Up on the tightrope natural killer cell activation and inhibition

There are no other known signaling motifs in the rather short cytoplasmic domains of these proteins, and mutation of the ITAM tyrosine residues abolishes their signaling function.

Typically, the ITAM-bearing signaling subunits and their receptors assemble through oppositely charged amino acids in their transmembrane regions. The ITAM-bearing proteins all possess an aspartate residue within their transmembrane region. Further studies are needed to understand the basis for this interaction. ITAM-bearing signaling subunits contain aspartate residues D within their transmembrane segments that associate noncovalently with oppositely charged lysine or arginine residues within the transmembrane of the receptors, an exception being CD16, which also has an aspartate residue within its transmembrane.

Y, tyrosine residues within ITAM domains. Engagement of these receptors causes the phosphorylation of the ITAM tyrosines, presumably by Src family kinases. NK cells express an abundance of Src family kinases, including Lck, Fyn, Src, Yes, Lyn and Fgr, and studies using gene-deficient mice suggest that their activities are probably redundant. In the mouse, Syk might be the predominant kinase 10 , whereas in humans both Syk and ZAP seem to couple efficiently with these ITAM-bearing subunits unpublished observations.

Therefore, the two major functions of NK cells—cytokine secretion and cytotoxicity—are regulated differentially downstream of ITAM-containing NK receptors. ITAM-bearing signaling subunits are phosphorylated, probably by Src family kinases, after receptor engagement.

The signaling pathways depicted are hypothetical and were deduced by synthesizing results from many studies investigating ITAM-coupled receptor signaling in human and mouse NK cells. ITAM-receptor activation induces actin cytoskeleton reorganization, which is required for cell polarization and release of the cytolytic granules containing perforin and granzymes, and results in the transcription of many cytokine and chemokine genes.

After activation, the human CD16 receptor complex is polyubiquitinated and degraded in the lysosomes and proteasomes by a process that might involve the action of the c-Cbl E3 ubiquitin ligase The fate of other stimulated ITAM-containing NK receptors has not been extensively investigated, and further studies should provide insights into how signaling through these receptors is regulated and terminated.

An arginine residue centrally located within the transmembrane region of NKG2D associates with the aspartate residue within the transmembrane domain of the DAP10 signaling subunit 25 Fig. The extracellular region of DAP10, a 10 kDa protein expressed as a disulfide-bonded homodimer, contains the cysteine residues required for dimer formation, as well as serine residues that serve as sites for O-linked glycosylation DAPmediated signaling in NK cells.

These events trigger distal signaling cascades as depicted. Because these two binding sites overlap, a single DAP10 chain will bind either p85 or Grb2, but not both. Mutagenesis studies have demonstrated that the tyrosine residue is essential for all signaling through DAP10, whereas mutation of the asparagine residue, which ablates Grb2 binding, or the methionine residue, which disrupts association with p85, results in different effects.

Using chimeric receptors transfected into primary NK cells, Leibson and colleagues demonstrated that mutation of either the pbinding site or the Grb2-binding site totally abrogates cytolytic activity initiated through DAP10 ref. The downstream consequences of the recruitment of Grb2 versus p85 after DAP10 cross-linking, however, are distinct. In accordance with these in vitro studies with human NK cells, Vav1-deficient mice show impaired DAPdependent, NKG2D-mediated cytotoxicity, and Vav1-deficient NK cells are unable to effectively reorganize their actin cytoskeleton or polarize their microtubule organizing center toward target cells 17 , 34 , Recent data indicate that DAP10 binds to p85 and induces production of phosphatidylinositol-3,4,5-trisphosphate in the immune synapse formed between NK cells and NKG2D ligand—expressing target cells Interestingly, recruitment of Grb2 to the immune synapse is not prevented by mutation of the Grb2-binding site in DAP10, but it is prevented when p85 signaling is blocked.

This suggests that Grb2 might be recruited to the immune synapse by DAP10 in two ways: directly, by binding to DAP10, and indirectly, possibly by recruitment of Sos1-Vav1-Grb2 complexes to phosphatidylinositol-3,4,5-trisphosphate—rich sites generated by activated PI3K at the immune synapse. The recruitment of Grb2 and p85 requires tyrosine phosphorylation of DAP Another candidate proposed for the phosphorylation of DAP10 is the kinase Jak3 ref.

In our own studies, we have observed that cytokines can be induced by cross-linking NKG2D on human NK cells, or on NK cells from DAPdeficient mice, with monoclonal antibodies to NKG2D, provided that plates are coated with a sufficiently high concentration of the agonist antibody unpublished observations.

