Search

Allotypes from four divergent HLA-B families (B8, B15, B16, and B27) were compared for their inhibition of cytolysis by NK cells expressing the NKB1 receptor. Allotypes differing solely at the Bw4/Bw6 region were examined as were a more divergent subset of B15 allotypes. The capacity to interact with NKB1 correlated precisely with possession of a Bw4 sequence motif at residues 77-83, whereas no correlation was made with the peptide-binding specificities of two Bw4 and four Bw6 allotypes of the B15 family. HLA-B allotypes having four different Bw4 motifs were examined and all interact with NKB1. In contrast, HLA-A allotypes, which have a Bw4 motif identical with one of those present in HLA-B, do not. Mutation at leucine 82 and arginine 83, the residues common to Bw4 motifs, shows they contribute to NKB1 interaction but are not essential. Three types of polymorphism are implicated in formation of the ligand recognized by NKB1: ones shared by Bw4 motifs; ones distinguishing Bw4 motifs; and ones outside the Bw4/Bw6 region that distinguish HLA-B from HLA-A.

NKG2 is a small family of type II transmembrane proteins possessing extracellular C-type lectin domains expressed primarily on NK cells. The function of these proteins is unknown. We have developed mAbs that recognize NKG2 family members on Western blots. Examination of cell extracts from NK cell clones demonstrates that individual NKG2 family members are expressed on subsets of NK cells. Because the anti-NKG2 mAb do not react with the cell surface Ag, signaling function was studied by generating cell lines that express a chimeric receptor consisting of a cytoplasmic and transmembrane domain from NKG2-A or NKG2-C fused to the extracellular segment of mouse NKR-P1C (NK1.1 Ag). Transfectants of the rat NK cell line, RNK-16, were tested for the effect of anti-NK1.1 mAb on cytolytic activity against the FcR+ target cell lines, P388D1 and P815. The anti-NK1.1 mAb stimulated lytic activity and calcium mobilization in the NK cell line expressing the NKG2-C/NKR-P1C chimeric receptor. By contrast, anti-NK1.1 inhibited the lytic activity and failed to stimulate a calcium response in cells expressing the NKG2-A/NKR-P1C chimeric receptor. Immunoprecipitation experiments demonstrated that the inhibitory cytoplasmic tyrosine phosphatase SHP-1 selectively associates with the NKG2-A/NKR-P1C chimeric receptor, but not with the stimulatory NKG2-C/NKR-P1C receptor. These data indicate that NKG2-A and NKG2-C deliver, respectively, inhibitory and activating transmembrane signals in NK cells.

CD94 receptors expressed on NK cells have been implicated in the recognition of certain HLA class I allotypes. We now demonstrate that CD94 glycoproteins form disulfide-bonded heterodimers with the NKG2A/B, NKG2C, and NKG2E glycoproteins. NKG2A/B possesses two immunoreceptor tyrosine-based inhibition motif (ITIM) sequences in its cytoplasmic domain, which may be responsible for the inhibitory function of these receptors, whereas other NKG2 proteins lack ITIMs and may potentially transmit positive signals. Structural heterogeneity in the NKG2 gene family and the formation of heterodimers with CD94 provides for the creation of a diverse NK cell repertoire.

IL-15 is a newly described cytokine produced by monocytes and other non-T cells that utilizes the IL-2R beta- and common gamma-chains, thereby stimulating many NK cell functions previously ascribed to IL-2. Thus, IL-15 may promote NK cell activity during innate immune responses, before the activation of T lymphocytes and subsequent production of IL-2. This study investigated the ability of rIL-15 to substitute for rIL-2 in initiating proliferation of resting human NK cells cocultured with various stimulator cells. NK cell proliferation could not be initiated with rIL-15 as the sole costimulatory cytokine. However, NK cell proliferation was initiated with rIL-15 and either rIL-10 or rIL-12, cytokines also produced by monocytes and other APC and implicated in innate immune responses. Individually, rIL-10, rIL-12, and rIL-15 are effective initiators of NK cell proliferation when combined with submitogenic concentrations of rIL-2, indicating their potential involvement in NK cell proliferation at early stages of an Ag-specific T cell immune response. NK cells proliferating in the different cytokine combinations or optimum concentrations of rIL-2 were indistinguishable in terms of phenotype and cytotoxic activity, but differed in whether they secreted IFN-gamma or IL-5. IFN-gamma was secreted in cultures containing rIL-12, whereas IL-5 secretion was dependent upon interaction of IL-2 with the high affinity IL-2R. These results support the notion that NK cell proliferation occurs at different phases of the immune response with the particular cytokine milieu influencing the repertoire of NK cell-secreted cytokines.

The expression of certain MHC class I allotypes by potential target cells can inhibit NK cell-mediated cytotoxicity. We recently identified the NKB1 surface Ag, expressed on T and NK cell subsets, as a putative inhibitory receptor for HLA-B class I molecules possessing the Bw4 serologic epitope. NKB1 is a 70-kDa glycoprotein that after deglycosylation migrates as a 50-kDa protein as determined by SDS-PAGE. A cDNA encoding the NKB1 receptor was cloned from a NKB1+T cell cDNA library by expression in COS-7 cells using the anti-NKB1 mAb DX9. NKB1 is a member of the lg superfamily containing three lg-like domains in the extracellular region and is related to the recently identified family (p58/NKAT) of human NK and T cell surface molecules that appear to function as inhibitory receptors for HLA class I.

In rodents, the NKR-P1 family of glycoproteins are preferentially expressed on NK cells and have been implicated in NK cell function. In this study, we describe the characterization and cloning of a human homologue. Human (h)NKR-P1A cDNA was cloned from a NK cell cDNA library by expression in COS7 cells with the use of the DX1 mAb. hNKR-P1A is a type II membrane glycoprotein with characteristic properties of the C-type lectin superfamily. Comparison of the predicted amino acid of human NKR-P1A with rat and mouse NKR-P1 indicates 46% homology. NKR-P1A is on human chromosome 12, the syntenic of mouse chromosome 6, where the murine NKR-P1 genes are located. All rat NK cells express NKR-P1; however, hNKR-P1A is present on only a subset of human NK cells. Although rodent T cells only infrequently express NKR-P1, hNKR-P1A is present on approximately 25% of adult peripheral blood T cells, including both CD4+ and CD8+ T cells, and is expressed preferentially on adult T cells with a "memory" antigenic phenotype. The anti-hNKR-P1A mAb failed to affect lysis of NK-sensitive targets; however, the spontaneous cytotoxicity mediated by certain NK cell clones against the murine P815 cell target was blocked by anti-hNKR-P1A mAb. Our findings demonstrate that NKR-P1A is a human homologue of the rodent NKR-P1 genes and suggest that this molecule may be involved in NK cell function.

