −/+, discrete − and + populations detected.

c

F, figures in this article; NS, data reported here as not shown; ∗, IL-2 or IL-15-induced expression of NKp44 was attenuated by IL-12.

FIGURE 3.

Surface phenotype of CD56+high and CD56+low NK cells within freshly separated PBL. Phenotypic analysis with mAb to the Ags indicated at the bottom was performed on freshly separated PBL (four-color IFL). TC, Tricolor. The histograms in A and C and density plots in B are on gated CD56+low (top) and CD56+high (bottom) cells (gated regions boxed in the bottom left density plots) within the gated CD3CD56+ NK cells (density plots, top left). Vertical lines, Peak fluorescence intensity of the CD56+low population. Experiment representative of at least five for each mAb with adult PBL and three with neonatal lymphocytes.

FIGURE 3.

Surface phenotype of CD56+high and CD56+low NK cells within freshly separated PBL. Phenotypic analysis with mAb to the Ags indicated at the bottom was performed on freshly separated PBL (four-color IFL). TC, Tricolor. The histograms in A and C and density plots in B are on gated CD56+low (top) and CD56+high (bottom) cells (gated regions boxed in the bottom left density plots) within the gated CD3CD56+ NK cells (density plots, top left). Vertical lines, Peak fluorescence intensity of the CD56+low population. Experiment representative of at least five for each mAb with adult PBL and three with neonatal lymphocytes.

Close modal

Comparison of the relative expression levels of the Ags modulated by IL-12 (four-color IFL) indicated higher levels of CD2, CD161, CD244, and NKp46, detectable in CD56+high and CD56+low NK cells in freshly isolated PBL, in the CD56+high cells (Fig. 3). NKp44, undetectable on CD56+low NK cells, was present at low levels in most CD56+high cells; it was induced at low levels in some CD56+low NK cells (Fig. 2) but rapidly induced to high levels on most CD56+high NK cells within 15 h of stimulation with IL-2 or IL-15 (not shown). The CD56+high cells contained significantly higher proportions of CD62L+ cells (61 ± 16%), with expression density higher than that on the positive CD56+low population (15 ± 7%, n = 11, p < 1 × 10−5). Conversely, the proportions of CD162R+ cells were significantly higher among the CD56+low (57 ± 3%) than the CD56+high (12 ± 8%, n = 4, p < 0.002) NK cells. Similar proportions of CD56+low and CD56+high NK cells coexpressed CD62L and CD162R (6 ± 5% and 3 ± 2%, respectively, n = 4) and expressed neither (32 ± 6% and 29 ± 17%, respectively, n = 4). A fraction of CD56+high, but not CD56+low NK cells, higher among adult than neonatal NK cells, also expressed CD45RO (Fig. 3 C).

The activation markers CD25 (IL-2Rα) and HLA-DR, undetectable on CD56+low NK cells, were also expressed at low levels on most CD56+high NK cells, and minor proportions of CD56+high NK cells expressed high HLA-DR levels. Expression of these activation markers, often linked to ongoing or occurred proliferation, suggested the possibility that a portion of the CD56+high cells might be actively proliferating. Indeed, ∼30% CD56+high NK cells within freshly isolated PBL cultured for 3 days without exogenous stimuli and with BrdU added on day 0, were BrdU+ (Fig. 4). A small cell proportion also incorporated BrdU in the presence of mimosine, suggesting that such cells were in S phase before culture. These results indicate that freshly ex vivo-derived CD56+high NK cells contain actively cycling cells, likely as a result of recent in vivo activation. These data do not contrast previous reports demonstrating no detectable [3H]TdR incorporation on days 3–4 of culture of CD56+high NK cells in the absence of exogenous stimulation (8, 11), because analysis of proliferation at the single-cell level after adding BrdU at the start of culture allows one to detect active cell proliferation that may wane quickly and thus be undetectable, at a population level, after culture without stimuli. Many of the CD56+high NK cells that did not proliferate autonomously did so in response to IL-12 or low IL-2 doses, whereas almost all proliferated in response to high IL-2 doses. CD56+low NK cells did not proliferate in the absence of exogenous stimulation, and lower percentages of these cells than of CD56+high cells proliferated in response to IL-2 or IL-12.

FIGURE 4.

Proliferative capability of CD56+high and CD56+low NK cells within freshly separated PBL. The percentages of BrdU+ cells were determined in gated CD3 (PE-Texas Red)CD56 (PE)+lowCD16 (CyC)+ (top) and CD3CD56+high (bottom) NK cells from a 3-day culture of freshly isolated adult PBL in medium without (none) or with added the indicated cytokines (or mimosine, control) and BrdU on day 0. Experiment representative of three.

FIGURE 4.

Proliferative capability of CD56+high and CD56+low NK cells within freshly separated PBL. The percentages of BrdU+ cells were determined in gated CD3 (PE-Texas Red)CD56 (PE)+lowCD16 (CyC)+ (top) and CD3CD56+high (bottom) NK cells from a 3-day culture of freshly isolated adult PBL in medium without (none) or with added the indicated cytokines (or mimosine, control) and BrdU on day 0. Experiment representative of three.

