A balance between positive and negative signals in cytotoxic lymphocytes regulates the polarization of lipid rafts during the development of cell-mediated killing.

Plasma membrane microdomains containing sphingolipids and cholesterol (lipid rafts) are enriched in signaling molecules. The cross-linking of certain types of cell surface receptors initiates the redistribution of these lipid rafts, resulting in the formation of signaling complexes. However, little is known about the regulation of the initial raft redistribution and whether negative regulatory signaling pathways target this phase of cellular activation. We used natural killer (NK) cells as a model to investigate the regulation of raft redistribution, as both positive and negative signals have been implicated in the development of their cellular function. Here we show that after NK cells form conjugates with sensitive tumor cells, rafts become polarized to the site of target recognition. This redistribution of lipid rafts requires the activation of both Src and Syk family protein tyrosine kinases. In contrast, engagement of major histocompatibility complex (MHC)-recognizing killer cell inhibitory receptors (KIRs) on NK cells by resistant, MHC-bearing tumor targets blocks raft redistribution. This inhibition is dependent on the catalytic activity of KIR-associated SHP-1, a Src homology 2 (SH2) domain containing tyrosine phosphatase. These results suggest that the influence of integrated positive and negative signals on raft redistribution critically influences the development of cell-mediated cytotoxicity.

The Rho family guanine nucleotide exchange factor Vav-2 regulates the development of cell-mediated cytotoxicity.

Previous pharmacologic and genetic studies have demonstrated a critical role for the low molecular weight GTP-binding protein RhoA in the regulation of cell-mediated killing by cytotoxic lymphocytes. However, a specific Rho family guanine nucleotide exchange factor (GEF) that activates this critical regulator of cellular cytotoxicity has not been identified. In this study, we provide evidence that the Rho family GEF, Vav-2, is present in cytotoxic lymphocytes, and becomes tyrosine phosphorylated after the cross-linking of activating receptors on cytotoxic lymphocytes and during the generation of cell-mediated killing. In addition, we show that overexpression of Vav-2 in cytotoxic lymphocytes enhances cellular cytotoxicity, and this enhancement requires a functional Dbl homology and Src homology 2 domain. Interestingly, the pleckstrin homology domain of Vav-2 was found to be required for enhancement of killing through some, but not all activating receptors on cytotoxic lymphocytes. Lastly, although Vav and Vav-2 share significant structural homology, only Vav is able to enhance nuclear factor of activated T cells-activator protein 1-mediated gene transcription downstream of the T cell receptor. These data demonstrate that Vav-2, a Rho family GEF, differs from Vav in the control of certain lymphocyte functions and participates in the control of cell-mediated killing by cytotoxic lymphocytes.

Fc gamma receptor activation induces the tyrosine phosphorylation of both phospholipase C (PLC)-gamma 1 and PLC-gamma 2 in natural killer cells.

Crosslinking of the low affinity immunoglobulin G (IgG) Fc receptor (Fc gamma R type III) on natural killer (NK) cells initiates antibody-dependent cellular cytotoxicity. During this process, Fc gamma R stimulation results in the rapid activation of phospholipase C (PLC), which hydrolyzes membrane phosphoinositides, generating inositol-1,4,5-trisphosphate and sn-1,2-diacylglycerol as second messengers. We have recently reported that PLC activation after Fc gamma R stimulation can be inhibited by a protein tyrosine kinase (PTK) inhibitor. Based on the paradigm provided by the receptor tyrosine kinases, we investigated whether PLC-gamma 1 and/or PLC-gamma 2 are expressed in NK cells, and whether the PLC-gamma isoforms are tyrosine phosphorylated in response to Fc gamma R stimulation. Immunoblotting analyses with PLC-gamma 1- and PLC-gamma 2-specific antisera demonstrate that both isoforms are expressed in human NK cells. Furthermore, Fc gamma R crosslinking triggers the tyrosine phosphorylation of both PLC-gamma 1 and PLC-gamma 2 in these cells. Phosphorylation of both isoforms is detectable within 1 min, and returns to basal level within 30 min. Pretreatment with herbimycin A, a PTK inhibitor, blocked the Fc gamma R-induced tyrosine phosphorylation of PLC-gamma 1 and PLC-gamma 2, and the subsequent release of inositol phosphates. These results suggest that Fc gamma R-initiated phosphoinositide turnover in human NK cells is regulated by the tyrosine phosphorylation of PLC-gamma. More broadly, these observations demonstrate that nonreceptor PTK(s) activated by crosslinkage of a multisubunit receptor can phosphorylate both PLC-gamma isoforms.