CD40 is a membrane differentiation antigen constitutively expressed on B cells that induces B cell growth and Ig synthesis after ligation with anti-CD40 mAb or with the recently identified CD40 ligand (CD40L). CD40L is rapidly induced on T cells after activation with anti-CD3 mAb or mitogens. While CD40-CD40L interactions are clearly beneficial to B cells, we speculated that a reciprocal costimulation of T cells might also occur. We have used genetic transfection to demonstrate that interactions between human small, resting T cells and CD40+ murine transfectants substantially augmented anti-CD3 induced T cell proliferation and resulted in the generation of CTL. T cell proliferation costimulated by CD40 was IL-2 dependent. The ability of CD40+ transfectants to costimulate T cell proliferation was specific in that VCAM-1+, CD54+, CD72+, CD56+, CD31+, and fas+ transfectants in the same host cells were inactive. CD4+ T cells preferentially responded to CD40 costimulation, whereas CD8+ T cells were substantially less reactive. By contrast, costimulation with B7 transfectants induced equivalent proliferation in the CD4+ and CD8+ T cell subsets. In addition, adult naive and memory T cells, as well as cord blood T cells, were responsive to CD40. These findings suggest that the CD40-CD40L costimulation pathway may allow for selective expansion of CD4+ T cells after interaction with CD40-bearing APC. The relatively restricted expression of CD40 on APC, as well as on medullary and cortical thymic epithelium, indicates a possible role for this interaction in T cell differentiation and activation.

A subset of CD8+ alpha beta-TCR/CD3+ T lymphocytes in adult human peripheral blood lacks expression of CD28, a membrane receptor for the B7/BB1 B cell differentiation Ag that is involved in T cell activation. CD28-8+ T cells were not observed in the thymus and were present at only low frequency in cord blood, suggesting that these cells may represent a type of "memory" population. Consistent with this interpretation, CD28-8+ T cells were morphologically large, granular lymphocytes and expressed CD54 (intercellular adhesion molecule-1), CD58 (lymphocyte function-associated Ag-3 (LFA)), and high levels of CD11a (LFA-1), but did not express Ag associated with acute activation (e.g., HLA-DR, CD25, CD69). Freshly isolated CD28-8+ T lymphocytes mediated potent anti-CD3 redirected cytotoxicity against FcR-bearing targets, demonstrating that the CD3/TCR complex is functional and that these cells possess cytolytic activity. However, the anti-CD3-induced proliferative response of CD28-8+ T cells was substantially less than CD28+8+ T cells and this deficiency could not be overcome by addition of exogenous IL-2. A large panel of T cell clones were produced from the CD28+8+ and CD28-8+ T cell populations by single cell sorting using a flow cytometer. CD28 expression was stable on clones derived from CD28+8+ T lymphocytes, whereas CD28 expression was quite variable and apparently reexpressed on some clones generated from the CD28-8+ T cell population. As with the freshly isolated cells, CD28-8+ T cell clones were cytotoxic, but anergic to anti-CD3- induced proliferation and could not be costimulated using B7 or CD58 (LFA-3) murine L cell transfectants. These results indicate that CD28- and CD28+ T cells may play different roles in an immune response.

Two genes encode the CD16 low affinity IgG FcR. CD16-I (Fc gamma RIII-1) is expressed on PMN as a phosphatidylinositol-glycan anchored glycoprotein. CD16-II (Fc gamma RIII-2) is expressed on NK cells and macrophages as a transmembrane glycoprotein associated with CD3 zeta or Fc epsilon RI-gamma. NK cells spontaneously release soluble CD16-II from the cell surface and this is enhanced by activation with phorbol ester. In this study, we demonstrate that a metalloprotease is involved in the spontaneous and PMA-induced release of CD16-II from NK cells. 1,10-phenanthroline, an inhibitor of Zn(2+)-dependent metalloproteases, efficiently inhibits CD16-II release. 1,7-phenanthroline, an inactive analogue that doesn't chelate Zn2+ or other divalent metal cations, and inhibitors of serine proteases do not affect spontaneous or PMA-induced release of CD16-II. Murine P815 mastocytoma cells transfected with human CD16-II cDNA shed membrane CD16, and 1,10-phenanthroline inhibits this process. P815 transfectants expressing CD16-II molecules with truncated cytoplasmic domains also release soluble receptors, indicating that the cytoplasmic segment of CD16-II is not required for interaction with the protease or the cytoskeleton. By contrast, 1,10-phenanthroline does not inhibit PMA-induced release of CD16-I glycoprotein from PMN, indicating a different mechanism of release for this phosphatidylinositol-glycan anchored molecule. Prior studies have demonstrated that NK cells are activated via the inositol phosphate pathway after engagement of CD16-II by immune complexes or Ig-coated tumor cell targets. A membrane metalloprotease with substrate specificity for CD16-II that is activated by PKC stimulation may provide a mechanism for releasing the immune complex or target from the effector cells and halting signal transduction.

Two genes encode Fc gamma RIII (CD16), a low affinity FcR for IgG. CD16-I is expressed as a phosphatidylinositol glycan-anchored membrane glycoprotein on neutrophils, whereas CD16-II is a transmembrane-linked glycoprotein on NK cells. Membrane anchoring is determined by codon 203. Site-directed mutation of codon 203 and transient expression of these cDNA in COS-7 cells indicated that Phe, Ile, Leu, and Val permit transmembrane expression, whereas Ser, Thr, Tyr, Asn, Gly, Ala, Asp and Lys enable phosphatidylinositol-glycan attachment. Thus, the involvement of amino acid 203 in membrane anchoring cannot be explained simply on the basis of size, charge, or polarity of the amino acid side groups at this site. Efficient expression of CD16-II in COS-7 cells requires co-transfection with either CD3 zeta or Fc epsilon RI-gamma. Truncation of the cytoplasmic segment of CD16 failed to affect association with CD3 zeta. CD3 zeta and Fc epsilon RI-gamma with truncated cytoplasmic segments were also able to facilitate membrane expression of CD16-II, implicating the transmembrane segments as the interaction site between CD16-II and CD3 zeta or Fc epsilon RI-gamma. Prior studies have suggested that the acidic residue in the CD3 zeta transmembrane segment may be important for the association of CD3 zeta complexes. Although site-directed mutation of CD3 zeta-Asp36 to Glu, Leu, or Val retained the ability to permit membrane expression of CD16-II, quantitatively the wild-type CD3 zeta-Asp36 provided optimal levels of expression, consistent with conservation of this amino acid in mouse and human CD3 zeta.