Close modal

The combined findings of active proliferation, high proliferative response to IL-12, and activated phenotype confirm that peripheral blood CD56+high NK cells are recently activated cells.

CD56+high NK cells, unlike most mature NK cells, have characteristically low CD16 expression. The contention that the CD56+highCD16 phenotype is stable derives from the observation that CD16 expression is not inducible on CD56+high CD16 NK cells upon culture with IL-2 (8). Although experiments (not shown) confirmed these data, low level CD16 expression was induced in isolated CD56+high CD16 NK cells after a 5-day culture with IL-2 and gamma-ray-irradiated autologous PBMC (TNF-α- and IL-12-neutralizing mAb added to prevent cell death; Ref. 1) with low proportions of cells expressing relatively higher levels (Fig. 5 A). Thus, factors other than IL-2, related to the interaction with accessory cells, can and are necessary to induce NK cell CD16 expression on CD16 NK cells.

FIGURE 5.

IL-12-mediated modulation of CD16 expression. A, CD3CD56+highCD16 cells were separated on day 0 (three-color IFL, cell sorting) from freshly isolated lymphocytes (top, gated regions highlighted) and cultured for 5 days in the presence of CFSE-labeled autologous irradiated PBMC and IL-2 (see Materials and Methods). At this time (bottom), CD56 and CD16 expression was analyzed (density plots at right) within gated CFSECD3 cells (left histogram). B and C, PBL from two adult donors were depleted of B cells (EAET rosetting) and left nontreated (bottom) or allowed to adhere on EA monolayers (EA+), from which nonadherent cells were discarded and the adherent ones recovered after a 2-h incubation at 37°C (top). Both samples were then cultured for the indicated times in the presence of the indicated cytokines (anti-TNF-α-mAb added to all cultures). CD16 expression was analyzed (three-color IFL) on the derived cells gated on CD3 (PE)CD161 (CyC)+ cells. Each experiment is representative of two.

FIGURE 5.

IL-12-mediated modulation of CD16 expression. A, CD3CD56+highCD16 cells were separated on day 0 (three-color IFL, cell sorting) from freshly isolated lymphocytes (top, gated regions highlighted) and cultured for 5 days in the presence of CFSE-labeled autologous irradiated PBMC and IL-2 (see Materials and Methods). At this time (bottom), CD56 and CD16 expression was analyzed (density plots at right) within gated CFSECD3 cells (left histogram). B and C, PBL from two adult donors were depleted of B cells (EAET rosetting) and left nontreated (bottom) or allowed to adhere on EA monolayers (EA+), from which nonadherent cells were discarded and the adherent ones recovered after a 2-h incubation at 37°C (top). Both samples were then cultured for the indicated times in the presence of the indicated cytokines (anti-TNF-α-mAb added to all cultures). CD16 expression was analyzed (three-color IFL) on the derived cells gated on CD3 (PE)CD161 (CyC)+ cells. Each experiment is representative of two.

Close modal

Taking this into account, we tested the possibility that the low CD16 expression on the CD56+high NK cells might result, in part, from in vivo activation of cells previously CD16+ by ligands and cytokines modulating FcγR expression (e.g., immune complexes (20), and IL-12; Ref. 6). For this, we analyzed CD16 re-expression after its down-modulation had been induced by immune complexes on cells CD16+ (Fig. 5, B and C). Provided IL-12 was neutralized to avoid CD16 down-modulation and inhibition of its re-expression, IL-15 allowed CD16 re-expression on CD16+ NK cells, although at levels lower than those on control untreated cells, after its complete loss was induced on engagement by immune complexes (Fig. 5,B), and induced it at high levels on a portion of cells similar to that in nonstimulated cells after it was significantly lowered, but not lost completely (Fig. 5 C).

Most adult peripheral blood NK cells produce TNF-α and low levels of IFN-γ, whereas variable or low proportions produce high levels IFN-γ or IL-13, respectively (10). Immature CD56 NK cells produce no IFN-γ and higher TNF-α levels, and most produce IL-13 (1, 9). To determine whether CD56+high NK cells produce cytokines like activated, mature NK cells, IFN-γ and TNF-α production in response to various stimuli was analyzed at the single cell level in CD56+high and CD56+low NK cells in the same PBL population, freshly isolated or after a 15-h culture with IL-2 and IL-12 (Fig. 6). The IFN-γ levels accumulated on maximal (PMA plus Ca2+ ionophore plus IL-2) stimulation were similar in CD56+low (27 ± 26%) and CD56+high NK cells (30 ± 31%, n = 7) in freshly isolated PBL but were higher in the latter when stimulated after a 15-h culture with IL-12. On maximal stimulation, CD56+low NK cells produced higher levels of TNF-α than CD56+high cells, whereas after a 15-h culture with IL-12 greater proportions of CD56+high than CD56+low cells produced TNF-α at higher levels. Also, after 15-h culture with IL-12, CD56+high NK cells produced higher levels of IFN-γ and TNF-α than the CD56+low ones in the absence of additional stimulation. In fresh PBL, target cells (K562) induced IFN-γ production in a small portion of CD56+high but not CD56+low NK cells; consistent with only ∼25% CD16+ cells in the CD56+high population, CD16 engagement induced a higher proportion of CD56+low than CD56+high NK cells to produce IFN-γ. In cells from 15-h culture with IL-12, both K562 and CD16 engagement induced IFN-γ accumulation similar to that induced upon maximal stimulation, i.e., greater than observed without stimulation, in CD56+low NK cells, whereas IFN-γ accumulation in CD56+high NK cells after culture was maximal regardless of stimulation.

