Activity of human natural killer cells against target cells possessing different sensitivity to interferon]

The cytotoxic activity of peripheral blood natural killers (NK) against target cells (TC) J-96 and L-929 with high sensitivity to interferon (IFN) action, J-41 and MCB resistant to IFN action and line K-562 labelled by H3-uridine was studied in 14 hrs cytotoxic test. It has been shown that human TC J-96 didn't differ from the J-41 in their sensitivity to NK cytotoxicity and they are strongly resistant to NK than TC K-562. The murine TC L-929 as the human TC didn't differ from the MCB in their sensitivity to NK lysis and had also the same sensitivity to NK as the K-562 cells.     

Characteristics of natural killer cells (NK cells) and features of the cytotoxic test in Syrian hamsters

The cytotoxic test in vitro with the use of xenogeneic target cells of human myeloma, strain K-562, labeled with 51Cr has demonstrated natural cytotoxicity of lymphoid cells from noninbred Syrian hamsters. This cytotoxicity occurs at the cost of non-adherent splenocytes. NK may be isolated over the gradient density of ficoll (1.078), selective for large granular lymphocytes. To detect the maximal lytic activity of NK from Syrian hamsters in the cytotoxic test in vitro, they should be brought into 10-12 hour contact with sensitive target cells K-562. In Syrian hamsters, the highest natural cytotoxicity is shown by the cells of the blood and spleen. In the bone marrow and thymus, it is little pronounced and is virtually absent from the peripheral lymph nodes.     

Interferon and butyrate treatment leads to a decreased sensitivity of NK target cells to lysis by homologous but not by heterologous effector cells

Human K-562 and HHMS cells were pretreated with human recombinant interferon (IFN)-gamma and used as targets in NK assays against human and murine effector cells. A protective effect against NK lysis was observed only in the homologous assay, whereas no change or even a slight increase in NK sensitivity against heterologous effector cells was found. In cold target inhibition experiments IFN-treatment of K-562 cells led to a decrease in their capacity to act as competitors in the homologous NK assay, leaving their inhibitory capacity unaltered in the heterologous assay. In accordance with results observed using human NK targets, murine YAC-1 cells treated with mouse recombinant IFN-gamma did not lose their susceptibility to human NK cells. However, they were markedly less susceptible to lysis mediated by murine effectors. Butyrate, another compound causing decreased sensitivity of K-562 cells for human natural killing, also failed to reduce the susceptibility against murine NK cells. The results indicate that the NK-resistant tumor target phenotype caused by IFN or differentiation-inducing agents can only be detected by homologous but not by heterologous effector cells. This suggests that major differences exist between the inter- and intraspecies NK killing mechanisms.     

Natural killers in pregnancy: the problem of heterogeneity and the regulation of activity

Natural killer activity of pregnant women's peripheral blood lymphocytes and of cord blood was investigated. 3H-uridine labeled K-562 and human embryo fibroblasts (HEF) were used as target cells in cytotoxic test. The results of competitive inhibition test led us to a conclusion about the presence of some common K-562 and HEF surface structures recognized by NK cells. It was shown that the decline of NK activity of pregnant women and a low NK activity of cord blood were not associated with the influence of T-lymphocytes or adherent cells.     

Role of interferon in natural kill of HSV-1-infected fibroblasts

The production of interferon during natural killer (NK) assays against HSV-1-infected fibroblasts (NK(HSV-1)) was studied to determine whether this interferon was responsible for inducing the preferential lysis of herpes-virus-infected target cells over uninfected target cells. The interferon produced during NK(HSV-1) assays was analyzed and found to have the properties of HU-IFN-alpha. Little or no IFN was produced during NK assays against uninfected fibroblasts (NK(FS)) or K562 (NK(K562)) cells. Although the appearance of interferon in the culture supernatants seemed to parallel the development of cytotoxicity during NK(HSV-1) assays, the levels of cytotoxicity and IFN generated did not correlate, arguing against a strict quantitative dependence of cytotoxicity upon IFN production. NK(K562) and NK(FS) cytotoxicity developed with little or no production of IFN. When IFN-pretreated effector cells were used, there was still a preferential lysis of infected over uninfected target cells. This preferential lysis by IFN-treated effector cells of infected over uninfected targets was seen as early as 2 hr into the assay. Anti-IFN antibodies added to the NK assays, although neutralizing all the IFN produced during the assays, had no effect on NK(FS) or NK(K562) cytotoxic activity and caused a slightly reduction of NK(HSV-1) activity only in one of three experiments. We conclude that although IFN is generated during NK(HSV-1) assays, this IFN cannot solely account for the increased lysis of infected over uninfected cells and that NK(HSV-1) activity is in some other way dependent on the virus infection.     