Fc receptor stimulation of phosphatidylinositol 3-kinase in natural killer cells is associated with protein kinase C-independent granule release and cell-mediated cytotoxicity.

Although diverse signaling events are initiated by stimulation of multichain immune recognition receptors on lymphocytes, it remains unclear as to which specific signal transduction pathways are functionally linked to granule exocytosis and cellular cytotoxicity. In the case of natural killer (NK) cells, it has been presumed that the rapid activation of protein kinase C (PKC) enables them to mediate antibody-dependent cellular cytotoxicity (ADCC) and "natural" cytotoxicity toward tumor cells. However, using cloned human NK cells, we determined here that Fc receptor stimulation triggers granule release and ADCC through a PKC-independent pathway. Specifically, pretreatment of NK cells with the selective PKC inhibitor, GF109203X (using concentrations that fully blocked phorbol myristate acetate/ionomycin-induced secretion) had no effect on FcR-initiated granule release or ADCC. In contrast, FcR ligation led to the rapid activation of phosphatidylinositol 3-kinase (PI 3-kinase), and inhibition of this enzyme with the selective inhibitor, wortmannin, blocked FcR-induced granule release and ADCC. Additional experiments showed that, whereas FcR-initiated killing was wortmannin sensitive and GF109203X insensitive, natural cytotoxic activity toward the tumor cell line K562 was wortmannin insensitive and GF109203X sensitive. Taken together, these results suggest that: (a) PI 3-kinase activation induced by FcR ligation is functionally coupled to granule exocytosis and ADCC; and (b) the signaling pathways involved in ADCC vs natural cytotoxicity are distinct.

The Vav-Rac1 pathway in cytotoxic lymphocytes regulates the generation of cell-mediated killing.

The Rac1 guanine nucleotide exchange factor, Vav, is activated in hematopoietic cells in response to a large variety of stimuli. The downstream signaling events derived from Vav have been primarily characterized as leading to transcription or transformation. However, we report here that Vav and Rac1 in natural killer (NK) cells regulate the development of cell-mediated killing. There is a rapid increase in Vav tyrosine phosphorylation during the development of antibody-dependent cellular cytotoxicity and natural killing. In addition, overexpression of Vav, but not of a mutant lacking exchange factor activity, enhances both forms of killing by NK cells. Furthermore, dominant-negative Rac1 inhibits the development of NK cell-mediated cytotoxicity by two mechanisms: (a) conjugate formation between NK cells and target cells is decreased; and (b) those NK cells that do form conjugates have decreased ability to polarize their granules toward the target cell. Therefore, our results suggest that in addition to participating in the regulation of transcription, Vav and Rac1 are pivotal regulators of adhesion, granule exocytosis, and cellular cytotoxicity.

Functional role for Syk tyrosine kinase in natural killer cell-mediated natural cytotoxicity.

Natural killer (NK) cells are named based on their natural cytotoxic activity against a variety of target cells. However, the mechanisms by which sensitive targets activate killing have been difficult to study due to the lack of a prototypic NK cell triggering receptor. Pharmacologic evidence has implicated protein tyrosine kinases (PTKs) in natural killing; however, Lck-deficient, Fyn-deficient, and ZAP-70-deficient mice do not exhibit defects in natural killing despite demonstrable defects in T cell function. This discrepancy implies the involvement of other tyrosine kinases. Here, using combined biochemical, pharmacologic, and genetic approaches, we demonstrate a central role for the PTK Syk in natural cytotoxicity. Biochemical analyses indicate that Syk is tyrosine phosphorylated after stimulation with a panel of NK-sensitive target cells. Pharmacologic exposure to piceatannol, a known Syk family kinase inhibitor, inhibits natural cytotoxicity. In addition, gene transfer of dominant-negative forms of Syk to NK cells inhibits natural cytotoxicity. Furthermore, sensitive targets that are rendered NK-resistant by major histocompatibility complex (MHC) class I transfection no longer activate Syk. These data suggest that Syk activation is an early and requisite signaling event in the development of natural cytotoxicity directed against a variety of cellular targets.