FIGURE 6.

Cytokine production by CD56+high and CD56+low NK cells. Intracellular accumulation of IFN-γ (top) and TNF-α (bottom) was analyzed (intracellular IFl), after a 5-h stimulation with the stimuli indicated on top, within gated CD3/CD5 (FITC), CD56 (CyC)+high, and CD56+low NK cells in adult PBL freshly isolated (left) or cultured for 18 h with IL-2 and IL-12 (right). Experiment representative of three.

FIGURE 6.

Cytokine production by CD56+high and CD56+low NK cells. Intracellular accumulation of IFN-γ (top) and TNF-α (bottom) was analyzed (intracellular IFl), after a 5-h stimulation with the stimuli indicated on top, within gated CD3/CD5 (FITC), CD56 (CyC)+high, and CD56+low NK cells in adult PBL freshly isolated (left) or cultured for 18 h with IL-2 and IL-12 (right). Experiment representative of three.

Close modal

Immature CD56 NK cells lose ability to produce IL-13 as they become mature CD56+ cells. As predicted if CD56+high NK cells derive from CD56+low NK cells, analysis at the single-cell level detected no IL-13 accumulation in CD56+high NK cells (Fig. 7, A and B) and no IL-13 was detectable by ELISA in culture supernatants from the same cells after purification (Fig. 7 A), whereas IL-13 accumulation and production were confined, as expected (1, 10), to the CD56 NK cells.

FIGURE 7.

IL-13 and IL-10 production by CD56+high and CD56+low NK cells. Intracellular accumulation of the indicated cytokines was analyzed (intracellular IFL), after a 5-h stimulation with PMA plus Ca2+ ionophore plus IL-2, within gated CD3CD161+CD56 lymphocytes in freshly separated EAETEA PBL (left), and CD3CD56+low and CD56+high populations in the EAET+ cells derived from them (middle and right panels; A); and the gated CD3/CD5 (FITC) population indicated at top (detected by CyC), in adult PBL freshly separated (B), or cultured for 18 h (C) or 8 days (D; gated CD161+ cells) in medium without (none) or with IL-2+IL-12, as indicated (anti-TNF-α-mAb added throughout). The top numbers in A are IL-13 (picograms per milliliter; mean ± SD, ELISA) detected in the supernatants after an 18-h stimulation of the respective populations isolated by cell sorting. The right histograms in B report cytokine accumulation in cells/particles, not gated for forward and side scatter, within the cells with high side light scatter (SS = apoptotic cells; gated region indicated in the density plots). Each experiment is representative of at least three.

FIGURE 7.

IL-13 and IL-10 production by CD56+high and CD56+low NK cells. Intracellular accumulation of the indicated cytokines was analyzed (intracellular IFL), after a 5-h stimulation with PMA plus Ca2+ ionophore plus IL-2, within gated CD3CD161+CD56 lymphocytes in freshly separated EAETEA PBL (left), and CD3CD56+low and CD56+high populations in the EAET+ cells derived from them (middle and right panels; A); and the gated CD3/CD5 (FITC) population indicated at top (detected by CyC), in adult PBL freshly separated (B), or cultured for 18 h (C) or 8 days (D; gated CD161+ cells) in medium without (none) or with IL-2+IL-12, as indicated (anti-TNF-α-mAb added throughout). The top numbers in A are IL-13 (picograms per milliliter; mean ± SD, ELISA) detected in the supernatants after an 18-h stimulation of the respective populations isolated by cell sorting. The right histograms in B report cytokine accumulation in cells/particles, not gated for forward and side scatter, within the cells with high side light scatter (SS = apoptotic cells; gated region indicated in the density plots). Each experiment is representative of at least three.

Close modal

Upon maximal stimulation, almost no IL-10 was detected in viable CD56+low or CD56+high NK cells in freshly isolated PBL, whereas IFN-γ accumulated at similar levels in both populations. Consistent with previous reports associating IL-10 production with dying cells in NK and T cells (1, 23), IL-10 and IFN-γ, but not IL-13, accumulated at low levels in CD3 particles with high side scatter characteristics of lymphocytes undergoing apoptosis. In PBL cultured with IL-12 for 15 h, a small portion of the CD56+high, but not CD56+low NK cells, produced detectable IL-10 levels (Fig. 7,C), whereas most NK cells produced low levels IL-10 after an 8-day culture with IL-12 (Fig. 7 D), with concomitantly increased CD56 expression (not shown).