Natural killer (NK) cells and cytomegalovirus have been locked in an evolutionary arms race for millions of years in an attempt to overwhelm each other. Cytomegaloviruses deploy cunning disguises to avoid detection by NK cells. Studies of the mouse model of infection have shown that NK cells deploy multiple mechanisms to deal with mouse cytomegalovirus (MCMV) infection, which involve receptors of the C-lectin type superfamily. Remarkably, these receptors have two additional common features: They map to the same genetic region, known as the NK cell gene complex; and they recognize MHC class I-related structures. While reviewing these attack-counterattack measures, this chapter points to the central role that recognition of the MCMV-infected cells by NK cells plays in host resistance to infection.

NKG2D is a potent activating receptor on natural killer cells, T cells, and macrophages. Mouse NKG2D interacts with two cell surface ligands related to class I MHC molecules: RAE1 and H60. We used soluble versions of NKG2D, RAE1, and H60 to characterize their interactions. RAE1 and H60 each bind NKG2D with nanomolar affinities, indicating tighter binding than most cell surface immune interactions, but NKG2D binds to H60 with approximately 25-fold higher affinity than to RAE1. RAE1 and H60 compete directly for occupancy of NKG2D, and, thus, NKG2D can be occupied by only one ligand at a time. The NKG2D-H60 interaction is more temperature dependent and makes greater use of electrostatic interactions than the NKG2D-RAE1 interaction. The distinct thermodynamic profiles provide insights into the different molecular mechanisms of the binding interactions.

Several inhibitory and activating receptors involved in natural killer cell activation have been characterized. The increasing knowledge about their ligands, including classical MHC class I molecules, non-classical MHC class I molecules and MHC class I-related molecules, is shedding new light on the targets of innate immune recognition. While classical MHC class I molecules are constitutively expressed, some MHC class I-related (MIC) molecules, however, are stress-induced by ill-defined stimuli. Two families of ligands for the human activating NKG2D receptor have been identified. These are the MIC proteins encoded by two highly polymorphic genes within the MHC class I and the retinoic acid-inducible early gene-1-like (also designated UL16-binding) proteins encoded by genes outside the MHC. For the mouse NKG2D receptor, one family, containing at least five distinct ligands, has been described. A better understanding about how targets signal their distress, which renders them susceptible to natural killer (NK)-cell attack, will help to define the role of NK cells in antimicrobial and antitumor immunity and transplantation.

The presence of a negatively charged residue in the transmembrane domain of DAP12 precludes its cell surface expression in the absence of a partner receptor containing a positive charge in its transmembrane domain. We utilized this property of DAP12 to screen a BALB / c macrophage cDNA library for novel molecules that induce cell surface expression of DAP12. By this method, we cloned a cell surface receptor with a single Ig (V) domain, a transmembrane lysine residue, and a short cytoplasmic domain. By homology screening of BALB / c macrophage libraries, we identified a second cDNA for a highly homologous receptor. These receptors appear to be the mouse orthologues of a recently identified human cDNA, TREM-2, so we have designated the receptors as mouse TREM-2a and TREM-2b. By Northern blotting, transcripts for TREM-2 were found in each of three macrophage cell lines but not in a variety of other hematopoietic cell lines. We further demonstrate that TREM-2a is associated with endogenous DAP12 in macrophage cells, and cross-linking of TREM-2a on the surface of macrophages leads to the release of nitric oxide. Our studies define TREM-2 as a receptor family in mouse macrophages and demonstrate the capacity of these receptors to activate macrophage function through DAP12.

Optimal T cell activation and expansion require engagement of the TCR plus costimulatory signals delivered through accessory molecules. SLAM (signaling lymphocytic activation molecule), a 70-kDa costimulatory molecule belonging to the Ig superfamily, was defined as a human cell surface molecule that mediated CD28-independent proliferation of human T cells and IFN-gamma production by human Th1 and Th2 clones. In this study, we describe the cloning of mouse SLAM and the production of mAb against it which reveal its expression on primary mouse T and B cells. Mouse SLAM is expressed on highly polarized Th1 and Th2 populations, and is maintained on Th1, but not on Th2 clones. Anti-mouse SLAM mAb augmented IFN-gamma production by Th1 cells and Th1 clones stimulated through the TCR, but did not induce IFN-gamma production by Th2 cells, nor their production of IL-4 or their proliferation. Mouse SLAM is a 75-kDa glycoprotein that upon tyrosine phosphorylation associates with the src homology 2-domain-containing protein tyrosine phosphatase SHP-2, but not SHP-1. Mouse SLAM also associates with the recently described human SLAM-associated protein. These studies may provide new insights into the regulation of Th1 responses.

The genetic defect in X-linked lymphoproliferative syndrome (XLP) is the Src homology 2 domain-containing protein SAP. SAP constitutively associates with the cell surface molecule, signaling lymphocytic activation molecule (SLAM), and competes with SH2-domain containing protein tyrosine phosphatase-2 (SHP-2) for recruitment to SLAM. SLAM exhibits homology with the mouse cell surface receptor 2B4. The human homologue of 2B4 has now been identified. It is recognized by the c1.7 mAb, a mAb capable of activating human NK cells. Human 2B4 became tyrosine phosphorylated following pervanadate-treatment of transfected cells and recruited SHP-2. SAP was also recruited to 2B4 in activated cells. Importantly, the 2B4-SAP interaction prevented the association between 2B4 and SHP-2. These results suggest that the phenotype of XLP may result from perturbed signaling not only through SLAM, but also other cell surface molecules that utilize SAP as a signaling adaptor protein.

