Segregation of HLA-C from ICAM-1 at NK cell immune synapses is controlled by its cell surface density

NK cell activity is controlled by the integration of signals from numerous activating and inhibitory receptors at the immunological synapse (IS). However, the importance of segregation and patterning of proteins at the NK cell IS is unknown. In this study, we report that the level of expression of HLA-C on target cells determined its supramolecular organization and segregation from ICAM-1 at the NK cell IS, as well as its capacity to inhibit NK cell cytotoxicity. At YTS NK cell synapses formed with target cells expressing low levels of HLA-C (i.e., 10(4)/cell surface), a multifocal patterning of MHC class I protein predominated, whereas for higher levels of expression (10(5)/cell surface), clusters of HLA-C were more commonly homogeneous, ring-shaped, or containing multiple exclusions. This correlation of protein density with its patterning at the IS was independent of ATP- or actin-driven processes. Importantly, ICAM-1 and HLA-C segregated only at synapses involving target cells expressing high levels of MHC protein. For peripheral blood NK clones, there were specific thresholds in the level of target cell HLA-C needed to inhibit cytotoxicity and to cause segregation of HLA-C from ICAM-1 at the synapse. Thus, the synapse organization of HLA-C, determined by its level of expression, could directly influence NK cell inhibition, e.g., by regulating the proximity of activating and inhibitory receptors. For the first time, this suggests an important function for the assembly of an inhibitory NK cell IS. More broadly, segregation of proteins at intercellular contacts could transmit information about protein expression levels between cells.     

Matched sizes of activating and inhibitory receptor/ligand pairs are required for optimal signal integration by human natural killer cells

It has been suggested that receptor-ligand complexes segregate or co-localise within immune synapses according to their size, and this is important for receptor signaling. Here, we set out to test the importance of receptor-ligand complex dimensions for immune surveillance of target cells by human Natural Killer (NK) cells. NK cell activation is regulated by integrating signals from activating receptors, such as NKG2D, and inhibitory receptors, such as KIR2DL1. Elongating the NKG2D ligand MICA reduced its ability to trigger NK cell activation. Conversely, elongation of KIR2DL1 ligand HLA-C reduced its ability to inhibit NK cells. Whereas normal-sized HLA-C was most effective at inhibiting activation by normal-length MICA, only elongated HLA-C could inhibit activation by elongated MICA. Moreover, HLA-C and MICA that were matched in size co-localised, whereas HLA-C and MICA that were different in size were segregated. These results demonstrate that receptor-ligand dimensions are important in NK cell recognition, and suggest that optimal integration of activating and inhibitory receptor signals requires the receptor-ligand complexes to have similar dimensions.     

Cutting edge: engagement of CD160 by its HLA-C physiological ligand triggers a unique cytokine profile secretion in the cytotoxic peripheral blood NK cell subset

CD160 is an Ig-like activating NK cell receptor expressed on the majority of circulating NK cells. This population corresponds to the nonproliferating, highly cytolytic, CD56dimCD16+ subset. CD160 engagement by HLA-C molecules mediates cytotoxic function. In this study, we report that upon specific activation by the physiological ligand HLA-C, or Ab cross-linking, CD160+ peripheral blood NK cells produce IFN-gamma, TNF-alpha, and IL-6. This unique CD160-mediated cytokine production differs from the one observed after CD16 engagement whose expression is also restricted to the CD56dim cytotoxic NK cell subset. As already reported for the CD160-mediated cytotoxic effector function, CD160-mediated cytokine production by peripheral blood-NK cells is negatively controlled by the killer Ig-like receptor CD158b. Thus, the CD160 receptor represents a unique triggering surface molecule expressed by cytotoxic NK cells that participates in the inflammatory response and determines the type of subsequent specific immunity.     