We report that, within CD56+ NK cells in freshly separated lymphocytes, those CD56+high have a phenotype identical with that of mature (CD3CD56+) NK cells that have been stimulated with combinations of IL-15 (or IL-2) and IL-12 (Table I), including the highest expression of activating receptors (NKp44 and NKp46), costimulatory receptors and adhesion molecules (CD2, CD62L, CD244), and the inhibitory receptor CD94, with concomitantly lowered expression of CD16 and CD162R. Also, unlike resting CD56+low and like activated NK cells, the CD56+high NK cells in freshly isolated PBL expressed IL-2Rα (CD25), CD45RO, HLA-DR, and NKp44; responded to IL-12 with fast kinetics; and proliferated in the absence of exogenous IL-2 and IL-15. These results strongly support the conclusion that, rather than an immature or distinct specialized subset, CD56+high NK cells are terminally differentiated cells that, within the general pool of mature CD56+ NK cells, have been recently activated in vivo. Known progenitors to CD56+ NK cells, including CD34+ hemopoietic precursors (31), Lin leukocytes, and CD161+CD56 NK cells (10), are all found at higher proportions in neonatal than adult blood. The similar proportions of CD56+high NK cells in neonatal and adult lymphocytes further detract from the hypothesis (7, 8) that they are immature progenitors to CD56+low NK cells.

As determined here and consistent with our previous reports (1, 10), immature CD3CD161+CD56 NK cells produce mainly type 2 cytokines (IL-5 and IL-13), relatively high GM-CSF and TNF-α levels, and no IFN-γ. Instead, mature CD3CD161+CD56+low NK cells produce low-to-null type 2 cytokines, low-to-high IFN-γ levels, and relatively moderate GM-CSF and TNF-α levels. CD56+high NK cells have a functional phenotype consistent with that of terminally differentiated, activated NK cells. On maximal, nonspecific stimulation, they produce IFN-γ levels similar to those produced by CD56+low NK cells; IL-10; GM-CSF, and TNF-α levels lower than those produced by the immature CD56 NK cells; and no type 2 cytokines. However, upon specific stimulation (e.g., IL-12, NKp46, K562, CD16) and like CD56+ NK cells cultured with IL-12, CD56+high NK cells produced higher levels of IFN-γ and TNF-α than CD56+low NK cells. The immediate responsiveness of CD56+high NK cells to IL-12 is consistent with the report of IFN-γ and TNF-β levels higher in culture supernatants of purified CD56+high than of CD56+low NK cells (32). The similar cytokine production by CD56+high and CD56+low NK cells in response to nonspecific, maximal stimulation before culture with IL-12 confirms that CD56+high NK cells are not constitutively capable of producing higher cytokine levels than the CD56+low cells but, having the potential to produce IFN-γ and TNF-α similar to that of the CD56+low cells, respond more rapidly to selected specific stimuli.

The present data confirm our previous reports (1, 10) that maximal IL-13 production, exclusive of IFN-γ, is restricted to immature CD56 NK cells, with almost no CD56+high or CD56+low NK cells producing IL-13 on maximal stimulation. They dismiss a previous conclusion that, based on analysis at the population level disregarding minor but detectable contaminant CD56 NK cells among the sorted CD56+high population, proposed that the CD56+high NK cells are the sole or major producers of IL-13 (32). Based on the analysis at the single-cell level reported here, the low (0–80 pg/ml) IL-13 amounts detected, in that report, in 72-h supernatants from CD56+high cells, compared with the 103-fold higher levels determined here to be produced in 15 h by CD56 NK cells, are accounted for by IL-13 produced by few contaminating CD56 cells. Instead, IL-10 is produced mostly by CD56+high NK cells at low levels within 18 h in response to IL-2 and IL-12, most prominently in apoptosing cells. After longer (8-day) culture with IL-12, most NK cells produced detectable IL-10 and concomitantly expressed increased CD56 levels. These results may explain why IL-10, accumulating in only minimal proportions of viable NK cells, is detectable in culture supernatants of CD56+high NK cells stimulated with IL-12 (32). The same occurs in T cells, with IL-10 and IL-13 detected exclusively in CD56+ and CD56 cells, respectively, and induced and lost, respectively, concomitant to induced CD56 expression (33).

The reportedly low cytotoxicity levels of the CD56+high NK cells (30) might reflect loss of cytotoxic granules in recently activated NK cells, rather than CD56+high NK cells being a separate specialized (32) or immature subset mediating cytokine-dependent functions and almost no cytotoxicity (8), in contrast to CD56+low cells mediating primarily cytotoxic functions. Absence of cytotoxic granules in many, but not all, CD56+high NK cells (30) and significant cytotoxicity levels mediated by these cells within 15 h of stimulation with IL-2 (8) are inconsistent with CD56+high NK cells being as developmentally immature as CD161+CD56 NK cells. Indeed, the latter do not express mature NK cell differentiation Ags and activating receptors and do not mediate granule exocytosis-dependent cytotoxicity on short term stimulation (34) but contain functional granules (35). Rather, this phenotype is consistent with degranulation of mature NK cells following recent triggering in vivo.