Long-term killing of natural killer-resistant target cells by interferon-alpha-, interferon-gamma-, and interleukin-2-activated natural killer cells

The sensitivity of target cells to natural killer (NK) cell-mediated cytotoxicity was investigated. Five target cell lines were examined for susceptibility to killing by activated NK cells in a 4-hour cytotoxicity assay: one of them (K562) was highly sensitive, while the other four were resistant. However, the four NK-resistant target cell lines were fully susceptible to lysis when the assay was extended to 24 h. The cytotoxic cells that killed the NK-resistant target cells in a 24-hour assay were plastic- and nylon wool-nonadherent human peripheral blood mononuclear cells (PBMC) and their cytotoxicity was increased by interferon-alpha, interferon-gamma, and interleukin-2. Further, the cytotoxic activity of PBMC in the long-term assay was associated with large granular lymphocytes purified on a Percoll gradient, that killed the NK-sensitive cell line K562 in a 4-hour assay. All of the above are general criteria to qualify the cytotoxic cells as NK cells. Thus, the NK-resistant phenotype may not reflect absolute immunity to NK-mediated lysis, but it may reflect the different rates at which various target cell lines can be killed.     

Expression of conformationally constrained adhesion peptide in an antibody CDR loop and inhibition of natural killer cell cytotoxic activity by an antibody antigenized with the RGD motif.

We report that an antibody engineered to express three Arg-Gly-Asp (RGD) repeats in the third complementarity-determining region of the heavy chain (antigenized antibody) efficiently inhibits the lysis of human erythroleukemia K-562 cells by natural killer (NK) cells. Synthetic peptides containing RGD did not inhibit. Inhibition was specific for the (RGD)3-containing loop and required simultaneous occupancy of the Fc receptor (CD16) on effector cells. The antigenized antibody inhibited other forms of cytotoxicity mediated by NK cells but not cytotoxicity mediated by major histocompatibility complex-restricted cytotoxic T lymphocytes (CTL). A three-dimensional model of the engineered antibody loop shows the structure and physicochemical characteristics probably required for the ligand activity. The results indicate that an RGD motif is involved in the productive interaction between NK and target cells. Moreover, they show that peptide expression in the hypervariable loops of an antibody molecule is an efficient procedure for stabilizing oligopeptides within a limited spectrum of tertiary structures. This is a new approach towards imparting ligand properties to antibody molecules and can be used to study the biological function and specificity of short peptide motifs, including those involved in cell adhesion.     

Surface receptors and immune activity of purified human circulating mononuclear cells. III. The mechanisms of lysis, by lymphocytes and monocytes, of target cells in the lymphocyte-dependent and monocyte-dependent ADCC, NK, NOCC, and MICC cytotoxic reactions

The abilities of unfractionated mononuclear cells (MNC), monocytes (98-99% pure), and lymphocytes (98-99% pure) to carry out the lysis of target cells in the ADCC, NK, NOCC, and MICC assays were compared. Lymphocytes by themselves were able to lyse the CRBC (ADCC), K-562 (NK), and RRBC (MICC) target cells. The monocytes were very effective in the lysis of the CRBC (MICC) target cells. However, the lysis of two other target cells--RRBC (NOCC) and HRBC (ADCC)--required the simultaneous presence of both lymphocytes and monocytes in order to effect optimal lysis. Soluble factor(s) secreted by the cytotoxic cells capable of lysing the target cells were detected only in the NK assay. The activity of the soluble cytotoxic factor (NKCF) was only 25-40% of that exhibited by the cytotoxic NK cells and it was secreted by the cytotoxic cells after 48 hr of culture and not 24 hr of culture which is the usual assay condition. The NKCF was cytotoxic only to the NK target cells and not to the target cells used in the ADCC, NOCC, and MICC cytotoxic assays. Different classes of lymphocytes were cytotoxic in the monocyte-independent assays [ADCC (CRBC), NK (K-562), and MICC (RRBC)]. The null lymphocytes and the T lymphocytes were the primary cytotoxic cells in the ADCC and MICC assays, respectively, whereas the T, B, and null cells were almost equally cytotoxic in the NK assay. With respect to the monocyte-dependent assays [ADCC (HRBC), NOCC (RRBC), and MICC (CRBC)], the cytotoxic activity of any one class of lymphocytes failed to approach that of the unfractionated MNC. The T cells were the most cytotoxic; the B cells exhibited limited cytotoxic activity in only the ADCC assay and the null cells showed no cytotoxic activity. However, the combination of T and non-T cells and, to a lesser extent, T and B cells, exhibited much greater cytotoxic activity than the individual cells and together were as cytotoxic as the unfractionated MNC. It is concluded that, depending upon the selection of the target cells, lysis in the ADCC, NK, NOCC, and MICC assays may be effected by lymphocytes only, by monocytes only, by both monocytes and lymphocytes, or as a result of lymphocyte-monocyte collaboration. In the latter instance more than one class of lymphocytes must be present in order for maximum cytotoxic activity to be expressed.     