T cell activation represents a balance between positive and negative signals delivered via distinct cell surface molecules. Many cytoplasmic protein tyrosine phosphatases are involved in regulating cellular responses by antagonizing the action of protein tyrosine kinases. CD148 is a receptor-type protein tyrosine phosphatase expressed by all human mononuclear cells. We have investigated the effect of CD148 on TCR-mediated activation of human T cells. Overexpression of wild-type, but not a phosphatase-deficient, CD148 in Jurkat T cells inhibited TCR-mediated activation, evidenced by reduced expression of the early activation Ag CD69, inhibition of tyrosine phosphorylation of many intracellular proteins including the critical protein tyrosine kinase ZAP-70, and impairment of mitogen-activated protein kinase activation. Taken together, these results suggest that CD148 is an important phosphatase involved in negatively regulating the proximal signaling events during activation of Ag-specific T cells.

Following ligation of the TCR and costimulatory molecules such as CD28, T cells proliferate and secrete cytokines. Several other cell surface molecules have been identified that are capable of augmenting activation mediated via the TCR. These include CD2, CD27, CD40 ligand, and signaling lymphocytic activation molecule. Here, we have characterized the expression and function of CD148, a recently identified receptor-type protein tyrosine phosphatase. CD148 is expressed at low levels on resting T cells, but is up-regulated following in vitro activation. Cross-linking CD148 with immobilized anti-CD148 mAb induced vigorous proliferation of anti-CD3 mAb-activated, highly purified peripheral blood T cells in an IL-2-dependent, cyclosporin A-sensitive manner. This effect was greatest after 8 days of in vitro culture, suggesting that this molecule is involved in the latter stages of a T cell response. CD148-induced proliferation was significantly greater for CD8+ T cells than for CD4+ T cells. Thus, CD148 is a receptor-type protein tyrosine phosphatase involved in the activation of T lymphocytes.

Several members of the Ly-49 receptor family inhibit NK cell-mediated lysis of targets expressing appropriate MHC class I molecules. Ly-49D and Ly-49H, two Ly-49 molecules that lack immunoreceptor tyrosine-based inhibitory motifs (ITIM) in their cytoplasmic domains, associate with mouse DAP12, a molecule that possesses an immunoreceptor tyrosine-based activation motif (ITAM). Cotransfection of either Ly-49D or Ly-49H with DAP12 induces surface expression of both Ly-49 and DAP12. The Ly-49/DAP12 complex was coimmunoprecipitated from the transfected cells, demonstrating a physical association of DAP12 with Ly-49D or Ly-49H in the plasma membrane. Stimulation of transfectants with Abs recognizing either Ly-49D or Ly-49H results in cellular activation, as assessed by induction of tyrosine phosphorylation of multiple cellular substrates.

NK cells and T cells express killer cell inhibitory receptors (KIR) recognizing polymorphic MHC class I molecules. Although prior studies have established that MHC class I can protect normal and transformed hematopoietic cells from NK cell lysis, the role of MHC class I on the recognition of solid tumors has been controversial. In this study, we investigated whether interactions of KIR with their ligands on melanoma tumor cells could inhibit tumor cell lysis by NK and gamma delta T cell clones. Ligation of the NK cell receptor KIR3DL1 by HLA-Bw4 allotypes resulted in inhibition of cytotoxicity against HLA-B*4403-transfected melanomas as well as against melanomas endogenously expressing HLA-Bw4 allotypes. Similarly, interactions of KIR2DL2 or KIR2DL3 (KIR2DL2/3) with HLA-Cw3-related allotypes on melanomas resulted in decreased tumor cell lysis. We also investigated whether signaling via KIR affected melanoma recognition by CTL. Introduction of KIR3DL1 molecules into HLA-A*0201-restricted gp100-specific CTL resulted in inhibition of lysis of gp100+ melanomas co-expressing HLA-A*0201 and HLA-Bw4 allotypes. These results suggest that disrupting interactions of KIR with their ligands on tumor cells in vivo may enhance antitumor responses mediated by both innate and adaptive immune effector cells.

Prior studies demonstrated that NK cells isolated from adult peripheral blood kill the HLA-A-, HLA-B-, and HLA-C-deficient B lymphoblastoid cell line 721.221, but many are unable to kill 721.221 cells transfected with HLA-G, a molecule expressed preferentially on fetal cytotrophoblasts. To determine the biologic relevance of this recognition, we established NK cell clones from the placenta and demonstrate that these NK cells also were unable to kill 721.221 cells expressing HLA-G. Recognition of HLA-G by NK cells was prevented in the presence of anti-CD94 mAb, implicating CD94/NKG2 as the predominant inhibitory NK cell receptor for HLA-G used by decidual NK cells. In contrast, mAbs against the killer cell inhibitory receptors recognizing HLA-Cw3-related, HLA-Cw4-related, or HLA-Bw4 ligands did not affect NK cell killing of the HLA-G transfectants.

The presence of T and NK cells in the human fetal liver and the fact that fetal liver hemopoietic progenitor cells develop into T and NK cells suggest a role for the fetal liver compartment in T and NK cell development. In this work, we show that the capacity of fetal liver progenitors to develop into T cells, in a human/mouse fetal thymic organ culture system, is restricted to an immature subset of CD34+ CD38- cells. No T cell-committed precursors are contained within the more differentiated CD34+ CD38+ population. This conclusion is supported by the observations that no TCR-delta gene rearrangements and no pre-TCR-alpha expression can be detected in this population. However, NK cells were derived from CD34+ CD38- and CD34+ CD38+ fetal liver cells cultured in the presence of IL-15, IL-7, and Flt-3 ligand. Eighty to ninety percent of cells arising from the CD34+ CD38+ population expressed the NK cell-associated markers CD56, CD16, CD94, and NKR-P1A. Several subpopulations of NK cell precursors were identified by differential expression of these receptors. Based on the detection of populations with a similar antigenic profile in freshly isolated fetal liver cells, we propose a model of NK cell differentiation. Collectively, our findings suggest that CD34+ cells differentiate into NK cells, but not into mature T cells, in the human fetal liver.