Ligand binding to inhibitory killer cell Ig-like receptors induce colocalization with Src homology domain 2-containing protein tyrosine phosphatase 1 and interruption of ongoing activation signals

Interaction of NK cells with target cells leads to formation of an immunological synapse (IS) at the contact site. NK cells form two distinctly different IS, the inhibitory NK cell IS (NKIS) and the cytolytic NKIS. Cognate ligand binding is sufficient to induce clustering of inhibitory killer cell Ig-like receptors (KIR) and phosphorylation of both the receptor and the phosphatase Src homology domain 2-containing protein tyrosine phosphatase 1 (SHP-1). Recruitment and activation of SHP-1 by a signaling competent inhibitory receptor are essential early events for NK cell inhibition. We have in the present study used three-dimensional immunofluorescence microscopy to analyze distribution of inhibitory KIR, SHP-1, LFA-1, and lipid rafts within the NKIS during cytolytic and noncytolytic interactions. NK clones retrovirally transduced with the inhibitory KIR2DL3 gene fused to GFP demonstrate colocalization of KIR2DL3 with SHP-1 in the center of early inhibitory NKIS. Ligand binding translocates the receptor to the center of the IS where activation signals are accumulating and provides a docking site for SHP-1. SHP-1 and rafts cluster in the center of early inhibitory NKIS and late cytolytic NKIS, and whereas rafts continue to increase in size in cytolytic conjugates, they are rapidly dissolved in inhibitory conjugates. Furthermore, rafts are essential only for cytolytic, not for inhibitory, outcome. These results indicate that the outcome of NK cell-target cell interactions is dictated by early quantitative differences in cumulative activating and inhibitory signals.     

Immune synapse formation requires ZAP-70 recruitment by ezrin and CD43 removal by moesin

Immunological synapse (IS) formation involves receptor–ligand pair clustering and intracellular signaling molecule recruitment with a coincident removal of other membrane proteins away from the IS. As microfilament–membrane linkage is critical to this process, we investigated the involvement of ezrin and moesin, the two ezrin/radixin/moesin proteins expressed in T cells. We demonstrate that ezrin and moesin, which are generally believed to be functionally redundant, are differentially localized and have important and complementary functions in IS formation. Specifically, we find that ezrin directly interacts with and recruits the signaling kinase ZAP-70 to the IS. Furthermore, the activation of ezrin by phosphorylation is essential for this process. In contrast, moesin dephosphorylation and removal, along with CD43, are necessary to prepare a region of the cell cortex for IS. Thus, ezrin and moesin have distinct and critical functions in the T cell cortex during IS formation.     

CD48 stimulation by 2B4 (CD244)-expressing targets activates human NK cells

Human NK cells can be activated by a variety of different cell surface receptors. Members of the SLAM-related receptors (SRR) are important modulators of NK cell activity. One interesting feature of the SRR is their homophilic interaction, combining receptor and ligand in the same molecule. Therefore, SRR cannot only function as activating NK cell receptors, but also as activating NK cell ligands. 2B4 (CD244) is the only SRR that does not show homophilic interaction. Instead, 2B4 is activated by binding to CD48, a GPI-anchored surface molecule that is widely expressed in the hemopoietic system. In this study, we show that 2B4 also can function as an activating NK cell ligand. 2B4-expressing target cells can efficiently stimulate NK cell cytotoxicity and IFN-gamma production. Using soluble receptor fusion proteins and SRR-transfected cells, we show that 2B4 does not bind to any other SRR expressed on NK cells, but only interacts with CD48. Lysis of 2B4-expressing target cells can be blocked by anti-CD48 Abs and triggering of CD48 in a redirected lysis assay can stimulate NK cell cytotoxicity. This demonstrates that 2B4 can stimulate NK cell cytotoxicity and cytokine production by interacting with NK cell expressed CD48 and adds CD48 to the growing number of activating NK cell receptors.     

Cytotoxicity of CD56(bright) NK cells towards autologous activated CD4+ T cells is mediated through NKG2D, LFA-1 and TRAIL and dampened via CD94/NKG2A

In mouse models of chronic inflammatory diseases, Natural Killer (NK) cells can play an immunoregulatory role by eliminating chronically activated leukocytes. Indirect evidence suggests that NK cells may also be immunoregulatory in humans. Two subsets of human NK cells can be phenotypically distinguished as CD16(+)CD56(dim) and CD16(dim/-)CD56(bright). An expansion in the CD56(bright) NK cell subset has been associated with clinical responses to therapy in various autoimmune diseases, suggesting an immunoregulatory role for this subset in vivo. Here we compared the regulation of activated human CD4(+) T cells by CD56(dim) and CD56(bright) autologous NK cells in vitro. Both subsets efficiently killed activated, but not resting, CD4(+) T cells. The activating receptor NKG2D, as well as the integrin LFA-1 and the TRAIL pathway, played important roles in this process. Degranulation by NK cells towards activated CD4(+) T cells was enhanced by IL-2, IL-15, IL-12+IL-18 and IFN-α. Interestingly, IL-7 and IL-21 stimulated degranulation by CD56(bright) NK cells but not by CD56(dim) NK cells. NK cell killing of activated CD4(+) T cells was suppressed by HLA-E on CD4(+) T cells, as blocking the interaction between HLA-E and the inhibitory CD94/NKG2A NK cell receptor enhanced NK cell degranulation. This study provides new insight into CD56(dim) and CD56(bright) NK cell-mediated elimination of activated autologous CD4(+) T cells, which potentially may provide an opportunity for therapeutic treatment of chronic inflammation.     