Essentially all immature and mature human and mouse NK cells express CD Initially, mouse CD was considered an activating receptor on NK cells 51 ; however, subsequent studies suggested that in some circumstances CD might also inhibit NK cell effector functions In mice, two CD isoforms, designated long and short, are generated by alternative RNA splicing 52 ; however, these distinct isoforms do not exist in humans and their functional properties in mice remain unclear.

The ligand for CD is CD48, a cell surface glycoprotein expressed broadly on hematopoietic cells. CD receptor complexes in NK cells. SAP binds to Fyn to mediate signal transduction. Why in some situations cross-linking of CD in the absence of SAP causes no activation of human NK cells and in other cases causes inhibition has not been explained.

Tyrosine residues in the C-terminal region of EAT2 and ERT are phosphorylated in NK cells, and mutation of these tyrosines to phenylalanine abrogates the inhibitory function of these adaptors Although the model proposed by Veillette and colleagues to explain the divergent functions of CD is well supported by evidence, including functional studies in the SAP-and EATdeficient animal models, a recent study has reported that Fyn can interact with either SAP or EAT-2, questioning the conclusion that association of EAT-2 with CD results in exclusively inhibitory function by this receptor complex Studies of CDdeficient mice also suggest that the situation may be more complicated.

NK cells from CDdeficient mice demonstrate enhanced cytotoxicity and cytokine production when challenged with CDbearing tumor targets in vitro and in vivo Curiously, CDdeficient mice more efficiently reject B16 melanoma cell lines expressing CD48, but this phenomenon is only observed in male mice These results suggest that the inhibitory form of CD is predominant on mouse NK cells.

Notably, if activated NK cells mediate fratricide, this might explain the seemingly paradoxical findings that, in other situations, CDdeficient mice show impaired, rather than enhanced, NK cell responses The predominant CD receptor expressed on mature human NK cells is clearly activating or coactivating. Transfecting certain mouse cell lines with human CD48 renders them more sensitive to lysis by human NK cells 47 , and antibodies neutralizing CD or CD48 partially block the killing of target cells expressing CD48 ref.

Therefore, although an inhibitory role for CD has been clearly demonstrated in mouse NK cells and, in relatively infrequent situations, in human NK cells, the molecular basis of this inhibition has not been established. Both mouse and human CD have been reported to associate with LAT in some studies 71 , 73 , 74 but not others Consistent findings indicate that cross-linking CD on human NK cells induces phosphorylation of CD, Vav1, SHIP and c-Cbl; however, other aspects of the signaling pathway downstream of CD vary considerably depending on the experimental conditions or cell type used.

In this situation, KIRs block tyrosine phosphorylation of CD and also prevent actin-dependent reorientation of CD in the cell membrane. Therefore, the net contribution of CD to NK cell signaling in humans and mice is tightly regulated by the interplay between inhibiting and activating receptors and can be influenced by the density at which these receptors and their ligands are expressed on NK cells and on their targets, respectively. Table Click here to view.

Selective rejection of Hdeficient lymphoma variants suggests alternative immune defense strategy. Immune inhibitory receptors. Tessmer MS, et al. Stebbins CC, et al. Vav1 dephosphorylation by the tyrosine phosphatase SHP-1 as a mechanism for inhibition of cellular cytotoxicity. Protein kinase C regulates expression and function of inhibitory killer cell Ig-like receptors in NK cells.

Synergy among receptors on resting NK cells for the activation of natural cytotoxicity and cytokine secretion. Convergence on a distinctive assembly mechanism by unrelated families of activating immune receptors. McVicar DW, et al. DAPmediated signal transduction in natural killer cells. A dominant role for the Syk protein-tyrosine kinase. Colucci F, et al. Natural cytotoxicity uncoupled from the Syk and ZAP intracellular kinases.

Chiesa S, et al. Multiplicity and plasticity of natural killer cell signaling pathways. Dysregulation of signaling pathways in CDdeficient NK cells leads to differentially regulated cytotoxicity and cytokine production.

A requirement for CD45 distinguishes Ly49D-mediated cytokine and chemokine production from killing in primary NK cells. Tassi I, et al.

Cella M, et al. Billadeau DD, et al. The Vav-Rac1 pathway in cytotoxic lymphocytes regulates the generation of cell-mediated killing. Galandrini R, et al. Bauer S, et al. Jamieson AM, et al. The role of the NKG2D immunoreceptor in immune cell activation and natural killing.

Wu J, et al. Diefenbach A, et al. Selective associations with signaling proteins determine stimulatory versus costimulatory activity of NKG2D. Rosen DB, et al. Andre P, et al. The activating NKG2D receptor assembles in the membrane with two signaling dimers into a hexameric structure. DAP10 and DAP12 form distinct, but functionally cooperative, receptor complexes in natural killer cells. Upshaw JL, et al. Graham DB, et al. Vav1 controls DAPmediated natural cytotoxicity by regulating actin and microtubule dynamics.