MHC class I glycoproteins possess an invariant site for N-linked oligosaccharide addition at position 86 of the heavy chain. For human HLA-A, -B, and -C class I glycoproteins, position 86 is the only site of N-linked glycosylation. Comparison of the size and relative abundance of oligosaccharides associated with nine HLA-A, -B, or -C allotypes isolated from EBV-transformed B cell lines and mixtures of HLA-A, -B, and -C allotypes isolated from pooled PBLs revealed a very restricted set of structures. Allotypes encoded by the HLA-A and -B loci have two predominant glycan structures that were almost exclusively di-sialylated. In contrast, HLA-C allotypes have four glycan structures, comprising those associated with HLA-A and -B and two additional glycans. Identical oligosaccharides were present on different allotypes of a class I HLA locus, and in particular, HLA-C allotypes defining two inhibitory specificities for NK cells were shown to possess the same set of oligosaccharides. The uniformity of oligosaccharide structure associated with different HLA-A, -B, and -C products and the relative lack of heterogeneity for any given allotype are unusual features for a mammalian glycoprotein. Particularly striking is that such conserved oligosaccharide structures juxtapose the major regions of amino acid sequence variation within the Ag recognition site, including the polymorphisms of the alpha 1 helix that determine the inhibitory ligands for human NK cells.

We have recently established a novel expression cloning system using retroviral vectors. The system is based on a high-efficiency packaging cell line, BOSC23, and a simplified retroviral vector, pBabeX, carrying no selection marker. cDNA libraries, constructed in the pBabeX vector, are transiently transfected into BOSC23 cells. The supernatant contains more than 3X10(6)/mL, which would cover large complexities of cDNA libraries. The retrovirus stock gave 100% infection efficiency in NIH3T3 cells and 5-40% infection efficiency in various hematopoietic cell lines. In contrast to the conventional expression cloning system, in which it is necessary to transfect cDNA libraries transiently into particular cell types such as COS cells, retrovirus-mediated expression cloning allows us to transduce cDNAs into a wide variety of cell types. This method therefore makes it possible to select cells expressing a cDNA of interest by various functional assays. When combined with polymerase chain reaction (PCR)-driven random mutagenesis, this system is also useful in searching for mutations of various molecules that will result in alterations of their functions.

Human NK cells produce IFN-gamma, TNF-alpha, and granulocyte macrophage-CSF when stimulated with susceptible target cells or through the CD16 and CD94 cell surface molecules. This study reports that NK cells also produce IL-5, a cytokine typically produced by Th2 cells, which mediates mobilization and differentiation of eosinophils. Polyclonal NK cell populations and NK cell clones produce IL-5 when stimulated to proliferate with gamma-irradiated MM-170 melanoma cells or JY B-lymphoblastoid cells and rIL-2. IL-5 is produced in cultures generated from freshly isolated NK cells (primary cultures) and when quiescent NK cells from primary cultures are restimulated to proliferate (secondary cultures). Production of IL-5 is on average 8.8-fold greater in secondary cultures compared with primary cultures (n18), suggesting that the ability of NK cells to produce IL-5 matures during primary stimulation. IL-5 secretion, particularly in primary cultures, is augmented by IL-4 and is inhibited by IL-12 and IL-10. By contrast, IL-4 and IL-12 have the reverse effects on IFN-gamma secretion. Cultured NK cells that no longer secrete cytokines can be restimulated to do so with either phorbol 12, 13 dibutyrate and ionomycin or with susceptible target cells in the presence of rIL-2. IL-5 production in these cultures occurs only when NK cells are in an exponential growth phase, whereas IFN-gamma, TNF-alpha, and granulocyte macrophage-CSF are produced also by stimulation of quiescent cells, although to a lesser extent. Furthermore, cytokine production is unrelated to the cytolytic activity of NK cells. In conclusion, proliferating human NK cells have the potential to produce IL-5 with secretion regulated by IL-4, IL-10, and IL-12.

NK cells lyse hematopoietic cells that lack expression of MHC class I molecules on the cell surface. Transfection of certain MHC class I negative cell lines with MHC class I genes renders these cells resistant to NK cell-mediated cytotoxicity. Recently, we described an NK cell receptor, NKB1, that inhibits NK cells from killing target cells expressing Bw4-reactive HLA-B molecules (-B*2705, -B*5101, -B*5801). In this study, we have demonstrated that another structurally distinct NK cell membrane glycoprotein, HP-3E4, is involved in the recognition of certain polymorphic HLA-C molecules (-Cw*0401 and -Cw*1503). NK cell clones co-expressing both the NKB1 and HP-3E4 receptors fail to lyse targets expressing HLA-Cw*0401 and -B*5801, but are able to kill the transfectants in the presence of mAbs against both receptors. These studies demonstrate that a single NK cell clone may possess multiple structurally distinct receptors for different polymorphic HLA class I molecules that function independently.

Signals initiated through both the TCR complex and CD28 are required for optimal activation of T lymphocytes. Recently, it has been demonstrated that CD28 interacts with two different ligands, designated CD80 (B7/B7-1) and CD86 (B70/B7-2). We have produced stable transfectants that express CD80, CD86, or both ligands and have examined their ability to costimulate T cell proliferation, cytokine production, and the generation of CTL. When we used small, resting human peripheral blood T cells as responders, both CD80 and CD86 transfectants efficiently costimulated anti-CD3 mAb-induced proliferation and the secretion of IL-2 and IFN-gamma. Additionally, both CD80 and CD86 transfectants were able to generate functional CTL. The magnitude and kinetics of these responses were similar, which indicates that both ligands provide efficient costimulatory signals. Because many APCs coexpress both CD80 and CD86, we compared the ability of anti-CD80 and anti-CD86 mAbs to inhibit allogeneic MLR stimulated with B lymphoblastoid cell lines and showed that it is necessary to inhibit interactions with both ligands to optimally block CD28-dependent proliferation. Given the limited homology of CD80 and CD86, it was surprising that the binding of CD28-Ig fusion protein to CD80 and that to CD86 transfectants were essentially indistinguishable. Binding of CTLA-4-Ig fusion protein to both transfectants also was quite similar, but was of higher affinity than CD28-Ig binding. Results from these studies indicate that both CD80 and CD86 are potent and similar costimulators of T lymphocytes. Therefore, the role of CD80 and CD86 in an immune response may be determined primarily by their differential expression on APC.

