NK cells are the first line of defense against foreign cells, virally infected cells, and tumors. The mechanisms whereby NK cells accumulate in extralymphoid sites in response to pathogenic stimuli are not well understood. The L-selectin adhesion molecule (CD62L) plays a primary role in mediating the initial interaction of leukocytes with vascular endothelium, a crucial step in the extravasation of immune effector cells into tissues. In this report, we show L-selectin to be uniquely expressed on a subset of resting human NK cells (CD56bright). Notably, CD56bright NK cells expressed L-selectin at a higher density than all other peripheral blood leukocytes. NK activation by PMA, IL-2, IL-15, or TGF-β down-regulated L-selectin on the CD56bright subset, while increased L-selectin levels were observed in both the CD56bright and CD56dim NK subsets in response to IL-12, IL-10, or IFN-α. Moreover, CD56bright NK cells bound with high efficiency to physiologic L-selectin ligands on peripheral lymph node high endothelial venules (HEV). In sharp contrast, CD56dim NK cells adhered poorly to HEV and were predominantly L-selectin or expressed L-selectin only at low density. In CD56bright cells and a subpopulation of CD56dim cells, L-selectin ligation by mAb cross-linking activated lymphocyte function-associated Ag 1 (LFA-1), a second adhesion molecule required for leukocyte extravasation. LFA-1 was expressed on both NK subsets, although its density was constitutively higher on CD56dim cells. Taken together, evidence of differential expression of L-selectin and LFA-1 on CD56bright and CD56dim NK subsets strongly suggests unique migratory properties and functions of these cells during the early immune response to foreign pathogens.

The NK cell population of large granular lymphocytes comprises 10 to 15% of the peripheral blood lymphocyte population. NK cells are functionally defined by their ability to lyse malignant cells and virally infected cells without prior sensitization or MHC restriction and are thought to be important in the early control of many infectious pathogens (1, 2). These cells are phenotypically identified by the expression of CD56, an isoform of the human neural cell adhesion molecule, as well as CD16 (i.e., FcRγIII). Approximately 10% of NK cells express CD56 at high density (CD56bright CD16negative/dim), while the more abundant NK cell subset expresses low levels of CD56 (CD56dim CD16bright) (3, 4, 5).

CD56bright NK cells are unique among unstimulated human lymphocytes in their constitutive expression of the high affinity heterotrimeric IL-2Rαβγ (6, 7) and the c-kit tyrosine kinase receptor (8). Thus, picomolar concentrations of IL-2 initiate a strong proliferative response by CD56bright NK cells, which can be potentiated by IL-15 or c-kit ligand (5, 6, 7, 8, 9, 10, 11). In contrast, CD56dim NK cells do not express the c-kit receptor and express only the intermediate affinity IL-2Rβγ (7, 8, 10). Therefore, nanomolar concentrations of IL-2 are required to stimulate CD56dim NK cell cytotoxicity, although these cells fail to proliferate in response to IL-2 (5, 6, 7, 8, 9, 11). The CD56bright NK subset also appears to produce significantly greater amounts of IFN-γ, TNF-α, and granulocyte/macrophage-CSF than CD56dim NK cells following stimulation with IL-2 and IL-12, even at nanomolar concentrations of IL-2 (12). Evidence that CD56bright NK cells selectively respond to low concentrations of lymphokines and monokines has led to the hypothesis that this NK cell subset has a distinct role in the early stages of the immune response when these cytokines are present in limiting concentrations (3, 13).

NK cells are localized in high numbers in the spleen, where they have ready access to foreign pathogens. In addition, there is increasing evidence that NK cells can be actively recruited to peripheral tissues at sites of infection, injury, or tumor growth (14, 15, 16, 17, 18, 19). Leukocyte emigration from the blood into tissues is dependent on a multistep adhesion cascade (20, 21). The L-selectin adhesion molecule (CD62L) on the surface of B and T lymphocytes, neutrophils, and monocytes mediates the initial attachment and slow rolling of these cells along the luminal surface of specialized high endothelial venules (HEV),3 a crucial first step in the extravasation of immune effector cells at sites of inflammation or injury as well as in lymph nodes and Peyer’s patches. Firm adhesion and transendothelial migration of lymphocytes are dependent on the interaction of the leukocyte integrin LFA-1 with ICAM-1 and ICAM-2 on endothelial cells. Recent studies have shown that L-selectin engagement by mAb or a physiologic ligand, GlyCAM-1 (glycosylation-dependent cell adhesion molecule-1) up-regulates LFA-1 affinity in neutrophils and T lymphocytes (22, 23, 24), suggesting that the L-selectin pathway functions as an intravascular trigger to facilitate leukocyte extravasation. The α4β1 and α4β7 integrins have also been implicated in recruitment of lymphocytes to tissues through binding to endothelial counterreceptors, VCAM-1 and MAdCAM-1 (mucosal vascular addressin-1) (20, 21).

NK cells adhere to and transmigrate across endothelial cells in vitro at a faster rate than T cells (25, 26, 27), supporting the purported role of these immune effector cells during the early stages of an immune response before mobilization of Ag-restricted T lymphocytes. The contribution of L-selectin to NK cell homing has not been fully resolved, in part, because of the wide discrepancy among reports with respect to the constitutive level of L-selectin expression on circulating NK cells. In this regard, the reported frequency of peripheral blood NK cells expressing L-selectin has ranged from 10 to 60% (5, 28, 29, 30, 31, 32). Possible explanations for these conflicting results include 1) heterogeneous expression of L-selectin by NK cell subsets, as suggested previously (5, 29), 2) variability in L-selectin expression among different donor NK cell populations, and 3) modulation of L-selectin levels during lymphocyte isolation from peripheral blood. The relationship between L-selectin expression and function on CD56bright and CD56dim NK cell subsets has not been previously addressed. Moreover, although LFA-1 and α4β1/7 integrins have been strongly implicated in NK cell recruitment to tissues and NK cytolytic function (25, 27, 33, 34), the relative level of expression of these molecules by NK cell subsets has not been investigated.

In this report, we have examined the expression and function of leukocyte homing receptors on NK cell subsets. These studies demonstrate for the first time differential utilization of the L-selectin adhesion pathway by CD56bright and CD56dim NK cells. L-selectin was found to be expressed at the highest density on CD56bright NK cells relative to other peripheral blood leukocyte subsets, including CD56dim NK cells, T cells, B cells, neutrophils, and monocytes. Consistent with these observations, CD56bright NK cells exhibit highly efficient L-selectin-dependent adhesion to HEV in lymph node tissue sections when compared with the CD56dim NK subset. Moreover, mAb-induced L-selectin ligation was shown to increase LFA-1 function in the CD56bright NK cell subset as well as in a subset of CD56dim NK cells. Taken together, these results predict that the CD56bright NK cell subset has a selective advantage, compared with CD56dim NK cells, in extravasating across HEV via the L-selectin pathway.

PBMC were isolated from normal donor buffy coat leukocyte concentrates (American Red Cross, Buffalo, NY) by Ficoll/Hypaque centrifugation as described (8, 35, 36). Following removal of adherent cells, the PBL population was enriched for NK cells by depleting T cells, B cells, and the remaining monocytes using goat anti-mouse Ig-conjugated immunomagnetic beads (Per Septive Biosystems, Framinghom, MA), and a combination of murine mAb reactive against human CD3, CD4, and HLA-DR (8). NK-enriched populations typically contained 80 to 90% CD56+ cells. PBL and NK-enriched populations were cultured at a final concentration of 4 × 106 cells/ml in RPMI 1640 medium (Life Technologies, Grand Island, NY) containing 10% FCS (Life Technologies), 2 mM l-glutamine, 100 U/ml penicillin, and 50 μg/ml streptomycin.