Development and function of CD94-deficient natural killer cells

The CD94 transmembrane-anchored glycoprotein forms disulfide-bonded heterodimers with the NKG2A subunit to form an inhibitory receptor or with the NKG2C or NKG2E subunits to assemble a receptor complex with activating DAP12 signaling proteins. CD94 receptors expressed on human and mouse NK cells and T cells have been proposed to be important in NK cell tolerance to self, play an important role in NK cell development, and contribute to NK cell-mediated immunity to certain infections including human cytomegalovirus. We generated a gene-targeted CD94-deficient mouse to understand the role of CD94 receptors in NK cell biology. CD94-deficient NK cells develop normally and efficiently kill NK cell-susceptible targets. Lack of these CD94 receptors does not alter control of mouse cytomegalovirus, lymphocytic choriomeningitis virus, vaccinia virus, or Listeria monocytogenes. Thus, the expression of CD94 and its associated NKG2A, NKG2C, and NKG2E subunits is dispensable for NK cell development, education, and many NK cell functions.     

Live Cell Linear Dichroism Imaging Reveals Extensive Membrane Ruffling within the Docking Structure of Natural Killer Cell Immune Synapses

We have applied fluorescence imaging of two-photon linear dichroism to measure the subresolution organization of the cell membrane during formation of the activating (cytolytic) natural killer (NK) cell immune synapse (IS). This approach revealed that the NK cell plasma membrane is convoluted into ruffles at the periphery, but not in the center of a mature cytolytic NK cell IS. Time-lapse imaging showed that the membrane ruffles formed at the initial point of contact between NK cells and target cells and then spread radialy across the intercellular contact as the size of the IS increased, becoming absent from the center of the mature synapse. Understanding the role of such extensive membrane ruffling in the assembly of cytolytic synapses is an intriguing new goal.     

CD1d1 displayed on cell size beads identifies and enriches an NK cell population negatively regulated by CD1d1

NK cells destroy microbe-infected cells while sparing healthy cells, and are controlled, in part, by inhibitory receptors specific for class I Ag-presenting molecules. CD1d1, a beta(2)-microglobulin-associated class I-like molecule, binds glycolipids and stimulates NKT cells. We previously demonstrated that target cell lysis by IL-2-activated mouse NK cells is inhibited by target cell expression of CD1d1, suggesting that IL-2-activated NK cells may express a CD1d1-specific inhibitory receptor. We now report that a significant subset of mouse IL-2-activated NK cells specifically binds cell size beads displaying either naturally expressed or recombinant CD1d1. In contrast, although tetramers of soluble recombinant CD1d1 loaded with alpha-galactosylceramide identify NKT cells, binding of this reagent to resting or IL-2-activated NK cells was undetectable, even with activated NK cells sorted with CD1d1 beads. Cytotoxicity by the CD1d1 bead-separated NK subset was strongly inhibited by CD1d1, compared with the NK cell subset not bound to CD1d1 beads. An Ab that blocks NKT cell recognition of CD1d1 also reverses CD1d1 inhibition of NK lysis, suggesting that TCRs of NKT cells and NK inhibitory receptor(s) may interact with a similar site on CD1d1. These results provide direct evidence for a physical interaction of NK cells with CD1d1, mediated by a functional, CD1d1-specific low-affinity inhibitory NK receptor. Display of ligands on cell size beads to maximize multivalent interaction may offer an alternative approach to examine NK cell receptor-ligand interactions, particularly those of lower expression and/or lower affinity/avidity that may go undetected using tetrameric reagents.     