Small, resting human peripheral blood T cells are able to mediate anti-CD3 redirected lysis against murine P815 cells transfected with human B7, a ligand of CD28. We demonstrate that cytotoxicity is mediated by preexisting cytotoxic effectors within the small, resting "memory" T cell population and by the de novo generation of additional CTL within both the "memory" and "virgin" T subsets. This conclusion is based on analysis of the kinetics of the response and the effects of metabolic inhibitors on the generation of CTL function. Memory CD45RO+ T cells demonstrated cytotoxicity within 4 h of coculture with anti-CD3 mAb and B7+ P815 cells and cytolysis was only partially prevented by inhibitors of protein synthesis. By contrast, virgin CD45RO- T cells demonstrated anti-CD3-induced lysis against B7+ P815 targets only after 6 or 8 h of coculture and cytotoxicity was completely prevented by inhibiting protein synthesis. Induction of cytotoxicity was B7 dependent in that parental P815 cells and P815 cells transfected with CD72 and vascular cell adhesion molecule-1, ligands for T cell-associated membrane receptors CD5 and very late activation antigen-4, respectively, did not initiate cytotoxicity. Our studies also revealed cooperation between the CD28/B7 and lymphocyte function-associated Ag-1/intercellular adhesion molecule-1 pathways in the generation of CTL from small, resting T cells. However, after CTL generation, the CD28-B7 interaction was not required for cytotoxic effector cell function. These observations may have important physiologic implications because this would permit activated CTL to lyse targets in vivo that do not express B7, after the CTL were generated by APC that do express B7 or possibly other costimulatory molecules.

NK cells have been defined as CD3-, CD16+, and/or CD56+ lymphocytes that mediate MHC-unrestricted cytotoxicity against certain tumors and virus-infected cells. Although CD3 epsilon transcripts have been detected in some NK clones, it has generally been thought that NK cells do not express CD3 proteins other than zeta which is associated with CD16 (Fc gamma RIII). We demonstrate that adult peripheral blood NK cell lines and clones express cytoplasmic CD3 epsilon proteins, but not CD3 delta or gamma. CD3 epsilon proteins were detected by immunoprecipitation, Western blot analysis, and immunofluorescence using antiserum directed against the cytoplasmic domain of CD3 epsilon. Although resting, adult peripheral blood NK cells have essentially undetectable levels of CD3 epsilon protein, expression was increased substantially after activation. In contrast to adult NK cells, NK cell clones established from human fetal liver express CD3 gamma, delta, and epsilon protein subunits that associate and form CD3 epsilon, gamma and CD3 epsilon, delta complexes in the cytoplasm, but are apparently unable to be transported to the cell surface. These results indicate that expression of CD3 gamma delta epsilon subunits is not restricted to T lymphocytes and supports the possibility that NK and T cells may be derived from common origins.

NK cells and certain CTL can recognize and lyse targets without restriction by the MHC. NK cells do not express CD3/TCR complexes and the membrane receptors participating in MHC-unrestricted cytotoxicity are largely unknown. We demonstrate that YT2C2, a human NK leukemia cell line, expresses the CD28 differentiation Ag and can spontaneously lyse both murine and human cell lines expressing B7, a B cell- activation Ag that is a ligand for CD28. The participation of CD28/B7 interactions in MHC-unrestricted cytotoxicity mediated by YT2C2 cells was demonstrated by correlation of target sensitivity with levels of B7 expression, inhibition of cytotoxicity by anti-CD28 or anti-B7 mAb, and by making both murine and human cell lines susceptible to YT2C2-mediated lysis by genetic transfection with expression vectors containing B7 cDNA. However, CD28/B7 interactions alone were insufficient to initiate cytotoxicity. mAb inhibition experiments and selection of CD54- (intercellular adhesion molecule-1) deficient B cell targets indicated that CD11a/18 (lymphocyte function-associated Ag-1) also cooperated in CD28/B7-dependent cytotoxicity. The requirement for both CD28/B7 and lymphocyte function-associated Ag-1/intercellular adhesion molecule-1 interactions in YT2C2-mediated MHC-unrestricted cytotoxicity was confirmed by demonstrating that efficient lysis of murine L cells required cotransfection with both B7 and intercellular adhesion molecule-1. These findings support the concept that MHC-unrestricted cytotoxicity may not be due to a unique receptor, but may result from interactions between an appropriate array of "adhesion" molecules with their ligands.

Co-stimulation via the CD28 pathway permits small, resting human peripheral-blood T lymphocytes to mediate anti-CD3 monoclonal antibody (MAb) "re-directed" cytotoxicity. The effector cells are contained with the "memory" population of T lymphocytes, identified by expression of the CD45RO differentiation antigen. In this article, we review the requirements for initiating a cytolytic response and speculate on the physiological consequences of this process.

The neural cell adhesion molecule (N-CAM/CD56) is a member of the Ig supergene family that has been shown to mediate homophilic binding. Several isoforms of N-CAM have been identified that are expressed preferentially in different tissues and stages of embryonic development. To examine the primary structure of N-CAM expressed in leukocytes, N-CAM cDNA were generated by polymerase chain reaction from RNA isolated from normal human NK cells and the KG1a hematopoietic leukemia cell line. The sequence of leukocyte-derived N-CAM cDNA was essentially identical with N-CAM cDNA from human neuroblastoma cells that encode the 140-kDa isoform of N-CAM. Inasmuch as N-CAM is preferentially expressed on human NK cells and a subset of T lymphocytes that mediate MHC-unrestricted cell-mediated cytotoxicity, we examined the potential role of N-CAM in cell-mediated cytotoxicity and heterotypic lymphocyte-tumor cell adhesion. N-CAM loss mutants were established from the human N-CAM+ KG1a leukemia cell line, and N-CAM cDNA was transfected into a human colon carcinoma cell line and murine L cells. Using this panel of mutants and transfectants, it was determined that expression of N-CAM on these target cells does not affect susceptibility to resting or IL-2-activated NK cell-mediated cytotoxicity. Moreover, expression of N-CAM in these transfectants failed to induce homotypic or heterotypic cellular adhesion. Collectively, these studies indicate that homophilic N-CAM interactions probably do not mediate a major role in the cytolytic interaction between NK cells and N-CAM+ tumor cell targets.