Murine 3T3 cells stably transfected with human ICAM-1 and 3T3-Neo controls have been described previously (36, 37) and were generously provided by Dr. W. Muller (Rockefeller University, New York, NY). These cells were maintained in culture in DMEM (Life Technologies) with 10% FCS, 1 mg/ml G418 (Life Technologies), and 0.4 mg/ml hygromycin (Calbiochem, La Jolla, CA). The ICAM-1 transfectant population was >95% positive for ICAM-1. 3T3-Neo controls were >99% ICAM-1-negative.

The following recombinant human cytokines were used: IL-2 (Hoffmann-La Roche, Nutley, NJ; sp. act., 1.53 × 107 U/mg); IL-10 (Schering-Plough, Kenilworth, NJ; sp. act., 11.0 × 106 IRU/mg); IL-12 (Genetics Institute, Andover, MA; sp. act., 4.5 × 106 U/mg); IL-15 (Immunex Corp., Seattle, WA); TNF-α (Ashahi Chemical, Tokyo, Japan; sp. act., 2 × 103 U/mg); IFN-α (Shering-Plough; sp. act., 2.2 × 108 IU/mg); and TGF-β2 (R&D Systems, Minneapolis, MN). The L-selectin-specific mAb DREG-56 (38) was a generous gift from Dr. E. Butcher (Stanford University, Stanford, CA). Anti-β1-integrin mAb was kindly provided by Dr. Richard Bankert (Roswell Park Cancer Institute, Buffalo, NY). The anti-HLA-DR mAb has been described previously (8). The following mAb were obtained commercially: L-selectin-specific mAb Leu-8-FITC, anti-CD3 (Leu-4)-PE, anti-CD16-PE and FITC- or phycoerythrin (PE)-conjugated isotype-matched control mAb (Becton Dickinson Immunocytometry, San Jose, CA); anti-CD56-PE and L-selectin-specific TQ1 mAb (Coulter Immunology, Hialeah, FL); FITC-labeled anti-human LFA-1α (CD11a)-FITC (Endogen, Boston, MA); unconjugated anti-LFA-1 (TS1/22), anti-CD3 (OKT3) and anti-CD4 (OKT4) (American Type Culture Collection, Manassas, VA); anti-CD49d-FITC (α4-integrin) (Immunotech, Westbrook, ME); and anti-CD64-TriColor and anti-CD3-TriColor (Caltag, Burlingame, CA). Goat-anti-mouse IgG-RITC (rhodamine isothiocyanate) was obtained from Southern Biotechnology (Birmingham, AL) and sheep anti-mouse IgG was from Organon Teknika (West Chester, PA). PMA and fibronectin were purchased from Sigma Chemical (St. Louis, MO).

Enriched PBL or NK cell populations were analyzed for simultaneous two-color or three-color immunofluorescence expression of adhesion molecules (L-selectin, LFA-1, and α4- or β1-integrins) and either CD56 (NK cells) or CD3 (T cells) as described (8, 35). For analysis of leukocyte-enriched whole blood, samples were washed twice with PBS containing 10 U/ml heparin, incubated with goat serum for 10 min at 4°C to block Fc receptor sites, and then incubated for 15 min simultaneously with FITC-labeled mAb specific for adhesion molecules and PE-labeled lymphocyte-specific mAb. RBC were lysed (150 mM ammonium chloride, 10 mM potassium bicarbonate, 97 μM EDTA) before fixation in 1% formaldehyde/PBS. Background fluorescence was determined on cells stained with fluorochrome-labeled isotype-matched nonreactive mAb. A total of 5000 events were collected for each leukocyte population on a FACScan (Becton Dickinson) in the Roswell Park Flow Cytometry facility, and analysis was performed on WinList 1.0 (Verity Software House, Topsham, ME). T cells or NK cell subsets were analyzed using a lymphocyte gate; the FL-2 amplifier gain was adjusted to trigger selectively on either CD3+, CD56dim, or CD56bright cells. Monocytes and neutrophils were discriminated by forward and side scatter parameters and on the basis of CD16 (neutrophils) or CD64 (monocytes) fluorescence.

Analysis of L-selectin-dependent binding of lymphocytes to PPME, the phosphomonoester core from Hansenula hostii phosphomannan, was performed as described (35, 39, 40). Enriched NK cell populations were initially stained with CD56-PE, washed once in RPMI 1640 medium, resuspended at 5 × 106 cells/ml, and then incubated for 15 min at 4°C either in medium alone or with the L-selectin-specific TQ1 blocking mAb (10 μg/ml). Without washing, cells were then incubated with a 1:200 dilution of fluorescein-conjugated PPME (generous gift of Dr. L. Stoolman, University of Michigan, Ann Arbor, MI) for 30 min at 4°C and analyzed immediately by flow cytometry.

Lymphocyte binding to HEV was assessed essentially as described with the following modifications (35, 36, 41). Enriched NK cell populations were initially stained at 4°C with anti-CD56-PE primary mAb and goat anti-mouse IgG-RITC secondary Ab to permit analysis of NK cell subset adhesion to HEV. In pilot experiments, it was determined that identical numbers of cells were bound to HEV regardless of whether CD56 was fluorescently labeled on NK cells. NK-enriched populations or PBL were then washed and resuspended at 5 × 107 cells/ml in RPMI 1640 medium containing 10% FCS and incubated for 30 min at room temperature with or without saturating amounts of L-selectin-specific blocking mAb (DREG-56). Lymphocytes (5 × 106 cells in 100 μl) were overlaid onto 12-μm-thick cryosections of BALB/c lymph nodes mounted on glass slides. Previous studies have established that L-selectin binding specificity is maintained during assay of human lymphocyte adhesion to mouse lymph node HEV (38). Slides were rotated at 112 rpm (Labline Instrument, Melrose Park, IL) at 4°C for 30 min, and nonadherent cells were removed by gentle washing in cold PBS. Slides were fixed vertically in 3% formaldehyde (Ernest F. Fullam, Catham, NY)/PBS for 1 h, then either rinsed in PBS (in the case of NK-enriched samples) or permeabilized in 70% ethanol and stained with 0.5% toluidine/absolute ethanol (PBL samples). Adhesion of CD56bright and CD56dim NK cells was quantified under double blind conditions using an Olympus BH2-RFL fluorescence microscope (Olympus Optical, Tokyo, Japan). PBL adhesion to HEV was evaluated by light microscopy. A total of 300 to 500 HEV were examined, and data are expressed as the mean number of lymphocytes bound per HEV ± SD; each sample was quantified in triplicate.

LFA-1-dependent adhesion of PBL to ICAM-1-transfected 3T3-fibroblasts was evaluated essentially as described (36, 37), with the following modifications to allow the discrimination of adherent NK cell subsets. Immediately before initiation of the assay, 2 × 106 enriched-NK cells were stained with anti-CD56 mAb and goat anti-mouse IgG-RITC secondary Abs. L-selectin cross-linking was performed by incubating cells with 5 μg/ml of DREG-56 mAb (30 min, room temperature), followed by 10 μg/ml sheep anti-mouse IgG (2 h at 37°C). In parallel samples, cells were incubated either in RPMI 1640 medium/10% FCS alone or in the presence of 100 ng/ml PMA to activate LFA-1 function. Cells were then washed and incubated in the absence or presence of saturating concentrations of anti-LFA-1 blocking mAb (TS1/22) for 30 min at 37°C. A total of 106 cells (1 ml) were added to confluent monolayers of ICAM-1-transfected 3T3 cells or Neo transfectants grown on fibronectin-coated tissue culture chamber slides (Miles Laboratories, Naperville, IL). Following incubation for 30 min at 37°C, nonadherent cells were removed by gentle washing, and adherent cells were fixed in 3% formaldehyde. The number of adherent CD56+ NK cells was quantified in five low power fields (20×) in replicate by fluorescence microscopy. Since the CD56 Ag became capped on the cell surface under these conditions, it was not possible to visually distinguish CD56bright from CD56dim cells by conventional fluorescence microscopy. Therefore, the proportion of CD56bright:CD56dim NK cells was determined by quantitating the fluorescence intensity of single cells by confocal fluorescence microscopy (MRC 600 CSLM; Bio-Rad, Palo Alto, CA). A total of 100 cells was examined at random from each condition, and total fluorescence was determined by electronic integration of capped regions of fluorescence using COMOS 6.03 (Bio-Rad).