Scanning electron microscopic study of natural killer cell-mediated cytotoxicity

The interaction between human natural killer (NK) cells and NK-susceptible target cells, as well as the mechanism involved in target cell lysis, were studied with scanning electron microscopy (SEM). Low density human peripheral blood lymphocytes, highly enriched with large granular lymphocytes (LGL), were used as effector cells, and K562-cells were used as NK-susceptible target cells. The surface features of LGL/NK cells were examined under SEM. In the area of interaction, NK/target-cell conjugates showed microvilli and/or filipodia, and extensive areas of intercellular contact. In addition, the effector cells in some NK/target-cell conjugates were polarized toward the target cell. Changes in target cell surface features included loss of microvilli, large surface blebs and the appearance of small pore-like lesions on the cell membrane. Our findings show that target cell lysis occurred by apoptosis and plasma membrane lesions analogous to those seen during complement-mediated cytotoxicity.     

Monocyte- and natural killer cell-mediated spontaneous cytotoxicity against human noncultured solid tumor cells

Unstimulated human peripheral blood mononuclear cells from healthy donors exhibited spontaneous cytotoxicity against noncultured solid tumor targets in a 12- to 24-hr 51Cr release or 111In release assay. Both purified monocytes (greater than 99% monocytes) and natural killer (NK)-enriched lymphocytes exhibited comparable levels of spontaneous cytotoxicity against fresh melanoma tumor targets. This cytotoxicity was observed under endotoxin-free conditions. NK-depleted lymphocytes did not lyse the melanoma targets. Culture supernatants of monocytes incubated with the melanoma tumor cells did not exhibit cytotoxic activity against these targets. Purified monocytes lacked NK activity against the K562 targets in a 4-hr 51Cr release assay. Treatment of the monocytes with anti-Leu 1 1b and anti-Leu7 monoclonal antibodies plus complement did not reduce monocyte-mediated lysis of the melanoma targets, demonstrating that contaminating NK cells, if any, were not responsible for the lysis of noncultured melanoma targets by monocytes. In contrast, Leu 1 1b+ NK cells were responsible for the lysis of the melanoma targets by NK-enriched lymphocytes. The addition of recombinant interferon-gamma (rIFN-gamma), but not lipopolysaccharide, into the 51Cr release assay or pretreatment of monocytes with rIFN-gamma significantly increased their cytotoxicity against noncultured solid tumor cells. Monocytes cultured for 3 days with medium alone lost their cytotoxic activity. The addition of rIFN-gamma from the beginning of these cultures prevented the loss of the cytotoxic activity of monocytes. In summary, both unstimulated monocytes and NK-enriched lymphocytes exhibit comparable levels of spontaneous cytotoxicity against fresh solid tumor targets.     