DAP12 and or FcRγ ITAM signals are required during normal osteoclast (OC) development in vivo and in vitro. To determine whether ITAM deficient mice are resistant to bone loss induced by estrogen deficiency, we performed ovariectomy (OVX) or SHAM surgery on 10–12 week female DAP12−/−, FcRγ−/−, DAP12−/−FcRγ−/− and control C57BL/6 (B6) mice. MicroCT of the distal femur showed statistically significant losses of trabecular bone in DAP12−/−FcRγ−/−, DAP12−/−, FcRγ−/− and B6 OVX groups compared to SHAM groups, BV/TV dropping from 56% to 35% (p

Natural killer (NK) cell have been implicated in immune responses to tumor and viral antigens. We describe here a monoclonal antibody, anti-KC-1, that blocks lysis of NK targets by fresh but not activated NK cells. Anti-KC-1 has no effect on cytotoxic T lymphocyte activity or on antibody-dependent cellular cytotoxicity. This antibody may be useful in the analysis of NK cell activation and the mechanism of lysis.

We examined the antigenic and functional characteristics of human peripheral blood lymphocytes that differentially express the CD16 (Leu-11) and Leu-19 (NKH-1) antigens. Leu-19 is a approximately 220,000 daltons protein expressed on approximately 15% of freshly isolated peripheral blood lymphocytes. Within the Leu-19+ subset, three distinct populations were identified: CD3-,CD16+,Leu-19+ cells; CD3+,CD16-,Leu-19+ cells; and CD3-,CD16-,Leu-19bright+ cells. Both the CD3+,CD16-,Leu-19+ and CD3-,CD16+,Leu-19+ populations mediated non-major histocompatibility complex (MHC)-restricted cytotoxicity against the NK-sensitive tumor cell K562 and were large granular lymphocytes. CD3-,CD16+,Leu-19+ NK cells were the most abundant (comprising approximately 10% of peripheral blood lymphocytes) and the most efficient cytotoxic effectors. The finding that CD3+,Leu 19+ lymphocytes mediated cytotoxicity against K562 unequivocally demonstrates that a unique subset of non-MHC-restricted cytotoxic CD3+ T lymphocytes are present in the peripheral blood of unprimed, normal individuals. However, CD3+,CD16-,Leu-19+ cells comprised less than 5% of peripheral blood lymphocytes, and the cytotoxic activity of this subset was significantly less than CD3-,CD16+,Leu-19+ NK cells. Most CD3+,Leu-19+ T cells co-expressed the CD2, CD8, and CD5 differentiation antigens. The antigenic and functional phenotype of peripheral blood CD3+,Leu-19+ cytotoxic T lymphocytes corresponds to the interleukin 2-dependent CD3+ cell lines that mediate non-MHC-restricted cytotoxicity against NK-sensitive tumor cell targets. A small population of Leu-19bright+ lymphocytes lacking both CD3 and CD16 was also observed. This population (comprising less than 2% of peripheral blood lymphocytes) contained both large agranular lymphocytes and large granular lymphocytes. CD3-,CD16-,Leu-19bright+ lymphocytes also mediate non-MHC-restricted cytotoxicity. The relationship of these CD3-CD16-,Leu-19bright+ lymphocytes to CD3+ T cells or CD16+ NK cells is unknown.

The functional and phenotypic characteristics of cells in human peripheral blood that mediate "natural killer" (NK) cytolysis have been examined with the use of multiparameter flow cytometry analysis and cell sorting. Essentially, all lymphocytes expressing NK and ADCC activity reacted with the anti-Leu-11a monoclonal antibody. The Leu-11a antigen was expressed on cytotoxic large granular lymphocytes (LGL), neutrophils, and basophils, but was not present on B cells, mitogen-activated T lymphoblasts, or Leu-1+ and Leu 4+ resting T cells. Anti-Leu-11a antibody selectively inhibited the binding of FITC heat-aggregated IgG complexes to granulocytes and LGL, and it may recognize a type of Fc receptor on these cells. Two-color FACS cell sorting indicated the existence of four lymphocyte subsets defined by the expression of Leu-11a and Leu-7 antigens. The Leu-11a+, -7- cells were highly active in 4-hr NK assays with the use of 51Cr-labeled K562 as the target. In contrast, the Leu-11a-, -7+ cells demonstrated weak activity and the Leu-11a-, -7- cells demonstrated no activity. The function of the Leu 11a+, -7+ cells varied considerably among several individuals examined. Multiparameter analysis with the use of two-color flow cytometry was used to determine the relationship between the expression of these NK-associated antigens and T and B cell-associated markers. These data indicate that considerable heterogeneity exists within human peripheral lymphocytes with regard to cell phenotype and function, but that several defined cellular subsets can be clearly revealed by using multiparameter FACS analysis and sorting.

In the present study we have identified and characterized three subpopulations of peripheral blood NK cells based on the surface expression of CD56 and CD16. We have designated these subsets CD16neg, CD16dim, and CD16bright according to the relative surface density of CD16. The CD16bright subset comprised about 10% to 15% of PBL, whereas the CD16dim and CD16neg subsets comprise less than 1% of the total lymphocytes. A detailed characterization of these subsets revealed both similarities and differences. The three subsets shared a great deal of phenotypic similarity, expressing CD2, CD7, CD11b, CD38, CD45R, CD18, and the p75 IL-2R on the majority of the cells in each subset. There were, however, several prominent phenotypic differences, particularly in the expression of CD57, CD11c, CD44, CD25, Leu-8, L263, and L265. The CD16neg cells were morphologically large agranular lymphocytes and demonstrated low levels of non-MHC restricted cytolysis of NK-sensitive tumor lines. The CD16dim and CD16bright subsets were large granular lymphocytes and revealed potent cytotoxicity against NK-sensitive targets. All subsets demonstrated IL-2-dependent activation and proliferation; however, the CD16dim and CD16neg subsets were preferentially responsive to very low concentrations of rIL-2. Although rIL-4 effectively inhibited the IL-2-induced cytolytic activation of all three NK cell subsets, only the CD16bright cells showed rIL-4 inhibition of IL-2 dependent proliferation. Cytokine transcription was also differentially regulated in the NK cell subsets after rIL-2 activation. Although TNF-alpha was equally transcribed in each subsets, IFN-gamma and serine protease-HF were preferentially transcribed in the CD16bright NK cells. Based on these results, we propose that these NK cell subsets represent portions of the NK cell differentiation pathway present in the peripheral blood.