A comparative analysis of adhesion molecule expression on NK cell subsets was performed by multiparameter flow cytometry using unseparated whole blood to avoid modulation of the cell surface levels of these molecules during the isolation of leukocyte subpopulations. Several striking differences were detected in L-selectin and LFA-1 expression by CD56bright and CD56dim peripheral blood NK cells that are suggestive of a differential ability of these cells to extravasate across HEV in lymphoid tissues and inflamed tissues.

In 20 donors examined, L-selectin was universally found to be highly expressed on the CD56bright NK cell subset (Figs. 1 and 2). Notably, the density of L-selectin on this NK cell subset was markedly higher than on CD56dim NK cells, T cells, B cells (not shown), neutrophils, or monocytes. In sharp contrast, in the majority of donors examined, L-selectin was expressed only on a minor proportion of CD56dim NK cells (<25%) at a density considerably lower than on CD56bright NK cells (Fig. 1 and 2,A). However, in ∼30% of donors, a different pattern of L-selectin expression was observed on CD56dim NK cells, indicating that considerable variability exists among donors with regard to the constitutive levels of L-selectin on this NK cell subset. In these individuals, L-selectin was detected on up to 50% of CD56dim NK cells, albeit at a low density compared with CD56 bright NK cells or T cells (Fig. 2 B). Discrimination of CD56+ NK cells from the CD56+CD3+ T cell subset that is present in peripheral blood (42) by three-color flow cytometric analysis further revealed that both L-selectin and CD56 were consistently expressed at a lower density on this T cell subpopulation than on CD56dim NK cells (n = 15 donors, data not shown). L-selectin levels on either NK subset remained unchanged during their isolation from whole blood or after culture for 1 day (not shown).

FIGURE 1.

Differential expression of adhesion molecules by peripheral blood leukocytes. Freshly isolated whole blood was stained with FITC-conjugated anti-adhesion molecule mAb (shaded histograms) or isotype-matched control mAb (white histogram) and analyzed by flow cytometry. For analysis of NK cells or T lymphocytes, samples were simultaneously stained with PE-labeled anti-CD56 mAb (detecting CD56bright and CD56dim NK subsets) or anti-CD3 (detecting T cells), and data were gated on PE-positive cells as described (8, 35). Monocytes and neutrophils were analyzed on the basis of forward- and side-scatter parameters and CD16 (neutrophil) or CD64 (monocyte) fluorescence. The frequency of leukocyte subtypes expressing the indicated adhesion molecules is shown. For comparative analysis, arrows indicate the peak fluorescence of mAb staining of CD56bright NK cells.

FIGURE 1.

Differential expression of adhesion molecules by peripheral blood leukocytes. Freshly isolated whole blood was stained with FITC-conjugated anti-adhesion molecule mAb (shaded histograms) or isotype-matched control mAb (white histogram) and analyzed by flow cytometry. For analysis of NK cells or T lymphocytes, samples were simultaneously stained with PE-labeled anti-CD56 mAb (detecting CD56bright and CD56dim NK subsets) or anti-CD3 (detecting T cells), and data were gated on PE-positive cells as described (8, 35). Monocytes and neutrophils were analyzed on the basis of forward- and side-scatter parameters and CD16 (neutrophil) or CD64 (monocyte) fluorescence. The frequency of leukocyte subtypes expressing the indicated adhesion molecules is shown. For comparative analysis, arrows indicate the peak fluorescence of mAb staining of CD56bright NK cells.

Close modal
FIGURE 2.

Heterogeneous L-selectin expression on NK cell subsets. Enriched populations of NK cells were stained with FITC-conjugated anti-Leu-8 (detecting L-selectin) and PE-labeled anti-CD56 (detecting CD56bright and CD56dim NK subsets) and analyzed by flow cytometry. In donor A, L-selectin was expressed on 86% of CD56bright and 16% of CD56dim NK cells. This phenotype is representative of approximately 70% of donors examined. In donor B, which is representative of 25 to 30% of the donors, L-selectin was expressed on 98% of CD56bright and 46% of CD56dim NK cells.

FIGURE 2.

Heterogeneous L-selectin expression on NK cell subsets. Enriched populations of NK cells were stained with FITC-conjugated anti-Leu-8 (detecting L-selectin) and PE-labeled anti-CD56 (detecting CD56bright and CD56dim NK subsets) and analyzed by flow cytometry. In donor A, L-selectin was expressed on 86% of CD56bright and 16% of CD56dim NK cells. This phenotype is representative of approximately 70% of donors examined. In donor B, which is representative of 25 to 30% of the donors, L-selectin was expressed on 98% of CD56bright and 46% of CD56dim NK cells.

Close modal

LFA-1 was also found to be differentially distributed on NK cell subsets. In all of the donors examined, LFA-1 was expressed on the majority of CD56dim NK cells (>90%) at significantly higher levels than on the CD56bright NK cell subset or T lymphocytes (Fig. 1). These results are consistent with the high expression of LFA-1 previously described on total NK cell populations (34). The density of LFA-1 detected on CD56bright NK cells was comparable with T lymphocytes and neutrophils. No difference was observed in the expression of the α4-integrin on NK cell subsets. β1-integrins were consistently found to be expressed at a moderately higher density on the CD56bright NK subset when compared with CD56dim cells.

L-selectin is rapidly shed from the plasma membrane of T and B lymphocytes, neutrophils, and monocytes in direct response to stimulation of protein kinase C by phorbol esters (i.e., PMA), although the kinetics vary in different leukocyte populations (20, 35, 38). Therefore, it was of interest to examine the effects of PMA on the kinetics of L-selectin down-regulation in CD56bright and CD56dim NK cells. Following incubation of leukocyte-enriched whole blood with PMA, a gradual decrease in L-selectin surface levels occurred over the course of 1 h in the CD56bright and CD56dim NK cell subsets (Fig. 3). The kinetics of L-selectin down-regulation in NK cells closely paralleled the PMA response of T lymphocytes (Fig. 3) and B lymphocytes (data not shown). These observations contrast sharply with the rapid loss of L-selectin detected within 1 min following PMA activation of neutrophils and monocytes.

FIGURE 3.

Kinetics of L-selectin down-regulation in leukocyte subsets following PMA stimulation. Whole blood was incubated for the indicated time intervals in the presence of 50 ng/ml PMA, then stained with FITC-conjugated anti-L-selectin mAb (anti-Leu-8) and PE-labeled anti-CD56 mAb (detecting CD56bright and CD56dim NK subsets) or anti-CD3 mAb (detecting T cells) and analyzed by flow cytometry. Monocytes and neutrophils were gated based on forward- and side-scatter parameters. Data are representative of three independent experiments.

FIGURE 3.

Kinetics of L-selectin down-regulation in leukocyte subsets following PMA stimulation. Whole blood was incubated for the indicated time intervals in the presence of 50 ng/ml PMA, then stained with FITC-conjugated anti-L-selectin mAb (anti-Leu-8) and PE-labeled anti-CD56 mAb (detecting CD56bright and CD56dim NK subsets) or anti-CD3 mAb (detecting T cells) and analyzed by flow cytometry. Monocytes and neutrophils were gated based on forward- and side-scatter parameters. Data are representative of three independent experiments.