Signal transduction through the T-cell antigen receptor.

The mechanism by which ligand binding to the T-cell antigen receptor triggers the T-cell activation program has long been one of the most fascinating questions in lymphocyte biology. Here, we review recent insights into the transmembrane signaling functions of this multisubunit receptor complex.

The expression of Fas Ligand by macrophages and its upregulation by human immunodeficiency virus infection.

Fas/Fas Ligand (FasL) interactions play a significant role in peripheral T lymphocyte homeostasis and in certain pathological states characterized by T cell depletion. In this study, we demonstrate that antigen-presenting cells such as monocyte-derived human macrophages (MDM) but not monocyte-derived dendritic cells express basal levels of FasL. HIV infection of MDM increases FasL protein expression independent of posttranslational mechanisms, thus highlighting the virus-induced transcriptional upregulation of FasL. The in vitro relevance of these observations is confirmed in human lymphoid tissue. FasL protein expression is constitutive and restricted to tissue macrophages and not dendritic cells. Moreover, a significant increase in macrophage-associated FasL is observed in lymphoid tissue from HIV (+) individuals (P < 0.001), which is further supported by increased levels of FasL mRNA using in situ hybridization. The degree of FasL protein expression in vivo correlates with the degree of tissue apoptosis (r = 0.761, P < 0. 001), which is significantly increased in tissue from HIV-infected patients (P < 0.001). These results identify human tissue macrophages as a relevant source for FasL expression in vitro and in vivo and highlight the potential role of FasL expression in the immunopathogenesis of HIV infection.

Effects of p56lck deficiency on the growth and cytolytic effector function of an interleukin-2-dependent cytotoxic T-cell line.

The growth, differentiation, and functional activities of antigen-stimulated T lymphocytes are regulated by the interaction of the T-cell-derived cytokine, interleukin-2 (IL-2), with the high-affinity IL-2 receptor (IL-2R). IL-2R occupancy initiates a rapid increase in intracellular protein tyrosine phosphorylation, suggesting that a receptor-coupled protein tyrosine kinase (PTK) serves as a proximal signaling element for the IL-2R. Previous studies implicated the src-family kinase, p56lck, as a potential IL-2R-linked signal transducer. In this study, we have characterized a spontaneous variant of the IL-2-dependent cytotoxic T-cell line, CTLL-2, which contains no detectable lck-derived mRNA transcripts, protein, or PTK activity. The p56lck-deficient CTLL-2 cells retained strict dependence on IL-2 for both viability and growth, indicating that p56lck activity was not required for the transduction of IL-2-mediated mitogenic signals. However, the p56lck-deficient cells exhibited a moderate decrease in their rate of IL-2-dependent proliferation. In contrast to this relatively modest proliferative defect, the p56lck-deficient cell line displayed a profound reduction in T-cell antigen receptor-dependent cytolytic effector functions. Both the proliferative and the cytolytic defects observed in the p56lck-deficient cells were completely reversed by transfection of these cells with a wild-type lck expression vector. These results indicate that p56lck expression is not obligatory for IL-2-mediated T-cell growth stimulation; however, this PTK plays a central role in the generation T-cell-mediated cytotoxic responses.

Genetic evidence for differential coupling of Syk family kinases to the T-cell receptor: reconstitution studies in a ZAP-70-deficient Jurkat T-cell line.

T-cell antigen receptor (TCR) engagement activates multiple protein tyrosine kinases (PTKs), including the Src family member, Lck, and the Syk-related PTK, ZAP-70. Studies in ZAP-70-deficient humans have demonstrated that ZAP-70 plays crucial roles in T-cell activation and development. However, progress toward a detailed understanding of the regulation and function of ZAP-70 during TCR signaling has been hampered by the lack of a suitable T-cell model for biochemical and genetic analyses. In this report, we describe the isolation and phenotypic characterization of a Syk- and ZAP-70-negative somatic mutant derived from the Jurkat T-cell line. The P116 cell line displays severe defects in TCR-induced signaling functions, including protein tyrosine phosphorylation, intracellular Ca2+ mobilization, and interleukin-2 promoter-driven transcription. These signaling defects were fully reversed by reintroduction of catalytically active versions of either Syk or ZAP-70 into the P116 cells. However, in contrast to ZAP-70 expression, Syk expression triggered a significant degree of cellular activation in the absence of TCR ligation. Transfection experiments with ZAP-70-Syk chimeric proteins indicated that both the amino-terminal regulatory regions and the carboxy-terminal catalytic domains of Syk and ZAP-70 contribute to the distinctive functional properties of these PTKs. These studies underscore the crucial role of ZAP-70 in TCR signaling and offer a powerful genetic model for further analyses of ZAP-70 regulation and function in T cells.