The CD69 (Leu-23) activation Ag is a phosphorylated 28 to 32-kDa disulfide-linked homodimer that is rapidly induced after lymphocyte activation. CD69 is not present on the surface of peripheral blood resting T cells, but is constitutively expressed by CD3bright thymocytes. Activation of protein kinase C (PKC) by stimulation of the TCR/CD3 or by phorbol esters directly induces CD69 expression on T cells. In the attempt to elucidate the function of CD69 we investigated the ability of the CD69 glycoprotein to transmit an activation signal. Cross-linking of CD69 by mAb induced a prolonged elevation of intracellular [Ca2+], mostly due to an influx of extracellular Ca2+. This signal alone was unable to effectively activate PKC. When PKC was simultaneously activated by PMA, stimulation of CD69 induced IL-2 and IFN-gamma gene expression, enhancement of CD25 expression, and ultimately IL-2-dependent T cell proliferation. Both CD4+ and CD8+ peripheral T cells responded to CD69-mediated activation. Stimulation of CD69 induced proliferation of thymocytes as well as peripheral T cells, but both required independent PKC activation by PMA. Cyclosporin A, which does not prevent PKC-induced CD69 expression, completely suppressed CD69-induced IL-2 and IFN-gamma gene expression. Although the signal delivered by the CD69 initiates T cell proliferation, it is unable to trigger cytotoxicity programs in CD69+-activated T cells or T cell clones.

Two clonal A-MuLV lymphoma cell lines have the capacity to generate phenotypic variants when grown in vivo as ascites tumors. Variant lines differed from parental lymphoma cells in their expression of enzymatic or cell surface differentiation markers. Parental lines expressed the B220 and Lyb-2 glycoproteins characteristic of pre-B cells and bound B220-specific monoclonal antibodies such as 14.8. The parental cells expressed low levels of TdT activity but did not synthesize detectable mu-heavy chain, a cellular phenotype that may correspond to lymphoid progenitor cells. Three classes of phenotypic variants were recovered from the Thy-1- parental lines: 1) 14.8+, Lyt-1+, Thy-1- cells; 2) 14.8 +/-, Lyt-1+, Thy-1+ cells, and 3) 14.8-, Lyt-1+, Thy-1+ cells. Cell cloning experiments indicated that Thy-1+ variant cells can be recovered within 14 days of in vivo inoculation as a minor proportion (1/10(6] of the tumor cell population and subsequently become the predominant tumor cell population. These clonal tumor lines provide a model for the study of cellular and molecular alterations that occur during neoplastic differentiation and progression in the lymphoid system.

Culture of human thymocytes in interleukin 2 (IL 2) results in the generation of cytotoxic T lymphocytes (CTL) that kill tumor cell targets without major histocompatibility complex (MHC) restriction. Thymic non-MHC restricted CTL expressed Leu-19 antigen, but were generated from thymic precursor cells that lacked expression of Leu-19. In contrast, short term culture in Il 2 of peripheral blood lymphocytes depleted of Leu-19+ lymphocytes did not result in the generation of cytotoxic activity. IL 2 was necessary and sufficient for the generation of cytotoxic thymocytes and induction of Leu-19 antigen expression. Thymic non-MHC restricted CTL were generated from precursor cells expressing CD1, an antigen present on the majority of thymocytes. Furthermore, cytotoxic activity was detected in IL 2 cultured thymocyte populations with an "immature" antigenic phenotype, i.e. CD1+ and CD4+, CD8+. Upon subsequent culture, thymic non-MHC restricted CTL lost expression of CD1, and developed an antigenic phenotype similar to peripheral blood non-MHC-restricted CTL, suggesting that peripheral non-MHC-restricted CTL may originate from these thymic precursors.

Hybridomas were prepared from the spleen of a BALB/c mouse immunized with EL-4 T lymphoma cells. One, designated A1, was found to secrete a monoclonal antibody that reacted with two T lymphoma cells of C57BL origin, EL-4 and C6VLB, but not with normal C57BL/6 splenocytes or thymocytes, C57BL/6 T cell clones, or other T or B lymphomas by complement-mediated cytotoxicity or indirect immunofluorescent staining. Monoclonal antibody (MAb) A1 precipitated a protein that migrated at 85 kD under nonreducing and 43 kD under reducing conditions. The fact that the antigen defined by MAb A1 was a disulfide-linked dimer, together with the essentially clone-specific distribution of the reactive epitope, raised the possibility that the antibody defined an epitope of the antigen receptor. However, several additional observations revealed that the antibody defined a distinct and novel T cell surface structure. MAb 124-40, previously shown to react with the antigen receptor of C6VLB cells, reacted with variants of C6VLB that failed to express the A1 epitope. Sequential immunoprecipitation indicated that MAb A1 and MAb 124-40 reacted with distinct molecular species on C6VLB cells. Endoglycosidase digestion showed that the structure reactive with MAb A1 was not derived from that reactive with MAb 124-40 by addition of N-linked oligosaccharide residues. Two-dimensional gel electrophoretic analysis of precipitates obtained from radioiodinated C6VLB cells with MAb 124-40 resolved the alpha and beta subunits of the antigen receptor. Similar analysis of precipitates obtained with MAb A1 revealed only a single basic chain under reducing conditions, although anomalous mobility suggestive of a second, more acidic chain was observed under nonreducing conditions. Two-dimensional maps of tyrosine-containing chymotryptic peptides of the proteins isolated with MAb A1 and MAb 124-40 were completely different, suggesting that the molecules shared no peptides and were distinct in primary structure. Finally, cross-linking studies performed with a cleavable reagent indicated that the A1 molecule, unlike the antigen receptor defined with MAb 124-40, was not associated with additional, T3-like structures on the surface of C6VLB cells. Although the MAb A1 was unreactive with normal cells in cytotoxicity or staining assays, a molecule of the appropriate size was immunoprecipitated in small amounts from lysates of radioiodinated normal spleen and thymus cells. These data indicate that MAb A1 defines a novel disulfide-linked T cell surface molecule distinct from the antigen receptor.