Giurisato E, et al. Phosphatidylinositol 3-kinase activation is required to form the NKG2D immunological synapse. Zompi S, et al. NKG2D signaling is coupled to the interleukin 15 receptor signaling pathway. Meresse B, et al. Sutherland CL, et al. Kubin M, et al. Ehrlich LIR, et al. Veillette A. Valiante NM, Trinchieri G.

Identification of a novel signal transduction surface molecule on human cytotoxic lymphocytes. Tangye SG, et al. Bottino C, et al. Mathew PA, et al. Cloning and characterization of the 2B4 gene encoding a molecule associated with non-MHC-restricted killing mediated by activated natural killercells and T cells.

Schatzle JD, et al. Characterization of inhibitory and stimulatory forms of the murine natural killer cell receptor 2B4. Roncagalli R, et al. Calpe S, et al. Identification and characterization of two related murine genes, Eat2a and Eat2b , encoding single SH2-domain adapters. Cutting edge: functional requirement for SAP in 2B4-mediated activation of human natural killer cells as revealed by the X-linked lymphoproliferative syndrome.

Nakajima H, et al. Patients with X-linked lymphoproliferative disease have a defect in 2B4 receptor-mediated NK cell cytotoxicity. Parolini S, et al. X-linked lymphoproliferative disease. Bloch-Queyrat C, et al.

Lee KM, et al. Vaidya SV, et al. Targeted disruption of the 2B4 gene in mice reveals an in vivo role of 2B4 CD in the rejection of B16 melanoma cells. Eissmann P, et al. Calpe, S. Identification and characterization of two related murine genes, Eat2a and Eat2b , encoding single SH2-domain adapters. Immunogenetics 58 , 15—25 Clarkson, N. Cutting edge: functional requirement for SAP in 2B4-mediated activation of human natural killer cells as revealed by the X-linked lymphoproliferative syndrome.

Nakajima, H. Patients with X-linked lymphoproliferative disease have a defect in 2B4 receptor-mediated NK cell cytotoxicity. Parolini, S. X-linked lymphoproliferative disease. Bloch-Queyrat, C. Lee, K. Vaidya, S.

Targeted disruption of the 2B4 gene in mice reveals an in vivo role of 2B4 CD in the rejection of B16 melanoma cells. Taniguchi, R. Eissmann, P. Molecular basis for positive and negative signaling by the natural killer cell receptor 2B4 CD Molecular dissection of 2B4 signaling: implications for signal transduction by SLAM-related receptors. Watzl, C. NK cell inhibitory receptors prevent tyrosine phosphorylation of the activation receptor 2B4 CD Saborit-Villarroya, I.

Aoukaty, A. Role for glycogen synthase kinase-3 in NK cell cytotoxicity and X-linked lymphoproliferative disease. Chuang, S. Klem, J. Analysis of the molecular mechanism involved in 2B4-mediated NK cell activation: evidence that human 2B4 is physically and functionally associated with the linker for activation of T cells. Natural killer cell inhibitory receptors block actin cytoskeletal-dependent recruitment of 2B4 CD to lipid rafts.

Wahle, J. Cutting edge: dominance by an MHC-independent inhibitory receptor compromises NK killing of complex targets. Bhat, R. Serial killing of tumor cells by human natural killer cells — enhancement by therapeutic antibodies.

Download references. I thank D. Billadeau, A. Weiss, J. Djeu, J. Carlyle, C. Chang, J. Orange, S. Tangye and N. Bezman for discussions and help with the illustrations. Dedicated to the memory of my friend Paul Leibson who made significant contributions to our understanding of NK cell receptor signaling. You can also search for this author in PubMed Google Scholar. Correspondence to Lewis L Lanier. Natural killer NK cell function is regulated by a balance between activating and inhibitory receptors, but the details of this receptor interplay are not extensively understood.

Furthermore, both activating and inhibitory receptors operate in an additive way, suggesting that a fine-tuned balance between activating and inhibitory receptors regulate proximal NK cell signaling. We also demonstrate that murine NK cell expression of H2D d lowered the capacity of Ly49A to deliver inhibitory signals after antibody crosslinking, suggesting that the cis interaction between H2D d and Ly49A reduce the signaling capacity of Ly49A in this setting.