Close modal

NK cell expansion and function is regulated by numerous immunomodulatory cytokines including IL-2, IL-10, IL-12, IL-15, and TGF-β (1, 2, 6, 7, 8, 9, 11, 43, 44). Lymphocyte activation in response to IL-2, mitogens, or CD3 cross-linking has been shown to down-regulate L-selectin cell surface levels on T cells and total NK cell populations (28, 30, 32, 35, 45), although the effects of IL-2 or other cytokines on L-selectin expression by CD56bright and CD56dim NK cell subsets has not been previously examined. Moreover, the effect of proinflammatory cytokines that are present in tissues during acute inflammation or injury (e.g., TNF-α, IFN-α) on L-selectin expression by NK cell subsets is not known. To address these issues, enriched populations of NK cells were cultured for 24 h in the presence of recombinant cytokines, and the expression of L-selectin on NK cell subsets was determined by two-color flow cytometric analysis (Fig. 4). IL-12 caused a marked increase in L-selectin expression on both CD56bright and CD56 dim NK cell subsets, as indicated by an increase in both L-selectin density (i.e., indicated by the mean channel fluorescence) and the frequency of L-selectin-positive cells. A moderate increase in L-selectin density was also observed in response to IL-10 (Fig. 4) and IFN-α in both NK cell subsets (data not shown). In contrast, L-selectin levels were markedly down-regulated in the CD56bright NK cell subset following culture with IL-2, IL-15, or TGF-β. Although IL-2, IL-15, or TGF-β did not cause a major change in the CD56dim subset in terms of the frequency of L-selectin-positive cells or L-selectin density, (as would be expected since these values are heavily weighted by the large L-selectin-negative population), a moderate decrease in fluorescence intensity was detected in response to these cytokines (note the downward shift on the log scale). TNF-α had essentially no effect on L-selectin expression by NK cell subsets.

FIGURE 4.

Modulation of L-selectin levels on CD56bright and CD56dim NK subsets following stimulation with immunomodulatory cytokines. Enriched NK cells were cultured for 24 h in medium alone or in the presence of 15 U/ml IL-12, 50 ng/ml IL-10, 1 nM IL-2, 100 ng/ml IL-15, 2 ng/ml TGF-β, or 1000 U/ml TNF-α. Cells were then stained with PE-labeled anti-CD56 and FITC-labeled anti-L-selectin mAb (anti-Leu-8) and analyzed by flow cytometry. L-selectin expression on CD56bright and CD56dim NK cells is indicated by black histograms. The frequency of L-selectin-positive NK cells and the mean channel fluorescence (MCF) of the total NK population are also shown. The shaded region represents background fluorescence of cells stained with a FITC-labeled isotype-matched control mAb. Data are representative of three independent experiments.

FIGURE 4.

Modulation of L-selectin levels on CD56bright and CD56dim NK subsets following stimulation with immunomodulatory cytokines. Enriched NK cells were cultured for 24 h in medium alone or in the presence of 15 U/ml IL-12, 50 ng/ml IL-10, 1 nM IL-2, 100 ng/ml IL-15, 2 ng/ml TGF-β, or 1000 U/ml TNF-α. Cells were then stained with PE-labeled anti-CD56 and FITC-labeled anti-L-selectin mAb (anti-Leu-8) and analyzed by flow cytometry. L-selectin expression on CD56bright and CD56dim NK cells is indicated by black histograms. The frequency of L-selectin-positive NK cells and the mean channel fluorescence (MCF) of the total NK population are also shown. The shaded region represents background fluorescence of cells stained with a FITC-labeled isotype-matched control mAb. Data are representative of three independent experiments.

Close modal

Studies were performed to determine whether the high density of L-selectin detected on CD56bright NK cells corresponded with efficient binding of this NK cell subset to soluble L-selectin carbohydrate ligands, i.e, PPME, and to physiologic ligands expressed on peripheral lymph node HEV. These functional assays are highly predictive of the L-selectin-dependent homing potential of leukocytes (20, 21). For these studies, an enriched population of NK cells (>90% CD56+CD3) was used in which 91% of CD56bright NK cells expressed high levels of L-selectin and 20% of CD56dim cells expressed this adhesion molecule at low density (i.e., the NK cells from this donor were phenotypically similar to the NK cell subsets shown in Fig. 1). The data shown in Figure 5 indicate that CD56bright NK cells bound high levels of PPME in an L-selectin-dependent manner that was strongly inhibited by the L-selectin-specific TQ1 mAb, whereas the majority of the CD56dim NK population bound low to non-detectable amounts of PPME.

FIGURE 5.

Preferential L-selectin-dependent binding of CD56bright NK cells to PPME. Enriched NK cells were initially stained with PE-labeled anti-CD56, then incubated for 15 min at 4°C either in medium alone or with the L-selectin-specific TQ1 blocking mAb (10 μg/ml) before the addition of FITC-conjugated PPME. Fluorescence intensity of PE-labeled cells was determined by flow cytometry with logarithmic amplification. The frequency of cells in each quadrant is indicated. Cells from the same donor were stained with anti-Leu-8-FITC; CD56bright and CD56dim populations, respectively, were found to be 91% and 20% L-selectin+ (not shown). Data are representative of four independent experiments.

FIGURE 5.

Preferential L-selectin-dependent binding of CD56bright NK cells to PPME. Enriched NK cells were initially stained with PE-labeled anti-CD56, then incubated for 15 min at 4°C either in medium alone or with the L-selectin-specific TQ1 blocking mAb (10 μg/ml) before the addition of FITC-conjugated PPME. Fluorescence intensity of PE-labeled cells was determined by flow cytometry with logarithmic amplification. The frequency of cells in each quadrant is indicated. Cells from the same donor were stained with anti-Leu-8-FITC; CD56bright and CD56dim populations, respectively, were found to be 91% and 20% L-selectin+ (not shown). Data are representative of four independent experiments.

Close modal

Adhesion of NK cell subsets to physiologic L-selectin ligands on lymph node HEV was assessed in an in vitro adhesion assay under nonstatic conditions to simulate blood flow dynamics in vivo. In the representative experiment shown in Figure 6, an enriched population of NK cells comprised of 75% CD56dim CD3negative cells and 10% CD56bright cells was initially labeled at 4°C with anti-CD56 primary mAb and goat-anti-mouse IgG-RITC to allow the discrimination of these NK cell subsets in the adhesion assay. L-selectin was expressed on >98% of the CD56bright cells and on <25% of the CD56 dim NK cells of this particular donor. Cells were then overlaid onto frozen sections of murine lymph nodes, and adherence of CD56bright and CD56dim NK cells to specialized HEV within tissues was quantified by fluorescence microscopy. Despite the fact that CD56bright NK cells were present in low numbers in the initial NK-enriched population, this NK cell subset represented a high proportion (>90%) of NK cells bound to HEV in a manner that could be inhibited specifically by the L-selectin-blocking mAb, DREG-56. In contrast, only low levels of L-selectin-dependent adhesion of CD56dim NK cells to lymph node HEV were detected. Parallel analysis of PBL adhesion to HEV indicated that comparable numbers of PBL and CD56bright NK cells bound to HEV (Fig. 6), further indicating that this numerically limited NK cell subset is capable of highly efficient adhesion to HEV under shear conditions. Immunofluorescence microscopy revealed that >70% of bound PBL were CD3+ T cells (data not shown).

FIGURE 6.