The observation that CD56+high NK cells are indistinguishable from mature NK cells activated by IL-15 (or IL-2), IL-12, and NK cell triggering stimuli (e.g., immune complex) supports the conclusion that these cells are generated in vivo after similar stimulation and explains their presence at sites of inflammation. CD56+high NK cells are found in the affected synovial tissues in rheumatoid arthritis (12), together with the monokines IL-12 (36) and IL-15 (37), and systemic immune complexes. CD62L expression on CD56+high NK cells explains their presence in lymph nodes, where they may produce IFN-γ and drive effective cellular immune responses (38). Interestingly, expression of CD162R, detectable on CD56+low but not CD56+high NK cells, is inversely regulated by IL-15 plus IL-12 compared with CD62L, a potential ligand for CD162R. The significance of this reciprocal expression is to be determined.

The anatomical site(s) for in vivo NK cell activation to become CD56+high remain(s) to be defined. Unlike T and B cells, minimal NK cell numbers are retained in the spleen, explaining their lower proportions in spleen vs peripheral blood in mice and humans. During murine CMV infection in mice, the number of splenic NK cells increases in correlation with and depending on IL-12 and IL-18 levels (39). In humans, transit through the spleen, or through capillaries and postcapillary venules containing monokines released as a result of inflammation from surrounding tissues, may induce mature NK cell activation and a CD56+high phenotype. The activated, CD56+high NK cells, induced to express CD62L, may consequently accumulate in lymph nodes, where they could affect the adaptive arm of immune responses via IFN-γ release. IL-10, produced as the cells die, likely limits the ongoing cellular response while promoting plasma cell generation, thus providing Ab for future protective immunity. The possibility will need to be considered that, among the CD56+low CD16+ NK cells, those that express CD62L (15% ± 7%, n = 11), possibly acquired in a fashion similar to that discussed above, may also transit to the lymph nodes and complete the transition to CD56+high cells under inflammatory conditions in which monokines are produced. Alternatively, the CD56+lowCD16+ NK cells not expressing CD62L may acquire the CD56highCD16−/low phenotype on interaction with immune complexes bound to the FcγR of the peripheral monocytes and be simultaneously triggered by monocyte-released monokines (IL-12, IL-15, IL-18). Although CD16 expression is down-modulated in response to IL-12 and immune complexes, the combined stimulation via CD16 engagement and monokines would result in CD62L expression, increased CD56 expression, among the other NK cell receptors described here, increased sensitivity to IL-2 and IL-12, and thus increased IFN-γ production and proliferative response. Simultaneous CD94 (if not other KIR) engagement, expected in most NK cells concomitant to CD16 engagement, may, however, limit their accumulation.

A major characteristic of CD56+high NK cells is their reduced expression of CD16. Consistent with the above scenario, and in agreement with reports of CD16 down-modulation and re-expression in response to IL-12 (6) or immune complex (40, 41) stimulation, this CD16−/low phenotype likely reflects, at least in part, occurred activation of mature CD16+high NK cells. CD16 surface expression slowly returned to low levels, but not necessarily on all cells, after engagement of CD16+ NK cells by immune complexes, but only after culture with IL-15, whereas IL-12 prevented CD16 re-expression. This contrasts the full CD16 recovery within 3 days in the absence of exogenous cytokines after CD16 removal by Pronase treatment (40). Analogous results of low-to-null CD16 expression by sorted CD56+highCD16 NK cells after culture with IL-15 and IL-12 confirmed that these cells behave like CD56+lowCD16+ cells induced to become CD56+highCD16. The CD16 cells could accumulate preferentially in inflammatory conditions where immune complexes and IL-12 are present, because they can respond to IL-15 and other monokines and acquire CD56+high phenotype, avoiding CD16-triggered activation-induced cell death.

In alternative, or addition to the above, at least part of the CD56+highCD16 NK cells may be cells that exited the bone marrow at a phenotypically incomplete developmental stage. NK cells might develop in the bone marrow to the stage when, like mature CD56+ NK cells, they express NKp46, CD2, and CD94; produce low levels of IFN-γ but not type 2 cytokines; and have cytotoxic potential, being activated upon activating receptor stimulation and inhibited upon CD94 engagement by MHC class I+ cells. A small fraction of these cells may exit the bone marrow before acquiring CD16. Although functionally mature and capable of acquiring a CD56+high phenotype, these cells may express or not KIR and CD8, which are acquired coincidentally to CD16 (1). If the conditions for complete phenotypic differentiation are not met in the periphery, these cells may never express these latest differentiation Ags. Consistent with the possibility that at least a portion of the CD56+high NK cells have indeed reached terminal development is that 25 ± 11% of them are CD8+, like the mature CD56+low NK cells (28 ± 13%, n = 7). Preliminary results indicate that KIR triggering on NK cells by mAb leads to loss of proliferative response to IL-15 (or IL-2) and death when the cells are preactivated by IL-15 (or IL-2), with little effect on nonactivated cells. Thus, basal activation by most (MHC class I+) cells in the periphery may allow better survival of peripheral CD16+KIR NK cells after CD16 down-modulation, allowing them to complete more easily the transition to the final, terminally differentiated, activated CD56+high stage. However, lack of KIR expression is unlikely a prerequisite to the CD56+high phenotype because the (variable) proportions of KIR3DL1+ cells in CD56+high NK cells (10 ± 7%) are significantly lower than those in the CD56+low cells (13 ± 4%) only within the CD16 population (2.5 ± 0.9%, p ≤ 0.05, n = 5).