Changes in susceptibility to cytotoxic antibody among tumor cells surviving exposure to chemotherapeutic agents.

We determined that myeloma cells which survived drug treatment had an altered sensitivity to cytotoxic antibody. The effects of several chemotherapeutic agents differing in drug action were compared. The antiserum was directed against viral determinants on the surface of S107 myeloma cells. This antiserum was then used to inhibit the colony formation of drug-treated myeloma cells. Tumor cells were treated with melphalan (200 ng/ml), methotrexate (40 ng/ml), actinomycin D (5 ng/ml), or 0.5 mM thymidine for 24 hr and then washed and resuspended in fresh medium. On various days after drug treatment, aliquots of these cells were exposed to complement and dilutions of antiserum and then cloned in soft agar in order to quantify the degree of antibody-mediated inhibition. Melphalan, methotrexate, and thymidine caused a severalfold increased resistance of the tumor cells to the antiserum during the first 1 or 2 days after drug treatment. During the following days, however, myeloma cells showed markedly increased susceptibility to the antiserum. The biphasic effect of methotrexate or thymidine treatment was similar to that previously observed after melphalan treatment and differed from the effect of actinomycin D. Actinomycin D caused only an increased susceptibility of the tumor cells to the antibody for a period of 4 to 5 days immediately following drug treatment. Our studies indicate that several chemotherapeutic drugs at concentrations comparable to those used in humans have long-lasting antagonistic as well as synergistic effects on the sensitivity of tumor cells to antibody and complement, depending on the particular drug used and on the time interval between drug exposure and subsequent treatment with antibody. These results suggest a model for evaluating the use of antibody in the elimination of malignant cells which, despite exposure to chemotherapy, remain clonogenic and proliferative.

Nonrandom escape of tumor cells from immune lysis due to intraclonal fluctuations in antigen expression.

Considerable heterogeneity in the amount of surface antigen can regularly be demonstrated by cytofluorometric analysis among genetically identical cells in a tumor clone. We have used monoclonal idiotype-specific antibodies to investigate the patterns of change in amounts of an idiotypic tumor antigen and how such changes affect the immune escape of malignant B cells expressing this antigen. By the use of fluorescence-activated cell sorting, we separated cells expressing either very large or very small amounts of the idiotypic target antigen and then analyzed these subpopulations of tumor cells at various times after isolation for expression of idiotype. We found that the differences in amount of antigen expression were not heritable, and that over a period of about 7 days of continuous growth in vitro, the fluorescence-activated cell sorted populations gradually came to express normal amounts of idiotype. The mechanisms regulating the quantity of surface idiotype were independent of those affecting the amounts of other membrane-associated molecules such as H-2 antigen. Furthermore, these nonheritable intraclonal differences in amounts of antigen expression were unrelated to stages of the cell cycle but clearly did affect the susceptibility of the cells from immune lysis. Thus, tumor cells expressing the lowest amount of surface idiotype were much more resistant to the lytic effects of antiidiotypic antibody and complement but had the same cell cycle distribution as did unseparated cells. These results demonstrate that nonheritable, non-cell cycle-related heterogeneity in amount of tumor antigen expression can significantly determine which cells in a cloned malignant cell population preferentially escape monoclonal tumor-specific antibody therapy.

Signal transduction during NK cell activation: balancing opposing forces.