Preferential L-selectin-dependent binding of CD56bright NK cells to peripheral lymph node HEV. NK-enriched cells were initially stained with anti-CD56 mAb and goat anti-mouse IgG-RITC secondary Ab to permit discrimination of NK cell subsets. The initial NK-enriched population was comprised of 75% CD56dim cells and 10% CD56bright cells (not shown). Moreover, flow cytometric analysis indicated that L-selectin was expressed on >98% of CD56bright cells and on <25% of CD56dim cells (not shown). PBL or NK-enriched cells were incubated for 30 min with the L-selectin function blocking mAb, DREG-56 (50 μg/ml), or with an isotype-matched control mAb, and lymphocyte adhesion to cryosections of BALB/c peripheral lymph node HEV was assayed as described in Materials and Methods. Error bars denote SD of three replicate sections. Data are representative of three independent experiments.

FIGURE 6.

Preferential L-selectin-dependent binding of CD56bright NK cells to peripheral lymph node HEV. NK-enriched cells were initially stained with anti-CD56 mAb and goat anti-mouse IgG-RITC secondary Ab to permit discrimination of NK cell subsets. The initial NK-enriched population was comprised of 75% CD56dim cells and 10% CD56bright cells (not shown). Moreover, flow cytometric analysis indicated that L-selectin was expressed on >98% of CD56bright cells and on <25% of CD56dim cells (not shown). PBL or NK-enriched cells were incubated for 30 min with the L-selectin function blocking mAb, DREG-56 (50 μg/ml), or with an isotype-matched control mAb, and lymphocyte adhesion to cryosections of BALB/c peripheral lymph node HEV was assayed as described in Materials and Methods. Error bars denote SD of three replicate sections. Data are representative of three independent experiments.

Close modal

Previous studies have demonstrated that Ab-mediated cross-linking of L-selectin on neutrophils and T lymphocytes up-regulates the function of β2-integrins (22, 23, 24, 46), suggesting that cooperation between distinct leukocyte-endothelial cell adhesion events facilitate leukocyte extravasation. To address whether L-selectin also stimulates NK cell adhesion through a β2-integrin pathway, an enriched population of NK cells (>90% CD56+CD3) was initially labeled with anti-CD56 mAb and goat-anti-mouse IgG-RITC secondary Ab to allow identification of NK cell subsets. The frequency of CD56bright and CD56dim cells expressing L-selectin in the representative experiment shown in Figure 7 and Table I was 94% and 18%, respectively, while the ratio of CD56bright to CD56dim cells that expressed L-selectin was 1:2.4. Following stimulation of cells under various conditions, LFA-1-dependent adhesion of fluorochrome-labeled CD56+. NK cells to monolayers of ICAM-1-transfected 3T3 cells or Neo-transfectant control cells was quantified by fluorescence microscopy. The CD56bright and CD56dim NK cell subsets could not be distinguished visually by conventional fluorescence microscopy, since a majority of the CD56 Ag became capped during the adhesion assay. Therefore, the fluorescence intensity of individual CD56-positive cells was quantified by confocal fluorescence microscopy and computer assisted digital analysis (Table I).

FIGURE 7.

mAb-induced L-selectin cross-linking activates LFA-1-mediated adhesion of CD56bright and CD56dim NK cells to 3T3-ICAM-1 transfectants. Enriched NK cells were initially stained with anti-CD56 mAb and goat anti-mouse IgG-RITC secondary Ab to permit the identification of NK cell subsets. The frequency of CD56bright and CD56dim cells expressing L-selectin in this experiment was 94% and 18%, respectively (not shown). After CD56 labeling, cells were incubated for 2 h at 37°C either in medium alone (untreated), 100 ng/ml PMA, or with 5 μg/ml DREG-56 in combination with sheep-anti-mouse IgG to cross-link L-selectin. Where indicated, lymphocytes were treated for 30 min at 37°C with anti-LFA-1 function blocking mAb (TS1/22) or control mAb immediately before the adhesion assay. Analysis of the adhesion of fluorescent-labeled CD56+ NK cells to 3T3-ICAM-1 transfectants or 3T3-Neo was performed as described in the Materials and Methods. Error bars denote SD of replicate microscopic fields. Data are representative of three independent experiments. The relative proportion of CD56bright and CD56dim cells in the initial NK cell population and in adherent populations is shown in Table I for this experiment.

FIGURE 7.

mAb-induced L-selectin cross-linking activates LFA-1-mediated adhesion of CD56bright and CD56dim NK cells to 3T3-ICAM-1 transfectants. Enriched NK cells were initially stained with anti-CD56 mAb and goat anti-mouse IgG-RITC secondary Ab to permit the identification of NK cell subsets. The frequency of CD56bright and CD56dim cells expressing L-selectin in this experiment was 94% and 18%, respectively (not shown). After CD56 labeling, cells were incubated for 2 h at 37°C either in medium alone (untreated), 100 ng/ml PMA, or with 5 μg/ml DREG-56 in combination with sheep-anti-mouse IgG to cross-link L-selectin. Where indicated, lymphocytes were treated for 30 min at 37°C with anti-LFA-1 function blocking mAb (TS1/22) or control mAb immediately before the adhesion assay. Analysis of the adhesion of fluorescent-labeled CD56+ NK cells to 3T3-ICAM-1 transfectants or 3T3-Neo was performed as described in the Materials and Methods. Error bars denote SD of replicate microscopic fields. Data are representative of three independent experiments. The relative proportion of CD56bright and CD56dim cells in the initial NK cell population and in adherent populations is shown in Table I for this experiment.

Close modal
Table I.

Analysis of CD56bright and CD56dim NK cell adhesion to ICAM-1-3T3 transfectantsa

No. NK Subsets/100 Total CD56+ NK CellsRatio of CD56bright:CD56dim
CD56bright NK cellsCD56dim NK cells
Enriched NK population 12 86 1:8.3 
Adherent NK cells    
Untreated control 12 86 1:8.3 
PMA stimulated 14 84 1:7.1 
DREG-56 cross-linked 37 63 1:2.7 
No. NK Subsets/100 Total CD56+ NK CellsRatio of CD56bright:CD56dim
CD56bright NK cellsCD56dim NK cells
Enriched NK population 12 86 1:8.3 
Adherent NK cells    
Untreated control 12 86 1:8.3 
PMA stimulated 14 84 1:7.1 
DREG-56 cross-linked 37 63 1:2.7 
a

Adhesion of fluorescent-labeled CD56+ NK cells to 3T3-ICAM-1 transfectants was assayed as described in Figure 7 and in Materials and Methods. The relative proportion of CD56bright and CD56dim cells in the initial enriched NK cell population and in adherent populations was determined by quantifying the fluorescence intensity of a total of 100 CD56+ cells by confocal fluorescence microscopy and computer-assisted digital analysis. The relative number of CD56+ NK cells that adhered to 3T3-ICAM-1 transfectants in this experiment is shown in Figure 7.

In the absence of stimulation, NK cell adhesion to ICAM-1 transfectants was only marginally above the background level observed in Neo transfectants (Fig. 7), consistent with the reported constitutive low affinity of LFA-1 on circulating leukocytes (20, 21). Thus, initial CD56 labeling did not directly activate LFA-1 function. Following stimulation with PMA, a known activator of LFA-1 affinity (20, 21), a threefold increase in LFA-1-dependent NK cell adhesion to ICAM-1 transfectants was detected. PMA-stimulated NK cell adhesion to ICAM-1 transfectants was strongly inhibited by an LFA-1 function-inhibiting mAb. Notably, despite higher expression of LFA-1 by CD56dim cells (Fig. 1), the proportions of CD56bright and CD56dim cells present in the initial NK-enriched population and in PMA-activated adherent NK cells were similar (Table I), indicating that preferential activation of LFA-1 activity does not occur in either NK cell subset in response to phorbol ester stimulation. L-selectin cross-linking via DREG-56 mAb stimulated a threefold increase in LFA-1-dependent adhesion of NK cells to ICAM-1 transfectants (Fig. 7). While L-selectin ligation increased LFA-1-dependent adhesion of both CD56bright and CD56dim NK cells (combined data in Fig. 7 and Table I), considerable enrichment of CD56bright NK cells in the adherent population was observed (i.e., CD56bright cells comprised 37% of adherent NK cells in DREG-56-stimulated cultures compared with 12% of the initial NK-enriched population), signifying highly efficient L-selectin-mediated activation of LFA-1 function in this numerically rare NK cell subset. Notably, the proportion of CD56bright cells to CD56dim cells in the adherent populations (i.e., 1:2.7, Table I) closely paralleled the ratio of these subsets that expressed L-selectin in the initial populations (i.e., 1:2.4), suggesting that LFA-1 activation by DREG-56 mAb occurred selectively in L-selectin+ NK cells.