On the basis of the definition of the latest stages of T cell development, we proposed that terminally differentiated, activated CD56+ T cells might be equivalent to regulatory TR1 cells (33). Like TR1 (42) and CD56+ T cells, CD56+high NK cells produce IL-10, express CD25, rapidly proliferate in response to IL-2, and have activated phenotype. TR1 cells are reportedly likely generated in vivo in nonpathological conditions in response to self-Ags. The same may be the case for CD56+high NK cells, in that the ligands for the triggering receptors expressed on most NK cells (e.g., MIC-A/B for NKG2D; Ref. 43) are surmised to be self proteins, either constitutively expressed or induced on stress. A thorough study of the factors (stromal and accessory cell interactions, soluble factors, temporal windows of opportunity) modulating CD16 and KIR expression during development in the bone marrow and periphery is needed to define which mechanism(s) are responsible for accumulation of CD16 and KIR CD56+high NK cells in the periphery and to guide rational approaches to the ex vivo generation and manipulation of these cells for possible therapeutic intervention.

We thank Elden Santos for help in starting this study; V. Berghella and the Staff of the Obstetrics and Gynecology Division, Thomas Jefferson Hospital, for providing the umbilical cord blood samples; and the personnel in the Kimmel Cancer Center Flow Cytometry Facility for assistance.

1

This work was supported in part by United States Public Health Service Grants AI055842 and CA56036.

3

Based on antigen density, the CD3CD161+ NK cells expressing relatively high or relatively lower CD56 levels are referred to here as CD56+high and CD56+low, respectively. However, for correctness, we do maintain the term CD56bright, referring to the results with the same cells in reports that have used this term.

4

Abbreviations used in this paper: CyC, CyChrome; EAET, sheep E treated with 2-aminoethylisothiouronium bromide; IFL, immunofluorescence; KIR, killer Ig-like receptor; +high, expressing relatively high levels; +low, expressing positive, but relatively low levels.