Significant progress has been made in our understanding of the basic signaling mechanisms regulating NK cell activation. Advances have been fueled in part by the molecular characterization of specific activating receptors (e.g., the Fc gamma RIII multi-subunit complex) and inhibitory receptors (e.g., novel MHC-recognizing inhibitory receptors). However, certain aspects of these analyses are complicated by the heterogeneous nature of the receptor-ligand interactions utilized during the development of a cytotoxic response. Future advances will depend in part on the further molecular characterization of the involved receptors and second messengers and on the development of experimental models for genetically manipulating the signaling elements. It will remain important to understand both activating and inhibitory signaling pathways as the emerging theme is that the balance of these two opposing forces determines the functional outcome of an NK cells interaction with its target.

Demonstration of clonality, by X-linked DNA analysis, in chronic natural killer cell lymphocytosis and successful therapy with oral cyclophosphamide.

The expanded lymphocyte population in large granular lymphocyte (LGL)-leukemia carries the phenotypic characteristics of either cytotoxic T lymphocytes (CD3+,CD8+) or natural killer (NK) cells (CD3-,CD15+). In the former subset, clonality has been demonstrated by T-cell receptor gene rearrangement studies. Since NK cells do not rearrange T-cell receptor genes, the neoplastic nature of chronic NK cell lymphocytosis has not been well defined. We used X-linked DNA analysis to study the clonal nature of an expanded NK cell population in a patient with a 3-year history of relative lymphocytosis associated with anemia and neutropenia. Southern blot analysis showed no clonal T-cell receptor gene rearrangement. The majority of the circulating lymphocytes had a NK cell phenotype and demonstrated both direct NK cell-mediated cytotoxicity and antibody-dependent cellular cytotoxicity. However, the in vitro growth characteristics of these cells did not suggest that they were polyclonal expansions of normal NK cells. To determine directly the clonal origin of these cells, we performed X-linked DNA analysis. Density gradient centrifugation methods were used to isolate mononuclear cells, and NK cells were positively selected by CD16-immunoconjugated magnetic beads. The DNA of these cells was analyzed by restriction fragment length polymorphism-methylation strategy and showed a monoclonal pattern of X-chromosome inactivation while a polyclonal pattern was obtained in corresponding skin tissue. Treatment of the patient with oral cyclophosphamide resulted in complete hematologic remission. We conclude that chronic NK lymphocytosis may be clonal and responsive to immunosuppressive therapy.

Interleukin-2 signal transduction in human NK cells: multisite phosphorylation and activation of the tyrosine kinase p56lck.

Interleukin-2 (IL-2) potently stimulates natural killer (NK) cell proliferation and cytotoxic function. However, the molecular mechanisms by which IL-2 delivers activation signals from the IL-2 receptor to the NK cell interior are incompletely understood. Previous studies demonstrated that IL-2 stimulation induced the tyrosine phosphorylation of multiple proteins in NK cells, together with a prominent reduction in the electrophoretic mobility of p56lck. The present studies indicate that IL-2 induces a rapid (< or = 1 min) increase in the catalytic activity of p56lck, as measured by increases in protein tyrosine kinase activity in vitro. Furthermore, in response to IL-2, p56lck itself undergoes complex alterations in serine and tyrosine phosphorylation. Cyanogen bromide cleavage maps indicate that IL-2 stimulates a pronounced increase in the phosphorylation of the NH2-terminal region of p56lck containing multiple known sites of serine phosphorylation. In addition, IL-2 induced a marked increase in the phosphorylation of a COOH-terminal peptide containing the regulatory Tyr-505 residue of p56lck. These results suggest that p56lck serves as a substrate for both protein serine and tyrosine kinases activated during stimulation of this cell type with IL-2. Furthermore, these results indicate that the pleiotropic effects of IL-2 on NK cell physiology are initiated and regulated by a complex and multitiered interaction of different protein kinases including p56lck.

Natural killer cells and the syndrome of chronic natural killer cell lymphocytosis.

Natural killer (NK) cells provide anti-infectious, anti-neoplastic, and immunomodulatory function effected by both cytokine production and direct cellular cytotoxicity that is not major histocompatibility complex-restricted. NK cells lack truly specific cell surface determinants as well as antigen-specific receptors. Recent information suggests a variety of receptor-ligand interactions that underlie recognition and treatment of target cells by NK cells. Primary NK cell disorders in humans are currently classified into NK cell lymphomas and chronic NK cell lymphocytosis (CNKL). In this review, we summarize current understanding of the biology of NK cells and describe the clinical manifestations of CNKL.