The obligate role of NK cell-produced cytokines in the early phase of immune responses to various pathogens before mobilization of Ag-specific T cells is well established (1, 2). This report addresses the potential mechanisms by which NK cells are recruited from the blood to tissues in response to pathogenic stimuli. Localization of NK cells in tissues serves to tightly regulate and promote the immune response by ensuring that NK cells receive activation signals for cytokine production within infected tissues. Moreover, local release of NK cell-derived cytokines focuses the immune response to affected tissues, obviating the need for high systemic concentrations of proinflammatory cytokines.

Data presented in this report demonstrate for the first time a direct correlation between L-selectin expression and function in CD56bright and CD56dim NK subsets, suggesting distinct migratory pathways of these leukocytes. L-selectin was shown to be highly expressed on resting CD56bright NK cells, at a greater density than on all other peripheral blood leukocytes. These data are consistent with the high frequency of L-selectin expression noted on CD56bright (CD16negative or CD16dim NK) cells in two previous reports (5, 29). In the current study, the CD56bright subset was further shown to bind with high efficiency to physiologic L-selectin ligands on specialized HEV. In sharp contrast, CD56dim NK cells bound poorly to HEV and were predominantly L-selectin-negative or expressed L-selectin only at low density. Based on the significant enrichment of CD56bright cells in the HEV-adherent NK fraction, of which >90% represented the CD56bright subset, it appears that the low density of L-selectin on CD56dim cells is not sufficient to optimally mediate adhesion to HEV under physiologic shear conditions. Taken together, these results suggest a model whereby L-selectin preferentially facilitates trafficking of the CD56bright NK cell subset from the blood into secondary sites during the early phase of infection or injury.

While L-selectin was found to be consistently expressed at high levels on the CD56bright cells, considerable heterogeneity was observed among donors in L-selectin expression on the numerically major CD56dim NK subset. These results may explain the variable findings of several studies regarding the constitutive level of L-selectin expression on NK cells in which the contributions of CD56bright and CD56 dim cells were not determined (28, 30, 31, 32). The reason for the variability in L-selectin expression on CD56dim cells is not known but may reflect differences among donors in the activation status of circulating NK cells. While healthy donors were used in these studies, subclinical or trivial infections could have lead to variations in L-selectin expression. As indicated in the present study, lymphocyte activation (i.e., in response to PMA or stimulatory cytokines such as IL-2) leads to rapid loss of L-selectin surface expression (20, 35, 38). In addition, heterogeneity in L-selectin expression may reflect the differentiation stage of circulating NK cells such that in some individuals, a higher percentage of the mature CD56dim population (5, 7, 13) transiently maintains some of the phenotypic characteristics (i.e., L-selectin levels) of the less mature CD56bright subset.

Emerging evidence that the CD56bright NK cell subset is unique in its high constitutive expression of multiple adhesion molecules that mediate cell-cell and cell-matrix interactions, such as L-selectin (this report and Refs. 5 and 29), CD2, CD44, CD11c, and ICAM-1 (5, 29, 34), suggests that this NK subset circulates in a heightened state of functional readiness. L-selectin potentially functions at several discrete levels to recruit CD56bright NK cells to secondary sites. In addition to mediating leukocyte-endothelium binding, L-selectin has recently been shown to support leukocyte-leukocyte interactions involving neutrophils (47, 48), suggesting similar functions for L-selectin in lymphocyte population such as CD56bright NK cells. The ability of L-selectin to mediate leukocyte rolling on adherent neutrophil monolayers has been proposed as a mechanism to provide access of circulating immune effector cells to tissues in the early stages of an immune response when newly recruited neutrophils effectively form a barrier to underlying endothelium (47, 48).

L-selectin was also shown in the present study to regulate LFA-1 function in NK cells, similar to its reported function in neutrophils and T lymphocytes (22, 23, 24, 46, 49). LFA-1 plays a crucial role in strengthening leukocyte-endothelial adhesion and promotes transendothelial migration (20, 21). Notably, recent studies by Berman et al. have demonstrated that CD56bright NK cells migrate at a significantly faster rate across cytokine-activated endothelium than CD56dim NK cells (37), although the adhesion pathway was not identified in this prior study. Our data demonstrate that the DREG-56 mAb, used in place of L-selectin ligand, strongly stimulates LFA-1 activation in CD56bright cells as well as a subset of CD56dim cells, presumably the CD56dim/L-selectindim population. However, in a physiologic setting in which L-selectin adhesion to HEV ligands is expected to occur preferentially in CD56bright cells, L-selectin-mediated activation of LFA-1 may have only a limited role in the emigration of the CD56dim subset. CD44 and CD11c are also present at a higher surface density on CD56bright NK cells than on the CD56dim subset (5, 29). These adhesion molecules mediate leukocyte adhesion to vascular endothelium and to extracellular matrix proteins, i.e., hyaluronate and fibrinogen, and thus, potentially function to facilitate NK cell extravasation and migration through the interstitial matrix in tissues (21). Within tissues, high constitutive levels of ICAM-1 on the CD56bright NK subset (5, 29) could mediate interactions with accessory cells such as monocytes via binding to its ligands LFA-1 and Mac-1.

The selective migration of the CD56bright NK subset via L-selectin into inflammatory sites may be advantageous at the initiation of infection, when cytokine levels are low but sufficient to cause proliferation of this subset. CD56bright NK cells are unique among circulating lymphocytes because of their constitutive expression of the high affinity IL-2R (6, 7). Thus, limited concentrations of T cell-derived IL-2 in tissues before extensive T cell infiltration would theoretically favor proliferation of CD56bright NK cells over other unstimulated lymphocyte populations. Recent evidence (9) that the monocyte-derived cytokine IL-15, which also binds to the IL-2βγR, stimulates CD56bright NK cell proliferation is particularly intriguing, since monocytes are known to function in the earliest phase of the immune response to bacterial or viral pathogens. Two monocyte-produced cytokines, IL-12 and IL-15, have been shown to act synergistically to stimulate NK cell production of proinflammatory cytokines including IFN-γ, TNF-α, and granulocyte/macrophage-CSF (1, 2, 9, 11, 13, 50, 51), suggesting a paracrine loop in which NK cell-produced cytokines regulate monocyte differentiation and effector function. Significantly, the CD56bright NK subset produces markedly higher levels of IFN-γ than CD56dim cells following stimulation by IL-12, in combination with either IL-15, IL-1β, or TNF-α (12).

NK-derived cytokines may also recruit other leukocyte types to the site. The role of IFN-γ and TNF-α in stimulating endothelial cell expression of ICAM-1, VCAM-1, and E-selectin is well established (20, 21). Thus, high level production of these cytokines by CD56bright NK cells early in the immune response feasibly potentiates the recruitment of T lymphocytes, CD56dim NK cells, and monocytes to tissues. The high density of LFA-1 on CD56dim NK cells demonstrated in this report may facilitate extravasation of this immune effector subset through interactions with ICAM-1 on cytokine-activated endothelium. Alternatively, high density LFA-1 may potentiate the cytolytic effector function of NK cells, as suggested previously (34). In addition, production by CD56bright NK cells of the C-C chemokine MIP-1α (macrophage inflammatory protein-1α) in response to IL-12 in combination with IL-15, as described recently (52), could enhance immune effector cell migration within inflammatory sites as well as their emigration from the blood.