1
Loza, M. J., B. Perussia.
2001
. Final steps of natural killer cells maturation: a general model for type 1-type 2 differentiation?.
Nat. Immunol.
2
:
917
.
2
Chan, S. H., M. Kobayashi, D. Santoli, B. Perussia, G. Trinchieri.
1992
. Mechanisms of IFN-γ induction by natural killer cell stimulatory factor (NKSF/IL-12): role of transcription and mRNA stability in the synergistic interaction between NKSF and IL-2.
J. Immunol.
148
:
92
.
3
Perussia, B., S. H. Chan, A. D’Andrea, K. Tsuji, D. Santoli, M. Pospisil, D. Young, S. F. Wolf, G. Trinchieri.
1992
. Natural killer (NK) cell stimulatory factor or IL-12 has differential effects on the proliferation of TCR-αβ+, TCR-γδ+ T lymphocytes, and NK cells.
J. Immunol.
149
:
3495
.
4
Kobayashi, M., L. Fitz, M. Ryan, R. M. Hewick, S. C. Clark, S. Chan, R. Loudon, F. Sherman, B. Perussia, G. Trinchieri.
1989
. Identification and purification of natural killer cell stimulatory factor (NKSF), a cytokine with multiple biologic effects on human lymphocytes.
J. Exp. Med.
170
:
827
.
5
Chehimi, J., N. M. Valiante, A. D’Andrea, M. Rengaraju, Z. Rosado, M. Kobayashi, B. Perussia, S. F. Wolf, S. E. Starr, G. Trinchieri.
1993
. Enhancing effect of natural killer cell stimulatory factor (NKSF/interleukin-12) on cell-mediated cytotoxicity against tumor-derived and virus-infected cells.
Eur. J. Immunol.
23
:
1826
.
6
Azzoni, L., O. Zatsepina, B. Abebe, I. M. Bennett, P. Kanakaraj, B. Perussia.
1998
. Differential transcriptional regulation of CD161 and a novel gene, 197/15a, by IL-2, IL-15 and IL-12 in NK and T cells.
J. Immunol.
161
:
3493
.
7
Carson, W. E., T. A. Fehniger, M. A. Caligiuri.
1997
. CD56bright natural killer cell subsets: characterization of distinct functional responses to interleukin-2 and the c-kit ligand.
Eur. J. Immunol.
27
:
354
.
8
Nagler, A., L. L. Lanier, S. Cwirla, J. H. Phillips.
1989
. Comparative studies of human FcRIII-positive and negative natural killer cells.
J. Immunol.
143
:
3183
.
9
Loza, M. J., L. Zamai, L. Azzoni, E. Rosati, B. Perussia.
2002
. Expression of type 1 (IFN-γ) and type 2 (IL-13, IL-5) cytokines at distinct stages of NK cell differentiation from progenitor cells.
Blood
99
:
1273
.
10
Loza, M. J., S. P. Peters, J. G. Zangrilli, B. Perussia.
2002
. Distinction between IL-13+ and IFN-γ+ NK cells and regulation of their pool size by IL-4.
Eur. J. Immunol.
32
:
413
.
11
Caligiuri, M. A., A. Zmuidzinas, T. J. Manley, H. Levine, K. A. Smith, J. Ritz.
1990
. Functional consequences of interleukin 2 receptor expression on resting human lymphocytes: identification of a novel natural killer cell subset with high affinity receptors.
J. Exp. Med.
171
:
1509
.
12
Dalbeth, N., M. F. C. Callan.
2002
. A subset of natural killer cells is greatly expanded within inflamed joints.
Arthritis Rheum.
46
:
1763
.
13
Moore, T. A., M. Bennett, V. Kumar.
1996
. Murine natural killer cell differentiation; past, present, and future.
Immunol. Res.
15
:
151
.
14
Unkeless, J. C..
1979
. Characterization of a monoclonal antibody directed against mouse macrophage and lymphocyte Fc receptors.
J. Exp. Med.
150
:
580
.
15
Perussia, B., M. M. Tutt, W. Q. Qiu, W. A. Kuziel, P. W. Tucker, G. Trinchieri, M. Bennett, J. V. Ravetch, V. Kumar.
1989
. Murine natural killer cells express functional Fcγ receptor II encoded by the FcγRα gene.
J. Exp. Med.
170
:
73
.
16
Vance, R. E., D. M. Tanamachi, T. Hanke, D. H. Raulet.
1997
. Cloning of a mouse homolog of CD94 extends the family of C-type lectins on murine natural killer cells.
Eur. J. Immunol.
27
:
3236
.
17
Raulet, D. H., W. Held, I. Correa, J. R. Dorfman, M. F. Wu, L. Corral.
1997
. Specificity, tolerance and developmental regulation of natural killer cells defined by expression of class I-specific Ly49 receptors.
Immunol. Rev.
155
:
41
.
18
Colonna, M..
1997
. Specificity and function of immunoglobulin superfamily NK cell inhibitory and stimulatory receptors.
Immunol. Rev.
155
:
127
.
19
Bennett, I. M., B. Perussia.
1996
. Purification of peripheral blood natural killer cells. G. E. Jones, ed.
Methods in Molecular Medicine: Human Cell Culture Protocols
161
. Humana Press, Fotoma, NJ.
20
Perussia, B., G. Trinchieri, J. C. Cerottini.
1979
. Functional studies of Fc receptor-bearing human lymphocytes: effect of treatment with proteolytic enzymes.
J. Immunol.
123
:
681
.
21
Speiser, D. E., M. Colonna, M. Ayyoub, M. Cella, M. J. Pittet, P. Batard, D. Valmori, P. Guillame, D. Lienard, J. C. Cerottini, P. Romero.
2001
. The activatory receptor 2B4 is expressed in vivo by human CD8+ effector αβ T cells.
J. Immunol.
167
:
6165
.
22
Loza, M., J. S. Faust, B. Perussia.
2003
. Multiple color immunofluorescence for cytokine detection at the single cell level.
Mol. Biotechnol.
23
:
245
.
23
Loza, M. J., B. Perussia.
2002
. Peripheral immature CD2−/lo T cell development from type 2 to type 1 cytokine production.
J. Immunol.
169
:
3061
.
24
Mehta, B. A., V. C. Maino.
1997
. Simultaneous detection of DNA synthesis and cytokine production in staphylococcal enterotoxin B activated CD4+ T lymphocytes by flow cytometry.
J. Immunol. Methods
208
:
49
.
25
Holm, M., M. Thomsen, M. Hoyer, P. Hokland.
1998