Spontaneous cell-mediated cytotoxicity in patients with transitional cell carcinoma of the urinary bladder

Summary Lymphocytes isolated from the blood of TCC patients, like those of control patients, were capable of mediating spontaneous cell-mediated cytotoxicity against K-562 cells. When this natural cytotoxicity was analyzed with regard to the effector cell type it was found that in TCC patients the SLMC was mostly displayed by E-rosetting T lymphocytes, whereas compared with controls, a significant decline in the SLMC of the non-T lymphocytes was observed. The SLMC of the T lymphocytes derived from TCC patients was further demonstrated on a T leukemia target cell (Peer). When the SLMC on K-562 and on Peer target cells was compared, a specificity difference was observed between TCC and the control patients' effector cells. The SLMC activity of the TCC patients' T cells was not abolished after depletion of Fc receptor-positive cells or following treatment with monoclonal antibodies OKT 8 or OKT 4 and complement (C'). These NK-like cells are therefore distinguished from cytotoxic T lymphocytes and NK cells.     

Anti-idiotype antibodies to the 9.1C3 blocking antibody used to probe the lethal hit stage of NK cell-mediated cytolysis

We previously described a monoclonal antibody, 9.1C3, which blocked natural killer (NK) cell-mediated cytolysis by acting on effector cells during a late step in the lethal hit stage. The present work describes the production in rabbits of anti-idiotypic (anti-id) antibodies to the 9.1C3 antibody. In addition to reacting specifically with the 9.1C3 antibody, the anti-id antibodies bound strongly to the K562 target cell. The anti-id antibodies blocked killing of K562 targets by NK, antibody-dependent cellular cytotoxicity, and NK-like cells but did not inhibit killing by cytotoxic T lymphocytes (CTL). Pretreatment of cells and washing before assay indicated that blocking occurred at the target cell level. Of particular interest, single cell assays with Percoll-enriched large granular lymphocytes demonstrated that the antibodies caused no reduction in binding. These data are consistent with a model for NK cell-mediated lysis that involves a secondary target cell receptor independent of the primary NK-target cell interaction. The anti-id antibodies immunoprecipitated cell surface proteins of relative m.w. 79K and 62K unreduced, and 94K and 79K reduced from K562 target cells. The development of anti-id antibodies may be a useful procedure to explore the structure and function of cellular receptors involved in NK cell-mediated cytolysis.     

K562 killing by K, IL 2-responsive NK, and T cells involves different effector cell post-binding trigger mechanisms

The monoclonal antibody 13.3 specifically blocks the trigger process of the NK-K562 cytolytic sequence at a post-binding effector cell level. This antibody was used to define differences in the lytic trigger processes of NK and other mechanisms of K562 lysis. Monoclonal antibody 13.3 inhibited lysis of K562 target cells by freshly isolated peripheral blood lymphocytes (PBL) and purified large granular lymphocytes (LGL), but had no inhibitory effect on antibody-dependent cell-mediated cytotoxicity to K562 by these effectors. Lectin-dependent cellular cytotoxicity (LDCC) to this target cell was also unresponsive to 13.3. The 13.3-induced inhibition of NK-K562 lytic activity persisted when PBL were activated in culture with interleukin 2 (IL 2) for periods up to 48 hr. After 48 hr of culture, the degree of inhibition diminished progressively in medium containing fetal calf serum but not in medium containing autologous serum. This 13.3-unresponsive lytic activity in cultured PBL could be attributed to more than one cell type and was present in both the LGL and Fc gamma receptor-depleted T cell fraction. Thus, K562 lysis by freshly isolated human lymphocytes via NK, K, and LDCC mechanisms is characterized by heterogeneity of the post-binding effector cell trigger mechanism. K562 lysis by lymphocytes cultured with IL 2 is similarly heterogeneous.     

Cytotoxic activity against trophoblast and choriocarcinoma cells of large granular lymphocytes from human early pregnancy decidua

Large granular lymphocytes (LGL) are the most abundant cell type in first trimester human pregnancy decidua. We have shown previously that CD56-positive decidual LGL have cytotoxic activity against the natural killer (NK) target K562, and that this cytotoxicity is augmented by pretreatment with interleukin-2 (IL-2). We now report that flow cytometrically purified populations of CD56-positive decidual LGL have no cytotoxic activity against either the BeWo choriocarcinoma cell line or freshly isolated term trophoblast. Incubation of unfractionated decidual cells with IL-2 induced cytotoxicity against BeWo, but term trophoblast remained resistant to lysis. Both BeWo and trophoblast showed much lower binding frequencies to decidual or peripheral blood cells than K56 targets, and excess trophoblast did not inhibit cytotoxic activity against K562. This suggests that the resistance of trophoblast to lysis by either decidual or peripheral blood LGL is due to the lack of accessible NK target structures on the surface of trophoblast.     