In this report, L-selectin was shown to be down-regulated following stimulation of CD56bright and CD56dim NK cells by IL-2, IL-15, TGF-β, or PMA in vitro. Similar changes in L-selectin expression in response to IL-2 or PMA have been described for total NK cell populations (28, 30, 32). These data suggest that L-selectin would be rapidly lost on activated CD56bright cells shortly after their infiltration into tissues and activation in situ, although it is difficult to predict the outcome in tissues containing multiple cytokines that exert opposing effects on L-selectin expression (e.g., IL-12, IL-10, IFN-α vs IL-2, IL-15, and TGF-β). In this regard, we have found in preliminary studies that inclusion of IL-12 in IL-15-stimulated cultures partially protects L-selectin expression on the CD56bright NK cell population (n = 2). The observed effect of cytokines on L-selectin expression by NK cells may have important implications for clinical studies in which cytolytic NK cells are expanded in vitro by IL-2 before use in adoptive immunotherapy in cancer patients. Under these conditions, cytokine-activated NK cells would likely be excluded from tissues requiring L-selectin in the extravasation process. Further studies are required to determine the value of using cytokines such as IL-12, IL-10, or IFNs, as suggested here and in previous studies (40, 53), to restore L-selectin levels on expanded populations of immune effector cells immediately before their transfer to patients.

In conclusion, data presented in this report strongly implicate L-selectin in the dissemination of the CD56bright NK cell subset to tissues in the early phase of an immune response to foreign pathogens. Based on these results, we speculate that the L-selectin adhesion pathway is involved in the rapid extralymphoid accumulation of NK cells as described previously at sites of viral infection (1, 2, 15, 16, 17), inoculation with bacterial agents (14), allograft rejection (18), or tumor growth (19). The finding that CD56bright NK cells bind efficiently to well-established sites of L-selectin ligand on HEV within lymph node tissues further raises the possibility that this NK cell subset, like naive T lymphocytes (20, 21), recirculates through peripheral lymphoid tissues as part of physiologic immune surveillance mechanisms.

We thank Drs. Eugene Butcher, Lloyd Stoolman, Richard Bankert, and William Muller for graciously providing the mAb, PPME-FITC reagent, and cell lines used in these studies; Sigrid Stewart for technical assistance; Edward Hurley for assistance with confocal microscopy; Dr. Elizabeth Repasky for the use of the fluorescence microscope; and Dr. Jennifer Black and Michelle Appenheimer for critical reading of the manuscript.

1

This work was supported in part by grants from the National Institutes of Health (RR08926 to S.S.E.; CA68456 and CA65670 to M.A.C.; CA60200 to C.C.S.), the Roswell Park Alliance Foundation (S.S.E.), the Cancer Center Support Grant (P30 CA16056-21) at Roswell Park Cancer Institute; and the Dr. Louis Sklarow Memorial Fund (S.S.E.). T.A.F. is the recipient of a Howard Hughes Medical Institute Research Fellowship for Medical Students and the Bennett Fellowship from The Ohio State University, College of Medicine.

3

Abbreviations used in this paper: HEV, high endothelial venule; PPME, phosphomonoester core polysaccharide; RITC, rhodamine isothiocyanate; PE, phycoerythrin; LFA-1, lymphocyte function-associated Ag 1.