Membranous structures transfer cell surface proteins across NK cell immune synapses

Intercellular transfer of cell surface proteins is widespread and facilitates several recently discovered means for immune cell communication. Here, we examined the molecular mechanism for intercellular exchange of the natural killer (NK) cell receptor KIR2DL1 and HLA-C, prototypical proteins that swap between NK cells and target cells. Transfer was contact dependent and enhanced for cells expressing cognate receptor/ligand pairs but did not depend on KIR2DL1 signaling. To a lesser extent, proteins transferred independent from specific recognition. Intracellular domains of transferred proteins were not exposed to the extracellular environment and transferred proteins were removed by brief exposure to low pH. By fluorescence microscopy, transferred proteins localized to discrete regions on the recipient cell surface. Higher resolution scanning electron micrographs revealed that transferred proteins were located within specific membranous structures. Transmission electron microscopy of the immune synapse revealed that membrane protrusions from one cell interacted with the apposing cell surface within the synaptic cleft. These data, coupled with previous observations, lead us to propose that intercellular protein transfer is mediated by membrane protrusions within and surrounding the immunological synapse.     

Human immunodeficiency virus type 1 infection is associated with increased NK cell polyfunctionality and higher levels of KIR3DL1+ NK cells in ugandans carrying the HLA-B Bw4 motif

Natural killer (NK) cells are important innate effector cells controlled by an array of activating and inhibitory receptors. Some alleles of the inhibitory killer-cell immunoglobulin-like receptor KIR3DL1 in combination with its HLA class I ligand Bw4 have been genetically associated with slower HIV-1 disease progression. Here, we observed that the presence of HLA-B Bw4 was associated with elevated frequencies of KIR3DL1(+) CD56(dim) NK cells in chronically HIV-1-infected individuals from the rural district of Kayunga, Uganda. In contrast, levels of KIR2DL1(+) CD56(dim) NK cells were decreased, and levels of KIR2DL3(+) CD56(dim) NK cells were unchanged in infected subjects carrying their respective HLA-C ligands. Furthermore, the size of the KIR3DL1(+) NK cell subset correlated directly with viral load, and this effect occurred only in HLA-B Bw4(+) patients, suggesting that these cells expand in response to viral replication but may have relatively poor antiviral capacity. In contrast, no association with viral load was present for KIR2DL1(+) and KIR2DL3(+) NK cells. Interestingly, chronic HIV-1 infection was associated with an increased polyfunctional response in the NK cell compartment, and, upon further investigation, KIR3DL1(+) CD56(dim) NK cells exhibited a significantly increased functional response in the patients carrying HLA-B Bw4. These results indicate that chronic HIV-1 infection is associated with increased NK cell polyfunctionality and elevated levels of KIR3DL1(+) NK cells in Ugandans carrying the HLA-B Bw4 motif.     

An unusual CD56(bright) CD16(low) NK cell subset dominates the early posttransplant period following HLA-matched hematopoietic stem cell transplantation

The expansion of the cytokine-producing CD56(bright) NK cell subset is a main feature of lymphocyte reconstitution after allogeneic hematopoietic stem cell transplantation (HSCT). We investigated phenotypes and functions of CD56(bright) and CD56(dim) NK subsets from 43 HLA-matched non-T cell-depleted HSCT donor-recipient pairs. The early expansion of CD56(bright) NK cells gradually declined in the posttransplant period but still persisted for at least 1 year and was characterized by the emergence of an unusual CD56(bright)CD16(low) subset with an intermediate maturation profile. The activating receptors NKG2D and NKp46, but also the inhibitory receptor NKG2A, were overexpressed compared with donor CD56(bright) populations. Recipient CD56(bright) NK cells produced higher amounts of IFN-gamma than did their respective donors and were competent for degranulation. Intracellular perforin content was increased in CD56(bright) NK cells as well as in T cells compared with donors. IL-15, the levels of which were increased in the posttransplant period, is a major candidate to mediate these changes. IL-15 serum levels and intracellular T cell perforin were significantly higher in recipients with acute graft-vs-host disease. Altogether, CD56(bright) NK cells postallogeneic HSCT exhibit peculiar phenotypic and functional properties. Functional interactions between this subset and T cells may be important in shaping the immune response after HSCT.     