An exoglycosidase-sensitive triggering site on NK cells which is coupled to transmethylation of membrane phospholipids

Glycosidic enzymes were used as probes to analyze the mechanism of NK cell-mediated cytotoxicity. Pretreatment of nylon wool-enriched CBA/J spleen cells, a murine NK clone, or human peripheral blood lymphocytes (PBL) with alpha-mannosidase, an exoglycosidase, led to a marked dose-dependent inhibition of NK lytic activity against YAC-1.2 or K562 tumor cells. Maximal inhibition occurred after a 60-min pretreatment of murine effectors at 37 degrees C, and the kinetics of NK inhibition by alpha-mannosidase was similar to the reported kinetics for enzymatic activity. Released hexose was detected chemically in the supernatant of mouse spleen cells treated with NK inhibitory dose of alpha-mannosidase, and inactivation of enzymatic function with EDTA reversed the NK inhibitory effect. These results suggest that alpha-mannosidase inhibited NK function by virtue of its enzymatic action. Culture of human PBL for 20-hr after treatment with this enzyme led to a greater than 70% recovery in NK lytic function. Recovery was blocked by incorporating tunicamycin, a glycosylation inhibitor of asparagine-linked glycoproteins, into the culture medium. These results suggest that the alpha-mannosidase-sensitive site may be de novo synthesized glycoprotein. Neuraminidase, beta-galactosidase, endo-beta-N-acetylglucosaminidase-D and H, and peptide-N-glycosidase treatments did not inhibit human NK cell lysis of K562 cells. Pretreatment of nylon wool-enriched CBA/J spleen cells or Percoll-enriched human LGL with alpha-mannosidase did not influence their capacity to bind YAC 1.2 target cells or K562 target cells, respectively, Ca++ pulse experiments revealed that the alpha-mannosidase-sensitive site on the NK cells was involved after target-effector binding but before the Ca++ influx. Pretreatment of effector cells with this enzyme which normally occurs after effector-target cell interaction. These results suggest that the phospholipid methylation reaction is coupled to the alpha-mannosidase-sensitive site on the NK cells. By analogy to other physiologic systems, such as histamine release in mast cells, the triggering of phospholipid methylation in the NK cells may serve as a mechanism for signal transduction across the plasma membrane.     

The elevated natural killer sensitivity of targets carrying surface-attached C3 fragments require the availability of the iC3b receptor (CR3) on the effectors

Human serum-treated Raji and Daudi cells were shown to bind C3 fragments on their surface as a consequence of their capacity to activate C via the alternative pathway. C3 molecules were detectable on the cell surfaces up to 24 h after serum exposure. The C3 fragment-coated cells showed increased sensitivity to spontaneous lymphocyte-mediated cytotoxicity. The effector lymphocytes involved in the enhanced cytotoxicity were NK cells with low buoyant density, carrying both CR3 and FcR. Blocking of the FcR and CR3 with F(ab)2 fragments from Leu-11c or Leu-15 mAb, respectively, did not influence the lysis of targets that did not carry C3 fragments. In contrast, the accessibility of CR3 on the effector lymphocytes was essential for the C3 fragment-mediated enhancement of cytotoxicity. In addition to the Leu-15 antibody, N-acetyl-D-Glucosamine, a compound known to block iC3b binding to CR3, also abrogated the C3 fragment-imposed effect. Our previous experiments showed that the C3 fragments bind to acceptor sites on target cells. The present experiments show that the C3 fragments fixed onto the target bind to CR3 on effector cells. These data substantiate the hypothesis that the bivalent C3 fragments, which are fixed on the targets, promote their interaction with lytic lymphocytes by bridging the two cells.     