1
Biron, C. A..
1997
. Activation and function of natural killer cell responses during viral infections.
Curr. Opin. Immunol.
9
:
24
2
Trinchieri, G..
1997
. Cytokines acting on or secreted by macrophages during intracellular infection (IL-10, IL-12, IFN-γ).
Curr. Opin. Immunol.
9
:
17
3
Robertson, M. J., J. Ritz.
1990
. Biology and clinical relevance of human natural killer cells.
Blood
76
:
2421
4
Ellis, T. M., R. I. Fisher.
1989
. Functional heterogeneity of Leu 19“bright”+ and Leu 19“dim”+ lymphokine-activated killer cells.
J. Immunol.
142
:
2949
5
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
6
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
7
Nagler, A., L. L. Lanier, J. H. Phillips.
1990
. Constitutive expression of high affinity interleukin 2 receptors on human CD16 natural killer cells in vivo.
J. Exp. Med.
171
:
1527
8
Matos, M. E., G. S. Schnier, M. S. Beecher, L. K. Ashman, D. E. Williams, M. A. Caligiuri.
1993
. Expression of a functional c-kit receptor on a subset of natural killer cells.
J. Exp. Med.
178
:
1079
9
Carson, W. E., J. G. Giri, M. J. Lindemann, M. L. Linett, M. Ahdieh, R. Paxton, D. Anderson, J. Eisenmann, K. Grabstein, M. A. Caligiuri.
1994
. Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor.
J. Exp. Med.
180
:
1395
10
Voss, S. D., P. M. Sondel, R. J. Robb.
1992
. Characterization of the interleukin 2 receptors (IL-2R) expressed on human natural killer cells activated in vivo by IL-2: association of the p64 IL-2R γ chain with the IL-2R β chain in functional intermediate-affinity IL-2R.
J. Exp. Med.
176
:
531
11
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
12
Fehniger, T. A., W. E. Carson, M. A. Caligiuri.
1997
. The CD56bright human NK cell subset is the most potent NK producer of macrophage activating factors.
Proc. Am. Assoc. Cancer Res.
38
:
32
13
Carson, W., M. Caligiuri.
1996
. Natural killer cell subsets and development.
Methods Companion Methods Enzymol.
9
:
327
14
Holmberg, L. A., T. A. Springer, K. A. Ault.
1981
. Natural killer activity in the peritoneal exudate of mice infected with listeria monocytogenes: characterization of the NK cells by using a monoclonal rat anti-munne antibody (MI/70). (MI/70).
J. Immunol.
127
:
1792
15
Natuk, R. J., R. M. Welsh.
1987
. Accumulation and chemotaxis of natural killer/large granular lymphocytes at site of virus replication.
J. Immunol.
138
:
877
16
McIntyre, K. W., R. M. Welsh.
1986
. Accumulation of natural killer and cytotoxic T large granular lymphocytes in the liver during virus infection.
J. Exp. Med.
164
:
1667
17
Biron, C. A., L. R. Turgiss, R. M. Welsh.
1983
. Increase in NK cell number and turnover rate during acute viral infection.
J. Immunol.
131
:
1539
18
Nemlander, A., E. Saksela, P. Hayry.
1983
. Are natural killer cells involved in allograft rejection?.
Eur. J. Immunol.
13
:
348
19
Melder, R. J., G. C. Koenig, L. L. Munn, R. K. Jain.
1996
. Adhesion of activated natural killer cells to tumor necrosis factor-α-treated endothelium under physiological flow conditions.
Nat. Immun.
15
:
154
20
Butcher, E. C., L. J. Picker.
1996
. Lymphocyte homing and homeostasis.
Science
272
:
60
21
Springer, T. A..
1995
. Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration.
Annu. Rev. Physiol.
57
:
827
22
Simon, S. I., A. R. Burns, A. D. Taylor, P. K. Gopalan, E. B. Lynam, L. A. Sklar, C. W. Smith.
1995
. L-selectin (CD62L) cross-linking signals neutrophil adhesive functions via the Mac-1 (CD11b/CD18) β2-integrin.
J. Immunol.
155
:
1502
23
Hwang, S. T., M. S. Singer, P. A. Giblin, T. A. Yednock, K. B. Bacon, S. I. Simon, S. D. Rosen.
1996
. GlyCAM-1, a physiologic ligand for L-selectin, activates β2 integrins on naive peripheral lymphocytes.
J. Exp. Med.
184
:
1343
24
Gopalan, P. K., C. W. Smith, H. Lu, E. L. Berg, L. V. McIntire, S. I. Simon.
1997
. Neutrophil CD18-dependent arrest on intercellular adhesion molecule 1 (ICAM-1) in shear flow can be activated through L-selectin.
J. Immunol.
158
:
367
25
Bianchi, G., M. Sironi, E. Ghibaudi, C. Selvaggini, M. Elices, P. Allavena, A. Mantovani.
1993
. Migration of natural killer cells across endothelial cell monolayers.
J. Immunol.
151
:
5135
26
Bender, J. R., R. Pardi, M. A. Karasek, E. G. Engleman.
1987
. Phenotypic and functional characterization of lymphoctes that bind human microvascular endothelial cells in vitro: evidence for preferential binding of natural killer cells.
J. Clin. Invest.
79
:
1679
27
Pinola, M., E. Saksela, S. Tiisala, R. Renkonen.
1994
. Human NK cells expressing α417 adhere to VCAM-1 without preactivaiton.
Scand. J. Immunol.
39
:
131
28
Uksila, J., M. Salmi, E. C. Butcher, J. Tarkkanen, S. Jalkanen.
1997
. Function of lymphocyte homing-associated adhesion molecules on human natural killer and lymphokine-activated killer cells.
J. Immunol.
158
:
1610
29
Sedlmayr, P., L. Schallhammer, A. Hammer, M. Wilders-Truschnig, R. Wintersteiger, G. Dohr.
1996
. Differential phenotypic properties of human peripheral blood CD56dim+ and CD56bright+ natural killer cell subpopulations.
Int. Arch. Allergy Immunol.
110
:
308
30
Buhrer, C., C. Berlin, H. G. Thiele, A. Hamann.
1990
. Lymphocyte activation and expression of the human leukocyte-endothelial cell adhesion molecule 1 (Leu-8/TQ1 antigen).
Immunology
71
:
442
31
Tedder, T. F., A. C. Penta, H. B. Levine, A. S. Freedman.
1990
. Expression of the human leukocyte adhesion molecule, LAM1: identity with the TQ1 and Leu-8 differentiation antigens.
J. Immunol.
144
:
532
32
Maenpaa, A., J. Jaaskelainen, O. Carpen, M. Patarroyo, T. Timonen.
1993
. Expression of integrins and other adhesion molecules on NK cells: impact of IL-2 on short- and long-term cultures.
Int. J. Cancer
53
:
850
33
Allavena, P., C. Paganin, I. Martin-Padura, G. Peri, M. Gaboli, E. Dejana, P. C. Marchisio, A. Mantovani.
1991
. Molecules and structures involved in the adhesion of natural killer cells to vascular endothelium.
J. Exp. Med.
173
:
439
34
Robertson, M. J., M. A. Caligiuri, T. J. Manley, H. Levine, J. Ritz.
1990
. Human natural killer cell adhesion molecules: differential expression after activation and participation in cytolysis.
J. Immunol.
145
:
3194
35
Frey, M., M. M. Appenheimer, S. S. Evans.
1997
. Tyrosine kinase-dependent regulation of L-selectin expression through the Leu-13 signal transduction molecule: evidence for a protein kinaseC-independent mechanism of L-selectin shedding.
J. Immunol.
158
:
5424
36
Wang, W. C., L. M. Goldman, D. M. Schleider, M. M. Appenheimer, J. R. Subjeck, E. A. Repasky, S. S. Evans.
1998
. Fever-range hyperthermia enhances L-selectin-dependent adhesion of lymphocytes to vascular endothelium.
J. Immunol.
160
:
961
37
Berman, M. E., Y. Xie, W. A. Muller.
1996
. Roles of platelet/endothelial cell adhesion molecule-1 (PECAM-1, CD31) in natural killer cell transendothelial migration and β2 integrin activation.
J. Immunol.
156
:
1515
38
Kishimoto, T. K., M. A. Jutila, E. C. Butcher.
1990
. Identification of a human peripheral lymph node homing receptor: a rapidly down-regulated adhesion molecule.
Proc. Natl. Acad. Sci. USA
87
:
2244
39
Stoolman, L. M., E. Kaldjian.
1992
. Adhesion molecules involved in the trafficking of normal and malignant leukocytes.
Invasion Metastasis
12
:
101
40
Evans, S. S., R. P. Collea, M. M. Appenheimer, S. O. Gollnick.
1993
. Interferon-α induces the expression of the L-selectin homing receptor in human B lymphoid cells.
J. Cell Biol.
123
:
1889
41
Stamper, H. B., Jr, J. J. Woodruff.
1976
. Lymphocyte homing into lymph nodes: in vitro demonstration of the selective affinity of recirculating lymphocytes for high-endothelial venules.
J. Exp. Med.
144
:
828
42
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.
136
:
4480
43
Carson, W. E., M. J. Lindemann, R. Baiocchi, M. Linett, J. C. Tan, C. C. Chou, S. Narula, M. A. Caligiuri.
1995
. The functional characterization of interleukin-10 receptor expression on human natural killer cells.
Blood
85
:
3577
44
Rook, A. H., J. H. Kehrl, L. M. Wakefield, A. B. Roberts, M. B. Sporn, D. B. Burlington, H. C. Lane, A. S. Fauci.
1986
. Effects of transforming growth factor β on the functions of natural killer cells: depressed cytolytic activity and blunting of interferon responsiveness.
J. Immunol.
136
:
3916
45
Steen, P. D., J. R. McGregor, C. M. Lehman, W. E. Samlowski.
1989
. Changes in homing receptor expression on murine lymphokine-activated killer cells during IL-2 exposure.
J. Immunol.
143
:
4324
46
Bowman, L., J. D. Black, S. S. Evans.
1994
. Association of L-selectin with the lymphocyte cytoskeletal matrix.
FASEB J.
8
:
A236
47
Bargatze, R. F., S. Kurk, E. C. Butcher, M. A. Jutila.
1994
. Neutrophils roll on adherent neutrophils bound to cytokine-induced endothelial cells via L-selectin on the rolling cells.
J. Exp. Med.
180
:
1785
48
Alon, R., R. C. Fuhlbrigge, E. B. Finger, T. A. Springer.
1996
. Interactions through L-selectin between leukocytes and adherent leukocytes nucleate rolling adhesions on selectins and VCAM-1 in shear flow.
J. Cell Biol.
135
:
849
49
Steeber, D. A., P. Engel, A. S. Miller, M. P. Sheetz, T. F. Tedder.
1997
. Ligation of L-selectin through conserved regions within the lectin domain activates signal transduction pathways and integrin function in human, mouse, and rat leukocytes.
J. Immunol.
159
:
952
50
Carson, W. E., M. E. Ross, R. A. Baiocchi, M. J. Marien, N. Boiani, K. Grabstein, M. A. Caligiuri.
1995
. Endogenous production of interleukin 15 by activated human monocytes is critical for optimal production of interferon-γ by natural killer cells in vitro.
J. Clin. Invest.
96
:
2578
51
Orange, J. S., C. A. Biron.
1996
. An absolute and restricted requirement for interleukin-12 (IL-12) in natural killer (NK) cell interferon-γ production and antiviral defense: studies of NK and T cell responses in contrasting viral infections.
J. Immunol.
156
:
1138
52
Bluman, E. M., K. J. Bartynski, B. R. Avalos, M. A. Caligiuri.
1996
. Human natural killer cells produce abundant macrophage inflammatory protein-1α in response to monocyte-derived cytokines.
J. Clin. Invest.
97
:
2722
53
McRae, B. L., L. J. Picker, G. A. van Seventer.
1997
. Human recombinant interferon-β influences T helper subset differentiation by regulating cytokine secretion pattern and expression of homing receptors.
Eur. J. Immunol.
27
:
2650