Immunology of pregnancy: renewed interest

The long-standing question of pregnancy immunological paradox has been generating renewed interest. Recent insights have emerged from studies in pregnant mice and humans demonstrating a number of mechanisms that prevent potentially harmful effects of maternal anti-paternal allo-antibodies (complement inhibition, partial deletion of maternal B cells specific of paternal antigens), cytotoxic CD8+ T cells (lack of HLA-A and HLA-B expression on trophoblast, local immunosuppressive molecules, transient tolerance of paternal allo-antigens specific T cells) and uterine NK cells directed against fetal-derived trophoblast cells (limited NK cytotoxic potential, trophoblast resistance to NK killing). Interestingly, it appears that not only decidual NK cell/trophoblast interactions are not harmful for the fetus but are beneficial for the placental vascularization and its subsequent development. A recent report has indeed demonstrated that during pregnancy most of the combinations of uterine KIR (killer cell immunoglobulin-like receptor) NK cell receptors and fetal HLA-C molecules expressed by trophoblast led to normal pregnancies, whereas mothers lacking activating KIR of the AA genotype when the fetus possessed HLA-C of the C2 group were at a greatly increased risk of severe preeclampsia pathology.     

Polar redistribution of the sialoglycoprotein CD43: implications for T cell function

Contact between T cells and APCs results in the orchestrated segregation of molecules at the cell-cell interface and formation of a specialized structure termed the immunological synapse. This model predicts the topological seclusion of large molecules such as CD43 from the site of closest contact between the T cell and APC, allowing for the close apposition of cell membranes and effective TCR engagement. Similarly, during T cell migration segregation of CD43 to the uropod is thought to aid integrin adhesion at the leading edge of the cell by removing steric hindrance. We show in this work that CD43 distribution on T cells is regulated by a membrane proximal ezrin binding site and that failure to displace CD43 from the immunological synapse has no inhibitory effects on primary T cell activation. We also report that CD43 expression at the contact zone between T cells and matrix does not negatively regulate motility but may regulate LFA-1 de-adhesion. These results suggest that the steric barrier model of CD43 is inadequate and that alternative mechanisms account for the negative regulatory properties of CD43.     

CD94/NKG2A inhibits NK cell activation by disrupting the actin network at the immunological synapse

An adequate immune response is the result of the fine balance between activation and inhibitory signals. The exact means by which inhibitory signals obviate activation signals in immune cells are not totally elucidated. Human CD94/NKG2A is an ITIM-containing inhibitory receptor expressed by NK cells and some CD8+ T cells that recognize HLA-E. We show that the engagement of this receptor prevents NK cell activation by disruption of the actin network and exclusion of lipid rafts at the point of contact with its ligand (inhibitory NK cell immunological synapse, iNKIS). CD94/NKG2A engagement leads to recruitment and activation of src homology 2 domain-bearing tyrosine phosphatase 1. This likely explains the observed dephosphorylation of guanine nucleotide exchange factor and regulator of actin, Vav1, as well as ezrin-radixin-moesin proteins that connect actin filaments to membrane structures. In contrast, NK cell activation by NKG2D induced Vav1 and ezrin-radixin-moesin phosphorylation. Thus, CD94/NKG2A prevents actin-dependent recruitment of raft-associated activation receptors complexes to the activating synapse. This was further substantiated by showing that inhibition of actin polymerization abolished lipid rafts exclusion at the iNKIS, whereas cholesterol depletion had no effect on actin disruption at the iNKIS. These data indicate that the lipid rafts exclusion at the iNKIS is an active process which requires an intact cytoskeleton to maintain lipid rafts outside the inhibitory synapse. The net effect is to maintain an inhibitory state in the proximity of the iNKIS, while allowing the formation of activation synapse at distal points within the same NK cell.     

Exclusion of lipid rafts and decreased mobility of CD94/NKG2A receptors at the inhibitory NK cell synapse

CD94/NKG2A is an inhibitory receptor expressed by most human natural killer (NK) cells and a subset of T cells that recognizes human leukocyte antigen E (HLA-E) on potential target cells. To elucidate the cell surface dynamics of CD94/NKG2A receptors, we have expressed CD94/NKG2A-EGFP receptors in the rat basophilic leukemia (RBL) cell line. Photobleaching experiments revealed that CD94/NKG2A-EGFP receptors move freely within the plasma membrane and accumulate at the site of contact with ligand. The enriched CD94/NKG2A-EGFP is markedly less mobile than the nonligated receptor. We observed that not only are lipid rafts not required for receptor polarization, they are excluded from the site of receptor contact with the ligand. Furthermore, the lipid raft patches normally observed at the sites where FcepsilonR1 activation receptors are cross-linked were not observed when CD94/NKG2A was coengaged along with the activation receptor. These results suggest that immobilization of the CD94/NKG2A receptors at ligation sites not only promote sustenance of the inhibitory signal, but by lipid rafts exclusion prevent formation of activation signaling complexes.     