Demonstration of NK cell-mediated lysis of varicella-zoster virus (VZV)-infected cells: characterization of the effector cells

Infection with varicella-zoster virus (VZV) rendered RAJI cells more susceptible to lysis by non-adherent blood lymphocytes. At an effector to target ratio of 80:1 the mean percentage of 51Cr release of VZV-infected RAJI cells was 41 +/- 12%, whereas that of uninfected RAJI cells was 15 +/- 6%. The increased susceptibility to lysis was associated with increased effector to target conjugate formation in immunofluorescence binding assays. The effector cells cytotoxic for VZV-infected RAJI cells were predominantly Leu-11a+ Leu-4- granular lymphocytes as demonstrated by fluorescence-activated cell sorting. The effector cell active against VZV-infected RAJI cells appeared similar to those active against herpes simplex virus (HSV)-infected cells, because in cold target competition experiments the lysis of 51Cr-labeled VZV-infected RAJI cells was efficiently inhibited by either unlabeled VZV-infected RAJI cells (mean 71% inhibition, 2:1 ratio unlabeled to labeled target) or HSV-infected RAJI cells (mean 69% inhibition) but not by uninfected RAJI cells (mean 10% inhibition). In contrast, competition experiments revealed donor heterogeneity in the overlap between effector cells for VZV- or HSV-infected RAJI vs K-562 cells.     

Inhibition of human natural killer activity by monolayers of primary cell cultures

Some primary and continuous cell cultures were tested for their capacity to regulate human natural killer (NK) activity. Primary cultures of endothelial cells, fetal fibroblasts, adult fibroblasts, amnion epithelial cells, renal parenchymal cells, and ovarian carcinoma cells inhibited NK activity when peripheral blood lymphocytes were preincubated on target cell monolayers for 18 h before testing the cytotoxicity against K-562. The supernatants of the inhibiting cell cultures were not suppressive. Prostaglandins or suppressive lymphocytes were not involved in the phenomenon. The binding capacity of the effector cells was not changed, suggesting that the suppressive signal was targeted at the cytolytic machinery of NK cells. The down-regulating capacity of the cell cultures weakened significantly during subculturing in vitro, and continuous cell lines were not inhibitory. The inactivation of NK cells may be one of the mechanisms by which target cells are protected from NK activity.     

Augmentation of human natural cytotoxic cell activity by leukotriene B4 mediated by enhanced effector-target cell binding and increased lytic efficiency

The addition of leukotriene B4 (LTB4) to cytotoxicity assays measuring natural killer (NK) or natural cytotoxic (NC) cell activities resulted in significantly augmented killing of K562 or herpes simplex virus (HSV)-infected target cells, respectively. Since the mechanism of cytotoxicity implies several steps, including the binding of effectors to targets which is Mg2+-dependent and the programming of lysis of the target which is Ca2+-dependent, we undertook to define the step(s) at which LTB4 acted in augmenting cytotoxicity. Our results showed that LTB4 significantly increased the percentage of effector-target conjugates when K562- or HSV-infected targets were incubated with lymphocytes. Maximal binding occurred at a concentration of LTB4 of 1 X 10(-10) M. Preincubation of lymphocytes and not target cells with LTB4 was sufficient to observe the increased binding. PBML binding to and killing of the NK-resistant target clone I, derived from K562, was not enhanced by LTB4. In the absence of Ca2+, cytotoxicity was impaired and LTB4 could not restore it. Use of a single cell lytic assay demonstrated augmented efficiency of lysis of both K562 and HSV-infected targets in the presence of LTB4. These findings suggest that LTB4 may augment natural cytotoxicity by enhancing target cell recognition by cytotoxic effector cells and subsequently by augmenting their lytic efficiency.     

Analysis of the mechanisms involved in NK resistance induced by a new tumor factor NK-RIF

The mechanisms involved in susceptibility or resistance of neoplasic cells to lysis by NK cells are not well known. We have recently described a 12-kDa factor (NK-RIF), produced and released by different tumor cell lines, making K562 resistant to NK lysis without affecting the cytotoxic function of NK effector cells. In this paper we further study the mechanism involved in NK resistance of K562 mediated by NK-RIF and its biological implications. The results show that NK-RIF does not affect the binding capacity of target and effector cells nor the levels of HLA class I antigen expression on the target cells, as a proof that resistance to NK-mediated lysis is not always associated with a defect in target effector binding or with an increased MHC class I antigen expression. However NK-RIF-treated K562 loses its capacity to induce NK cell activation and the subsequent capacity to release NKCF and makes K562 resistant to lysis by NKCF. Therefore our results show that induction of resistance to NK cytotoxicity can be the result of the modulation of target structures responsible for inducing effector cell activation without affecting target/effector binding molecules. This indicates that the structures involved in adherence and activation of NK cells have a different nature and that molecules other than HLA participate in NK resistance.