. Optimization of a flow cytometric method for the simultaneous measurement of cell surface antigen, DNA content, and in vitro BrdU incorporation into normal and malignant hematopoietic cells.
Cytometry
32
:
28
.
26
Trotta, R., C. Puorro, M. Paroli, L. Azzoni, B. Abebe, L. C. Eisenlohr, B. Perussia.
1998
. Dependence of both spontaneous and antibody-dependent, granule exocytosis-mediated, NK cell cytotoxicity on extracellular signal-regulated kinases.
J. Immunol.
152
:
6648
.
27
Andre, P., O. Spertini, S. Guia, P. Rihet, F. Dignat-George, H. Brailly, J. Sampol, P. J. Anderson, E. Vivier.
2000
. Modification of P-selectin glycoprotein ligand-1 with a natural killer cell-restricted sulfated lactosamine creates an alternate ligand for L-selectin.
Proc. Natl. Acad. Sci. USA
97
:
3400
.
28
Vitale, M., C. Bottino, S. Sivori, L. Sanseverino, R. Castriconi, E. Marcenaro, R. Augugliaro, L. Moretta, A. Moretta.
1998
. NKp44, a novel triggering surface molecule specifically expressed by activated natural killer cells, is involved in non-major histocompatibility complex-restricted tumor cell lysis.
J. Exp. Med.
187
:
2065
.
29
Perussia, B., C. Ramoni, I. Anegon, M. C. Cuturi, J. Faust, G. Trinchieri.
1987
. Preferential proliferation of natural killer cells among peripheral blood mononuclear cells co-cultured with B lymphoblastoid cell lines.
Nat. Immunol. Cell Growth Regul.
6
:
171
.
30
Lanier, L. L., A. M. Le, C. I. Civin, M. R. Loken, J. H. Phillips.
1986
. The relationship of CD16 (Leu-11) and Leu-19 (NKH-1) antigen expression on human peripheral blood NK cells and cytotoxic T lymphocytes.
J. Immunol.
86
:
4480
.
31
Saeland, S., V. Duvert, C. Caux, D. Pandrau, C. Favre, A. Valle, I. Durand, P. Charbord, J. de Vries, J. Banchereau.
1992
. Distribution of surface-membrane molecules on bone marrow and cord blood CD34+ hematopoietic cells.
Exp. Hematol.
92
:
24
.
32
Cooper, M. A., T. A. Fehniger, S. C. Turner, S. Chen, B. A. Ghahert, 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
.
33
Loza, M. J., L. M. Metelitsa, B. Perussia.
2002
. NKT and T cells: coordinate regulation of NK-like phenotype and cytokine production.
Eur. J. Immunol.
32
:
3543
.
34
Zamai, L., M. Ahmad, I. M. Bennett, L. Azzoni, E. S. Alnemri, B. Perussia.
1998
. Natural killer (NK) cell-mediated cytotoxicity: differential use of TRAIL and Fas ligand by immature and mature primary human NK cells.
J. Exp. Med.
188
:
2375
.
35
Bennett, I. M., O. Zatsepina, L. Zamai, L. Azzoni, T. Mikheeva, B. Perussia.
1996
. Definition of a natural killer NKR-P1A+/CD56/CD16 functionally immature human NK cell subset that differentiates in vitro in the presence of interleukin 12.
J. Exp. Med.
184
:
1845
.
36
Morita, Y., M. Yamamura, K. Nishida, S. Harada, H. Okamoto, H. Inoue, Y. Ohmoto, R. L. Modlin, H. Makino.
1998
. Expression of interleukin-12 in synovial tissue from patients with rheumatoid arthritis.
Arthritis Rheum.
41
:
306
.
37
McInnes, I. B., J. al-Mughales, M. Field, B. P. Leung, F. P. Huang, R. Dixon, R. D. Sturrock, P. C. Wilkinson, F. Y. Liew.
1996
. The role of interleukin-15 in T-cell migration and activation in rheumatoid arthritis.
Nat. Med.
2
:
175
.
38
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 humanlymph nodes and are activated by T cell-derived IL-2: a potential link between adaptive and innate immunity.
Blood
101
:
3052
.
39
Andrews, D. M., A. A. Scalzo, W. M. Yokoyama, M. J. Smyth, M. A. Degli-Espositi.
2003
. Functional interactions between dendritic cells and NK cells during viral infection.
Nat. Immunol.
4
:
175
.
40
Perussia, B., G. Trinchieri, J. C. Cerottini.
1979
. Analysis of human lymphocyte subpopulations responsible for antibody-dependent and spontaneous cell-mediated cytotoxicity.
Transplant. Proc.
11
:
793
.
41
Perussia, B., O. Acuto, C. Terhorst, J. Faust, V. Fanning, R. Lazarus, G. Trinchieri.
1983
. Human natural killer cell analyzed by B73.1, a monoclonal antibody blocking FcR functions. II. Studies of B73.1 antibody-antigen interaction on the lymphocyte membrane..
J. Immunol.
130
:
2142
.
42
Roncarolo, M. G., R. Bacchetta, C. Bordignon, S. Narula, M. K. Levings.
2001
. Type 1 regulatory cells.
Immunol. Rev.
182
:
68
.
43
Bauer, S., J. G. V., A. Wu, J. H. Steinle, L. L. Phillips, L. L. Lanier, T. Spies.
1999
. Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA.
Science (Wash. DC)
285
:
727
.
44
Voss, S. D., J. Daley, J. Ritz, M. J. Robertson.
1998
. Participation of the CD94 receptor complex in costimulation of human natural killer cells.
J. Immunol.
160
:
1618
.
45
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
.
46
Robertson, M. J., R. J. Soiffer, S. F. Wolf, T. J. Manley, C. Donahue, D. Young, S. H. Hebermann, J. Ritz.
1992
. Response of human natural killer (NK) cells to NK cell stimulatory factor (NKSF): cytolytic activity and proliferation of NK cells are differentially regulated by NKSF.
J. Exp. Med.
75
:
779
.
47
Frey, M., N. B. Packianathan, T. A. Fehinger, M. E. Ross, W. C. Wang, C. C. Stewart, M. A. Caligiuri, S. S. Evans.
1998
. Differential expression and function of L-selectin on CD56bright and CD56dim natural killer cell subsets.
J. Immunol.
161
:
400
.