Kinetics and peptide dependency of the binding of the inhibitory NK receptor CD94/NKG2-A and the activating receptor CD94/NKG2-C to HLA-E.

The lytic function of human natural killer (NK) cells is markedly influenced by recognition of class I major histocompatibility complex (MHC) molecules, a process mediated by several types of activating and inhibitory receptors expressed on the NK cell. One of the most important of these mechanisms of regulation is the recognition of the non-classical class I MHC molecule HLA-E, in complex with nonamer peptides derived from the signal sequences of certain class I MHC molecules, by heterodimers of the C-type lectin-like proteins CD94 and NKG2. Using soluble, recombinant HLA-E molecules assembled with peptides derived from different leader sequences and soluble CD94/NKG2-A and CD94/NKG2-C proteins, the binding of these receptor-ligand pairs has been analysed. We show first that these interactions have very fast association and dissociation rate constants, secondly, that the inhibitory CD94/NKG2-A receptor has a higher binding affinity for HLA-E than the activating CD94/NKG2-C receptor and, finally, that recognition of HLA-E by both CD94/NKG2-A and CD94/NKG2-C is peptide dependent. There appears to be a strong, direct correlation between the binding affinity of the peptide-HLA-E complexes for the CD94/NKG2 receptors and the triggering of a response by the NK cell. These data may help to understand the balance of signals that control cytotoxicity by NK cells.     

2B4 (CD244) signaling via chimeric receptors costimulates tumor-antigen specific proliferation and in vitro expansion of human T cells

Regulatory NK cell receptors can contribute to antigen-specific adaptive immune responses by modulating T cell receptor (TCR)-induced T cell activation. We investigated the potential of the NK cell receptor 2B4 (CD244) to enhance tumor antigen-induced activation of human T cells. 2B4 is a member of the CD2 receptor subfamily with both activating and inhibitory functions in NK cells. In T cells, its expression is positively associated with the acquisition of a cytolytic effector memory phenotype. Recombinant chimeric receptors that link extracellular single-chain Fv fragments specific for the tumor-associated surface antigens CD19 and G_D2 to the signaling domains of human 2B4 and/or TCRζ were expressed in non-specifically activated peripheral blood T cells by retroviral gene transfer. While 2B4 signaling alone failed to induce T cell effector functions or proliferation, it significantly augmented the antigen-specific activation responses induced by TCRζ. 2B4 costimulation did not affect the predominant effector memory phenotype of expanding T cells, nor did it increase the proportion of T cells with regulatory phenotype (CD4+CD25^hiFoxP3+). These data support a costimulatory role for 2B4 in human T cell subpopulations. As an amplifier of TCR-mediated signals, 2B4 may provide a powerful new tool for immunotherapy of cancer, promoting sustained activation and proliferation of gene-modified antitumor T cells.     

Molecular basis of the dual functions of 2B4 (CD244)

2B4 belongs to the CD2 family of molecules and is expressed on all NK, gammadelta, and memory CD8(+) (alphabeta) T cells. The murine NK receptor 2B4 exhibits both inhibitory and activating functions, whereas human 2B4 has been reported to be an activating molecule. How murine 2B4 can act both as an activating and inhibitory receptor and what distinguishes its function from human 2B4 have remained largely unknown. We use here a model system that allows the study of human and murine 2B4 under identical and controlled conditions. These studies reveal that both human and mouse 2B4 can activate or inhibit NK cells. We show here that the level of 2B4 expression and the degree of 2B4 cross-linking play a significant role in the regulation of signaling lymphocyte activation molecule-associated protein-mediated activation by 2B4. A high level of 2B4 expression, heavy cross-linking, and relative paucity of signaling lymphocyte activation molecule-associated protein promote inhibitory function. Our studies demonstrate how a single receptor can have opposing function depending on the degree of receptor expression, extent of its ligation, and the relative abundance of certain adaptor molecules. Because the levels of 2B4 and CD48 are dynamically regulated, these findings have implications for the regulation of NK cell function.