Our data shed new light on NK cell inhibition and raises new questions for further studies. Natural Killer NK cells are innate lymphocytes involved in the immune defense against viral infections and transformed cells 1 , and they also exert control over adaptive immune responses 2 , 3. NK cells mediate killing of target cells through the secretion of perforin, granzymes and the interaction of Fas-Fas-ligands 4 , 5 , and they also secrete a multitude of cytokines with immunomodulatory properties 6.

The effector functions of NK cells are regulated by signals coming from activating and inhibitory cell surface receptors, which are integrated inside the cells to determine the functional output 7 , 8.

In contrast to T cells, each of which possess a unique T cell receptor generated by somatic gene recombination, NK cells express several distinct types of germline-encoded activating receptors that are differentially expressed, signal differently and bind unique ligands on target cells 9.

A similar diversity can be found for inhibitory receptors for MHC class I. In mice these belong to the lectin-like Ly49 receptor family, whereas in humans they constitute immunoglobulin superfamily killer immunoglobulin-like receptors KIR The mechanisms underlying NK cell inhibition is complex, but the most studied mechanism involves a de-phosphorylation event acting on the intracellular signaling mediator VAV, which act downstream of activating NK cell receptors 15 , Furthermore, a recent study showed that inhibitory receptor engagement increased ubiquitilation of LAT, suggesting that inhibitory receptor signaling may also function by limiting the availability of signaling substrates inside the NK cells Given the central role for inhibitory receptors in NK cell function, surprisingly little is known about the nature of the activating and inhibitory signal cross-talk in NK cells.

It is assumed that intracellular signals transduced by the triggering of activating and inhibitory receptors are integrated through a cross-talk proximal to the cell membrane 8 , 17 , 18 , but this cross-talk has not been studied in much detail.

Furthermore, whether or not the inhibitory signals are binary all or none or dynamic and quantitative, if they are affected by the cis interaction with MHC class I or if they are influenced by cytokines that regulate NK cell function, such as IL, are remaining questions.

Using this assay, we provide several novel insights of relevance to the way by which inhibitory receptors may control NK cell function. It has been demonstrated that inhibitory receptor ligation exert proximal down-modulatory effects on signaling pathways downstream of activating receptors, but the nature of these inhibitory influences have not been extensively studied. We reasoned that this setup would allow us to investigate if inhibitory receptor triggering quantitatively downregulates NK cell activation, or if inhibition would operate in a threshold mode.

In a first step, we identified reagents that could be used to identify subsets of mouse NK cells and at the same time be used to cross-link activating and inhibitory receptors simultaneously Table S1. Following crosslinking of NK1. For both NK1. In line with a quantitative response to signaling strength, when these two activating receptors were co-crosslinked, NK cells displayed an additive enhanced calcium flux response, with both earlier onset and higher peak value in NK cell crosslinked via the two receptors simultaneously Figures 1E,F.

Figure 1. The colored lines depict various concentrations of the primary antibody. One representative experiment.

B Total area-under-the-curve AUC values after baseline correction see Materials and Methods from four independent experiments. Different colors indicate different experiments. Statistics calculated using a one-way paired Student's T -test.

C Same setup as in A but with an antibody against NKp D Same setup as in B but for NKp46 antibody. E Calciums flux response kinetics plot of the between Flou-4 and Fura-red after individual and simultaneous crosslinking of NK1.

Different crosslinkers were used for NK11 and NKp46 primary antibodies. Note lack of cross-reactivity beteen the crosslinkers flat lines. F Total AUC values after baseline correction see Material and Methods from 5 to 7 independent experiments, which are color-coded. Some experiments did not inluce all groups. We next tested if co-crosslinking NK1. To test this, we designed a co-staining protocol in which NK cells were double-stained with antibodies against activating receptors NK1.

In order for this setup to work as intended, it was necessary to make sure that flourochrome-labeled primary antibodies against inhibitory receptors could be used Figure S1. Because all inhibitory receptors in the mouse are expressed on subsets of NK cells, this setup would allow detection of a control subset and a test subset in the same sample, providing a critical test if inhibitory receptor triggering quantitatively downregulates NK cell triggering, or if it operates by means of a threshold effect.

NKG2A was not possible to use in these titration experiments, since staining was rapidly lost after titration and the cells could not be identified clearly using lower concentrations than Figure 2. Red line represents the response on the NK cell subset expressing the inhibitory receptor in question and the blue line represents the inhibitory receptor-negative subset in the same sample. Gates were set on plots in the top panel.

Corresponding concentrations of the antibodies in the various dilutions are indicated in Table S1. B Reduction of inhibitory influence on activating signals after titration of inibitory antibody. Top: staining of Ly49G2 at various antibody dilutions. Bottom: Kinetics plot of Fluo-4 fluorescence after NK1.